Preparing for release of 3.3.0 [buildall]

Fortified (-D_FORTIFY_SOURCE=2 for gcc) builds make strlcpy() crash when
its third parameter (size) is larger than the buffer:
  $ rsync -FFXHav '--filter=merge global-rsync-filter' Align-37-43/ xxx
  sending incremental file list
  *** buffer overflow detected ***: terminated

It's in the exclude code in setup_merge_file():
  strlcpy(y, save, MAXPATHLEN);

Note the 'y' pointer was incremented, so it no longer points to memory
with MAXPATHLEN "owned" bytes.

Fix it by remembering the number of copied bytes into the 'save' buffer
and use that instead of MAXPATHLEN which is clearly incorrect.

Fixes #511.

.cirrus.yml

@@ -1,13 +1,14 @@
 freebsd_task:
   name: FreeBSD
   freebsd_instance:
-    image: freebsd-12-1-release-amd64
+    image_family: freebsd-13-1
   env:
     PATH: /usr/local/bin:$PATH
   prep_script:
     - dd if=/dev/zero of=/tmp/zpool bs=1M count=1024
     - zpool create -m `pwd`/testtmp zpool /tmp/zpool
-    - pkg install -y autotools xxhash zstd liblz4
+    - pkg install -y bash autotools m4 xxhash zstd liblz4 wget
+    - wget -O git-version.h https://gist.githubusercontent.com/WayneD/c11243fa374fc64d4e42f2855c8e3827/raw/rsync-git-version.h
   configure_script:
     - CPPFLAGS=-I/usr/local/include/ LDFLAGS=-L/usr/local/lib/ ./configure --disable-md2man
   make_script:
@@ -17,6 +18,6 @@ freebsd_task:
   info_script:
     - rsync --version
   test_script:
-    - RSYNC_MAX_SKIPPED=3 make check
+    - RSYNC_EXPECT_SKIPPED=acls-default,acls,crtimes,protected-regular make check
   ssl_file_list_script:
     - rsync-ssl --no-motd download.samba.org::rsyncftp/ || true

.github/workflows/build.yml

@@ -15,13 +15,14 @@ jobs:
   ubuntu-build:
     runs-on: ubuntu-20.04
     steps:
-    - uses: actions/checkout@v2
+    - uses: actions/checkout@v3
     - name: prep
       run: |
-        sudo apt-get install acl libacl1-dev attr libattr1-dev liblz4-dev libzstd-dev libxxhash-dev python3-cmarkgfm openssl
-        echo "::add-path::/usr/local/bin"
+        sudo apt-get install acl libacl1-dev attr libattr1-dev liblz4-dev libzstd-dev libxxhash-dev python3-cmarkgfm openssl wget
+        wget -O git-version.h https://gist.githubusercontent.com/WayneD/c11243fa374fc64d4e42f2855c8e3827/raw/rsync-git-version.h
+        echo "/usr/local/bin" >>$GITHUB_PATH
     - name: configure
-      run: ./configure
+      run: ./configure --with-rrsync
     - name: make
       run: make
     - name: install
@@ -29,15 +30,15 @@ jobs:
     - name: info
       run: rsync --version
     - name: check
-      run: sudo RSYNC_MAX_SKIPPED=1 make check
+      run: sudo RSYNC_EXPECT_SKIPPED=crtimes make check
     - name: check30
-      run: sudo RSYNC_MAX_SKIPPED=1 make check30
+      run: sudo RSYNC_EXPECT_SKIPPED=crtimes make check30
     - name: check29
-      run: sudo RSYNC_MAX_SKIPPED=1 make check29
+      run: sudo RSYNC_EXPECT_SKIPPED=crtimes make check29
     - name: ssl file list
       run: rsync-ssl --no-motd download.samba.org::rsyncftp/ || true
     - name: save artifact
-      uses: actions/upload-artifact@v2
+      uses: actions/upload-artifact@v3
       with:
         name: ubuntu-bin
         path: |
@@ -46,18 +47,21 @@ jobs:
           rsync.1
           rsync-ssl.1
           rsyncd.conf.5
+          rrsync.1
+          rrsync
 
   macos-build:
     runs-on: macos-latest
     steps:
-    - uses: actions/checkout@v2
+    - uses: actions/checkout@v3
     - name: prep
       run: |
-        brew install automake openssl xxhash zstd lz4
+        brew install automake openssl xxhash zstd lz4 wget
         sudo pip3 install commonmark
-        echo "::add-path::/usr/local/bin"
+        wget -O git-version.h https://gist.githubusercontent.com/WayneD/c11243fa374fc64d4e42f2855c8e3827/raw/rsync-git-version.h
+        echo "/usr/local/bin" >>$GITHUB_PATH
     - name: configure
-      run: CPPFLAGS=-I/usr/local/opt/openssl/include/ LDFLAGS=-L/usr/local/opt/openssl/lib/ ./configure
+      run: CPPFLAGS=-I/usr/local/opt/openssl/include/ LDFLAGS=-L/usr/local/opt/openssl/lib/ ./configure --with-rrsync
     - name: make
       run: make
     - name: install
@@ -65,11 +69,11 @@ jobs:
     - name: info
       run: rsync --version
     - name: check
-      run: sudo make check
+      run: sudo RSYNC_EXPECT_SKIPPED=acls-default,chmod-temp-dir,chown-fake,devices-fake,dir-sgid,protected-regular,xattrs-hlink,xattrs make check
     - name: ssl file list
       run: rsync-ssl --no-motd download.samba.org::rsyncftp/ || true
     - name: save artifact
-      uses: actions/upload-artifact@v2
+      uses: actions/upload-artifact@v3
       with:
         name: macos-bin
         path: |
@@ -78,21 +82,25 @@ jobs:
           rsync.1
           rsync-ssl.1
           rsyncd.conf.5
+          rrsync.1
+          rrsync
 
   cygwin-build:
-    runs-on: windows-latest
+    runs-on: windows-2022
     if: (github.event_name == 'schedule' || contains(github.event.head_commit.message, '[buildall]'))
     steps:
-    - uses: actions/checkout@v2
-    - uses: crazy-max/ghaction-chocolatey@v1.2.2
-      with:
-        args: install -y --no-progress cygwin cyg-get
+    - uses: actions/checkout@v3
+    - name: cygwin
+      run: choco install -y --no-progress cygwin cyg-get
     - name: prep
       run: |
-        cyg-get make autoconf automake gcc-core attr libattr-devel python3 python36-commonmark libzstd-devel liblz4-devel libssl-devel
-        echo "::add-path::C:/tools/cygwin/bin"
+        cyg-get make autoconf automake gcc-core attr libattr-devel python39 python39-pip libzstd-devel liblz4-devel libssl-devel libxxhash0 libxxhash-devel
+        curl.exe -o git-version.h https://gist.githubusercontent.com/WayneD/c11243fa374fc64d4e42f2855c8e3827/raw/rsync-git-version.h
+        echo "C:/tools/cygwin/bin" >>$Env:GITHUB_PATH
+    - name: commonmark
+      run: bash -c 'python3 -mpip install --user commonmark'
     - name: configure
-      run: bash -c './configure --disable-xxhash'
+      run: bash -c './configure --with-rrsync'
     - name: make
       run: bash -c 'make'
     - name: install
@@ -100,11 +108,11 @@ jobs:
     - name: info
       run: bash -c '/usr/local/bin/rsync --version'
     - name: check
-      run: bash -c 'make check'
+      run: bash -c 'RSYNC_EXPECT_SKIPPED=acls-default,acls,chown,devices,dir-sgid,protected-regular make check'
     - name: ssl file list
       run: bash -c 'PATH="/usr/local/bin:$PATH" rsync-ssl --no-motd download.samba.org::rsyncftp/ || true'
     - name: save artifact
-      uses: actions/upload-artifact@v2
+      uses: actions/upload-artifact@v3
       with:
         name: cygwin-bin
         path: |
@@ -113,3 +121,5 @@ jobs:
           rsync.1
           rsync-ssl.1
           rsyncd.conf.5
+          rrsync.1
+          rrsync

.gitignore

@@ -15,9 +15,11 @@ config.status
 aclocal.m4
 /proto.h
 /proto.h-tstamp
-/rsync*.1
-/rsync*.5
+/rsync*.[15]
+/rrsync
+/rrsync*.1
 /rsync*.html
+/rrsync*.html
 /help-rsync*.h
 /default-cvsignore.h
 /default-dont-compress.h
@@ -28,10 +30,12 @@ aclocal.m4
 /conftest*
 /dox
 /getgroups
+/gists
 /gmon.out
 /rsync
 /stunnel-rsyncd.conf
 /shconfig
+/git-version.h
 /testdir
 /tests-dont-exist
 /testtmp

COPYING

@@ -1,7 +1,16 @@
+REGARDING OPENSSL AND XXHASH
+
+In addition, as a special exception, the copyright holders give
+permission to dynamically link rsync with the OpenSSL and xxhash
+libraries when those libraries are being distributed in compliance
+with their license terms, and to distribute a dynamically linked
+combination of rsync and these libraries.  This is also considered
+to be covered under the GPL's System Libraries exception.
+
                     GNU GENERAL PUBLIC LICENSE
                        Version 3, 29 June 2007
 
- Copyright (C) 2007 Free Software Foundation, Inc. <http://fsf.org/>
+ Copyright (C) 2007 Free Software Foundation, Inc. <https://fsf.org/>
  Everyone is permitted to copy and distribute verbatim copies
  of this license document, but changing it is not allowed.
 
@@ -645,7 +654,7 @@ the "copyright" line and a pointer to where the full notice is found.
     GNU General Public License for more details.
 
     You should have received a copy of the GNU General Public License
-    along with this program.  If not, see <http://www.gnu.org/licenses/>.
+    along with this program.  If not, see <https://www.gnu.org/licenses/>.
 
 Also add information on how to contact you by electronic and paper mail.
 
@@ -664,20 +673,11 @@ might be different; for a GUI interface, you would use an "about box".
   You should also get your employer (if you work as a programmer) or school,
 if any, to sign a "copyright disclaimer" for the program, if necessary.
 For more information on this, and how to apply and follow the GNU GPL, see
-<http://www.gnu.org/licenses/>.
+<https://www.gnu.org/licenses/>.
 
   The GNU General Public License does not permit incorporating your program
 into proprietary programs.  If your program is a subroutine library, you
 may consider it more useful to permit linking proprietary applications with
 the library.  If this is what you want to do, use the GNU Lesser General
 Public License instead of this License.  But first, please read
-<http://www.gnu.org/philosophy/why-not-lgpl.html>.
-
-REGARDING OPENSSL AND XXHASH
-
-In addition, as a special exception, the copyright holders give
-permission to dynamically link rsync with the OpenSSL and xxhash
-libraries when those libraries are being distributed in compliance
-with their license terms, and to distribute a dynamically linked
-combination of rsync and these libraries.  This is also considered
-to be covered under the GPL's System Libraries exception.
+<https://www.gnu.org/licenses/why-not-lgpl.html>.

INSTALL.md

@@ -11,16 +11,30 @@ features (the impatient can skip down to the package summary):
 
 You need to have a C compiler installed and optionally a C++ compiler in order
 to try to build some hardware-accelerated checksum routines.  Rsync also needs
-a modern awk, which might be provides via gawk or nawk on some OSes.
+a modern awk, which might be provided via gawk or nawk on some OSes.
+
+## Autoconf & manpages
 
 If you're installing from the git repo (instead of a release tar file) you'll
 also need the GNU autotools (autoconf & automake) and your choice of 2 python3
-markdown libraries: cmarkgfm or commonmark (needed to generate the man pages).
+markdown libraries: cmarkgfm or commonmark (needed to generate the manpages).
 If your OS doesn't provide a python3-cmarkgfm or python3-commonmark package,
 you can run the following to install the commonmark python library for your
 build user (after installing python3's pip package):
 
->     pip3 install --user commonmark
+>     python3 -mpip install --user commonmark
+
+You can test if you've got it fixed by running (from the rsync checkout):
+
+>     ./md-convert --test rsync-ssl.1.md
+
+Alternately, you can avoid generating the manpages by fetching the very latest
+versions (that match the latest git source) from the [generated-files][6] dir.
+One way to do that is to run:
+
+>     ./prepare-source fetchgen
+
+[6]: https://download.samba.org/pub/rsync/generated-files/
 
 ## ACL support
 
@@ -30,7 +44,7 @@ installed to manipulate ACLs and to run the rsync testsuite.
 
 ## Xattr support
 
-To support copying xattr file information, make sure you have an xattr
+To support copying xattr file information, make sure you have an attr
 development library installed. It also helps to have the helper programs
 installed to manipulate xattrs and to run the rsync testsuite.
 
@@ -39,13 +53,14 @@ installed to manipulate xattrs and to run the rsync testsuite.
 The [xxHash library][1] provides extremely fast checksum functions that can
 make the "rsync algorithm" run much more quickly, especially when matching
 blocks in large files.  Installing this development library adds xxhash
-checksums as the default checksum algorithm.
+checksums as the default checksum algorithm.  You'll need at least v0.8.0
+if you want rsync to include the full range of its checksum algorithms.
 
 [1]: https://cyan4973.github.io/xxHash/
 
 ## zstd
 
-The [zstd library][2] compression algorithm that uses a lot less CPU than
+The [zstd library][2] compression algorithm that uses less CPU than
 the default zlib algorithm at the same compression level.  Note that you
 need at least version 1.4, so you might need to skip the zstd compression if
 you can only install a 1.3 release.  Installing this development library
@@ -86,7 +101,7 @@ like.
     >     sudo apt install -y attr libattr1-dev
     >     sudo apt install -y libxxhash-dev
     >     sudo apt install -y libzstd-dev
-    >     sudo apt install -y libzlz4-dev
+    >     sudo apt install -y liblz4-dev
     >     sudo apt install -y libssl-dev
 
  -  For CentOS (use EPEL for python3-pip):
@@ -99,7 +114,16 @@ like.
     >     sudo yum -y install libzstd-devel
     >     sudo yum -y install lz4-devel
     >     sudo yum -y install openssl-devel
-    >     pip3 install --user commonmark
+    >     python3 -mpip install --user commonmark
+
+ -  For Fedora 33:
+
+    >     sudo dnf -y install acl libacl-devel
+    >     sudo dnf -y install attr libattr-devel
+    >     sudo dnf -y install xxhash-devel
+    >     sudo dnf -y install libzstd-devel
+    >     sudo dnf -y install lz4-devel
+    >     sudo dnf -y install openssl-devel
 
  -  For FreeBSD (this assumes that the python3 version is 3.7):
 
@@ -118,11 +142,17 @@ like.
 
  -  For Cygwin (with all cygwin programs stopped, run the appropriate setup program from a cmd shell):
 
-    >     .\setup-x86_64.exe --quiet-mode -P make,gawk,autoconf,automake,gcc-core,python3,python36-commonmark
-    >     .\setup-x86_64.exe --quiet-mode -P attr,libattr-devel
-    >     .\setup-x86_64.exe --quiet-mode -P libzstd-devel
-    >     .\setup-x86_64.exe --quiet-mode -P liblz4-devel
-    >     .\setup-x86_64.exe --quiet-mode -P libssl-devel
+    >     setup-x86_64 --quiet-mode -P make,gawk,autoconf,automake,gcc-core,python38,python38-pip
+    >     setup-x86_64 --quiet-mode -P attr,libattr-devel
+    >     setup-x86_64 --quiet-mode -P libzstd-devel
+    >     setup-x86_64 --quiet-mode -P liblz4-devel
+    >     setup-x86_64 --quiet-mode -P libssl-devel
+
+    Sometimes cygwin has commonmark packaged and sometimes it doesn't. Now that
+    its python38 has stabilized, you could install python38-commonmark. Or just
+    avoid the issue by running this from a bash shell as your build user:
+
+    >     python3 -mpip install --user commonmark
 
 ## Build and install
 
@@ -148,9 +178,9 @@ config.h, or just override them in your /etc/rsyncd.conf file.
 As of 2.4.7, rsync uses Eric Troan's popt option-parsing library.  A
 cut-down copy of a recent release is included in the rsync distribution,
 and will be used if there is no popt library on your build host, or if
-the --with-included-popt option is passed to ./configure.
+the `--with-included-popt` option is passed to ./configure.
 
-If you configure using --enable-maintainer-mode, then rsync will try
+If you configure using `--enable-maintainer-mode`, then rsync will try
 to pop up an xterm on DISPLAY=:0 if it crashes.  You might find this
 useful, but it should be turned off for production builds.
 
@@ -164,7 +194,7 @@ This is helpful when using the branch-from-patch and patch-update scripts
 to maintain the official rsync patches.  If you ever need to build from
 a "detached head" git position then you'll need to manually chdir into
 the build dir to run make.  I also like to create 2 more symlinks in the
-source dir:  ln -s build/rsync . ; ln -s build/testtmp .
+source dir:  `ln -s build/rsync . ; ln -s build/testtmp .`
 
 ## Make compatibility
 
@@ -198,7 +228,7 @@ Some versions of Mac OS X (Darwin) seem to have an IPv6 stack, but do
 not completely implement the "New Sockets" API.
 
 [This site][5] says that Apple started to support IPv6 in 10.2 (Jaguar).  If
-your build fails, try again after running configure with --disable-ipv6.
+your build fails, try again after running configure with `--disable-ipv6`.
 
 [5]: http://www.ipv6.org/impl/mac.html
 

Makefile.in

@@ -6,6 +6,7 @@ exec_prefix=@exec_prefix@
 bindir=@bindir@
 libdir=@libdir@/rsync
 mandir=@mandir@
+with_rrsync=@with_rrsync@
 
 LIBS=@LIBS@
 CC=@CC@
@@ -29,28 +30,30 @@ SHELL=/bin/sh
 .SUFFIXES:
 .SUFFIXES: .c .o
 
-SIMD_x86_64=simd-checksum-x86_64.o
-ASM_x86_64=lib/md5-asm-x86_64.o
+ROLL_SIMD_x86_64=simd-checksum-x86_64.o
+ROLL_ASM_x86_64=simd-checksum-avx2.o
+MD5_ASM_x86_64=lib/md5-asm-x86_64.o
 
-GENFILES=configure.sh aclocal.m4 config.h.in proto.h proto.h-tstamp rsync.1 rsync.1.html \
-	 rsync-ssl.1 rsync-ssl.1.html rsyncd.conf.5 rsyncd.conf.5.html
+GENFILES=configure.sh aclocal.m4 config.h.in rsync.1 rsync.1.html \
+	 rsync-ssl.1 rsync-ssl.1.html rsyncd.conf.5 rsyncd.conf.5.html \
+	 @GEN_RRSYNC@
 HEADERS=byteorder.h config.h errcode.h proto.h rsync.h ifuncs.h itypes.h inums.h \
-	lib/pool_alloc.h lib/mdigest.h lib/md-defines.h version.h
+	lib/pool_alloc.h lib/mdigest.h lib/md-defines.h
 LIBOBJ=lib/wildmatch.o lib/compat.o lib/snprintf.o lib/mdfour.o lib/md5.o \
 	lib/permstring.o lib/pool_alloc.o lib/sysacls.o lib/sysxattrs.o @LIBOBJS@
 zlib_OBJS=zlib/deflate.o zlib/inffast.o zlib/inflate.o zlib/inftrees.o \
 	zlib/trees.o zlib/zutil.o zlib/adler32.o zlib/compress.o zlib/crc32.o
 OBJS1=flist.o rsync.o generator.o receiver.o cleanup.o sender.o exclude.o \
-	util.o util2.o main.o checksum.o match.o syscall.o log.o backup.o delete.o
+	util1.o util2.o main.o checksum.o match.o syscall.o log.o backup.o delete.o
 OBJS2=options.o io.o compat.o hlink.o token.o uidlist.o socket.o hashtable.o \
-	fileio.o batch.o clientname.o chmod.o acls.o xattrs.o
-OBJS3=progress.o pipe.o @ASM@
+	usage.o fileio.o batch.o clientname.o chmod.o acls.o xattrs.o
+OBJS3=progress.o pipe.o @MD5_ASM@ @ROLL_SIMD@ @ROLL_ASM@
 DAEMON_OBJ = params.o loadparm.o clientserver.o access.o connection.o authenticate.o
 popt_OBJS=popt/findme.o  popt/popt.o  popt/poptconfig.o \
 	popt/popthelp.o popt/poptparse.o
-OBJS=$(OBJS1) $(OBJS2) $(OBJS3) @SIMD@ $(DAEMON_OBJ) $(LIBOBJ) @BUILD_ZLIB@ @BUILD_POPT@
+OBJS=$(OBJS1) $(OBJS2) $(OBJS3) $(DAEMON_OBJ) $(LIBOBJ) @BUILD_ZLIB@ @BUILD_POPT@
 
-TLS_OBJ = tls.o syscall.o t_stub.o lib/compat.o lib/snprintf.o lib/permstring.o lib/sysxattrs.o @BUILD_POPT@
+TLS_OBJ = tls.o syscall.o util2.o t_stub.o lib/compat.o lib/snprintf.o lib/permstring.o lib/sysxattrs.o @BUILD_POPT@
 
 # Programs we must have to run the test cases
 CHECK_PROGS = rsync$(EXEEXT) tls$(EXEEXT) getgroups$(EXEEXT) getfsdev$(EXEEXT) \
@@ -67,8 +70,10 @@ CHECK_OBJS=tls.o testrun.o getgroups.o getfsdev.o t_stub.o t_unsafe.o trimslash.
 	$(CC) -I. -I$(srcdir) $(CFLAGS) $(CPPFLAGS) -c $< @CC_SHOBJ_FLAG@
 @OBJ_RESTORE@
 
+# NOTE: consider running "packaging/smart-make" instead of "make" to auto-handle
+# any changes to configure.sh and the main Makefile prior to a "make all".
+all: Makefile rsync$(EXEEXT) stunnel-rsyncd.conf @MAKE_RRSYNC@ @MAKE_MAN@
 .PHONY: all
-all: Makefile rsync$(EXEEXT) stunnel-rsyncd.conf @MAKE_MAN@
 
 .PHONY: install
 install: all
@@ -80,6 +85,10 @@ install: all
 	if test -f rsync.1; then $(INSTALLMAN) -m 644 rsync.1 $(DESTDIR)$(mandir)/man1; fi
 	if test -f rsync-ssl.1; then $(INSTALLMAN) -m 644 rsync-ssl.1 $(DESTDIR)$(mandir)/man1; fi
 	if test -f rsyncd.conf.5; then $(INSTALLMAN) -m 644 rsyncd.conf.5 $(DESTDIR)$(mandir)/man5; fi
+	if test "$(with_rrsync)" = yes; then \
+	    $(INSTALLCMD) -m 755 rrsync $(DESTDIR)$(bindir); \
+	    if test -f rrsync.1; then $(INSTALLMAN) -m 644 rrsync.1 $(DESTDIR)$(mandir)/man1; fi; \
+	fi
 
 install-ssl-daemon: stunnel-rsyncd.conf
 	-$(MKDIR_P) $(DESTDIR)/etc/stunnel
@@ -96,11 +105,13 @@ install-strip:
 rsync$(EXEEXT): $(OBJS)
 	$(CC) $(CFLAGS) $(LDFLAGS) -o $@ $(OBJS) $(LIBS)
 
+rrsync: support/rrsync
+	cp -p $(srcdir)/support/rrsync rrsync
+
 $(OBJS): $(HEADERS)
 $(CHECK_OBJS): $(HEADERS)
 tls.o xattrs.o: lib/sysxattrs.h
-options.o: latest-year.h help-rsync.h help-rsyncd.h
-exclude.o: default-cvsignore.h
+usage.o: version.h latest-year.h help-rsync.h help-rsyncd.h git-version.h default-cvsignore.h
 loadparm.o: default-dont-compress.h daemon-parm.h
 
 flist.o: rounding.h
@@ -132,11 +143,20 @@ rounding.h: rounding.c rsync.h proto.h
 	fi
 	@rm -f rounding.out
 
-simd-checksum-x86_64.o: simd-checksum-x86_64.cpp
-	@$(srcdir)/cmdormsg disable-simd $(CXX) -I. $(CXXFLAGS) $(CPPFLAGS) -c -o $@ $(srcdir)/simd-checksum-x86_64.cpp
+git-version.h: ALWAYS_RUN
+	$(srcdir)/mkgitver
 
-lib/md5-asm-x86_64.o: lib/md5-asm-x86_64.S config.h lib/md-defines.h
-	@$(srcdir)/cmdormsg disable-asm $(CC) -I. @NOEXECSTACK@ -c -o $@ $(srcdir)/lib/md5-asm-x86_64.S
+.PHONY: ALWAYS_RUN
+ALWAYS_RUN:
+
+simd-checksum-x86_64.o: simd-checksum-x86_64.cpp
+	@$(srcdir)/cmd-or-msg disable-roll-simd $(CXX) -I. $(CXXFLAGS) $(CPPFLAGS) -c -o $@ $(srcdir)/simd-checksum-x86_64.cpp
+
+simd-checksum-avx2.o: simd-checksum-avx2.S
+	@$(srcdir)/cmd-or-msg disable-roll-asm $(CC) $(CFLAGS) -I. @NOEXECSTACK@ -c -o $@ $(srcdir)/simd-checksum-avx2.S
+
+lib/md5-asm-x86_64.o: lib/md5-asm-x86_64.S lib/md-defines.h
+	@$(srcdir)/cmd-or-msg disable-md5-asm $(CC) -I. @NOEXECSTACK@ -c -o $@ $(srcdir)/lib/md5-asm-x86_64.S
 
 tls$(EXEEXT): $(TLS_OBJ)
 	$(CC) $(CFLAGS) $(LDFLAGS) -o $@ $(TLS_OBJ) $(LIBS)
@@ -150,11 +170,11 @@ getgroups$(EXEEXT): getgroups.o
 getfsdev$(EXEEXT): getfsdev.o
 	$(CC) $(CFLAGS) $(LDFLAGS) -o $@ getfsdev.o $(LIBS)
 
-TRIMSLASH_OBJ = trimslash.o syscall.o t_stub.o lib/compat.o lib/snprintf.o
+TRIMSLASH_OBJ = trimslash.o syscall.o util2.o t_stub.o lib/compat.o lib/snprintf.o
 trimslash$(EXEEXT): $(TRIMSLASH_OBJ)
 	$(CC) $(CFLAGS) $(LDFLAGS) -o $@ $(TRIMSLASH_OBJ) $(LIBS)
 
-T_UNSAFE_OBJ = t_unsafe.o syscall.o util.o util2.o t_stub.o lib/compat.o lib/snprintf.o lib/wildmatch.o
+T_UNSAFE_OBJ = t_unsafe.o syscall.o util1.o util2.o t_stub.o lib/compat.o lib/snprintf.o lib/wildmatch.o
 t_unsafe$(EXEEXT): $(T_UNSAFE_OBJ)
 	$(CC) $(CFLAGS) $(LDFLAGS) -o $@ $(T_UNSAFE_OBJ) $(LIBS)
 
@@ -162,11 +182,15 @@ t_unsafe$(EXEEXT): $(T_UNSAFE_OBJ)
 conf: configure.sh config.h.in
 
 .PHONY: gen
-gen: conf proto.h man
+gen: conf proto.h man git-version.h
 
 .PHONY: gensend
 gensend: gen
-	rsync -aic $(GENFILES) $${SAMBA_HOST-samba.org}:/home/ftp/pub/rsync/generated-files/
+	if ! diff git-version.h $(srcdir)/gists/rsync-git-version.h >/dev/null; then \
+	    ./rsync -ai git-version.h $(srcdir)/gists/rsync-git-version.h && \
+	    (cd $(srcdir)/gists && git commit --allow-empty-message -m '' rsync-git-version.h && git push) ; \
+	fi
+	rsync -aic $(GENFILES) git-version.h $${SAMBA_HOST-samba.org}:/home/ftp/pub/rsync/generated-files/ || true
 
 aclocal.m4: $(srcdir)/m4/*.m4
 	aclocal -I $(srcdir)/m4
@@ -188,7 +212,7 @@ configure.sh config.h.in: configure.ac aclocal.m4
 	else \
 	    echo "config.h.in has CHANGED."; \
 	fi
-	@if test -f configure.sh.old -o -f config.h.in.old; then \
+	@if test -f configure.sh.old || test -f config.h.in.old; then \
 	    if test "$(MAKECMDGOALS)" = reconfigure; then \
 		echo 'Continuing with "make reconfigure".'; \
 	    else \
@@ -235,22 +259,25 @@ proto.h-tstamp: $(srcdir)/*.c $(srcdir)/lib/compat.c daemon-parm.h
 	$(AWK) -f $(srcdir)/mkproto.awk $(srcdir)/*.c $(srcdir)/lib/compat.c daemon-parm.h
 
 .PHONY: man
-man: rsync.1 rsync-ssl.1 rsyncd.conf.5
+man: rsync.1 rsync-ssl.1 rsyncd.conf.5 @MAKE_RRSYNC_1@
 
-rsync.1: rsync.1.md md2man version.h Makefile
-	@$(srcdir)/maybe-make-man $(srcdir) rsync.1.md
+rsync.1: rsync.1.md md-convert version.h Makefile
+	@$(srcdir)/maybe-make-man rsync.1.md
 
-rsync-ssl.1: rsync-ssl.1.md md2man version.h Makefile
-	@$(srcdir)/maybe-make-man $(srcdir) rsync-ssl.1.md
+rsync-ssl.1: rsync-ssl.1.md md-convert version.h Makefile
+	@$(srcdir)/maybe-make-man rsync-ssl.1.md
 
-rsyncd.conf.5: rsyncd.conf.5.md md2man version.h Makefile
-	@$(srcdir)/maybe-make-man $(srcdir) rsyncd.conf.5.md
+rsyncd.conf.5: rsyncd.conf.5.md md-convert version.h Makefile
+	@$(srcdir)/maybe-make-man rsyncd.conf.5.md
+
+rrsync.1: support/rrsync.1.md md-convert Makefile
+	@$(srcdir)/maybe-make-man support/rrsync.1.md
 
 .PHONY: clean
 clean: cleantests
-	rm -f *~ $(OBJS) $(CHECK_PROGS) $(CHECK_OBJS) $(CHECK_SYMLINKS) \
-		rounding rounding.h *.old rsync*.1 rsync*.5 rsync*.html \
-		daemon-parm.h help-*.h default-*.h
+	rm -f *~ $(OBJS) $(CHECK_PROGS) $(CHECK_OBJS) $(CHECK_SYMLINKS) @MAKE_RRSYNC@ \
+		git-version.h rounding rounding.h *.old rsync*.1 rsync*.5 @MAKE_RRSYNC_1@ \
+		*.html daemon-parm.h help-*.h default-*.h proto.h proto.h-tstamp
 
 .PHONY: cleantests
 cleantests:
@@ -261,16 +288,11 @@ cleantests:
 # the source directory.
 .PHONY: distclean
 distclean: clean
-	rm -f Makefile config.h config.status
-	rm -f stunnel-rsyncd.conf
-	rm -f lib/dummy popt/dummy zlib/dummy
-	rm -f $(srcdir)/Makefile $(srcdir)/config.h $(srcdir)/config.status
-	rm -f $(srcdir)/lib/dummy $(srcdir)/popt/dummy $(srcdir)/zlib/dummy
-	rm -f config.cache config.log
-	rm -f $(srcdir)/config.cache $(srcdir)/config.log
-	rm -f shconfig $(srcdir)/shconfig
-	rm -f $(GENFILES)
-	rm -rf autom4te.cache
+	for dir in $(srcdir) . ; do \
+	    (cd "$$dir" && rm -rf Makefile config.h config.status stunnel-rsyncd.conf \
+	     lib/dummy popt/dummy zlib/dummy config.cache config.log shconfig \
+	     $(GENFILES) autom4te.cache) ; \
+	done
 
 # this target is really just for my use. It only works on a limited
 # range of machines and is used to produce a list of potentially
@@ -280,6 +302,7 @@ finddead:
 	nm *.o */*.o |grep 'U ' | awk '{print $$2}' | sort -u > nmused.txt
 	nm *.o */*.o |grep 'T ' | awk '{print $$3}' | sort -u > nmfns.txt
 	comm -13 nmused.txt nmfns.txt
+	@rm nmused.txt nmfns.txt
 
 # 'check' is the GNU name, 'test' is the name for everybody else :-)
 .PHONY: test

README.md

@@ -23,8 +23,9 @@ options.  To get a complete list of supported options type:
 
     rsync --help
 
-See the manpage for more detailed information.
+See the [manpage][0] for more detailed information.
 
+[0]: https://download.samba.org/pub/rsync/rsync.1
 
 BUILDING AND INSTALLING
 -----------------------
@@ -64,8 +65,8 @@ RSYNC DAEMONS
 -------------
 
 Rsync can also talk to "rsync daemons" which can provide anonymous or
-authenticated rsync.  See the rsyncd.conf(5) man page for details on how
-to setup an rsync daemon.  See the rsync(1) man page for info on how to
+authenticated rsync.  See the rsyncd.conf(5) manpage for details on how
+to setup an rsync daemon.  See the rsync(1) manpage for info on how to
 connect to an rsync daemon.
 
 

SECURITY.md

@@ -0,0 +1,12 @@
+# Security Policy
+
+## Supported Versions
+
+Only the current release of the software is actively supported.  If you need
+help backporting fixes into an older release, feel free to ask.
+
+## Reporting a Vulnerability
+
+Email your vulnerability information to rsync's maintainer:
+
+  Wayne Davison <wayne@opencoder.net>

access.c

@@ -2,7 +2,7 @@
  * Routines to authenticate access to a daemon (hosts allow/deny).
  *
  * Copyright (C) 1998 Andrew Tridgell
- * Copyright (C) 2004-2020 Wayne Davison
+ * Copyright (C) 2004-2022 Wayne Davison
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -20,6 +20,9 @@
 
 #include "rsync.h"
 #include "ifuncs.h"
+#ifdef HAVE_NETGROUP_H
+#include <netgroup.h>
+#endif
 
 static int allow_forward_dns;
 

acls.c

@@ -3,7 +3,7 @@
  *
  * Copyright (C) 1996 Andrew Tridgell
  * Copyright (C) 1996 Paul Mackerras
- * Copyright (C) 2006-2020 Wayne Davison
+ * Copyright (C) 2006-2022 Wayne Davison
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -519,6 +519,7 @@ static int get_rsync_acl(const char *fname, rsync_acl *racl,
 
 		sys_acl_free_acl(sacl);
 		if (!ok) {
+			rsyserr(FERROR_XFER, errno, "get_acl: unpack_smb_acl(%s)", fname);
 			return -1;
 		}
 	} else if (no_acl_syscall_error(errno)) {
@@ -763,6 +764,7 @@ static int recv_rsync_acl(int f, item_list *racl_list, SMB_ACL_TYPE_T type, mode
 #ifdef HAVE_OSX_ACLS
 	/* If we received a superfluous mask, throw it away. */
 	duo_item->racl.mask_obj = NO_ENTRY;
+	(void)mode;
 #else
 	if (duo_item->racl.names.count && duo_item->racl.mask_obj == NO_ENTRY) {
 		/* Mask must be non-empty with lists. */

authenticate.c

@@ -2,7 +2,7 @@
  * Support rsync daemon authentication.
  *
  * Copyright (C) 1998-2000 Andrew Tridgell
- * Copyright (C) 2002-2020 Wayne Davison
+ * Copyright (C) 2002-2022 Wayne Davison
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -24,6 +24,7 @@
 
 extern int read_only;
 extern char *password_file;
+extern struct name_num_obj valid_auth_checksums;
 
 /***************************************************************************
 encode a buffer using base64 - simple and slow algorithm. null terminates
@@ -72,9 +73,9 @@ static void gen_challenge(const char *addr, char *challenge)
 	SIVAL(input, 20, tv.tv_usec);
 	SIVAL(input, 24, getpid());
 
-	sum_init(-1, 0);
+	len = sum_init(valid_auth_checksums.negotiated_nni, 0);
 	sum_update(input, sizeof input);
-	len = sum_end(digest);
+	sum_end(digest);
 
 	base64_encode(digest, len, challenge, 0);
 }
@@ -86,10 +87,10 @@ static void generate_hash(const char *in, const char *challenge, char *out)
 	char buf[MAX_DIGEST_LEN];
 	int len;
 
-	sum_init(-1, 0);
+	len = sum_init(valid_auth_checksums.negotiated_nni, 0);
 	sum_update(in, strlen(in));
 	sum_update(challenge, strlen(challenge));
-	len = sum_end(buf);
+	sum_end(buf);
 
 	base64_encode(buf, len, out, 0);
 }
@@ -238,6 +239,7 @@ char *auth_server(int f_in, int f_out, int module, const char *host,
 	if (!users || !*users)
 		return "";
 
+	negotiate_daemon_auth(f_out, 0);
 	gen_challenge(addr, challenge);
 
 	io_printf(f_out, "%s%s\n", leader, challenge);
@@ -350,6 +352,7 @@ void auth_client(int fd, const char *user, const char *challenge)
 
 	if (!user || !*user)
 		user = "nobody";
+	negotiate_daemon_auth(-1, 1);
 
 	if (!(pass = getpassf(password_file))
 	 && !(pass = getenv("RSYNC_PASSWORD"))) {

backup.c

@@ -2,7 +2,7 @@
  * Backup handling code.
  *
  * Copyright (C) 1999 Andrew Tridgell
- * Copyright (C) 2003-2020 Wayne Davison
+ * Copyright (C) 2003-2022 Wayne Davison
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -304,7 +304,8 @@ int make_backup(const char *fname, BOOL prefer_rename)
 #endif
 
 	if (!ret && !S_ISREG(file->mode)) {
-		rprintf(FINFO, "make_bak: skipping non-regular file %s\n", fname);
+		if (INFO_GTE(NONREG, 1))
+			rprintf(FINFO, "make_bak: skipping non-regular file %s\n", fname);
 		unmake_file(file);
 #ifdef SUPPORT_ACLS
 		uncache_tmp_acls();

batch.c

@@ -3,7 +3,7 @@
  *
  * Copyright (C) 1999 Weiss
  * Copyright (C) 2004 Chris Shoemaker
- * Copyright (C) 2004-2020 Wayne Davison
+ * Copyright (C) 2004-2022 Wayne Davison
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -194,7 +194,7 @@ static int write_opt(const char *opt, const char *arg)
 {
 	int len = strlen(opt);
 	int err = write(batch_sh_fd, " ", 1) != 1;
-	err = write(batch_sh_fd, opt, len) != len ? 1 : 0; 
+	err = write(batch_sh_fd, opt, len) != len ? 1 : 0;
 	if (arg) {
 		err |= write(batch_sh_fd, "=", 1) != 1;
 		err |= write_arg(arg);

byteorder.h

@@ -2,7 +2,7 @@
  * Simple byteorder handling.
  *
  * Copyright (C) 1992-1995 Andrew Tridgell
- * Copyright (C) 2007-2020 Wayne Davison
+ * Copyright (C) 2007-2022 Wayne Davison
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -129,4 +129,3 @@ SIVAL(char *buf, int pos, uint32 val)
 {
 	SIVALu((uchar*)buf, pos, val);
 }
-

checksum.c

@@ -3,7 +3,7 @@
  *
  * Copyright (C) 1996 Andrew Tridgell
  * Copyright (C) 1996 Paul Mackerras
- * Copyright (C) 2004-2020 Wayne Davison
+ * Copyright (C) 2004-2023 Wayne Davison
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -42,41 +42,94 @@ extern int protocol_version;
 extern int proper_seed_order;
 extern const char *checksum_choice;
 
-struct name_num_obj valid_checksums = {
-	"checksum", NULL, NULL, 0, 0, {
+#define NNI_BUILTIN (1<<0)
+#define NNI_EVP (1<<1)
+#define NNI_EVP_OK (1<<2)
+
+struct name_num_item valid_checksums_items[] = {
 #ifdef SUPPORT_XXH3
-		{ CSUM_XXH3_128, "xxh128", NULL },
-		{ CSUM_XXH3_64, "xxh3", NULL },
+	{ CSUM_XXH3_128, 0, "xxh128", NULL },
+	{ CSUM_XXH3_64, 0, "xxh3", NULL },
 #endif
 #ifdef SUPPORT_XXHASH
-		{ CSUM_XXH64, "xxh64", NULL },
-		{ CSUM_XXH64, "xxhash", NULL },
+	{ CSUM_XXH64, 0, "xxh64", NULL },
+	{ CSUM_XXH64, 0, "xxhash", NULL },
 #endif
-		{ CSUM_MD5, "md5", NULL },
-		{ CSUM_MD4, "md4", NULL },
-		{ CSUM_NONE, "none", NULL },
-		{ 0, NULL, NULL }
-	}
+	{ CSUM_MD5, NNI_BUILTIN|NNI_EVP, "md5", NULL },
+	{ CSUM_MD4, NNI_BUILTIN|NNI_EVP, "md4", NULL },
+#ifdef SHA_DIGEST_LENGTH
+	{ CSUM_SHA1, NNI_EVP, "sha1", NULL },
+#endif
+	{ CSUM_NONE, 0, "none", NULL },
+	{ 0, 0, NULL, NULL }
 };
 
-int xfersum_type = 0; /* used for the file transfer checksums */
-int checksum_type = 0; /* used for the pre-transfer (--checksum) checksums */
+struct name_num_obj valid_checksums = {
+	"checksum", NULL, 0, 0, valid_checksums_items
+};
 
-int parse_csum_name(const char *name, int len)
+struct name_num_item valid_auth_checksums_items[] = {
+#ifdef SHA512_DIGEST_LENGTH
+	{ CSUM_SHA512, NNI_EVP, "sha512", NULL },
+#endif
+#ifdef SHA256_DIGEST_LENGTH
+	{ CSUM_SHA256, NNI_EVP, "sha256", NULL },
+#endif
+#ifdef SHA_DIGEST_LENGTH
+	{ CSUM_SHA1, NNI_EVP, "sha1", NULL },
+#endif
+	{ CSUM_MD5, NNI_BUILTIN|NNI_EVP, "md5", NULL },
+	{ CSUM_MD4, NNI_BUILTIN|NNI_EVP, "md4", NULL },
+	{ 0, 0, NULL, NULL }
+};
+
+struct name_num_obj valid_auth_checksums = {
+	"daemon auth checksum", NULL, 0, 0, valid_auth_checksums_items
+};
+
+/* These cannot make use of openssl, so they're marked just as built-in */
+struct name_num_item implied_checksum_md4 =
+    { CSUM_MD4, NNI_BUILTIN, "md4", NULL };
+struct name_num_item implied_checksum_md5 =
+    { CSUM_MD5, NNI_BUILTIN, "md5", NULL };
+
+struct name_num_item *xfer_sum_nni; /* used for the transfer checksum2 computations */
+int xfer_sum_len;
+struct name_num_item *file_sum_nni; /* used for the pre-transfer --checksum computations */
+int file_sum_len, file_sum_extra_cnt;
+
+#ifdef USE_OPENSSL
+const EVP_MD *xfer_sum_evp_md;
+const EVP_MD *file_sum_evp_md;
+EVP_MD_CTX *ctx_evp = NULL;
+#endif
+
+static int initialized_choices = 0;
+
+struct name_num_item *parse_csum_name(const char *name, int len)
 {
 	struct name_num_item *nni;
 
 	if (len < 0 && name)
 		len = strlen(name);
 
+	init_checksum_choices();
+
 	if (!name || (len == 4 && strncasecmp(name, "auto", 4) == 0)) {
-		if (protocol_version >= 30)
-			return CSUM_MD5;
-		if (protocol_version >= 27)
-			return CSUM_MD4_OLD;
-		if (protocol_version >= 21)
-			return CSUM_MD4_BUSTED;
-		return CSUM_MD4_ARCHAIC;
+		if (protocol_version >= 30) {
+			if (!proper_seed_order)
+				return &implied_checksum_md5;
+			name = "md5";
+			len = 3;
+		} else {
+			if (protocol_version >= 27)
+				implied_checksum_md4.num = CSUM_MD4_OLD;
+			else if (protocol_version >= 21)
+				implied_checksum_md4.num = CSUM_MD4_BUSTED;
+			else
+				implied_checksum_md4.num = CSUM_MD4_ARCHAIC;
+			return &implied_checksum_md4;
+		}
 	}
 
 	nni = get_nni_by_name(&valid_checksums, name, len);
@@ -86,44 +139,74 @@ int parse_csum_name(const char *name, int len)
 		exit_cleanup(RERR_UNSUPPORTED);
 	}
 
-	return nni->num;
+	return nni;
 }
 
-static const char *checksum_name(int num)
+#ifdef USE_OPENSSL
+static const EVP_MD *csum_evp_md(struct name_num_item *nni)
 {
-	struct name_num_item *nni = get_nni_by_num(&valid_checksums, num);
+	const EVP_MD *emd;
+	if (!(nni->flags & NNI_EVP))
+		return NULL;
 
-	return nni ? nni->name : num < CSUM_MD4 ? "md4" : "UNKNOWN";
+#ifdef USE_MD5_ASM
+	if (nni->num == CSUM_MD5)
+		emd = NULL;
+	else
+#endif
+		emd = EVP_get_digestbyname(nni->name);
+	if (emd && !(nni->flags & NNI_EVP_OK)) { /* Make sure it works before we advertise it */
+		if (!ctx_evp && !(ctx_evp = EVP_MD_CTX_create()))
+			out_of_memory("csum_evp_md");
+		/* Some routines are marked as legacy and are not enabled in the openssl.cnf file.
+		 * If we can't init the emd, we'll fall back to our built-in code. */
+		if (EVP_DigestInit_ex(ctx_evp, emd, NULL) == 0)
+			emd = NULL;
+		else
+			nni->flags = (nni->flags & ~NNI_BUILTIN) | NNI_EVP_OK;
+	}
+	if (!emd)
+		nni->flags &= ~NNI_EVP;
+	return emd;
 }
+#endif
 
 void parse_checksum_choice(int final_call)
 {
-	if (valid_checksums.negotiated_name)
-		xfersum_type = checksum_type = valid_checksums.negotiated_num;
+	if (valid_checksums.negotiated_nni)
+		xfer_sum_nni = file_sum_nni = valid_checksums.negotiated_nni;
 	else {
 		char *cp = checksum_choice ? strchr(checksum_choice, ',') : NULL;
 		if (cp) {
-			xfersum_type = parse_csum_name(checksum_choice, cp - checksum_choice);
-			checksum_type = parse_csum_name(cp+1, -1);
+			xfer_sum_nni = parse_csum_name(checksum_choice, cp - checksum_choice);
+			file_sum_nni = parse_csum_name(cp+1, -1);
 		} else
-			xfersum_type = checksum_type = parse_csum_name(checksum_choice, -1);
+			xfer_sum_nni = file_sum_nni = parse_csum_name(checksum_choice, -1);
 		if (am_server && checksum_choice)
-			validate_choice_vs_env(NSTR_CHECKSUM, xfersum_type, checksum_type);
+			validate_choice_vs_env(NSTR_CHECKSUM, xfer_sum_nni->num, file_sum_nni->num);
 	}
+	xfer_sum_len = csum_len_for_type(xfer_sum_nni->num, 0);
+	file_sum_len = csum_len_for_type(file_sum_nni->num, 0);
+#ifdef USE_OPENSSL
+	xfer_sum_evp_md = csum_evp_md(xfer_sum_nni);
+	file_sum_evp_md = csum_evp_md(file_sum_nni);
+#endif
 
-	if (xfersum_type == CSUM_NONE)
+	file_sum_extra_cnt = (file_sum_len + EXTRA_LEN - 1) / EXTRA_LEN;
+
+	if (xfer_sum_nni->num == CSUM_NONE)
 		whole_file = 1;
 
 	/* Snag the checksum name for both write_batch's option output & the following debug output. */
-	if (valid_checksums.negotiated_name)
-		checksum_choice = valid_checksums.negotiated_name;
+	if (valid_checksums.negotiated_nni)
+		checksum_choice = valid_checksums.negotiated_nni->name;
 	else if (checksum_choice == NULL)
-		checksum_choice = checksum_name(xfersum_type);
+		checksum_choice = xfer_sum_nni->name;
 
 	if (final_call && DEBUG_GTE(NSTR, am_server ? 3 : 1)) {
 		rprintf(FINFO, "%s%s checksum: %s\n",
 			am_server ? "Server" : "Client",
-			valid_checksums.negotiated_name ? " negotiated" : "",
+			valid_checksums.negotiated_nni ? " negotiated" : "",
 			checksum_choice);
 	}
 }
@@ -143,6 +226,18 @@ int csum_len_for_type(int cst, BOOL flist_csum)
 		return MD4_DIGEST_LEN;
 	  case CSUM_MD5:
 		return MD5_DIGEST_LEN;
+#ifdef SHA_DIGEST_LENGTH
+	  case CSUM_SHA1:
+		return SHA_DIGEST_LENGTH;
+#endif
+#ifdef SHA256_DIGEST_LENGTH
+	  case CSUM_SHA256:
+		return SHA256_DIGEST_LENGTH;
+#endif
+#ifdef SHA512_DIGEST_LENGTH
+	  case CSUM_SHA512:
+		return SHA512_DIGEST_LENGTH;
+#endif
 	  case CSUM_XXH64:
 	  case CSUM_XXH3_64:
 		return 64/8;
@@ -168,6 +263,9 @@ int canonical_checksum(int csum_type)
 		break;
 	  case CSUM_MD4:
 	  case CSUM_MD5:
+	  case CSUM_SHA1:
+	  case CSUM_SHA256:
+	  case CSUM_SHA512:
 		return -1;
 	  case CSUM_XXH64:
 	  case CSUM_XXH3_64:
@@ -179,7 +277,7 @@ int canonical_checksum(int csum_type)
 	return 0;
 }
 
-#ifndef HAVE_SIMD /* See simd-checksum-*.cpp. */
+#ifndef USE_ROLL_SIMD /* See simd-checksum-*.cpp. */
 /*
   a simple 32 bit checksum that can be updated from either end
   (inspired by Mark Adler's Adler-32 checksum)
@@ -202,9 +300,25 @@ uint32 get_checksum1(char *buf1, int32 len)
 }
 #endif
 
+/* The "sum" buffer must be at least MAX_DIGEST_LEN bytes! */
 void get_checksum2(char *buf, int32 len, char *sum)
 {
-	switch (xfersum_type) {
+#ifdef USE_OPENSSL
+	if (xfer_sum_evp_md) {
+		static EVP_MD_CTX *evp = NULL;
+		uchar seedbuf[4];
+		if (!evp && !(evp = EVP_MD_CTX_create()))
+			out_of_memory("get_checksum2");
+		EVP_DigestInit_ex(evp, xfer_sum_evp_md, NULL);
+		if (checksum_seed) {
+			SIVALu(seedbuf, 0, checksum_seed);
+			EVP_DigestUpdate(evp, seedbuf, 4);
+		}
+		EVP_DigestUpdate(evp, (uchar *)buf, len);
+		EVP_DigestFinal_ex(evp, (uchar *)sum, NULL);
+	} else
+#endif
+	switch (xfer_sum_nni->num) {
 #ifdef SUPPORT_XXHASH
 	  case CSUM_XXH64:
 		SIVAL64(sum, 0, XXH64(buf, len, checksum_seed));
@@ -222,40 +336,26 @@ void get_checksum2(char *buf, int32 len, char *sum)
 	  }
 #endif
 	  case CSUM_MD5: {
-		MD5_CTX m5;
+		md_context m5;
 		uchar seedbuf[4];
-		MD5_Init(&m5);
+		md5_begin(&m5);
 		if (proper_seed_order) {
 			if (checksum_seed) {
 				SIVALu(seedbuf, 0, checksum_seed);
-				MD5_Update(&m5, seedbuf, 4);
+				md5_update(&m5, seedbuf, 4);
 			}
-			MD5_Update(&m5, (uchar *)buf, len);
+			md5_update(&m5, (uchar *)buf, len);
 		} else {
-			MD5_Update(&m5, (uchar *)buf, len);
+			md5_update(&m5, (uchar *)buf, len);
 			if (checksum_seed) {
 				SIVALu(seedbuf, 0, checksum_seed);
-				MD5_Update(&m5, seedbuf, 4);
+				md5_update(&m5, seedbuf, 4);
 			}
 		}
-		MD5_Final((uchar *)sum, &m5);
+		md5_result(&m5, (uchar *)sum);
 		break;
 	  }
 	  case CSUM_MD4:
-#ifdef USE_OPENSSL
-	  {
-		MD4_CTX m4;
-		MD4_Init(&m4);
-		MD4_Update(&m4, (uchar *)buf, len);
-		if (checksum_seed) {
-			uchar seedbuf[4];
-			SIVALu(seedbuf, 0, checksum_seed);
-			MD4_Update(&m4, seedbuf, 4);
-		}
-		MD4_Final((uchar *)sum, &m4);
-		break;
-	  }
-#endif
 	  case CSUM_MD4_OLD:
 	  case CSUM_MD4_BUSTED:
 	  case CSUM_MD4_ARCHAIC: {
@@ -288,7 +388,7 @@ void get_checksum2(char *buf, int32 len, char *sum)
 		 * are multiples of 64.  This is fixed by calling mdfour_update()
 		 * even when there are no more bytes.
 		 */
-		if (len - i > 0 || xfersum_type > CSUM_MD4_BUSTED)
+		if (len - i > 0 || xfer_sum_nni->num > CSUM_MD4_BUSTED)
 			mdfour_update(&m, (uchar *)(buf1+i), len-i);
 
 		mdfour_result(&m, (uchar *)sum);
@@ -306,15 +406,33 @@ void file_checksum(const char *fname, const STRUCT_STAT *st_p, char *sum)
 	int32 remainder;
 	int fd;
 
-	memset(sum, 0, MAX_DIGEST_LEN);
-
 	fd = do_open(fname, O_RDONLY, 0);
-	if (fd == -1)
+	if (fd == -1) {
+		memset(sum, 0, file_sum_len);
 		return;
+	}
 
 	buf = map_file(fd, len, MAX_MAP_SIZE, CHUNK_SIZE);
 
-	switch (checksum_type) {
+#ifdef USE_OPENSSL
+	if (file_sum_evp_md) {
+		static EVP_MD_CTX *evp = NULL;
+		if (!evp && !(evp = EVP_MD_CTX_create()))
+			out_of_memory("file_checksum");
+
+		EVP_DigestInit_ex(evp, file_sum_evp_md, NULL);
+
+		for (i = 0; i + CHUNK_SIZE <= len; i += CHUNK_SIZE)
+			EVP_DigestUpdate(evp, (uchar *)map_ptr(buf, i, CHUNK_SIZE), CHUNK_SIZE);
+
+		remainder = (int32)(len - i);
+		if (remainder > 0)
+			EVP_DigestUpdate(evp, (uchar *)map_ptr(buf, i, remainder), remainder);
+
+		EVP_DigestFinal_ex(evp, (uchar *)sum, NULL);
+	} else
+#endif
+	switch (file_sum_nni->num) {
 #ifdef SUPPORT_XXHASH
 	  case CSUM_XXH64: {
 		static XXH64_state_t* state = NULL;
@@ -374,38 +492,21 @@ void file_checksum(const char *fname, const STRUCT_STAT *st_p, char *sum)
 	  }
 #endif
 	  case CSUM_MD5: {
-		MD5_CTX m5;
+		md_context m5;
 
-		MD5_Init(&m5);
+		md5_begin(&m5);
 
 		for (i = 0; i + CHUNK_SIZE <= len; i += CHUNK_SIZE)
-			MD5_Update(&m5, (uchar *)map_ptr(buf, i, CHUNK_SIZE), CHUNK_SIZE);
+			md5_update(&m5, (uchar *)map_ptr(buf, i, CHUNK_SIZE), CHUNK_SIZE);
 
 		remainder = (int32)(len - i);
 		if (remainder > 0)
-			MD5_Update(&m5, (uchar *)map_ptr(buf, i, remainder), remainder);
+			md5_update(&m5, (uchar *)map_ptr(buf, i, remainder), remainder);
 
-		MD5_Final((uchar *)sum, &m5);
+		md5_result(&m5, (uchar *)sum);
 		break;
 	  }
 	  case CSUM_MD4:
-#ifdef USE_OPENSSL
-	  {
-		MD4_CTX m4;
-
-		MD4_Init(&m4);
-
-		for (i = 0; i + CHUNK_SIZE <= len; i += CHUNK_SIZE)
-			MD4_Update(&m4, (uchar *)map_ptr(buf, i, CHUNK_SIZE), CHUNK_SIZE);
-
-		remainder = (int32)(len - i);
-		if (remainder > 0)
-			MD4_Update(&m4, (uchar *)map_ptr(buf, i, remainder), remainder);
-
-		MD4_Final((uchar *)sum, &m4);
-		break;
-	  }
-#endif
 	  case CSUM_MD4_OLD:
 	  case CSUM_MD4_BUSTED:
 	  case CSUM_MD4_ARCHAIC: {
@@ -413,15 +514,15 @@ void file_checksum(const char *fname, const STRUCT_STAT *st_p, char *sum)
 
 		mdfour_begin(&m);
 
-		for (i = 0; i + CHUNK_SIZE <= len; i += CHUNK_SIZE)
-			mdfour_update(&m, (uchar *)map_ptr(buf, i, CHUNK_SIZE), CHUNK_SIZE);
+		for (i = 0; i + CSUM_CHUNK <= len; i += CSUM_CHUNK)
+			mdfour_update(&m, (uchar *)map_ptr(buf, i, CSUM_CHUNK), CSUM_CHUNK);
 
 		/* Prior to version 27 an incorrect MD4 checksum was computed
 		 * by failing to call mdfour_tail() for block sizes that
 		 * are multiples of 64.  This is fixed by calling mdfour_update()
 		 * even when there are no more bytes. */
 		remainder = (int32)(len - i);
-		if (remainder > 0 || checksum_type > CSUM_MD4_BUSTED)
+		if (remainder > 0 || file_sum_nni->num > CSUM_MD4_BUSTED)
 			mdfour_update(&m, (uchar *)map_ptr(buf, i, remainder), remainder);
 
 		mdfour_result(&m, (uchar *)sum);
@@ -429,7 +530,7 @@ void file_checksum(const char *fname, const STRUCT_STAT *st_p, char *sum)
 	  }
 	  default:
 		rprintf(FERROR, "Invalid checksum-choice for --checksum: %s (%d)\n",
-			checksum_name(checksum_type), checksum_type);
+			file_sum_nni->name, file_sum_nni->num);
 		exit_cleanup(RERR_UNSUPPORTED);
 	}
 
@@ -438,30 +539,43 @@ void file_checksum(const char *fname, const STRUCT_STAT *st_p, char *sum)
 }
 
 static int32 sumresidue;
-static union {
-	md_context md;
-#ifdef USE_OPENSSL
-	MD4_CTX m4;
-#endif
-	MD5_CTX m5;
-} ctx;
+static md_context ctx_md;
 #ifdef SUPPORT_XXHASH
 static XXH64_state_t* xxh64_state;
 #endif
 #ifdef SUPPORT_XXH3
 static XXH3_state_t* xxh3_state;
 #endif
-static int cursum_type;
+static struct name_num_item *cur_sum_nni;
+int cur_sum_len;
 
-void sum_init(int csum_type, int seed)
+#ifdef USE_OPENSSL
+static const EVP_MD *cur_sum_evp_md;
+#endif
+
+/* Initialize a hash digest accumulator.  Data is supplied via
+ * sum_update() and the resulting binary digest is retrieved via
+ * sum_end().  This only supports one active sum at a time. */
+int sum_init(struct name_num_item *nni, int seed)
 {
 	char s[4];
 
-	if (csum_type < 0)
-		csum_type = parse_csum_name(NULL, 0);
-	cursum_type = csum_type;
+	if (!nni)
+		nni = parse_csum_name(NULL, 0);
+	cur_sum_nni = nni;
+	cur_sum_len = csum_len_for_type(nni->num, 0);
+#ifdef USE_OPENSSL
+	cur_sum_evp_md = csum_evp_md(nni);
+#endif
 
-	switch (csum_type) {
+#ifdef USE_OPENSSL
+	if (cur_sum_evp_md) {
+		if (!ctx_evp && !(ctx_evp = EVP_MD_CTX_create()))
+			out_of_memory("file_checksum");
+		EVP_DigestInit_ex(ctx_evp, cur_sum_evp_md, NULL);
+	} else
+#endif
+	switch (cur_sum_nni->num) {
 #ifdef SUPPORT_XXHASH
 	  case CSUM_XXH64:
 		if (!xxh64_state && !(xxh64_state = XXH64_createState()))
@@ -482,20 +596,16 @@ void sum_init(int csum_type, int seed)
 		break;
 #endif
 	  case CSUM_MD5:
-		MD5_Init(&ctx.m5);
+		md5_begin(&ctx_md);
 		break;
 	  case CSUM_MD4:
-#ifdef USE_OPENSSL
-		MD4_Init(&ctx.m4);
-#else
-		mdfour_begin(&ctx.md);
+		mdfour_begin(&ctx_md);
 		sumresidue = 0;
-#endif
 		break;
 	  case CSUM_MD4_OLD:
 	  case CSUM_MD4_BUSTED:
 	  case CSUM_MD4_ARCHAIC:
-		mdfour_begin(&ctx.md);
+		mdfour_begin(&ctx_md);
 		sumresidue = 0;
 		SIVAL(s, 0, seed);
 		sum_update(s, 4);
@@ -505,19 +615,19 @@ void sum_init(int csum_type, int seed)
 	  default: /* paranoia to prevent missing case values */
 		exit_cleanup(RERR_UNSUPPORTED);
 	}
+
+	return cur_sum_len;
 }
 
-/**
- * Feed data into an MD4 accumulator, md.  The results may be
- * retrieved using sum_end().  md is used for different purposes at
- * different points during execution.
- *
- * @todo Perhaps get rid of md and just pass in the address each time.
- * Very slightly clearer and slower.
- **/
+/* Feed data into a hash digest accumulator. */
 void sum_update(const char *p, int32 len)
 {
-	switch (cursum_type) {
+#ifdef USE_OPENSSL
+	if (cur_sum_evp_md) {
+		EVP_DigestUpdate(ctx_evp, (uchar *)p, len);
+	} else
+#endif
+	switch (cur_sum_nni->num) {
 #ifdef SUPPORT_XXHASH
 	  case CSUM_XXH64:
 		XXH64_update(xxh64_state, p, len);
@@ -532,39 +642,35 @@ void sum_update(const char *p, int32 len)
 		break;
 #endif
 	  case CSUM_MD5:
-		MD5_Update(&ctx.m5, (uchar *)p, len);
+		md5_update(&ctx_md, (uchar *)p, len);
 		break;
 	  case CSUM_MD4:
-#ifdef USE_OPENSSL
-		MD4_Update(&ctx.m4, (uchar *)p, len);
-		break;
-#endif
 	  case CSUM_MD4_OLD:
 	  case CSUM_MD4_BUSTED:
 	  case CSUM_MD4_ARCHAIC:
 		if (len + sumresidue < CSUM_CHUNK) {
-			memcpy(ctx.md.buffer + sumresidue, p, len);
+			memcpy(ctx_md.buffer + sumresidue, p, len);
 			sumresidue += len;
 			break;
 		}
 
 		if (sumresidue) {
 			int32 i = CSUM_CHUNK - sumresidue;
-			memcpy(ctx.md.buffer + sumresidue, p, i);
-			mdfour_update(&ctx.md, (uchar *)ctx.md.buffer, CSUM_CHUNK);
+			memcpy(ctx_md.buffer + sumresidue, p, i);
+			mdfour_update(&ctx_md, (uchar *)ctx_md.buffer, CSUM_CHUNK);
 			len -= i;
 			p += i;
 		}
 
 		while (len >= CSUM_CHUNK) {
-			mdfour_update(&ctx.md, (uchar *)p, CSUM_CHUNK);
+			mdfour_update(&ctx_md, (uchar *)p, CSUM_CHUNK);
 			len -= CSUM_CHUNK;
 			p += CSUM_CHUNK;
 		}
 
 		sumresidue = len;
 		if (sumresidue)
-			memcpy(ctx.md.buffer, p, sumresidue);
+			memcpy(ctx_md.buffer, p, sumresidue);
 		break;
 	  case CSUM_NONE:
 		break;
@@ -573,13 +679,18 @@ void sum_update(const char *p, int32 len)
 	}
 }
 
-/* NOTE: all the callers of sum_end() pass in a pointer to a buffer that is
- * MAX_DIGEST_LEN in size, so even if the csum-len is shorter that that (i.e.
- * CSUM_MD4_ARCHAIC), we don't have to worry about limiting the data we write
- * into the "sum" buffer. */
-int sum_end(char *sum)
+/* The sum buffer only needs to be as long as the current checksum's digest
+ * len, not MAX_DIGEST_LEN. Note that for CSUM_MD4_ARCHAIC that is the full
+ * MD4_DIGEST_LEN even if the file-list code is going to ignore all but the
+ * first 2 bytes of it. */
+void sum_end(char *sum)
 {
-	switch (cursum_type) {
+#ifdef USE_OPENSSL
+	if (cur_sum_evp_md) {
+		EVP_DigestFinal_ex(ctx_evp, (uchar *)sum, NULL);
+	} else
+#endif
+	switch (cur_sum_nni->num) {
 #ifdef SUPPORT_XXHASH
 	  case CSUM_XXH64:
 		SIVAL64(sum, 0, XXH64_digest(xxh64_state));
@@ -597,22 +708,18 @@ int sum_end(char *sum)
 	  }
 #endif
 	  case CSUM_MD5:
-		MD5_Final((uchar *)sum, &ctx.m5);
+		md5_result(&ctx_md, (uchar *)sum);
 		break;
 	  case CSUM_MD4:
-#ifdef USE_OPENSSL
-		MD4_Final((uchar *)sum, &ctx.m4);
-		break;
-#endif
 	  case CSUM_MD4_OLD:
-		mdfour_update(&ctx.md, (uchar *)ctx.md.buffer, sumresidue);
-		mdfour_result(&ctx.md, (uchar *)sum);
+		mdfour_update(&ctx_md, (uchar *)ctx_md.buffer, sumresidue);
+		mdfour_result(&ctx_md, (uchar *)sum);
 		break;
 	  case CSUM_MD4_BUSTED:
 	  case CSUM_MD4_ARCHAIC:
 		if (sumresidue)
-			mdfour_update(&ctx.md, (uchar *)ctx.md.buffer, sumresidue);
-		mdfour_result(&ctx.md, (uchar *)sum);
+			mdfour_update(&ctx_md, (uchar *)ctx_md.buffer, sumresidue);
+		mdfour_result(&ctx_md, (uchar *)sum);
 		break;
 	  case CSUM_NONE:
 		*sum = '\0';
@@ -620,6 +727,78 @@ int sum_end(char *sum)
 	  default: /* paranoia to prevent missing case values */
 		exit_cleanup(RERR_UNSUPPORTED);
 	}
-
-	return csum_len_for_type(cursum_type, 0);
+}
+
+#if defined SUPPORT_XXH3 || defined USE_OPENSSL
+static void verify_digest(struct name_num_item *nni, BOOL check_auth_list)
+{
+#ifdef SUPPORT_XXH3
+	static int xxh3_result = 0;
+#endif
+#ifdef USE_OPENSSL
+	static int prior_num = 0, prior_flags = 0, prior_result = 0;
+#endif
+
+#ifdef SUPPORT_XXH3
+	if (nni->num == CSUM_XXH3_64 || nni->num == CSUM_XXH3_128) {
+		if (!xxh3_result) {
+			char buf[32816];
+			int j;
+			for (j = 0; j < (int)sizeof buf; j++)
+				buf[j] = ' ' + (j % 96);
+			sum_init(nni, 0);
+			sum_update(buf, 32816);
+			sum_update(buf, 31152);
+			sum_update(buf, 32474);
+			sum_update(buf, 9322);
+			xxh3_result = XXH3_64bits_digest(xxh3_state) != 0xadbcf16d4678d1de ? -1 : 1;
+		}
+		if (xxh3_result < 0)
+			nni->num = CSUM_gone;
+		return;
+	}
+#endif
+
+#ifdef USE_OPENSSL
+	if (BITS_SETnUNSET(nni->flags, NNI_EVP, NNI_BUILTIN|NNI_EVP_OK)) {
+		if (nni->num == prior_num && nni->flags == prior_flags) {
+			nni->flags = prior_result;
+			if (!(nni->flags & NNI_EVP))
+				nni->num = CSUM_gone;
+		} else {
+			prior_num = nni->num;
+			prior_flags = nni->flags;
+			if (!csum_evp_md(nni))
+				nni->num = CSUM_gone;
+			prior_result = nni->flags;
+			if (check_auth_list && (nni = get_nni_by_num(&valid_auth_checksums, prior_num)) != NULL)
+				verify_digest(nni, False);
+		}
+	}
+#endif
+}
+#endif
+
+void init_checksum_choices()
+{
+#if defined SUPPORT_XXH3 || defined USE_OPENSSL
+	struct name_num_item *nni;
+#endif
+
+	if (initialized_choices)
+		return;
+
+#if defined USE_OPENSSL && OPENSSL_VERSION_NUMBER < 0x10100000L
+	OpenSSL_add_all_algorithms();
+#endif
+
+#if defined SUPPORT_XXH3 || defined USE_OPENSSL
+	for (nni = valid_checksums.list; nni->name; nni++)
+		verify_digest(nni, True);
+
+	for (nni = valid_auth_checksums.list; nni->name; nni++)
+		verify_digest(nni, False);
+#endif
+
+	initialized_choices = 1;
 }

clientname.c

@@ -3,7 +3,7 @@
  *
  * Copyright (C) 1992-2001 Andrew Tridgell <tridge@samba.org>
  * Copyright (C) 2001, 2002 Martin Pool <mbp@samba.org>
- * Copyright (C) 2002-2020 Wayne Davison
+ * Copyright (C) 2002-2022 Wayne Davison
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -49,7 +49,7 @@ static char ipaddr_buf[100];
 
 static void client_sockaddr(int fd, struct sockaddr_storage *ss, socklen_t *ss_len);
 static int check_name(const char *ipaddr, const struct sockaddr_storage *ss, char *name_buf, size_t name_buf_size);
-static int valid_ipaddr(const char *s);
+static int valid_ipaddr(const char *s, int allow_scope);
 
 /* Return the IP addr of the client as a string. */
 char *client_addr(int fd)
@@ -73,7 +73,7 @@ char *client_addr(int fd)
 			if ((p = strchr(ipaddr_buf, ' ')) != NULL)
 				*p = '\0';
 		}
-		if (valid_ipaddr(ipaddr_buf))
+		if (valid_ipaddr(ipaddr_buf, True))
 			return ipaddr_buf;
 	}
 
@@ -213,12 +213,14 @@ int read_proxy_protocol_header(int fd)
 				if (size != sizeof hdr.v2.addr.ip4)
 					return 0;
 				inet_ntop(AF_INET, hdr.v2.addr.ip4.src_addr, ipaddr_buf, sizeof ipaddr_buf);
-				return valid_ipaddr(ipaddr_buf);
+				return valid_ipaddr(ipaddr_buf, False);
+#ifdef INET6
 			case PROXY_FAM_TCPv6:
 				if (size != sizeof hdr.v2.addr.ip6)
 					return 0;
 				inet_ntop(AF_INET6, hdr.v2.addr.ip6.src_addr, ipaddr_buf, sizeof ipaddr_buf);
-				return valid_ipaddr(ipaddr_buf);
+				return valid_ipaddr(ipaddr_buf, False);
+#endif
 			default:
 				break;
 			}
@@ -274,7 +276,7 @@ int read_proxy_protocol_header(int fd)
 		if ((sp = strchr(p, ' ')) == NULL)
 			return 0;
 		*sp = '\0';
-		if (!valid_ipaddr(p))
+		if (!valid_ipaddr(p, False))
 			return 0;
 		strlcpy(ipaddr_buf, p, sizeof ipaddr_buf); /* It will always fit when valid. */
 
@@ -282,7 +284,7 @@ int read_proxy_protocol_header(int fd)
 		if ((sp = strchr(p, ' ')) == NULL)
 			return 0;
 		*sp = '\0';
-		if (!valid_ipaddr(p))
+		if (!valid_ipaddr(p, False))
 			return 0;
 		/* Ignore destination address. */
 
@@ -464,7 +466,7 @@ static int check_name(const char *ipaddr, const struct sockaddr_storage *ss, cha
 }
 
 /* Returns 1 for a valid IPv4 or IPv6 addr, or 0 for a bad one. */
-static int valid_ipaddr(const char *s)
+static int valid_ipaddr(const char *s, int allow_scope)
 {
 	int i;
 
@@ -482,6 +484,11 @@ static int valid_ipaddr(const char *s)
 		for (count = 0; count < 8; count++) {
 			if (!*s)
 				return saw_double_colon;
+			if (allow_scope && *s == '%') {
+				if (saw_double_colon)
+					break;
+				return 0;
+			}
 
 			if (strchr(s, ':') == NULL && strchr(s, '.') != NULL) {
 				if ((!saw_double_colon && count != 6) || (saw_double_colon && count > 6))
@@ -507,8 +514,11 @@ static int valid_ipaddr(const char *s)
 			}
 		}
 
-		if (!ipv4_at_end)
-			return !*s;
+		if (!ipv4_at_end) {
+			if (allow_scope && *s == '%')
+				for (s++; isAlNum(s); s++) { }
+			return !*s && s[-1] != '%';
+		}
 	}
 
 	/* IPv4 */

clientserver.c

@@ -3,7 +3,7 @@
  *
  * Copyright (C) 1998-2001 Andrew Tridgell <tridge@samba.org>
  * Copyright (C) 2001-2002 Martin Pool <mbp@samba.org>
- * Copyright (C) 2002-2020 Wayne Davison
+ * Copyright (C) 2002-2022 Wayne Davison
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -47,6 +47,7 @@ extern int protocol_version;
 extern int io_timeout;
 extern int no_detach;
 extern int write_batch;
+extern int old_style_args;
 extern int default_af_hint;
 extern int logfile_format_has_i;
 extern int logfile_format_has_o_or_i;
@@ -66,6 +67,7 @@ extern uid_t our_uid;
 extern gid_t our_gid;
 
 char *auth_user;
+char *daemon_auth_choices;
 int read_only = 0;
 int module_id = -1;
 int pid_file_fd = -1;
@@ -148,13 +150,9 @@ int start_socket_client(char *host, int remote_argc, char *remote_argv[],
 static int exchange_protocols(int f_in, int f_out, char *buf, size_t bufsiz, int am_client)
 {
 	int remote_sub = -1;
-#if SUBPROTOCOL_VERSION != 0
-	int our_sub = protocol_version < PROTOCOL_VERSION ? 0 : SUBPROTOCOL_VERSION;
-#else
-	int our_sub = 0;
-#endif
+	int our_sub = get_subprotocol_version();
 
-	io_printf(f_out, "@RSYNCD: %d.%d\n", protocol_version, our_sub);
+	output_daemon_greeting(f_out, am_client);
 	if (!am_client) {
 		char *motd = lp_motd_file();
 		if (motd && *motd) {
@@ -186,16 +184,30 @@ static int exchange_protocols(int f_in, int f_out, char *buf, size_t bufsiz, int
 	}
 
 	if (remote_sub < 0) {
-		if (remote_protocol == 30) {
+		if (remote_protocol >= 30) {
 			if (am_client)
-				rprintf(FERROR, "rsync: server is speaking an incompatible beta of protocol 30\n");
+				rprintf(FERROR, "rsync: the server omitted the subprotocol value: %s\n", buf);
 			else
-				io_printf(f_out, "@ERROR: your client is speaking an incompatible beta of protocol 30\n");
+				io_printf(f_out, "@ERROR: your client omitted the subprotocol value: %s\n", buf);
 			return -1;
 		}
 		remote_sub = 0;
 	}
 
+	daemon_auth_choices = strchr(buf + 9, ' ');
+	if (daemon_auth_choices) {
+		char *cp;
+		daemon_auth_choices = strdup(daemon_auth_choices + 1);
+		if ((cp = strchr(daemon_auth_choices, '\n')) != NULL)
+			*cp = '\0';
+	} else if (remote_protocol > 31) {
+		if (am_client)
+			rprintf(FERROR, "rsync: the server omitted the digest name list: %s\n", buf);
+		else
+			io_printf(f_out, "@ERROR: your client omitted the digest name list: %s\n", buf);
+		return -1;
+	}
+
 	if (protocol_version > remote_protocol) {
 		protocol_version = remote_protocol;
 		if (remote_sub)
@@ -288,20 +300,45 @@ int start_inband_exchange(int f_in, int f_out, const char *user, int argc, char
 
 	sargs[sargc++] = ".";
 
+	if (!old_style_args)
+		snprintf(line, sizeof line, " %.*s/", modlen, modname);
+
 	while (argc > 0) {
 		if (sargc >= MAX_ARGS - 1) {
 		  arg_overflow:
 			rprintf(FERROR, "internal: args[] overflowed in do_cmd()\n");
 			exit_cleanup(RERR_SYNTAX);
 		}
-		if (strncmp(*argv, modname, modlen) == 0
-		 && argv[0][modlen] == '\0')
+		if (strncmp(*argv, modname, modlen) == 0 && argv[0][modlen] == '\0')
 			sargs[sargc++] = modname; /* we send "modname/" */
-		else if (**argv == '-') {
-			if (asprintf(sargs + sargc++, "./%s", *argv) < 0)
-				out_of_memory("start_inband_exchange");
-		} else
-			sargs[sargc++] = *argv;
+		else {
+			char *arg = *argv;
+			int extra_chars = *arg == '-' ? 2 : 0; /* a leading dash needs a "./" prefix. */
+			/* If --old-args was not specified, make sure that the arg won't split at a mod name! */
+			if (!old_style_args && (p = strstr(arg, line)) != NULL) {
+				do {
+					extra_chars += 2;
+				} while ((p = strstr(p+1, line)) != NULL);
+			}
+			if (extra_chars) {
+				char *f = arg;
+				char *t = arg = new_array(char, strlen(arg) + extra_chars + 1);
+				if (*f == '-') {
+					*t++ = '.';
+					*t++ = '/';
+				}
+				while (*f) {
+					if (*f == ' ' && strncmp(f, line, modlen+2) == 0) {
+						*t++ = '[';
+						*t++ = *f++;
+						*t++ = ']';
+					} else
+						*t++ = *f++;
+				}
+				*t = '\0';
+			}
+			sargs[sargc++] = arg;
+		}
 		argv++;
 		argc--;
 	}
@@ -355,7 +392,7 @@ int start_inband_exchange(int f_in, int f_out, const char *user, int argc, char
 
 	if (rl_nulls) {
 		for (i = 0; i < sargc; i++) {
-			if (!sargs[i]) /* stop at --protect-args NULL */
+			if (!sargs[i]) /* stop at --secluded-args NULL */
 				break;
 			write_sbuf(f_out, sargs[i]);
 			write_byte(f_out, 0);
@@ -380,7 +417,7 @@ int start_inband_exchange(int f_in, int f_out, const char *user, int argc, char
 	return 0;
 }
 
-#ifdef HAVE_PUTENV
+#if defined HAVE_SETENV || defined HAVE_PUTENV
 static int read_arg_from_pipe(int fd, char *buf, int limit)
 {
 	char *bp = buf, *eob = buf + limit - 1;
@@ -403,25 +440,59 @@ static int read_arg_from_pipe(int fd, char *buf, int limit)
 }
 #endif
 
-static void set_env_str(const char *var, const char *str)
+void set_env_str(const char *var, const char *str)
 {
+#ifdef HAVE_SETENV
+	if (setenv(var, str, 1) < 0)
+		out_of_memory("set_env_str");
+#else
 #ifdef HAVE_PUTENV
 	char *mem;
 	if (asprintf(&mem, "%s=%s", var, str) < 0)
 		out_of_memory("set_env_str");
 	putenv(mem);
+#else
+	(void)var;
+	(void)str;
+#endif
 #endif
 }
 
+#if defined HAVE_SETENV || defined HAVE_PUTENV
+
+static void set_envN_str(const char *var, int num, const char *str)
+{
+#ifdef HAVE_SETENV
+	char buf[128];
+	(void)snprintf(buf, sizeof buf, "%s%d", var, num);
+	if (setenv(buf, str, 1) < 0)
+		out_of_memory("set_env_str");
+#else
 #ifdef HAVE_PUTENV
+	char *mem;
+	if (asprintf(&mem, "%s%d=%s", var, num, str) < 0)
+		out_of_memory("set_envN_str");
+	putenv(mem);
+#endif
+#endif
+}
+
 void set_env_num(const char *var, long num)
 {
+#ifdef HAVE_SETENV
+	char val[64];
+	(void)snprintf(val, sizeof val, "%ld", num);
+	if (setenv(var, val, 1) < 0)
+		out_of_memory("set_env_str");
+#else
+#ifdef HAVE_PUTENV
 	char *mem;
 	if (asprintf(&mem, "%s=%ld", var, num) < 0)
 		out_of_memory("set_env_num");
 	putenv(mem);
-}
 #endif
+#endif
+}
 
 /* Used for "early exec", "pre-xfer exec", and the "name converter" script. */
 static pid_t start_pre_exec(const char *cmd, int *arg_fd_ptr, int *error_fd_ptr)
@@ -451,15 +522,13 @@ static pid_t start_pre_exec(const char *cmd, int *arg_fd_ptr, int *error_fd_ptr)
 		set_env_str("RSYNC_REQUEST", buf);
 
 		for (j = 0; ; j++) {
-			char *p;
 			len = read_arg_from_pipe(arg_fd, buf, BIGPATHBUFLEN);
 			if (len <= 0) {
 				if (!len)
 					break;
 				_exit(1);
 			}
-			if (asprintf(&p, "RSYNC_ARG%d=%s", j, buf) >= 0)
-				putenv(p);
+			set_envN_str("RSYNC_ARG", j, buf);
 		}
 
 		dup2(arg_fd, STDIN_FILENO);
@@ -490,6 +559,8 @@ static pid_t start_pre_exec(const char *cmd, int *arg_fd_ptr, int *error_fd_ptr)
 	return pid;
 }
 
+#endif
+
 static void write_pre_exec_args(int write_fd, char *request, char **early_argv, char **argv, int exec_type)
 {
 	int j = 0;
@@ -630,7 +701,7 @@ static int rsync_module(int f_in, int f_out, int i, const char *addr, const char
 	int set_uid;
 	char *p, *err_msg = NULL;
 	char *name = lp_name(i);
-	int use_chroot = lp_use_chroot(i);
+	int use_chroot = lp_use_chroot(i); /* might be 1 (yes), 0 (no), or -1 (unset) */
 	int ret, pre_exec_arg_fd = -1, pre_exec_error_fd = -1;
 	int save_munge_symlinks;
 	pid_t pre_exec_pid = 0;
@@ -755,6 +826,20 @@ static int rsync_module(int f_in, int f_out, int i, const char *addr, const char
 		io_printf(f_out, "@ERROR: no path setting.\n");
 		return -1;
 	}
+	if (use_chroot < 0) {
+		if (strstr(module_dir, "/./") != NULL)
+			use_chroot = 1; /* The module is expecting a chroot inner & outer path. */
+		else if (chroot("/") < 0) {
+			rprintf(FLOG, "chroot test failed: %s. "
+				      "Switching 'use chroot' from unset to false.\n",
+				      strerror(errno));
+			use_chroot = 0;
+		} else {
+			if (chdir("/") < 0)
+			    rsyserr(FLOG, errno, "chdir(\"/\") failed");
+			use_chroot = 1;
+		}
+	}
 	if (use_chroot) {
 		if ((p = strstr(module_dir, "/./")) != NULL) {
 			*p = '\0'; /* Temporary... */
@@ -809,7 +894,7 @@ static int rsync_module(int f_in, int f_out, int i, const char *addr, const char
 
 	log_init(1);
 
-#ifdef HAVE_PUTENV
+#if defined HAVE_SETENV || defined HAVE_PUTENV
 	if ((*lp_early_exec(module_id) || *lp_prexfer_exec(module_id)
 	  || *lp_postxfer_exec(module_id) || *lp_name_converter(module_id))
 	 && !getenv("RSYNC_NO_XFER_EXEC")) {
@@ -891,20 +976,8 @@ static int rsync_module(int f_in, int f_out, int i, const char *addr, const char
 	}
 
 	if (use_chroot) {
-		/*
-		 * XXX: The 'use chroot' flag is a fairly reliable
-		 * source of confusion, because it fails under two
-		 * important circumstances: running as non-root,
-		 * running on Win32 (or possibly others).  On the
-		 * other hand, if you are running as root, then it
-		 * might be better to always use chroot.
-		 *
-		 * So, perhaps if we can't chroot we should just issue
-		 * a warning, unless a "require chroot" flag is set,
-		 * in which case we fail.
-		 */
 		if (chroot(module_chdir)) {
-			rsyserr(FLOG, errno, "chroot %s failed", module_chdir);
+			rsyserr(FLOG, errno, "chroot(\"%s\") failed", module_chdir);
 			io_printf(f_out, "@ERROR: chroot failed\n");
 			return -1;
 		}
@@ -913,7 +986,7 @@ static int rsync_module(int f_in, int f_out, int i, const char *addr, const char
 
 	if (!change_dir(module_chdir, CD_NORMAL))
 		return path_failure(f_out, module_chdir, True);
-	if (module_dirlen || (!use_chroot && !*lp_daemon_chroot()))
+	if (module_dirlen)
 		sanitize_paths = 1;
 
 	if ((munge_symlinks = lp_munge_symlinks(module_id)) < 0)
@@ -1228,8 +1301,12 @@ int start_daemon(int f_in, int f_out)
 	p = lp_daemon_chroot();
 	if (*p) {
 		log_init(0); /* Make use we've initialized syslog before chrooting. */
-		if (chroot(p) < 0 || chdir("/") < 0) {
-			rsyserr(FLOG, errno, "daemon chroot %s failed", p);
+		if (chroot(p) < 0) {
+			rsyserr(FLOG, errno, "daemon chroot(\"%s\") failed", p);
+			return -1;
+		}
+		if (chdir("/") < 0) {
+			rsyserr(FLOG, errno, "daemon chdir(\"/\") failed");
 			return -1;
 		}
 	}
@@ -1451,7 +1528,7 @@ int daemon_main(void)
 	log_init(0);
 
 	rprintf(FLOG, "rsyncd version %s starting, listening on port %d\n",
-		RSYNC_VERSION, rsync_port);
+		rsync_version(), rsync_port);
 	/* TODO: If listening on a particular address, then show that
 	 * address too.  In fact, why not just do getnameinfo on the
 	 * local address??? */

compat.c

@@ -3,7 +3,7 @@
  *
  * Copyright (C) Andrew Tridgell 1996
  * Copyright (C) Paul Mackerras 1996
- * Copyright (C) 2004-2020 Wayne Davison
+ * Copyright (C) 2004-2022 Wayne Davison
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -21,6 +21,7 @@
 
 #include "rsync.h"
 #include "itypes.h"
+#include "ifuncs.h"
 
 extern int am_server;
 extern int am_sender;
@@ -51,19 +52,24 @@ extern int need_messages_from_generator;
 extern int delete_mode, delete_before, delete_during, delete_after;
 extern int do_compression;
 extern int do_compression_level;
+extern int saw_stderr_opt;
+extern int msgs2stderr;
 extern char *shell_cmd;
 extern char *partial_dir;
 extern char *files_from;
 extern char *filesfrom_host;
 extern const char *checksum_choice;
 extern const char *compress_choice;
+extern char *daemon_auth_choices;
 extern filter_rule_list filter_list;
 extern int need_unsorted_flist;
 #ifdef ICONV_OPTION
 extern iconv_t ic_send, ic_recv;
 extern char *iconv_opt;
 #endif
-extern struct name_num_obj valid_checksums;
+extern struct name_num_obj valid_checksums, valid_auth_checksums;
+
+extern struct name_num_item *xfer_sum_nni;
 
 int remote_protocol = 0;
 int file_extra_cnt = 0; /* count of file-list extras that everyone gets */
@@ -76,6 +82,9 @@ int inplace_partial = 0;
 int do_negotiated_strings = 0;
 int xmit_id0_names = 0;
 
+struct name_num_item *xattr_sum_nni;
+int xattr_sum_len = 0;
+
 /* These index values are for the file-list's extra-attribute array. */
 int pathname_ndx, depth_ndx, atimes_ndx, crtimes_ndx, uid_ndx, gid_ndx, acls_ndx, xattrs_ndx, unsort_ndx;
 
@@ -88,19 +97,21 @@ int filesfrom_convert = 0;
 
 #define MAX_NSTR_STRLEN 256
 
-struct name_num_obj valid_compressions = {
-	"compress", NULL, NULL, 0, 0, {
+struct name_num_item valid_compressions_items[] = {
 #ifdef SUPPORT_ZSTD
-		{ CPRES_ZSTD, "zstd", NULL },
+	{ CPRES_ZSTD, 0, "zstd", NULL },
 #endif
 #ifdef SUPPORT_LZ4
-		{ CPRES_LZ4, "lz4", NULL },
+	{ CPRES_LZ4, 0, "lz4", NULL },
 #endif
-		{ CPRES_ZLIBX, "zlibx", NULL },
-		{ CPRES_ZLIB, "zlib", NULL },
-		{ CPRES_NONE, "none", NULL },
-		{ 0, NULL, NULL }
-	}
+	{ CPRES_ZLIBX, 0, "zlibx", NULL },
+	{ CPRES_ZLIB, 0, "zlib", NULL },
+	{ CPRES_NONE, 0, "none", NULL },
+	{ 0, 0, NULL, NULL }
+};
+
+struct name_num_obj valid_compressions = {
+	"compress", NULL, 0, 0, valid_compressions_items
 };
 
 #define CF_INC_RECURSE	 (1<<0)
@@ -122,11 +133,7 @@ static void check_sub_protocol(void)
 {
 	char *dot;
 	int their_protocol, their_sub;
-#if SUBPROTOCOL_VERSION != 0
-	int our_sub = protocol_version < PROTOCOL_VERSION ? 0 : SUBPROTOCOL_VERSION;
-#else
-	int our_sub = 0;
-#endif
+	int our_sub = get_subprotocol_version();
 
 	/* client_info starts with VER.SUB string if client is a pre-release. */
 	if (!(their_protocol = atoi(client_info))
@@ -153,7 +160,13 @@ static void check_sub_protocol(void)
 
 void set_allow_inc_recurse(void)
 {
-	client_info = shell_cmd ? shell_cmd : "";
+	if (!local_server)
+		client_info = shell_cmd ? shell_cmd : "";
+	else if (am_server) {
+		char buf[64];
+		maybe_add_e_option(buf, sizeof buf);
+		client_info = *buf ? strdup(buf+1) : ""; /* The +1 skips the leading "e". */
+	}
 
 	if (!recurse || use_qsort)
 		allow_inc_recurse = 0;
@@ -161,15 +174,14 @@ void set_allow_inc_recurse(void)
 	 && (delete_before || delete_after
 	  || delay_updates || prune_empty_dirs))
 		allow_inc_recurse = 0;
-	else if (am_server && !local_server
-	 && (strchr(client_info, 'i') == NULL))
+	else if (am_server && strchr(client_info, 'i') == NULL)
 		allow_inc_recurse = 0;
 }
 
 void parse_compress_choice(int final_call)
 {
-	if (valid_compressions.negotiated_name)
-		do_compression = valid_compressions.negotiated_num;
+	if (valid_compressions.negotiated_nni)
+		do_compression = valid_compressions.negotiated_nni->num;
 	else if (compress_choice) {
 		struct name_num_item *nni = get_nni_by_name(&valid_compressions, compress_choice, -1);
 		if (!nni) {
@@ -191,8 +203,8 @@ void parse_compress_choice(int final_call)
 		compress_choice = NULL;
 
 	/* Snag the compression name for both write_batch's option output & the following debug output. */
-	if (valid_compressions.negotiated_name)
-		compress_choice = valid_compressions.negotiated_name;
+	if (valid_compressions.negotiated_nni)
+		compress_choice = valid_compressions.negotiated_nni->name;
 	else if (compress_choice == NULL) {
 		struct name_num_item *nni = get_nni_by_num(&valid_compressions, do_compression);
 		compress_choice = nni ? nni->name : "UNKNOWN";
@@ -202,7 +214,7 @@ void parse_compress_choice(int final_call)
 	 && (do_compression != CPRES_NONE || do_compression_level != CLVL_NOT_SPECIFIED)) {
 		rprintf(FINFO, "%s%s compress: %s (level %d)\n",
 			am_server ? "Server" : "Client",
-			valid_compressions.negotiated_name ? " negotiated" : "",
+			valid_compressions.negotiated_nni ? " negotiated" : "",
 			compress_choice, do_compression_level);
 	}
 }
@@ -215,6 +227,8 @@ struct name_num_item *get_nni_by_name(struct name_num_obj *nno, const char *name
 		len = strlen(name);
 
 	for (nni = nno->list; nni->name; nni++) {
+		if (nni->num == CSUM_gone)
+			continue;
 		if (strncasecmp(name, nni->name, len) == 0 && nni->name[len] == '\0')
 			return nni;
 	}
@@ -249,10 +263,12 @@ static void init_nno_saw(struct name_num_obj *nno, int val)
 	if (!nno->saw) {
 		nno->saw = new_array0(uchar, nno->saw_len);
 
-		/* We'll take this opportunity to make sure that the main_name values are set right. */
+		/* We'll take this opportunity to set the main_nni values for duplicates. */
 		for (cnt = 1, nni = nno->list; nni->name; nni++, cnt++) {
+			if (nni->num == CSUM_gone)
+				continue;
 			if (nno->saw[nni->num])
-				nni->main_name = nno->list[nno->saw[nni->num]-1].name;
+				nni->main_nni = &nno->list[nno->saw[nni->num]-1];
 			else
 				nno->saw[nni->num] = cnt;
 		}
@@ -278,8 +294,8 @@ static int parse_nni_str(struct name_num_obj *nno, const char *from, char *tobuf
 				struct name_num_item *nni = get_nni_by_name(nno, tok, to - tok);
 				if (nni && !nno->saw[nni->num]) {
 					nno->saw[nni->num] = ++cnt;
-					if (nni->main_name) {
-						to = tok + strlcpy(tok, nni->main_name, tobuf_len - (tok - tobuf));
+					if (nni->main_nni) {
+						to = tok + strlcpy(tok, nni->main_nni->name, tobuf_len - (tok - tobuf));
 						if (to - tobuf >= tobuf_len) {
 							to = tok - 1;
 							break;
@@ -313,13 +329,44 @@ static int parse_nni_str(struct name_num_obj *nno, const char *from, char *tobuf
 	return to - tobuf;
 }
 
+static int parse_negotiate_str(struct name_num_obj *nno, char *tmpbuf)
+{
+	struct name_num_item *nni, *ret = NULL;
+	int best = nno->saw_len; /* We want best == 1 from the client list, so start with a big number. */
+	char *space, *tok = tmpbuf;
+	while (tok) {
+		while (*tok == ' ') tok++; /* Should be unneeded... */
+		if (!*tok)
+			break;
+		if ((space = strchr(tok, ' ')) != NULL)
+			*space = '\0';
+		nni = get_nni_by_name(nno, tok, -1);
+		if (space) {
+			*space = ' ';
+			tok = space + 1;
+		} else
+			tok = NULL;
+		if (!nni || !nno->saw[nni->num] || best <= nno->saw[nni->num])
+			continue;
+		ret = nni;
+		best = nno->saw[nni->num];
+		if (best == 1 || am_server) /* The server side stops at the first acceptable client choice */
+			break;
+	}
+	if (ret) {
+		free(nno->saw);
+		nno->saw = NULL;
+		nno->negotiated_nni = ret->main_nni ? ret->main_nni : ret;
+		return 1;
+	}
+	return 0;
+}
+
 /* This routine is always called with a tmpbuf of MAX_NSTR_STRLEN length, but the
  * buffer may be pre-populated with a "len" length string to use OR a len of -1
  * to tell us to read a string from the fd. */
 static void recv_negotiate_str(int f_in, struct name_num_obj *nno, char *tmpbuf, int len)
 {
-	struct name_num_item *ret = NULL;
-
 	if (len < 0)
 		len = read_vstring(f_in, tmpbuf, MAX_NSTR_STRLEN);
 
@@ -330,37 +377,8 @@ static void recv_negotiate_str(int f_in, struct name_num_obj *nno, char *tmpbuf,
 			rprintf(FINFO, "Server %s list (on client): %s\n", nno->type, tmpbuf);
 	}
 
-	if (len > 0) {
-		struct name_num_item *nni;
-		int best = nno->saw_len; /* We want best == 1 from the client list, so start with a big number. */
-		char *space, *tok = tmpbuf;
-		while (tok) {
-			while (*tok == ' ') tok++; /* Should be unneeded... */
-			if (!*tok)
-				break;
-			if ((space = strchr(tok, ' ')) != NULL)
-				*space = '\0';
-			nni = get_nni_by_name(nno, tok, -1);
-			if (space) {
-				*space = ' ';
-				tok = space + 1;
-			} else
-				tok = NULL;
-			if (!nni || !nno->saw[nni->num] || best <= nno->saw[nni->num])
-				continue;
-			ret = nni;
-			best = nno->saw[nni->num];
-			if (best == 1 || am_server) /* The server side stops at the first acceptable client choice */
-				break;
-		}
-		if (ret) {
-			free(nno->saw);
-			nno->saw = NULL;
-			nno->negotiated_name = ret->main_name ? ret->main_name : ret->name;
-			nno->negotiated_num = ret->num;
-			return;
-		}
-	}
+	if (len > 0 && parse_negotiate_str(nno, tmpbuf))
+		return;
 
 	if (!am_server || !do_negotiated_strings) {
 		char *cp = tmpbuf;
@@ -392,7 +410,7 @@ static const char *getenv_nstr(int ntype)
 	const char *env_str = getenv(ntype == NSTR_COMPRESS ? "RSYNC_COMPRESS_LIST" : "RSYNC_CHECKSUM_LIST");
 
 	/* When writing a batch file, we always negotiate an old-style choice. */
-	if (write_batch) 
+	if (write_batch)
 		env_str = ntype == NSTR_COMPRESS ? "zlib" : protocol_version >= 30 ? "md5" : "md4";
 
 	if (am_server && env_str) {
@@ -425,7 +443,7 @@ void validate_choice_vs_env(int ntype, int num1, int num2)
 		nno->saw[CSUM_MD4_ARCHAIC] = nno->saw[CSUM_MD4_BUSTED] = nno->saw[CSUM_MD4_OLD] = nno->saw[CSUM_MD4];
 
 	if (!nno->saw[num1] || (num2 >= 0 && !nno->saw[num2])) {
-		rprintf(FERROR, "Your --%s-choice value (%s) was refused by the server.\n", 
+		rprintf(FERROR, "Your --%s-choice value (%s) was refused by the server.\n",
 			ntype == NSTR_COMPRESS ? "compress" : "checksum",
 			ntype == NSTR_COMPRESS ? compress_choice : checksum_choice);
 		exit_cleanup(RERR_UNSUPPORTED);
@@ -456,8 +474,10 @@ int get_default_nno_list(struct name_num_obj *nno, char *to_buf, int to_buf_len,
 	init_nno_saw(nno, 0);
 
 	for (nni = nno->list, len = 0; nni->name; nni++) {
-		if (nni->main_name) {
-			if (!dup_markup)
+		if (nni->num == CSUM_gone)
+			continue;
+		if (nni->main_nni) {
+			if (!dup_markup || nni->main_nni->num == CSUM_gone)
 				continue;
 			delim = dup_markup;
 		}
@@ -515,6 +535,8 @@ static void negotiate_the_strings(int f_in, int f_out)
 {
 	/* We send all the negotiation strings before we start to read them to help avoid a slow startup. */
 
+	init_checksum_choices();
+
 	if (!checksum_choice)
 		send_negotiate_str(f_out, &valid_checksums, NSTR_CHECKSUM);
 
@@ -544,7 +566,7 @@ static void negotiate_the_strings(int f_in, int f_out)
 	/* If the other side is too old to negotiate, the above steps just made sure that
 	 * the env didn't disallow the old algorithm. Mark things as non-negotiated. */
 	if (!do_negotiated_strings)
-		valid_checksums.negotiated_name = valid_compressions.negotiated_name = NULL;
+		valid_checksums.negotiated_nni = valid_compressions.negotiated_nni = NULL;
 }
 
 void setup_protocol(int f_out,int f_in)
@@ -558,7 +580,7 @@ void setup_protocol(int f_out,int f_in)
 		atimes_ndx = (file_extra_cnt += EXTRA64_CNT);
 	if (preserve_crtimes)
 		crtimes_ndx = (file_extra_cnt += EXTRA64_CNT);
-	if (am_sender) /* This is most likely in the in64 union as well. */
+	if (am_sender) /* This is most likely in the file_extras64 union as well. */
 		pathname_ndx = (file_extra_cnt += PTR_EXTRA_CNT);
 	else
 		depth_ndx = ++file_extra_cnt;
@@ -596,7 +618,7 @@ void setup_protocol(int f_out,int f_in)
 	if (remote_protocol < MIN_PROTOCOL_VERSION
 	 || remote_protocol > MAX_PROTOCOL_VERSION) {
 		rprintf(FERROR,"protocol version mismatch -- is your shell clean?\n");
-		rprintf(FERROR,"(see the rsync man page for an explanation)\n");
+		rprintf(FERROR,"(see the rsync manpage for an explanation)\n");
 		exit_cleanup(RERR_PROTOCOL);
 	}
 	if (remote_protocol < OLD_PROTOCOL_VERSION) {
@@ -616,6 +638,9 @@ void setup_protocol(int f_out,int f_in)
 	if (read_batch)
 		check_batch_flags();
 
+	if (!saw_stderr_opt && protocol_version <= 28 && am_server)
+		msgs2stderr = 0; /* The client side may not have stderr setup for us. */
+
 #ifndef SUPPORT_PREALLOCATION
 	if (preallocate_files && !am_sender) {
 		rprintf(FERROR, "preallocation is not supported on this %s\n",
@@ -691,17 +716,17 @@ void setup_protocol(int f_out,int f_in)
 #ifdef ICONV_OPTION
 			compat_flags |= CF_SYMLINK_ICONV;
 #endif
-			if (local_server || strchr(client_info, 'f') != NULL)
+			if (strchr(client_info, 'f') != NULL)
 				compat_flags |= CF_SAFE_FLIST;
-			if (local_server || strchr(client_info, 'x') != NULL)
+			if (strchr(client_info, 'x') != NULL)
 				compat_flags |= CF_AVOID_XATTR_OPTIM;
-			if (local_server || strchr(client_info, 'C') != NULL)
+			if (strchr(client_info, 'C') != NULL)
 				compat_flags |= CF_CHKSUM_SEED_FIX;
-			if (local_server || strchr(client_info, 'I') != NULL)
+			if (strchr(client_info, 'I') != NULL)
 				compat_flags |= CF_INPLACE_PARTIAL_DIR;
-			if (local_server || strchr(client_info, 'u') != NULL)
+			if (strchr(client_info, 'u') != NULL)
 				compat_flags |= CF_ID0_NAMES;
-			if (local_server || strchr(client_info, 'v') != NULL) {
+			if (strchr(client_info, 'v') != NULL) {
 				do_negotiated_strings = 1;
 				compat_flags |= CF_VARINT_FLIST_FLAGS;
 			}
@@ -737,7 +762,7 @@ void setup_protocol(int f_out,int f_in)
 #endif
 #ifdef ICONV_OPTION
 		sender_symlink_iconv = iconv_opt && (am_server
-		    ? local_server || strchr(client_info, 's') != NULL
+		    ? strchr(client_info, 's') != NULL
 		    : !!(compat_flags & CF_SYMLINK_ICONV));
 #endif
 		if (inc_recurse && !allow_inc_recurse) {
@@ -790,11 +815,77 @@ void setup_protocol(int f_out,int f_in)
 		checksum_seed = read_int(f_in);
 	}
 
-	parse_checksum_choice(1); /* Sets checksum_type & xfersum_type */
+	parse_checksum_choice(1); /* Sets file_sum_nni & xfer_sum_nni */
 	parse_compress_choice(1); /* Sets do_compression */
 
+	/* TODO in the future allow this algorithm to be chosen somehow, but it can't get too
+	 * long or the size starts to cause a problem in the xattr abbrev/non-abbrev code. */
+	xattr_sum_nni = parse_csum_name(NULL, 0);
+	xattr_sum_len = csum_len_for_type(xattr_sum_nni->num, 0);
+
 	if (write_batch && !am_server)
 		write_batch_shell_file();
 
 	init_flist();
 }
+
+void output_daemon_greeting(int f_out, int am_client)
+{
+	char tmpbuf[MAX_NSTR_STRLEN];
+	int our_sub = get_subprotocol_version();
+
+	init_checksum_choices();
+
+	get_default_nno_list(&valid_auth_checksums, tmpbuf, MAX_NSTR_STRLEN, '\0');
+
+	io_printf(f_out, "@RSYNCD: %d.%d %s\n", protocol_version, our_sub, tmpbuf);
+
+	if (am_client && DEBUG_GTE(NSTR, 2))
+		rprintf(FINFO, "Client %s list (on client): %s\n", valid_auth_checksums.type, tmpbuf);
+}
+
+void negotiate_daemon_auth(int f_out, int am_client)
+{
+	char tmpbuf[MAX_NSTR_STRLEN];
+	int save_am_server = am_server;
+	int md4_is_old = 0;
+
+	if (!am_client)
+		am_server = 1;
+
+	if (daemon_auth_choices)
+		strlcpy(tmpbuf, daemon_auth_choices, MAX_NSTR_STRLEN);
+	else {
+		strlcpy(tmpbuf, protocol_version >= 30 ? "md5" : "md4", MAX_NSTR_STRLEN);
+		md4_is_old = 1;
+	}
+
+	if (am_client) {
+		recv_negotiate_str(-1, &valid_auth_checksums, tmpbuf, strlen(tmpbuf));
+		if (DEBUG_GTE(NSTR, 1)) {
+			rprintf(FINFO, "Client negotiated %s: %s\n", valid_auth_checksums.type,
+				valid_auth_checksums.negotiated_nni->name);
+		}
+	} else {
+		if (!parse_negotiate_str(&valid_auth_checksums, tmpbuf)) {
+			get_default_nno_list(&valid_auth_checksums, tmpbuf, MAX_NSTR_STRLEN, '\0');
+			io_printf(f_out, "@ERROR: your client does not support one of our daemon-auth checksums: %s\n",
+				  tmpbuf);
+			exit_cleanup(RERR_UNSUPPORTED);
+		}
+	}
+	am_server = save_am_server;
+	if (md4_is_old && valid_auth_checksums.negotiated_nni->num == CSUM_MD4) {
+		valid_auth_checksums.negotiated_nni->num = CSUM_MD4_OLD;
+		valid_auth_checksums.negotiated_nni->flags = 0;
+	}
+}
+
+int get_subprotocol_version()
+{
+#if SUBPROTOCOL_VERSION != 0
+	return protocol_version < PROTOCOL_VERSION ? 0 : SUBPROTOCOL_VERSION;
+#else
+	return 0;
+#endif
+}

configure.ac

@@ -2,12 +2,27 @@ dnl Process this file with autoconf to produce a configure script.
 
 AC_INIT([rsync],[ ],[https://rsync.samba.org/bug-tracking.html])
 
+AC_C_BIGENDIAN
+AC_HEADER_DIRENT
+AC_HEADER_SYS_WAIT
+AC_CHECK_HEADERS(sys/fcntl.h sys/select.h fcntl.h sys/time.h sys/unistd.h \
+    unistd.h utime.h compat.h sys/param.h ctype.h sys/wait.h sys/stat.h \
+    sys/ioctl.h sys/filio.h string.h stdlib.h sys/socket.h sys/mode.h grp.h \
+    sys/un.h sys/attr.h arpa/inet.h arpa/nameser.h locale.h sys/types.h \
+    netdb.h malloc.h float.h limits.h iconv.h libcharset.h langinfo.h mcheck.h \
+    sys/acl.h acl/libacl.h attr/xattr.h sys/xattr.h sys/extattr.h dl.h \
+    popt.h popt/popt.h linux/falloc.h netinet/in_systm.h netgroup.h \
+    zlib.h xxhash.h openssl/md4.h openssl/md5.h zstd.h lz4.h sys/file.h \
+    bsd/string.h)
+AC_CHECK_HEADERS([netinet/ip.h], [], [], [[#include <netinet/in.h>]])
+AC_HEADER_MAJOR_FIXED
+
 AC_CONFIG_MACRO_DIR([m4])
 AC_CONFIG_SRCDIR([byteorder.h])
-AC_CONFIG_HEADER(config.h)
+AC_CONFIG_HEADERS([config.h])
 AC_PREREQ([2.69])
 
-PACKAGE_VERSION=`sed 's/.*"\(.*\)".*/\1/' <$srcdir/version.h`
+PACKAGE_VERSION=`sed -n 's/.*RSYNC_VERSION.*"\(.*\)".*/\1/p' <$srcdir/version.h`
 
 AC_MSG_NOTICE([Configuring rsync $PACKAGE_VERSION])
 
@@ -26,7 +41,7 @@ AC_CONFIG_LIBOBJ_DIR([lib])
 
 AC_MSG_CHECKING([whether to include debugging symbols])
 AC_ARG_ENABLE(debug,
-	AS_HELP_STRING([--disable-debug],[disable debugging symbols and features]))
+	AS_HELP_STRING([--disable-debug],[disable to omit debugging symbols and features]))
 
 if test x"$enable_debug" = x"no"; then
     AC_MSG_RESULT(no)
@@ -45,7 +60,6 @@ AC_PROG_AWK
 AC_PROG_EGREP
 AC_PROG_INSTALL
 AC_PROG_MKDIR_P
-AC_PROG_CC_STDC
 AC_SUBST(SHELL)
 AC_PATH_PROG([PERL], [perl])
 AC_PATH_PROG([PYTHON3], [python3])
@@ -57,13 +71,18 @@ if test x"$ac_cv_prog_cc_stdc" = x"no"; then
 	AC_MSG_WARN([rsync requires an ANSI C compiler and you do not seem to have one])
 fi
 
+no_lib=''
+err_msg=''
+nl='
+'
+
 AC_ARG_ENABLE(profile,
-	AS_HELP_STRING([--enable-profile],[turn on CPU profiling]))
+	AS_HELP_STRING([--enable-profile],[enable to turn on CPU profiling]))
 if test x"$enable_profile" = x"yes"; then
 	CFLAGS="$CFLAGS -pg"
 fi
 
-AC_MSG_CHECKING([if md2man can create man pages])
+AC_MSG_CHECKING([if md2man can create manpages])
 if test x"$ac_cv_path_PYTHON3" = x; then
     AC_MSG_RESULT(no - python3 not found)
     md2man_works=no
@@ -81,26 +100,27 @@ fi
 
 AC_MSG_CHECKING([if we require man-page building])
 AC_ARG_ENABLE([md2man],
-	AS_HELP_STRING([--disable-md2man],[disable md2man for man page creation]))
+	AS_HELP_STRING([--disable-md2man],[disable to omit manpage creation]))
 if test x"$enable_md2man" != x"no"; then
     if test -f "$srcdir/rsync.1"; then
 	AC_MSG_RESULT(optional)
     else
 	AC_MSG_RESULT(required)
 	if test x"$md2man_works" = x"no"; then
-	    AC_MSG_ERROR(You need python3 and the cmarkgfm OR commonmark python3 lib in order to build man pages.
-You can specify --disable-md2man if you want to skip building them.)
+	    err_msg="$err_msg$nl- You need python3 and either the cmarkgfm OR commonmark python3 lib in order"
+	    err_msg="$err_msg$nl  to build manpages based on the git source (manpages are included in the"
+	    err_msg="$err_msg$nl  official release tar files)."
+	    no_lib="$no_lib md2man"
 	fi
     fi
     MAKE_MAN=man
 else
     AC_MSG_RESULT(no)
-    MAKE_MAN=''
 fi
 
 # Specifically, this turns on panic_action handling.
 AC_ARG_ENABLE(maintainer-mode,
-	AS_HELP_STRING([--enable-maintainer-mode],[turn on extra debug features]))
+	AS_HELP_STRING([--enable-maintainer-mode],[enable to turn on extra debug features]))
 if test x"$enable_maintainer_mode" = x"yes"; then
 	CFLAGS="$CFLAGS -DMAINTAINER_MODE"
 fi
@@ -114,16 +134,37 @@ if test x"$GCC" = x"yes"; then
 	CFLAGS="$CFLAGS -Wall -W"
 fi
 
+AC_ARG_WITH(openssl-conf,
+	AS_HELP_STRING([--with-openssl-conf=PATH],[set default OPENSSL_CONF path for rsync]))
+case "$with_openssl_conf" in
+    *[^-/a-zA-Z0-9.,=@+_]*) AC_MSG_ERROR([Invalid path given to --with-openssl-conf]) ;;
+    /*) CFLAGS="$CFLAGS -DSET_OPENSSL_CONF=$with_openssl_conf" ;;
+    no|'') ;;
+    yes) AC_MSG_ERROR([No path given to --with-openssl-conf]) ;;
+    *) AC_MSG_ERROR([Non absolute path given to --with-openssl-conf]) ;;
+esac
+
+AC_ARG_WITH(rrsync,
+        AS_HELP_STRING([--with-rrsync],[also install the rrsync script and its manpage]))
+if test x"$with_rrsync" != x"yes"; then
+    with_rrsync=no
+else
+    MAKE_RRSYNC='rrsync'
+    MAKE_RRSYNC_1='rrsync.1'
+    GEN_RRSYNC='rrsync.1 rrsync.1.html'
+fi
+AC_SUBST(with_rrsync)
+
 AC_ARG_WITH(included-popt,
         AS_HELP_STRING([--with-included-popt],[use bundled popt library, not from system]))
 
 AC_ARG_WITH(included-zlib,
         AS_HELP_STRING([--with-included-zlib],[use bundled zlib library, not from system]))
 
-AC_ARG_WITH(protected-args,
-        AS_HELP_STRING([--with-protected-args],[make --protected-args option the default]))
-if test x"$with_protected_args" = x"yes"; then
-	AC_DEFINE_UNQUOTED(RSYNC_USE_PROTECTED_ARGS, 1, [Define to 1 if --protected-args should be the default])
+AC_ARG_WITH(secluded-args,
+        AS_HELP_STRING([--with-secluded-args],[make --secluded-args option the default]))
+if test x"$with_secluded_args" = x"yes"; then
+	AC_DEFINE_UNQUOTED(RSYNC_USE_SECLUDED_ARGS, 1, [Define to 1 if --secluded-args should be the default])
 fi
 
 AC_ARG_WITH(rsync-path,
@@ -173,6 +214,11 @@ AC_DEFINE_UNQUOTED(RSYNC_RSH, "$RSYNC_RSH", [default -e command])
 AC_PATH_PROG(SHELL_PATH, sh, /bin/sh, [/usr/xpg4/bin$PATH_SEPARATOR$PATH])
 AC_PATH_PROG(FAKEROOT_PATH, fakeroot, /usr/bin/fakeroot, [/usr/xpg4/bin$PATH_SEPARATOR$PATH])
 
+AC_ARG_WITH(nobody-user,
+    AS_HELP_STRING([--with-nobody-user=USER],[set the default unprivileged user (default nobody)]),
+    [ NOBODY_USER="$with_nobody_user" ],
+    [ NOBODY_USER="nobody" ])
+
 AC_ARG_WITH(nobody-group,
     AS_HELP_STRING([--with-nobody-group=GROUP],[set the default unprivileged group (default nobody or nogroup)]),
     [ NOBODY_GROUP="$with_nobody_group" ])
@@ -189,24 +235,22 @@ if test x"$with_nobody_group" = x; then
     AC_MSG_RESULT($NOBODY_GROUP)
 fi
 
-AC_DEFINE_UNQUOTED(NOBODY_USER, "nobody", [unprivileged user--e.g. nobody])
+AC_DEFINE_UNQUOTED(NOBODY_USER, "$NOBODY_USER", [unprivileged user--e.g. nobody])
 AC_DEFINE_UNQUOTED(NOBODY_GROUP, "$NOBODY_GROUP", [unprivileged group for unprivileged user])
 
-# SIMD optimizations
-SIMD=
+# rolling-checksum SIMD optimizations
+ROLL_SIMD=
 
-AC_MSG_CHECKING([whether to enable SIMD optimizations])
-AC_ARG_ENABLE(simd,
-    AS_HELP_STRING([--disable-simd],[disable SIMD optimizations (requires c++)]))
+AC_MSG_CHECKING([whether to enable rolling-checksum SIMD optimizations])
+AC_ARG_ENABLE(roll-simd,
+    AS_HELP_STRING([--enable-roll-simd],[enable/disable to control rolling-checksum SIMD optimizations (requires c++)]))
 
 # Clag is crashing with -g -O2, so we'll get rid of -g for now.
 CXXFLAGS=`echo "$CXXFLAGS" | sed 's/-g //'`
-
-if test x"$enable_simd" != x"no"; then
-    # For x86-64 SIMD, g++ >=5 or clang++ >=7 is required
-    if test x"$build_cpu" = x"x86_64"; then
-	AC_LANG(C++)
-	AC_RUN_IFELSE([AC_LANG_PROGRAM([[#include <stdio.h>
+m4_define(SIMD_X86_64_TEST, [[#include <stdio.h>
+#if HAVE_STDLIB_H
+#include <stdlib.h>
+#endif
 #include <immintrin.h>
 __attribute__ ((target("default"))) int test_ssse3(int x) { return x; }
 __attribute__ ((target("default"))) int test_sse2(int x) { return x; }
@@ -226,27 +270,45 @@ __attribute__ ((target("ssse3"))) void more_testing(char* buf, int len)
 	in8_2 = _mm_lddqu_si128((__m128i_u*)&buf[i + 16]);
     }
 }
-]], [[if (test_ssse3(42) != 42 || test_sse2(42) != 42 || test_avx2(42) != 42) exit(1);]])],[CXX_OK=yes],[CXX_OK=no])
+]])
+
+if test x"$enable_roll_simd" = x""; then
+    case "$host_os" in
+	*linux*) ;;
+	*) enable_roll_simd=no ;;
+    esac
+fi
+
+if test x"$enable_roll_simd" != x"no"; then
+    # For x86-64 SIMD, g++ >=5 or clang++ >=7 is required
+    if test x"$host_cpu" = x"x86_64" || test x"$host_cpu" = x"amd64"; then
+	AC_LANG(C++)
+	if test x"$host" = x"$build"; then
+	    AC_RUN_IFELSE([AC_LANG_PROGRAM([SIMD_X86_64_TEST],[[if (test_ssse3(42) != 42 || test_sse2(42) != 42 || test_avx2(42) != 42) exit(1);]])],
+		[CXX_OK=yes],[CXX_OK=no])
+	else
+	    AC_COMPILE_IFELSE([AC_LANG_PROGRAM([SIMD_X86_64_TEST])],[CXX_OK=yes],[CXX_OK=no])
+	fi
 	AC_LANG(C)
 	if test x"$CXX_OK" = x"yes"; then
 	    # AC_MSG_RESULT() is called below.
-	    SIMD="x86_64"
-	elif test x"$enable_simd" = x"yes"; then
+	    ROLL_SIMD="$host_cpu"
+	elif test x"$enable_roll_simd" = x"yes"; then
 	    AC_MSG_RESULT(error)
-	    AC_MSG_ERROR(The SIMD compilation test failed.
-Omit --enable-simd to continue without it.)
+	    AC_MSG_ERROR(The rolling-checksum SIMD compilation test failed.
+Omit --enable-roll-simd to continue without it.)
 	fi
-    elif test x"$enable_simd" = x"yes"; then
+    elif test x"$enable_roll_simd" = x"yes"; then
         AC_MSG_RESULT(unavailable)
-        AC_MSG_ERROR(The SIMD optimizations are currently x86_64 only.
-Omit --enable-simd to continue without it.)
+        AC_MSG_ERROR(The rolling-checksum SIMD optimizations are currently x86_64|amd64 only.
+Omit --enable-roll-simd to continue without it.)
     fi
 fi
 
-if test x"$SIMD" != x""; then
-    AC_MSG_RESULT([yes ($SIMD)])
-    AC_DEFINE(HAVE_SIMD, 1, [Define to 1 to enable SIMD optimizations])
-    SIMD='$(SIMD_'"$SIMD)"
+if test x"$ROLL_SIMD" != x""; then
+    AC_MSG_RESULT([yes ($ROLL_SIMD)])
+    AC_DEFINE(USE_ROLL_SIMD, 1, [Define to 1 to enable rolling-checksum SIMD optimizations])
+    ROLL_SIMD='$(ROLL_SIMD_'"$ROLL_SIMD)"
     # We only use c++ for its target attribute dispatching, disable unneeded bulky features
     CXXFLAGS="$CXXFLAGS -fno-exceptions -fno-rtti"
     # Apple often has "g++" as a symlink for clang. Try to find out the truth.
@@ -258,7 +320,7 @@ else
     AC_MSG_RESULT(no)
 fi
 
-AC_SUBST(SIMD)
+AC_SUBST(ROLL_SIMD)
 
 AC_MSG_CHECKING([if assembler accepts noexecstack])
 OLD_CFLAGS="$CFLAGS"
@@ -269,43 +331,19 @@ AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[ ]], [[return 0;]])],
 CFLAGS="$OLD_CFLAGS"
 AC_SUBST(NOEXECSTACK)
 
-ASM=
-
-AC_MSG_CHECKING([whether to enable ASM optimizations])
-AC_ARG_ENABLE(asm,
-    AS_HELP_STRING([--disable-asm],[disable ASM optimizations]))
-
-if test x"$enable_asm" != x"no"; then
-    if test x"$build_cpu" = x"x86_64"; then
-	ASM="$build_cpu"
-    elif test x"$enable_asm" = x"yes"; then
-        AC_MSG_RESULT(unavailable)
-        AC_MSG_ERROR(The ASM optimizations are currently x86_64 only.
-Omit --enable-asm to continue without it.)
-    fi
-fi
-
-if test x"$ASM" != x""; then
-    AC_MSG_RESULT([yes ($ASM)])
-    AC_DEFINE(HAVE_ASM, 1, [Define to 1 to enable ASM optimizations])
-    ASM='$(ASM_'"$ASM)"
-else
-    AC_MSG_RESULT(no)
-fi
-
-AC_SUBST(ASM)
-
 # arrgh. libc in some old debian version screwed up the largefile
 # stuff, getting byte range locking wrong
 AC_CACHE_CHECK([for broken largefile support],rsync_cv_HAVE_BROKEN_LARGEFILE,[
 AC_RUN_IFELSE([AC_LANG_SOURCE([[
 #define _FILE_OFFSET_BITS 64
-#include <stdio.h>
-#include <fcntl.h>
-#include <sys/types.h>
+$ac_includes_default
+#ifdef HAVE_FCNTL_H
+# include <fcntl.h>
+#elif defined HAVE_SYS_FCNTL_H
+# include <sys/fcntl.h>
+#endif
+#ifdef HAVE_SYS_WAIT_H
 #include <sys/wait.h>
-#if HAVE_UNISTD_H
-#include <unistd.h>
 #endif
 
 int main(void)
@@ -338,132 +376,39 @@ if test x"$rsync_cv_HAVE_BROKEN_LARGEFILE" != x"yes"; then
    AC_SYS_LARGEFILE
 fi
 
-ipv6type=unknown
-ipv6lib=none
-ipv6trylibc=yes
-
+AC_MSG_CHECKING([whether to enable ipv6])
 AC_ARG_ENABLE(ipv6,
-	AS_HELP_STRING([--disable-ipv6],[turn off IPv6 support]))
-if test x"$enable_ipv6" != x"no"; then
-	AC_MSG_CHECKING([ipv6 stack type])
-	for i in inria kame linux-glibc linux-inet6 solaris toshiba v6d zeta cygwin TANDEM; do
-		case $i in
-		inria)
-			# http://www.kame.net/
-			AC_EGREP_CPP(yes, [
-#include <netinet/in.h>
-#ifdef IPV6_INRIA_VERSION
-yes
-#endif],
-				[ipv6type=$i;
-				AC_DEFINE(INET6, 1, [true if you have IPv6])
-				])
-			;;
-		kame)
-			# http://www.kame.net/
-			AC_EGREP_CPP(yes, [
-#include <netinet/in.h>
-#ifdef __KAME__
-yes
-#endif],
-				[ipv6type=$i;
-				AC_DEFINE(INET6, 1, [true if you have IPv6])])
-			;;
-		linux-glibc)
-			# http://www.v6.linux.or.jp/
-			AC_EGREP_CPP(yes, [
-#include <features.h>
-#if defined(__GLIBC__) && __GLIBC__ >= 2 && __GLIBC_MINOR__ >= 1
-yes
-#endif],
-				[ipv6type=$i;
-AC_DEFINE(INET6, 1, [true if you have IPv6])])
-			;;
-		linux-inet6)
-			# http://www.v6.linux.or.jp/
-			if test -d /usr/inet6 -o -f /usr/inet6/lib/libinet6.a; then
-				ipv6type=$i
-				ipv6lib=inet6
-				ipv6libdir=/usr/inet6/lib
-				ipv6trylibc=yes;
-				AC_DEFINE(INET6, 1, [true if you have IPv6])
-				CFLAGS="-I/usr/inet6/include $CFLAGS"
-			fi
-			;;
-		solaris)
-			# http://www.sun.com
-			AC_EGREP_CPP(yes, [
-#include <netinet/ip6.h>
-#ifdef __sun
-yes
-#endif],
-				[ipv6type=$i;
-				AC_DEFINE(INET6, 1, [true if you have IPv6])])
-			;;
-		toshiba)
-			AC_EGREP_CPP(yes, [
-#include <sys/param.h>
-#ifdef _TOSHIBA_INET6
-yes
-#endif],
-				[ipv6type=$i;
-				ipv6lib=inet6;
-				ipv6libdir=/usr/local/v6/lib;
-				AC_DEFINE(INET6, 1, [true if you have IPv6])])
-			;;
-		v6d)
-			AC_EGREP_CPP(yes, [
-#include </usr/local/v6/include/sys/v6config.h>
-#ifdef __V6D__
-yes
-#endif],
-				[ipv6type=$i;
-				ipv6lib=v6;
-				ipv6libdir=/usr/local/v6/lib;
-				AC_DEFINE(INET6, 1, [true if you have IPv6])])
-			;;
-		zeta)
-			AC_EGREP_CPP(yes, [
-#include <sys/param.h>
-#ifdef _ZETA_MINAMI_INET6
-yes
-#endif],
-				[ipv6type=$i;
-				ipv6lib=inet6;
-				ipv6libdir=/usr/local/v6/lib;
-				AC_DEFINE(INET6, 1, [true if you have IPv6])])
-			;;
-		cygwin)
-			AC_EGREP_CPP(yes, [
-#include <netinet/in.h>
-#ifdef _CYGWIN_IN6_H
-yes
-#endif],
-				[ipv6type=$i;
-				AC_DEFINE(INET6, 1, [true if you have IPv6])])
-			;;
-		TANDEM)
-			AC_EGREP_CPP(yes, [
-#include <netinet/ip6.h>
-#ifdef __TANDEM
-yes
-#endif],
-				[ipv6type=$i;
-				AC_DEFINE(INET6, 1, [true if you have IPv6])])
-			;;
-		esac
-		if test "$ipv6type" != "unknown"; then
-			break
-		fi
-	done
-	AC_MSG_RESULT($ipv6type)
+AS_HELP_STRING([--disable-ipv6],[disable to omit ipv6 support]),
+[ case "$enableval" in
+  no)
+	AC_MSG_RESULT(no)
+	;;
+  *)	AC_MSG_RESULT(yes)
+	AC_DEFINE(INET6, 1, [true if you have IPv6])
+	;;
+  esac ],
 
-	AC_SEARCH_LIBS(getaddrinfo, inet6)
-fi
+  AC_RUN_IFELSE([AC_LANG_SOURCE([[ /* AF_INET6 availability check */
+#include <stdlib.h>
+#include <sys/types.h>
+#include <sys/socket.h>
+main()
+{
+   if (socket(AF_INET6, SOCK_STREAM, 0) < 0)
+     exit(1);
+   else
+     exit(0);
+}
+]])],
+  [AC_MSG_RESULT(yes)
+  AC_DEFINE(INET6, 1, true if you have IPv6)],
+  [AC_MSG_RESULT(no)],
+  [AC_MSG_RESULT(no)]
+))
 
 dnl Do you want to disable use of locale functions
 AC_ARG_ENABLE([locale],
-	AS_HELP_STRING([--disable-locale],[disable locale features]))
+	AS_HELP_STRING([--disable-locale],[disable to omit locale features]))
 AH_TEMPLATE([CONFIG_LOCALE],
 [Undefine if you do not want locale features.  By default this is defined.])
 if test x"$enable_locale" != x"no"; then
@@ -479,44 +424,88 @@ case $host_os in
 	       * ) AC_MSG_RESULT(no);;
 esac
 
-AC_C_BIGENDIAN
-AC_HEADER_DIRENT
-AC_HEADER_TIME
-AC_HEADER_SYS_WAIT
-AC_CHECK_HEADERS(sys/fcntl.h sys/select.h fcntl.h sys/time.h sys/unistd.h \
-    unistd.h utime.h grp.h compat.h sys/param.h ctype.h sys/wait.h \
-    sys/ioctl.h sys/filio.h string.h stdlib.h sys/socket.h sys/mode.h \
-    sys/un.h sys/attr.h mcheck.h arpa/inet.h arpa/nameser.h locale.h \
-    netdb.h malloc.h float.h limits.h iconv.h libcharset.h langinfo.h \
-    sys/acl.h acl/libacl.h attr/xattr.h sys/xattr.h sys/extattr.h dl.h \
-    popt.h popt/popt.h linux/falloc.h netinet/in_systm.h netinet/ip.h \
-    zlib.h xxhash.h openssl/md4.h openssl/md5.h zstd.h lz4.h sys/file.h)
-AC_HEADER_MAJOR_FIXED
-
 AC_MSG_CHECKING([whether to enable use of openssl crypto library])
 AC_ARG_ENABLE([openssl],
-	AS_HELP_STRING([--disable-openssl],[disable openssl crypto library]))
+	AS_HELP_STRING([--disable-openssl],[disable to omit openssl crypto library]))
 AH_TEMPLATE([USE_OPENSSL],
 [Undefine if you do not want to use openssl crypto library.  By default this is defined.])
 if test x"$enable_openssl" != x"no"; then
     if test x"$ac_cv_header_openssl_md4_h" = x"yes" && test x"$ac_cv_header_openssl_md5_h" = x"yes"; then
       AC_MSG_RESULT(yes)
-      AC_SEARCH_LIBS(MD5_Init, crypto,
-          [AC_DEFINE(USE_OPENSSL)],
-          [AC_MSG_ERROR(Failed to find MD5_Init function in openssl crypto lib.
-Use --disable-openssl to continue without openssl crypto lib support.)])
+      AC_SEARCH_LIBS(EVP_MD_CTX_copy, crypto,
+          [AC_DEFINE(USE_OPENSSL)
+	   enable_openssl=yes],
+          [err_msg="$err_msg$nl- Failed to find EVP_MD_CTX_copy function in openssl crypto lib.";
+	   no_lib="$no_lib openssl"])
     else
         AC_MSG_RESULT(no)
-        AC_MSG_ERROR(Failed to find openssl/md4.h and openssl/md5.h for openssl crypto lib support.
-Use --disable-openssl to continue without it.)
+	err_msg="$err_msg$nl- Failed to find openssl/md4.h and openssl/md5.h for openssl crypto lib support."
+	no_lib="$no_lib openssl"
+    fi
+    if test x"$enable_md5_asm" != x"yes"; then
+	enable_md5_asm=no
     fi
 else
     AC_MSG_RESULT(no)
 fi
 
+MD5_ASM=
+
+AC_MSG_CHECKING([whether to enable MD5 ASM optimizations])
+AC_ARG_ENABLE(md5-asm,
+    AS_HELP_STRING([--enable-md5-asm],[enable/disable to control MD5 ASM optimizations]))
+
+if test x"$enable_md5_asm" = x""; then
+    case "$host_os" in
+	*linux*) ;;
+	*) enable_md5_asm=no ;;
+    esac
+fi
+
+if test x"$enable_md5_asm" != x"no"; then
+    if test x"$host_cpu" = x"x86_64" || test x"$host_cpu" = x"amd64"; then
+	MD5_ASM="$host_cpu"
+    elif test x"$enable_md5_asm" = x"yes"; then
+        AC_MSG_RESULT(unavailable)
+        AC_MSG_ERROR(The ASM optimizations are currently x86_64|amd64 only.
+Omit --enable-md5-asm to continue without it.)
+    fi
+fi
+
+if test x"$MD5_ASM" != x""; then
+    AC_MSG_RESULT([yes ($MD5_ASM)])
+    AC_DEFINE(USE_MD5_ASM, 1, [Define to 1 to enable MD5 ASM optimizations])
+    MD5_ASM='$(MD5_ASM_'"$MD5_ASM)"
+else
+    AC_MSG_RESULT(no)
+fi
+
+AC_SUBST(MD5_ASM)
+
+ROLL_ASM=
+
+AC_MSG_CHECKING([whether to enable rolling-checksum ASM optimizations])
+AC_ARG_ENABLE(roll-asm,
+    AS_HELP_STRING([--enable-roll-asm],[enable/disable to control rolling-checksum ASM optimizations (requires --enable-roll-simd)]))
+
+if test x"$ROLL_SIMD" = x""; then
+    enable_roll_asm=no
+fi
+
+if test x"$enable_roll_asm" = x"yes"; then
+    ROLL_ASM="$host_cpu"
+    AC_MSG_RESULT([yes ($ROLL_ASM)])
+    AC_DEFINE(USE_ROLL_ASM, 1, [Define to 1 to enable rolling-checksum ASM optimizations (requires --enable-roll-simd)])
+    ROLL_ASM='$(ROLL_ASM_'"$ROLL_ASM)"
+else
+    AC_MSG_RESULT(no)
+fi
+
+AC_SUBST(ROLL_ASM)
+
 AC_MSG_CHECKING([whether to enable xxhash checksum support])
 AC_ARG_ENABLE([xxhash],
-	AS_HELP_STRING([--disable-xxhash],[disable xxhash checksums]))
+	AS_HELP_STRING([--disable-xxhash],[disable to omit xxhash checksums]))
 AH_TEMPLATE([SUPPORT_XXHASH],
 [Undefine if you do not want xxhash checksums.  By default this is defined.])
 if test x"$enable_xxhash" != x"no"; then
@@ -524,12 +513,12 @@ if test x"$enable_xxhash" != x"no"; then
 	AC_MSG_RESULT(yes)
 	AC_SEARCH_LIBS(XXH64_createState, xxhash,
 	    [AC_DEFINE(SUPPORT_XXHASH)],
-	    [AC_MSG_ERROR(Failed to find XXH64_createState function in xxhash lib.
-Use --disable-xxhash to continue without xxhash checksums.)])
+	    [err_msg="$err_msg$nl- Failed to find XXH64_createState function in xxhash lib.";
+	     no_lib="$no_lib xxhash"])
     else
 	AC_MSG_RESULT(no)
-	AC_MSG_ERROR(Failed to find xxhash.h for xxhash checksum support.
-Use --disable-xxhash to continue without it.)
+	err_msg="$err_msg$nl- Failed to find xxhash.h for xxhash checksum support.";
+	no_lib="$no_lib xxhash"
     fi
 else
     AC_MSG_RESULT(no)
@@ -537,7 +526,7 @@ fi
 
 AC_MSG_CHECKING([whether to enable zstd compression])
 AC_ARG_ENABLE([zstd],
-        AC_HELP_STRING([--disable-zstd], [disable zstd compression]))
+        AS_HELP_STRING([--disable-zstd], [disable to omit zstd compression]))
 AH_TEMPLATE([SUPPORT_ZSTD],
 [Undefine if you do not want zstd compression.  By default this is defined.])
 if test x"$enable_zstd" != x"no"; then
@@ -545,12 +534,12 @@ if test x"$enable_zstd" != x"no"; then
 	AC_MSG_RESULT(yes)
 	AC_SEARCH_LIBS(ZSTD_minCLevel, zstd,
 	    [AC_DEFINE(SUPPORT_ZSTD)],
-	    [AC_MSG_ERROR(Failed to find ZSTD_minCLevel function in zstd lib.
-Use --disable-zstd to continue without zstd compression.)])
+	    [err_msg="$err_msg$nl- Failed to find ZSTD_minCLevel function in zstd lib.";
+	     no_lib="$no_lib zstd"])
     else
 	AC_MSG_RESULT(no)
-	AC_MSG_ERROR(Failed to find zstd.h for zstd compression support.
-Use --disable-zstd to continue without it.)
+	err_msg="$err_msg$nl- Failed to find zstd.h for zstd compression support.";
+	no_lib="$no_lib zstd"
     fi
 else
     AC_MSG_RESULT(no)
@@ -558,7 +547,7 @@ fi
 
 AC_MSG_CHECKING([whether to enable LZ4 compression])
 AC_ARG_ENABLE([lz4],
-        AC_HELP_STRING([--disable-lz4], [disable LZ4 compression]))
+        AS_HELP_STRING([--disable-lz4], [disable to omit LZ4 compression]))
 AH_TEMPLATE([SUPPORT_LZ4],
 [Undefine if you do not want LZ4 compression.  By default this is defined.])
 if test x"$enable_lz4" != x"no"; then
@@ -566,20 +555,39 @@ if test x"$enable_lz4" != x"no"; then
 	AC_MSG_RESULT(yes)
 	AC_SEARCH_LIBS(LZ4_compress_default, lz4,
 	    [AC_DEFINE(SUPPORT_LZ4)],
-	    [AC_MSG_ERROR(Failed to find LZ4_compress_default function in lz4 lib.
-Use --disable-lz4 to continue without lz4 compression.)])
+	    [err_msg="$err_msg$nl- Failed to find LZ4_compress_default function in lz4 lib.";
+	     no_lib="$no_lib lz4"])
     else
 	AC_MSG_RESULT(no)
-	AC_MSG_ERROR(Failed to find lz4.h for lz4 compression support.
-Use --disable-lz4 to continue without it.)
+	err_msg="$err_msg$nl- Failed to find lz4.h for lz4 compression support."
+	no_lib="$no_lib lz4"
     fi
 else
     AC_MSG_RESULT(no)
 fi
 
+if test x"$no_lib" != x; then
+    echo ""
+    echo "Configure found the following issues:"
+    echo "$err_msg"
+    echo ""
+    echo "See the INSTALL file for hints on how to install the missing libraries and/or"
+    echo "how to generate (or fetch) manpages:"
+    echo "    https://github.com/WayneD/rsync/blob/master/INSTALL.md"
+    echo ""
+    echo "To disable one or more features, the relevant configure options are:"
+    for lib in $no_lib; do
+	echo "    --disable-$lib"
+    done
+    echo ""
+    AC_MSG_ERROR(Aborting configure run)
+fi
+
 AC_CACHE_CHECK([if makedev takes 3 args],rsync_cv_MAKEDEV_TAKES_3_ARGS,[
 AC_RUN_IFELSE([AC_LANG_SOURCE([[
+#ifdef HAVE_SYS_TYPES_H
 #include <sys/types.h>
+#endif
 #ifdef MAJOR_IN_MKDEV
 #include <sys/mkdev.h>
 # if !defined makedev && (defined mkdev || defined _WIN32 || defined __WIN32__)
@@ -625,7 +633,11 @@ fi
 
 AC_TYPE_UID_T
 AC_CHECK_TYPES([mode_t,off_t,size_t,pid_t,id_t])
-AC_TYPE_GETGROUPS
+if test "$cross_compiling" = no; then
+    AC_TYPE_GETGROUPS
+else
+    AC_DEFINE([GETGROUPS_T],[gid_t],[Define to the type of elements in the array set by `getgroups'. Usually this is either `int' or `gid_t'.])
+fi
 AC_CHECK_MEMBERS([struct stat.st_rdev,
 		  struct stat.st_mtimensec,
 		  struct stat.st_mtimespec.tv_nsec,
@@ -695,7 +707,9 @@ AC_SEARCH_LIBS(libiconv_open, iconv)
 AC_MSG_CHECKING([for iconv declaration])
 AC_CACHE_VAL(am_cv_proto_iconv, [
     AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[
+#if HAVE_STDLIB_H
 #include <stdlib.h>
+#endif
 #include <iconv.h>
 extern
 #ifdef __cplusplus
@@ -719,31 +733,47 @@ dnl AC_MSG_NOTICE([Looking in libraries: $LIBS])
 AC_REPLACE_FUNCS([inet_ntop inet_pton])
 
 AC_HAVE_TYPE([struct addrinfo], [#include <netdb.h>])
-AC_HAVE_TYPE([struct sockaddr_storage], [#include <sys/types.h>
-#include <sys/socket.h>])
+AC_HAVE_TYPE([struct sockaddr_storage], [
+#ifdef HAVE_SYS_TYPES_H
+#include <sys/types.h>
+#endif
+#ifdef HAVE_SYS_SOCKET_H
+#include <sys/socket.h>
+#endif])
 
 # Irix 6.5 has getaddrinfo but not the corresponding defines, so use
 #   builtin getaddrinfo if one of the defines don't exist
 AC_CACHE_CHECK([whether defines needed by getaddrinfo exist],
                rsync_cv_HAVE_GETADDR_DEFINES,[
 			AC_EGREP_CPP(yes, [
-			#include <sys/types.h>
-			#include <sys/socket.h>
-			#include <netdb.h>
-			#ifdef AI_PASSIVE
-			yes
-			#endif],
+#ifdef HAVE_SYS_TYPES_H
+#include <sys/types.h>
+#endif
+#ifdef HAVE_SYS_SOCKET_H
+#include <sys/socket.h>
+#endif
+#ifdef HAVE_NETDB_H
+#include <netdb.h>
+#endif
+#ifdef AI_PASSIVE
+yes
+#endif],
 			rsync_cv_HAVE_GETADDR_DEFINES=yes,
 			rsync_cv_HAVE_GETADDR_DEFINES=no)])
-AS_IF([test x"$rsync_cv_HAVE_GETADDR_DEFINES" = x"yes" -a x"$ac_cv_type_struct_addrinfo" = x"yes"],[
+AS_IF([test x"$rsync_cv_HAVE_GETADDR_DEFINES" = x"yes" && test x"$ac_cv_type_struct_addrinfo" = x"yes"],[
 	# Tru64 UNIX has getaddrinfo() but has it renamed in libc as
 	# something else so we must include <netdb.h> to get the
 	# redefinition.
 	AC_CHECK_FUNCS(getaddrinfo, ,
 		[AC_MSG_CHECKING([for getaddrinfo by including <netdb.h>])
-		AC_LINK_IFELSE([AC_LANG_PROGRAM([[#include <sys/types.h>
-		#include <sys/socket.h>
-		#include <netdb.h>]], [[getaddrinfo(NULL, NULL, NULL, NULL);]])],[AC_MSG_RESULT([yes])
+		AC_LINK_IFELSE([AC_LANG_PROGRAM([[
+#ifdef HAVE_SYS_TYPES_H
+#include <sys/types.h>
+#endif
+#ifdef HAVE_SYS_SOCKET_H
+#include <sys/socket.h>
+#endif
+#include <netdb.h>]], [[getaddrinfo(NULL, NULL, NULL, NULL);]])],[AC_MSG_RESULT([yes])
 			AC_DEFINE(HAVE_GETADDRINFO, 1,
 				[Define to 1 if you have the "getaddrinfo" function and required types.])],[AC_MSG_RESULT([no])
 			AC_LIBOBJ([getaddrinfo])])])
@@ -753,16 +783,24 @@ AC_CHECK_MEMBER([struct sockaddr.sa_len],
 		[ AC_DEFINE(HAVE_SOCKADDR_LEN, 1, [Do we have sockaddr.sa_len?]) ],
 		[],
 		[
+#ifdef HAVE_SYS_TYPES_H
 #include <sys/types.h>
+#endif
+#ifdef HAVE_SYS_SOCKET_H
 #include <sys/socket.h>
+#endif
 ])
 
 AC_CHECK_MEMBER([struct sockaddr_in.sin_len],
 		[ AC_DEFINE(HAVE_SOCKADDR_IN_LEN, 1, [Do we have sockaddr_in.sin_len?]) ],
 		[],
 		[
+#ifdef HAVE_SYS_TYPES_H
 #include <sys/types.h>
+#endif
+#ifdef HAVE_SYS_SOCKET_H
 #include <sys/socket.h>
+#endif
 #include <netinet/in.h>
 ])
 
@@ -770,8 +808,12 @@ AC_CHECK_MEMBER([struct sockaddr_un.sun_len],
 		[ AC_DEFINE(HAVE_SOCKADDR_UN_LEN, 1, [Do we have sockaddr_un.sun_len?]) ],
 		[],
 		[
+#ifdef HAVE_SYS_TYPES_H
 #include <sys/types.h>
+#endif
+#ifdef HAVE_SYS_SOCKET_H
 #include <sys/socket.h>
+#endif
 #include <netinet/in.h>
 ])
 
@@ -779,8 +821,12 @@ AC_CHECK_MEMBER([struct sockaddr_in6.sin6_scope_id],
 		[ AC_DEFINE(HAVE_SOCKADDR_IN6_SCOPE_ID, 1, [Do we have sockaddr_in6.sin6_scope_id?]) ],
 		[],
 		[
+#ifdef HAVE_SYS_TYPES_H
 #include <sys/types.h>
+#endif
+#ifdef HAVE_SYS_SOCKET_H
 #include <sys/socket.h>
+#endif
 #include <netinet/in.h>
 ])
 
@@ -824,11 +870,12 @@ AC_CHECK_FUNCS(waitpid wait4 getcwd chown chmod lchmod mknod mkfifo \
     fchmod fstat ftruncate strchr readlink link utime utimes lutimes strftime \
     chflags getattrlist mktime innetgr linkat \
     memmove lchown vsnprintf snprintf vasprintf asprintf setsid strpbrk \
-    strlcat strlcpy strtol mallinfo getgroups setgroups geteuid getegid \
+    strlcat strlcpy strtol mallinfo mallinfo2 getgroups setgroups geteuid getegid \
     setlocale setmode open64 lseek64 mkstemp64 mtrace va_copy __va_copy \
     seteuid strerror putenv iconv_open locale_charset nl_langinfo getxattr \
     extattr_get_link sigaction sigprocmask setattrlist getgrouplist \
-    initgroups utimensat posix_fallocate attropen setvbuf nanosleep usleep)
+    initgroups utimensat posix_fallocate attropen setvbuf nanosleep usleep \
+    setenv unsetenv)
 
 dnl cygwin iconv.h defines iconv_open as libiconv_open
 if test x"$ac_cv_func_iconv_open" != x"yes"; then
@@ -839,7 +886,9 @@ dnl Preallocation stuff (also fallocate, posix_fallocate function tests above):
 
 AC_CACHE_CHECK([for useable fallocate],rsync_cv_have_fallocate,[
 AC_LINK_IFELSE([AC_LANG_PROGRAM([[#include <fcntl.h>
-#include <sys/types.h>]], [[fallocate(0, 0, 0, 0);]])],[rsync_cv_have_fallocate=yes],[rsync_cv_have_fallocate=no])])
+#ifdef HAVE_SYS_TYPES_H
+#include <sys/types.h>
+#endif]], [[fallocate(0, 0, 0, 0);]])],[rsync_cv_have_fallocate=yes],[rsync_cv_have_fallocate=no])])
 if test x"$rsync_cv_have_fallocate" = x"yes"; then
     AC_DEFINE(HAVE_FALLOCATE, 1, [Define to 1 if you have the fallocate function and it compiles and links without error])
 fi
@@ -876,8 +925,12 @@ AC_PREPROC_IFELSE([AC_LANG_SOURCE([[
 
 AC_CACHE_CHECK([for SYS_fallocate],rsync_cv_have_sys_fallocate,[
 AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[#include <sys/syscall.h>
+#ifdef HAVE_UNISTD_H
 #include <unistd.h>
-#include <sys/types.h>]], [[syscall(SYS_fallocate, 0, 0, (loff_t)0, (loff_t)0);]])],[rsync_cv_have_sys_fallocate=yes],[rsync_cv_have_sys_fallocate=no])])
+#endif
+#ifdef HAVE_SYS_TYPES_H
+#include <sys/types.h>
+#endif]], [[syscall(SYS_fallocate, 0, 0, (loff_t)0, (loff_t)0);]])],[rsync_cv_have_sys_fallocate=yes],[rsync_cv_have_sys_fallocate=no])])
 if test x"$rsync_cv_have_sys_fallocate" = x"yes"; then
     AC_DEFINE(HAVE_SYS_FALLOCATE, 1, [Define to 1 if you have the SYS_fallocate syscall number])
 fi
@@ -904,7 +957,7 @@ if test $ac_cv_func_getpgrp = yes; then
 fi
 
 AC_ARG_ENABLE(iconv-open,
-    AS_HELP_STRING([--disable-iconv-open],[disable all use of iconv_open() function]),
+    AS_HELP_STRING([--disable-iconv-open],[disable to avoid all use of iconv_open()]),
     [], [enable_iconv_open=$ac_cv_func_iconv_open])
 
 if test x"$enable_iconv_open" != x"no"; then
@@ -912,7 +965,7 @@ if test x"$enable_iconv_open" != x"no"; then
 fi
 
 AC_ARG_ENABLE(iconv,
-    AS_HELP_STRING([--disable-iconv],[disable rsync's --iconv option]),
+    AS_HELP_STRING([--disable-iconv],[disable to omit the --iconv option]),
     [], [enable_iconv=$enable_iconv_open])
 AH_TEMPLATE([ICONV_OPTION],
 [Define if you want the --iconv option.  Specifying a value will set the
@@ -996,8 +1049,12 @@ fi
 
 AC_CACHE_CHECK([for working socketpair],rsync_cv_HAVE_SOCKETPAIR,[
 AC_RUN_IFELSE([AC_LANG_SOURCE([[
+#ifdef HAVE_SYS_TYPES_H
 #include <sys/types.h>
+#endif
+#ifdef HAVE_SYS_SOCKET_H
 #include <sys/socket.h>
+#endif
 
 int main(void) {
        int fd[2];
@@ -1058,13 +1115,16 @@ else
 fi
 
 AC_CACHE_CHECK([for unsigned char],rsync_cv_SIGNED_CHAR_OK,[
-AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[]], [[signed char *s = ""]])],[rsync_cv_SIGNED_CHAR_OK=yes],[rsync_cv_SIGNED_CHAR_OK=no])])
+AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[]], [[signed char *s = (signed char *)""]])],[rsync_cv_SIGNED_CHAR_OK=yes],[rsync_cv_SIGNED_CHAR_OK=no])])
 if test x"$rsync_cv_SIGNED_CHAR_OK" = x"yes"; then
     AC_DEFINE(SIGNED_CHAR_OK, 1, [Define to 1 if "signed char" is a valid type])
 fi
 
 AC_CACHE_CHECK([for broken readdir],rsync_cv_HAVE_BROKEN_READDIR,[
-AC_RUN_IFELSE([AC_LANG_SOURCE([[#include <sys/types.h>
+AC_RUN_IFELSE([AC_LANG_SOURCE([[
+#ifdef HAVE_SYS_TYPES_H
+#include <sys/types.h>
+#endif
 #include <dirent.h>
 int main(void) { struct dirent *di; DIR *d = opendir("."); di = readdir(d);
 if (di && di->d_name[-2] == '.' && di->d_name[-1] == 0 &&
@@ -1074,7 +1134,10 @@ if test x"$rsync_cv_HAVE_BROKEN_READDIR" = x"yes"; then
 fi
 
 AC_CACHE_CHECK([for utimbuf],rsync_cv_HAVE_STRUCT_UTIMBUF,[
-AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[#include <sys/types.h>
+AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[
+#ifdef HAVE_SYS_TYPES_H
+#include <sys/types.h>
+#endif
 #include <utime.h>]], [[struct utimbuf tbuf;  tbuf.actime = 0; tbuf.modtime = 1; return utime("foo.c",&tbuf);]])],[rsync_cv_HAVE_STRUCT_UTIMBUF=yes],[rsync_cv_HAVE_STRUCT_UTIMBUF=no])])
 if test x"$rsync_cv_HAVE_STRUCT_UTIMBUF" = x"yes"; then
     AC_DEFINE(HAVE_STRUCT_UTIMBUF, 1, [Define to 1 if you have the "struct utimbuf" type])
@@ -1082,17 +1145,23 @@ fi
 
 AC_CACHE_CHECK([if gettimeofday takes tz argument],rsync_cv_HAVE_GETTIMEOFDAY_TZ,[
 AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[#include <sys/time.h>
-#include <unistd.h>]], [[struct timeval tv; return gettimeofday(&tv, NULL);]])],[rsync_cv_HAVE_GETTIMEOFDAY_TZ=yes],[rsync_cv_HAVE_GETTIMEOFDAY_TZ=no])])
+#ifdef HAVE_UNISTD_H
+#include <unistd.h>
+#endif]], [[struct timeval tv; return gettimeofday(&tv, NULL);]])],[rsync_cv_HAVE_GETTIMEOFDAY_TZ=yes],[rsync_cv_HAVE_GETTIMEOFDAY_TZ=no])])
 if test x"$rsync_cv_HAVE_GETTIMEOFDAY_TZ" != x"no"; then
     AC_DEFINE(HAVE_GETTIMEOFDAY_TZ, 1, [Define to 1 if gettimeofday() takes a time-zone arg])
 fi
 
 AC_CACHE_CHECK([for C99 vsnprintf],rsync_cv_HAVE_C99_VSNPRINTF,[
 AC_RUN_IFELSE([AC_LANG_SOURCE([[
+#ifdef HAVE_SYS_TYPES_H
 #include <sys/types.h>
+#endif
 #include <stdarg.h>
 #include <stdio.h>
+#if HAVE_STDLIB_H
 #include <stdlib.h>
+#endif
 #include <string.h>
 void foo(const char *format, ...) {
        va_list ap;
@@ -1115,9 +1184,13 @@ fi
 
 AC_CACHE_CHECK([for secure mkstemp],rsync_cv_HAVE_SECURE_MKSTEMP,[
 AC_RUN_IFELSE([AC_LANG_SOURCE([[#include <stdlib.h>
+#ifdef HAVE_SYS_TYPES_H
 #include <sys/types.h>
+#endif
 #include <sys/stat.h>
+#ifdef HAVE_UNISTD_H
 #include <unistd.h>
+#endif
 int main(void) {
   struct stat st;
   char tpl[20]="/tmp/test.XXXXXX";
@@ -1206,6 +1279,9 @@ AC_SUBST(OBJ_RESTORE)
 AC_SUBST(CC_SHOBJ_FLAG)
 AC_SUBST(BUILD_POPT)
 AC_SUBST(BUILD_ZLIB)
+AC_SUBST(MAKE_RRSYNC)
+AC_SUBST(MAKE_RRSYNC_1)
+AC_SUBST(GEN_RRSYNC)
 AC_SUBST(MAKE_MAN)
 
 AC_CHECK_FUNCS(_acl __acl _facl __facl)
@@ -1214,7 +1290,7 @@ AC_CHECK_FUNCS(_acl __acl _facl __facl)
 
 AC_MSG_CHECKING([whether to support ACLs])
 AC_ARG_ENABLE(acl-support,
-	AS_HELP_STRING([--disable-acl-support],[disable ACL support]))
+	AS_HELP_STRING([--disable-acl-support],[disable to omit ACL support]))
 
 if test x"$enable_acl_support" = x"no"; then
     AC_MSG_RESULT(no)
@@ -1260,16 +1336,26 @@ else
 	AC_MSG_RESULT(running tests:)
 	AC_CHECK_LIB(acl,acl_get_file)
 	AC_CACHE_CHECK([for ACL support],samba_cv_HAVE_POSIX_ACLS,[
-	    AC_LINK_IFELSE([AC_LANG_PROGRAM([[#include <sys/types.h>
-#include <sys/acl.h>]], [[ acl_t acl; int entry_id; acl_entry_t *entry_p; return acl_get_entry( acl, entry_id, entry_p);]])],[samba_cv_HAVE_POSIX_ACLS=yes],[samba_cv_HAVE_POSIX_ACLS=no])])
+	    AC_LINK_IFELSE([AC_LANG_PROGRAM([[
+#ifdef HAVE_SYS_TYPES_H
+#include <sys/types.h>
+#endif
+#ifdef HAVE_SYS_ACL_H
+#include <sys/acl.h>
+#endif]], [[ acl_t acl; int entry_id; acl_entry_t *entry_p; return acl_get_entry( acl, entry_id, entry_p);]])],[samba_cv_HAVE_POSIX_ACLS=yes],[samba_cv_HAVE_POSIX_ACLS=no])])
 	AC_MSG_CHECKING(ACL test results)
 	if test x"$samba_cv_HAVE_POSIX_ACLS" = x"yes"; then
 	    AC_MSG_RESULT(Using posix ACLs)
 	    AC_DEFINE(HAVE_POSIX_ACLS, 1, [true if you have posix ACLs])
 	    AC_DEFINE(SUPPORT_ACLS, 1)
 	    AC_CACHE_CHECK([for acl_get_perm_np],samba_cv_HAVE_ACL_GET_PERM_NP,[
-		AC_LINK_IFELSE([AC_LANG_PROGRAM([[#include <sys/types.h>
-#include <sys/acl.h>]], [[ acl_permset_t permset_d; acl_perm_t perm; return acl_get_perm_np( permset_d, perm);]])],[samba_cv_HAVE_ACL_GET_PERM_NP=yes],[samba_cv_HAVE_ACL_GET_PERM_NP=no])])
+		AC_LINK_IFELSE([AC_LANG_PROGRAM([[
+#ifdef HAVE_SYS_TYPES_H
+#include <sys/types.h>
+#endif
+#ifdef HAVE_SYS_ACL_H
+#include <sys/acl.h>
+#endif]], [[ acl_permset_t permset_d; acl_perm_t perm; return acl_get_perm_np( permset_d, perm);]])],[samba_cv_HAVE_ACL_GET_PERM_NP=yes],[samba_cv_HAVE_ACL_GET_PERM_NP=no])])
 	    if test x"$samba_cv_HAVE_ACL_GET_PERM_NP" = x"yes"; then
 		AC_DEFINE(HAVE_ACL_GET_PERM_NP, 1, [true if you have acl_get_perm_np])
 	    fi
@@ -1288,7 +1374,7 @@ fi
 # check for extended attribute support
 AC_MSG_CHECKING(whether to support extended attributes)
 AC_ARG_ENABLE(xattr-support,
-    AS_HELP_STRING([--disable-xattr-support],[disable extended attributes]),
+    AS_HELP_STRING([--disable-xattr-support],[disable to omit extended attributes]),
     [], [case "$ac_cv_func_getxattr$ac_cv_func_extattr_get_link$ac_cv_func_attropen" in
 	*yes*) enable_xattr_support=maybe ;;
 	*) enable_xattr_support=no ;;
@@ -1334,7 +1420,7 @@ else
     esac
 fi
 
-if test x"$enable_acl_support" = x"no" -o x"$enable_xattr_support" = x"no" -o x"$enable_iconv" = x"no"; then
+if test x"$enable_acl_support" = x"no" || test x"$enable_xattr_support" = x"no" || test x"$enable_iconv" = x"no"; then
     AC_MSG_CHECKING([whether $CC supports -Wno-unused-parameter])
     OLD_CFLAGS="$CFLAGS"
     CFLAGS="$CFLAGS -Wno-unused-parameter"

csprotocol.txt

@@ -7,39 +7,54 @@ basically a summary of clientserver.c and authenticate.c.
 This is the protocol used for rsync --daemon; i.e. connections to port
 873 rather than invocations over a remote shell.
 
-When the server accepts a connection, it prints a greeting
+When the server accepts a connection, it prints a newline-terminated
+greeting line:
 
-  @RSYNCD: <version>.<subprotocol>
+  @RSYNCD: <version>.<subprotocol> <digest1> <digestN>
 
-where <version> is the numeric version (see PROTOCOL_VERSION in rsync.h)
-'.' is a literal period, and <subprotocol> is the numeric subprotocol
-version (see SUBPROTOCOL_VERSION -- it will be 0 for final releases).
-Protocols prior to 30 only output <version> alone.  The daemon expects
-to see a similar greeting back from the client.  For protocols prior to
-30, an absent ".<subprotocol>" value is assumed to be 0.  For protocol
-30, an absent value is a fatal error.  The daemon then follows this line
-with a free-format text message-of-the-day (if any is defined).
+The <version> is the numeric version (see PROTOCOL_VERSION in rsync.h)
+The <subprotocol> is the numeric subprotocol version (which is 0 for a
+final protocol version, as the SUBPROTOCOL_VERSION define discusses).
+The <digestN> names are the authentication digest algorithms that the
+daemon supports, listed in order of preference.
+
+An rsync prior to 3.2.7 omits the digest names.  An rsync prior to 3.0.0
+also omits the period and the <subprotocol> value.  Since a final
+protocol has a subprotocol value of 0, a missing subprotocol value is
+assumed to be 0 for any protocol prior to 30.  It is considered a fatal
+error for protocol 30 and above to omit it.  It is considered a fatal
+error for protocol 32 and above to omit the digest name list (currently
+31 is the newest protocol).
+
+The daemon expects to see a similar greeting line back from the client.
+Once received, the daemon follows the opening line with a free-format
+text message-of-the-day (if any is defined).
 
 The server is now in the connected state.  The client can either send
-the command
+the command:
 
   #list
 
-to get a listing of modules, or the name of a module.  After this, the
+(to get a listing of modules) or the name of a module.  After this, the
 connection is now bound to a particular module.  Access per host for
 this module is now checked, as is per-module connection limits.
 
-If authentication is required to use this module, the server will say
+If authentication is required to use this module, the server will say:
 
   @RSYNCD: AUTHREQD <challenge>
 
 where <challenge> is a random string of base64 characters.  The client
-must respond with
+must respond with:
 
   <user> <response>
 
-where <user> is the username they claim to be, and <response> is the
-base64 form of the MD4 hash of challenge+password.
+The <user> is the username they claim to be. The <response> is the
+base64 form of the digest hash of the challenge+password string. The
+chosen digest method is the most preferred client method that is also in
+the server's list.  If no digest list was explicitly provided, the side
+expecting a list assumes the other side provided either the single name
+"md5" (for a negotiated protocol 30 or 31), or the single name "md4"
+(for an older protocol).
 
 At this point the server applies all remaining constraints before
 handing control to the client, including switching uid/gid, setting up
@@ -76,6 +91,13 @@ stay tuned (or write it yourself!).
 ------------
 Protocol version changes
 
+31	(2013-09-28, 3.1.0)
+
+	Initial release of protocol 31 had no changes.  Rsync 3.2.7
+	introduced the suffixed list of digest names on the greeting
+	line.  The presence of the list is allowed even if the greeting
+	indicates an older protocol version number.
+
 30	(2007-10-04, 3.0.0pre1)
 
 	The use of a ".<subprotocol>" number was added to

daemon-parm.txt

@@ -60,9 +60,9 @@ BOOL	read_only		True
 BOOL	reverse_lookup		True
 BOOL	strict_modes		True
 BOOL	transfer_logging	False
-BOOL	use_chroot		True
 BOOL	write_only		False
 
 BOOL3	munge_symlinks		Unset
 BOOL3	numeric_ids		Unset
 BOOL3	open_noatime		Unset
+BOOL3	use_chroot		Unset

delete.c

@@ -4,7 +4,7 @@
  * Copyright (C) 1996-2000 Andrew Tridgell
  * Copyright (C) 1996 Paul Mackerras
  * Copyright (C) 2002 Martin Pool <mbp@samba.org>
- * Copyright (C) 2003-2020 Wayne Davison
+ * Copyright (C) 2003-2024 Wayne Davison
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -188,7 +188,7 @@ enum delret delete_item(char *fbuf, uint16 mode, uint16 flags)
 				stats.deleted_symlinks++;
 #endif
 			else if (IS_DEVICE(mode))
-				stats.deleted_symlinks++;
+				stats.deleted_devices++;
 			else
 				stats.deleted_specials++;
 		}

exclude.c

@@ -4,7 +4,7 @@
  * Copyright (C) 1996-2001 Andrew Tridgell <tridge@samba.org>
  * Copyright (C) 1996 Paul Mackerras
  * Copyright (C) 2002 Martin Pool
- * Copyright (C) 2003-2020 Wayne Davison
+ * Copyright (C) 2003-2024 Wayne Davison
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -21,21 +21,25 @@
  */
 
 #include "rsync.h"
-#include "default-cvsignore.h"
 #include "ifuncs.h"
 
 extern int am_server;
 extern int am_sender;
+extern int am_generator;
 extern int eol_nulls;
 extern int io_error;
+extern int xfer_dirs;
+extern int recurse;
 extern int local_server;
 extern int prune_empty_dirs;
 extern int ignore_perishable;
+extern int relative_paths;
 extern int delete_mode;
 extern int delete_excluded;
 extern int cvs_exclude;
 extern int sanitize_paths;
 extern int protocol_version;
+extern int trust_sender_args;
 extern int module_id;
 
 extern char curr_dir[MAXPATHLEN];
@@ -45,8 +49,11 @@ extern unsigned int module_dirlen;
 filter_rule_list filter_list = { .debug_type = "" };
 filter_rule_list cvs_filter_list = { .debug_type = " [global CVS]" };
 filter_rule_list daemon_filter_list = { .debug_type = " [daemon]" };
+filter_rule_list implied_filter_list = { .debug_type = " [implied]" };
 
 int saw_xattr_filter = 0;
+int trust_sender_args = 0;
+int trust_sender_filter = 0;
 
 /* Need room enough for ":MODS " prefix plus some room to grow. */
 #define MAX_RULE_PREFIX (16)
@@ -71,6 +78,10 @@ static filter_rule **mergelist_parents;
 static int mergelist_cnt = 0;
 static int mergelist_size = 0;
 
+#define LOCAL_RULE   1
+#define REMOTE_RULE  2
+static uchar cur_elide_value = REMOTE_RULE;
+
 /* Each filter_list_struct describes a singly-linked list by keeping track
  * of both the head and tail pointers.  The list is slightly unusual in that
  * a parent-dir's content can be appended to the end of the local list in a
@@ -153,13 +164,17 @@ static void add_rule(filter_rule_list *listp, const char *pat, unsigned int pat_
 {
 	const char *cp;
 	unsigned int pre_len, suf_len, slash_cnt = 0;
+	char *mention_rule_suffix;
 
-	if (DEBUG_GTE(FILTER, 2)) {
-		rprintf(FINFO, "[%s] add_rule(%s%.*s%s)%s\n",
+	if (DEBUG_GTE(FILTER, 1) && pat_len && (pat[pat_len-1] == ' ' || pat[pat_len-1] == '\t'))
+		mention_rule_suffix = " -- CAUTION: trailing whitespace!";
+	else
+		mention_rule_suffix = DEBUG_GTE(FILTER, 2) ? "" : NULL;
+	if (mention_rule_suffix) {
+		rprintf(FINFO, "[%s] add_rule(%s%.*s%s)%s%s\n",
 			who_am_i(), get_rule_prefix(rule, pat, 0, NULL),
-			(int)pat_len, pat,
-			(rule->rflags & FILTRULE_DIRECTORY) ? "/" : "",
-			listp->debug_type);
+			(int)pat_len, pat, (rule->rflags & FILTRULE_DIRECTORY) ? "/" : "",
+			listp->debug_type, mention_rule_suffix);
 	}
 
 	/* These flags also indicate that we're reading a list that
@@ -209,6 +224,7 @@ static void add_rule(filter_rule_list *listp, const char *pat, unsigned int pat_
 				slash_cnt++;
 		}
 	}
+	rule->elide = 0;
 	strlcpy(rule->pattern + pre_len, pat, pat_len + 1);
 	pat_len += pre_len;
 	if (suf_len) {
@@ -289,6 +305,271 @@ static void add_rule(filter_rule_list *listp, const char *pat, unsigned int pat_
 	}
 }
 
+/* If the wildcards failed, the remote shell might give us a file matching the literal
+ * wildcards.  Since "*" & "?" already match themselves, this just needs to deal with
+ * failed "[foo]" idioms.
+ */
+static void maybe_add_literal_brackets_rule(filter_rule const *based_on, int arg_len)
+{
+	filter_rule *rule;
+	const char *arg = based_on->pattern, *cp;
+	char *p;
+	int cnt = 0;
+
+	if (arg_len < 0)
+		arg_len = strlen(arg);
+
+	for (cp = arg; *cp; cp++) {
+		if (*cp == '\\' && cp[1]) {
+			cp++;
+		} else if (*cp == '[')
+			cnt++;
+	}
+	if (!cnt)
+		return;
+
+	rule = new0(filter_rule);
+	rule->rflags = based_on->rflags;
+	rule->u.slash_cnt = based_on->u.slash_cnt;
+	p = rule->pattern = new_array(char, arg_len + cnt + 1);
+	for (cp = arg; *cp; ) {
+		if (*cp == '\\' && cp[1]) {
+			*p++ = *cp++;
+		} else if (*cp == '[')
+			*p++ = '\\';
+		*p++ = *cp++;
+	}
+	*p++ = '\0';
+
+	rule->next = implied_filter_list.head;
+	implied_filter_list.head = rule;
+	if (DEBUG_GTE(FILTER, 3)) {
+		rprintf(FINFO, "[%s] add_implied_include(%s%s)\n", who_am_i(), rule->pattern,
+			rule->rflags & FILTRULE_DIRECTORY ? "/" : "");
+	}
+}
+
+static char *partial_string_buf = NULL;
+static int partial_string_len = 0;
+void implied_include_partial_string(const char *s_start, const char *s_end)
+{
+	partial_string_len = s_end - s_start;
+	if (partial_string_len <= 0 || partial_string_len >= MAXPATHLEN) { /* too-large should be impossible... */
+		partial_string_len = 0;
+		return;
+	}
+	if (!partial_string_buf)
+		partial_string_buf = new_array(char, MAXPATHLEN);
+	memcpy(partial_string_buf, s_start, partial_string_len);
+}
+
+void free_implied_include_partial_string()
+{
+	if (partial_string_buf) {
+		if (partial_string_len)
+			add_implied_include("", 0);
+		free(partial_string_buf);
+		partial_string_buf = NULL;
+	}
+	partial_string_len = 0; /* paranoia */
+}
+
+/* Each arg the client sends to the remote sender turns into an implied include
+ * that the receiver uses to validate the file list from the sender. */
+void add_implied_include(const char *arg, int skip_daemon_module)
+{
+	int arg_len, saw_wild = 0, saw_live_open_brkt = 0, backslash_cnt = 0;
+	int slash_cnt = 0;
+	const char *cp;
+	char *p;
+	if (trust_sender_args)
+		return;
+	if (partial_string_len) {
+		arg_len = strlen(arg);
+		if (partial_string_len + arg_len >= MAXPATHLEN) {
+			partial_string_len = 0;
+			return; /* Should be impossible... */
+		}
+		memcpy(partial_string_buf + partial_string_len, arg, arg_len + 1);
+		partial_string_len = 0;
+		arg = partial_string_buf;
+	}
+	if (skip_daemon_module) {
+		if ((cp = strchr(arg, '/')) != NULL)
+			arg = cp + 1;
+		else
+			arg = "";
+	}
+	if (relative_paths) {
+		if ((cp = strstr(arg, "/./")) != NULL)
+			arg = cp + 3;
+	} else if ((cp = strrchr(arg, '/')) != NULL) {
+		arg = cp + 1;
+	}
+	if (*arg == '.' && arg[1] == '\0')
+		arg++;
+	arg_len = strlen(arg);
+	if (arg_len) {
+		char *new_pat;
+		if (strpbrk(arg, "*[?")) {
+			/* We need to add room to escape backslashes if wildcard chars are present. */
+			for (cp = arg; (cp = strchr(cp, '\\')) != NULL; cp++)
+				arg_len++;
+			saw_wild = 1;
+		}
+		arg_len++; /* Leave room for the prefixed slash */
+		p = new_pat = new_array(char, arg_len + 1);
+		*p++ = '/';
+		slash_cnt++;
+		for (cp = arg; *cp; ) {
+			switch (*cp) {
+			  case '\\':
+				if (cp[1] == ']') {
+					if (!saw_wild)
+						cp++; /* A \] in a non-wild filter causes a problem, so drop the \ . */
+				} else if (!strchr("*[?", cp[1])) {
+					backslash_cnt++;
+					if (saw_wild)
+						*p++ = '\\';
+				}
+				*p++ = *cp++;
+				break;
+			  case '/':
+				if (p[-1] == '/') { /* This is safe because of the initial slash. */
+					if (*++cp == '\0') {
+						slash_cnt--;
+						p--;
+					}
+				} else if (cp[1] == '\0') {
+					cp++;
+				} else {
+					slash_cnt++;
+					*p++ = *cp++;
+				}
+				break;
+			  case '.':
+				if (p[-1] == '/') {
+					if (cp[1] == '/') {
+						cp += 2;
+						if (!*cp) {
+							slash_cnt--;
+							p--;
+						}
+					} else if (cp[1] == '\0') {
+						cp++;
+						slash_cnt--;
+						p--;
+					} else
+						*p++ = *cp++;
+				} else
+					*p++ = *cp++;
+				break;
+			  case '[':
+				saw_live_open_brkt = 1;
+				*p++ = *cp++;
+				break;
+			  default:
+				*p++ = *cp++;
+				break;
+			}
+		}
+		*p = '\0';
+		arg_len = p - new_pat;
+		if (!arg_len)
+			free(new_pat);
+		else {
+			filter_rule *rule = new0(filter_rule);
+			rule->rflags = FILTRULE_INCLUDE + (saw_wild ? FILTRULE_WILD : 0);
+			rule->u.slash_cnt = slash_cnt;
+			arg = rule->pattern = new_pat;
+			if (!implied_filter_list.head)
+				implied_filter_list.head = implied_filter_list.tail = rule;
+			else {
+				rule->next = implied_filter_list.head;
+				implied_filter_list.head = rule;
+			}
+			if (DEBUG_GTE(FILTER, 3))
+				rprintf(FINFO, "[%s] add_implied_include(%s)\n", who_am_i(), arg);
+			if (saw_live_open_brkt)
+				maybe_add_literal_brackets_rule(rule, arg_len);
+			if (relative_paths && slash_cnt) {
+				int sub_slash_cnt = slash_cnt;
+				while ((p = strrchr(new_pat, '/')) != NULL && p != new_pat) {
+					filter_rule const *ent;
+					filter_rule *R_rule;
+					int found = 0;
+					*p = '\0';
+					for (ent = implied_filter_list.head; ent; ent = ent->next) {
+						if (ent != rule && strcmp(ent->pattern, new_pat) == 0) {
+							found = 1;
+							break;
+						}
+					}
+					if (found) {
+						*p = '/';
+						break; /* We added all parent dirs already */
+					}
+					R_rule = new0(filter_rule);
+					R_rule->rflags = FILTRULE_INCLUDE | FILTRULE_DIRECTORY;
+					/* Check if our sub-path has wildcards or escaped backslashes */
+					if (saw_wild && strpbrk(new_pat, "*[?\\"))
+						R_rule->rflags |= FILTRULE_WILD;
+					R_rule->pattern = strdup(new_pat);
+					R_rule->u.slash_cnt = --sub_slash_cnt;
+					R_rule->next = implied_filter_list.head;
+					implied_filter_list.head = R_rule;
+					if (DEBUG_GTE(FILTER, 3)) {
+						rprintf(FINFO, "[%s] add_implied_include(%s/)\n",
+							who_am_i(), R_rule->pattern);
+					}
+					if (saw_live_open_brkt)
+						maybe_add_literal_brackets_rule(R_rule, -1);
+				}
+				for (p = new_pat; sub_slash_cnt < slash_cnt; sub_slash_cnt++) {
+					p += strlen(p);
+					*p = '/';
+				}
+			}
+		}
+	}
+
+	if (recurse || xfer_dirs) {
+		/* Now create a rule with an added "/" & "**" or "*" at the end */
+		filter_rule *rule = new0(filter_rule);
+		rule->rflags = FILTRULE_INCLUDE | FILTRULE_WILD;
+		if (recurse)
+			rule->rflags |= FILTRULE_WILD2;
+		/* We must leave enough room for / * * \0. */
+		if (!saw_wild && backslash_cnt) {
+			/* We are appending a wildcard, so now the backslashes need to be escaped. */
+			p = rule->pattern = new_array(char, arg_len + backslash_cnt + 3 + 1);
+			for (cp = arg; *cp; ) { /* Note that arg_len != 0 because backslash_cnt > 0 */
+				if (*cp == '\\')
+					*p++ = '\\';
+				*p++ = *cp++;
+			}
+		} else {
+			p = rule->pattern = new_array(char, arg_len + 3 + 1);
+			if (arg_len) {
+				memcpy(p, arg, arg_len);
+				p += arg_len;
+			}
+		}
+		*p++ = '/';
+		*p++ = '*';
+		if (recurse)
+			*p++ = '*';
+		*p = '\0';
+		rule->u.slash_cnt = slash_cnt + 1;
+		rule->next = implied_filter_list.head;
+		implied_filter_list.head = rule;
+		if (DEBUG_GTE(FILTER, 3))
+			rprintf(FINFO, "[%s] add_implied_include(%s)\n", who_am_i(), rule->pattern);
+		if (saw_live_open_brkt)
+			maybe_add_literal_brackets_rule(rule, p - rule->pattern);
+	}
+}
+
 /* This frees any non-inherited items, leaving just inherited items on the list. */
 static void pop_filter_list(filter_rule_list *listp)
 {
@@ -439,7 +720,8 @@ static BOOL setup_merge_file(int mergelist_num, filter_rule *ex,
 	parent_dirscan = True;
 	while (*y) {
 		char save[MAXPATHLEN];
-		strlcpy(save, y, MAXPATHLEN);
+		/* copylen is strlen(y) which is < MAXPATHLEN. +1 for \0 */
+		size_t copylen = strlcpy(save, y, MAXPATHLEN) + 1;
 		*y = '\0';
 		dirbuf_len = y - dirbuf;
 		strlcpy(x, ex->pattern, MAXPATHLEN - (x - buf));
@@ -453,7 +735,7 @@ static BOOL setup_merge_file(int mergelist_num, filter_rule *ex,
 			lp->head = NULL;
 		}
 		lp->tail = NULL;
-		strlcpy(y, save, MAXPATHLEN);
+		strlcpy(y, save, copylen);
 		while ((*x++ = *y++) != '/') {}
 	}
 	parent_dirscan = False;
@@ -626,7 +908,7 @@ static int rule_matches(const char *fname, filter_rule *ex, int name_flags)
 	const char *strings[16]; /* more than enough */
 	const char *name = fname + (*fname == '/');
 
-	if (!*name)
+	if (!*name || ex->elide == cur_elide_value)
 		return 0;
 
 	if (!(name_flags & NAME_IS_XATTR) ^ !(ex->rflags & FILTRULE_XATTR))
@@ -703,11 +985,12 @@ static void report_filter_result(enum logcode code, char const *name,
 				 filter_rule const *ent,
 				 int name_flags, const char *type)
 {
+	int log_level = am_sender || am_generator ? 1 : 3;
+
 	/* If a trailing slash is present to match only directories,
 	 * then it is stripped out by add_rule().  So as a special
-	 * case we add it back in here. */
-
-	if (DEBUG_GTE(FILTER, 1)) {
+	 * case we add it back in the log output. */
+	if (DEBUG_GTE(FILTER, log_level)) {
 		static char *actions[2][2]
 		    = { {"show", "hid"}, {"risk", "protect"} };
 		const char *w = who_am_i();
@@ -715,7 +998,7 @@ static void report_filter_result(enum logcode code, char const *name,
 			      : name_flags & NAME_IS_DIR ? "directory"
 			      : "file";
 		rprintf(code, "[%s] %sing %s %s because of pattern %s%s%s\n",
-		    w, actions[*w!='s'][!(ent->rflags & FILTRULE_INCLUDE)],
+		    w, actions[*w=='g'][!(ent->rflags & FILTRULE_INCLUDE)],
 		    t, name, ent->pattern,
 		    ent->rflags & FILTRULE_DIRECTORY ? "/" : "", type);
 	}
@@ -741,6 +1024,15 @@ int name_is_excluded(const char *fname, int name_flags, int filter_level)
 	return 0;
 }
 
+int check_server_filter(filter_rule_list *listp, enum logcode code, const char *name, int name_flags)
+{
+	int ret;
+	cur_elide_value = LOCAL_RULE;
+	ret = check_filter(listp, code, name, name_flags);
+	cur_elide_value = REMOTE_RULE;
+	return ret;
+}
+
 /* Return -1 if file "name" is defined to be excluded by the specified
  * exclude list, 1 if it is included, and 0 if it was not matched. */
 int check_filter(filter_rule_list *listp, enum logcode code,
@@ -887,6 +1179,7 @@ static filter_rule *parse_rule_tok(const char **rulestr_ptr,
 		}
 		switch (ch) {
 		case ':':
+			trust_sender_filter = 1;
 			rule->rflags |= FILTRULE_PERDIR_MERGE
 				      | FILTRULE_FINISH_SETUP;
 			/* FALL THROUGH */
@@ -1053,7 +1346,7 @@ static void get_cvs_excludes(uint32 rflags)
 		return;
 	initialized = 1;
 
-	parse_filter_str(&cvs_filter_list, DEFAULT_CVSIGNORE,
+	parse_filter_str(&cvs_filter_list, default_cvsignore(),
 			 rule_template(rflags | (protocol_version >= 30 ? FILTRULE_PERISHABLE : 0)),
 			 0);
 
@@ -1295,7 +1588,7 @@ char *get_rule_prefix(filter_rule *rule, const char *pat, int for_xfer,
 
 static void send_rules(int f_out, filter_rule_list *flp)
 {
-	filter_rule *ent, *prev = NULL;
+	filter_rule *ent;
 
 	for (ent = flp->head; ent; ent = ent->next) {
 		unsigned int len, plen, dlen;
@@ -1310,21 +1603,15 @@ static void send_rules(int f_out, filter_rule_list *flp)
 		 * merge files as an optimization (since they can only have
 		 * include/exclude rules). */
 		if (ent->rflags & FILTRULE_SENDER_SIDE)
-			elide = am_sender ? 1 : -1;
+			elide = am_sender ? LOCAL_RULE : REMOTE_RULE;
 		if (ent->rflags & FILTRULE_RECEIVER_SIDE)
-			elide = elide ? 0 : am_sender ? -1 : 1;
+			elide = elide ? 0 : am_sender ? REMOTE_RULE : LOCAL_RULE;
 		else if (delete_excluded && !elide
 		 && (!(ent->rflags & FILTRULE_PERDIR_MERGE)
 		  || ent->rflags & FILTRULE_NO_PREFIXES))
-			elide = am_sender ? 1 : -1;
-		if (elide < 0) {
-			if (prev)
-				prev->next = ent->next;
-			else
-				flp->head = ent->next;
-		} else
-			prev = ent;
-		if (elide > 0)
+			elide = am_sender ? LOCAL_RULE : REMOTE_RULE;
+		ent->elide = elide;
+		if (elide == LOCAL_RULE)
 			continue;
 		if (ent->rflags & FILTRULE_CVS_IGNORE
 		    && !(ent->rflags & FILTRULE_MERGE_FILE)) {
@@ -1352,7 +1639,6 @@ static void send_rules(int f_out, filter_rule_list *flp)
 		if (dlen)
 			write_byte(f_out, '/');
 	}
-	flp->tail = prev;
 }
 
 /* This is only called by the client. */

fileio.c

@@ -3,7 +3,7 @@
  *
  * Copyright (C) 1998 Andrew Tridgell
  * Copyright (C) 2002 Martin Pool
- * Copyright (C) 2004-2020 Wayne Davison
+ * Copyright (C) 2004-2023 Wayne Davison
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -40,30 +40,34 @@ OFF_T preallocated_len = 0;
 static OFF_T sparse_seek = 0;
 static OFF_T sparse_past_write = 0;
 
-int sparse_end(int f, OFF_T size)
+int sparse_end(int f, OFF_T size, int updating_basis_or_equiv)
 {
-	int ret;
+	int ret = 0;
 
-	sparse_past_write = 0;
-
-	if (!sparse_seek)
-		return 0;
-
-#ifdef HAVE_FTRUNCATE
-	ret = do_ftruncate(f, size);
-#else
-	if (do_lseek(f, sparse_seek-1, SEEK_CUR) != size-1)
-		ret = -1;
-	else {
-		do {
-			ret = write(f, "", 1);
-		} while (ret < 0 && errno == EINTR);
-
-		ret = ret <= 0 ? -1 : 0;
-	}
+	if (updating_basis_or_equiv) {
+		if (sparse_seek && do_punch_hole(f, sparse_past_write, sparse_seek) < 0)
+			ret = -1;
+#ifdef HAVE_FTRUNCATE /* A compilation formality -- in-place requires ftruncate() */
+		else /* Just in case the original file was longer */
+			ret = do_ftruncate(f, size);
 #endif
+	} else if (sparse_seek) {
+#ifdef HAVE_FTRUNCATE
+		ret = do_ftruncate(f, size);
+#else
+		if (do_lseek(f, sparse_seek-1, SEEK_CUR) != size-1)
+			ret = -1;
+		else {
+			do {
+				ret = write(f, "", 1);
+			} while (ret < 0 && errno == EINTR);
 
-	sparse_seek = 0;
+			ret = ret <= 0 ? -1 : 0;
+		}
+#endif
+	}
+
+	sparse_past_write = sparse_seek = 0;
 
 	return ret;
 }

flist.c

@@ -4,7 +4,7 @@
  * Copyright (C) 1996 Andrew Tridgell
  * Copyright (C) 1996 Paul Mackerras
  * Copyright (C) 2001, 2002 Martin Pool <mbp@samba.org>
- * Copyright (C) 2002-2020 Wayne Davison
+ * Copyright (C) 2002-2023 Wayne Davison
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -33,7 +33,6 @@ extern int am_sender;
 extern int am_generator;
 extern int inc_recurse;
 extern int always_checksum;
-extern int checksum_type;
 extern int module_id;
 extern int ignore_errors;
 extern int numeric_ids;
@@ -43,6 +42,7 @@ extern int use_qsort;
 extern int xfer_dirs;
 extern int filesfrom_fd;
 extern int one_file_system;
+extern int copy_devices;
 extern int copy_dirlinks;
 extern int preserve_uid;
 extern int preserve_gid;
@@ -72,18 +72,20 @@ extern int need_unsorted_flist;
 extern int sender_symlink_iconv;
 extern int output_needs_newline;
 extern int sender_keeps_checksum;
+extern int trust_sender_filter;
 extern int unsort_ndx;
 extern uid_t our_uid;
 extern struct stats stats;
 extern char *filesfrom_host;
 extern char *usermap, *groupmap;
 
+extern struct name_num_item *file_sum_nni;
+
 extern char curr_dir[MAXPATHLEN];
 
 extern struct chmod_mode_struct *chmod_modes;
 
-extern filter_rule_list filter_list;
-extern filter_rule_list daemon_filter_list;
+extern filter_rule_list filter_list, implied_filter_list, daemon_filter_list;
 
 #ifdef ICONV_OPTION
 extern int filesfrom_convert;
@@ -144,7 +146,8 @@ void init_flist(void)
 		rprintf(FINFO, "FILE_STRUCT_LEN=%d, EXTRA_LEN=%d\n",
 			(int)FILE_STRUCT_LEN, (int)EXTRA_LEN);
 	}
-	flist_csum_len = csum_len_for_type(checksum_type, 1);
+	/* Note that this isn't identical to file_sum_len in the case of CSUM_MD4_ARCHAIC: */
+	flist_csum_len = csum_len_for_type(file_sum_nni->num, 1);
 
 	show_filelist_progress = INFO_GTE(FLIST, 1) && xfer_dirs && !am_server && !inc_recurse;
 }
@@ -295,6 +298,8 @@ static void flist_expand(struct file_list *flist, int extra)
 		flist->malloced = FLIST_START;
 	else if (flist->malloced >= FLIST_LINEAR)
 		flist->malloced += FLIST_LINEAR;
+	else if (flist->malloced < FLIST_START_LARGE/16)
+		flist->malloced *= 4;
 	else
 		flist->malloced *= 2;
 
@@ -305,7 +310,7 @@ static void flist_expand(struct file_list *flist, int extra)
 
 	new_ptr = realloc_array(flist->files, struct file_struct *, flist->malloced);
 
-	if (DEBUG_GTE(FLIST, 1) && flist->malloced != FLIST_START) {
+	if (DEBUG_GTE(FLIST, 1) && flist->files) {
 		rprintf(FCLIENT, "[%s] expand file_list pointer array to %s bytes, did%s move\n",
 		    who_am_i(),
 		    big_num(sizeof flist->files[0] * flist->malloced),
@@ -698,6 +703,7 @@ static struct file_struct *recv_file_entry(int f, struct file_list *flist, int x
 	int alloc_len, basename_len, linkname_len;
 	int extra_len = file_extra_cnt * EXTRA_LEN;
 	int first_hlink_ndx = -1;
+	char real_ISREG_entry;
 	int64 file_length;
 #ifdef CAN_SET_NSEC
 	uint32 modtime_nsec;
@@ -750,7 +756,7 @@ static struct file_struct *recv_file_entry(int f, struct file_list *flist, int x
 	if (*thisname
 	 && (clean_fname(thisname, CFN_REFUSE_DOT_DOT_DIRS) < 0 || (!relative_paths && *thisname == '/'))) {
 		rprintf(FERROR, "ABORTING due to unsafe pathname from sender: %s\n", thisname);
-		exit_cleanup(RERR_PROTOCOL);
+		exit_cleanup(RERR_UNSUPPORTED);
 	}
 
 	if (sanitize_paths)
@@ -812,6 +818,7 @@ static struct file_struct *recv_file_entry(int f, struct file_list *flist, int x
 				linkname_len = strlen(F_SYMLINK(first)) + 1;
 			else
 				linkname_len = 0;
+			real_ISREG_entry = S_ISREG(mode) ? 1 : 0;
 			goto create_object;
 		}
 	}
@@ -829,7 +836,7 @@ static struct file_struct *recv_file_entry(int f, struct file_list *flist, int x
 			}
 #endif
 		} else
-			modtime = read_int(f);
+			modtime = read_uint(f);
 	}
 	if (xflags & XMIT_MOD_NSEC)
 #ifndef CAN_SET_NSEC
@@ -939,10 +946,20 @@ static struct file_struct *recv_file_entry(int f, struct file_list *flist, int x
 #endif
 		linkname_len = 0;
 
+	if (copy_devices && IS_DEVICE(mode)) {
+		/* This is impossible in the official release, but some pre-release patches
+		 * didn't convert the device into a file before sending, so we'll do it here
+		 * (even though the length is typically 0 and any checksum data is zeros). */
+		mode = S_IFREG | (mode & ACCESSPERMS);
+		modtime = time(NULL); /* The mtime on the device is not up-to-date, so set it to "now". */
+		real_ISREG_entry = 0;
+	} else
+		real_ISREG_entry = S_ISREG(mode) ? 1 : 0;
+
 #ifdef SUPPORT_HARD_LINKS
   create_object:
 	if (preserve_hard_links) {
-		if (protocol_version < 28 && S_ISREG(mode))
+		if (protocol_version < 28 && real_ISREG_entry)
 			xflags |= XMIT_HLINKED;
 		if (xflags & XMIT_HLINKED)
 			extra_len += (inc_recurse+1) * EXTRA_LEN;
@@ -971,6 +988,19 @@ static struct file_struct *recv_file_entry(int f, struct file_list *flist, int x
 		exit_cleanup(RERR_UNSUPPORTED);
 	}
 
+	if (*thisname == '/' ? thisname[1] != '.' || thisname[2] != '\0' : *thisname != '.' || thisname[1] != '\0') {
+		int filt_flags = S_ISDIR(mode) ? NAME_IS_DIR : NAME_IS_FILE;
+		if (!trust_sender_filter /* a per-dir filter rule means we must trust the sender's filtering */
+		 && filter_list.head && check_server_filter(&filter_list, FINFO, thisname, filt_flags) < 0) {
+			rprintf(FERROR, "ERROR: rejecting excluded file-list name: %s\n", thisname);
+			exit_cleanup(RERR_UNSUPPORTED);
+		}
+		if (implied_filter_list.head && check_filter(&implied_filter_list, FINFO, thisname, filt_flags) <= 0) {
+			rprintf(FERROR, "ERROR: rejecting unrequested file-list name: %s\n", thisname);
+			exit_cleanup(RERR_UNSUPPORTED);
+		}
+	}
+
 	if (inc_recurse && S_ISDIR(mode)) {
 		if (one_file_system) {
 			/* Room to save the dir's device for -x */
@@ -1158,8 +1188,8 @@ static struct file_struct *recv_file_entry(int f, struct file_list *flist, int x
 	}
 #endif
 
-	if (always_checksum && (S_ISREG(mode) || protocol_version < 28)) {
-		if (S_ISREG(mode))
+	if (always_checksum && (real_ISREG_entry || protocol_version < 28)) {
+		if (real_ISREG_entry)
 			bp = F_SUM(file);
 		else {
 			/* Prior to 28, we get a useless set of nulls. */
@@ -1358,6 +1388,18 @@ struct file_struct *make_file(const char *fname, struct file_list *flist,
 	linkname_len = 0;
 #endif
 
+	if (copy_devices && am_sender && IS_DEVICE(st.st_mode)) {
+		if (st.st_size == 0) {
+			int fd = do_open(fname, O_RDONLY, 0);
+			if (fd >= 0) {
+				st.st_size = get_device_size(fd, fname);
+				close(fd);
+			}
+		}
+		st.st_mode = S_IFREG | (st.st_mode & ACCESSPERMS);
+		st.st_mtime = time(NULL); /* The mtime on the device is not up-to-date, so set it to "now". */
+	}
+
 #ifdef ST_MTIME_NSEC
 	if (st.ST_MTIME_NSEC && protocol_version >= 31)
 		extra_len += EXTRA_LEN;
@@ -1438,7 +1480,7 @@ struct file_struct *make_file(const char *fname, struct file_list *flist,
 		F_ATIME(file) = st.st_atime;
 #ifdef SUPPORT_CRTIMES
 	if (crtimes_ndx)
-		F_CRTIME(file) = get_create_time(fname);
+		F_CRTIME(file) = get_create_time(fname, &st);
 #endif
 
 	if (basename != thisname)
@@ -2186,8 +2228,10 @@ struct file_list *send_file_list(int f, int argc, char *argv[])
 #endif
 
 	flist = cur_flist = flist_new(0, "send_file_list");
+	flist_expand(flist, FLIST_START_LARGE);
 	if (inc_recurse) {
 		dir_flist = flist_new(FLIST_TEMP, "send_file_list");
+		flist_expand(dir_flist, FLIST_START_LARGE);
 		flags |= FLAG_DIVERT_DIRS;
 	} else
 		dir_flist = cur_flist;
@@ -2323,7 +2367,7 @@ struct file_list *send_file_list(int f, int argc, char *argv[])
 		}
 
 		dirlen = dir ? strlen(dir) : 0;
-		if (dirlen != lastdir_len || memcmp(lastdir, dir, dirlen) != 0) {
+		if (dirlen != lastdir_len || (dirlen && memcmp(lastdir, dir, dirlen) != 0)) {
 			if (!change_pathname(NULL, dir, -dirlen))
 				goto bad_path;
 			lastdir = pathname;
@@ -2541,10 +2585,13 @@ struct file_list *recv_file_list(int f, int dir_ndx)
 #endif
 
 	flist = flist_new(0, "recv_file_list");
+	flist_expand(flist, FLIST_START_LARGE);
 
 	if (inc_recurse) {
-		if (flist->ndx_start == 1)
+		if (flist->ndx_start == 1) {
 			dir_flist = flist_new(FLIST_TEMP, "recv_file_list");
+			flist_expand(dir_flist, FLIST_START_LARGE);
+		}
 		dstart = dir_flist->used;
 	} else {
 		dir_flist = flist;
@@ -2595,7 +2642,7 @@ struct file_list *recv_file_list(int f, int dir_ndx)
 					rprintf(FERROR,
 						"ABORTING due to invalid path from sender: %s/%s\n",
 						cur_dir, file->basename);
-					exit_cleanup(RERR_PROTOCOL);
+					exit_cleanup(RERR_UNSUPPORTED);
 				}
 				good_dirname = cur_dir;
 			}
@@ -2612,7 +2659,7 @@ struct file_list *recv_file_list(int f, int dir_ndx)
 		} else if (S_ISLNK(file->mode))
 			stats.num_symlinks++;
 		else if (IS_DEVICE(file->mode))
-			stats.num_symlinks++;
+			stats.num_devices++;
 		else
 			stats.num_specials++;
 

generator.c

@@ -4,7 +4,7 @@
  * Copyright (C) 1996-2000 Andrew Tridgell
  * Copyright (C) 1996 Paul Mackerras
  * Copyright (C) 2002 Martin Pool <mbp@samba.org>
- * Copyright (C) 2003-2020 Wayne Davison
+ * Copyright (C) 2003-2023 Wayne Davison
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -35,16 +35,18 @@ extern int inc_recurse;
 extern int relative_paths;
 extern int implied_dirs;
 extern int keep_dirlinks;
+extern int write_devices;
 extern int preserve_acls;
 extern int preserve_xattrs;
 extern int preserve_links;
 extern int preserve_devices;
-extern int write_devices;
 extern int preserve_specials;
 extern int preserve_hard_links;
 extern int preserve_executability;
 extern int preserve_perms;
-extern int preserve_times;
+extern int preserve_mtimes;
+extern int omit_dir_times;
+extern int omit_link_times;
 extern int delete_mode;
 extern int delete_before;
 extern int delete_during;
@@ -112,10 +114,6 @@ static int need_retouch_dir_times;
 static int need_retouch_dir_perms;
 static const char *solo_file = NULL;
 
-enum nonregtype {
-	TYPE_DIR, TYPE_SPECIAL, TYPE_DEVICE, TYPE_SYMLINK
-};
-
 /* Forward declarations. */
 #ifdef SUPPORT_HARD_LINKS
 static void handle_skipped_hlink(struct file_struct *file, int itemizing,
@@ -186,7 +184,8 @@ static int remember_delete(struct file_struct *file, const char *fname, int flag
 static int read_delay_line(char *buf, int *flags_p)
 {
 	static int read_pos = 0;
-	int j, len, mode;
+	unsigned int mode;
+	int j, len;
 	char *bp, *past_space;
 
 	while (1) {
@@ -270,7 +269,7 @@ static void do_delayed_deletions(char *delbuf)
  * MAXPATHLEN buffer with the name of the directory in it (the functions we
  * call will append names onto the end, but the old dir value will be restored
  * on exit). */
-static void delete_in_dir(char *fbuf, struct file_struct *file, dev_t *fs_dev)
+static void delete_in_dir(char *fbuf, struct file_struct *file, dev_t fs_dev)
 {
 	static int already_warned = 0;
 	static struct hashtable *dev_tbl;
@@ -305,12 +304,12 @@ static void delete_in_dir(char *fbuf, struct file_struct *file, dev_t *fs_dev)
 		if (!dev_tbl)
 			dev_tbl = hashtable_create(16, HT_KEY64);
 		if (file->flags & FLAG_TOP_DIR) {
-			hashtable_find(dev_tbl, *fs_dev+1, "");
-			filesystem_dev = *fs_dev;
-		} else if (filesystem_dev != *fs_dev) {
-			if (!hashtable_find(dev_tbl, *fs_dev+1, NULL))
+			hashtable_find(dev_tbl, fs_dev+1, "");
+			filesystem_dev = fs_dev;
+		} else if (filesystem_dev != fs_dev) {
+			if (!hashtable_find(dev_tbl, fs_dev+1, NULL))
 				return;
-			filesystem_dev = *fs_dev; /* it's a prior top-dir dev */
+			filesystem_dev = fs_dev; /* it's a prior top-dir dev */
 		}
 	}
 
@@ -379,9 +378,9 @@ static void do_delete_pass(void)
 		 || !S_ISDIR(st.st_mode))
 			continue;
 
-		delete_in_dir(fbuf, file, &st.st_dev);
+		delete_in_dir(fbuf, file, st.st_dev);
 	}
-	delete_in_dir(NULL, NULL, &dev_zero);
+	delete_in_dir(NULL, NULL, dev_zero);
 
 	if (INFO_GTE(FLIST, 2) && !am_server)
 		rprintf(FINFO, "                    \r");
@@ -402,7 +401,7 @@ static inline int any_time_differs(stat_x *sxp, struct file_struct *file, UNUSED
 #ifdef SUPPORT_CRTIMES
 	if (!differs && crtimes_ndx) {
 		if (sxp->crtime == 0)
-			sxp->crtime = get_create_time(fname);
+			sxp->crtime = get_create_time(fname, &sxp->st);
 		differs = !same_time(sxp->crtime, 0, F_CRTIME(file), 0);
 	}
 #endif
@@ -463,7 +462,7 @@ int unchanged_attrs(const char *fname, struct file_struct *file, stat_x *sxp)
 {
 	if (S_ISLNK(file->mode)) {
 #ifdef CAN_SET_SYMLINK_TIMES
-		if (preserve_times & PRESERVE_LINK_TIMES && any_time_differs(sxp, file, fname))
+		if (preserve_mtimes && !omit_link_times && any_time_differs(sxp, file, fname))
 			return 0;
 #endif
 #ifdef CAN_CHMOD_SYMLINK
@@ -483,7 +482,7 @@ int unchanged_attrs(const char *fname, struct file_struct *file, stat_x *sxp)
 			return 0;
 #endif
 	} else {
-		if (preserve_times && any_time_differs(sxp, file, fname))
+		if (preserve_mtimes && any_time_differs(sxp, file, fname))
 			return 0;
 		if (perms_differ(file, sxp))
 			return 0;
@@ -507,9 +506,9 @@ void itemize(const char *fnamecmp, struct file_struct *file, int ndx, int statre
 	     const char *xname)
 {
 	if (statret >= 0) { /* A from-dest-dir statret can == 1! */
-		int keep_time = !preserve_times ? 0
-		    : S_ISDIR(file->mode) ? preserve_times & PRESERVE_DIR_TIMES
-		    : S_ISLNK(file->mode) ? preserve_times & PRESERVE_LINK_TIMES
+		int keep_time = !preserve_mtimes ? 0
+		    : S_ISDIR(file->mode) ? !omit_dir_times
+		    : S_ISLNK(file->mode) ? !omit_link_times
 		    : 1;
 
 		if (S_ISREG(file->mode) && F_LENGTH(file) != sxp->st.st_size)
@@ -528,12 +527,12 @@ void itemize(const char *fnamecmp, struct file_struct *file, int ndx, int statre
 #ifdef SUPPORT_CRTIMES
 		if (crtimes_ndx) {
 			if (sxp->crtime == 0)
-				sxp->crtime = get_create_time(fnamecmp);
+				sxp->crtime = get_create_time(fnamecmp, &sxp->st);
 			if (!same_time(sxp->crtime, 0, F_CRTIME(file), 0))
 				iflags |= ITEM_REPORT_CRTIME;
 		}
 #endif
-#if !defined HAVE_LCHMOD && !defined HAVE_SETATTRLIST
+#ifndef CAN_CHMOD_SYMLINK
 		if (S_ISLNK(file->mode)) {
 			;
 		} else
@@ -599,30 +598,77 @@ void itemize(const char *fnamecmp, struct file_struct *file, int ndx, int statre
 	}
 }
 
-
-/* Perform our quick-check heuristic for determining if a file is unchanged. */
-int unchanged_file(char *fn, struct file_struct *file, STRUCT_STAT *st)
+static enum filetype get_file_type(mode_t mode)
 {
-	if (st->st_size != F_LENGTH(file))
-		return 0;
-
-	/* if always checksum is set then we use the checksum instead
-	   of the file time to determine whether to sync */
-	if (always_checksum > 0 && S_ISREG(st->st_mode)) {
-		char sum[MAX_DIGEST_LEN];
-		file_checksum(fn, st, sum);
-		return memcmp(sum, F_SUM(file), flist_csum_len) == 0;
-	}
-
-	if (size_only > 0)
-		return 1;
-
-	if (ignore_times)
-		return 0;
-
-	return !mtime_differs(st, file);
+	if (S_ISREG(mode))
+		return FT_REG;
+	if (S_ISLNK(mode))
+		return FT_SYMLINK;
+	if (S_ISDIR(mode))
+		return FT_DIR;
+	if (IS_SPECIAL(mode))
+		return FT_SPECIAL;
+	if (IS_DEVICE(mode))
+		return FT_DEVICE;
+	return FT_UNSUPPORTED;
 }
 
+/* Perform our quick-check heuristic for determining if a file is unchanged. */
+int quick_check_ok(enum filetype ftype, const char *fn, struct file_struct *file, STRUCT_STAT *st)
+{
+	switch (ftype) {
+	  case FT_REG:
+		if (st->st_size != F_LENGTH(file))
+			return 0;
+
+		/* If always_checksum is set then we use the checksum instead
+		 * of the file mtime to determine whether to sync. */
+		if (always_checksum > 0) {
+			char sum[MAX_DIGEST_LEN];
+			file_checksum(fn, st, sum);
+			return memcmp(sum, F_SUM(file), flist_csum_len) == 0;
+		}
+
+		if (size_only > 0)
+			return 1;
+
+		if (ignore_times)
+			return 0;
+
+		if (mtime_differs(st, file))
+			return 0;
+		break;
+	  case FT_DIR:
+		break;
+	  case FT_SYMLINK: {
+#ifdef SUPPORT_LINKS
+		char lnk[MAXPATHLEN];
+		int len = do_readlink(fn, lnk, MAXPATHLEN-1);
+		if (len <= 0)
+			return 0;
+		lnk[len] = '\0';
+		if (strcmp(lnk, F_SYMLINK(file)) != 0)
+			return 0;
+		break;
+#else
+		return -1;
+#endif
+	  }
+	  case FT_SPECIAL:
+		if (!BITS_EQUAL(file->mode, st->st_mode, _S_IFMT))
+			return 0;
+		break;
+	  case FT_DEVICE: {
+		uint32 *devp = F_RDEV_P(file);
+		if (st->st_rdev != MAKEDEV(DEV_MAJOR(devp), DEV_MINOR(devp)))
+			return 0;
+		break;
+	  }
+	  case FT_UNSUPPORTED:
+		return -1;
+	}
+	return 1;
+}
 
 /*
  * set (initialize) the size entries in the per-file sum_struct
@@ -737,7 +783,7 @@ static int generate_and_send_sums(int fd, OFF_T len, int f_out, int f_copy)
 	for (i = 0; i < sum.count; i++) {
 		int32 n1 = (int32)MIN(len, (OFF_T)sum.blength);
 		char *map = map_ptr(mapbuf, offset, n1);
-		char sum2[SUM_LENGTH];
+		char sum2[MAX_DIGEST_LEN];
 		uint32 sum1;
 
 		len -= n1;
@@ -829,9 +875,12 @@ static struct file_struct *find_fuzzy(struct file_struct *file, struct file_list
 			len = strlen(name);
 			suf = find_filename_suffix(name, len, &suf_len);
 
-			dist = fuzzy_distance(name, len, fname, fname_len);
-			/* Add some extra weight to how well the suffixes match. */
-			dist += fuzzy_distance(suf, suf_len, fname_suf, fname_suf_len) * 10;
+			dist = fuzzy_distance(name, len, fname, fname_len, lowest_dist);
+			/* Add some extra weight to how well the suffixes match unless we've already disqualified
+			 * this file based on a heuristic. */
+			if (dist < 0xFFFF0000U) {
+				dist += fuzzy_distance(suf, suf_len, fname_suf, fname_suf_len, 0xFFFF0000U) * 10;
+			}
 			if (DEBUG_GTE(FUZZY, 2)) {
 				rprintf(FINFO, "fuzzy distance for %s = %d.%05d\n",
 					f_name(fp, NULL), (int)(dist>>16), (int)(dist&0xFFFF));
@@ -907,7 +956,7 @@ static int try_dests_reg(struct file_struct *file, char *fname, int ndx,
 			best_match = j;
 			match_level = 1;
 		}
-		if (!unchanged_file(cmpbuf, file, &sxp->st))
+		if (!quick_check_ok(FT_REG, cmpbuf, file, &sxp->st))
 			continue;
 		if (match_level == 1) {
 			best_match = j;
@@ -1006,29 +1055,14 @@ static int try_dests_non(struct file_struct *file, char *fname, int ndx,
 {
 	int best_match = -1;
 	int match_level = 0;
-	enum nonregtype type;
-	uint32 *devp;
-#ifdef SUPPORT_LINKS
-	char lnk[MAXPATHLEN];
-	int len;
-#endif
+	enum filetype ftype = get_file_type(file->mode);
 	int j = 0;
 
 #ifndef SUPPORT_LINKS
-	if (S_ISLNK(file->mode))
+	if (ftype == FT_SYMLINK)
 		return -1;
 #endif
-	if (S_ISDIR(file->mode)) {
-		type = TYPE_DIR;
-	} else if (IS_SPECIAL(file->mode))
-		type = TYPE_SPECIAL;
-	else if (IS_DEVICE(file->mode))
-		type = TYPE_DEVICE;
-#ifdef SUPPORT_LINKS
-	else if (S_ISLNK(file->mode))
-		type = TYPE_SYMLINK;
-#endif
-	else {
+	if (ftype == FT_REG || ftype == FT_UNSUPPORTED) {
 		rprintf(FERROR,
 			"internal: try_dests_non() called with invalid mode (%o)\n",
 			(int)file->mode);
@@ -1039,53 +1073,14 @@ static int try_dests_non(struct file_struct *file, char *fname, int ndx,
 		pathjoin(cmpbuf, MAXPATHLEN, basis_dir[j], fname);
 		if (link_stat(cmpbuf, &sxp->st, 0) < 0)
 			continue;
-		switch (type) {
-		case TYPE_DIR:
-			if (!S_ISDIR(sxp->st.st_mode))
-				continue;
-			break;
-		case TYPE_SPECIAL:
-			if (!IS_SPECIAL(sxp->st.st_mode))
-				continue;
-			break;
-		case TYPE_DEVICE:
-			if (!IS_DEVICE(sxp->st.st_mode))
-				continue;
-			break;
-		case TYPE_SYMLINK:
-#ifdef SUPPORT_LINKS
-			if (!S_ISLNK(sxp->st.st_mode))
-				continue;
-			break;
-#else
-			return -1;
-#endif
-		}
+		if (ftype != get_file_type(sxp->st.st_mode))
+			continue;
 		if (match_level < 1) {
 			match_level = 1;
 			best_match = j;
 		}
-		switch (type) {
-		case TYPE_DIR:
-		case TYPE_SPECIAL:
-			break;
-		case TYPE_DEVICE:
-			devp = F_RDEV_P(file);
-			if (sxp->st.st_rdev != MAKEDEV(DEV_MAJOR(devp), DEV_MINOR(devp)))
-				continue;
-			break;
-		case TYPE_SYMLINK:
-#ifdef SUPPORT_LINKS
-			if ((len = do_readlink(cmpbuf, lnk, MAXPATHLEN-1)) <= 0)
-				continue;
-			lnk[len] = '\0';
-			if (strcmp(lnk, F_SYMLINK(file)) != 0)
-				continue;
-			break;
-#else
-			return -1;
-#endif
-		}
+		if (!quick_check_ok(ftype, cmpbuf, file, &sxp->st))
+			continue;
 		if (match_level < 2) {
 			match_level = 2;
 			best_match = j;
@@ -1130,14 +1125,14 @@ static int try_dests_non(struct file_struct *file, char *fname, int ndx,
 			match_level = 2;
 		if (itemizing && stdout_format_has_i
 		 && (INFO_GTE(NAME, 2) || stdout_format_has_i > 1)) {
-			int chg = alt_dest_type == COMPARE_DEST && type != TYPE_DIR ? 0
+			int chg = alt_dest_type == COMPARE_DEST && ftype != FT_DIR ? 0
 			    : ITEM_LOCAL_CHANGE + (match_level == 3 ? ITEM_XNAME_FOLLOWS : 0);
 			char *lp = match_level == 3 ? "" : NULL;
 			itemize(cmpbuf, file, ndx, 0, sxp, chg + ITEM_MATCHED, 0, lp);
 		}
 		if (INFO_GTE(NAME, 2) && maybe_ATTRS_REPORT) {
 			rprintf(FCLIENT, "%s%s is uptodate\n",
-				fname, type == TYPE_DIR ? "/" : "");
+				fname, ftype == FT_DIR ? "/" : "");
 		}
 		return -2;
 	}
@@ -1231,7 +1226,8 @@ static void recv_generator(char *fname, struct file_struct *file, int ndx,
 	char fnamecmpbuf[MAXPATHLEN];
 	uchar fnamecmp_type;
 	int del_opts = delete_mode || force_delete ? DEL_RECURSE : 0;
-	int is_dir = !S_ISDIR(file->mode) ? 0
+	enum filetype stype, ftype = get_file_type(file->mode);
+	int is_dir = ftype != FT_DIR ? 0
 		   : inc_recurse && ndx != cur_flist->ndx_start - 1 ? -1
 		   : 1;
 
@@ -1278,7 +1274,6 @@ static void recv_generator(char *fname, struct file_struct *file, int ndx,
 			return;
 		}
 	}
-	sx.crtime = 0;
 
 	if (dry_run > 1 || (dry_missing_dir && is_below(file, dry_missing_dir))) {
 		int i;
@@ -1380,10 +1375,27 @@ static void recv_generator(char *fname, struct file_struct *file, int ndx,
 	 && !am_root && sx.st.st_uid == our_uid)
 		del_opts |= DEL_NO_UID_WRITE;
 
+	if (statret == 0)
+		stype = get_file_type(sx.st.st_mode);
+	else
+		stype = FT_UNSUPPORTED;
+
 	if (ignore_existing > 0 && statret == 0
-	 && (!is_dir || !S_ISDIR(sx.st.st_mode))) {
-		if (INFO_GTE(SKIP, 1) && is_dir >= 0)
-			rprintf(FINFO, "%s exists\n", fname);
+	 && (!is_dir || stype != FT_DIR)) {
+		if (INFO_GTE(SKIP, 1) && is_dir >= 0) {
+			const char *suf = "";
+			if (INFO_GTE(SKIP, 2)) {
+				if (ftype != stype)
+					suf = " (type change)";
+				else if (!quick_check_ok(ftype, fname, file, &sx.st))
+					suf = always_checksum ? " (sum change)" : " (file change)";
+				else if (!unchanged_attrs(fname, file, &sx))
+					suf = " (attr change)";
+				else
+					suf = " (uptodate)";
+			}
+			rprintf(FINFO, "%s exists%s\n", fname, suf);
+		}
 #ifdef SUPPORT_HARD_LINKS
 		if (F_IS_HLINKED(file))
 			handle_skipped_hlink(file, itemizing, code, f_out);
@@ -1404,7 +1416,7 @@ static void recv_generator(char *fname, struct file_struct *file, int ndx,
 		} else
 			added_perms = 0;
 		if (is_dir < 0) {
-			if (!(preserve_times & PRESERVE_DIR_TIMES))
+			if (!preserve_mtimes || omit_dir_times)
 				goto cleanup;
 			/* In inc_recurse mode we want to make sure any missing
 			 * directories get created while we're still processing
@@ -1412,7 +1424,7 @@ static void recv_generator(char *fname, struct file_struct *file, int ndx,
 			 * dir's mtime right away).  We will handle the dir in
 			 * full later (right before we handle its contents). */
 			if (statret == 0
-			 && (S_ISDIR(sx.st.st_mode)
+			 && (stype == FT_DIR
 			  || delete_item(fname, sx.st.st_mode, del_opts | DEL_FOR_DIR) != 0))
 				goto cleanup; /* Any errors get reported later. */
 			if (do_mkdir(fname, (file->mode|added_perms) & 0700) == 0)
@@ -1424,7 +1436,7 @@ static void recv_generator(char *fname, struct file_struct *file, int ndx,
 		 * file of that name and it is *not* a directory, then
 		 * we need to delete it.  If it doesn't exist, then
 		 * (perhaps recursively) create it. */
-		if (statret == 0 && !S_ISDIR(sx.st.st_mode)) {
+		if (statret == 0 && stype != FT_DIR) {
 			if (delete_item(fname, sx.st.st_mode, del_opts | DEL_FOR_DIR) != 0)
 				goto skipping_dir_contents;
 			statret = -1;
@@ -1508,7 +1520,7 @@ static void recv_generator(char *fname, struct file_struct *file, int ndx,
 		else if (delete_during && f_out != -1 && !phase
 		    && !(file->flags & FLAG_MISSING_DIR)) {
 			if (file->flags & FLAG_CONTENT_DIR)
-				delete_in_dir(fname, file, &real_sx.st.st_dev);
+				delete_in_dir(fname, file, real_sx.st.st_dev);
 			else
 				change_local_filter_dir(fname, strlen(fname), F_DEPTH(file));
 		}
@@ -1519,7 +1531,7 @@ static void recv_generator(char *fname, struct file_struct *file, int ndx,
 	/* If we're not preserving permissions, change the file-list's
 	 * mode based on the local permissions and some heuristics. */
 	if (!preserve_perms) {
-		int exists = statret == 0 && !S_ISDIR(sx.st.st_mode);
+		int exists = statret == 0 && stype != FT_DIR;
 		file->mode = dest_mode(file->mode, sx.st.st_mode, dflt_perms, exists);
 	}
 
@@ -1529,7 +1541,7 @@ static void recv_generator(char *fname, struct file_struct *file, int ndx,
 		goto cleanup;
 #endif
 
-	if (preserve_links && S_ISLNK(file->mode)) {
+	if (preserve_links && ftype == FT_SYMLINK) {
 #ifdef SUPPORT_LINKS
 		const char *sl = F_SYMLINK(file);
 		if (safe_symlinks && unsafe_symlink(sl, fname)) {
@@ -1546,12 +1558,7 @@ static void recv_generator(char *fname, struct file_struct *file, int ndx,
 			goto cleanup;
 		}
 		if (statret == 0) {
-			char lnk[MAXPATHLEN];
-			int len;
-
-			if (S_ISLNK(sx.st.st_mode)
-			 && (len = do_readlink(fname, lnk, MAXPATHLEN-1)) > 0
-			 && strncmp(lnk, sl, len) == 0 && sl[len] == '\0') {
+			if (stype == FT_SYMLINK && quick_check_ok(stype, fname, file, &sx.st)) {
 				/* The link is pointing to the right place. */
 				set_file_attrs(fname, file, &sx, NULL, maybe_ATTRS_REPORT);
 				if (itemizing)
@@ -1584,7 +1591,7 @@ static void recv_generator(char *fname, struct file_struct *file, int ndx,
 		if (atomic_create(file, fname, sl, NULL, MAKEDEV(0, 0), &sx, statret == 0 ? DEL_FOR_SYMLINK : 0)) {
 			set_file_attrs(fname, file, NULL, NULL, 0);
 			if (itemizing) {
-				if (statret == 0 && !S_ISLNK(sx.st.st_mode))
+				if (statret == 0 && stype != FT_SYMLINK)
 					statret = -1;
 				itemize(fnamecmp, file, ndx, statret, &sx,
 					ITEM_LOCAL_CHANGE|ITEM_REPORT_CHANGE, 0, NULL);
@@ -1605,28 +1612,22 @@ static void recv_generator(char *fname, struct file_struct *file, int ndx,
 		goto cleanup;
 	}
 
-	if ((am_root && preserve_devices && IS_DEVICE(file->mode))
-	 || (preserve_specials && IS_SPECIAL(file->mode))) {
+	if ((am_root && preserve_devices && ftype == FT_DEVICE)
+	 || (preserve_specials && ftype == FT_SPECIAL)) {
 		dev_t rdev;
-		int del_for_flag = 0;
-		if (IS_DEVICE(file->mode)) {
+		int del_for_flag;
+		if (ftype == FT_DEVICE) {
 			uint32 *devp = F_RDEV_P(file);
 			rdev = MAKEDEV(DEV_MAJOR(devp), DEV_MINOR(devp));
-		} else
+			del_for_flag = DEL_FOR_DEVICE;
+		} else {
 			rdev = 0;
+			del_for_flag = DEL_FOR_SPECIAL;
+		}
 		if (statret == 0) {
-			if (IS_DEVICE(file->mode)) {
-				if (!IS_DEVICE(sx.st.st_mode))
-					statret = -1;
-				del_for_flag = DEL_FOR_DEVICE;
-			} else {
-				if (!IS_SPECIAL(sx.st.st_mode))
-					statret = -1;
-				del_for_flag = DEL_FOR_SPECIAL;
-			}
-			if (statret == 0
-			 && BITS_EQUAL(sx.st.st_mode, file->mode, _S_IFMT)
-			 && (IS_SPECIAL(sx.st.st_mode) || sx.st.st_rdev == rdev)) {
+			if (ftype != stype)
+				statret = -1;
+			else if (quick_check_ok(ftype, fname, file, &sx.st)) {
 				/* The device or special file is identical. */
 				set_file_attrs(fname, file, &sx, NULL, maybe_ATTRS_REPORT);
 				if (itemizing)
@@ -1679,10 +1680,12 @@ static void recv_generator(char *fname, struct file_struct *file, int ndx,
 		goto cleanup;
 	}
 
-	if (!S_ISREG(file->mode)) {
-		if (solo_file)
-			fname = f_name(file, NULL);
-		rprintf(FINFO, "skipping non-regular file \"%s\"\n", fname);
+	if (ftype != FT_REG) {
+		if (INFO_GTE(NONREG, 1)) {
+			if (solo_file)
+				fname = f_name(file, NULL);
+			rprintf(FINFO, "skipping non-regular file \"%s\"\n", fname);
+		}
 		goto cleanup;
 	}
 
@@ -1703,7 +1706,7 @@ static void recv_generator(char *fname, struct file_struct *file, int ndx,
 		goto cleanup;
 	}
 
-	if (update_only > 0 && statret == 0 && file->modtime - sx.st.st_mtime <= modify_window) {
+	if (update_only > 0 && statret == 0 && file->modtime - sx.st.st_mtime < modify_window) {
 		if (INFO_GTE(SKIP, 1))
 			rprintf(FINFO, "%s is newer\n", fname);
 #ifdef SUPPORT_HARD_LINKS
@@ -1715,7 +1718,7 @@ static void recv_generator(char *fname, struct file_struct *file, int ndx,
 
 	fnamecmp_type = FNAMECMP_FNAME;
 
-	if (statret == 0 && !(S_ISREG(sx.st.st_mode) || (write_devices && IS_DEVICE(sx.st.st_mode)))) {
+	if (statret == 0 && !(stype == FT_REG || (write_devices && stype == FT_DEVICE))) {
 		if (delete_item(fname, sx.st.st_mode, del_opts | DEL_FOR_FILE) != 0)
 			goto cleanup;
 		statret = -1;
@@ -1749,7 +1752,7 @@ static void recv_generator(char *fname, struct file_struct *file, int ndx,
 		partialptr = NULL;
 
 	if (statret != 0 && fuzzy_basis) {
-		if (need_fuzzy_dirlist && S_ISREG(file->mode)) {
+		if (need_fuzzy_dirlist) {
 			const char *dn = file->dirname ? file->dirname : ".";
 			int i;
 			strlcpy(fnamecmpbuf, dn, sizeof fnamecmpbuf);
@@ -1793,11 +1796,17 @@ static void recv_generator(char *fname, struct file_struct *file, int ndx,
 		goto cleanup;
 	}
 
+	if (write_devices && IS_DEVICE(sx.st.st_mode) && sx.st.st_size == 0) {
+		/* This early open into fd skips the regular open below. */
+		if ((fd = do_open(fnamecmp, O_RDONLY, 0)) >= 0)
+			real_sx.st.st_size = sx.st.st_size = get_device_size(fd, fnamecmp);
+	}
+
 	if (fnamecmp_type <= FNAMECMP_BASIS_DIR_HIGH)
 		;
 	else if (fnamecmp_type >= FNAMECMP_FUZZY)
 		;
-	else if (unchanged_file(fnamecmp, file, &sx.st)) {
+	else if (quick_check_ok(FT_REG, fnamecmp, file, &sx.st)) {
 		if (partialptr) {
 			do_unlink(partialptr);
 			handle_partial_dir(partialptr, PDIR_DELETE);
@@ -1813,7 +1822,7 @@ static void recv_generator(char *fname, struct file_struct *file, int ndx,
 			goto cleanup;
 	  return_with_success:
 		if (!dry_run)
-			send_msg_int(MSG_SUCCESS, ndx);
+			send_msg_success(fname, ndx);
 		goto cleanup;
 	}
 
@@ -1858,7 +1867,7 @@ static void recv_generator(char *fname, struct file_struct *file, int ndx,
 	}
 
 	/* open the file */
-	if ((fd = do_open(fnamecmp, O_RDONLY, 0)) < 0) {
+	if (fd < 0 && (fd = do_open(fnamecmp, O_RDONLY, 0)) < 0) {
 		rsyserr(FERROR, errno, "failed to open %s, continuing",
 			full_fname(fnamecmp));
 	  pretend_missing:
@@ -1875,11 +1884,9 @@ static void recv_generator(char *fname, struct file_struct *file, int ndx,
 
 	if (inplace && make_backups > 0 && fnamecmp_type == FNAMECMP_FNAME) {
 		if (!(backupptr = get_backup_name(fname))) {
-			close(fd);
 			goto cleanup;
 		}
 		if (!(back_file = make_file(fname, NULL, NULL, 0, NO_FILTERS))) {
-			close(fd);
 			goto pretend_missing;
 		}
 		if (robust_unlink(backupptr) && errno != ENOENT) {
@@ -1887,14 +1894,12 @@ static void recv_generator(char *fname, struct file_struct *file, int ndx,
 				full_fname(backupptr));
 			unmake_file(back_file);
 			back_file = NULL;
-			close(fd);
 			goto cleanup;
 		}
 		if ((f_copy = do_open(backupptr, O_WRONLY | O_CREAT | O_TRUNC | O_EXCL, 0600)) < 0) {
 			rsyserr(FERROR_XFER, errno, "open %s", full_fname(backupptr));
 			unmake_file(back_file);
 			back_file = NULL;
-			close(fd);
 			goto cleanup;
 		}
 		fnamecmp_type = FNAMECMP_BACKUP;
@@ -1945,7 +1950,6 @@ static void recv_generator(char *fname, struct file_struct *file, int ndx,
 		write_sum_head(f_out, NULL);
 	else if (sx.st.st_size <= 0) {
 		write_sum_head(f_out, NULL);
-		close(fd);
 	} else {
 		if (generate_and_send_sums(fd, sx.st.st_size, f_out, f_copy) < 0) {
 			rprintf(FWARNING,
@@ -1953,10 +1957,11 @@ static void recv_generator(char *fname, struct file_struct *file, int ndx,
 				fnamecmp);
 			write_sum_head(f_out, NULL);
 		}
-		close(fd);
 	}
 
   cleanup:
+	if (fd >= 0)
+		close(fd);
 	if (back_file) {
 		int save_preserve_xattrs = preserve_xattrs;
 		if (f_copy >= 0)
@@ -2242,7 +2247,7 @@ void generate_files(int f_out, const char *local_name)
 	}
 	solo_file = local_name;
 	dir_tweaking = !(list_only || solo_file || dry_run);
-	need_retouch_dir_times = preserve_times & PRESERVE_DIR_TIMES;
+	need_retouch_dir_times = preserve_mtimes && !omit_dir_times;
 	loopchk_limit = allowed_lull ? allowed_lull * 5 : 200;
 	symlink_timeset_failed_flags = ITEM_REPORT_TIME
 	    | (protocol_version >= 30 || !am_server ? ITEM_REPORT_TIMEFAIL : 0);
@@ -2295,7 +2300,7 @@ void generate_files(int f_out, const char *local_name)
 						dirdev = MAKEDEV(DEV_MAJOR(devp), DEV_MINOR(devp));
 					} else
 						dirdev = MAKEDEV(0, 0);
-					delete_in_dir(fbuf, fp, &dirdev);
+					delete_in_dir(fbuf, fp, dirdev);
 				} else
 					change_local_filter_dir(fbuf, strlen(fbuf), F_DEPTH(fp));
 			}
@@ -2342,7 +2347,7 @@ void generate_files(int f_out, const char *local_name)
 	} while ((cur_flist = cur_flist->next) != NULL);
 
 	if (delete_during)
-		delete_in_dir(NULL, NULL, &dev_zero);
+		delete_in_dir(NULL, NULL, dev_zero);
 	phase++;
 	if (DEBUG_GTE(GENR, 1))
 		rprintf(FINFO, "generate_files phase=%d\n", phase);

hashtable.c

@@ -1,7 +1,7 @@
 /*
  * Routines to provide a memory-efficient hashtable.
  *
- * Copyright (C) 2007-2020 Wayne Davison
+ * Copyright (C) 2007-2022 Wayne Davison
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -350,6 +350,9 @@ void *hashtable_find(struct hashtable *tbl, int64 key, void *data_when_new)
  -------------------------------------------------------------------------------
 */
 
+#define NON_ZERO_32(x) ((x) ? (x) : (uint32_t)1)
+#define NON_ZERO_64(x, y) ((x) || (y) ? (y) | (int64)(x) << 32 | (y) : (int64)1)
+
 uint32_t hashlittle(const void *key, size_t length)
 {
   uint32_t a,b,c;                                          /* internal state */
@@ -390,7 +393,7 @@ uint32_t hashlittle(const void *key, size_t length)
     case 3 : a+=((uint32_t)k8[2])<<16;   /* fall through */
     case 2 : a+=((uint32_t)k8[1])<<8;    /* fall through */
     case 1 : a+=k8[0]; break;
-    case 0 : return c;
+    case 0 : return NON_ZERO_32(c);
     }
   } else if (HASH_LITTLE_ENDIAN && ((u.i & 0x1) == 0)) {
     const uint16_t *k = (const uint16_t *)key;         /* read 16-bit chunks */
@@ -436,7 +439,7 @@ uint32_t hashlittle(const void *key, size_t length)
              break;
     case 1 : a+=k8[0];
              break;
-    case 0 : return c;                     /* zero length requires no mixing */
+    case 0 : return NON_ZERO_32(c);         /* zero length requires no mixing */
     }
 
   } else {                        /* need to read the key one byte at a time */
@@ -489,10 +492,171 @@ uint32_t hashlittle(const void *key, size_t length)
 	     /* FALLTHROUGH */
     case 1 : a+=k[0];
              break;
-    case 0 : return c;
+    case 0 : return NON_ZERO_32(c);
     }
   }
 
   final(a,b,c);
-  return c;
+  return NON_ZERO_32(c);
 }
+
+#if SIZEOF_INT64 >= 8
+/*
+ * hashlittle2: return 2 32-bit hash values joined into an int64.
+ *
+ * This is identical to hashlittle(), except it returns two 32-bit hash
+ * values instead of just one.  This is good enough for hash table
+ * lookup with 2^^64 buckets, or if you want a second hash if you're not
+ * happy with the first, or if you want a probably-unique 64-bit ID for
+ * the key.  *pc is better mixed than *pb, so use *pc first.  If you want
+ * a 64-bit value do something like "*pc + (((uint64_t)*pb)<<32)".
+ */
+int64 hashlittle2(const void *key, size_t length)
+{
+  uint32_t a,b,c;                                          /* internal state */
+  union { const void *ptr; size_t i; } u;     /* needed for Mac Powerbook G4 */
+
+  /* Set up the internal state */
+  a = b = c = 0xdeadbeef + ((uint32_t)length);
+
+  u.ptr = key;
+  if (HASH_LITTLE_ENDIAN && ((u.i & 0x3) == 0)) {
+    const uint32_t *k = (const uint32_t *)key;         /* read 32-bit chunks */
+    const uint8_t  *k8;
+
+    /*------ all but last block: aligned reads and affect 32 bits of (a,b,c) */
+    while (length > 12)
+    {
+      a += k[0];
+      b += k[1];
+      c += k[2];
+      mix(a,b,c);
+      length -= 12;
+      k += 3;
+    }
+
+    /*----------------------------- handle the last (probably partial) block */
+    k8 = (const uint8_t *)k;
+    switch(length)
+    {
+    case 12: c+=k[2]; b+=k[1]; a+=k[0]; break;
+    case 11: c+=((uint32_t)k8[10])<<16;  /* fall through */
+    case 10: c+=((uint32_t)k8[9])<<8;    /* fall through */
+    case 9 : c+=k8[8];                   /* fall through */
+    case 8 : b+=k[1]; a+=k[0]; break;
+    case 7 : b+=((uint32_t)k8[6])<<16;   /* fall through */
+    case 6 : b+=((uint32_t)k8[5])<<8;    /* fall through */
+    case 5 : b+=k8[4];                   /* fall through */
+    case 4 : a+=k[0]; break;
+    case 3 : a+=((uint32_t)k8[2])<<16;   /* fall through */
+    case 2 : a+=((uint32_t)k8[1])<<8;    /* fall through */
+    case 1 : a+=k8[0]; break;
+    case 0 : return NON_ZERO_64(b, c);
+    }
+  } else if (HASH_LITTLE_ENDIAN && ((u.i & 0x1) == 0)) {
+    const uint16_t *k = (const uint16_t *)key;         /* read 16-bit chunks */
+    const uint8_t  *k8;
+
+    /*--------------- all but last block: aligned reads and different mixing */
+    while (length > 12)
+    {
+      a += k[0] + (((uint32_t)k[1])<<16);
+      b += k[2] + (((uint32_t)k[3])<<16);
+      c += k[4] + (((uint32_t)k[5])<<16);
+      mix(a,b,c);
+      length -= 12;
+      k += 6;
+    }
+
+    /*----------------------------- handle the last (probably partial) block */
+    k8 = (const uint8_t *)k;
+    switch(length)
+    {
+    case 12: c+=k[4]+(((uint32_t)k[5])<<16);
+             b+=k[2]+(((uint32_t)k[3])<<16);
+             a+=k[0]+(((uint32_t)k[1])<<16);
+             break;
+    case 11: c+=((uint32_t)k8[10])<<16;     /* fall through */
+    case 10: c+=k[4];
+             b+=k[2]+(((uint32_t)k[3])<<16);
+             a+=k[0]+(((uint32_t)k[1])<<16);
+             break;
+    case 9 : c+=k8[8];                      /* fall through */
+    case 8 : b+=k[2]+(((uint32_t)k[3])<<16);
+             a+=k[0]+(((uint32_t)k[1])<<16);
+             break;
+    case 7 : b+=((uint32_t)k8[6])<<16;      /* fall through */
+    case 6 : b+=k[2];
+             a+=k[0]+(((uint32_t)k[1])<<16);
+             break;
+    case 5 : b+=k8[4];                      /* fall through */
+    case 4 : a+=k[0]+(((uint32_t)k[1])<<16);
+             break;
+    case 3 : a+=((uint32_t)k8[2])<<16;      /* fall through */
+    case 2 : a+=k[0];
+             break;
+    case 1 : a+=k8[0];
+             break;
+    case 0 : return NON_ZERO_64(b, c);  /* zero length strings require no mixing */
+    }
+
+  } else {                        /* need to read the key one byte at a time */
+    const uint8_t *k = (const uint8_t *)key;
+
+    /*--------------- all but the last block: affect some 32 bits of (a,b,c) */
+    while (length > 12)
+    {
+      a += k[0];
+      a += ((uint32_t)k[1])<<8;
+      a += ((uint32_t)k[2])<<16;
+      a += ((uint32_t)k[3])<<24;
+      b += k[4];
+      b += ((uint32_t)k[5])<<8;
+      b += ((uint32_t)k[6])<<16;
+      b += ((uint32_t)k[7])<<24;
+      c += k[8];
+      c += ((uint32_t)k[9])<<8;
+      c += ((uint32_t)k[10])<<16;
+      c += ((uint32_t)k[11])<<24;
+      mix(a,b,c);
+      length -= 12;
+      k += 12;
+    }
+
+    /*-------------------------------- last block: affect all 32 bits of (c) */
+    switch(length)                   /* all the case statements fall through */
+    {
+    case 12: c+=((uint32_t)k[11])<<24;
+	     /* FALLTHROUGH */
+    case 11: c+=((uint32_t)k[10])<<16;
+	     /* FALLTHROUGH */
+    case 10: c+=((uint32_t)k[9])<<8;
+	     /* FALLTHROUGH */
+    case 9 : c+=k[8];
+	     /* FALLTHROUGH */
+    case 8 : b+=((uint32_t)k[7])<<24;
+	     /* FALLTHROUGH */
+    case 7 : b+=((uint32_t)k[6])<<16;
+	     /* FALLTHROUGH */
+    case 6 : b+=((uint32_t)k[5])<<8;
+	     /* FALLTHROUGH */
+    case 5 : b+=k[4];
+	     /* FALLTHROUGH */
+    case 4 : a+=((uint32_t)k[3])<<24;
+	     /* FALLTHROUGH */
+    case 3 : a+=((uint32_t)k[2])<<16;
+	     /* FALLTHROUGH */
+    case 2 : a+=((uint32_t)k[1])<<8;
+	     /* FALLTHROUGH */
+    case 1 : a+=k[0];
+             break;
+    case 0 : return NON_ZERO_64(b, c);
+    }
+  }
+
+  final(a,b,c);
+  return NON_ZERO_64(b, c);
+}
+#else
+#define hashlittle2(key, len) hashlittle(key, len)
+#endif

hlink.c

@@ -4,7 +4,7 @@
  * Copyright (C) 1996 Andrew Tridgell
  * Copyright (C) 1996 Paul Mackerras
  * Copyright (C) 2002 Martin Pool <mbp@samba.org>
- * Copyright (C) 2004-2020 Wayne Davison
+ * Copyright (C) 2004-2022 Wayne Davison
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -406,7 +406,7 @@ int hard_link_check(struct file_struct *file, int ndx, char *fname,
 				}
 				break;
 			}
-			if (!unchanged_file(cmpbuf, file, &alt_sx.st))
+			if (!quick_check_ok(FT_REG, cmpbuf, file, &alt_sx.st))
 				continue;
 			statret = 1;
 			if (unchanged_attrs(cmpbuf, file, &alt_sx))
@@ -446,7 +446,7 @@ int hard_link_check(struct file_struct *file, int ndx, char *fname,
 		return -1;
 
 	if (remove_source_files == 1 && do_xfers)
-		send_msg_int(MSG_SUCCESS, ndx);
+		send_msg_success(fname, ndx);
 
 	return 1;
 }
@@ -519,7 +519,7 @@ void finish_hard_link(struct file_struct *file, const char *fname, int fin_ndx,
 		if (val < 0)
 			continue;
 		if (remove_source_files == 1 && do_xfers)
-			send_msg_int(MSG_SUCCESS, ndx);
+			send_msg_success(fname, ndx);
 	}
 
 	if (inc_recurse) {

ifuncs.h

@@ -1,6 +1,6 @@
 /* Inline functions for rsync.
  *
- * Copyright (C) 2007-2020 Wayne Davison
+ * Copyright (C) 2007-2022 Wayne Davison
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -75,6 +75,7 @@ d_name(struct dirent *di)
 static inline void
 init_stat_x(stat_x *sx_p)
 {
+	sx_p->crtime = 0;
 #ifdef SUPPORT_ACLS
 	sx_p->acc_acl = sx_p->def_acl = NULL;
 #endif
@@ -105,7 +106,7 @@ free_stat_x(stat_x *sx_p)
 static inline char *my_strdup(const char *str, const char *file, int line)
 {
     int len = strlen(str)+1;
-    char *buf = my_alloc(do_malloc, len, 1, file, line);
+    char *buf = my_alloc(NULL, len, 1, file, line);
     memcpy(buf, str, len);
     return buf;
 }

install-sh

@@ -115,7 +115,7 @@ else
 # might cause directories to be created, which would be especially bad 
 # if $src (and thus $dsttmp) contains '*'.
 
-	if [ -f $src -o -d $src ]
+	if [ -f $src ] || [ -d $src ]
 	then
 		true
 	else

io.c

@@ -4,7 +4,7 @@
  * Copyright (C) 1996-2001 Andrew Tridgell
  * Copyright (C) 1996 Paul Mackerras
  * Copyright (C) 2001, 2002 Martin Pool <mbp@samba.org>
- * Copyright (C) 2003-2020 Wayne Davison
+ * Copyright (C) 2003-2022 Wayne Davison
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -41,6 +41,7 @@ extern int am_server;
 extern int am_sender;
 extern int am_receiver;
 extern int am_generator;
+extern int local_server;
 extern int msgs2stderr;
 extern int inc_recurse;
 extern int io_error;
@@ -84,6 +85,8 @@ int sock_f_out = -1;
 int64 total_data_read = 0;
 int64 total_data_written = 0;
 
+char num_dev_ino_buf[4 + 8 + 8];
+
 static struct {
 	xbuf in, out, msg;
 	int in_fd;
@@ -264,15 +267,18 @@ static size_t safe_read(int fd, char *buf, size_t len)
 			rprintf(FINFO, "select exception on fd %d\n", fd); */
 
 		if (FD_ISSET(fd, &r_fds)) {
-			int n = read(fd, buf + got, len - got);
-			if (DEBUG_GTE(IO, 2))
-				rprintf(FINFO, "[%s] safe_read(%d)=%ld\n", who_am_i(), fd, (long)n);
+			ssize_t n = read(fd, buf + got, len - got);
+			if (DEBUG_GTE(IO, 2)) {
+				rprintf(FINFO, "[%s] safe_read(%d)=%" SIZE_T_FMT_MOD "d\n",
+					who_am_i(), fd, (SIZE_T_FMT_CAST)n);
+			}
 			if (n == 0)
 				break;
 			if (n < 0) {
 				if (errno == EINTR)
 					continue;
-				rsyserr(FERROR, errno, "safe_read failed to read %ld bytes", (long)len);
+				rsyserr(FERROR, errno, "safe_read failed to read %" SIZE_T_FMT_MOD "d bytes",
+					(SIZE_T_FMT_CAST)len);
 				exit_cleanup(RERR_STREAMIO);
 			}
 			if ((got += (size_t)n) == len)
@@ -304,7 +310,7 @@ static const char *what_fd_is(int fd)
  * is not used on the socket except very early in the transfer. */
 static void safe_write(int fd, const char *buf, size_t len)
 {
-	int n;
+	ssize_t n;
 
 	assert(fd != iobuf.out_fd);
 
@@ -315,8 +321,8 @@ static void safe_write(int fd, const char *buf, size_t len)
 		if (errno != EINTR && errno != EWOULDBLOCK && errno != EAGAIN) {
 		  write_failed:
 			rsyserr(FERROR, errno,
-				"safe_write failed to write %ld bytes to %s",
-				(long)len, what_fd_is(fd));
+				"safe_write failed to write %" SIZE_T_FMT_MOD "d bytes to %s",
+				(SIZE_T_FMT_CAST)len, what_fd_is(fd));
 			exit_cleanup(RERR_STREAMIO);
 		}
 	} else {
@@ -362,7 +368,7 @@ static void safe_write(int fd, const char *buf, size_t len)
  * a chunk of data and put it into the output buffer. */
 static void forward_filesfrom_data(void)
 {
-	int len;
+	ssize_t len;
 
 	len = read(ff_forward_fd, ff_xb.buf + ff_xb.len, ff_xb.size - ff_xb.len);
 	if (len <= 0) {
@@ -373,12 +379,15 @@ static void forward_filesfrom_data(void)
 			free_xbuf(&ff_xb);
 			if (ff_reenable_multiplex >= 0)
 				io_start_multiplex_out(ff_reenable_multiplex);
+			free_implied_include_partial_string();
 		}
 		return;
 	}
 
-	if (DEBUG_GTE(IO, 2))
-		rprintf(FINFO, "[%s] files-from read=%ld\n", who_am_i(), (long)len);
+	if (DEBUG_GTE(IO, 2)) {
+		rprintf(FINFO, "[%s] files-from read=%" SIZE_T_FMT_MOD "d\n",
+			who_am_i(), (SIZE_T_FMT_CAST)len);
+	}
 
 #ifdef ICONV_OPTION
 	len += ff_xb.len;
@@ -414,6 +423,7 @@ static void forward_filesfrom_data(void)
 		while (s != eob) {
 			if (*s++ == '\0') {
 				ff_xb.len = s - sob - 1;
+				add_implied_include(sob, 0);
 				if (iconvbufs(ic_send, &ff_xb, &iobuf.out, flags) < 0)
 					exit_cleanup(RERR_PROTOCOL); /* impossible? */
 				write_buf(iobuf.out_fd, s-1, 1); /* Send the '\0'. */
@@ -429,6 +439,7 @@ static void forward_filesfrom_data(void)
 			ff_lastchar = '\0';
 		else {
 			/* Handle a partial string specially, saving any incomplete chars. */
+			implied_include_partial_string(sob, s);
 			flags &= ~ICB_INCLUDE_INCOMPLETE;
 			if (iconvbufs(ic_send, &ff_xb, &iobuf.out, flags) < 0) {
 				if (errno == E2BIG)
@@ -445,13 +456,17 @@ static void forward_filesfrom_data(void)
 		char *f = ff_xb.buf + ff_xb.pos;
 		char *t = ff_xb.buf;
 		char *eob = f + len;
+		char *cur = t;
 		/* Eliminate any multi-'\0' runs. */
 		while (f != eob) {
 			if (!(*t++ = *f++)) {
+				add_implied_include(cur, 0);
+				cur = t;
 				while (f != eob && *f == '\0')
 					f++;
 			}
 		}
+		implied_include_partial_string(cur, t);
 		ff_lastchar = f[-1];
 		if ((len = t - ff_xb.buf) != 0) {
 			/* This will not circle back to perform_io() because we only get
@@ -562,52 +577,59 @@ static char *perform_io(size_t needed, int flags)
 	case PIO_NEED_INPUT:
 		/* We never resize the circular input buffer. */
 		if (iobuf.in.size < needed) {
-			rprintf(FERROR, "need to read %ld bytes, iobuf.in.buf is only %ld bytes.\n",
-				(long)needed, (long)iobuf.in.size);
+			rprintf(FERROR, "need to read %" SIZE_T_FMT_MOD "d bytes,"
+					" iobuf.in.buf is only %" SIZE_T_FMT_MOD "d bytes.\n",
+				(SIZE_T_FMT_CAST)needed, (SIZE_T_FMT_CAST)iobuf.in.size);
 			exit_cleanup(RERR_PROTOCOL);
 		}
 
 		if (msgs2stderr == 1 && DEBUG_GTE(IO, 3)) {
-			rprintf(FINFO, "[%s] perform_io(%ld, %sinput)\n",
-				who_am_i(), (long)needed, flags & PIO_CONSUME_INPUT ? "consume&" : "");
+			rprintf(FINFO, "[%s] perform_io(%" SIZE_T_FMT_MOD "d, %sinput)\n",
+				who_am_i(), (SIZE_T_FMT_CAST)needed, flags & PIO_CONSUME_INPUT ? "consume&" : "");
 		}
 		break;
 
 	case PIO_NEED_OUTROOM:
 		/* We never resize the circular output buffer. */
 		if (iobuf.out.size - iobuf.out_empty_len < needed) {
-			fprintf(stderr, "need to write %ld bytes, iobuf.out.buf is only %ld bytes.\n",
-				(long)needed, (long)(iobuf.out.size - iobuf.out_empty_len));
+			fprintf(stderr, "need to write %" SIZE_T_FMT_MOD "d bytes,"
+					" iobuf.out.buf is only %" SIZE_T_FMT_MOD "d bytes.\n",
+				(SIZE_T_FMT_CAST)needed, (SIZE_T_FMT_CAST)(iobuf.out.size - iobuf.out_empty_len));
 			exit_cleanup(RERR_PROTOCOL);
 		}
 
 		if (msgs2stderr == 1 && DEBUG_GTE(IO, 3)) {
-			rprintf(FINFO, "[%s] perform_io(%ld, outroom) needs to flush %ld\n",
-				who_am_i(), (long)needed,
+			rprintf(FINFO, "[%s] perform_io(%" SIZE_T_FMT_MOD "d,"
+				       " outroom) needs to flush %" SIZE_T_FMT_MOD "d\n",
+				who_am_i(), (SIZE_T_FMT_CAST)needed,
 				iobuf.out.len + needed > iobuf.out.size
-				? (long)(iobuf.out.len + needed - iobuf.out.size) : 0L);
+				? (SIZE_T_FMT_CAST)(iobuf.out.len + needed - iobuf.out.size) : (SIZE_T_FMT_CAST)0);
 		}
 		break;
 
 	case PIO_NEED_MSGROOM:
 		/* We never resize the circular message buffer. */
 		if (iobuf.msg.size < needed) {
-			fprintf(stderr, "need to write %ld bytes, iobuf.msg.buf is only %ld bytes.\n",
-				(long)needed, (long)iobuf.msg.size);
+			fprintf(stderr, "need to write %" SIZE_T_FMT_MOD "d bytes,"
+					" iobuf.msg.buf is only %" SIZE_T_FMT_MOD "d bytes.\n",
+				(SIZE_T_FMT_CAST)needed, (SIZE_T_FMT_CAST)iobuf.msg.size);
 			exit_cleanup(RERR_PROTOCOL);
 		}
 
 		if (msgs2stderr == 1 && DEBUG_GTE(IO, 3)) {
-			rprintf(FINFO, "[%s] perform_io(%ld, msgroom) needs to flush %ld\n",
-				who_am_i(), (long)needed,
+			rprintf(FINFO, "[%s] perform_io(%" SIZE_T_FMT_MOD "d,"
+				       " msgroom) needs to flush %" SIZE_T_FMT_MOD "d\n",
+				who_am_i(), (SIZE_T_FMT_CAST)needed,
 				iobuf.msg.len + needed > iobuf.msg.size
-				? (long)(iobuf.msg.len + needed - iobuf.msg.size) : 0L);
+				? (SIZE_T_FMT_CAST)(iobuf.msg.len + needed - iobuf.msg.size) : (SIZE_T_FMT_CAST)0);
 		}
 		break;
 
 	case 0:
-		if (msgs2stderr == 1 && DEBUG_GTE(IO, 3))
-			rprintf(FINFO, "[%s] perform_io(%ld, %d)\n", who_am_i(), (long)needed, flags);
+		if (msgs2stderr == 1 && DEBUG_GTE(IO, 3)) {
+			rprintf(FINFO, "[%s] perform_io(%" SIZE_T_FMT_MOD "d, %d)\n",
+				who_am_i(), (SIZE_T_FMT_CAST)needed, flags);
+		}
 		break;
 
 	default:
@@ -665,8 +687,8 @@ static char *perform_io(size_t needed, int flags)
 					      ((MPLEX_BASE + (int)MSG_DATA)<<24) + iobuf.out.len - 4);
 
 					if (msgs2stderr == 1 && DEBUG_GTE(IO, 1)) {
-						rprintf(FINFO, "[%s] send_msg(%d, %ld)\n",
-							who_am_i(), (int)MSG_DATA, (long)iobuf.out.len - 4);
+						rprintf(FINFO, "[%s] send_msg(%d, %" SIZE_T_FMT_MOD "d)\n",
+							who_am_i(), (int)MSG_DATA, (SIZE_T_FMT_CAST)iobuf.out.len - 4);
 					}
 
 					/* reserve room for the next MSG_DATA header */
@@ -757,7 +779,7 @@ static char *perform_io(size_t needed, int flags)
 
 		if (iobuf.in_fd >= 0 && FD_ISSET(iobuf.in_fd, &r_fds)) {
 			size_t len, pos = iobuf.in.pos + iobuf.in.len;
-			int n;
+			ssize_t n;
 			if (pos >= iobuf.in.size) {
 				pos -= iobuf.in.size;
 				len = iobuf.in.size - iobuf.in.len;
@@ -784,15 +806,13 @@ static char *perform_io(size_t needed, int flags)
 					exit_cleanup(RERR_SOCKETIO);
 				}
 			}
-			if (msgs2stderr == 1 && DEBUG_GTE(IO, 2))
-				rprintf(FINFO, "[%s] recv=%ld\n", who_am_i(), (long)n);
+			if (msgs2stderr == 1 && DEBUG_GTE(IO, 2)) {
+				rprintf(FINFO, "[%s] recv=%" SIZE_T_FMT_MOD "d\n",
+					who_am_i(), (SIZE_T_FMT_CAST)n);
+			}
 
-			if (io_timeout || stop_at_utime) {
+			if (io_timeout) {
 				last_io_in = time(NULL);
-				if (stop_at_utime && last_io_in >= stop_at_utime) {
-					rprintf(FERROR, "stopping at requested limit\n");
-					exit_cleanup(RERR_TIMEOUT);
-				}
 				if (io_timeout && flags & PIO_NEED_INPUT)
 					maybe_send_keepalive(last_io_in, 0);
 			}
@@ -801,9 +821,14 @@ static char *perform_io(size_t needed, int flags)
 			iobuf.in.len += n;
 		}
 
+		if (stop_at_utime && time(NULL) >= stop_at_utime) {
+			rprintf(FERROR, "stopping at requested limit\n");
+			exit_cleanup(RERR_TIMEOUT);
+		}
+
 		if (out && FD_ISSET(iobuf.out_fd, &w_fds)) {
 			size_t len = iobuf.raw_flushing_ends_before ? iobuf.raw_flushing_ends_before - out->pos : out->len;
-			int n;
+			ssize_t n;
 
 			if (bwlimit_writemax && len > bwlimit_writemax)
 				len = bwlimit_writemax;
@@ -824,8 +849,8 @@ static char *perform_io(size_t needed, int flags)
 				}
 			}
 			if (msgs2stderr == 1 && DEBUG_GTE(IO, 2)) {
-				rprintf(FINFO, "[%s] %s sent=%ld\n",
-					who_am_i(), out == &iobuf.out ? "out" : "msg", (long)n);
+				rprintf(FINFO, "[%s] %s sent=%" SIZE_T_FMT_MOD "d\n",
+					who_am_i(), out == &iobuf.out ? "out" : "msg", (SIZE_T_FMT_CAST)n);
 			}
 
 			if (io_timeout)
@@ -945,8 +970,10 @@ int send_msg(enum msgcode code, const char *buf, size_t len, int convert)
 	if (!OUT_MULTIPLEXED)
 		return 0;
 
-	if (want_debug)
-		rprintf(FINFO, "[%s] send_msg(%d, %ld)\n", who_am_i(), (int)code, (long)len);
+	if (want_debug) {
+		rprintf(FINFO, "[%s] send_msg(%d, %" SIZE_T_FMT_MOD "d)\n",
+			who_am_i(), (int)code, (SIZE_T_FMT_CAST)len);
+	}
 
 	/* When checking for enough free space for this message, we need to
 	 * make sure that there is space for the 4-byte header, plus we'll
@@ -962,9 +989,9 @@ int send_msg(enum msgcode code, const char *buf, size_t len, int convert)
 #endif
 		needed = len + 4 + 3;
 	if (iobuf.msg.len + needed > iobuf.msg.size) {
-		if (!am_receiver)
+		if (am_sender)
 			perform_io(needed, PIO_NEED_MSGROOM);
-		else { /* We allow the receiver to increase their iobuf.msg size to avoid a deadlock. */
+		else { /* We sometimes allow the iobuf.msg size to increase to avoid a deadlock. */
 			size_t old_size = iobuf.msg.size;
 			restore_iobuf_size(&iobuf.msg);
 			realloc_xbuf(&iobuf.msg, iobuf.msg.size * 2);
@@ -1021,8 +1048,10 @@ int send_msg(enum msgcode code, const char *buf, size_t len, int convert)
 
 	SIVAL(hdr, 0, ((MPLEX_BASE + (int)code)<<24) + len);
 
-	if (want_debug && convert > 0)
-		rprintf(FINFO, "[%s] converted msg len=%ld\n", who_am_i(), (long)len);
+	if (want_debug && convert > 0) {
+		rprintf(FINFO, "[%s] converted msg len=%" SIZE_T_FMT_MOD "d\n",
+			who_am_i(), (SIZE_T_FMT_CAST)len);
+	}
 
 	return 1;
 }
@@ -1038,6 +1067,24 @@ void send_msg_int(enum msgcode code, int num)
 	send_msg(code, numbuf, 4, -1);
 }
 
+void send_msg_success(const char *fname, int num)
+{
+	if (local_server) {
+		STRUCT_STAT st;
+
+		if (DEBUG_GTE(IO, 1))
+			rprintf(FINFO, "[%s] send_msg_success(%d)\n", who_am_i(), num);
+
+		if (stat(fname, &st) < 0)
+			memset(&st, 0, sizeof (STRUCT_STAT));
+		SIVAL(num_dev_ino_buf, 0, num);
+		SIVAL64(num_dev_ino_buf, 4, st.st_dev);
+		SIVAL64(num_dev_ino_buf, 4+8, st.st_ino);
+		send_msg(MSG_SUCCESS, num_dev_ino_buf, sizeof num_dev_ino_buf, -1);
+	} else
+		send_msg_int(MSG_SUCCESS, num);
+}
+
 static void got_flist_entry_status(enum festatus status, int ndx)
 {
 	struct file_list *flist = flist_for_ndx(ndx, "got_flist_entry_status");
@@ -1052,8 +1099,12 @@ static void got_flist_entry_status(enum festatus status, int ndx)
 
 	switch (status) {
 	case FES_SUCCESS:
-		if (remove_source_files)
-			send_msg_int(MSG_SUCCESS, ndx);
+		if (remove_source_files) {
+			if (local_server)
+				send_msg(MSG_SUCCESS, num_dev_ino_buf, sizeof num_dev_ino_buf, -1);
+			else
+				send_msg_int(MSG_SUCCESS, ndx);
+		}
 		/* FALL THROUGH */
 	case FES_NO_SEND:
 #ifdef SUPPORT_HARD_LINKS
@@ -1436,8 +1487,10 @@ static void read_a_msg(void)
 	msg_bytes = tag & 0xFFFFFF;
 	tag = (tag >> 24) - MPLEX_BASE;
 
-	if (msgs2stderr == 1 && DEBUG_GTE(IO, 1))
-		rprintf(FINFO, "[%s] got msg=%d, len=%ld\n", who_am_i(), (int)tag, (long)msg_bytes);
+	if (msgs2stderr == 1 && DEBUG_GTE(IO, 1)) {
+		rprintf(FINFO, "[%s] got msg=%d, len=%" SIZE_T_FMT_MOD "d\n",
+			who_am_i(), (int)tag, (SIZE_T_FMT_CAST)msg_bytes);
+	}
 
 	switch (tag) {
 	case MSG_DATA:
@@ -1546,14 +1599,15 @@ static void read_a_msg(void)
 		}
 		break;
 	case MSG_SUCCESS:
-		if (msg_bytes != 4) {
+		if (msg_bytes != (local_server ? 4+8+8 : 4)) {
 		  invalid_msg:
 			rprintf(FERROR, "invalid multi-message %d:%lu [%s%s]\n",
 				tag, (unsigned long)msg_bytes, who_am_i(),
 				inc_recurse ? "/inc" : "");
 			exit_cleanup(RERR_STREAMIO);
 		}
-		val = raw_read_int();
+		raw_read_buf(num_dev_ino_buf, msg_bytes);
+		val = IVAL(num_dev_ino_buf, 0);
 		iobuf.in_multiplexed = 1;
 		if (am_generator)
 			got_flist_entry_status(FES_SUCCESS, val);
@@ -1613,8 +1667,10 @@ static void read_a_msg(void)
 		else
 			goto invalid_msg;
 		iobuf.in_multiplexed = 1;
-		if (DEBUG_GTE(EXIT, 3))
-			rprintf(FINFO, "[%s] got MSG_ERROR_EXIT with %ld bytes\n", who_am_i(), (long)msg_bytes);
+		if (DEBUG_GTE(EXIT, 3)) {
+			rprintf(FINFO, "[%s] got MSG_ERROR_EXIT with %" SIZE_T_FMT_MOD "d bytes\n",
+					who_am_i(), (SIZE_T_FMT_CAST)msg_bytes);
+		}
 		if (msg_bytes == 0) {
 			if (!am_sender && !am_generator) {
 				if (DEBUG_GTE(EXIT, 3)) {
@@ -1728,6 +1784,13 @@ int32 read_int(int f)
 	return num;
 }
 
+uint32 read_uint(int f)
+{
+	char b[4];
+	read_buf(f, b, 4);
+	return IVAL(b, 0);
+}
+
 int32 read_varint(int f)
 {
 	union {
@@ -1820,6 +1883,7 @@ int64 read_longint(int f)
 #endif
 }
 
+/* Debugging note: this will be named read_buf_() when using an external zlib. */
 void read_buf(int f, char *buf, size_t len)
 {
 	if (f != iobuf.in_fd) {

itypes.h

@@ -1,6 +1,6 @@
 /* Inline functions for rsync.
  *
- * Copyright (C) 2007-2020 Wayne Davison
+ * Copyright (C) 2007-2022 Wayne Davison
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -40,6 +40,12 @@ isSpace(const char *ptr)
 	return isspace(*(unsigned char *)ptr);
 }
 
+static inline int
+isAlNum(const char *ptr)
+{
+	return isalnum(*(unsigned char *)ptr);
+}
+
 static inline int
 isLower(const char *ptr)
 {

latest-year.h

@@ -1 +1 @@
-#define LATEST_YEAR "2020"
+#define LATEST_YEAR "2024"

lib/md-defines.h

@@ -1,11 +1,28 @@
 /* Keep this simple so both C and ASM can use it */
 
+/* These allow something like CFLAGS=-DDISABLE_SHA512_DIGEST */
+#ifdef DISABLE_SHA256_DIGEST
+#undef SHA256_DIGEST_LENGTH
+#endif
+#ifdef DISABLE_SHA512_DIGEST
+#undef SHA512_DIGEST_LENGTH
+#endif
+
 #define MD4_DIGEST_LEN 16
 #define MD5_DIGEST_LEN 16
+#if defined SHA512_DIGEST_LENGTH
+#define MAX_DIGEST_LEN SHA512_DIGEST_LENGTH
+#elif defined SHA256_DIGEST_LENGTH
+#define MAX_DIGEST_LEN SHA256_DIGEST_LENGTH
+#elif defined SHA_DIGEST_LENGTH
+#define MAX_DIGEST_LEN SHA_DIGEST_LENGTH
+#else
 #define MAX_DIGEST_LEN MD5_DIGEST_LEN
+#endif
 
 #define CSUM_CHUNK 64
 
+#define CSUM_gone -1
 #define CSUM_NONE 0
 #define CSUM_MD4_ARCHAIC 1
 #define CSUM_MD4_BUSTED 2
@@ -15,3 +32,6 @@
 #define CSUM_XXH64 6
 #define CSUM_XXH3_64 7
 #define CSUM_XXH3_128 8
+#define CSUM_SHA1 9
+#define CSUM_SHA256 10
+#define CSUM_SHA512 11

lib/md5-asm-x86_64.S

@@ -27,7 +27,7 @@
 #include "config.h"
 #include "md-defines.h"
 
-#if !defined USE_OPENSSL && CSUM_CHUNK == 64
+#ifdef USE_MD5_ASM /* { */
 
 #ifdef __APPLE__
 #define md5_process_asm _md5_process_asm
@@ -698,4 +698,4 @@ md5_process_asm:
 	pop	%rbp
 	ret
 
-#endif /* !USE_OPENSSL ... */
+#endif /* } USE_MD5_ASM */

lib/md5.c

@@ -2,7 +2,7 @@
  * RFC 1321 compliant MD5 implementation
  *
  * Copyright (C) 2001-2003 Christophe Devine
- * Copyright (C) 2007-2020 Wayne Davison
+ * Copyright (C) 2007-2022 Wayne Davison
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -20,7 +20,6 @@
 
 #include "rsync.h"
 
-#ifndef USE_OPENSSL
 void md5_begin(md_context *ctx)
 {
 	ctx->A = 0x67452301;
@@ -148,7 +147,10 @@ static void md5_process(md_context *ctx, const uchar data[CSUM_CHUNK])
 	ctx->D += D;
 }
 
-#if defined HAVE_ASM && CSUM_CHUNK == 64
+#ifdef USE_MD5_ASM
+#if CSUM_CHUNK != 64
+#error The MD5 ASM code does not support CSUM_CHUNK != 64
+#endif
 extern void md5_process_asm(md_context *ctx, const void *data, size_t num);
 #endif
 
@@ -176,20 +178,20 @@ void md5_update(md_context *ctx, const uchar *input, uint32 length)
 		left = 0;
 	}
 
-#if defined HAVE_ASM && CSUM_CHUNK == 64
+#ifdef USE_MD5_ASM /* { */
 	if (length >= CSUM_CHUNK) {
 		uint32 chunks = length / CSUM_CHUNK;
 		md5_process_asm(ctx, input, chunks);
 		length -= chunks * CSUM_CHUNK;
 		input += chunks * CSUM_CHUNK;
 	}
-#else
+#else /* } { */
 	while (length >= CSUM_CHUNK) {
 		md5_process(ctx, input);
 		length -= CSUM_CHUNK;
 		input  += CSUM_CHUNK;
 	}
-#endif
+#endif /* } */
 
 	if (length)
 		memcpy(ctx->buffer + left, input, length);
@@ -221,9 +223,8 @@ void md5_result(md_context *ctx, uchar digest[MD5_DIGEST_LEN])
 	SIVALu(digest, 8, ctx->C);
 	SIVALu(digest, 12, ctx->D);
 }
-#endif
 
-#ifdef TEST_MD5
+#ifdef TEST_MD5 /* { */
 
 void get_md5(uchar *out, const uchar *input, int n)
 {
@@ -317,4 +318,4 @@ int main(int argc, char *argv[])
 	return 0;
 }
 
-#endif
+#endif /* } */

lib/mdigest.h

@@ -1,8 +1,8 @@
 /* The include file for both the MD4 and MD5 routines. */
 
 #ifdef USE_OPENSSL
-#include "openssl/md4.h"
-#include "openssl/md5.h"
+#include <openssl/sha.h>
+#include <openssl/evp.h>
 #endif
 #include "md-defines.h"
 
@@ -17,13 +17,6 @@ void mdfour_begin(md_context *md);
 void mdfour_update(md_context *md, const uchar *in, uint32 length);
 void mdfour_result(md_context *md, uchar digest[MD4_DIGEST_LEN]);
 
-#ifndef USE_OPENSSL
-#define MD5_CTX md_context
-#define MD5_Init md5_begin
-#define MD5_Update md5_update
-#define MD5_Final(digest, cptr) md5_result(cptr, digest)
-
 void md5_begin(md_context *ctx);
 void md5_update(md_context *ctx, const uchar *input, uint32 length);
 void md5_result(md_context *ctx, uchar digest[MD5_DIGEST_LEN]);
-#endif

lib/pool_alloc.c

@@ -9,8 +9,7 @@ struct alloc_pool
 	size_t			size;		/* extent size		*/
 	size_t			quantum;	/* allocation quantum	*/
 	struct pool_extent	*extents;	/* top extent is "live" */
-	void			(*bomb)();	/* function to call if
-						 * malloc fails		*/
+	void			(*bomb)();	/* called if malloc fails */
 	int			flags;
 
 	/* statistical data */
@@ -49,7 +48,7 @@ pool_create(size_t size, size_t quantum, void (*bomb)(const char*, const char*,
 {
 	struct alloc_pool *pool;
 
-	if ((MINALIGN & (MINALIGN - 1)) != 0) {
+	if ((MINALIGN & (MINALIGN - 1)) != (0)) {
 		if (bomb)
 			(*bomb)("Compiler error: MINALIGN is not a power of 2", __FILE__, __LINE__);
 		return NULL;

lib/snprintf.c

@@ -20,7 +20,7 @@
  * for string length.  This covers a nasty loophole.
  *
  * The other functions are there to prevent NULL pointers from
- * causing nast effects.
+ * causing nasty effects.
  *
  * More Recently:
  *  Brandon Long <blong@fiction.net> 9/15/96 for mutt 0.43

lib/sysacls.c

@@ -2,7 +2,7 @@
  * Unix SMB/CIFS implementation.
  * Based on the Samba ACL support code.
  * Copyright (C) Jeremy Allison 2000.
- * Copyright (C) 2007-2020 Wayne Davison
+ * Copyright (C) 2007-2022 Wayne Davison
  *
  * The permission functions have been changed to get/set all bits via
  * one call.  Some functions that rsync doesn't need were also removed.
@@ -175,7 +175,7 @@ int sys_acl_delete_def_file(const char *name)
 	return acl_delete_def_file(name);
 }
 
-int sys_acl_free_acl(SMB_ACL_T the_acl) 
+int sys_acl_free_acl(SMB_ACL_T the_acl)
 {
 	return acl_free(the_acl);
 }
@@ -185,7 +185,7 @@ int sys_acl_free_acl(SMB_ACL_T the_acl)
  * The interface to DEC/Compaq Tru64 UNIX ACLs
  * is based on Draft 13 of the POSIX spec which is
  * slightly different from the Draft 16 interface.
- * 
+ *
  * Also, some of the permset manipulation functions
  * such as acl_clear_perm() and acl_add_perm() appear
  * to be broken on Tru64 so we have to manipulate
@@ -310,7 +310,7 @@ int sys_acl_delete_def_file(const char *name)
 	return acl_delete_def_file((char *)name);
 }
 
-int sys_acl_free_acl(SMB_ACL_T the_acl) 
+int sys_acl_free_acl(SMB_ACL_T the_acl)
 {
 	return acl_free(the_acl);
 }
@@ -457,7 +457,7 @@ SMB_ACL_T sys_acl_get_file(const char *path_p, SMB_ACL_TYPE_T type)
 			break;
 	}
 	ndefault = count - naccess;
-	
+
 	/*
 	 * if the caller wants the default ACL we have to copy
 	 * the entries down to the start of the acl[] buffer
@@ -517,7 +517,7 @@ SMB_ACL_T sys_acl_get_fd(int fd)
 		if (acl_d->acl[naccess].a_type & ACL_DEFAULT)
 			break;
 	}
-	
+
 	acl_d->count = naccess;
 
 	return acl_d;
@@ -532,7 +532,7 @@ int sys_acl_get_info(SMB_ACL_ENTRY_T entry, SMB_ACL_TAG_T *tag_type_p, uint32 *b
 
 	if (*tag_type_p == SMB_ACL_USER || *tag_type_p == SMB_ACL_GROUP)
 		*u_g_id_p = entry->a_id;
-	
+
 	return 0;
 }
 
@@ -633,7 +633,7 @@ static int acl_sort(SMB_ACL_T acl_d)
 	}
 	return 0;
 }
- 
+
 int sys_acl_valid(SMB_ACL_T acl_d)
 {
 	return acl_sort(acl_d);
@@ -755,11 +755,11 @@ int sys_acl_delete_def_file(const char *path)
 	ret = acl(path, SETACL, acl_d->count, acl_d->acl);
 
 	sys_acl_free_acl(acl_d);
-	
+
 	return ret;
 }
 
-int sys_acl_free_acl(SMB_ACL_T acl_d) 
+int sys_acl_free_acl(SMB_ACL_T acl_d)
 {
 	SAFE_FREE(acl_d);
 	return 0;
@@ -895,10 +895,10 @@ SMB_ACL_T sys_acl_get_file(const char *path_p, SMB_ACL_TYPE_T type)
 	int		ndefault;	/* # of default ACL entries	*/
 
 	if (hpux_acl_call_presence() == False) {
-		/* Looks like we don't have the acl() system call on HPUX. 
+		/* Looks like we don't have the acl() system call on HPUX.
 		 * May be the system doesn't have the latest version of JFS.
 		 */
-		return NULL; 
+		return NULL;
 	}
 
 	if (type != SMB_ACL_TYPE_ACCESS && type != SMB_ACL_TYPE_DEFAULT) {
@@ -949,7 +949,7 @@ SMB_ACL_T sys_acl_get_file(const char *path_p, SMB_ACL_TYPE_T type)
 			break;
 	}
 	ndefault = count - naccess;
-	
+
 	/*
 	 * if the caller wants the default ACL we have to copy
 	 * the entries down to the start of the acl[] buffer
@@ -1109,9 +1109,9 @@ struct hpux_acl_types {
  * aclp           - Array of ACL structures.
  * acl_type_count - Pointer to acl_types structure. Should already be
  *                  allocated.
- * Output: 
+ * Output:
  *
- * acl_type_count - This structure is filled up with counts of various 
+ * acl_type_count - This structure is filled up with counts of various
  *                  acl types.
  */
 
@@ -1123,28 +1123,28 @@ static void hpux_count_obj(int acl_count, struct acl *aclp, struct hpux_acl_type
 
 	for (i = 0; i < acl_count; i++) {
 		switch (aclp[i].a_type) {
-		case USER: 
+		case USER:
 			acl_type_count->n_user++;
 			break;
-		case USER_OBJ: 
+		case USER_OBJ:
 			acl_type_count->n_user_obj++;
 			break;
-		case DEF_USER_OBJ: 
+		case DEF_USER_OBJ:
 			acl_type_count->n_def_user_obj++;
 			break;
-		case GROUP: 
+		case GROUP:
 			acl_type_count->n_group++;
 			break;
-		case GROUP_OBJ: 
+		case GROUP_OBJ:
 			acl_type_count->n_group_obj++;
 			break;
-		case DEF_GROUP_OBJ: 
+		case DEF_GROUP_OBJ:
 			acl_type_count->n_def_group_obj++;
 			break;
-		case OTHER_OBJ: 
+		case OTHER_OBJ:
 			acl_type_count->n_other_obj++;
 			break;
-		case DEF_OTHER_OBJ: 
+		case DEF_OTHER_OBJ:
 			acl_type_count->n_def_other_obj++;
 			break;
 		case CLASS_OBJ:
@@ -1159,14 +1159,14 @@ static void hpux_count_obj(int acl_count, struct acl *aclp, struct hpux_acl_type
 		case DEF_GROUP:
 			acl_type_count->n_def_group++;
 			break;
-		default: 
+		default:
 			acl_type_count->n_illegal_obj++;
 			break;
 		}
 	}
 }
 
-/* swap_acl_entries:  Swaps two ACL entries. 
+/* swap_acl_entries:  Swaps two ACL entries.
  *
  * Inputs: aclp0, aclp1 - ACL entries to be swapped.
  */
@@ -1189,25 +1189,25 @@ static void hpux_swap_acl_entries(struct acl *aclp0, struct acl *aclp1)
 }
 
 /* prohibited_duplicate_type
- * Identifies if given ACL type can have duplicate entries or 
+ * Identifies if given ACL type can have duplicate entries or
  * not.
  *
  * Inputs: acl_type - ACL Type.
  *
- * Outputs: 
+ * Outputs:
  *
- * Return.. 
+ * Return..
  *
  * True - If the ACL type matches any of the prohibited types.
  * False - If the ACL type doesn't match any of the prohibited types.
- */ 
+ */
 
 static BOOL hpux_prohibited_duplicate_type(int acl_type)
 {
 	switch (acl_type) {
 	case USER:
 	case GROUP:
-	case DEF_USER: 
+	case DEF_USER:
 	case DEF_GROUP:
 		return True;
 	default:
@@ -1217,7 +1217,7 @@ static BOOL hpux_prohibited_duplicate_type(int acl_type)
 
 /* get_needed_class_perm
  * Returns the permissions of a ACL structure only if the ACL
- * type matches one of the pre-determined types for computing 
+ * type matches one of the pre-determined types for computing
  * CLASS_OBJ permissions.
  *
  * Inputs: aclp - Pointer to ACL structure.
@@ -1226,17 +1226,17 @@ static BOOL hpux_prohibited_duplicate_type(int acl_type)
 static int hpux_get_needed_class_perm(struct acl *aclp)
 {
 	switch (aclp->a_type) {
-	case USER: 
-	case GROUP_OBJ: 
-	case GROUP: 
-	case DEF_USER_OBJ: 
+	case USER:
+	case GROUP_OBJ:
+	case GROUP:
+	case DEF_USER_OBJ:
 	case DEF_USER:
-	case DEF_GROUP_OBJ: 
+	case DEF_GROUP_OBJ:
 	case DEF_GROUP:
 	case DEF_CLASS_OBJ:
-	case DEF_OTHER_OBJ: 
+	case DEF_OTHER_OBJ:
 		return aclp->a_perm;
-	default: 
+	default:
 		return 0;
 	}
 }
@@ -1267,15 +1267,15 @@ static int hpux_acl_sort(int acl_count, int calclass, struct acl *aclp)
 #if !defined(HAVE_HPUX_ACLSORT)
 	/*
 	 * The aclsort() system call is available on the latest HPUX General
-	 * Patch Bundles. So for HPUX, we developed our version of acl_sort 
-	 * function. Because, we don't want to update to a new 
+	 * Patch Bundles. So for HPUX, we developed our version of acl_sort
+	 * function. Because, we don't want to update to a new
 	 * HPUX GR bundle just for aclsort() call.
 	 */
 
 	struct hpux_acl_types acl_obj_count;
 	int n_class_obj_perm = 0;
 	int i, j;
- 
+
 	if (!acl_count) {
 		DEBUG(10, ("Zero acl count passed. Returning Success\n"));
 		return 0;
@@ -1290,8 +1290,8 @@ static int hpux_acl_sort(int acl_count, int calclass, struct acl *aclp)
 
 	hpux_count_obj(acl_count, aclp, &acl_obj_count);
 
-	/* There should be only one entry each of type USER_OBJ, GROUP_OBJ, 
-	 * CLASS_OBJ and OTHER_OBJ 
+	/* There should be only one entry each of type USER_OBJ, GROUP_OBJ,
+	 * CLASS_OBJ and OTHER_OBJ
 	 */
 
 	if (acl_obj_count.n_user_obj != 1
@@ -1313,15 +1313,15 @@ or DEF_USER_OBJ or DEF_GROUP_OBJ or DEF_OTHER_OBJ\n"));
 		return -1;
 	}
 
-	/* We now have proper number of OBJ and DEF_OBJ entries. Now sort the acl 
-	 * structures.  
+	/* We now have proper number of OBJ and DEF_OBJ entries. Now sort the acl
+	 * structures.
 	 *
 	 * Sorting crieteria - First sort by ACL type. If there are multiple entries of
 	 * same ACL type, sort by ACL id.
 	 *
-	 * I am using the trivial kind of sorting method here because, performance isn't 
+	 * I am using the trivial kind of sorting method here because, performance isn't
 	 * really effected by the ACLs feature. More over there aren't going to be more
-	 * than 17 entries on HPUX. 
+	 * than 17 entries on HPUX.
 	 */
 
 	for (i = 0; i < acl_count; i++) {
@@ -1390,7 +1390,7 @@ static int acl_sort(SMB_ACL_T acl_d)
 	}
 	return 0;
 }
- 
+
 int sys_acl_valid(SMB_ACL_T acl_d)
 {
 	return acl_sort(acl_d);
@@ -1405,11 +1405,11 @@ int sys_acl_set_file(const char *name, SMB_ACL_TYPE_T type, SMB_ACL_T acl_d)
 	int		ret;
 
 	if (hpux_acl_call_presence() == False) {
-		/* Looks like we don't have the acl() system call on HPUX. 
+		/* Looks like we don't have the acl() system call on HPUX.
 		 * May be the system doesn't have the latest version of JFS.
 		 */
 		errno=ENOSYS;
-		return -1; 
+		return -1;
 	}
 
 	if (type != SMB_ACL_TYPE_ACCESS && type != SMB_ACL_TYPE_DEFAULT) {
@@ -1538,11 +1538,11 @@ int sys_acl_delete_def_file(const char *path)
 	ret = acl(path, ACL_SET, acl_d->count, acl_d->acl);
 
 	sys_acl_free_acl(acl_d);
-	
+
 	return ret;
 }
 
-int sys_acl_free_acl(SMB_ACL_T acl_d) 
+int sys_acl_free_acl(SMB_ACL_T acl_d)
 {
 	free(acl_d);
 	return 0;
@@ -1723,7 +1723,7 @@ int sys_acl_delete_def_file(const char *name)
 	return acl_delete_def_file(name);
 }
 
-int sys_acl_free_acl(SMB_ACL_T acl_d) 
+int sys_acl_free_acl(SMB_ACL_T acl_d)
 {
 	if (acl_d->freeaclp) {
 		acl_free(acl_d->aclp);
@@ -1739,7 +1739,6 @@ int sys_acl_free_acl(SMB_ACL_T acl_d)
 int sys_acl_get_entry(SMB_ACL_T theacl, int entry_id, SMB_ACL_ENTRY_T *entry_p)
 {
 	struct acl_entry_link *link;
-	struct new_acl_entry *entry;
 	int keep_going;
 
 	if (entry_id == SMB_ACL_FIRST_ENTRY)
@@ -1765,10 +1764,15 @@ int sys_acl_get_entry(SMB_ACL_T theacl, int entry_id, SMB_ACL_ENTRY_T *entry_p)
 	for (keep_going = 0; keep_going < theacl->count; keep_going++)
 		link = link->nextp;
 
-	entry = *entry_p =  link->entryp;
+	*entry_p =  link->entryp;
 
-	DEBUG(10, ("*entry_p is %d\n", entry_p));
-	DEBUG(10, ("*entry_p->ace_access is %d\n", entry->ace_access));
+#if 0
+	{
+		struct new_acl_entry *entry = *entry_p;
+		DEBUG(10, ("*entry_p is %lx\n", (long)entry));
+		DEBUG(10, ("*entry_p->ace_access is %d\n", entry->ace_access));
+	}
+#endif
 
 	/* Increment count */
 	theacl->count++;
@@ -1830,12 +1834,12 @@ SMB_ACL_T sys_acl_get_file(const char *path_p, SMB_ACL_TYPE_T type)
 	}
 
 	/* Get the acl using statacl */
- 
+
 	DEBUG(10, ("Entering sys_acl_get_file\n"));
 	DEBUG(10, ("path_p is %s\n", path_p));
 
 	file_acl = (struct acl *)SMB_MALLOC(BUFSIZ);
- 
+
 	if (file_acl == NULL) {
 		errno=ENOMEM;
 		DEBUG(0, ("Error in AIX sys_acl_get_file: %d\n", errno));
@@ -1927,9 +1931,9 @@ SMB_ACL_T sys_acl_get_file(const char *path_p, SMB_ACL_TYPE_T type)
 			 * to be specified but, it's better than leaving it 0 */
 
 			acl_entry_link->entryp->ace_type = acl_entry->ace_type;
- 
+
 			acl_entry_link->entryp->ace_access = acl_entry->ace_access;
- 
+
 			memcpy(acl_entry_link->entryp->ace_id, idp, sizeof (struct ace_id));
 
 			/* The access in the acl entries must be left shifted by *
@@ -1958,7 +1962,7 @@ SMB_ACL_T sys_acl_get_file(const char *path_p, SMB_ACL_TYPE_T type)
 
 			DEBUG(10, ("acl_entry = %d\n", acl_entry));
 			DEBUG(10, ("The ace_type is %d\n", acl_entry->ace_type));
- 
+
 			acl_entry = acl_nxt(acl_entry);
 		}
 	} /* end of if enabled */
@@ -2010,12 +2014,12 @@ SMB_ACL_T sys_acl_get_file(const char *path_p, SMB_ACL_TYPE_T type)
 			new_acl_entry->ace_access = file_acl->o_access << 6;
 			idp->id_type = SMB_ACL_OTHER;
 			break;
- 
+
 		case 1:
 			new_acl_entry->ace_access = file_acl->u_access << 6;
 			idp->id_type = SMB_ACL_USER_OBJ;
 			break;
- 
+
 		default:
 			return NULL;
 
@@ -2044,7 +2048,7 @@ SMB_ACL_T sys_acl_get_fd(int fd)
 	int rc = 0;
 
 	/* Get the acl using fstatacl */
-   
+
 	DEBUG(10, ("Entering sys_acl_get_fd\n"));
 	DEBUG(10, ("fd is %d\n", fd));
 	file_acl = (struct acl *)SMB_MALLOC(BUFSIZ);
@@ -2091,12 +2095,12 @@ SMB_ACL_T sys_acl_get_fd(int fd)
 
 	DEBUG(10, ("acl_entry is %d\n", acl_entry));
 	DEBUG(10, ("acl_last(file_acl) id %d\n", acl_last(file_acl)));
- 
+
 	/* Check if the extended acl bit is on.   *
 	 * If it isn't, do not show the           *
 	 * contents of the acl since AIX intends  *
 	 * the extended info to remain unused     */
- 
+
 	if (file_acl->acl_mode & S_IXACL){
 		/* while we are not pointing to the very end */
 		while (acl_entry < acl_last(file_acl)) {
@@ -2111,7 +2115,7 @@ SMB_ACL_T sys_acl_get_fd(int fd)
 			}
 
 			idp = acl_entry->ace_id;
- 
+
 			/* Check if this is the first entry in the linked list. *
 			 * The first entry needs to keep prevp pointing to NULL *
 			 * and already has entryp allocated.                 */
@@ -2173,7 +2177,7 @@ SMB_ACL_T sys_acl_get_fd(int fd)
 
 			DEBUG(10, ("acl_entry = %d\n", acl_entry));
 			DEBUG(10, ("The ace_type is %d\n", acl_entry->ace_type));
- 
+
 			acl_entry = acl_nxt(acl_entry);
 		}
 	} /* end of if enabled */
@@ -2206,43 +2210,43 @@ SMB_ACL_T sys_acl_get_fd(int fd)
 		}
 
 		acl_entry_link->nextp = NULL;
- 
+
 		new_acl_entry = acl_entry_link->entryp;
 		idp = new_acl_entry->ace_id;
- 
+
 		new_acl_entry->ace_len = sizeof (struct acl_entry);
 		new_acl_entry->ace_type = ACC_PERMIT;
 		idp->id_len = sizeof (struct ace_id);
 		DEBUG(10, ("idp->id_len = %d\n", idp->id_len));
 		memset(idp->id_data, 0, sizeof (uid_t));
- 
+
 		switch (i) {
 		case 2:
 			new_acl_entry->ace_access = file_acl->g_access << 6;
 			idp->id_type = SMB_ACL_GROUP_OBJ;
 			break;
- 
+
 		case 3:
 			new_acl_entry->ace_access = file_acl->o_access << 6;
 			idp->id_type = SMB_ACL_OTHER;
 			break;
- 
+
 		case 1:
 			new_acl_entry->ace_access = file_acl->u_access << 6;
 			idp->id_type = SMB_ACL_USER_OBJ;
 			break;
- 
+
 		default:
 			return NULL;
 		}
- 
+
 		acl_entry_link_head->count++;
 		DEBUG(10, ("new_acl_entry->ace_access = %d\n", new_acl_entry->ace_access));
 	}
 
 	acl_entry_link_head->count = 0;
 	SAFE_FREE(file_acl);
- 
+
 	return acl_entry_link_head;
 }
 #endif
@@ -2270,7 +2274,7 @@ int sys_acl_get_info(SMB_ACL_ENTRY_T entry, SMB_ACL_TAG_T *tag_type_p, uint32 *b
 SMB_ACL_T sys_acl_init(int count)
 {
 	struct acl_entry_link *theacl = NULL;
- 
+
 	if (count < 0) {
 		errno = EINVAL;
 		return NULL;
@@ -2297,7 +2301,7 @@ int sys_acl_create_entry(SMB_ACL_T *pacl, SMB_ACL_ENTRY_T *pentry)
 {
 	struct acl_entry_link *theacl;
 	struct acl_entry_link *acl_entryp;
-	struct acl_entry_link *temp_entry;
+	struct acl_entry_link *temp_entry = NULL;
 	int counting;
 
 	DEBUG(10, ("Entering the sys_acl_create_entry\n"));
@@ -2379,9 +2383,9 @@ int sys_acl_valid(SMB_ACL_T theacl)
 	}
 
 	DEBUG(10, ("user_obj=%d, group_obj=%d, other_obj=%d\n", user_obj, group_obj, other_obj));
- 
+
 	if (user_obj != 1 || group_obj != 1 || other_obj != 1)
-		return -1; 
+		return -1;
 
 	return 0;
 }
@@ -2400,7 +2404,7 @@ int sys_acl_set_file(const char *name, SMB_ACL_TYPE_T acltype, SMB_ACL_T theacl)
 
 	DEBUG(10, ("Entering sys_acl_set_file\n"));
 	DEBUG(10, ("File name is %s\n", name));
- 
+
 	/* AIX has no default ACL */
 	if (acltype == SMB_ACL_TYPE_DEFAULT)
 		return 0;
@@ -2445,7 +2449,7 @@ int sys_acl_set_file(const char *name, SMB_ACL_TYPE_T acltype, SMB_ACL_T theacl)
 				errno = ENOMEM;
 				DEBUG(0, ("Error in sys_acl_set_file is %d\n", errno));
 				return -1;
-			}  
+			}
 
 			memcpy(file_acl_temp, file_acl, file_acl->acl_len);
 			SAFE_FREE(file_acl);
@@ -2456,15 +2460,15 @@ int sys_acl_set_file(const char *name, SMB_ACL_TYPE_T acltype, SMB_ACL_T theacl)
 		file_acl->acl_len += sizeof (struct acl_entry);
 		acl_entry->ace_len = acl_entry_link->entryp->ace_len;
 		acl_entry->ace_access = acl_entry_link->entryp->ace_access;
- 
+
 		/* In order to use this, we'll need to wait until we can get denies */
 		/* if (!acl_entry->ace_access && acl_entry->ace_type == ACC_PERMIT)
 			acl_entry->ace_type = ACC_SPECIFY; */
 
 		acl_entry->ace_type = ACC_SPECIFY;
- 
+
 		ace_id = acl_entry->ace_id;
- 
+
 		ace_id->id_type = acl_entry_link->entryp->ace_id->id_type;
 		DEBUG(10, ("The id type is %d\n", ace_id->id_type));
 		ace_id->id_len = acl_entry_link->entryp->ace_id->id_len;
@@ -2492,7 +2496,7 @@ int sys_acl_set_fd(int fd, SMB_ACL_T theacl)
 	uint user_id;
 	uint acl_length;
 	uint rc;
- 
+
 	DEBUG(10, ("Entering sys_acl_set_fd\n"));
 	acl_length = BUFSIZ;
 	file_acl = (struct acl *)SMB_MALLOC(BUFSIZ);
@@ -2504,7 +2508,7 @@ int sys_acl_set_fd(int fd, SMB_ACL_T theacl)
 	}
 
 	memset(file_acl, 0, BUFSIZ);
- 
+
 	file_acl->acl_len = ACL_SIZ;
 	file_acl->acl_mode = S_IXACL;
 
@@ -2546,22 +2550,22 @@ int sys_acl_set_fd(int fd, SMB_ACL_T theacl)
 		file_acl->acl_len += sizeof (struct acl_entry);
 		acl_entry->ace_len = acl_entry_link->entryp->ace_len;
 		acl_entry->ace_access = acl_entry_link->entryp->ace_access;
- 
+
 		/* In order to use this, we'll need to wait until we can get denies */
 		/* if (!acl_entry->ace_access && acl_entry->ace_type == ACC_PERMIT)
 			acl_entry->ace_type = ACC_SPECIFY; */
- 
+
 		acl_entry->ace_type = ACC_SPECIFY;
- 
+
 		ace_id = acl_entry->ace_id;
- 
+
 		ace_id->id_type = acl_entry_link->entryp->ace_id->id_type;
 		DEBUG(10, ("The id type is %d\n", ace_id->id_type));
 		ace_id->id_len = acl_entry_link->entryp->ace_id->id_len;
 		memcpy(&user_id, acl_entry_link->entryp->ace_id->id_data, sizeof (uid_t));
 		memcpy(ace_id->id_data, &user_id, sizeof (uid_t));
 	}
- 
+
 	rc = fchacl(fd, file_acl, file_acl->acl_len);
 	DEBUG(10, ("errno is %d\n", errno));
 	DEBUG(10, ("return code is %d\n", rc));
@@ -2590,7 +2594,7 @@ int sys_acl_free_acl(SMB_ACL_T posix_acl)
 	SAFE_FREE(acl_entry_link->prevp);
 	SAFE_FREE(acl_entry_link->entryp);
 	SAFE_FREE(acl_entry_link);
- 
+
 	return 0;
 }
 

lib/sysacls.h

@@ -3,7 +3,7 @@
  * Version 2.2.x
  * Portable SMB ACL interface
  * Copyright (C) Jeremy Allison 2000
- * Copyright (C) 2007-2020 Wayne Davison
+ * Copyright (C) 2007-2022 Wayne Davison
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -232,7 +232,7 @@ struct new_acl_entry{
 
 #define SMB_ACL_ENTRY_T		struct new_acl_entry*
 #define SMB_ACL_T		struct acl_entry_link*
- 
+
 #define SMB_ACL_TAG_T		unsigned short
 #define SMB_ACL_TYPE_T		int
 

lib/sysxattrs.c

@@ -2,7 +2,7 @@
  * Extended attribute support for rsync.
  *
  * Copyright (C) 2004 Red Hat, Inc.
- * Copyright (C) 2003-2019 Wayne Davison
+ * Copyright (C) 2003-2022 Wayne Davison
  * Written by Jay Fenlason.
  *
  * This program is free software; you can redistribute it and/or modify
@@ -67,7 +67,7 @@ ssize_t sys_lgetxattr(const char *path, const char *name, void *value, size_t si
 		u_int32_t offset = len;
 		size_t data_retrieved = len;
 		while (data_retrieved < size) {
-			len = getxattr(path, name, value + offset, size - data_retrieved, offset, XATTR_NOFOLLOW);
+			len = getxattr(path, name, (char*)value + offset, size - data_retrieved, offset, XATTR_NOFOLLOW);
 			if (len <= 0)
 				break;
 			data_retrieved += len;
@@ -167,7 +167,7 @@ static ssize_t read_xattr(int attrfd, void *buf, size_t buflen)
 	} else {
 		size_t bufpos;
 		for (bufpos = 0; bufpos < sb.st_size; ) {
-			ssize_t cnt = read(attrfd, buf + bufpos, sb.st_size - bufpos);
+			ssize_t cnt = read(attrfd, (char*)buf + bufpos, sb.st_size - bufpos);
 			if (cnt <= 0) {
 				if (cnt < 0 && errno == EINTR)
 					continue;
@@ -218,7 +218,7 @@ int sys_lsetxattr(const char *path, const char *name, const void *value, size_t
 		return -1;
 
 	for (bufpos = 0; bufpos < size; ) {
-		ssize_t cnt = write(attrfd, value+bufpos, size);
+		ssize_t cnt = write(attrfd, (char*)value + bufpos, size);
 		if (cnt <= 0) {
 			if (cnt < 0 && errno == EINTR)
 				continue;
@@ -274,7 +274,8 @@ ssize_t sys_llistxattr(const char *path, char *list, size_t size)
 		 && (dp->d_name[10] == 'o' || dp->d_name[10] == 'w'))
 			continue;
 
-		if ((ret += len+1) > size) {
+		ret += len + 1;
+		if ((size_t)ret > size) {
 			if (size == 0)
 				continue;
 			ret = -1;

loadparm.c

@@ -437,7 +437,7 @@ static BOOL do_parameter(char *parmname, char *parmvalue)
 		break;
 
 	case P_OCTAL:
-		sscanf(parmvalue, "%o", (int *)parm_ptr);
+		sscanf(parmvalue, "%o", (unsigned int *)parm_ptr);
 		break;
 
 	case P_PATH:

log.c

@@ -3,7 +3,7 @@
  *
  * Copyright (C) 1998-2001 Andrew Tridgell <tridge@samba.org>
  * Copyright (C) 2000-2001 Martin Pool <mbp@samba.org>
- * Copyright (C) 2003-2020 Wayne Davison
+ * Copyright (C) 2003-2022 Wayne Davison
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -34,10 +34,8 @@ extern int module_id;
 extern int allow_8bit_chars;
 extern int protocol_version;
 extern int always_checksum;
-extern int preserve_times;
+extern int preserve_mtimes;
 extern int msgs2stderr;
-extern int xfersum_type;
-extern int checksum_type;
 extern int stdout_format_has_i;
 extern int stdout_format_has_o_or_i;
 extern int logfile_format_has_i;
@@ -62,6 +60,8 @@ extern unsigned int module_dirlen;
 extern char sender_file_sum[MAX_DIGEST_LEN];
 extern const char undetermined_hostname[];
 
+extern struct name_num_item *xfer_sum_nni, *file_sum_nni;
+
 static int log_initialised;
 static int logfile_was_closed;
 static FILE *logfile_fp;
@@ -680,12 +680,12 @@ static void log_formatted(enum logcode code, const char *format, const char *op,
 			n = NULL;
 			if (S_ISREG(file->mode)) {
 				if (always_checksum)
-					n = sum_as_hex(checksum_type, F_SUM(file), 1);
+					n = sum_as_hex(file_sum_nni->num, F_SUM(file), 1);
 				else if (iflags & ITEM_TRANSFER)
-					n = sum_as_hex(xfersum_type, sender_file_sum, 0);
+					n = sum_as_hex(xfer_sum_nni->num, sender_file_sum, 0);
 			}
 			if (!n) {
-				int sum_len = csum_len_for_type(always_checksum ? checksum_type : xfersum_type,
+				int sum_len = csum_len_for_type(always_checksum ? file_sum_nni->num : xfer_sum_nni->num,
 								always_checksum);
 				memset(buf2, ' ', sum_len*2);
 				buf2[sum_len*2] = '\0';
@@ -706,7 +706,7 @@ static void log_formatted(enum logcode code, const char *format, const char *op,
 				c[1] = 'L';
 				c[3] = '.';
 				c[4] = !(iflags & ITEM_REPORT_TIME) ? '.'
-				     : !preserve_times || !receiver_symlink_times
+				     : !preserve_mtimes || !receiver_symlink_times
 				    || (iflags & ITEM_REPORT_TIMEFAIL) ? 'T' : 't';
 			} else {
 				c[1] = S_ISDIR(file->mode) ? 'd'
@@ -714,7 +714,7 @@ static void log_formatted(enum logcode code, const char *format, const char *op,
 				     : IS_DEVICE(file->mode) ? 'D' : 'f';
 				c[3] = !(iflags & ITEM_REPORT_SIZE) ? '.' : 's';
 				c[4] = !(iflags & ITEM_REPORT_TIME) ? '.'
-				     : !preserve_times ? 'T' : 't';
+				     : !preserve_mtimes ? 'T' : 't';
 			}
 			c[2] = !(iflags & ITEM_REPORT_CHANGE) ? '.' : 'c';
 			c[5] = !(iflags & ITEM_REPORT_PERMS) ? '.' : 'p';
@@ -838,14 +838,24 @@ void maybe_log_item(struct file_struct *file, int iflags, int itemizing, const c
 
 void log_delete(const char *fname, int mode)
 {
-	static struct {
-		union file_extras ex[4]; /* just in case... */
-		struct file_struct file;
-	} x; /* Zero-initialized due to static declaration. */
+	static struct file_struct *file = NULL;
 	int len = strlen(fname);
 	const char *fmt;
 
-	x.file.mode = mode;
+	if (!file) {
+		int extra_len = (file_extra_cnt + 2) * EXTRA_LEN;
+		char *bp;
+#if EXTRA_ROUNDING > 0
+		if (extra_len & (EXTRA_ROUNDING * EXTRA_LEN))
+			extra_len = (extra_len | (EXTRA_ROUNDING * EXTRA_LEN)) + EXTRA_LEN;
+#endif
+
+		bp = new_array0(char, FILE_STRUCT_LEN + extra_len + 1);
+		bp += extra_len;
+		file = (struct file_struct *)bp;
+	}
+
+	file->mode = mode;
 
 	if (am_server && protocol_version >= 29 && len < MAXPATHLEN) {
 		if (S_ISDIR(mode))
@@ -855,14 +865,14 @@ void log_delete(const char *fname, int mode)
 		;
 	else {
 		fmt = stdout_format_has_o_or_i ? stdout_format : "deleting %n";
-		log_formatted(FCLIENT, fmt, "del.", &x.file, fname, ITEM_DELETED, NULL);
+		log_formatted(FCLIENT, fmt, "del.", file, fname, ITEM_DELETED, NULL);
 	}
 
 	if (!logfile_name || dry_run || !logfile_format)
 		return;
 
 	fmt = logfile_format_has_o_or_i ? logfile_format : "deleting %n";
-	log_formatted(FLOG, fmt, "del.", &x.file, fname, ITEM_DELETED, NULL);
+	log_formatted(FLOG, fmt, "del.", file, fname, ITEM_DELETED, NULL);
 }
 
 /*
@@ -891,10 +901,10 @@ void log_exit(int code, const char *file, int line)
 		/* VANISHED is not an error, only a warning */
 		if (code == RERR_VANISHED) {
 			rprintf(FWARNING, "rsync warning: %s (code %d) at %s(%d) [%s=%s]\n",
-				name, code, src_file(file), line, who_am_i(), RSYNC_VERSION);
+				name, code, src_file(file), line, who_am_i(), rsync_version());
 		} else {
 			rprintf(FERROR, "rsync error: %s (code %d) at %s(%d) [%s=%s]\n",
-				name, code, src_file(file), line, who_am_i(), RSYNC_VERSION);
+				name, code, src_file(file), line, who_am_i(), rsync_version());
 		}
 	}
 }

m4/have_type.m4

@@ -1,6 +1,5 @@
 dnl AC_HAVE_TYPE(TYPE,INCLUDES)
 AC_DEFUN([AC_HAVE_TYPE], [
-AC_REQUIRE([AC_HEADER_STDC])
 cv=`echo "$1" | sed 'y%./+- %__p__%'`
 AC_MSG_CHECKING(for $1)
 AC_CACHE_VAL([ac_cv_type_$cv],

main.c

@@ -4,7 +4,7 @@
  * Copyright (C) 1996-2001 Andrew Tridgell <tridge@samba.org>
  * Copyright (C) 1996 Paul Mackerras
  * Copyright (C) 2001, 2002 Martin Pool <mbp@samba.org>
- * Copyright (C) 2003-2020 Wayne Davison
+ * Copyright (C) 2003-2022 Wayne Davison
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -48,6 +48,7 @@ extern int called_from_signal_handler;
 extern int need_messages_from_generator;
 extern int kluge_around_eof;
 extern int got_xfer_error;
+extern int old_style_args;
 extern int msgs2stderr;
 extern int module_id;
 extern int read_only;
@@ -87,6 +88,9 @@ extern BOOL shutting_down;
 extern int backup_dir_len;
 extern int basis_dir_cnt;
 extern int default_af_hint;
+extern int stdout_format_has_i;
+extern int trust_sender_filter;
+extern int trust_sender_args;
 extern struct stats stats;
 extern char *stdout_format;
 extern char *logfile_format;
@@ -102,7 +106,7 @@ extern char curr_dir[MAXPATHLEN];
 extern char backup_dir_buf[MAXPATHLEN];
 extern char *basis_dir[MAX_BASIS_DIRS+1];
 extern struct file_list *first_flist;
-extern filter_rule_list daemon_filter_list;
+extern filter_rule_list daemon_filter_list, implied_filter_list;
 
 uid_t our_uid;
 gid_t our_gid;
@@ -466,38 +470,33 @@ static void output_summary(void)
  **/
 static void show_malloc_stats(void)
 {
-#ifdef HAVE_MALLINFO
-	struct mallinfo mi;
-
-	mi = mallinfo();
+#ifdef MEM_ALLOC_INFO
+	struct MEM_ALLOC_INFO mi = MEM_ALLOC_INFO(); /* mallinfo or mallinfo2 */
 
 	rprintf(FCLIENT, "\n");
 	rprintf(FINFO, RSYNC_NAME "[%d] (%s%s%s) heap statistics:\n",
 		(int)getpid(), am_server ? "server " : "",
 		am_daemon ? "daemon " : "", who_am_i());
-	rprintf(FINFO, "  arena:     %10ld   (bytes from sbrk)\n",
-		(long)mi.arena);
-	rprintf(FINFO, "  ordblks:   %10ld   (chunks not in use)\n",
-		(long)mi.ordblks);
-	rprintf(FINFO, "  smblks:    %10ld\n",
-		(long)mi.smblks);
-	rprintf(FINFO, "  hblks:     %10ld   (chunks from mmap)\n",
-		(long)mi.hblks);
-	rprintf(FINFO, "  hblkhd:    %10ld   (bytes from mmap)\n",
-		(long)mi.hblkhd);
-	rprintf(FINFO, "  allmem:    %10ld   (bytes from sbrk + mmap)\n",
-		(long)mi.arena + mi.hblkhd);
-	rprintf(FINFO, "  usmblks:   %10ld\n",
-		(long)mi.usmblks);
-	rprintf(FINFO, "  fsmblks:   %10ld\n",
-		(long)mi.fsmblks);
-	rprintf(FINFO, "  uordblks:  %10ld   (bytes used)\n",
-		(long)mi.uordblks);
-	rprintf(FINFO, "  fordblks:  %10ld   (bytes free)\n",
-		(long)mi.fordblks);
-	rprintf(FINFO, "  keepcost:  %10ld   (bytes in releasable chunk)\n",
-		(long)mi.keepcost);
-#endif /* HAVE_MALLINFO */
+
+#define PRINT_ALLOC_NUM(title, descr, num) \
+	rprintf(FINFO, "  %-11s%10" SIZE_T_FMT_MOD "d   (" descr ")\n", \
+		title ":", (SIZE_T_FMT_CAST)(num));
+
+	PRINT_ALLOC_NUM("arena", "bytes from sbrk", mi.arena);
+	PRINT_ALLOC_NUM("ordblks", "chunks not in use", mi.ordblks);
+	PRINT_ALLOC_NUM("smblks", "free fastbin blocks", mi.smblks);
+	PRINT_ALLOC_NUM("hblks", "chunks from mmap", mi.hblks);
+	PRINT_ALLOC_NUM("hblkhd", "bytes from mmap", mi.hblkhd);
+	PRINT_ALLOC_NUM("allmem", "bytes from sbrk + mmap", mi.arena + mi.hblkhd);
+	PRINT_ALLOC_NUM("usmblks", "always 0", mi.usmblks);
+	PRINT_ALLOC_NUM("fsmblks", "bytes in freed fastbin blocks", mi.fsmblks);
+	PRINT_ALLOC_NUM("uordblks", "bytes used", mi.uordblks);
+	PRINT_ALLOC_NUM("fordblks", "bytes free", mi.fordblks);
+	PRINT_ALLOC_NUM("keepcost", "bytes in releasable chunk", mi.keepcost);
+
+#undef PRINT_ALLOC_NUM
+
+#endif /* MEM_ALLOC_INFO */
 }
 
 
@@ -611,11 +610,7 @@ static pid_t do_cmd(char *cmd, char *machine, char *user, char **remote_argv, in
 				rprintf(FERROR, "internal: args[] overflowed in do_cmd()\n");
 				exit_cleanup(RERR_SYNTAX);
 			}
-			if (**remote_argv == '-') {
-				if (asprintf(args + argc++, "./%s", *remote_argv++) < 0)
-					out_of_memory("do_cmd");
-			} else
-				args[argc++] = *remote_argv++;
+			args[argc++] = safe_arg(NULL, *remote_argv++);
 			remote_argc--;
 		}
 	}
@@ -667,6 +662,16 @@ static pid_t do_cmd(char *cmd, char *machine, char *user, char **remote_argv, in
 	return pid;
 }
 
+/* Older versions turn an empty string as a reference to the current directory.
+ * We now treat this as an error unless --old-args was used. */
+static char *dot_dir_or_error()
+{
+	if (old_style_args || am_server)
+		return ".";
+	rprintf(FERROR, "Empty destination arg specified (use \".\" or see --old-args).\n");
+	exit_cleanup(RERR_SYNTAX);
+}
+
 /* The receiving side operates in one of two modes:
  *
  * 1. it receives any number of files into a destination directory,
@@ -694,9 +699,8 @@ static char *get_local_name(struct file_list *flist, char *dest_path)
 	if (!dest_path || list_only)
 		return NULL;
 
-	/* Treat an empty string as a copy into the current directory. */
 	if (!*dest_path)
-		dest_path = ".";
+		dest_path = dot_dir_or_error();
 
 	if (daemon_filter_list.head) {
 		char *slash = strrchr(dest_path, '/');
@@ -721,18 +725,21 @@ static char *get_local_name(struct file_list *flist, char *dest_path)
 	trailing_slash = cp && !cp[1];
 
 	if (mkpath_dest_arg && statret < 0 && (cp || file_total > 1)) {
+		int save_errno = errno;
 		int ret = make_path(dest_path, file_total > 1 && !trailing_slash ? 0 : MKP_DROP_NAME);
 		if (ret < 0)
 			goto mkdir_error;
-		if (INFO_GTE(NAME, 1)) {
+		if (ret && (INFO_GTE(NAME, 1) || stdout_format_has_i)) {
 			if (file_total == 1 || trailing_slash)
 				*cp = '\0';
 			rprintf(FINFO, "created %d director%s for %s\n", ret, ret == 1 ? "y" : "ies", dest_path);
 			if (file_total == 1 || trailing_slash)
 				*cp = '/';
 		}
-		if (file_total > 1 || trailing_slash)
+		if (ret)
 			statret = do_stat(dest_path, &st);
+		else
+			errno = save_errno;
 	}
 
 	if (statret == 0) {
@@ -788,7 +795,7 @@ static char *get_local_name(struct file_list *flist, char *dest_path)
 		 && strcmp(flist->files[flist->low]->basename, ".") == 0)
 			flist->files[0]->flags |= FLAG_DIR_CREATED;
 
-		if (INFO_GTE(NAME, 1))
+		if (INFO_GTE(NAME, 1) || stdout_format_has_i)
 			rprintf(FINFO, "created directory %s\n", dest_path);
 
 		if (dry_run) {
@@ -1080,6 +1087,7 @@ static int do_recv(int f_in, int f_out, char *local_name)
 	}
 
 	am_generator = 1;
+	implied_filter_list.head = implied_filter_list.tail = NULL;
 	flist_receiving_enabled = True;
 
 	io_end_multiplex_in(MPLX_SWITCHING);
@@ -1375,15 +1383,6 @@ int client_run(int f_in, int f_out, pid_t pid, int argc, char *argv[])
 	return MAX(exit_code, exit_code2);
 }
 
-static void dup_argv(char *argv[])
-{
-	int i;
-
-	for (i = 0; argv[i]; i++)
-		argv[i] = strdup(argv[i]);
-}
-
-
 /* Start a client for either type of remote connection.  Work out
  * whether the arguments request a remote shell or rsyncd connection,
  * and call the appropriate connection function, then run_client.
@@ -1399,10 +1398,6 @@ static int start_client(int argc, char *argv[])
 	int ret;
 	pid_t pid;
 
-	/* Don't clobber argv[] so that ps(1) can still show the right
-	 * command line. */
-	dup_argv(argv);
-
 	if (!read_batch) { /* for read_batch, NO source is specified */
 		char *path = check_for_hostspec(argv[0], &shell_machine, &rsync_port);
 		if (path) { /* source is remote */
@@ -1435,6 +1430,8 @@ static int start_client(int argc, char *argv[])
 
 			if (argc > 1) {
 				p = argv[--argc];
+				if (!*p)
+					p = dot_dir_or_error();
 				remote_argv = argv + argc;
 			} else {
 				static char *dotarg[1] = { "." };
@@ -1475,6 +1472,12 @@ static int start_client(int argc, char *argv[])
 		rsync_port = 0;
 	}
 
+	/* A local transfer doesn't unbackslash anything, so leave the args alone. */
+	if (local_server) {
+		old_style_args = 2;
+		trust_sender_args = trust_sender_filter = 1;
+	}
+
 	if (!rsync_port && remote_argc && !**remote_argv) /* Turn an empty arg into a dot dir. */
 		*remote_argv = ".";
 
@@ -1500,6 +1503,8 @@ static int start_client(int argc, char *argv[])
 		char *dummy_host;
 		int dummy_port = rsync_port;
 		int i;
+		if (filesfrom_fd < 0)
+			add_implied_include(remote_argv[0], daemon_connection);
 		/* For remote source, any extra source args must have either
 		 * the same hostname or an empty hostname. */
 		for (i = 1; i < remote_argc; i++) {
@@ -1523,6 +1528,7 @@ static int start_client(int argc, char *argv[])
 			if (!rsync_port && !*arg) /* Turn an empty arg into a dot dir. */
 				arg = ".";
 			remote_argv[i] = arg;
+			add_implied_include(arg, daemon_connection);
 		}
 	}
 
@@ -1564,6 +1570,8 @@ static int start_client(int argc, char *argv[])
 #ifdef HAVE_PUTENV
 	if (daemon_connection)
 		set_env_num("RSYNC_PORT", env_port);
+#else
+	(void)env_port;
 #endif
 
 	pid = do_cmd(shell_cmd, shell_machine, shell_user, remote_argv, remote_argc, &f_in, &f_out);
@@ -1636,7 +1644,6 @@ void remember_children(UNUSED(int val))
 #endif
 }
 
-
 /**
  * This routine catches signals and tries to send them to gdb.
  *
@@ -1660,7 +1667,6 @@ const char *get_panic_action(void)
 	return "xterm -display :0 -T Panic -n Panic -e gdb /proc/%d/exe %d";
 }
 
-
 /**
  * Handle a fatal signal by launching a debugger, controlled by $RSYNC_PANIC_ACTION.
  *
@@ -1684,6 +1690,22 @@ static void rsync_panic_handler(UNUSED(int whatsig))
 }
 #endif
 
+static void unset_env_var(const char *var)
+{
+#ifdef HAVE_UNSETENV
+	unsetenv(var);
+#else
+#ifdef HAVE_PUTENV
+	char *mem;
+	if (asprintf(&mem, "%s=", var) < 0)
+		out_of_memory("unset_env_var");
+	putenv(mem);
+#else
+	(void)var;
+#endif
+#endif
+}
+
 
 int main(int argc,char *argv[])
 {
@@ -1721,6 +1743,19 @@ int main(int argc,char *argv[])
 	our_gid = MY_GID();
 	am_root = our_uid == ROOT_UID;
 
+	unset_env_var("DISPLAY");
+
+#if defined USE_OPENSSL && defined SET_OPENSSL_CONF
+#define TO_STR2(x) #x
+#define TO_STR(x) TO_STR2(x)
+	/* ./configure --with-openssl-conf=/etc/ssl/openssl-rsync.cnf
+	 * defines SET_OPENSSL_CONF as that unquoted pathname. */
+	if (!getenv("OPENSSL_CONF")) /* Don't override it if it's already set. */
+		set_env_str("OPENSSL_CONF", TO_STR(SET_OPENSSL_CONF));
+#undef TO_STR
+#undef TO_STR2
+#endif
+
 	memset(&stats, 0, sizeof(stats));
 
 	/* Even a non-daemon runs needs the default config values to be set, e.g.
@@ -1739,6 +1774,7 @@ int main(int argc,char *argv[])
 
 #if defined CONFIG_LOCALE && defined HAVE_SETLOCALE
 	setlocale(LC_CTYPE, "");
+	setlocale(LC_NUMERIC, "");
 #endif
 
 	if (!parse_arguments(&argc, (const char ***) &argv)) {

match.c

@@ -3,7 +3,7 @@
  *
  * Copyright (C) 1996 Andrew Tridgell
  * Copyright (C) 1996 Paul Mackerras
- * Copyright (C) 2003-2020 Wayne Davison
+ * Copyright (C) 2003-2023 Wayne Davison
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -24,7 +24,9 @@
 
 extern int checksum_seed;
 extern int append_mode;
-extern int xfersum_type;
+
+extern struct name_num_item *xfer_sum_nni;
+extern int xfer_sum_len;
 
 int updating_basis_file;
 char sender_file_sum[MAX_DIGEST_LEN];
@@ -140,7 +142,7 @@ static void hash_search(int f,struct sum_struct *s,
 {
 	OFF_T offset, aligned_offset, end;
 	int32 k, want_i, aligned_i, backup;
-	char sum2[SUM_LENGTH];
+	char sum2[MAX_DIGEST_LEN];
 	uint32 s1, s2, sum;
 	int more;
 	schar *map;
@@ -356,15 +358,13 @@ static void hash_search(int f,struct sum_struct *s,
  **/
 void match_sums(int f, struct sum_struct *s, struct map_struct *buf, OFF_T len)
 {
-	int sum_len;
-
 	last_match = 0;
 	false_alarms = 0;
 	hash_hits = 0;
 	matches = 0;
 	data_transfer = 0;
 
-	sum_init(xfersum_type, checksum_seed);
+	sum_init(xfer_sum_nni, checksum_seed);
 
 	if (append_mode > 0) {
 		if (append_mode == 2) {
@@ -405,22 +405,22 @@ void match_sums(int f, struct sum_struct *s, struct map_struct *buf, OFF_T len)
 		matched(f, s, buf, len, -1);
 	}
 
-	sum_len = sum_end(sender_file_sum);
+	sum_end(sender_file_sum);
 
 	/* If we had a read error, send a bad checksum.  We use all bits
 	 * off as long as the checksum doesn't happen to be that, in
 	 * which case we turn the last 0 bit into a 1. */
 	if (buf && buf->status != 0) {
 		int i;
-		for (i = 0; i < sum_len && sender_file_sum[i] == 0; i++) {}
-		memset(sender_file_sum, 0, sum_len);
-		if (i == sum_len)
+		for (i = 0; i < xfer_sum_len && sender_file_sum[i] == 0; i++) {}
+		memset(sender_file_sum, 0, xfer_sum_len);
+		if (i == xfer_sum_len)
 			sender_file_sum[i-1]++;
 	}
 
 	if (DEBUG_GTE(DELTASUM, 2))
 		rprintf(FINFO,"sending file_sum\n");
-	write_buf(f, sender_file_sum, sum_len);
+	write_buf(f, sender_file_sum, xfer_sum_len);
 
 	if (DEBUG_GTE(DELTASUM, 2)) {
 		rprintf(FINFO, "false_alarms=%d hash_hits=%d matches=%d\n",

maybe-make-man

@@ -1,22 +1,18 @@
 #!/bin/sh
 
-if [ x"$2" = x ]; then
-    echo "Usage: $0 SRC_DIR NAME.NUM.md" 1>&2
+if [ $# != 1 ]; then
+    echo "Usage: $0 NAME.NUM.md" 1>&2
     exit 1
 fi
 
-srcdir="$1"
-inname="$2"
+inname="$1"
+srcdir=`dirname "$0"`
 flagfile="$srcdir/.md2man-works"
-
-if [ ! -d "$srcdir" ]; then
-    echo "The specified SRC_DIR is not a directory: $srcdir" 1>&2
-    exit 1
-fi
+force_flagfile="$srcdir/.md2man-force"
 
 if [ ! -f "$flagfile" ]; then
     # We test our smallest manpage just to see if the python setup works.
-    if "$srcdir/md2man" --test "$srcdir/rsync-ssl.1.md" >/dev/null 2>&1; then
+    if "$srcdir/md-convert" --test "$srcdir/rsync-ssl.1.md" >/dev/null 2>&1; then
 	touch $flagfile
     else
 	outname=`echo "$inname" | sed 's/\.md$//'`
@@ -37,4 +33,10 @@ if [ ! -f "$flagfile" ]; then
     fi
 fi
 
-"$srcdir/md2man" "$srcdir/$inname"
+if [ -f "$force_flagfile" ]; then
+    opt='--force-link-text'
+else
+    opt=''
+fi
+
+"$srcdir/md-convert" $opt "$srcdir/$inname"

md-convert

@@ -0,0 +1,662 @@
+#!/usr/bin/env python3
+
+# This script transforms markdown files into html and (optionally) nroff. The
+# output files are written into the current directory named for the input file
+# without the .md suffix and either the .html suffix or no suffix.
+#
+# If the input .md file has a section number at the end of the name (e.g.,
+# rsync.1.md) a nroff file is also output (PROJ.NUM.md -> PROJ.NUM).
+#
+# The markdown input format has one extra extension: if a numbered list starts
+# at 0, it is turned into a description list. The dl's dt tag is taken from the
+# contents of the first tag inside the li, which is usually a p, code, or
+# strong tag.
+#
+# The cmarkgfm or commonmark lib is used to transforms the input file into
+# html.  Then, the html.parser is used as a state machine that lets us tweak
+# the html and (optionally) output nroff data based on the html tags.
+#
+# If the string @USE_GFM_PARSER@ exists in the file, the string is removed and
+# a github-flavored-markup parser is used to parse the file.
+#
+# The man-page .md files also get the vars @VERSION@, @BINDIR@, and @LIBDIR@
+# substituted.  Some of these values depend on the Makefile $(prefix) (see the
+# generated Makefile).  If the maintainer wants to build files for /usr/local
+# while creating release-ready man-page files for /usr, use the environment to
+# set RSYNC_OVERRIDE_PREFIX=/usr.
+
+# Copyright (C) 2020 - 2021 Wayne Davison
+#
+# This program is freely redistributable.
+
+import os, sys, re, argparse, subprocess, time
+from html.parser import HTMLParser
+
+VALID_PAGES = 'README INSTALL COPYING rsync.1 rrsync.1 rsync-ssl.1 rsyncd.conf.5'.split()
+
+CONSUMES_TXT = set('h1 h2 h3 p li pre'.split())
+
+HTML_START = """\
+<html><head>
+<title>%TITLE%</title>
+<meta charset="UTF-8"/>
+<link href="https://fonts.googleapis.com/css2?family=Roboto&family=Roboto+Mono&display=swap" rel="stylesheet">
+<style>
+body {
+  max-width: 50em;
+  margin: auto;
+}
+body, b, strong, u {
+  font-family: 'Roboto', sans-serif;
+}
+a.tgt { font-face: symbol; font-weight: 400; font-size: 70%; visibility: hidden; text-decoration: none; color: #ddd; padding: 0 4px; border: 0; }
+a.tgt:after { content: '🔗'; }
+a.tgt:hover { color: #444; background-color: #eaeaea; }
+h1:hover > a.tgt, h2:hover > a.tgt, h3:hover > a.tgt, dt:hover > a.tgt { visibility: visible; }
+code {
+  font-family: 'Roboto Mono', monospace;
+  font-weight: bold;
+  white-space: pre;
+}
+pre code {
+  display: block;
+  font-weight: normal;
+}
+blockquote pre code {
+  background: #f1f1f1;
+}
+dd p:first-of-type {
+  margin-block-start: 0em;
+}
+</style>
+</head><body>
+"""
+
+TABLE_STYLE = """\
+table {
+  border-color: grey;
+  border-spacing: 0;
+}
+tr {
+  border-top: 1px solid grey;
+}
+tr:nth-child(2n) {
+  background-color: #f6f8fa;
+}
+th, td {
+  border: 1px solid #dfe2e5;
+  text-align: center;
+  padding-left: 1em;
+  padding-right: 1em;
+}
+"""
+
+MAN_HTML_END = """\
+<div style="float: right"><p><i>%s</i></p></div>
+"""
+
+HTML_END = """\
+</body></html>
+"""
+
+MAN_START = r"""
+.TH "%s" "%s" "%s" "%s" "User Commands"
+.\" prefix=%s
+""".lstrip()
+
+MAN_END = """\
+"""
+
+NORM_FONT = ('\1', r"\fP")
+BOLD_FONT = ('\2', r"\fB")
+UNDR_FONT = ('\3', r"\fI")
+NBR_DASH = ('\4', r"\-")
+NBR_SPACE = ('\xa0', r"\ ")
+
+FILENAME_RE = re.compile(r'^(?P<fn>(?P<srcdir>.+/)?(?P<name>(?P<prog>[^/]+?)(\.(?P<sect>\d+))?)\.md)$')
+ASSIGNMENT_RE = re.compile(r'^(\w+)=(.+)')
+VER_RE = re.compile(r'^#define\s+RSYNC_VERSION\s+"(\d.+?)"', re.M)
+TZ_RE = re.compile(r'^#define\s+MAINTAINER_TZ_OFFSET\s+(-?\d+(\.\d+)?)', re.M)
+VAR_REF_RE = re.compile(r'\$\{(\w+)\}')
+VERSION_RE = re.compile(r' (\d[.\d]+)[, ]')
+BIN_CHARS_RE = re.compile(r'[\1-\7]+')
+SPACE_DOUBLE_DASH_RE = re.compile(r'\s--(\s)')
+NON_SPACE_SINGLE_DASH_RE = re.compile(r'(^|\W)-')
+WHITESPACE_RE = re.compile(r'\s')
+CODE_BLOCK_RE = re.compile(r'[%s]([^=%s]+)[=%s]' % (BOLD_FONT[0], NORM_FONT[0], NORM_FONT[0]))
+NBR_DASH_RE = re.compile(r'[%s]' % NBR_DASH[0])
+INVALID_TARGET_CHARS_RE = re.compile(r'[^-A-Za-z0-9._]')
+INVALID_START_CHAR_RE = re.compile(r'^([^A-Za-z0-9])')
+MANIFY_LINESTART_RE = re.compile(r"^(['.])", flags=re.M)
+
+md_parser = None
+env_subs = { }
+
+warning_count = 0
+
+def main():
+    for mdfn in args.mdfiles:
+        parse_md_file(mdfn)
+
+    if args.test:
+        print("The test was successful.")
+
+
+def parse_md_file(mdfn):
+    fi = FILENAME_RE.match(mdfn)
+    if not fi:
+        die('Failed to parse a md input file name:', mdfn)
+    fi = argparse.Namespace(**fi.groupdict())
+    fi.want_manpage = not not fi.sect
+    if fi.want_manpage:
+        fi.title = fi.prog + '(' + fi.sect + ') manpage'
+    else:
+        fi.title = fi.prog + ' for rsync'
+
+    if fi.want_manpage:
+        if not env_subs:
+            find_man_substitutions()
+        prog_ver = 'rsync ' + env_subs['VERSION']
+        if fi.prog != 'rsync':
+            prog_ver = fi.prog + ' from ' + prog_ver
+        fi.man_headings = (fi.prog, fi.sect, env_subs['date'], prog_ver, env_subs['prefix'])
+
+    with open(mdfn, 'r', encoding='utf-8') as fh:
+        txt = fh.read()
+
+    use_gfm_parser = '@USE_GFM_PARSER@' in txt
+    if use_gfm_parser:
+        txt = txt.replace('@USE_GFM_PARSER@', '')
+
+    if fi.want_manpage:
+        txt = (txt.replace('@VERSION@', env_subs['VERSION'])
+                  .replace('@BINDIR@', env_subs['bindir'])
+                  .replace('@LIBDIR@', env_subs['libdir']))
+
+    if use_gfm_parser:
+        if not gfm_parser:
+            die('Input file requires cmarkgfm parser:', mdfn)
+        fi.html_in = gfm_parser(txt)
+    else:
+        fi.html_in = md_parser(txt)
+    txt = None
+
+    TransformHtml(fi)
+
+    if args.test:
+        return
+
+    output_list = [ (fi.name + '.html', fi.html_out) ]
+    if fi.want_manpage:
+        output_list += [ (fi.name, fi.man_out) ]
+    for fn, txt in output_list:
+        if args.dest and args.dest != '.':
+            fn = os.path.join(args.dest, fn)
+        if os.path.lexists(fn):
+            os.unlink(fn)
+        print("Wrote:", fn)
+        with open(fn, 'w', encoding='utf-8') as fh:
+            fh.write(txt)
+
+
+def find_man_substitutions():
+    srcdir = os.path.dirname(sys.argv[0]) + '/'
+    mtime = 0
+
+    git_dir = srcdir + '.git'
+    if os.path.lexists(git_dir):
+        mtime = int(subprocess.check_output(['git', '--git-dir', git_dir, 'log', '-1', '--format=%at']))
+
+    # Allow "prefix" to be overridden via the environment:
+    env_subs['prefix'] = os.environ.get('RSYNC_OVERRIDE_PREFIX', None)
+
+    if args.test:
+        env_subs['VERSION'] = '1.0.0'
+        env_subs['bindir'] = '/usr/bin'
+        env_subs['libdir'] = '/usr/lib/rsync'
+        tz_offset = 0
+    else:
+        for fn in (srcdir + 'version.h', 'Makefile'):
+            try:
+                st = os.lstat(fn)
+            except OSError:
+                die('Failed to find', srcdir + fn)
+            if not mtime:
+                mtime = st.st_mtime
+
+        with open(srcdir + 'version.h', 'r', encoding='utf-8') as fh:
+            txt = fh.read()
+        m = VER_RE.search(txt)
+        env_subs['VERSION'] = m.group(1)
+        m = TZ_RE.search(txt) # the tzdata lib may not be installed, so we use a simple hour offset
+        tz_offset = float(m.group(1)) * 60 * 60
+
+        with open('Makefile', 'r', encoding='utf-8') as fh:
+            for line in fh:
+                m = ASSIGNMENT_RE.match(line)
+                if not m:
+                    continue
+                var, val = (m.group(1), m.group(2))
+                if var == 'prefix' and env_subs[var] is not None:
+                    continue
+                while VAR_REF_RE.search(val):
+                    val = VAR_REF_RE.sub(lambda m: env_subs[m.group(1)], val)
+                env_subs[var] = val
+                if var == 'srcdir':
+                    break
+
+    env_subs['date'] = time.strftime('%d %b %Y', time.gmtime(mtime + tz_offset)).lstrip('0')
+
+
+def html_via_commonmark(txt):
+    return commonmark.HtmlRenderer().render(commonmark.Parser().parse(txt))
+
+
+class TransformHtml(HTMLParser):
+    def __init__(self, fi):
+        HTMLParser.__init__(self, convert_charrefs=True)
+
+        self.fn = fi.fn
+
+        st = self.state = argparse.Namespace(
+                list_state = [ ],
+                p_macro = ".P\n",
+                at_first_tag_in_li = False,
+                at_first_tag_in_dd = False,
+                dt_from = None,
+                in_pre = False,
+                in_code = False,
+                html_out = [ HTML_START.replace('%TITLE%', fi.title) ],
+                man_out = [ ],
+                txt = '',
+                want_manpage = fi.want_manpage,
+                created_hashtags = set(),
+                derived_hashtags = set(),
+                referenced_hashtags = set(),
+                bad_hashtags = set(),
+                latest_targets = [ ],
+                opt_prefix = 'opt',
+                a_href = None,
+                a_href_external = False,
+                a_txt_start = None,
+                after_a_tag = False,
+                target_suf = '',
+                )
+
+        if st.want_manpage:
+            st.man_out.append(MAN_START % fi.man_headings)
+
+        if '</table>' in fi.html_in:
+            st.html_out[0] = st.html_out[0].replace('</style>', TABLE_STYLE + '</style>')
+
+        self.feed(fi.html_in)
+        fi.html_in = None
+
+        if st.want_manpage:
+            st.html_out.append(MAN_HTML_END % env_subs['date'])
+        st.html_out.append(HTML_END)
+        st.man_out.append(MAN_END)
+
+        fi.html_out = ''.join(st.html_out)
+        st.html_out = None
+
+        fi.man_out = ''.join(st.man_out)
+        st.man_out = None
+
+        for tgt, txt in st.derived_hashtags:
+            derived = txt2target(txt, tgt)
+            if derived not in st.created_hashtags:
+                txt = BIN_CHARS_RE.sub('', txt.replace(NBR_DASH[0], '-').replace(NBR_SPACE[0], ' '))
+                warn('Unknown derived hashtag link in', self.fn, 'based on:', (tgt, txt))
+
+        for bad in st.bad_hashtags:
+            if bad in st.created_hashtags:
+                warn('Missing "#" in hashtag link in', self.fn + ':', bad)
+            else:
+                warn('Unknown non-hashtag link in', self.fn + ':', bad)
+
+        for bad in st.referenced_hashtags - st.created_hashtags:
+            warn('Unknown hashtag link in', self.fn + ':', '#' + bad)
+
+    def handle_UE(self):
+        st = self.state
+        if st.txt.startswith(('.', ',', '!', '?', ';', ':')):
+            st.man_out[-1] = ".UE " + st.txt[0] + "\n"
+            st.txt = st.txt[1:]
+        st.after_a_tag = False
+
+    def handle_starttag(self, tag, attrs_list):
+        st = self.state
+        if args.debug:
+            self.output_debug('START', (tag, attrs_list))
+        if st.at_first_tag_in_li:
+            if st.list_state[-1] == 'dl':
+                st.dt_from = tag
+                if tag == 'p':
+                    tag = 'dt'
+                else:
+                    st.html_out.append('<dt>')
+            elif tag == 'p':
+                st.at_first_tag_in_dd = True # Kluge to suppress a .P at the start of an li.
+            st.at_first_tag_in_li = False
+        if tag == 'p':
+            if not st.at_first_tag_in_dd:
+                st.man_out.append(st.p_macro)
+        elif tag == 'li':
+            st.at_first_tag_in_li = True
+            lstate = st.list_state[-1]
+            if lstate == 'dl':
+                return
+            if lstate == 'o':
+                st.man_out.append(".IP o\n")
+            else:
+                st.man_out.append(".IP " + str(lstate) + ".\n")
+                st.list_state[-1] += 1
+        elif tag == 'blockquote':
+            st.man_out.append(".RS 4\n")
+        elif tag == 'pre':
+            st.in_pre = True
+            st.man_out.append(st.p_macro + ".nf\n")
+        elif tag == 'code' and not st.in_pre:
+            st.in_code = True
+            st.txt += BOLD_FONT[0]
+        elif tag == 'strong' or tag == 'b':
+            st.txt += BOLD_FONT[0]
+        elif tag == 'em' or  tag == 'i':
+            if st.want_manpage:
+                tag = 'u' # Change it into underline to be more like the manpage
+                st.txt += UNDR_FONT[0]
+        elif tag == 'ol':
+            start = 1
+            for var, val in attrs_list:
+                if var == 'start':
+                    start = int(val) # We only support integers.
+                    break
+            if st.list_state:
+                st.man_out.append(".RS\n")
+            if start == 0:
+                tag = 'dl'
+                attrs_list = [ ]
+                st.list_state.append('dl')
+            else:
+                st.list_state.append(start)
+            st.man_out.append(st.p_macro)
+            st.p_macro = ".IP\n"
+        elif tag == 'ul':
+            st.man_out.append(st.p_macro)
+            if st.list_state:
+                st.man_out.append(".RS\n")
+                st.p_macro = ".IP\n"
+            st.list_state.append('o')
+        elif tag == 'hr':
+            st.man_out.append(".l\n")
+            st.html_out.append("<hr />")
+            return
+        elif tag == 'a':
+            st.a_href = None
+            for var, val in attrs_list:
+                if var == 'href':
+                    if val.startswith(('https://', 'http://', 'mailto:', 'ftp:')):
+                        if st.after_a_tag:
+                            self.handle_UE()
+                        st.man_out.append(manify(st.txt.strip()) + "\n")
+                        st.man_out.append(".UR " + val + "\n")
+                        st.txt = ''
+                        st.a_href = val
+                        st.a_href_external = True
+                    elif '#' in val:
+                        pg, tgt = val.split('#', 1)
+                        if pg and pg not in VALID_PAGES or '#' in tgt:
+                            st.bad_hashtags.add(val)
+                        elif tgt in ('', 'opt', 'dopt'):
+                            st.a_href = val
+                            st.a_href_external = False
+                        elif pg == '':
+                            st.referenced_hashtags.add(tgt)
+                            if tgt in st.latest_targets:
+                                warn('Found link to the current section in', self.fn + ':', val)
+                    elif val not in VALID_PAGES:
+                        st.bad_hashtags.add(val)
+            st.a_txt_start = len(st.txt)
+        st.html_out.append('<' + tag + ''.join(' ' + var + '="' + htmlify(val) + '"' for var, val in attrs_list) + '>')
+        st.at_first_tag_in_dd = False
+
+
+    def handle_endtag(self, tag):
+        st = self.state
+        if args.debug:
+            self.output_debug('END', (tag,))
+        if st.after_a_tag:
+            self.handle_UE()
+        if tag in CONSUMES_TXT or st.dt_from == tag:
+            txt = st.txt.strip()
+            st.txt = ''
+        else:
+            txt = None
+        add_to_txt = None
+        if tag == 'h1':
+            tgt = txt
+            target_suf = ''
+            if tgt.startswith('NEWS for '):
+                m = VERSION_RE.search(tgt)
+                if m:
+                    tgt = m.group(1)
+                    st.target_suf = '-' + tgt
+            self.add_targets(tag, tgt)
+        elif tag == 'h2':
+            st.man_out.append(st.p_macro + '.SH "' + manify(txt) + '"\n')
+            self.add_targets(tag, txt, st.target_suf)
+            st.opt_prefix = 'dopt' if txt == 'DAEMON OPTIONS' else 'opt'
+        elif tag == 'h3':
+            st.man_out.append(st.p_macro + '.SS "' + manify(txt) + '"\n')
+            self.add_targets(tag, txt, st.target_suf)
+        elif tag == 'p':
+            if st.dt_from == 'p':
+                tag = 'dt'
+                st.man_out.append('.IP "' + manify(txt) + '"\n')
+                if txt.startswith(BOLD_FONT[0]):
+                    self.add_targets(tag, txt)
+                st.dt_from = None
+            elif txt != '':
+                st.man_out.append(manify(txt) + "\n")
+        elif tag == 'li':
+            if st.list_state[-1] == 'dl':
+                if st.at_first_tag_in_li:
+                    die("Invalid 0. -> td translation")
+                tag = 'dd'
+            if txt != '':
+                st.man_out.append(manify(txt) + "\n")
+            st.at_first_tag_in_li = False
+        elif tag == 'blockquote':
+            st.man_out.append(".RE\n")
+        elif tag == 'pre':
+            st.in_pre = False
+            st.man_out.append(manify(txt) + "\n.fi\n")
+        elif (tag == 'code' and not st.in_pre):
+            st.in_code = False
+            add_to_txt = NORM_FONT[0]
+        elif tag == 'strong' or tag == 'b':
+            add_to_txt = NORM_FONT[0]
+        elif tag == 'em' or  tag == 'i':
+            if st.want_manpage:
+                tag = 'u' # Change it into underline to be more like the manpage
+                add_to_txt = NORM_FONT[0]
+        elif tag == 'ol' or tag == 'ul':
+            if st.list_state.pop() == 'dl':
+                tag = 'dl'
+            if st.list_state:
+                st.man_out.append(".RE\n")
+            else:
+                st.p_macro = ".P\n"
+            st.at_first_tag_in_dd = False
+        elif tag == 'hr':
+            return
+        elif tag == 'a':
+            if st.a_href_external:
+                st.txt = st.txt.strip()
+                if args.force_link_text or st.a_href != st.txt:
+                    st.man_out.append(manify(st.txt) + "\n")
+                st.man_out.append(".UE\n") # This might get replaced with a punctuation version in handle_UE()
+                st.after_a_tag = True
+                st.a_href_external = False
+                st.txt = ''
+            elif st.a_href:
+                atxt = st.txt[st.a_txt_start:]
+                find = 'href="' + st.a_href + '"'
+                for j in range(len(st.html_out)-1, 0, -1):
+                    if find in st.html_out[j]:
+                        pg, tgt = st.a_href.split('#', 1)
+                        derived = txt2target(atxt, tgt)
+                        if pg == '':
+                            if derived in st.latest_targets:
+                                warn('Found link to the current section in', self.fn + ':', st.a_href)
+                            st.derived_hashtags.add((tgt, atxt))
+                        st.html_out[j] = st.html_out[j].replace(find, 'href="' + pg + '#' + derived + '"')
+                        break
+                else:
+                    die('INTERNAL ERROR: failed to find href in html data:', find)
+        st.html_out.append('</' + tag + '>')
+        if add_to_txt:
+            if txt is None:
+                st.txt += add_to_txt
+            else:
+                txt += add_to_txt
+        if st.dt_from == tag:
+            st.man_out.append('.IP "' + manify(txt) + '"\n')
+            st.html_out.append('</dt><dd>')
+            st.at_first_tag_in_dd = True
+            st.dt_from = None
+        elif tag == 'dt':
+            st.html_out.append('<dd>')
+            st.at_first_tag_in_dd = True
+
+
+    def handle_data(self, txt):
+        st = self.state
+        if '](' in txt:
+            warn('Malformed link in', self.fn + ':', txt)
+        if args.debug:
+            self.output_debug('DATA', (txt,))
+        if st.in_pre:
+            html = htmlify(txt)
+        else:
+            txt = SPACE_DOUBLE_DASH_RE.sub(NBR_SPACE[0] + r'--\1', txt).replace('--', NBR_DASH[0]*2)
+            txt = NON_SPACE_SINGLE_DASH_RE.sub(r'\1' + NBR_DASH[0], txt)
+            html = htmlify(txt)
+            if st.in_code:
+                txt = WHITESPACE_RE.sub(NBR_SPACE[0], txt)
+                html = html.replace(NBR_DASH[0], '-').replace(NBR_SPACE[0], ' ') # <code> is non-breaking in CSS
+        st.html_out.append(html.replace(NBR_SPACE[0], '&nbsp;').replace(NBR_DASH[0], '-&#8288;'))
+        st.txt += txt
+
+
+    def add_targets(self, tag, txt, suf=None):
+        st = self.state
+        tag = '<' + tag + '>'
+        targets = CODE_BLOCK_RE.findall(txt)
+        if not targets:
+            targets = [ txt ]
+        tag_pos = 0
+        for txt in targets:
+            txt = txt2target(txt, st.opt_prefix)
+            if not txt:
+                continue
+            if suf:
+                txt += suf
+            if txt in st.created_hashtags:
+                for j in range(2, 1000):
+                    chk = txt + '-' + str(j)
+                    if chk not in st.created_hashtags:
+                        print('Made link target unique:', chk)
+                        txt = chk
+                        break
+            if tag_pos == 0:
+                tag_pos -= 1
+                while st.html_out[tag_pos] != tag:
+                    tag_pos -= 1
+                st.html_out[tag_pos] = tag[:-1] + ' id="' + txt + '">'
+                st.html_out.append('<a href="#' + txt + '" class="tgt"></a>')
+                tag_pos -= 1 # take into account the append
+            else:
+                st.html_out[tag_pos] = '<span id="' + txt + '"></span>' + st.html_out[tag_pos]
+            st.created_hashtags.add(txt)
+        st.latest_targets = targets
+
+
+    def output_debug(self, event, extra):
+        import pprint
+        st = self.state
+        if args.debug < 2:
+            st = argparse.Namespace(**vars(st))
+            if len(st.html_out) > 2:
+                st.html_out = ['...'] + st.html_out[-2:]
+            if len(st.man_out) > 2:
+                st.man_out = ['...'] + st.man_out[-2:]
+        print(event, extra)
+        pprint.PrettyPrinter(indent=2).pprint(vars(st))
+
+
+def txt2target(txt, opt_prefix):
+    txt = txt.strip().rstrip(':')
+    m = CODE_BLOCK_RE.search(txt)
+    if m:
+        txt = m.group(1)
+    txt = NBR_DASH_RE.sub('-', txt)
+    txt = BIN_CHARS_RE.sub('', txt)
+    txt = INVALID_TARGET_CHARS_RE.sub('_', txt)
+    if opt_prefix and txt.startswith('-'):
+        txt = opt_prefix + txt
+    else:
+        txt = INVALID_START_CHAR_RE.sub(r't\1', txt)
+    return txt
+
+
+def manify(txt):
+    return MANIFY_LINESTART_RE.sub(r'\&\1', txt.replace('\\', '\\\\')
+            .replace(NBR_SPACE[0], NBR_SPACE[1])
+            .replace(NBR_DASH[0], NBR_DASH[1])
+            .replace(NORM_FONT[0], NORM_FONT[1])
+            .replace(BOLD_FONT[0], BOLD_FONT[1])
+            .replace(UNDR_FONT[0], UNDR_FONT[1]))
+
+
+def htmlify(txt):
+    return txt.replace('&', '&amp;').replace('<', '&lt;').replace('>', '&gt;').replace('"', '&quot;')
+
+
+def warn(*msg):
+    print(*msg, file=sys.stderr)
+    global warning_count
+    warning_count += 1
+
+
+def die(*msg):
+    warn(*msg)
+    sys.exit(1)
+
+
+if __name__ == '__main__':
+    parser = argparse.ArgumentParser(description="Convert markdown into html and (optionally) nroff. Each input filename must have a .md suffix, which is changed to .html for the output filename. If the input filename ends with .num.md (e.g. foo.1.md) then a nroff file is also output with the input filename's .md suffix removed (e.g. foo.1).", add_help=False)
+    parser.add_argument('--test', action='store_true', help="Just test the parsing without outputting any files.")
+    parser.add_argument('--dest', metavar='DIR', help="Create files in DIR instead of the current directory.")
+    parser.add_argument('--force-link-text', action='store_true', help="Don't remove the link text if it matches the link href. Useful when nroff doesn't understand .UR and .UE.")
+    parser.add_argument('--debug', '-D', action='count', default=0, help='Output copious info on the html parsing. Repeat for even more.')
+    parser.add_argument("--help", "-h", action="help", help="Output this help message and exit.")
+    parser.add_argument("mdfiles", metavar='FILE.md', nargs='+', help="One or more .md files to convert.")
+    args = parser.parse_args()
+
+    try:
+        import cmarkgfm
+        md_parser = cmarkgfm.markdown_to_html
+        gfm_parser = cmarkgfm.github_flavored_markdown_to_html
+    except:
+        try:
+            import commonmark
+            md_parser = html_via_commonmark
+        except:
+            die("Failed to find cmarkgfm or commonmark for python3.")
+        gfm_parser = None
+
+    main()
+    if warning_count:
+        sys.exit(1)

md2man

@@ -1,383 +0,0 @@
-#!/usr/bin/env python3
-
-# This script takes a manpage written in markdown and turns it into an html web
-# page and a nroff man page.  The input file must have the name of the program
-# and the section in this format: NAME.NUM.md.  The output files are written
-# into the current directory named NAME.NUM.html and NAME.NUM.  The input
-# format has one extra extension: if a numbered list starts at 0, it is turned
-# into a description list. The dl's dt tag is taken from the contents of the
-# first tag inside the li, which is usually a p, code, or strong tag.  The
-# cmarkgfm or commonmark lib is used to transforms the input file into html.
-# The html.parser is used as a state machine that both tweaks the html and
-# outputs the nroff data based on the html tags.
-#
-# Copyright (C) 2020 Wayne Davison
-#
-# This program is freely redistributable.
-
-import sys, os, re, argparse, subprocess, time
-from html.parser import HTMLParser
-
-CONSUMES_TXT = set('h1 h2 p li pre'.split())
-
-HTML_START = """\
-<html><head>
-<title>%s</title>
-<link href="https://fonts.googleapis.com/css2?family=Roboto&family=Roboto+Mono&display=swap" rel="stylesheet">
-<style>
-body {
-  max-width: 50em;
-  margin: auto;
-}
-body, b, strong, u {
-  font-family: 'Roboto', sans-serif;
-}
-code {
-  font-family: 'Roboto Mono', monospace;
-  font-weight: bold;
-  white-space: pre;
-}
-pre code {
-  display: block;
-  font-weight: normal;
-}
-blockquote pre code {
-  background: #f1f1f1;
-}
-dd p:first-of-type {
-  margin-block-start: 0em;
-}
-</style>
-</head><body>
-"""
-
-HTML_END = """\
-<div style="float: right"><p><i>%s</i></p></div>
-</body></html>
-"""
-
-MAN_START = r"""
-.TH "%s" "%s" "%s" "%s" "User Commands"
-""".lstrip()
-
-MAN_END = """\
-"""
-
-NORM_FONT = ('\1', r"\fP")
-BOLD_FONT = ('\2', r"\fB")
-UNDR_FONT = ('\3', r"\fI")
-NBR_DASH = ('\4', r"\-")
-NBR_SPACE = ('\xa0', r"\ ")
-
-md_parser = None
-
-def main():
-    fi = re.match(r'^(?P<fn>(?P<srcdir>.+/)?(?P<name>(?P<prog>[^/]+)\.(?P<sect>\d+))\.md)$', args.mdfile)
-    if not fi:
-        die('Failed to parse NAME.NUM.md out of input file:', args.mdfile)
-    fi = argparse.Namespace(**fi.groupdict())
-
-    if not fi.srcdir:
-        fi.srcdir = './'
-
-    fi.title = fi.prog + '(' + fi.sect + ') man page'
-    fi.mtime = 0
-
-    git_dir = fi.srcdir + '.git'
-    if os.path.lexists(git_dir):
-        fi.mtime = int(subprocess.check_output(['git', '--git-dir', git_dir, 'log', '-1', '--format=%at']))
-
-    env_subs = { 'prefix': os.environ.get('RSYNC_OVERRIDE_PREFIX', None) }
-
-    if args.test:
-        env_subs['VERSION'] = '1.0.0'
-        env_subs['libdir'] = '/usr'
-    else:
-        for fn in (fi.srcdir + 'version.h', 'Makefile'):
-            try:
-                st = os.lstat(fn)
-            except:
-                die('Failed to find', fi.srcdir + fn)
-            if not fi.mtime:
-                fi.mtime = st.st_mtime
-
-        with open(fi.srcdir + 'version.h', 'r', encoding='utf-8') as fh:
-            txt = fh.read()
-        m = re.search(r'"(.+?)"', txt)
-        env_subs['VERSION'] = m.group(1)
-
-        with open('Makefile', 'r', encoding='utf-8') as fh:
-            for line in fh:
-                m = re.match(r'^(\w+)=(.+)', line)
-                if not m:
-                    continue
-                var, val = (m.group(1), m.group(2))
-                if var == 'prefix' and env_subs[var] is not None:
-                    continue
-                while re.search(r'\$\{', val):
-                    val = re.sub(r'\$\{(\w+)\}', lambda m: env_subs[m.group(1)], val)
-                env_subs[var] = val
-                if var == 'srcdir':
-                    break
-
-    with open(fi.fn, 'r', encoding='utf-8') as fh:
-        txt = fh.read()
-
-    txt = re.sub(r'@VERSION@', env_subs['VERSION'], txt)
-    txt = re.sub(r'@LIBDIR@', env_subs['libdir'], txt)
-
-    fi.html_in = md_parser(txt)
-    txt = None
-
-    fi.date = time.strftime('%d %b %Y', time.localtime(fi.mtime))
-    fi.man_headings = (fi.prog, fi.sect, fi.date, fi.prog + ' ' + env_subs['VERSION'])
-
-    HtmlToManPage(fi)
-
-    if args.test:
-        print("The test was successful.")
-        return
-
-    for fn, txt in ((fi.name + '.html', fi.html_out), (fi.name, fi.man_out)):
-        print("Wrote:", fn)
-        with open(fn, 'w', encoding='utf-8') as fh:
-            fh.write(txt)
-
-
-def html_via_commonmark(txt):
-    return commonmark.HtmlRenderer().render(commonmark.Parser().parse(txt))
-
-
-class HtmlToManPage(HTMLParser):
-    def __init__(self, fi):
-        HTMLParser.__init__(self, convert_charrefs=True)
-
-        st = self.state = argparse.Namespace(
-                list_state = [ ],
-                p_macro = ".P\n",
-                at_first_tag_in_li = False,
-                at_first_tag_in_dd = False,
-                dt_from = None,
-                in_pre = False,
-                in_code = False,
-                html_out = [ HTML_START % fi.title ],
-                man_out = [ MAN_START % fi.man_headings ],
-                txt = '',
-                )
-
-        self.feed(fi.html_in)
-        fi.html_in = None
-
-        st.html_out.append(HTML_END % fi.date)
-        st.man_out.append(MAN_END)
-
-        fi.html_out = ''.join(st.html_out)
-        st.html_out = None
-
-        fi.man_out = ''.join(st.man_out)
-        st.man_out = None
-
-
-    def handle_starttag(self, tag, attrs_list):
-        st = self.state
-        if args.debug:
-            self.output_debug('START', (tag, attrs_list))
-        if st.at_first_tag_in_li:
-            if st.list_state[-1] == 'dl':
-                st.dt_from = tag
-                if tag == 'p':
-                    tag = 'dt'
-                else:
-                    st.html_out.append('<dt>')
-            elif tag == 'p':
-                st.at_first_tag_in_dd = True # Kluge to suppress a .P at the start of an li.
-            st.at_first_tag_in_li = False
-        if tag == 'p':
-            if not st.at_first_tag_in_dd:
-                st.man_out.append(st.p_macro)
-        elif tag == 'li':
-            st.at_first_tag_in_li = True
-            lstate = st.list_state[-1]
-            if lstate == 'dl':
-                return
-            if lstate == 'o':
-                st.man_out.append(".IP o\n")
-            else:
-                st.man_out.append(".IP " + str(lstate) + ".\n")
-                st.list_state[-1] += 1
-        elif tag == 'blockquote':
-            st.man_out.append(".RS 4\n")
-        elif tag == 'pre':
-            st.in_pre = True
-            st.man_out.append(st.p_macro + ".nf\n")
-        elif tag == 'code' and not st.in_pre:
-            st.in_code = True
-            st.txt += BOLD_FONT[0]
-        elif tag == 'strong' or tag == 'b':
-            st.txt += BOLD_FONT[0]
-        elif tag == 'em' or  tag == 'i':
-            tag = 'u' # Change it into underline to be more like the man page
-            st.txt += UNDR_FONT[0]
-        elif tag == 'ol':
-            start = 1
-            for var, val in attrs_list:
-                if var == 'start':
-                    start = int(val) # We only support integers.
-                    break
-            if st.list_state:
-                st.man_out.append(".RS\n")
-            if start == 0:
-                tag = 'dl'
-                attrs_list = [ ]
-                st.list_state.append('dl')
-            else:
-                st.list_state.append(start)
-            st.man_out.append(st.p_macro)
-            st.p_macro = ".IP\n"
-        elif tag == 'ul':
-            st.man_out.append(st.p_macro)
-            if st.list_state:
-                st.man_out.append(".RS\n")
-                st.p_macro = ".IP\n"
-            st.list_state.append('o')
-        st.html_out.append('<' + tag + ''.join(' ' + var + '="' + htmlify(val) + '"' for var, val in attrs_list) + '>')
-        st.at_first_tag_in_dd = False
-
-
-    def handle_endtag(self, tag):
-        st = self.state
-        if args.debug:
-            self.output_debug('END', (tag,))
-        if tag in CONSUMES_TXT or st.dt_from == tag:
-            txt = st.txt.strip()
-            st.txt = ''
-        else:
-            txt = None
-        add_to_txt = None
-        if tag == 'h1':
-            st.man_out.append(st.p_macro + '.SH "' + manify(txt) + '"\n')
-        elif tag == 'h2':
-            st.man_out.append(st.p_macro + '.SS "' + manify(txt) + '"\n')
-        elif tag == 'p':
-            if st.dt_from == 'p':
-                tag = 'dt'
-                st.man_out.append('.IP "' + manify(txt) + '"\n')
-                st.dt_from = None
-            elif txt != '':
-                st.man_out.append(manify(txt) + "\n")
-        elif tag == 'li':
-            if st.list_state[-1] == 'dl':
-                if st.at_first_tag_in_li:
-                    die("Invalid 0. -> td translation")
-                tag = 'dd'
-            if txt != '':
-                st.man_out.append(manify(txt) + "\n")
-            st.at_first_tag_in_li = False
-        elif tag == 'blockquote':
-            st.man_out.append(".RE\n")
-        elif tag == 'pre':
-            st.in_pre = False
-            st.man_out.append(manify(txt) + "\n.fi\n")
-        elif (tag == 'code' and not st.in_pre):
-            st.in_code = False
-            add_to_txt = NORM_FONT[0]
-        elif tag == 'strong' or tag == 'b':
-            add_to_txt = NORM_FONT[0]
-        elif tag == 'em' or  tag == 'i':
-            tag = 'u' # Change it into underline to be more like the man page
-            add_to_txt = NORM_FONT[0]
-        elif tag == 'ol' or tag == 'ul':
-            if st.list_state.pop() == 'dl':
-                tag = 'dl'
-            if st.list_state:
-                st.man_out.append(".RE\n")
-            else:
-                st.p_macro = ".P\n"
-            st.at_first_tag_in_dd = False
-        st.html_out.append('</' + tag + '>')
-        if add_to_txt:
-            if txt is None:
-                st.txt += add_to_txt
-            else:
-                txt += add_to_txt
-        if st.dt_from == tag:
-            st.man_out.append('.IP "' + manify(txt) + '"\n')
-            st.html_out.append('</dt><dd>')
-            st.at_first_tag_in_dd = True
-            st.dt_from = None
-        elif tag == 'dt':
-            st.html_out.append('<dd>')
-            st.at_first_tag_in_dd = True
-
-
-    def handle_data(self, txt):
-        st = self.state
-        if args.debug:
-            self.output_debug('DATA', (txt,))
-        if st.in_pre:
-            html = htmlify(txt)
-        else:
-            txt = re.sub(r'\s--(\s)', NBR_SPACE[0] + r'--\1', txt).replace('--', NBR_DASH[0]*2)
-            txt = re.sub(r'(^|\W)-', r'\1' + NBR_DASH[0], txt)
-            html = htmlify(txt)
-            if st.in_code:
-                txt = re.sub(r'\s', NBR_SPACE[0], txt)
-                html = html.replace(NBR_DASH[0], '-').replace(NBR_SPACE[0], ' ') # <code> is non-breaking in CSS
-        st.html_out.append(html.replace(NBR_SPACE[0], '&nbsp;').replace(NBR_DASH[0], '-&#8288;'))
-        st.txt += txt
-
-
-    def output_debug(self, event, extra):
-        import pprint
-        st = self.state
-        if args.debug < 2:
-            st = argparse.Namespace(**vars(st))
-            if len(st.html_out) > 2:
-                st.html_out = ['...'] + st.html_out[-2:]
-            if len(st.man_out) > 2:
-                st.man_out = ['...'] + st.man_out[-2:]
-        print(event, extra)
-        pprint.PrettyPrinter(indent=2).pprint(vars(st))
-
-
-def manify(txt):
-    return re.sub(r"^(['.])", r'\&\1', txt.replace('\\', '\\\\')
-            .replace(NBR_SPACE[0], NBR_SPACE[1])
-            .replace(NBR_DASH[0], NBR_DASH[1])
-            .replace(NORM_FONT[0], NORM_FONT[1])
-            .replace(BOLD_FONT[0], BOLD_FONT[1])
-            .replace(UNDR_FONT[0], UNDR_FONT[1]), flags=re.M)
-
-
-def htmlify(txt):
-    return txt.replace('&', '&amp;').replace('<', '&lt;').replace('>', '&gt;').replace('"', '&quot;')
-
-
-def warn(*msg):
-    print(*msg, file=sys.stderr)
-
-
-def die(*msg):
-    warn(*msg)
-    sys.exit(1)
-
-
-if __name__ == '__main__':
-    parser = argparse.ArgumentParser(description='Transform a NAME.NUM.md markdown file into a NAME.NUM.html web page & a NAME.NUM man page.', add_help=False)
-    parser.add_argument('--test', action='store_true', help='Test if we can parse the input w/o updating any files.')
-    parser.add_argument('--debug', '-D', action='count', default=0, help='Output copious info on the html parsing. Repeat for even more.')
-    parser.add_argument("--help", "-h", action="help", help="Output this help message and exit.")
-    parser.add_argument('mdfile', help="The NAME.NUM.md file to parse.")
-    args = parser.parse_args()
-
-    try:
-        import cmarkgfm
-        md_parser = cmarkgfm.markdown_to_html
-    except:
-        try:
-            import commonmark
-            md_parser = html_via_commonmark
-        except:
-            die("Failed to find cmarkgfm or commonmark for python3.")
-
-    main()

md2man

@@ -0,0 +1 @@
+md-convert

mkgitver

@@ -0,0 +1,22 @@
+#!/bin/sh
+
+srcdir=`dirname $0`
+
+if [ ! -f git-version.h ]; then
+    touch git-version.h
+fi
+
+if test -d "$srcdir/.git" || test -f "$srcdir/.git"; then
+    gitver=`git describe --abbrev=8 2>/dev/null`
+    # NOTE: I'm avoiding "|" in sed since I'm not sure if sed -r is portable and "\|" fails on some OSes.
+    verchk=`echo "$gitver-" | sed -n '/^v3\.[0-9][0-9]*\.[0-9][0-9]*\(pre[0-9]*\)*-/p'`
+    if [ -n "$verchk" ]; then
+	echo "#define RSYNC_GITVER \"$gitver\"" >git-version.h.new
+	if ! diff git-version.h.new git-version.h >/dev/null; then
+	    echo "Updating git-version.h"
+	    mv git-version.h.new git-version.h
+	else
+	    rm git-version.h.new
+	fi
+    fi
+fi

mkproto.awk

@@ -36,5 +36,5 @@ inheader {
 
 END {
     if (old_protos != protos) print protos > "proto.h"
-    printf "" > "proto.h-tstamp"
+    system("touch proto.h-tstamp")
 }

packaging/auto-Makefile

@@ -1,6 +1,6 @@
 TARGETS := all install install-ssl-daemon install-all install-strip conf gen gensend reconfigure restatus \
 	proto man clean cleantests distclean test check check29 check30 installcheck splint \
-	doxygen doxygen-upload
+	doxygen doxygen-upload finddead rrsync
 
 .PHONY: $(TARGETS) auto-prep
 

packaging/cull-options

@@ -0,0 +1,148 @@
+#!/usr/bin/env python3
+# This script outputs either perl or python code that parses all possible options
+# that the code in options.c might send to the server.  The resulting code is then
+# included in the rrsync script.
+
+import re, argparse
+
+short_no_arg = { }
+short_with_num = { '@': 1 }
+long_opts = { # These include some extra long-args that BackupPC uses:
+        'block-size': 1,
+        'daemon': -1,
+        'debug': 1,
+        'fake-super': 0,
+        'fuzzy': 0,
+        'group': 0,
+        'hard-links': 0,
+        'ignore-times': 0,
+        'info': 1,
+        'links': 0,
+        'log-file': 3,
+        'munge-links': 0,
+        'no-munge-links': -1,
+        'one-file-system': 0,
+        'owner': 0,
+        'perms': 0,
+        'recursive': 0,
+        'stderr': 1,
+        'times': 0,
+        'copy-devices': -1,
+        'write-devices': -1,
+        }
+
+def main():
+    last_long_opt = None
+
+    with open('../options.c') as fh:
+        for line in fh:
+            m = re.search(r"argstr\[x\+\+\] = '([^.ie])'", line)
+            if m:
+                short_no_arg[m.group(1)] = 1
+                last_long_opt = None
+                continue
+
+            m = re.search(r'asprintf\([^,]+, "-([a-zA-Z0-9])\%l?[ud]"', line)
+            if m:
+                short_with_num[m.group(1)] = 1
+                last_long_opt = None
+                continue
+
+            m = re.search(r'args\[ac\+\+\] = "--([^"=]+)"', line)
+            if m:
+                last_long_opt = m.group(1)
+                if last_long_opt not in long_opts:
+                    long_opts[last_long_opt] = 0
+                else:
+                    last_long_opt = None
+                continue
+
+            if last_long_opt:
+                m = re.search(r'args\[ac\+\+\] = safe_arg\("", ([^[("\s]+)\);', line)
+                if m:
+                    long_opts[last_long_opt] = 2
+                    last_long_opt = None
+                    continue
+                if 'args[ac++] = ' in line:
+                    last_long_opt = None
+
+            m = re.search(r'return "--([^"]+-dest)";', line)
+            if m:
+                long_opts[m.group(1)] = 2
+                last_long_opt = None
+                continue
+
+            m = re.search(r'asprintf\([^,]+, "--([^"=]+)=', line)
+            if not m:
+                m = re.search(r'args\[ac\+\+\] = "--([^"=]+)=', line)
+                if not m:
+                    m = re.search(r'args\[ac\+\+\] = safe_arg\("--([^"=]+)"', line)
+                    if not m:
+                        m = re.search(r'fmt = .*: "--([^"=]+)=', line)
+            if m:
+                long_opts[m.group(1)] = 1
+                last_long_opt = None
+
+    long_opts['files-from'] = 3
+
+    txt = """\
+### START of options data produced by the cull-options script. ###
+
+# To disable a short-named option, add its letter to this string:
+"""
+
+    txt += str_assign('short_disabled', 's') + "\n"
+    txt += '# These are also disabled when the restricted dir is not "/":\n'
+    txt += str_assign('short_disabled_subdir', 'KLk') + "\n"
+    txt += '# These are all possible short options that we will accept (when not disabled above):\n'
+    txt += str_assign('short_no_arg', ''.join(sorted(short_no_arg)), 'DO NOT REMOVE ANY')
+    txt += str_assign('short_with_num', ''.join(sorted(short_with_num)), 'DO NOT REMOVE ANY')
+   
+    txt += """
+# To disable a long-named option, change its value to a -1.  The values mean:
+# 0 = the option has no arg; 1 = the arg doesn't need any checking; 2 = only
+# check the arg when receiving; and 3 = always check the arg.
+"""
+
+    print(txt, end='')
+
+    if args.python:
+        print("long_opts = {")
+        sep = ':'
+    else:
+        print("our %long_opt = (")
+        sep = ' =>'
+
+    for opt in sorted(long_opts):
+        if opt.startswith(('min-', 'max-')):
+            val = 1
+        else:
+            val = long_opts[opt]
+        print(' ', repr(opt) + sep, str(val) + ',')
+
+    if args.python:
+        print("}")
+    else:
+        print(");")
+    print("\n### END of options data produced by the cull-options script. ###")
+
+
+def str_assign(name, val, comment=None):
+    comment = ' # ' + comment if comment else ''
+    if args.python:
+        return name + ' = ' + repr(val) + comment + "\n"
+    return 'our $' + name + ' = ' + repr(val) + ';' + comment + "\n"
+
+
+if __name__ == '__main__':
+    parser = argparse.ArgumentParser(description="Output culled rsync options for rrsync.", add_help=False)
+    out_group = parser.add_mutually_exclusive_group()
+    out_group.add_argument('--perl', action='store_true', help="Output perl code.")
+    out_group.add_argument('--python', action='store_true', help="Output python code (the default).")
+    parser.add_argument('--help', '-h', action='help', help="Output this help message and exit.")
+    args = parser.parse_args()
+    if not args.perl:
+        args.python = True
+    main()
+
+# vim: sw=4 et

packaging/cull_options

@@ -1,85 +0,0 @@
-#!/usr/bin/env perl
-# This script outputs some perl code that parses all possible options
-# that the code in options.c might send to the server.  This perl code
-# is included in the rrsync script.
-use strict;
-
-our %short_no_arg;
-our %short_with_num = ( '@' => 1 );
-our %long_opt = ( # These include some extra long-args that BackupPC uses:
-    'block-size' => 1,
-    'daemon' => -1,
-    'debug' => 1,
-    'fake-super' => 0,
-    'fuzzy' => 0,
-    'group' => 0,
-    'hard-links' => 0,
-    'ignore-times' => 0,
-    'info' => 1,
-    'links' => 0,
-    'log-file' => 3,
-    'one-file-system' => 0,
-    'owner' => 0,
-    'perms' => 0,
-    'recursive' => 0,
-    'times' => 0,
-    'write-devices' => -1,
-);
-our $last_long_opt;
-
-open(IN, '../options.c') or die "Unable to open ../options.c: $!\n";
-
-while (<IN>) {
-    if (/\Qargstr[x++]\E = '([^.ie])'/) {
-	$short_no_arg{$1} = 1;
-	undef $last_long_opt;
-    } elsif (/\Qasprintf(\E[^,]+, "-([a-zA-Z0-9])\%l?[ud]"/) {
-	$short_with_num{$1} = 1;
-	undef $last_long_opt;
-    } elsif (/\Qargs[ac++]\E = "--([^"=]+)"/) {
-	$last_long_opt = $1;
-	$long_opt{$1} = 0 unless exists $long_opt{$1};
-    } elsif (defined($last_long_opt)
-	&& /\Qargs[ac++]\E = ([^["\s]+);/) {
-	$long_opt{$last_long_opt} = 2;
-	undef $last_long_opt;
-    } elsif (/return "--([^"]+-dest)";/) {
-	$long_opt{$1} = 2;
-	undef $last_long_opt;
-    } elsif (/\Qasprintf(\E[^,]+, "--([^"=]+)=/ || /\Qargs[ac++]\E = "--([^"=]+)=/ || /fmt = .*: "--([^"=]+)=/) {
-	$long_opt{$1} = 1;
-	undef $last_long_opt;
-    }
-}
-close IN;
-
-my $short_no_arg = join('', sort keys %short_no_arg);
-my $short_with_num = join('', sort keys %short_with_num);
-
-print <<EOT;
-
-# These options are the only options that rsync might send to the server,
-# and only in the option format that the stock rsync produces.
-
-# To disable a short-named option, add its letter to this string:
-our \$short_disabled = 's';
-
-our \$short_no_arg = '$short_no_arg'; # DO NOT REMOVE ANY
-our \$short_with_num = '$short_with_num'; # DO NOT REMOVE ANY
-
-# To disable a long-named option, change its value to a -1.  The values mean:
-# 0 = the option has no arg; 1 = the arg doesn't need any checking; 2 = only
-# check the arg when receiving; and 3 = always check the arg.
-our \%long_opt = (
-EOT
-
-foreach my $opt (sort keys %long_opt) {
-    my $val = $long_opt{$opt};
-    $val = 1 if $opt =~ /^(max-|min-)/;
-    $val = 3 if $opt eq 'files-from';
-    $val = q"$only eq 'r' ? -1 : " . $val if $opt =~ /^(remove-|log-file)/;
-    $val = q"$only eq 'w' ? -1 : " . $val if $opt eq 'sender';
-    print "  '$opt' => $val,\n";
-}
-
-print ");\n\n";

packaging/lsb/rsync.spec

@@ -1,9 +1,9 @@
 Summary: A fast, versatile, remote (and local) file-copying tool
 Name: rsync
-Version: 3.2.3
-%define fullversion %{version}pre1
-Release: 0.1.pre1
-%define srcdir src-previews
+Version: 3.3.0
+%define fullversion %{version}
+Release: 1
+%define srcdir src
 Group: Applications/Internet
 License: GPL
 Source0: https://rsync.samba.org/ftp/rsync/%{srcdir}/rsync-%{fullversion}.tar.gz
@@ -79,8 +79,8 @@ rm -rf $RPM_BUILD_ROOT
 %dir /etc/rsync-ssl/certs
 
 %changelog
-* Mon Jul 27 2020 Wayne Davison <wayne@opencoder.net>
-Released 3.2.3pre1.
+* Sat Apr 06 2024 Wayne Davison <wayne@opencoder.net>
+Released 3.3.0.
 
 * Fri Mar 21 2008 Wayne Davison <wayne@opencoder.net>
 Added installation of /etc/xinetd.d/rsync file and some commented-out

packaging/md2html

@@ -1,104 +0,0 @@
-#!/usr/bin/env python3
-
-# Copyright (C) 2020 Wayne Davison
-#
-# This program is freely redistributable.
-
-import os, re, argparse
-
-HTML_START = """\
-<html><head>
-<title>%s</title>
-<link href="https://fonts.googleapis.com/css2?family=Roboto&family=Roboto+Mono&display=swap" rel="stylesheet">
-<style>
-body {
-  max-width: 50em;
-  margin: auto;
-}
-body, b, strong, u {
-  font-family: 'Roboto', sans-serif;
-}
-code {
-  font-family: 'Roboto Mono', monospace;
-  font-weight: bold;
-}
-pre code {
-  display: block;
-  font-weight: normal;
-}
-blockquote pre code {
-  background: #f1f1f1;
-}
-dd p:first-of-type {
-  margin-block-start: 0em;
-}
-table {
-  border-color: grey;
-  border-spacing: 0;
-}
-tr {
-  border-top: 1px solid grey;
-}
-tr:nth-child(2n) {
-  background-color: #f6f8fa;
-}
-th, td {
-  border: 1px solid #dfe2e5;
-  text-align: center;
-  padding-left: 1em;
-  padding-right: 1em;
-}
-</style>
-</head><body>
-"""
-
-HTML_END = """\
-</body></html>
-"""
-
-md_parser = None
-
-def main():
-    for mdfn in args.mdfiles:
-        if not mdfn.endswith('.md'):
-            print('Ignoring non-md input file:', mdfn)
-            continue
-        title = re.sub(r'.*/', '', mdfn).replace('.md', '')
-        htfn = mdfn.replace('.md', '.html')
-
-        print("Parsing", mdfn, '->', htfn)
-
-        with open(mdfn, 'r', encoding='utf-8') as fh:
-            txt = fh.read()
-
-        txt = re.sub(r'\s--\s', '\xa0-- ', txt)
-
-        html = md_parser(txt)
-
-        html = re.sub(r'(?<!<pre>)(<code>)([\s\S]*?)(</code>)', lambda m: m[1] + re.sub(r'\s', '\xa0', m[2]) + m[3], html)
-        html = html.replace('--', '&#8209;&#8209;').replace("\xa0-", '&nbsp;&#8209;').replace("\xa0", '&nbsp;')
-        html = re.sub(r'(\W)-', r'\1&#8209;', html)
-
-        if os.path.lexists(htfn):
-            os.unlink(htfn)
-
-        with open(htfn, 'w', encoding='utf-8') as fh:
-            fh.write(HTML_START % title)
-            fh.write(html)
-            fh.write(HTML_END)
-
-
-if __name__ == '__main__':
-    parser = argparse.ArgumentParser(description='Output html for md pages.', add_help=False)
-    parser.add_argument("--help", "-h", action="help", help="Output this help message and exit.")
-    parser.add_argument("mdfiles", nargs='+', help="The .md files to turn into .html files.")
-    args = parser.parse_args()
-
-    try:
-        import cmarkgfm
-        # Our NEWS.md file has a gfm table in it.
-        md_parser = cmarkgfm.github_flavored_markdown_to_html
-    except:
-        die("Failed to find cmarkgfm for python3.")
-
-    main()

packaging/openssl-rsync.cnf

@@ -0,0 +1,18 @@
+# This config file can be used with rsync to enable legacy digests
+# (such as MD4) by using the OPENSSL_CONF environment variable.
+# See rsync's configure --with-openssl-conf=/path/name option.
+
+openssl_conf = openssl_init
+
+[openssl_init]
+providers = provider_sect
+
+[provider_sect]
+default = default_sect
+legacy = legacy_sect
+
+[default_sect]
+activate = 1
+
+[legacy_sect]
+activate = 1

packaging/patch-update

@@ -33,7 +33,7 @@ def main():
 
     master_commit = latest_git_hash(args.base_branch)
 
-    if cmd_txt_chk(['packaging/prep-auto-dir']) == '':
+    if cmd_txt_chk(['packaging/prep-auto-dir']).out == '':
         die('You must setup an auto-build-save dir to use this script.')
 
     if args.gen:

packaging/pkglib.py

@@ -1,4 +1,4 @@
-import os, sys, re, subprocess
+import os, sys, re, subprocess, argparse
 
 # This python3 library provides a few helpful routines that are
 # used by the latest packaging scripts.
@@ -23,27 +23,32 @@ def set_default_encoding(enc):
 
 
 # Set shell=True if the cmd is a string; sets a default encoding unless raw=True was specified.
-def _tweak_opts(cmd, opts, **maybe_set):
-    # This sets any maybe_set value that isn't already set AND creates a copy of opts for us.
-    opts = {**maybe_set, **opts}
+def _tweak_opts(cmd, opts, **maybe_set_args):
+    def _maybe_set(o, **msa): # Only set a value if the user didn't already set it.
+        for var, val in msa.items():
+            if var not in o:
+                o[var] = val
 
-    if type(cmd) == str:
-        opts = {'shell': True, **opts}
+    opts = opts.copy()
+    _maybe_set(opts, **maybe_set_args)
+
+    if isinstance(cmd, str):
+        _maybe_set(opts, shell=True)
 
     want_raw = opts.pop('raw', False)
     if default_encoding and not want_raw:
-        opts = {'encoding': default_encoding, **opts}
+        _maybe_set(opts, encoding=default_encoding)
 
     capture = opts.pop('capture', None)
     if capture:
         if capture == 'stdout':
-            opts = {'stdout': subprocess.PIPE, **opts}
+            _maybe_set(opts, stdout=subprocess.PIPE)
         elif capture == 'stderr':
-            opts = {'stderr': subprocess.PIPE, **opts}
+            _maybe_set(opts, stderr=subprocess.PIPE)
         elif capture == 'output':
-            opts = {'stdout': subprocess.PIPE, 'stderr': subprocess.PIPE, **opts}
+            _maybe_set(opts, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
         elif capture == 'combined':
-            opts = {'stdout': subprocess.PIPE, 'stderr': subprocess.STDOUT, **opts}
+            _maybe_set(opts, stdout=subprocess.PIPE, stderr=subprocess.STDOUT)
 
     discard = opts.pop('discard', None)
     if discard:
@@ -66,30 +71,26 @@ def cmd_chk(cmd, **opts):
     return subprocess.run(cmd, **_tweak_opts(cmd, opts, check=True))
 
 
-# Capture stdout in a string and return the (output, return_code) tuple.
-# Use capture='combined' opt to get both stdout and stderr together.
-def cmd_txt_status(cmd, **opts):
+# Capture stdout in a string and return an object with out, err, and rc (return code).
+# It defaults to capture='stdout' (so err is empty) but can be overridden using
+# capture='combined' or capture='output' (the latter populates the err value).
+def cmd_txt(cmd, **opts):
     input = opts.pop('input', None)
     if input is not None:
         opts['stdin'] = subprocess.PIPE
     proc = subprocess.Popen(cmd, **_tweak_opts(cmd, opts, capture='stdout'))
-    out = proc.communicate(input=input)[0]
-    return (out, proc.returncode)
+    out, err = proc.communicate(input=input)
+    return argparse.Namespace(out=out, err=err, rc=proc.returncode)
 
 
-# Like cmd_txt_status() but just return the output.
-def cmd_txt(cmd, **opts):
-    return cmd_txt_status(cmd, **opts)[0]
-
-
-# Capture stdout in a string and return the output if the command has a 0 return code.
-# Otherwise it throws an exception that indicates the return code and the output.
+# Just like calling cmd_txt() except that it raises an error if the command has a non-0 return code.
+# The raised error includes the cmd, the return code, and the captured output.
 def cmd_txt_chk(cmd, **opts):
-    out, rc = cmd_txt_status(cmd, **opts)
-    if rc != 0:
-        cmd_err = f'Command "{cmd}" returned non-zero exit status "{rc}" and output:\n{out}'
+    ct = cmd_txt(cmd, **opts)
+    if ct.rc != 0:
+        cmd_err = f'Command "{cmd}" returned non-0 exit status "{ct.rc}" and output:\n{ct.out}{ct.err}'
         raise Exception(cmd_err)
-    return out
+    return ct
 
 
 # Starts a piped-output command of stdout (by default) and leaves it up to you to read
@@ -102,7 +103,7 @@ def cmd_pipe(cmd, **opts):
 # arg fatal_unless_clean can be used to make that non-fatal.  Returns a
 # tuple of the current branch, the is_clean flag, and the status text.
 def check_git_status(fatal_unless_clean=True, subdir='.'):
-    status_txt = cmd_txt_chk(f"cd '{subdir}' && git status")
+    status_txt = cmd_txt_chk(f"cd '{subdir}' && git status").out
     is_clean = re.search(r'\nnothing to commit.+working (directory|tree) clean', status_txt) != None
 
     if not is_clean and fatal_unless_clean:
@@ -149,7 +150,7 @@ def check_git_state(master_branch, fatal_unless_clean=True, check_extra_dir=None
 
 # Return the git hash of the most recent commit.
 def latest_git_hash(branch):
-    out = cmd_txt_chk(['git', 'log', '-1', '--no-color', branch])
+    out = cmd_txt_chk(['git', 'log', '-1', '--no-color', branch]).out
     m = re.search(r'^commit (\S+)', out, flags=re.M)
     if not m:
         die(f"Unable to determine commit hash for master branch: {branch}")
@@ -175,20 +176,20 @@ def mandate_gensend_hook():
         print('Creating hook file:', hook)
         cmd_chk(['./rsync', '-a', 'packaging/pre-push', hook])
     else:
-        out, rc = cmd_txt_status(['fgrep', 'make gensend', hook], discard='output')
-        if rc:
+        ct = cmd_txt(['grep', 'make gensend', hook], discard='output')
+        if ct.rc:
             die('Please add a "make gensend" into your', hook, 'script.')
 
 
-# Snag the GENFILES values out of the Makefile.in file and return them as a list.
+# Snag the GENFILES values out of the Makefile file and return them as a list.
 def get_gen_files(want_dir_plus_list=False):
     cont_re = re.compile(r'\\\n')
 
     gen_files = [ ]
 
-    auto_dir = os.path.join('auto-build-save', cmd_txt('git rev-parse --abbrev-ref HEAD').strip().replace('/', '%'))
+    auto_dir = os.path.join('auto-build-save', cmd_txt('git rev-parse --abbrev-ref HEAD').out.strip().replace('/', '%'))
 
-    with open('Makefile.in', 'r', encoding='utf-8') as fh:
+    with open(auto_dir + '/Makefile', 'r', encoding='utf-8') as fh:
         for line in fh:
             if not gen_files:
                 chk = re.sub(r'^GENFILES=', '', line)

packaging/pre-push

@@ -1,3 +1,16 @@
-#!/bin/sh
+#!/bin/bash -e
+
 cat >/dev/null # Just discard stdin data
-make gensend
+
+if [[ -f /proc/$PPID/cmdline ]]; then
+    while read -d $'\0' arg ; do
+	if [[ "$arg" == '--tags' ]] ; then
+	    exit 0
+	fi
+    done </proc/$PPID/cmdline
+fi
+
+branch=`git rev-parse --abbrev-ref HEAD`
+if [[ "$branch" = master && "$*" == *github* ]]; then
+    make gensend
+fi

packaging/prep-auto-dir

@@ -13,7 +13,7 @@
 # run "make distclean" before creating the auto-build-save dir.
 
 auto_top='auto-build-save'
-if test -d $auto_top -a -d .git; then
+if test -d $auto_top && test -d .git; then
     desired_branch=`git rev-parse --abbrev-ref HEAD | tr / %`
     if test "$desired_branch" = HEAD; then
 	echo "ERROR: switch to the right build dir manually when in detached HEAD mode." 1>&2

packaging/release-rsync

@@ -18,6 +18,9 @@ dest = os.environ['HOME'] + '/samba-rsync-ftp'
 ORIGINAL_PATH = os.environ['PATH']
 
 def main():
+    if not os.path.isfile('packaging/release-rsync'):
+        die('You must run this script from the top of your rsync checkout.')
+
     now = datetime.now()
     cl_today = now.strftime('* %a %b %d %Y')
     year = now.strftime('%Y')
@@ -30,7 +33,7 @@ def main():
 
     signal.signal(signal.SIGINT, signal_handler)
 
-    if cmd_txt_chk(['packaging/prep-auto-dir']) == '':
+    if cmd_txt_chk(['packaging/prep-auto-dir']).out == '':
         die('You must setup an auto-build-save dir to use this script.');
 
     auto_dir, gen_files = get_gen_files(True)
@@ -44,6 +47,17 @@ def main():
 {dash_line}
 """)
 
+    with open('build/rsync.1') as fh:
+        for line in fh:
+            if line.startswith(r'.\" prefix='):
+                doc_prefix = line.split('=')[1].strip()
+                if doc_prefix != '/usr':
+                    warn(f"*** The documentation was built with prefix {doc_prefix} instead of /usr ***")
+                    die("*** Read the md2man script for a way to override this. ***")
+                break
+            if line.startswith('.P'):
+                die("Failed to find the prefix comment at the start of the rsync.1 manpage.")
+
     if not os.path.isdir(dest):
         die(dest, "dest does not exist")
     if not os.path.isdir('.git'):
@@ -84,13 +98,15 @@ def main():
     if os.path.lexists(rsync_ver):
         die(f'"{rsync_ver}" must not exist in the current directory.')
 
-    out = cmd_txt_chk(['git', 'tag', '-l', v_ver])
+    out = cmd_txt_chk(['git', 'tag', '-l', v_ver]).out
     if out != '':
         print(f"Tag {v_ver} already exists.")
         ans = input("\nDelete tag or quit? [Q/del] ")
         if not re.match(r'^del', ans, flags=re.I):
             die("Aborted")
         cmd_chk(['git', 'tag', '-d', v_ver])
+        if os.path.isdir('patches/.git'):
+            cmd_chk(f"cd patches && git tag -d '{v_ver}'")
 
     version = re.sub(r'[-.]*pre[-.]*', 'pre', version)
     if 'pre' in version and not curversion.endswith('dev'):
@@ -179,7 +195,9 @@ About to:
         with open(fn, 'r', encoding='utf-8') as fh:
             old_txt = txt = fh.read()
         if fn == 'version.h':
-            txt = f'#define RSYNC_VERSION "{version}"\n'
+            x_re = re.compile(r'^(#define RSYNC_VERSION).*', re.M)
+            msg = f"Unable to update RSYNC_VERSION in {fn}"
+            txt = replace_or_die(x_re, r'\1 "%s"' % version, txt, msg)
         elif '.spec' in fn:
             for var, val in specvars.items():
                 x_re = re.compile(r'^%s .*' % re.escape(var), re.M)
@@ -192,10 +210,10 @@ About to:
             txt = replace_or_die(x_re, repl, txt, f"Unable to find SUBPROTOCOL_VERSION define in {fn}")
         elif fn == 'NEWS.md':
             efv = re.escape(finalversion)
-            x_re = re.compile(r'^<.+>\s+# NEWS for rsync %s \(UNRELEASED\)\s+## Changes in this version:\n' % efv
+            x_re = re.compile(r'^# NEWS for rsync %s \(UNRELEASED\)\s+## Changes in this version:\n' % efv
                     + r'(\n### PROTOCOL NUMBER:\s+- The protocol number was changed to \d+\.\n)?')
             rel_day = 'UNRELEASED' if pre else today
-            repl = (f'<a name="{finalversion}"></a>\n\n# NEWS for rsync {finalversion} ({rel_day})\n\n'
+            repl = (f'# NEWS for rsync {finalversion} ({rel_day})\n\n'
                 + '## Changes in this version:\n')
             if proto_changed:
                 repl += f'\n### PROTOCOL NUMBER:\n\n - The protocol number was changed to {protocol_version}.\n'
@@ -216,7 +234,7 @@ About to:
     cmd_chk(['packaging/year-tweak'])
 
     print(dash_line)
-    cmd_run("git diff --color | less -p '^diff .*'")
+    cmd_run("git diff".split())
 
     srctar_name = f"{rsync_ver}.tar.gz"
     pattar_name = f"rsync-patches-{version}.tar.gz"
@@ -231,20 +249,20 @@ About to:
 
 About to:
     - git commit all changes
-    - generate the manpages
+    - run a full build, ensuring that the manpages & configure.sh are up-to-date
     - merge the {args.master_branch} branch into the patch/{args.master_branch}/* branches
     - update the files in the "patches" dir and OPTIONALLY (if you type 'y') to
       run patch-update with the --make option (which opens a shell on error)
 """)
     ans = input("<Press Enter OR 'y' to continue> ")
 
-    s = cmd_run(['git', 'commit', '-a', '-m', f'Preparing for release of {version}'])
+    s = cmd_run(['git', 'commit', '-a', '-m', f'Preparing for release of {version} [buildall]'])
     if s.returncode:
         die('Aborting')
 
-    cmd_chk('make gen')
+    cmd_chk('touch configure.ac && packaging/smart-make && make gen')
 
-    print(f'Creating any missing patch branches.')
+    print('Creating any missing patch branches.')
     s = cmd_run(f'packaging/branch-from-patch --branch={args.master_branch} --add-missing')
     if s.returncode:
         die('Aborting')
@@ -282,13 +300,13 @@ About to:
     # TODO: is there a better way to ensure that our passphrase is in the agent?
     cmd_run("touch TeMp; gpg --sign TeMp; rm TeMp*")
 
-    out = cmd_txt(f"git tag -s -m 'Version {version}.' {v_ver}", capture='combined')
+    out = cmd_txt(f"git tag -s -m 'Version {version}.' {v_ver}", capture='combined').out
     print(out, end='')
     if 'bad passphrase' in out or 'failed' in out:
         die('Aborting')
 
     if os.path.isdir('patches/.git'):
-        out = cmd_txt(f"cd patches && git tag -s -m 'Version {version}.' {v_ver}", capture='combined')
+        out = cmd_txt(f"cd patches && git tag -s -m 'Version {version}.' {v_ver}", capture='combined').out
         print(out, end='')
         if 'bad passphrase' in out or 'failed' in out:
             die('Aborting')
@@ -327,7 +345,7 @@ About to:
     md_files = 'README.md NEWS.md INSTALL.md'.split()
     html_files = [ fn for fn in gen_pathnames if fn.endswith('.html') ]
     cmd_chk(['rsync', '-a', *md_files, *html_files, dest])
-    cmd_chk(["packaging/md2html"] + [ dest +'/'+ fn for fn in md_files ])
+    cmd_chk(["./md-convert", "--dest", dest, *md_files])
 
     cmd_chk(f"git log --name-status | gzip -9 >{dest}/ChangeLog.gz")
 

packaging/smart-make

@@ -4,8 +4,6 @@ set -e
 
 export LANG=C
 
-make=`which gmake 2>/dev/null` || make=`which make 2>/dev/null`
-
 branch=`packaging/prep-auto-dir`
 if test x"$branch" = x; then
     srcdir=.
@@ -40,8 +38,8 @@ fi
 
 ./config.status
 
-$make all
+make all
 
 if test x"$1" = x"check"; then
-    $make check
+    make check
 fi

packaging/systemd/rsync.service

@@ -7,6 +7,7 @@ Documentation=man:rsync(1) man:rsyncd.conf(5)
 [Service]
 ExecStart=/usr/bin/rsync --daemon --no-detach
 RestartSec=1
+Restart=on-failure
 
 # Citing README.md:
 #

packaging/var-checker

@@ -4,7 +4,7 @@
 # for vars that are defined but not used, and for inconsistent array
 # sizes.  Run it from inside the main rsync directory.
 
-import re, argparse, glob
+import os, sys, re, argparse, glob
 
 VARS_RE = re.compile(r'^(?!(?:extern|enum)\s)([a-zA-Z]\S*\s+.*);', re.M)
 EXTERNS_RE = re.compile(r'^extern\s+(.*);', re.M)
@@ -15,8 +15,15 @@ def main():
     add_syscall_c = set('t_stub.c t_unsafe.c tls.c trimslash.c'.split())
     add_util_c = set('t_stub.c t_unsafe.c'.split())
 
+    if not os.path.exists('syscall.c'):
+        if os.path.exists('var-checker'):
+            os.chdir('..')
+        else:
+            print("Couldn't find the source dir.")
+            sys.exit(1)
+
     syscall_c = slurp_file('syscall.c', True)
-    util_c = slurp_file('util.c', True)
+    util_c = slurp_file('util1.c', True)
 
     for fn in sorted(glob.glob('*.c')):
         txt = slurp_file(fn)

prepare-source

@@ -5,8 +5,8 @@
 # Specify one action or more than one to provide a fall-back:
 #
 #   build     build the config files [the default w/no arg]
-#   fetch     fetch the latest dev config files
-#   fetchgen  fetch all the latest dev generated files
+#   fetch     fetch the latest dev autoconfig files
+#   fetchgen  fetch all the latest dev generated files (including manpages)
 #   fetchSRC  fetch the latest dev source files [NON-GENERATED FILES]
 #
 # The script stops after the first successful action.
@@ -32,7 +32,7 @@ if test "$dir" != '.'; then
 	fi
     done
     for fn in configure.sh config.h.in aclocal.m4; do
-	test ! -f $fn -a -f "$dir/$fn" && cp -p "$dir/$fn" $fn
+	test ! -f $fn && test -f "$dir/$fn" && cp -p "$dir/$fn" $fn
     done
 fi
 

prepare-source.mak

@@ -1,7 +1,7 @@
 SHELL=/bin/sh
 
-.PHONY: conf
 conf: configure.sh config.h.in
+.PHONY: conf
 
 aclocal.m4: m4/*.m4
 	aclocal -I m4

progress.c

@@ -4,7 +4,7 @@
  * Copyright (C) 1996-2000 Andrew Tridgell
  * Copyright (C) 1996 Paul Mackerras
  * Copyright (C) 2001, 2002 Martin Pool <mbp@samba.org>
- * Copyright (C) 2003-2020 Wayne Davison
+ * Copyright (C) 2003-2022 Wayne Davison
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -115,13 +115,13 @@ static void rprint_progress(OFF_T ofs, OFF_T size, struct timeval *now, int is_l
 		units = "kB/s";
 	}
 
-	if (remain < 0)
+	if (remain < 0 || remain > 9999.0 * 3600.0)
 		strlcpy(rembuf, "  ??:??:??", sizeof rembuf);
 	else {
-		snprintf(rembuf, sizeof rembuf, "%4d:%02d:%02d",
-			 (int) (remain / 3600.0),
-			 (int) (remain / 60.0) % 60,
-			 (int) remain % 60);
+		snprintf(rembuf, sizeof rembuf, "%4u:%02u:%02u",
+			 (unsigned int) (remain / 3600.0),
+			 (unsigned int) (remain / 60.0) % 60,
+			 (unsigned int) remain % 60);
 	}
 
 	output_needs_newline = 0;

receiver.c

@@ -3,7 +3,7 @@
  *
  * Copyright (C) 1996-2000 Andrew Tridgell
  * Copyright (C) 1996 Paul Mackerras
- * Copyright (C) 2003-2020 Wayne Davison
+ * Copyright (C) 2003-2023 Wayne Davison
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -41,6 +41,7 @@ extern int preserve_hard_links;
 extern int preserve_perms;
 extern int write_devices;
 extern int preserve_xattrs;
+extern int do_fsync;
 extern int basis_dir_cnt;
 extern int make_backups;
 extern int cleanup_got_literal;
@@ -55,7 +56,6 @@ extern int inplace;
 extern int inplace_partial;
 extern int allowed_lull;
 extern int delay_updates;
-extern int xfersum_type;
 extern BOOL want_progress_now;
 extern mode_t orig_umask;
 extern struct stats stats;
@@ -67,6 +67,9 @@ extern struct file_list *cur_flist, *first_flist, *dir_flist;
 extern filter_rule_list daemon_filter_list;
 extern OFF_T preallocated_len;
 
+extern struct name_num_item *xfer_sum_nni;
+extern int xfer_sum_len;
+
 static struct bitbag *delayed_bits = NULL;
 static int phase = 0, redoing = 0;
 static flist_ndx_list batch_redo_list;
@@ -239,7 +242,6 @@ static int receive_data(int f_in, char *fname_r, int fd_r, OFF_T size_r,
 	static char file_sum1[MAX_DIGEST_LEN];
 	struct map_struct *mapbuf;
 	struct sum_struct sum;
-	int sum_len;
 	int32 len;
 	OFF_T total_size = F_LENGTH(file);
 	OFF_T offset = 0;
@@ -279,7 +281,7 @@ static int receive_data(int f_in, char *fname_r, int fd_r, OFF_T size_r,
 	} else
 		mapbuf = NULL;
 
-	sum_init(xfersum_type, checksum_seed);
+	sum_init(xfer_sum_nni, checksum_seed);
 
 	if (append_mode > 0) {
 		OFF_T j;
@@ -370,7 +372,7 @@ static int receive_data(int f_in, char *fname_r, int fd_r, OFF_T size_r,
 
 	if (fd != -1 && offset > 0) {
 		if (sparse_files > 0) {
-			if (sparse_end(fd, offset) != 0)
+			if (sparse_end(fd, offset, updating_basis_or_equiv) != 0)
 				goto report_write_error;
 		} else if (flush_write_file(fd) < 0) {
 		    report_write_error:
@@ -392,15 +394,20 @@ static int receive_data(int f_in, char *fname_r, int fd_r, OFF_T size_r,
 	if (INFO_GTE(PROGRESS, 1))
 		end_progress(total_size);
 
-	sum_len = sum_end(file_sum1);
+	sum_end(file_sum1);
+
+	if (do_fsync && fd != -1 && fsync(fd) != 0) {
+		rsyserr(FERROR, errno, "fsync failed on %s", full_fname(fname));
+		exit_cleanup(RERR_FILEIO);
+	}
 
 	if (mapbuf)
 		unmap_file(mapbuf);
 
-	read_buf(f_in, sender_file_sum, sum_len);
+	read_buf(f_in, sender_file_sum, xfer_sum_len);
 	if (DEBUG_GTE(DELTASUM, 2))
 		rprintf(FINFO,"got file_sum\n");
-	if (fd != -1 && memcmp(file_sum1, sender_file_sum, sum_len) != 0)
+	if (fd != -1 && memcmp(file_sum1, sender_file_sum, xfer_sum_len) != 0)
 		return 0;
 	return 1;
 }
@@ -433,9 +440,8 @@ static void handle_delayed_updates(char *local_name)
 					"rename failed for %s (from %s)",
 					full_fname(fname), partialptr);
 			} else {
-				if (remove_source_files
-				 || (preserve_hard_links && F_IS_HLINKED(file)))
-					send_msg_int(MSG_SUCCESS, ndx);
+				if (remove_source_files || (preserve_hard_links && F_IS_HLINKED(file)))
+					send_msg_success(fname, ndx);
 				handle_partial_dir(partialptr, PDIR_DELETE);
 			}
 		}
@@ -539,6 +545,9 @@ int recv_files(int f_in, int f_out, char *local_name)
 	if (delay_updates)
 		delayed_bits = bitbag_create(cur_flist->used + 1);
 
+	if (whole_file < 0)
+		whole_file = 0;
+
 	progress_init();
 
 	while (1) {
@@ -584,10 +593,13 @@ int recv_files(int f_in, int f_out, char *local_name)
 		if (DEBUG_GTE(RECV, 1))
 			rprintf(FINFO, "recv_files(%s)\n", fname);
 
-		if (daemon_filter_list.head && (*fname != '.' || fname[1] != '\0')
-		 && check_filter(&daemon_filter_list, FLOG, fname, 0) < 0) {
-			rprintf(FERROR, "attempt to hack rsync failed.\n");
-			exit_cleanup(RERR_PROTOCOL);
+		if (daemon_filter_list.head && (*fname != '.' || fname[1] != '\0')) {
+			int filt_flags = S_ISDIR(file->mode) ? NAME_IS_DIR : NAME_IS_FILE;
+			if (check_filter(&daemon_filter_list, FLOG, fname, filt_flags) < 0) {
+				rprintf(FERROR, "ERROR: rejecting file transfer request for daemon excluded file: %s\n",
+					fname);
+				exit_cleanup(RERR_PROTOCOL);
+			}
 		}
 
 #ifdef SUPPORT_XATTRS
@@ -686,7 +698,7 @@ int recv_files(int f_in, int f_out, char *local_name)
 			if (!am_server)
 				discard_receive_data(f_in, file);
 			if (inc_recurse)
-				send_msg_int(MSG_SUCCESS, ndx);
+				send_msg_success(fname, ndx);
 			continue;
 		}
 
@@ -799,14 +811,16 @@ int recv_files(int f_in, int f_out, char *local_name)
 			continue;
 		}
 
-		if (fd1 != -1 && !(S_ISREG(st.st_mode) || (write_devices && IS_DEVICE(st.st_mode)))) {
+		if (write_devices && IS_DEVICE(st.st_mode)) {
+			if (fd1 != -1 && st.st_size == 0)
+				st.st_size = get_device_size(fd1, fname);
+			/* Mark the file entry as a device so that we don't try to truncate it later on. */
+			file->mode = S_IFBLK | (file->mode & ACCESSPERMS);
+		} else if (fd1 != -1 && !(S_ISREG(st.st_mode))) {
 			close(fd1);
 			fd1 = -1;
 		}
 
-		if (fd1 != -1 && IS_DEVICE(st.st_mode) && st.st_size == 0)
-			st.st_size = get_device_size(fd1, fname);
-
 		/* If we're not preserving permissions, change the file-list's
 		 * mode based on the local permissions and some heuristics. */
 		if (!preserve_perms) {
@@ -826,6 +840,12 @@ int recv_files(int f_in, int f_out, char *local_name)
 		if (inplace || one_inplace)  {
 			fnametmp = one_inplace ? partialptr : fname;
 			fd2 = do_open(fnametmp, O_WRONLY|O_CREAT, 0600);
+#ifdef linux
+			if (fd2 == -1 && errno == EACCES) {
+				/* Maybe the error was due to protected_regular setting? */
+				fd2 = do_open(fname, O_WRONLY, 0600);
+			}
+#endif
 			if (fd2 == -1) {
 				rsyserr(FERROR_XFER, errno, "open %s failed",
 					full_fname(fnametmp));
@@ -878,7 +898,7 @@ int recv_files(int f_in, int f_out, char *local_name)
 					do_unlink(partialptr);
 				handle_partial_dir(partialptr, PDIR_DELETE);
 			}
-		} else if (keep_partial && partialptr && !one_inplace) {
+		} else if (keep_partial && partialptr && (!one_inplace || delay_updates)) {
 			if (!handle_partial_dir(partialptr, PDIR_CREATE)) {
 				rprintf(FERROR,
 					"Unable to create partial-dir for %s -- discarding %s.\n",
@@ -906,13 +926,12 @@ int recv_files(int f_in, int f_out, char *local_name)
 		case 2:
 			break;
 		case 1:
-			if (remove_source_files || inc_recurse
-			 || (preserve_hard_links && F_IS_HLINKED(file)))
-				send_msg_int(MSG_SUCCESS, ndx);
+			if (remove_source_files || inc_recurse || (preserve_hard_links && F_IS_HLINKED(file)))
+				send_msg_success(fname, ndx);
 			break;
 		case 0: {
 			enum logcode msgtype = redoing ? FERROR_XFER : FWARNING;
-			if (msgtype == FERROR_XFER || INFO_GTE(NAME, 1)) {
+			if (msgtype == FERROR_XFER || INFO_GTE(NAME, 1) || stdout_format_has_i) {
 				char *errstr, *redostr, *keptstr;
 				if (!(keep_partial && partialptr) && !inplace)
 					keptstr = "discarded";

rsync-ssl

@@ -73,8 +73,16 @@ function rsync_ssl_helper {
 	certopt=""
 	gnutls_cert_opt=""
     else
-	certopt="cert$optsep$RSYNC_SSL_CERT"
-	gnutls_cert_opt="--x509keyfile=$RSYNC_SSL_CERT"
+	certopt="-cert$optsep$RSYNC_SSL_CERT"
+	gnutls_cert_opt="--x509certfile=$RSYNC_SSL_CERT"
+    fi
+
+    if [[ -z "$RSYNC_SSL_KEY" ]]; then
+	keyopt=""
+	gnutls_key_opt=""
+    else
+	keyopt="-key$optsep$RSYNC_SSL_KEY"
+	gnutls_key_opt="--x509keyfile=$RSYNC_SSL_KEY"
     fi
 
     if [[ -z ${RSYNC_SSL_CA_CERT+x} ]]; then
@@ -129,9 +137,9 @@ function rsync_ssl_helper {
     fi
 
     if [[ $RSYNC_SSL_TYPE == openssl ]]; then
-	exec $RSYNC_SSL_OPENSSL s_client $caopt $certopt -quiet -verify_quiet -servername $hostname -connect $hostname:$port
+	exec $RSYNC_SSL_OPENSSL s_client $caopt $certopt $keyopt -quiet -verify_quiet -servername $hostname -verify_hostname $hostname -connect $hostname:$port
     elif [[ $RSYNC_SSL_TYPE == gnutls ]]; then
-	exec $RSYNC_SSL_GNUTLS --logfile=/dev/null $gnutls_cert_opt $gnutls_opts $hostname:$port
+	exec $RSYNC_SSL_GNUTLS --logfile=/dev/null $gnutls_cert_opt $gnutls_key_opt $gnutls_opts $hostname:$port
     else
 	# devzero@web.de came up with this no-tmpfile calling syntax:
 	exec $RSYNC_SSL_STUNNEL -fd 10 11<&0 <<EOF 10<&0 0<&11 11<&-

rsync-ssl.1.md

@@ -1,14 +1,17 @@
-# NAME
+## NAME
 
 rsync-ssl - a helper script for connecting to an ssl rsync daemon
 
-# SYNOPSIS
+## SYNOPSIS
 
 ```
 rsync-ssl [--type=SSL_TYPE] RSYNC_ARGS
 ```
 
-# DESCRIPTION
+The online version of this manpage (that includes cross-linking of topics)
+is available at <https://download.samba.org/pub/rsync/rsync-ssl.1>.
+
+## DESCRIPTION
 
 The rsync-ssl script helps you to run an rsync copy to/from an rsync daemon
 that requires ssl connections.
@@ -20,7 +23,7 @@ environment.  You can specify an overriding port via `--port` or by including
 it in the normal spot in the URL format, though both of those require your
 rsync version to be at least 3.2.0.
 
-# OPTIONS
+## OPTIONS
 
 If the **first** arg is a `--type=SSL_TYPE` option, the script will only use
 that particular program to open an ssl connection instead of trying to find an
@@ -32,33 +35,56 @@ required for this particular option.
 All the other options are passed through to the rsync command, so consult the
 **rsync**(1) manpage for more information on how it works.
 
-# ENVIRONMENT VARIABLES
+## ENVIRONMENT VARIABLES
 
 The ssl helper scripts are affected by the following environment variables:
 
-0.  `RSYNC_SSL_TYPE` Specifies the program type that should be used to open the
-    ssl connection.  It must be one of `openssl` or `stunnel`.  The
-    `--type=SSL_TYPE` option overrides this, when specified.
-0.  `RSYNC_SSL_PORT` If specified, the value is the port number that is used as
-    the default when the user does not specify a port in their rsync command.
-    When not specified, the default port number is 874.  (Note that older rsync
-    versions (prior to 3.2.0) did not communicate an overriding port number
-    value to the helper script.)
-0.  `RSYNC_SSL_CERT` If specified, the value is a filename that contains a
-    certificate to use for the connection.
-0.  `RSYNC_SSL_CA_CERT` If specified, the value is a filename that contains a
-    certificate authority certificate that is used to validate the connection.
-0.  `RSYNC_SSL_OPENSSL` Specifies the openssl executable to run when the
-    connection type is set to openssl.  If unspecified, the $PATH is searched
-    for "openssl".
-0.  `RSYNC_SSL_GNUTLS` Specifies the gnutls-cli executable to run when the
-    connection type is set to gnutls.  If unspecified, the $PATH is searched
-    for "gnutls-cli".
-0.  `RSYNC_SSL_STUNNEL` Specifies the stunnel executable to run when the
-    connection type is set to stunnel.  If unspecified, the $PATH is searched
-    first for "stunnel4" and then for "stunnel".
+0.  `RSYNC_SSL_TYPE`
 
-# EXAMPLES
+    Specifies the program type that should be used to open the ssl connection.
+    It must be one of `openssl` or `stunnel`.  The `--type=SSL_TYPE` option
+    overrides this, when specified.
+
+0.  `RSYNC_SSL_PORT`
+
+    If specified, the value is the port number that is used as the default when
+    the user does not specify a port in their rsync command.  When not
+    specified, the default port number is 874.  (Note that older rsync versions
+    (prior to 3.2.0) did not communicate an overriding port number value to the
+    helper script.)
+
+0.  `RSYNC_SSL_CERT`
+
+    If specified, the value is a filename that contains a certificate to use
+    for the connection.
+
+0.  `RSYNC_SSL_KEY`
+
+    If specified, the value is a filename that contains a key for the provided
+    certificate to use for the connection.
+
+0.  `RSYNC_SSL_CA_CERT`
+
+    If specified, the value is a filename that contains a certificate authority
+    certificate that is used to validate the connection.
+
+0.  `RSYNC_SSL_OPENSSL`
+
+    Specifies the openssl executable to run when the connection type is set to
+    openssl.  If unspecified, the $PATH is searched for "openssl".
+
+0.  `RSYNC_SSL_GNUTLS`
+
+    Specifies the gnutls-cli executable to run when the connection type is set
+    to gnutls.  If unspecified, the $PATH is searched for "gnutls-cli".
+
+0.  `RSYNC_SSL_STUNNEL`
+
+    Specifies the stunnel executable to run when the connection type is set to
+    stunnel.  If unspecified, the $PATH is searched first for "stunnel4" and
+    then for "stunnel".
+
+## EXAMPLES
 
 >     rsync-ssl -aiv example.com::mod/ dest
 
@@ -68,11 +94,16 @@ The ssl helper scripts are affected by the following environment variables:
 
 >     rsync-ssl -aiv rsync://example.com:9874/mod/ dest
 
-# SEE ALSO
+## THE SERVER SIDE
 
-**rsync**(1), **rsyncd.conf**(5)
+For help setting up an SSL/TLS supporting rsync, see the [instructions in
+rsyncd.conf](rsyncd.conf.5#SSL_TLS_Daemon_Setup).
 
-# CAVEATS
+## SEE ALSO
+
+[**rsync**(1)](rsync.1), [**rsyncd.conf**(5)](rsyncd.conf.5)
+
+## CAVEATS
 
 Note that using an stunnel connection requires at least version 4 of stunnel,
 which should be the case on modern systems.  Also, it does not verify a
@@ -85,23 +116,23 @@ release the gnutls-cli command was dropping output, making it unusable.  If
 that bug has been fixed in your version, feel free to put gnutls into an
 exported RSYNC_SSL_TYPE environment variable to make its use the default.
 
-# BUGS
+## BUGS
 
 Please report bugs! See the web site at <https://rsync.samba.org/>.
 
-# VERSION
+## VERSION
 
-This man page is current for version @VERSION@ of rsync.
+This manpage is current for version @VERSION@ of rsync.
 
-# CREDITS
+## CREDITS
 
-rsync is distributed under the GNU General Public License.  See the file
-COPYING for details.
+Rsync is distributed under the GNU General Public License.  See the file
+[COPYING](COPYING) for details.
 
 A web site is available at <https://rsync.samba.org/>.  The site includes an
 FAQ-O-Matic which may cover questions unanswered by this manual page.
 
-# AUTHOR
+## AUTHOR
 
 This manpage was written by Wayne Davison.
 

rsync.c

@@ -3,7 +3,7 @@
  *
  * Copyright (C) 1996 Andrew Tridgell
  * Copyright (C) 1996 Paul Mackerras
- * Copyright (C) 2003-2020 Wayne Davison
+ * Copyright (C) 2003-2022 Wayne Davison
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -32,7 +32,9 @@ extern int preserve_acls;
 extern int preserve_xattrs;
 extern int preserve_perms;
 extern int preserve_executability;
-extern int preserve_times;
+extern int preserve_mtimes;
+extern int omit_dir_times;
+extern int omit_link_times;
 extern int am_root;
 extern int am_server;
 extern int am_daemon;
@@ -63,8 +65,7 @@ extern char *iconv_opt;
 #define UPDATED_ATIME (1<<3)
 #define UPDATED_ACLS  (1<<4)
 #define UPDATED_MODE  (1<<5)
-
-#define UPDATED_TIMES (UPDATED_MTIME|UPDATED_ATIME)
+#define UPDATED_CRTIME (1<<6)
 
 #ifdef ICONV_CONST
 iconv_t ic_chck = (iconv_t)-1;
@@ -576,17 +577,20 @@ int set_file_attrs(const char *fname, struct file_struct *file, stat_x *sxp,
 		set_xattr(fname, file, fnamecmp, sxp);
 #endif
 
-	if (!preserve_times
-	 || (!(preserve_times & PRESERVE_DIR_TIMES) && S_ISDIR(sxp->st.st_mode))
-	 || (!(preserve_times & PRESERVE_LINK_TIMES) && S_ISLNK(sxp->st.st_mode)))
-		flags |= ATTRS_SKIP_MTIME | ATTRS_SKIP_ATIME;
-	else if (sxp != &sx2)
-		memcpy(&sx2.st, &sxp->st, sizeof (sx2.st));
-	if (!atimes_ndx || S_ISDIR(sxp->st.st_mode))
-		flags |= ATTRS_SKIP_ATIME;
-	/* Don't set the creation date on the root folder of an HFS+ volume. */
-	if (sxp->st.st_ino == 2 && S_ISDIR(sxp->st.st_mode))
-		flags |= ATTRS_SKIP_CRTIME;
+	if ((omit_dir_times && S_ISDIR(sxp->st.st_mode))
+	 || (omit_link_times && S_ISLNK(sxp->st.st_mode)))
+		flags |= ATTRS_SKIP_MTIME | ATTRS_SKIP_ATIME | ATTRS_SKIP_CRTIME;
+	else {
+		if (!preserve_mtimes)
+			flags |= ATTRS_SKIP_MTIME;
+		if (!atimes_ndx || S_ISDIR(sxp->st.st_mode))
+			flags |= ATTRS_SKIP_ATIME;
+		/* Don't set the creation date on the root folder of an HFS+ volume. */
+		if (sxp->st.st_ino == 2 && S_ISDIR(sxp->st.st_mode))
+			flags |= ATTRS_SKIP_CRTIME;
+	}
+	if (sxp != &sx2)
+		memcpy(&sx2.st, &sxp->st, sizeof sx2.st);
 	if (!(flags & ATTRS_SKIP_MTIME) && !same_mtime(file, &sxp->st, flags & ATTRS_ACCURATE_TIME)) {
 		sx2.st.st_mtime = file->modtime;
 #ifdef ST_MTIME_NSEC
@@ -604,33 +608,41 @@ int set_file_attrs(const char *fname, struct file_struct *file, stat_x *sxp,
 			updated |= UPDATED_ATIME;
 		}
 	}
-	if (updated & UPDATED_TIMES) {
-		int ret = set_times(fname, &sx2.st);
-		if (ret < 0) {
-			rsyserr(FERROR_XFER, errno, "failed to set times on %s",
-				full_fname(fname));
-			goto cleanup;
-		}
-		if (ret > 0) { /* ret == 1 if symlink could not be set */
-			updated &= ~UPDATED_TIMES;
-			file->flags |= FLAG_TIME_FAILED;
-		}
-	}
 #ifdef SUPPORT_CRTIMES
 	if (crtimes_ndx && !(flags & ATTRS_SKIP_CRTIME)) {
 		time_t file_crtime = F_CRTIME(file);
 		if (sxp->crtime == 0)
-			sxp->crtime = get_create_time(fname);
-		if (!same_time(sxp->crtime, 0L, file_crtime, 0L)
-		 && set_create_time(fname, file_crtime) == 0)
-			updated = 1;
+			sxp->crtime = get_create_time(fname, &sxp->st);
+		if (!same_time(sxp->crtime, 0L, file_crtime, 0L)) {
+			if (
+#ifdef HAVE_GETATTRLIST
+			     do_setattrlist_crtime(fname, file_crtime) == 0
+#elif defined __CYGWIN__
+			     do_SetFileTime(fname, file_crtime) == 0
+#else
+#error Unknown crtimes implementation
+#endif
+			)
+				updated |= UPDATED_CRTIME;
+		}
 	}
 #endif
+	if (updated & (UPDATED_MTIME|UPDATED_ATIME)) {
+		int ret = set_times(fname, &sx2.st);
+		if (ret < 0) {
+			rsyserr(FERROR_XFER, errno, "failed to set times on %s", full_fname(fname));
+			goto cleanup;
+		}
+		if (ret > 0) { /* ret == 1 if symlink could not be set */
+			updated &= ~(UPDATED_MTIME|UPDATED_ATIME);
+			file->flags |= FLAG_TIME_FAILED;
+		}
+	}
 
 #ifdef SUPPORT_ACLS
 	/* It's OK to call set_acl() now, even for a dir, as the generator
 	 * will enable owner-writability using chmod, if necessary.
-	 * 
+	 *
 	 * If set_acl() changes permission bits in the process of setting
 	 * an access ACL, it changes sxp->st.st_mode so we know whether we
 	 * need to chmod(). */

rsync.h

@@ -2,7 +2,7 @@
  * Copyright (C) 1996, 2000 Andrew Tridgell
  * Copyright (C) 1996 Paul Mackerras
  * Copyright (C) 2001, 2002 Martin Pool <mbp@samba.org>
- * Copyright (C) 2003-2020 Wayne Davison
+ * Copyright (C) 2003-2022 Wayne Davison
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -108,12 +108,22 @@
 #define BITS_EQUAL(b1,b2,mask) (((unsigned)(b1) & (unsigned)(mask)) \
 			     == ((unsigned)(b2) & (unsigned)(mask)))
 
-/* update this if you make incompatible changes */
+/* Update this if you make incompatible changes and ALSO update the
+ * SUBPROTOCOL_VERSION if it is not a final (official) release. */
 #define PROTOCOL_VERSION 31
 
-/* This is used when working on a new protocol version in CVS, and should
- * be a new non-zero value for each CVS change that affects the protocol.
- * It must ALWAYS be 0 when the protocol goes final (and NEVER before)! */
+/* This is used when working on a new protocol version or for any unofficial
+ * protocol tweaks.  It should be a non-zero value for each pre-release repo
+ * change that affects the protocol. The official pre-release versions should
+ * start with 1 (after incrementing the PROTOCOL_VERSION) and go up by 1 for
+ * each new protocol change.  For unofficial changes, pick a fairly large
+ * random number that will hopefully not collide with anyone else's unofficial
+ * protocol.  It must ALWAYS be 0 when the protocol goes final (and official)
+ * and NEVER before!  When rsync negotiates a protocol match, it will only
+ * allow the newest protocol to be used if the SUBPROTOCOL_VERSION matches.
+ * All older protocol versions MUST be compatible with the final, official
+ * release of the protocol, so don't tweak the code to change the protocol
+ * behavior for an older protocol version. */
 #define SUBPROTOCOL_VERSION 0
 
 /* We refuse to interoperate with versions that are not in this range.
@@ -267,6 +277,10 @@ enum msgcode {
 	MSG_NO_SEND=102,/* sender failed to open a file we wanted */
 };
 
+enum filetype {
+	FT_UNSUPPORTED, FT_REG, FT_DIR, FT_SYMLINK, FT_SPECIAL, FT_DEVICE
+};
+
 #define NDX_DONE -1
 #define NDX_FLIST_EOF -2
 #define NDX_DEL_STATS -3
@@ -300,7 +314,6 @@ enum delret {
 #include "errcode.h"
 
 #include "config.h"
-#include "version.h"
 
 /* The default RSYNC_RSH is always set in config.h. */
 
@@ -325,6 +338,9 @@ enum delret {
 # endif
 # include <string.h>
 #endif
+#ifdef HAVE_BSD_STRING_H
+# include <bsd/string.h>
+#endif
 #ifdef HAVE_STRINGS_H
 # include <strings.h>
 #endif
@@ -350,16 +366,10 @@ enum delret {
 #include <sys/socket.h>
 #endif
 
-#ifdef TIME_WITH_SYS_TIME
-#include <sys/time.h>
-#include <time.h>
-#else
 #ifdef HAVE_SYS_TIME_H
 #include <sys/time.h>
-#else
+#endif
 #include <time.h>
-#endif
-#endif
 
 #ifdef HAVE_FCNTL_H
 #include <fcntl.h>
@@ -478,7 +488,6 @@ enum delret {
 #ifndef __TANDEM
 #define MAKEDEV(devmajor,devminor) makedev(devmajor,devminor)
 #else
-# include <sys/stat.h>
 # define major DEV_TO_MAJOR
 # define minor DEV_TO_MINOR
 # define MAKEDEV MAJORMINOR_TO_DEV
@@ -570,7 +579,11 @@ typedef unsigned int size_t;
 #endif
 #endif
 
-#ifdef HAVE_GETATTRLIST
+#if !defined __APPLE__ || defined HAVE_GETATTRLIST
+#define SUPPORT_ATIMES 1
+#endif
+
+#if defined HAVE_GETATTRLIST || defined __CYGWIN__
 #define SUPPORT_CRTIMES 1
 #endif
 
@@ -762,6 +775,11 @@ struct ht_int64_node {
 
 #if defined __STDC_VERSION__ && __STDC_VERSION__ >= 199901L
 #define USE_FLEXIBLE_ARRAY 1
+#define SIZE_T_FMT_MOD "z" /* printf supports %zd */
+#define SIZE_T_FMT_CAST size_t
+#else
+#define SIZE_T_FMT_MOD "l" /* printf supports %ld */
+#define SIZE_T_FMT_CAST long
 #endif
 
 union file_extras {
@@ -802,6 +820,7 @@ extern int uid_ndx;
 extern int gid_ndx;
 extern int acls_ndx;
 extern int xattrs_ndx;
+extern int file_sum_extra_cnt;
 
 #ifdef USE_FLEXIBLE_ARRAY
 #define FILE_STRUCT_LEN (sizeof (struct file_struct))
@@ -812,7 +831,7 @@ extern int xattrs_ndx;
 #define DEV_EXTRA_CNT 2
 #define DIRNODE_EXTRA_CNT 3
 #define EXTRA64_CNT ((sizeof (union file_extras64) + EXTRA_LEN - 1) / EXTRA_LEN)
-#define SUM_EXTRA_CNT ((MAX_DIGEST_LEN + EXTRA_LEN - 1) / EXTRA_LEN)
+#define SUM_EXTRA_CNT file_sum_extra_cnt
 
 #define REQ_EXTRA(f,ndx) ((union file_extras*)(f) - (ndx))
 #define OPT_EXTRA(f,bump) ((union file_extras*)(f) - file_extra_cnt - 1 - (bump))
@@ -900,8 +919,9 @@ extern int xattrs_ndx;
  * Start the flist array at FLIST_START entries and grow it
  * by doubling until FLIST_LINEAR then grow by FLIST_LINEAR
  */
-#define FLIST_START	(32 * 1024)
-#define FLIST_LINEAR	(FLIST_START * 512)
+#define FLIST_START		(32)
+#define FLIST_START_LARGE	(32 * 1024)
+#define FLIST_LINEAR		(FLIST_START_LARGE * 512)
 
 /*
  * Extent size for allocation pools: A minimum size of 128KB
@@ -998,6 +1018,7 @@ typedef struct filter_struct {
 		int slash_cnt;
 		struct filter_list_struct *mergelist;
 	} u;
+	uchar elide;
 } filter_rule;
 
 typedef struct filter_list_struct {
@@ -1137,19 +1158,23 @@ typedef struct {
 #define NSTR_COMPRESS 1
 
 struct name_num_item {
-	int num;
-	const char *name, *main_name;
+	int num, flags;
+	const char *name;
+	struct name_num_item *main_nni;
 };
 
 struct name_num_obj {
 	const char *type;
-	const char *negotiated_name;
+	struct name_num_item *negotiated_nni;
 	uchar *saw;
 	int saw_len;
-	int negotiated_num;
-	struct name_num_item list[8]; /* A big-enough len (we'll get a compile error if it is ever too small) */
+	struct name_num_item *list;
 };
 
+#ifdef EXTERNAL_ZLIB
+#define read_buf read_buf_
+#endif
+
 #ifndef __cplusplus
 #include "proto.h"
 #endif
@@ -1308,10 +1333,6 @@ extern int errno;
 #define IS_SPECIAL(mode) (S_ISSOCK(mode) || S_ISFIFO(mode))
 #define IS_DEVICE(mode) (S_ISCHR(mode) || S_ISBLK(mode))
 
-#define PRESERVE_FILE_TIMES	(1<<0)
-#define PRESERVE_DIR_TIMES	(1<<1)
-#define PRESERVE_LINK_TIMES	(1<<2)
-
 /* Initial mask on permissions given to temporary files.  Mask off setuid
      bits and group access because of potential race-condition security
      holes, and mask other access because mode 707 is bizarre */
@@ -1320,15 +1341,15 @@ extern int errno;
 /* handler for null strings in printf format */
 #define NS(s) ((s)?(s):"<NULL>")
 
-extern char *do_malloc;
+extern char *do_calloc;
 
 /* Convenient wrappers for malloc and realloc.  Use them. */
-#define new(type) ((type*)my_alloc(do_malloc, sizeof (type), 1, __FILE__, __LINE__))
-#define new0(type) ((type*)my_alloc(NULL, sizeof (type), 1, __FILE__, __LINE__))
+#define new(type) ((type*)my_alloc(NULL, sizeof (type), 1, __FILE__, __LINE__))
+#define new0(type) ((type*)my_alloc(do_calloc, sizeof (type), 1, __FILE__, __LINE__))
 #define realloc_buf(ptr, num) my_alloc((ptr), (num), 1, __FILE__, __LINE__)
 
-#define new_array(type, num) ((type*)my_alloc(do_malloc, (num), sizeof (type), __FILE__, __LINE__))
-#define new_array0(type, num) ((type*)my_alloc(NULL, (num), sizeof (type), __FILE__, __LINE__))
+#define new_array(type, num) ((type*)my_alloc(NULL, (num), sizeof (type), __FILE__, __LINE__))
+#define new_array0(type, num) ((type*)my_alloc(do_calloc, (num), sizeof (type), __FILE__, __LINE__))
 #define realloc_array(ptr, type, num) ((type*)my_alloc((ptr), (num), sizeof (type), __FILE__, __LINE__))
 
 #undef strdup
@@ -1401,7 +1422,8 @@ extern short info_levels[], debug_levels[];
 #define INFO_MISC (INFO_FLIST+1)
 #define INFO_MOUNT (INFO_MISC+1)
 #define INFO_NAME (INFO_MOUNT+1)
-#define INFO_PROGRESS (INFO_NAME+1)
+#define INFO_NONREG (INFO_NAME+1)
+#define INFO_PROGRESS (INFO_NONREG+1)
 #define INFO_REMOVE (INFO_PROGRESS+1)
 #define INFO_SKIP (INFO_REMOVE+1)
 #define INFO_STATS (INFO_SKIP+1)
@@ -1456,3 +1478,9 @@ const char *get_panic_action(void);
     fprintf(stderr, "%s in %s at line %d\n", msg, __FILE__, __LINE__); \
     exit_cleanup(RERR_UNSUPPORTED); \
 } while (0)
+
+#ifdef HAVE_MALLINFO2
+#define MEM_ALLOC_INFO mallinfo2
+#elif defined HAVE_MALLINFO
+#define MEM_ALLOC_INFO mallinfo
+#endif

rsyncd.conf.5.md

@@ -1,12 +1,15 @@
-# NAME
+## NAME
 
 rsyncd.conf - configuration file for rsync in daemon mode
 
-# SYNOPSIS
+## SYNOPSIS
 
 rsyncd.conf
 
-# DESCRIPTION
+The online version of this manpage (that includes cross-linking of topics)
+is available at <https://download.samba.org/pub/rsync/rsyncd.conf.5>.
+
+## DESCRIPTION
 
 The rsyncd.conf file is the runtime configuration file for rsync when run as an
 rsync daemon.
@@ -14,7 +17,7 @@ rsync daemon.
 The rsyncd.conf file controls authentication, access, logging and available
 modules.
 
-# FILE FORMAT
+## FILE FORMAT
 
 The file consists of modules and parameters. A module begins with the name of
 the module in square brackets and continues until the next module begins.
@@ -40,10 +43,9 @@ The values following the equals sign in parameters are all either a string (no
 quotes needed) or a boolean, which may be given as yes/no, 0/1 or true/false.
 Case is not significant in boolean values, but is preserved in string values.
 
-# LAUNCHING THE RSYNC DAEMON
+## LAUNCHING THE RSYNC DAEMON
 
-The rsync daemon is launched by specifying the `--daemon` option to
-rsync.
+The rsync daemon is launched by specifying the `--daemon` option to rsync.
 
 The daemon must run with root privileges if you wish to use chroot, to bind to
 a port numbered under 1024 (as is the default 873), or to set file ownership.
@@ -60,45 +62,27 @@ When run via inetd you should add a line like this to /etc/services:
 
 and a single line something like this to /etc/inetd.conf:
 
->     rsync   stream  tcp     nowait  root   /usr/bin/rsync rsyncd --daemon
+>     rsync   stream  tcp     nowait  root   @BINDIR@/rsync rsyncd --daemon
 
-Replace "/usr/bin/rsync" with the path to where you have rsync installed on
+Replace "@BINDIR@/rsync" with the path to where you have rsync installed on
 your system.  You will then need to send inetd a HUP signal to tell it to
 reread its config file.
 
 Note that you should **not** send the rsync daemon a HUP signal to force it to
 reread the `rsyncd.conf` file. The file is re-read on each client connection.
 
-# GLOBAL PARAMETERS
+## GLOBAL PARAMETERS
 
 The first parameters in the file (before a [module] header) are the global
-parameters.  Rsync also allows for the use of a "[global]" module name to
-indicate the start of one or more global-parameter sections (the name must be
-lower case).
-
-You may also include any module parameters in the global part of the config
-file in which case the supplied value will override the default for that
-parameter.
-
-You may use references to environment variables in the values of parameters.
-String parameters will have %VAR% references expanded as late as possible (when
-the string is first used in the program), allowing for the use of variables
-that rsync sets at connection time, such as RSYNC_USER_NAME.  Non-string
-parameters (such as true/false settings) are expanded when read from the config
-file.  If a variable does not exist in the environment, or if a sequence of
-characters is not a valid reference (such as an un-paired percent sign), the
-raw characters are passed through unchanged.  This helps with backward
-compatibility and safety (e.g. expanding a non-existent %VAR% to an empty
-string in a path could result in a very unsafe path).  The safest way to insert
-a literal % into a value is to use %%.
+parameters:
 
 [comment]: # (An OL starting at 0 is converted into a DL by the parser.)
 
 0.  `motd file`
 
-    This parameter allows you to specify a "message of the day" to display to
-    clients on each connect. This usually contains site information and any
-    legal notices. The default is no motd file.  This can be overridden by the
+    This parameter allows you to specify a "message of the day" (MOTD) to display
+    to clients on each connect. This usually contains site information and any
+    legal notices. The default is no MOTD file.  This can be overridden by the
     `--dparam=motdfile=FILE` command-line option when starting the daemon.
 
 0.  `pid file`
@@ -126,7 +110,7 @@ a literal % into a value is to use %%.
 
     This parameter can provide endless fun for people who like to tune their
     systems to the utmost degree. You can set all sorts of socket options which
-    may make transfers faster (or slower!). Read the man page for the
+    may make transfers faster (or slower!). Read the manpage for the
     **setsockopt()** system call for details on some of the options you may be
     able to set. By default no special socket options are set.  These settings
     can also be specified via the `--sockopts` command-line option.
@@ -136,7 +120,23 @@ a literal % into a value is to use %%.
     You can override the default backlog value when the daemon listens for
     connections.  It defaults to 5.
 
-# MODULE PARAMETERS
+You may also include any [MODULE PARAMETERS](#) in the global part of the
+config file, in which case the supplied value will override the default for
+that parameter.
+
+You may use references to environment variables in the values of parameters.
+String parameters will have %VAR% references expanded as late as possible (when
+the string is first used in the program), allowing for the use of variables
+that rsync sets at connection time, such as RSYNC_USER_NAME.  Non-string
+parameters (such as true/false settings) are expanded when read from the config
+file.  If a variable does not exist in the environment, or if a sequence of
+characters is not a valid reference (such as an un-paired percent sign), the
+raw characters are passed through unchanged.  This helps with backward
+compatibility and safety (e.g. expanding a non-existent %VAR% to an empty
+string in a path could result in a very unsafe path).  The safest way to insert
+a literal % into a value is to use %%.
+
+## MODULE PARAMETERS
 
 After the global parameters you should define a number of modules, each module
 exports a directory tree as a symbolic name. Modules are exported by specifying
@@ -144,11 +144,17 @@ a module name in square brackets [module] followed by the parameters for that
 module.  The module name cannot contain a slash or a closing square bracket.
 If the name contains whitespace, each internal sequence of whitespace will be
 changed into a single space, while leading or trailing whitespace will be
-discarded.  Also, the name cannot be "global" as that exact name indicates that
-global parameters follow (see above).
+discarded.
 
-As with GLOBAL PARAMETERS, you may use references to environment variables in
-the values of parameters.  See the GLOBAL PARAMETERS section for more details.
+There is also a special module name of "[global]" that does not define a module
+but instead switches back to the global settings context where default
+parameters can be specified.  Because each defined module gets its full set of
+parameters as a combination of the default values that are set at that position
+in the config file plus its own parameter list, the use of a "[global]" section
+can help to maintain shared config values for multiple modules.
+
+As with [GLOBAL PARAMETERS](#), you may use references to environment variables
+in the values of parameters.  See that section for details.
 
 0.  `comment`
 
@@ -162,6 +168,16 @@ the values of parameters.  See the GLOBAL PARAMETERS section for more details.
     available in this module.  You must specify this parameter for each module
     in `rsyncd.conf`.
 
+    If the value contains a "/./" element then the path will be divided at that
+    point into a chroot dir and an inner-chroot subdir.  If [`use chroot`](#)
+    is set to false, though, the extraneous dot dir is just cleaned out of the
+    path.  An example of this idiom is:
+
+    >     path = /var/rsync/./module1
+
+    This will (when chrooting) chroot to "/var/rsync" and set the inside-chroot
+    path to "/module1".
+
     You may base the path's value off of an environment variable by surrounding
     the variable name with percent signs.  You can even reference a variable
     that is set by rsync when the user connects.  For example, this would use
@@ -177,7 +193,7 @@ the values of parameters.  See the GLOBAL PARAMETERS section for more details.
 
 0.  `use chroot`
 
-    If "use chroot" is true, the rsync daemon will chroot to the "path" before
+    If "use chroot" is true, the rsync daemon will chroot to the "[path](#)" before
     starting the file transfer with the client.  This has the advantage of
     extra protection against possible implementation security holes, but it has
     the disadvantages of requiring super-user privileges, of not being able to
@@ -185,30 +201,48 @@ the values of parameters.  See the GLOBAL PARAMETERS section for more details.
     path, and of complicating the preservation of users and groups by name (see
     below).
 
-    As an additional safety feature, you can specify a dot-dir in the module's
-    "path" to indicate the point where the chroot should occur.  This allows
-    rsync to run in a chroot with a non-"/" path for the top of the transfer
-    hierarchy.  Doing this guards against unintended library loading (since
-    those absolute paths will not be inside the transfer hierarchy unless you
-    have used an unwise pathname), and lets you setup libraries for the chroot
-    that are outside of the transfer.  For example, specifying
-    "/var/rsync/./module1" will chroot to the "/var/rsync" directory and set
-    the inside-chroot path to "/module1".  If you had omitted the dot-dir, the
-    chroot would have used the whole path, and the inside-chroot path would
-    have been "/".
+    If `use chroot` is not set, it defaults to trying to enable a chroot but
+    allows the daemon to continue (after logging a warning) if it fails. The
+    one exception to this is when a module's [`path`](#) has a "/./" chroot
+    divider in it -- this causes an unset value to be treated as true for that
+    module.
 
-    When both "use chroot" and "daemon chroot" are false, OR the inside-chroot
-    path of "use chroot" is not "/", rsync will: (1) munge symlinks by default
-    for security reasons (see "munge symlinks" for a way to turn this off, but
-    only if you trust your users), (2) substitute leading slashes in absolute
-    paths with the module's path (so that options such as `--backup-dir`,
-    `--compare-dest`, etc. interpret an absolute path as rooted in the module's
-    "path" dir), and (3) trim ".." path elements from args if rsync believes
-    they would escape the module hierarchy.  The default for "use chroot" is
-    true, and is the safer choice (especially if the module is not read-only).
+    Prior to rsync 3.2.7, the default value was "true".  The new "unset"
+    default makes it easier to setup an rsync daemon as a non-root user or to
+    run a daemon on a system where chroot fails.  Explicitly setting the value
+    to "true" in rsyncd.conf will always require the chroot to succeed.
 
-    When this parameter is enabled *and* the "name converter" parameter is
-    *not* set, the "numeric ids" parameter will default to being enabled
+    It is also possible to specify a dot-dir in the module's "[path](#)" to
+    indicate that you want to chdir to the earlier part of the path and then
+    serve files from inside the latter part of the path (with sanitizing and
+    default symlink munging).  This can be useful if you need some library dirs
+    inside the chroot (typically for uid & gid lookups) but don't want to put
+    the lib dir into the top of the served path (even though they can be hidden
+    with an [`exclude`](#) directive).  However, a better choice for a modern
+    rsync setup is to use a [`name converter`](#)" and try to avoid inner lib
+    dirs altogether.  See also the [`daemon chroot`](#) parameter, which causes
+    rsync to chroot into its own chroot area before doing any path-related
+    chrooting.
+
+    If the daemon is serving the "/" dir (either directly or due to being
+    chrooted to the module's path), rsync does not do any path sanitizing or
+    (default) munging.
+
+    When it has to limit access to a particular subdir (either due to chroot
+    being disabled or having an inside-chroot path set), rsync will munge
+    symlinks (by default) and sanitize paths.  Those that dislike munged
+    symlinks (and really, really trust their users to not break out of the
+    subdir) can disable the symlink munging via the "[munge symlinks](#)"
+    parameter.
+
+    When rsync is sanitizing paths, it trims ".." path elements from args that
+    it believes would escape the module hierarchy. It also substitutes leading
+    slashes in absolute paths with the module's path (so that options such as
+    `--backup-dir` & `--compare-dest` interpret an absolute path as rooted in
+    the module's "[path](#)" dir).
+
+    When a chroot is in effect *and* the "[name converter](#)" parameter is
+    *not* set, the "[numeric ids](#)" parameter will default to being enabled
     (disabling name lookups).  This means that if you manually setup
     name-lookup libraries in your chroot (instead of using a name converter)
     that you need to explicitly set `numeric ids = false` for rsync to do name
@@ -217,16 +251,16 @@ the values of parameters.  See the GLOBAL PARAMETERS section for more details.
     If you copy library resources into the module's chroot area, you should
     protect them through your OS's normal user/group or ACL settings (to
     prevent the rsync module's user from being able to change them), and then
-    hide them from the user's view via "exclude" (see how in the discussion of
+    hide them from the user's view via "[exclude](#)" (see how in the discussion of
     that parameter).  However, it's easier and safer to setup a name converter.
 
 0.  `daemon chroot`
 
     This parameter specifies a path to which the daemon will chroot before
-    beginning communication with clients. Module paths (and any "use chroot"
+    beginning communication with clients. Module paths (and any "[use chroot](#)"
     settings) will then be related to this one. This lets you choose if you
     want the whole daemon to be chrooted (with this setting), just the
-    transfers to be chrooted (with "use chroot"), or both.  Keep in mind that
+    transfers to be chrooted (with "[use chroot](#)"), or both.  Keep in mind that
     the "daemon chroot" area may need various OS/lib/etc files installed to
     allow the daemon to function.  By default the daemon runs without any
     chrooting.
@@ -284,11 +318,11 @@ the values of parameters.  See the GLOBAL PARAMETERS section for more details.
     transfer behave as if the client had passed the `--numeric-ids`
     command-line option.  By default, this parameter is enabled for chroot
     modules and disabled for non-chroot modules.  Also keep in mind that
-    uid/gid preservation requires the module to be running as root (see "uid")
-    or for "fake super" to be configured.
+    uid/gid preservation requires the module to be running as root (see "[uid](#)")
+    or for "[fake super](#)" to be configured.
 
     A chroot-enabled module should not have this parameter set to false unless
-    you're using a "name converter" program *or* you've taken steps to ensure
+    you're using a "[name converter](#)" program *or* you've taken steps to ensure
     that the module has the necessary resources it needs to translate names and
     that it is not possible for a user to change those resources.
 
@@ -298,12 +332,12 @@ the values of parameters.  See the GLOBAL PARAMETERS section for more details.
     (non-daemon-affecting) `--munge-links` command-line option (using a method
     described below).  This should help protect your files from user trickery
     when your daemon module is writable.  The default is disabled when
-    "use chroot" is on with an inside-chroot path of "/", OR if "daemon chroot"
+    "[use chroot](#)" is on with an inside-chroot path of "/", OR if "[daemon chroot](#)"
     is on, otherwise it is enabled.
 
     If you disable this parameter on a daemon that is not read-only, there are
     tricks that a user can play with uploaded symlinks to access
-    daemon-excluded items (if your module has any), and, if "use chroot" is
+    daemon-excluded items (if your module has any), and, if "[use chroot](#)" is
     off, rsync can even be tricked into showing or changing data that is
     outside the module's path (as access-permissions allow).
 
@@ -324,7 +358,7 @@ the values of parameters.  See the GLOBAL PARAMETERS section for more details.
     the source code named "munge-symlinks" that can be used to add or remove
     this prefix from your symlinks.
 
-    When this parameter is disabled on a writable module and "use chroot" is
+    When this parameter is disabled on a writable module and "[use chroot](#)" is
     off (or the inside-chroot path is not "/"), incoming symlinks will be
     modified to drop a leading slash and to remove ".." path elements that
     rsync believes will allow a symlink to escape the module's hierarchy.
@@ -340,10 +374,10 @@ the values of parameters.  See the GLOBAL PARAMETERS section for more details.
     conversion in a chroot module without extra files in the chroot area, and
     also ensures that name-translation is done in a consistent manner.  If the
     "charset" parameter is not set, the `--iconv` option is refused, just as if
-    "iconv" had been specified via "refuse options".
+    "iconv" had been specified via "[refuse options](#)".
 
     If you wish to force users to always use `--iconv` for a particular module,
-    add "no-iconv" to the "refuse options" parameter.  Keep in mind that this
+    add "no-iconv" to the "[refuse options](#)" parameter.  Keep in mind that this
     will restrict access to your module to very new rsync clients.
 
 0.  `max connections`
@@ -352,7 +386,7 @@ the values of parameters.  See the GLOBAL PARAMETERS section for more details.
     connections you will allow.  Any clients connecting when the maximum has
     been reached will receive a message telling them to try later.  The default
     is 0, which means no limit.  A negative value disables the module.  See
-    also the "lock file" parameter.
+    also the "[lock file](#)" parameter.
 
 0.  `log file`
 
@@ -381,7 +415,7 @@ the values of parameters.  See the GLOBAL PARAMETERS section for more details.
     facility name which is defined on your system. Common names are auth,
     authpriv, cron, daemon, ftp, kern, lpr, mail, news, security, syslog, user,
     uucp, local0, local1, local2, local3, local4, local5, local6 and local7.
-    The default is daemon.  This setting has no effect if the "log file"
+    The default is daemon.  This setting has no effect if the "[log file](#)"
     setting is a non-empty string (either set in the per-modules settings, or
     inherited from the global settings).
 
@@ -389,7 +423,7 @@ the values of parameters.  See the GLOBAL PARAMETERS section for more details.
 
     This parameter allows you to specify the syslog tag to use when logging
     messages from the rsync daemon. The default is "rsyncd".  This setting has
-    no effect if the "log file" setting is a non-empty string (either set in
+    no effect if the "[log file](#)" setting is a non-empty string (either set in
     the per-modules settings, or inherited from the global settings).
 
     For example, if you wanted each authenticated user's name to be included in
@@ -414,7 +448,7 @@ the values of parameters.  See the GLOBAL PARAMETERS section for more details.
 
 0.  `lock file`
 
-    This parameter specifies the file to use to support the "max connections"
+    This parameter specifies the file to use to support the "[max connections](#)"
     parameter. The rsync daemon uses record locking on this file to ensure that
     the max connections limit is not exceeded for the modules sharing the lock
     file.  The default is `/var/run/rsyncd.lock`.
@@ -426,7 +460,7 @@ the values of parameters.  See the GLOBAL PARAMETERS section for more details.
     "read only" is false then uploads will be possible if file permissions on
     the daemon side allow them. The default is for all modules to be read only.
 
-    Note that "auth users" can override this setting on a per-user basis.
+    Note that "[auth users](#)" can override this setting on a per-user basis.
 
 0.  `write only`
 
@@ -460,8 +494,8 @@ the values of parameters.  See the GLOBAL PARAMETERS section for more details.
     This parameter determines whether this module is listed when the client
     asks for a listing of available modules.  In addition, if this is false,
     the daemon will pretend the module does not exist when a client denied by
-    "hosts allow" or "hosts deny" attempts to access it.  Realize that if
-    "reverse lookup" is disabled globally but enabled for the module, the
+    "[hosts allow](#)" or "[hosts deny](#)" attempts to access it.  Realize that if
+    "[reverse lookup](#)" is disabled globally but enabled for the module, the
     resulting reverse lookup to a potentially client-controlled DNS server may
     still reveal to the client that it hit an existing module.  The default is
     for modules to be listable.
@@ -470,10 +504,10 @@ the values of parameters.  See the GLOBAL PARAMETERS section for more details.
 
     This parameter specifies the user name or user ID that file transfers to
     and from that module should take place as when the daemon was run as root.
-    In combination with the "gid" parameter this determines what file
+    In combination with the "[gid](#)" parameter this determines what file
     permissions are available. The default when run by a super-user is to
     switch to the system's "nobody" user.  The default for a non-super-user is
-    to not try to change the user.  See also the "gid" parameter.
+    to not try to change the user.  See also the "[gid](#)" parameter.
 
     The RSYNC_USER_NAME environment variable may be used to request that rsync
     run as the authorizing user.  For example, if you want a rsync to run as
@@ -489,7 +523,7 @@ the values of parameters.  See the GLOBAL PARAMETERS section for more details.
     accessing the module.  The first one will be the default group, and any
     extra ones be set as supplemental groups.  You may also specify a "`*`" as
     the first gid in the list, which will be replaced by all the normal groups
-    for the transfer's user (see "uid").  The default when run by a super-user
+    for the transfer's user (see "[uid](#)").  The default when run by a super-user
     is to switch to your OS's "nobody" (or perhaps "nogroup") group with no
     other supplementary groups.  The default for a non-super-user is to not
     change any group attributes (and indeed, your OS may not allow a
@@ -505,13 +539,13 @@ the values of parameters.  See the GLOBAL PARAMETERS section for more details.
 
     This parameter specifies a uid under which the daemon will run. The daemon
     usually runs as user root, and when this is left unset the user is left
-    unchanged. See also the "uid" parameter.
+    unchanged. See also the "[uid](#)" parameter.
 
 0.  `daemon gid`
 
     This parameter specifies a gid under which the daemon will run. The daemon
     usually runs as group root, and when this is left unset, the group is left
-    unchanged. See also the "gid" parameter.
+    unchanged. See also the "[gid](#)" parameter.
 
 0.  `fake super`
 
@@ -532,8 +566,8 @@ the values of parameters.  See the GLOBAL PARAMETERS section for more details.
     or tampering with private administrative files, such as files you may add
     to support uid/gid name translations.
 
-    The daemon filter chain is built from the "filter", "include from",
-    "include", "exclude from", and "exclude" parameters, in that order of
+    The daemon filter chain is built from the "filter", "[include from](#)",
+    "[include](#)", "[exclude from](#)", and "[exclude](#)" parameters, in that order of
     priority.  Anchored patterns are anchored at the root of the module.  To
     prevent access to an entire subtree, for example, "`/secret`", you **must**
     exclude everything in the subtree; the easiest way to do this is with a
@@ -560,8 +594,8 @@ the values of parameters.  See the GLOBAL PARAMETERS section for more details.
 
 0.  `include`
 
-    Use an "include" to override the effects of the "exclude" parameter.  Only
-    one "include" parameter can apply to a given module.  See the "filter"
+    Use an "include" to override the effects of the "[exclude](#)" parameter.  Only
+    one "include" parameter can apply to a given module.  See the "[filter](#)"
     parameter for a description of how excluded files affect the daemon.
 
 0.  `exclude from`
@@ -569,14 +603,14 @@ the values of parameters.  See the GLOBAL PARAMETERS section for more details.
     This parameter specifies the name of a file on the daemon that contains
     daemon exclude patterns, one per line.  Only one "exclude from" parameter
     can apply to a given module; if you have multiple exclude-from files, you
-    can specify them as a merge file in the "filter" parameter.  See the
-    "filter" parameter for a description of how excluded files affect the
+    can specify them as a merge file in the "[filter](#)" parameter.  See the
+    "[filter](#)" parameter for a description of how excluded files affect the
     daemon.
 
 0.  `include from`
 
-    Analogue of "exclude from" for a file of daemon include patterns.  Only one
-    "include from" parameter can apply to a given module.  See the "filter"
+    Analogue of "[exclude from](#)" for a file of daemon include patterns.  Only one
+    "include from" parameter can apply to a given module.  See the "[filter](#)"
     parameter for a description of how excluded files affect the daemon.
 
 0.  `incoming chmod`
@@ -611,7 +645,7 @@ the values of parameters.  See the GLOBAL PARAMETERS section for more details.
     to supply a username and password to connect to the module. A challenge
     response authentication protocol is used for this exchange. The plain text
     usernames and passwords are stored in the file specified by the
-    "secrets file" parameter. The default is for all users to be able to
+    "[secrets file](#)" parameter. The default is for all users to be able to
     connect without a password (this is called "anonymous rsync").
 
     In addition to username matching, you can specify groupname matching via a
@@ -623,7 +657,7 @@ the values of parameters.  See the GLOBAL PARAMETERS section for more details.
     Finally, options may be specified after a colon (:).  The options allow you
     to "deny" a user or a group, set the access to "ro" (read-only), or set the
     access to "rw" (read/write).  Setting an auth-rule-specific ro/rw setting
-    overrides the module's "read only" setting.
+    overrides the module's "[read only](#)" setting.
 
     Be sure to put the rules in the order you want them to be matched, because
     the checking stops at the first matching user or group, and that is the
@@ -661,7 +695,7 @@ the values of parameters.  See the GLOBAL PARAMETERS section for more details.
 
     This parameter specifies the name of a file that contains the
     username:password and/or @groupname:password pairs used for authenticating
-    this module. This file is only consulted if the "auth users" parameter is
+    this module. This file is only consulted if the "[auth users](#)" parameter is
     specified.  The file is line-based and contains one name:password pair per
     line.  Any line has a hash (#) as the very first character on the line is
     considered a comment and is skipped.  The passwords can contain any
@@ -675,14 +709,14 @@ the values of parameters.  See the GLOBAL PARAMETERS section for more details.
     "@groupname:password" line for the group that triggered the authentication.
 
     It is up to you what kind of password entries you want to include, either
-    users, groups, or both.  The use of group rules in "auth users" does not
+    users, groups, or both.  The use of group rules in "[auth users](#)" does not
     require that you specify a group password if you do not want to use shared
     passwords.
 
     There is no default for the "secrets file" parameter, you must choose a
     name (such as `/etc/rsyncd.secrets`).  The file must normally not be
-    readable by "other"; see "strict modes".  If the file is not found or is
-    rejected, no logins for a "user auth" module will be possible.
+    readable by "other"; see "[strict modes](#)".  If the file is not found or is
+    rejected, no logins for an "[auth users](#)" module will be possible.
 
 0.  `strict modes`
 
@@ -714,11 +748,11 @@ the values of parameters.  See the GLOBAL PARAMETERS section for more details.
       addresses which match the masked IP address will be allowed in.
     - a hostname pattern using wildcards. If the hostname of the connecting IP
       (as determined by a reverse lookup) matches the wildcarded name (using
-      the same rules as normal unix filename matching), the client is allowed
-      in.  This only works if "reverse lookup" is enabled (the default).
+      the same rules as normal Unix filename matching), the client is allowed
+      in.  This only works if "[reverse lookup](#)" is enabled (the default).
     - a hostname. A plain hostname is matched against the reverse DNS of the
-      connecting IP (if "reverse lookup" is enabled), and/or the IP of the
-      given hostname is matched against the connecting IP (if "forward lookup"
+      connecting IP (if "[reverse lookup](#)" is enabled), and/or the IP of the
+      given hostname is matched against the connecting IP (if "[forward lookup](#)"
       is enabled, as it is by default).  Any match will be allowed in.
     - an '@' followed by a netgroup name, which will match if the reverse DNS
       of the connecting IP is in the specified netgroup.
@@ -730,11 +764,11 @@ the values of parameters.  See the GLOBAL PARAMETERS section for more details.
     >     fe80::%link1/64
     >     fe80::%link1/ffff:ffff:ffff:ffff::
 
-    You can also combine "hosts allow" with "hosts deny" as a way to add
+    You can also combine "hosts allow" with "[hosts deny](#)" as a way to add
     exceptions to your deny list.  When both parameters are specified, the
     "hosts allow" parameter is checked first and a match results in the client
     being able to connect.  A non-allowed host is then matched against the
-    "hosts deny" list to see if it should be rejected.  A host that does not
+    "[hosts deny](#)" list to see if it should be rejected.  A host that does not
     match either list is allowed to connect.
 
     The default is no "hosts allow" parameter, which means all hosts can
@@ -745,7 +779,7 @@ the values of parameters.  See the GLOBAL PARAMETERS section for more details.
     This parameter allows you to specify a list of comma- and/or
     whitespace-separated patterns that are matched against a connecting clients
     hostname and IP address. If the pattern matches then the connection is
-    rejected. See the "hosts allow" parameter for more information.
+    rejected. See the "[hosts allow](#)" parameter for more information.
 
     The default is no "hosts deny" parameter, which means all hosts can
     connect.
@@ -753,8 +787,8 @@ the values of parameters.  See the GLOBAL PARAMETERS section for more details.
 0.  `reverse lookup`
 
     Controls whether the daemon performs a reverse lookup on the client's IP
-    address to determine its hostname, which is used for "hosts allow" &
-    "hosts deny" checks and the "%h" log escape.  This is enabled by default,
+    address to determine its hostname, which is used for "[hosts allow](#)" &
+    "[hosts deny](#)" checks and the "%h" log escape.  This is enabled by default,
     but you may wish to disable it to save time if you know the lookup will not
     return a useful result, in which case the daemon will use the name
     "UNDETERMINED" instead.
@@ -794,7 +828,7 @@ the values of parameters.  See the GLOBAL PARAMETERS section for more details.
     logs the transfer at the end, so if a transfer is aborted, no mention will
     be made in the log file.
 
-    If you want to customize the log lines, see the "log format" parameter.
+    If you want to customize the log lines, see the "[log format](#)" parameter.
 
 0.  `log format`
 
@@ -811,7 +845,7 @@ the values of parameters.  See the GLOBAL PARAMETERS section for more details.
     (e.g. "`%''l %'b %f`").
 
     The default log format is "`%o %h [%a] %m (%u) %f %l`", and a "`%t [%p] `"
-    is always prefixed when using the "log file" parameter.  (A perl script
+    is always prefixed when using the "[log file](#)" parameter.  (A perl script
     that will summarize this default log format is included in the rsync source
     code distribution in the "support" subdirectory: rsyncstats.)
 
@@ -892,7 +926,7 @@ the values of parameters.  See the GLOBAL PARAMETERS section for more details.
     >     refuse options = * !a !v !compress*
 
     Don't worry that the "`*`" will refuse certain vital options such as
-    `--dry-run`, `--server`, `--no-iconv`, `--protect-args`, etc. These
+    `--dry-run`, `--server`, `--no-iconv`, `--seclude-args`, etc. These
     important options are not matched by wild-card, so they must be overridden
     by their exact name.  For instance, if you're forcing iconv transfers you
     could use something like this:
@@ -922,17 +956,19 @@ the values of parameters.  See the GLOBAL PARAMETERS section for more details.
 
     >     refuse options = * !a !delete* delete-after
 
-    A note on refusing "compress" -- it is better to set the "dont compress"
-    daemon parameter to "`*`" because that disables compression silently
+    A note on refusing "compress": it may be better to set the "[dont compress](#)"
+    daemon parameter to "`*`" and ensure that `RSYNC_COMPRESS_LIST=zlib` is set
+    in the environment of the daemon in order to disable compression silently
     instead of returning an error that forces the client to remove the `-z`
     option.
 
-    If you are un-refusing the compress option, you probably want to match
-    "`!compress*`" so that you also accept the `--compress-level` option.
+    If you are un-refusing the compress option, you may want to match
+    "`!compress*`" if you also want to allow the `--compress-level` option.
 
-    Note that the "write-devices" option is refused by default, but can be
-    explicitly accepted with "`!write-devices`".  The options "log-file" and
-    "log-file-format" are forcibly refused and cannot be accepted.
+    Note that the "copy-devices" & "write-devices" options are refused by
+    default, but they can be explicitly accepted with "`!copy-devices`" and/or
+    "`!write-devices`".  The options "log-file" and "log-file-format" are
+    forcibly refused and cannot be accepted.
 
     Here are all the options that are not matched by wild-cards:
 
@@ -942,18 +978,22 @@ the values of parameters.  See the GLOBAL PARAMETERS section for more details.
       receiver.  While rsync passes the older alias `--log-format` for
       compatibility reasons, this options should not be confused with
       `--log-file-format`.
-    - `--sender`: Use "write only" parameter instead of refusing this.
+    - `--sender`: Use "[write only](#)" parameter instead of refusing this.
     - `--dry-run`, `-n`: Who would want to disable this?
-    - `--protect-args`, `-s`: This actually makes transfers safer.
+    - `--seclude-args`, `-s`: Is the oldest arg-protection method.
     - `--from0`, `-0`: Makes it easier to accept/refuse `--files-from` without
       affecting this helpful modifier.
-    - `--iconv`: This is auto-disabled based on "charset" parameter.
+    - `--iconv`: This is auto-disabled based on "[charset](#)" parameter.
     - `--no-iconv`: Most transfers use this option.
     - `--checksum-seed`: Is a fairly rare, safe option.
     - `--write-devices`: Is non-wild but also auto-disabled.
 
 0.  `dont compress`
 
+    **NOTE:** This parameter currently has no effect except in one instance: if
+    it is set to "`*`" then it minimizes or disables compression for all files
+    (for those that don't want to refuse the `--compress` option completely).
+
     This parameter allows you to select filenames based on wildcard patterns
     that should not be compressed when pulling files from the daemon (no
     analogous parameter exists to govern the pushing of files to a daemon).
@@ -964,14 +1004,14 @@ the values of parameters.  See the GLOBAL PARAMETERS section for more details.
     The "dont compress" parameter takes a space-separated list of
     case-insensitive wildcard patterns. Any source filename matching one of the
     patterns will be compressed as little as possible during the transfer.  If
-    the compression algorithm has an "off" level (such as zlib/zlibx) then no
-    compression occurs for those files.  Other algorithms have the level
-    minimized to reduces the CPU usage as much as possible.
+    the compression algorithm has an "off" level, then no compression occurs
+    for those files.  If an algorithms has the ability to change the level in
+    mid-stream, it will be minimized to reduce the CPU usage as much as
+    possible.
 
     See the `--skip-compress` parameter in the **rsync**(1) manpage for the
-    list of file suffixes that are not compressed by default.  Specifying a
-    value for the "dont compress" parameter changes the default when the daemon
-    is the sender.
+    list of file suffixes that are skipped by default if this parameter is not
+    set.
 
 0.  `early exec`, `pre-xfer exec`, `post-xfer exec`
 
@@ -983,7 +1023,7 @@ the values of parameters.  See the GLOBAL PARAMETERS section for more details.
     _not_ displayed if the script returns success.  The other programs cannot
     send any text to the user.  All output except for the `pre-xfer exec`
     stdout goes to the corresponding daemon's stdout/stderr, which is typically
-    discarded.  See the `--no-detatch` option for a way to see the daemon's
+    discarded.  See the `--no-detach` option for a way to see the daemon's
     output, which can assist with debugging.
 
     Note that the `early exec` command runs before any part of the transfer
@@ -1033,7 +1073,7 @@ the values of parameters.  See the GLOBAL PARAMETERS section for more details.
     **system()** call's default shell), and use RSYNC_NO_XFER_EXEC to disable
     both options completely.
 
-# CONFIG DIRECTIVES
+## CONFIG DIRECTIVES
 
 There are currently two config directives available that allow a config file to
 incorporate the contents of other files:  `&include` and `&merge`.  Both allow
@@ -1088,7 +1128,7 @@ This would merge any `/etc/rsyncd.d/*.inc` files (for global values that should
 stay in effect), and then include any `/etc/rsyncd.d/*.conf` files (defining
 modules without any global-value cross-talk).
 
-# AUTHENTICATION STRENGTH
+## AUTHENTICATION STRENGTH
 
 The authentication protocol used in rsync is a 128 bit MD4 based challenge
 response system. This is fairly weak protection, though (with at least one
@@ -1103,18 +1143,18 @@ authentication is provided. Use ssh as the transport if you want encryption.
 You can also make use of SSL/TLS encryption if you put rsync behind an
 SSL proxy.
 
-# SSL/TLS Daemon Setup
+## SSL/TLS Daemon Setup
 
 When setting up an rsync daemon for access via SSL/TLS, you will need to
-configure a proxy (such as haproxy or nginx) as the front-end that handles the
-encryption.
+configure a TCP proxy (such as haproxy or nginx) as the front-end that handles
+the encryption.
 
 - You should limit the access to the backend-rsyncd port to only allow the
   proxy to connect.  If it is on the same host as the proxy, then configuring
   it to only listen on localhost is a good idea.
-- You should consider turning on the `proxy protocol` parameter if your proxy
-  supports sending that information.  The examples below assume that this is
-  enabled.
+- You should consider turning on the `proxy protocol` rsync-daemon parameter if
+  your proxy supports sending that information.  The examples below assume that
+  this is enabled.
 
 An example haproxy setup is as follows:
 
@@ -1141,14 +1181,14 @@ An example nginx proxy setup is as follows:
 >        ssl_certificate_key /etc/letsencrypt/example.com/privkey.pem;
 >
 >        proxy_pass localhost:873;
->        proxy_protocol on; # Requires "proxy protocol = true"
+>        proxy_protocol on; # Requires rsyncd.conf "proxy protocol = true"
 >        proxy_timeout 1m;
 >        proxy_connect_timeout 5s;
 >    }
 > }
 > ```
 
-# EXAMPLES
+## DAEMON CONFIG EXAMPLES
 
 A simple rsyncd.conf file that allow anonymous rsync to a ftp area at
 `/home/ftp` would be:
@@ -1197,46 +1237,41 @@ The /etc/rsyncd.secrets file would look something like this:
 >     tridge:mypass
 >     susan:herpass
 
-# FILES
+## FILES
 
 /etc/rsyncd.conf or rsyncd.conf
 
-# SEE ALSO
+## SEE ALSO
 
-**rsync**(1), **rsync-ssl**(1)
+[**rsync**(1)](rsync.1), [**rsync-ssl**(1)](rsync-ssl.1)
 
-# BUGS
+## BUGS
 
 Please report bugs! The rsync bug tracking system is online at
 <https://rsync.samba.org/>.
 
-# VERSION
+## VERSION
 
-This man page is current for version @VERSION@ of rsync.
+This manpage is current for version @VERSION@ of rsync.
 
-# CREDITS
+## CREDITS
 
-rsync is distributed under the GNU General Public License.  See the file
-COPYING for details.
+Rsync is distributed under the GNU General Public License.  See the file
+[COPYING](COPYING) for details.
 
-The primary ftp site for rsync is <ftp://rsync.samba.org/pub/rsync>
+An rsync web site is available at <https://rsync.samba.org/> and its github
+project is <https://github.com/WayneD/rsync>.
 
-A web site is available at <https://rsync.samba.org/>.
-
-We would be delighted to hear from you if you like this program.
-
-This program uses the zlib compression library written by Jean-loup Gailly and
-Mark Adler.
-
-# THANKS
+## THANKS
 
 Thanks to Warren Stanley for his original idea and patch for the rsync daemon.
 Thanks to Karsten Thygesen for his many suggestions and documentation!
 
-# AUTHOR
+## AUTHOR
 
-rsync was written by Andrew Tridgell and Paul Mackerras.  Many people have
-later contributed to it.
+Rsync was originally written by Andrew Tridgell and Paul Mackerras.  Many
+people have later contributed to it. It is currently maintained by Wayne
+Davison.
 
 Mailing lists for support and development are available at
 <https://lists.samba.org/>.

runtests.sh

@@ -1,7 +1,7 @@
 #! /bin/sh
 
 # Copyright (C) 2001, 2002 by Martin Pool <mbp@samba.org>
-# Copyright (C) 2003-2020 Wayne Davison
+# Copyright (C) 2003-2022 Wayne Davison
 
 # This program is free software; you can redistribute it and/or modify
 # it under the terms of the GNU General Public License version
@@ -155,7 +155,7 @@ if test x"$TOOLDIR" = x; then
     TOOLDIR=`pwd`
 fi
 srcdir=`dirname $0`
-if test x"$srcdir" = x -o x"$srcdir" = x.; then
+if test x"$srcdir" = x || test x"$srcdir" = x.; then
     srcdir="$TOOLDIR"
 fi
 if test x"$rsync_bin" = x; then
@@ -167,10 +167,10 @@ RSYNC="$rsync_bin $*"
 #RSYNC="valgrind $rsync_bin $*"
 
 TLS_ARGS=''
-if egrep '^#define HAVE_LUTIMES 1' config.h >/dev/null; then
+if grep -E '^#define HAVE_LUTIMES 1' config.h >/dev/null; then
     TLS_ARGS="$TLS_ARGS -l"
 fi
-if egrep '#undef CHOWN_MODIFIES_SYMLINK' config.h >/dev/null; then
+if grep -E '#undef CHOWN_MODIFIES_SYMLINK' config.h >/dev/null; then
     TLS_ARGS="$TLS_ARGS -L"
 fi
 
@@ -226,6 +226,8 @@ if [ ! -d "$srcdir" ]; then
     exit 2
 fi
 
+expect_skipped="${RSYNC_EXPECT_SKIPPED-IGNORE}"
+skipped_list=''
 skipped=0
 missing=0
 passed=0
@@ -236,7 +238,7 @@ failed=0
 # failure to aid investigation.  We don't remove the testtmp subdir at
 # the end so that it can be configured as a symlink to a filesystem that
 # has ACLs and xattr support enabled (if desired).
-scratchbase="$TOOLDIR"/testtmp
+scratchbase="${scratchbase:-$TOOLDIR}"/testtmp
 echo "    scratchbase=$scratchbase"
 [ -d "$scratchbase" ] || mkdir "$scratchbase"
 
@@ -265,10 +267,12 @@ maybe_discard_scratch() {
 
 if [ "x$whichtests" = x ]; then
     whichtests="*.test"
+    full_run=yes
+else
+    full_run=no
 fi
 
-for testscript in $suitedir/$whichtests
-do
+for testscript in $suitedir/$whichtests; do
     testbase=`echo $testscript | sed -e 's!.*/!!' -e 's/.test\$//'`
     scratchdir="$scratchbase/$testbase"
 
@@ -284,7 +288,7 @@ do
     result=$?
     set -e
 
-    if [ "x$always_log" = xyes -o \( $result != 0 -a $result != 77 -a $result != 78 \) ]
+    if [ "x$always_log" = xyes ] || ( [ $result != 0 ] && [ $result != 77 ] && [ $result != 78 ] )
     then
 	echo "----- $testbase log follows"
 	cat "$scratchdir/test.log"
@@ -306,6 +310,7 @@ do
 	# backticks will fill the whole file onto one line, which is a feature
 	whyskipped=`cat "$scratchdir/whyskipped"`
 	echo "SKIP    $testbase ($whyskipped)"
+	skipped_list="$skipped_list,$testbase"
 	skipped=`expr $skipped + 1`
 	maybe_discard_scratch
 	;;
@@ -331,6 +336,15 @@ echo "      $passed passed"
 [ "$failed" -gt 0 ]  && echo "      $failed failed"
 [ "$skipped" -gt 0 ] && echo "      $skipped skipped"
 [ "$missing" -gt 0 ] && echo "      $missing missing"
+if [ "$full_run" = yes ] && [ "$expect_skipped" != IGNORE ]; then
+    skipped_list=`echo "$skipped_list" | sed 's/^,//'`
+    echo "----- skipped results:"
+    echo "      expected: $expect_skipped"
+    echo "      got:      $skipped_list"
+else
+    skipped_list=''
+    expect_skipped=''
+fi
 echo '------------------------------------------------------------'
 
 # OK, so expr exits with 0 if the result is neither null nor zero; and
@@ -339,7 +353,7 @@ echo '------------------------------------------------------------'
 # because -e is set.
 
 result=`expr $failed + $missing || true`
-if [ "$result" = 0 -a "$skipped" -gt "${RSYNC_MAX_SKIPPED:-9999}" ]; then
+if [ "$result" = 0 ] && [ "$skipped_list" != "$expect_skipped" ]; then
     result=1
 fi
 echo "overall result is $result"

sender.c

@@ -3,7 +3,7 @@
  *
  * Copyright (C) 1996 Andrew Tridgell
  * Copyright (C) 1996 Paul Mackerras
- * Copyright (C) 2003-2020 Wayne Davison
+ * Copyright (C) 2003-2022 Wayne Davison
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -25,6 +25,7 @@
 extern int do_xfers;
 extern int am_server;
 extern int am_daemon;
+extern int local_server;
 extern int inc_recurse;
 extern int log_before_transfer;
 extern int stdout_format_has_i;
@@ -35,7 +36,9 @@ extern int append_mode;
 extern int copy_links;
 extern int io_error;
 extern int flist_eof;
+extern int whole_file;
 extern int allowed_lull;
+extern int copy_devices;
 extern int preserve_xattrs;
 extern int protocol_version;
 extern int remove_source_files;
@@ -49,6 +52,7 @@ extern int file_old_total;
 extern BOOL want_progress_now;
 extern struct stats stats;
 extern struct file_list *cur_flist, *first_flist, *dir_flist;
+extern char num_dev_ino_buf[4 + 8 + 8];
 
 BOOL extra_flist_sending_enabled;
 
@@ -142,6 +146,13 @@ void successful_send(int ndx)
 		goto failed;
 	}
 
+	if (local_server
+	 && (int64)st.st_dev == IVAL64(num_dev_ino_buf, 4)
+	 && (int64)st.st_ino == IVAL64(num_dev_ino_buf, 4 + 8)) {
+		rprintf(FERROR_XFER, "ERROR: Skipping sender remove of destination file: %s\n", fname);
+		return;
+	}
+
 	if (st.st_size != F_LENGTH(file) || st.st_mtime != file->modtime
 #ifdef ST_MTIME_NSEC
 	 || (NSEC_BUMP(file) && (uint32)st.ST_MTIME_NSEC != F_MOD_NSEC(file))
@@ -204,6 +215,9 @@ void send_files(int f_in, int f_out)
 	if (DEBUG_GTE(SEND, 1))
 		rprintf(FINFO, "send_files starting\n");
 
+	if (whole_file < 0)
+		whole_file = 0;
+
 	progress_init();
 
 	while (1) {
@@ -362,6 +376,25 @@ void send_files(int f_in, int f_out)
 			exit_cleanup(RERR_FILEIO);
 		}
 
+		if (IS_DEVICE(st.st_mode)) {
+			if (!copy_devices) {
+				rprintf(FERROR, "attempt to copy device contents without --copy-devices\n");
+				exit_cleanup(RERR_PROTOCOL);
+			}
+			if (st.st_size == 0)
+				st.st_size = get_device_size(fd, fname);
+		}
+
+		if (append_mode > 0 && st.st_size < F_LENGTH(file)) {
+			rprintf(FWARNING, "skipped diminished file: %s\n",
+				full_fname(fname));
+			free_sums(s);
+			close(fd);
+			if (protocol_version >= 30)
+				send_msg_int(MSG_NO_SEND, ndx);
+			continue;
+		}
+
 		if (st.st_size) {
 			int32 read_size = MAX(s->blength * 3, MAX_MAP_SIZE);
 			mbuf = map_file(fd, st.st_size, read_size, s->blength);

simd-checksum-avx2.S

@@ -0,0 +1,177 @@
+#include "config.h"
+
+#ifdef USE_ROLL_ASM /* { */
+
+#define CHAR_OFFSET 0 /* Keep this the same as rsync.h, which isn't likely to change. */
+
+#ifdef __APPLE__
+#define get_checksum1_avx2_asm  _get_checksum1_avx2_asm
+#endif
+
+.intel_syntax noprefix
+.text
+
+	.p2align 5
+	.globl get_checksum1_avx2_asm
+
+# rdi=*buf, esi=len, edx=i, rcx= *ps1, r8= *ps2
+get_checksum1_avx2_asm:
+	vmovd	xmm6,[rcx] # load *ps1
+	lea	eax, [rsi-128] # at least 128 bytes to process?
+	cmp	edx, eax
+	jg	.exit
+	lea	rax, .mul_T2[rip]
+	vmovntdqa ymm7, [rax]   # load T2 multiplication constants
+	vmovntdqa ymm12,[rax+32]# from memory.
+	vpcmpeqd  ymm15, ymm15, ymm15 # set all elements to -1.
+
+#if CHAR_OFFSET != 0
+	mov	eax, 32*CHAR_OFFSET
+	vmovd	xmm10, eax
+	vpbroadcastd	ymm10, xmm10
+	mov	eax, 528*CHAR_OFFSET
+	vmovd	xmm13, eax
+	vpbroadcastd ymm13, xmm13
+#endif
+	vpabsb	ymm15, ymm15  # set all byte size elements to 1.
+	add	rdi, rdx
+	vmovdqu ymm2, [rdi]   # preload the first 64 bytes.
+	vmovdqu ymm3, [rdi+32]
+	and	esi, ~63   # only needed during final reduction, 
+			   # done here to avoid a longer nop for 
+			   # alignment below.
+	add	edx, esi
+	shr	rsi, 6	   # longer opcode for alignment
+	add	rdi, 64
+	vpxor	xmm1, xmm1, xmm1 # reset both partial sums accumulators.
+	vpxor	xmm4, xmm4, xmm4
+	mov	eax, [r8]
+	.p2align 4 		# should fit into the LSD allocation queue.
+.loop:
+	vpmaddubsw	ymm0, ymm15, ymm2  # s1 partial sums
+	vpmaddubsw	ymm5, ymm15, ymm3 
+	vmovdqu	ymm8,  [rdi]	# preload the next 
+	vmovdqu	ymm9,  [rdi+32] # 64 bytes.
+	add	rdi, 64
+	vpaddd	ymm4, ymm4, ymm6
+	vpaddw	ymm5, ymm5, ymm0
+	vpsrld	ymm0, ymm5, 16
+	vpaddw	ymm5, ymm0, ymm5
+	vpaddd	ymm6, ymm5, ymm6
+	vpmaddubsw	ymm2, ymm7, ymm2  # s2 partial sums
+	vpmaddubsw	ymm3, ymm12, ymm3
+	prefetcht0	[rdi+384] # prefetch 6 cachelines ahead.
+	vpaddw	ymm3, ymm2, ymm3
+	vpsrldq	ymm2, ymm3, 2
+	vpaddd	ymm3, ymm2, ymm3
+	vpaddd	ymm1, ymm1, ymm3
+
+#if CHAR_OFFSET != 0
+	vpaddd ymm6, ymm10, ymm6  #  32*CHAR_OFFSET
+	vpaddd ymm1, ymm13, ymm1  # 528*CHAR_OFFSET
+#endif
+	vmovdqa ymm2, ymm8   # move the next 64 bytes 
+	vmovdqa ymm3, ymm9   # into the right registers
+	sub	esi, 1
+	jnz	.loop
+
+	# now we reduce the partial sums.
+	vpslld	ymm3, ymm4, 6
+	vpsrldq	ymm2, ymm6, 4
+	
+	vpaddd  ymm0, ymm3, ymm1
+	vpaddd	ymm6, ymm2, ymm6
+	vpsrlq	ymm3, ymm0, 32
+	
+	vpsrldq	ymm2, ymm6, 8
+	vpaddd	ymm0, ymm3, ymm0
+	vpsrldq	ymm3, ymm0, 8
+	vpaddd	ymm6, ymm2, ymm6
+	vpaddd	ymm0, ymm3, ymm0
+	vextracti128	xmm2, ymm6, 0x1
+	vextracti128	xmm1, ymm0, 0x1
+	vpaddd	xmm6, xmm2, xmm6
+	vmovd	[rcx], xmm6
+	vpaddd	xmm1, xmm1, xmm0
+	vmovd	ecx, xmm1
+	add	eax, ecx
+	mov	[r8], eax
+.exit:
+	vzeroupper
+	mov	eax, edx
+	ret
+
+#ifdef __APPLE__
+.data
+	.align 6
+#else
+.section	.rodata
+	.p2align 6 
+#endif
+.mul_T2: 
+	.byte 64
+	.byte 63
+	.byte 62
+	.byte 61
+	.byte 60
+	.byte 59
+	.byte 58
+	.byte 57
+	.byte 56
+	.byte 55
+	.byte 54
+	.byte 53
+	.byte 52
+	.byte 51
+	.byte 50
+	.byte 49
+	.byte 48
+	.byte 47
+	.byte 46
+	.byte 45
+	.byte 44
+	.byte 43
+	.byte 42
+	.byte 41
+	.byte 40
+	.byte 39
+	.byte 38
+	.byte 37
+	.byte 36
+	.byte 35
+	.byte 34
+	.byte 33
+	.byte 32
+	.byte 31
+	.byte 30
+	.byte 29
+	.byte 28
+	.byte 27
+	.byte 26
+	.byte 25
+	.byte 24
+	.byte 23
+	.byte 22
+	.byte 21
+	.byte 20
+	.byte 19
+	.byte 18
+	.byte 17
+	.byte 16
+	.byte 15
+	.byte 14
+	.byte 13
+	.byte 12
+	.byte 11
+	.byte 10
+	.byte 9
+	.byte 8
+	.byte 7
+	.byte 6
+	.byte 5
+	.byte 4
+	.byte 3
+	.byte 2
+	.byte 1
+
+#endif /* } USE_ROLL_ASM */

simd-checksum-x86_64.cpp

@@ -51,12 +51,12 @@
  * GCC 4.x are not supported to ease configure.ac logic.
  */
 
-#ifdef __x86_64__
-#ifdef __cplusplus
+#ifdef __x86_64__ /* { */
+#ifdef __cplusplus /* { */
 
 #include "rsync.h"
 
-#ifdef HAVE_SIMD
+#ifdef USE_ROLL_SIMD /* { */
 
 #include <immintrin.h>
 
@@ -85,7 +85,9 @@ typedef long long __m256i_u __attribute__((__vector_size__(32), __may_alias__, _
 #define SSE2_HADDS_EPI16(a, b) _mm_adds_epi16(SSE2_INTERLEAVE_EVEN_EPI16(a, b), SSE2_INTERLEAVE_ODD_EPI16(a, b))
 #define SSE2_MADDUBS_EPI16(a, b) _mm_adds_epi16(SSE2_MULU_EVEN_EPI8(a, b), SSE2_MULU_ODD_EPI8(a, b))
 
+#ifndef USE_ROLL_ASM
 __attribute__ ((target("default"))) MVSTATIC int32 get_checksum1_avx2_64(schar* buf, int32 len, int32 i, uint32* ps1, uint32* ps2) { return i; }
+#endif
 __attribute__ ((target("default"))) MVSTATIC int32 get_checksum1_ssse3_32(schar* buf, int32 len, int32 i, uint32* ps1, uint32* ps2) { return i; }
 __attribute__ ((target("default"))) MVSTATIC int32 get_checksum1_sse2_32(schar* buf, int32 len, int32 i, uint32* ps1, uint32* ps2) { return i; }
 
@@ -311,6 +313,12 @@ __attribute__ ((target("sse2"))) MVSTATIC int32 get_checksum1_sse2_32(schar* buf
     return i;
 }
 
+#ifdef USE_ROLL_ASM /* { */
+
+extern "C" __attribute__ ((target("avx2"))) int32 get_checksum1_avx2_asm(schar* buf, int32 len, int32 i, uint32* ps1, uint32* ps2);
+
+#else /* } { */
+
 /*
   AVX2 loop per 64 bytes:
     int16 t1[16];
@@ -326,113 +334,107 @@ __attribute__ ((target("sse2"))) MVSTATIC int32 get_checksum1_sse2_32(schar* buf
     s1 += (uint32)(t1[0] + t1[1] + t1[2] + t1[3] + t1[4] + t1[5] + t1[6] + t1[7] + t1[8] + t1[9] + t1[10] + t1[11] + t1[12] + t1[13] + t1[14] + t1[15]) +
           64*CHAR_OFFSET;
  */
+
 __attribute__ ((target("avx2"))) MVSTATIC int32 get_checksum1_avx2_64(schar* buf, int32 len, int32 i, uint32* ps1, uint32* ps2)
 {
     if (len > 64) {
-        // Instructions reshuffled compared to SSE2 for slightly better performance
-        int aligned = ((uintptr_t)buf & 31) == 0;
 
-        uint32 x[8] = {0};
-        x[0] = *ps1;
-        __m256i ss1 = _mm256_lddqu_si256((__m256i_u*)x);
-        x[0] = *ps2;
-        __m256i ss2 = _mm256_lddqu_si256((__m256i_u*)x);
+        uint32 x[4] = {0};
+        __m128i ss1 = _mm_cvtsi32_si128(*ps1);
+        __m128i ss2 = _mm_cvtsi32_si128(*ps2);
 
-        // The order gets shuffled compared to SSE2
-        const int16 mul_t1_buf[16] = {60, 56, 52, 48, 28, 24, 20, 16, 44, 40, 36, 32, 12, 8, 4, 0};
-        __m256i mul_t1 = _mm256_lddqu_si256((__m256i_u*)mul_t1_buf);
+        const char mul_t1_buf[16] = {60, 56, 52, 48, 44, 40, 36, 32, 28, 24, 20, 16, 12, 8, 4, 0};
+	__m128i tmp = _mm_load_si128((__m128i*) mul_t1_buf);
+        __m256i mul_t1 = _mm256_cvtepu8_epi16(tmp);
+	__m256i mul_const = _mm256_broadcastd_epi32(_mm_cvtsi32_si128(4 | (3 << 8) | (2 << 16) | (1 << 24)));
+        __m256i mul_one;
+       	    mul_one = _mm256_abs_epi8(_mm256_cmpeq_epi16(mul_one,mul_one)); // set all vector elements to 1
 
         for (; i < (len-64); i+=64) {
-            // Load ... 2*[int8*32]
+            // Load ... 4*[int8*16]
             __m256i in8_1, in8_2;
-            if (!aligned) {
-                in8_1 = _mm256_lddqu_si256((__m256i_u*)&buf[i]);
-                in8_2 = _mm256_lddqu_si256((__m256i_u*)&buf[i + 32]);
-            } else {
-                in8_1 = _mm256_load_si256((__m256i_u*)&buf[i]);
-                in8_2 = _mm256_load_si256((__m256i_u*)&buf[i + 32]);
-            }
+	    __m128i in8_1_low, in8_2_low, in8_1_high, in8_2_high;
+	    in8_1_low = _mm_loadu_si128((__m128i_u*)&buf[i]);
+	    in8_2_low = _mm_loadu_si128((__m128i_u*)&buf[i+16]);
+	    in8_1_high = _mm_loadu_si128((__m128i_u*)&buf[i+32]);
+	    in8_2_high = _mm_loadu_si128((__m128i_u*)&buf[i+48]);
+	    in8_1 = _mm256_inserti128_si256(_mm256_castsi128_si256(in8_1_low), in8_1_high,1);
+	    in8_2 = _mm256_inserti128_si256(_mm256_castsi128_si256(in8_2_low), in8_2_high,1);
 
-            // Prefetch for next loops. This has no observable effect on the
-            // tested AMD but makes as much as 20% difference on the Intel.
-            // Curiously that same Intel sees no benefit from this with SSE2
-            // or SSSE3.
-            _mm_prefetch(&buf[i + 64], _MM_HINT_T0);
-            _mm_prefetch(&buf[i + 96], _MM_HINT_T0);
-            _mm_prefetch(&buf[i + 128], _MM_HINT_T0);
-            _mm_prefetch(&buf[i + 160], _MM_HINT_T0);
-
-            // (1*buf[i] + 1*buf[i+1]), (1*buf[i+2], 1*buf[i+3]), ... 2*[int16*16]
+            // (1*buf[i] + 1*buf[i+1]), (1*buf[i+2], 1*buf[i+3]), ... 2*[int16*8]
             // Fastest, even though multiply by 1
-            __m256i mul_one = _mm256_set1_epi8(1);
             __m256i add16_1 = _mm256_maddubs_epi16(mul_one, in8_1);
             __m256i add16_2 = _mm256_maddubs_epi16(mul_one, in8_2);
 
-            // (4*buf[i] + 3*buf[i+1]), (2*buf[i+2], buf[i+3]), ... 2*[int16*16]
-            __m256i mul_const = _mm256_set1_epi32(4 + (3 << 8) + (2 << 16) + (1 << 24));
+            // (4*buf[i] + 3*buf[i+1]), (2*buf[i+2], buf[i+3]), ... 2*[int16*8]
             __m256i mul_add16_1 = _mm256_maddubs_epi16(mul_const, in8_1);
             __m256i mul_add16_2 = _mm256_maddubs_epi16(mul_const, in8_2);
 
             // s2 += 64*s1
-            ss2 = _mm256_add_epi32(ss2, _mm256_slli_epi32(ss1, 6));
+            ss2 = _mm_add_epi32(ss2, _mm_slli_epi32(ss1, 6));
 
-            // [t1[0] + t1[1], t1[2] + t1[3] ...] [int16*16]
+            // [sum(t1[0]..t1[7]), X, X, X] [int32*4]; faster than multiple _mm_hadds_epi16
+            __m256i sum_add32 = _mm256_add_epi16(add16_1, add16_2);
+            sum_add32 = _mm256_add_epi16(sum_add32, _mm256_srli_epi32(sum_add32, 16));
+            sum_add32 = _mm256_add_epi16(sum_add32, _mm256_srli_si256(sum_add32, 4));
+            sum_add32 = _mm256_add_epi16(sum_add32, _mm256_srli_si256(sum_add32, 8));
+
+            // [sum(t2[0]..t2[7]), X, X, X] [int32*4]; faster than multiple _mm_hadds_epi16
+            __m256i sum_mul_add32 = _mm256_add_epi16(mul_add16_1, mul_add16_2);
+            sum_mul_add32 = _mm256_add_epi16(sum_mul_add32, _mm256_srli_epi32(sum_mul_add32, 16));
+            sum_mul_add32 = _mm256_add_epi16(sum_mul_add32, _mm256_srli_si256(sum_mul_add32, 4));
+            sum_mul_add32 = _mm256_add_epi16(sum_mul_add32, _mm256_srli_si256(sum_mul_add32, 8));
+
+            // s1 += t1[0] + t1[1] + t1[2] + t1[3] + t1[4] + t1[5] + t1[6] + t1[7]
+	    __m128i sum_add32_hi = _mm256_extracti128_si256(sum_add32, 0x1);
+            ss1 = _mm_add_epi32(ss1, _mm256_castsi256_si128(sum_add32));
+            ss1 = _mm_add_epi32(ss1, sum_add32_hi);
+
+            // s2 += t2[0] + t2[1] + t2[2] + t2[3] + t2[4] + t2[5] + t2[6] + t2[7]
+	    __m128i sum_mul_add32_hi = _mm256_extracti128_si256(sum_mul_add32, 0x1);
+            ss2 = _mm_add_epi32(ss2, _mm256_castsi256_si128(sum_mul_add32));
+            ss2 = _mm_add_epi32(ss2, sum_mul_add32_hi);
+
+            // [t1[0] + t1[1], t1[2] + t1[3] ...] [int16*8]
+            // We could've combined this with generating sum_add32 above and
+            // save an instruction but benchmarking shows that as being slower
             __m256i add16 = _mm256_hadds_epi16(add16_1, add16_2);
 
-            // [t1[0], t1[1], ...] -> [t1[0]*60 + t1[1]*56, ...] [int32*8]
+            // [t1[0], t1[1], ...] -> [t1[0]*28 + t1[1]*24, ...] [int32*4]
             __m256i mul32 = _mm256_madd_epi16(add16, mul_t1);
 
-            // [sum(t1[0]..t1[15]), X, X, X, X, X, X, X] [int32*8]
-            __m256i sum_add32 = _mm256_add_epi16(add16_1, add16_2);
-            sum_add32 = _mm256_add_epi16(sum_add32, _mm256_permute4x64_epi64(sum_add32, 2 + (3 << 2) + (0 << 4) + (1 << 6)));
-            sum_add32 = _mm256_add_epi16(sum_add32, _mm256_slli_si256(sum_add32, 2));
-            sum_add32 = _mm256_add_epi16(sum_add32, _mm256_slli_si256(sum_add32, 4));
-            sum_add32 = _mm256_add_epi16(sum_add32, _mm256_slli_si256(sum_add32, 8));
-            sum_add32 = _mm256_srai_epi32(sum_add32, 16);
-            sum_add32 = _mm256_shuffle_epi32(sum_add32, 3);
-
-            // s1 += t1[0] + t1[1] + t1[2] + t1[3] + t1[4] + t1[5] + t1[6] + t1[7] + t1[8] + t1[9] + t1[10] + t1[11] + t1[12] + t1[13] + t1[14] + t1[15]
-            ss1 = _mm256_add_epi32(ss1, sum_add32);
-
-            // [sum(t2[0]..t2[15]), X, X, X, X, X, X, X] [int32*8]
-            __m256i sum_mul_add32 = _mm256_add_epi16(mul_add16_1, mul_add16_2);
-            sum_mul_add32 = _mm256_add_epi16(sum_mul_add32, _mm256_permute4x64_epi64(sum_mul_add32, 2 + (3 << 2) + (0 << 4) + (1 << 6)));
-            sum_mul_add32 = _mm256_add_epi16(sum_mul_add32, _mm256_slli_si256(sum_mul_add32, 2));
-            sum_mul_add32 = _mm256_add_epi16(sum_mul_add32, _mm256_slli_si256(sum_mul_add32, 4));
-            sum_mul_add32 = _mm256_add_epi16(sum_mul_add32, _mm256_slli_si256(sum_mul_add32, 8));
-            sum_mul_add32 = _mm256_srai_epi32(sum_mul_add32, 16);
-            sum_mul_add32 = _mm256_shuffle_epi32(sum_mul_add32, 3);
-
-            // s2 += t2[0] + t2[1] + t2[2] + t2[3] + t2[4] + t2[5] + t2[6] + t2[7] + t2[8] + t2[9] + t2[10] + t2[11] + t2[12] + t2[13] + t2[14] + t2[15]
-            ss2 = _mm256_add_epi32(ss2, sum_mul_add32);
-
-            // [sum(mul32), X, X, X, X, X, X, X] [int32*8]
-            mul32 = _mm256_add_epi32(mul32, _mm256_permute2x128_si256(mul32, mul32, 1));
+            // [sum(mul32), X, X, X] [int32*4]; faster than multiple _mm_hadd_epi32
             mul32 = _mm256_add_epi32(mul32, _mm256_srli_si256(mul32, 4));
             mul32 = _mm256_add_epi32(mul32, _mm256_srli_si256(mul32, 8));
+	    // prefetch 2 cacheline ahead
+            _mm_prefetch(&buf[i + 160], _MM_HINT_T0);
 
-            // s2 += 60*t1[0] + 56*t1[1] + 52*t1[2] + 48*t1[3] + 44*t1[4] + 40*t1[5] + 36*t1[6] + 32*t1[7] + 28*t1[8] + 24*t1[9] + 20*t1[10] + 16*t1[11] + 12*t1[12] + 8*t1[13] + 4*t1[14]
-            ss2 = _mm256_add_epi32(ss2, mul32);
+            // s2 += 28*t1[0] + 24*t1[1] + 20*t1[2] + 16*t1[3] + 12*t1[4] + 8*t1[5] + 4*t1[6]
+	    __m128i mul32_hi = _mm256_extracti128_si256(mul32, 0x1);
+            ss2 = _mm_add_epi32(ss2, _mm256_castsi256_si128(mul32));
+            ss2 = _mm_add_epi32(ss2, mul32_hi);
 
 #if CHAR_OFFSET != 0
-            // s1 += 64*CHAR_OFFSET
-            __m256i char_offset_multiplier = _mm256_set1_epi32(64 * CHAR_OFFSET);
-            ss1 = _mm256_add_epi32(ss1, char_offset_multiplier);
+            // s1 += 32*CHAR_OFFSET
+            __m128i char_offset_multiplier = _mm_set1_epi32(32 * CHAR_OFFSET);
+            ss1 = _mm_add_epi32(ss1, char_offset_multiplier);
 
-            // s2 += 2080*CHAR_OFFSET
-            char_offset_multiplier = _mm256_set1_epi32(2080 * CHAR_OFFSET);
-            ss2 = _mm256_add_epi32(ss2, char_offset_multiplier);
+            // s2 += 528*CHAR_OFFSET
+            char_offset_multiplier = _mm_set1_epi32(528 * CHAR_OFFSET);
+            ss2 = _mm_add_epi32(ss2, char_offset_multiplier);
 #endif
         }
 
-        _mm256_store_si256((__m256i_u*)x, ss1);
+        _mm_store_si128((__m128i_u*)x, ss1);
         *ps1 = x[0];
-        _mm256_store_si256((__m256i_u*)x, ss2);
+        _mm_store_si128((__m128i_u*)x, ss2);
         *ps2 = x[0];
     }
     return i;
 }
 
+#endif /* } !USE_ROLL_ASM */
+
 static int32 get_checksum1_default_1(schar* buf, int32 len, int32 i, uint32* ps1, uint32* ps2)
 {
     uint32 s1 = *ps1;
@@ -459,7 +461,11 @@ static inline uint32 get_checksum1_cpp(char *buf1, int32 len)
     uint32 s2 = 0;
 
     // multiples of 64 bytes using AVX2 (if available)
+#ifdef USE_ROLL_ASM
+    i = get_checksum1_avx2_asm((schar*)buf1, len, i, &s1, &s2);
+#else
     i = get_checksum1_avx2_64((schar*)buf1, len, i, &s1, &s2);
+#endif
 
     // multiples of 32 bytes using SSSE3 (if available)
     i = get_checksum1_ssse3_32((schar*)buf1, len, i, &s1, &s2);
@@ -521,14 +527,18 @@ static int32 get_checksum1_auto(schar* buf, int32 len, int32 i, uint32* ps1, uin
 
 int main() {
     int i;
-    unsigned char* buf = (unsigned char*)malloc(BLOCK_LEN);
+    unsigned char* buf = (unsigned char*)aligned_alloc(64,BLOCK_LEN);
     for (i = 0; i < BLOCK_LEN; i++) buf[i] = (i + (i % 3) + (i % 11)) % 256;
 
     benchmark("Auto", get_checksum1_auto, (schar*)buf, BLOCK_LEN);
     benchmark("Raw-C", get_checksum1_default_1, (schar*)buf, BLOCK_LEN);
     benchmark("SSE2", get_checksum1_sse2_32, (schar*)buf, BLOCK_LEN);
     benchmark("SSSE3", get_checksum1_ssse3_32, (schar*)buf, BLOCK_LEN);
+#ifdef USE_ROLL_ASM
+    benchmark("AVX2-ASM", get_checksum1_avx2_asm, (schar*)buf, BLOCK_LEN);
+#else
     benchmark("AVX2", get_checksum1_avx2_64, (schar*)buf, BLOCK_LEN);
+#endif
 
     free(buf);
     return 0;
@@ -538,6 +548,6 @@ int main() {
 #pragma clang optimize on
 #endif /* BENCHMARK_SIMD_CHECKSUM1 */
 
-#endif /* HAVE_SIMD */
-#endif /* __cplusplus */
-#endif /* __x86_64__ */
+#endif /* } USE_ROLL_SIMD */
+#endif /* } __cplusplus */
+#endif /* } __x86_64__ */

support/atomic-rsync

@@ -1,92 +1,94 @@
-#!/usr/bin/env perl
-#
+#!/usr/bin/env python3
 # This script lets you update a hierarchy of files in an atomic way by
 # first creating a new hierarchy using rsync's --link-dest option, and
 # then swapping the hierarchy into place.  **See the usage message for
 # more details and some important caveats!**
 
-use strict;
-use warnings;
-use Cwd 'abs_path';
+import os, sys, re, subprocess, shutil
 
-my $RSYNC_PROG = '/usr/bin/rsync';
-my $RM_PROG = '/bin/rm';
+ALT_DEST_ARG_RE = re.compile('^--[a-z][^ =]+-dest(=|$)')
 
-my $dest_dir = $ARGV[-1];
-&usage if !defined $dest_dir || $dest_dir =~ /(^-|^$)/ || grep(/^--help/, @ARGV);
-$dest_dir =~ s{(?<=.)/+$} {};
+RSYNC_PROG = '/usr/bin/rsync'
 
-if (!-d $dest_dir) {
-    die "$dest_dir is not a directory.\nUse --help for help.\n";
-}
+def main():
+    cmd_args = sys.argv[1:]
+    if '--help' in cmd_args:
+        usage_and_exit()
 
-if (@_ = grep(/^--[a-z]+-dest\b/, @ARGV)) {
-    $_ = join(' or ', @_);
-    die "You cannot use the $_ option with atomic-rsync.\nUse --help for help.\n";
-}
+    if len(cmd_args) < 2:
+        usage_and_exit(True)
 
-my $symlink_content = readlink $dest_dir; # undef when a real dir
+    dest_dir = cmd_args[-1].rstrip('/')
+    if dest_dir == '' or dest_dir.startswith('-'):
+        usage_and_exit(True)
 
-my $dest_arg = $dest_dir;
-# This gives us the real destination dir, with all symlinks dereferenced.
-$dest_dir = abs_path($dest_dir);
-if ($dest_dir eq '/') {
-    die qq|You must not use "/" as the destination directory.\nUse --help for help.\n|;
-}
+    if not os.path.isdir(dest_dir):
+        die(dest_dir, "is not a directory or a symlink to a dir.\nUse --help for help.")
 
-my($old_dir, $new_dir);
-if (defined $symlink_content && $dest_dir =~ /-([12])$/) {
-    my $num = 3 - $1;
-    $old_dir = undef;
-    ($new_dir = $dest_dir) =~ s/-[12]$/-$num/;
-    $symlink_content =~ s/-[12]$/-$num/;
-} else {
-    $old_dir = "$dest_dir~old~";
-    $new_dir = "$dest_dir~new~";
-}
+    bad_args = [ arg for arg in cmd_args if ALT_DEST_ARG_RE.match(arg) ]
+    if bad_args:
+        die("You cannot use the", ' or '.join(bad_args), "option with atomic-rsync.\nUse --help for help.")
 
-$ARGV[-1] = "$new_dir/";
+    # We ignore exit-code 24 (file vanished) by default.
+    allowed_exit_codes = '0 ' + os.environ.get('ATOMIC_RSYNC_OK_CODES', '24')
+    try:
+        allowed_exit_codes = set(int(num) for num in re.split(r'[, ]+', allowed_exit_codes) if num != '')
+    except ValueError:
+        die('Invalid integer in ATOMIC_RSYNC_OK_CODES:', allowed_exit_codes[2:])
 
-system($RM_PROG, '-rf', $old_dir) if defined $old_dir && -d $old_dir;
-system($RM_PROG, '-rf', $new_dir) if -d $new_dir;
+    symlink_content = os.readlink(dest_dir) if os.path.islink(dest_dir) else None
 
-if (system($RSYNC_PROG, "--link-dest=$dest_dir", @ARGV)) {
-    if ($? == -1) {
-	print "failed to execute $RSYNC_PROG: $!\n";
-    } elsif ($? & 127) {
-	printf "child died with signal %d, %s coredump\n",
-	    ($? & 127),  ($? & 128) ? 'with' : 'without';
-    } else {
-	printf "child exited with value %d\n", $? >> 8;
-    }
-    exit 1;
-}
+    dest_arg = dest_dir
+    dest_dir = os.path.realpath(dest_dir) # The real destination dir with all symlinks dereferenced
+    if dest_dir == '/':
+        die('You must not use "/" as the destination directory.\nUse --help for help.')
 
-if (!defined $old_dir) {
-    atomic_symlink($symlink_content, $dest_arg);
-    exit;
-}
+    old_dir = new_dir = None
+    if symlink_content is not None and dest_dir.endswith(('-1','-2')):
+        if not symlink_content.endswith(dest_dir[-2:]):
+            die("Symlink suffix out of sync with dest_dir name:", symlink_content, 'vs', dest_dir)
+        num = 3 - int(dest_dir[-1]);
+        old_dir = None
+        new_dir = dest_dir[:-1] + str(num)
+        symlink_content = symlink_content[:-1] + str(num)
+    else:
+        old_dir = dest_dir + '~old~'
+        new_dir = dest_dir + '~new~'
 
-rename($dest_dir, $old_dir) or die "Unable to rename $dest_dir to $old_dir: $!";
-rename($new_dir, $dest_dir) or die "Unable to rename $new_dir to $dest_dir: $!";
+    cmd_args[-1] = new_dir + '/'
 
-exit;
+    if old_dir is not None and os.path.isdir(old_dir):
+        shutil.rmtree(old_dir)
+    if os.path.isdir(new_dir):
+        shutil.rmtree(new_dir)
 
-sub atomic_symlink
-{
-    my($target, $link) = @_;
-    my $newlink = "$link~new~";
+    child = subprocess.run([RSYNC_PROG, '--link-dest=' + dest_dir, *cmd_args])
+    if child.returncode not in allowed_exit_codes:
+        die('The rsync copy failed with code', child.returncode, exitcode=child.returncode)
 
-    unlink($newlink); # Just in case
-    symlink($target, $newlink) or die "Unable to symlink $newlink -> $target: $!\n";
-    rename($newlink, $link) or die "Unable to rename $newlink to $link: $!\n";
-}
+    if not os.path.isdir(new_dir):
+        die('The rsync copy failed to create:', new_dir)
+
+    if old_dir is None:
+        atomic_symlink(symlink_content, dest_arg)
+    else:
+        os.rename(dest_dir, old_dir)
+        os.rename(new_dir, dest_dir)
 
 
-sub usage
-{
-    die <<EOT;
-Usage: atomic-rsync [RSYNC-OPTIONS] HOST:/SOURCE/DIR/ /DEST/DIR/
+def atomic_symlink(target, link):
+    newlink = link + "~new~"
+    try:
+        os.unlink(newlink); # Just in case
+    except OSError:
+        pass
+    os.symlink(target, newlink)
+    os.rename(newlink, link)
+
+
+def usage_and_exit(use_stderr=False):
+    usage_msg = """\
+Usage: atomic-rsync [RSYNC-OPTIONS] [HOST:]/SOURCE/DIR/ /DEST/DIR/
        atomic-rsync [RSYNC-OPTIONS] HOST::MOD/DIR/ /DEST/DIR/
 
 This script lets you update a hierarchy of files in an atomic way by first
@@ -96,27 +98,41 @@ to a local directory, and that directory must already exist.  For example:
 
     mkdir /local/files-1
     ln -s files-1 /local/files
-    atomic-rsync -av host:/remote/files/ /local/files/
+    atomic-rsync -aiv host:/remote/files/ /local/files/
 
-If /local/files is a symlink to a directory that ends in -1 or -2, the
-copy will go to the alternate suffix and the symlink will be changed to
-point to the new dir.  This is a fully atomic update.  If the destination
-is not a symlink (or not a symlink to a *-1 or a *-2 directory), this
-will instead create a directory with "~new~" suffixed, move the current
-directory to a name with "~old~" suffixed, and then move the ~new~
-directory to the original destination name (this double rename is not
-fully atomic, but is rapid).  In both cases, the prior destintaion
-directory will be preserved until the next update, at which point it
-will be deleted.
+If /local/files is a symlink to a directory that ends in -1 or -2, the copy
+will go to the alternate suffix and the symlink will be changed to point to
+the new dir.  This is a fully atomic update.  If the destination is not a
+symlink (or not a symlink to a *-1 or a *-2 directory), this will instead
+create a directory with "~new~" suffixed, move the current directory to a
+name with "~old~" suffixed, and then move the ~new~ directory to the original
+destination name (this double rename is not fully atomic, but is rapid).  In
+both cases, the prior destintaion directory will be preserved until the next
+update, at which point it will be deleted.
 
-In all likelihood, you do NOT want to specify this command:
+By default, rsync exit-code 24 (file vanished) is allowed without halting the
+atomic update.  If you want to change that, specify the environment variable
+ATOMIC_RSYNC_OK_CODES with numeric values separated by spaces and/or commas.
+Specify an empty string to only allow a successful copy.  An override example:
 
-    atomic-rsync -av host:/remote/files /local/
+    ATOMIC_RSYNC_OK_CODES='23 24' atomic-rsync -aiv host:src/ dest/
 
-... UNLESS you want the entire /local dir to be swapped out!
+See the errcode.h file for a list of all the exit codes.
 
 See the "rsync" command for its list of options.  You may not use the
 --link-dest, --compare-dest, or --copy-dest options (since this script
 uses --link-dest to make the transfer efficient).
-EOT
-}
+"""
+    print(usage_msg, file=sys.stderr if use_stderr else sys.stdout)
+    sys.exit(1 if use_stderr else 0)
+
+
+def die(*args, exitcode=1):
+    print(*args, file=sys.stderr)
+    sys.exit(exitcode)
+
+
+if __name__ == '__main__':
+    main()
+
+# vim: sw=4 et

support/cvs2includes

@@ -1,42 +1,63 @@
-#!/usr/bin/env perl
-#
+#!/usr/bin/env python3
 # This script finds all CVS/Entries files in the current directory and below
 # and creates a local .cvsinclude file with non-inherited rules including each
 # checked-in file.  Then, use this option whenever using --cvs-exclude (-C):
 #
-# 	-f ': .cvsinclude'
+#    -f ': .cvsinclude'
 #
 # That ensures that all checked-in files/dirs are included in the transfer.
 # (You could alternately put ": .cvsinclude" into an .rsync-filter file and
 # use the -F option, which is easier to type.)
 #
 # The downside is that you need to remember to re-run cvs2includes whenever
-# you add a new file to the project.
-use strict;
+# CVS gets an added or removed file. Maybe just run it before every copy.
 
-open(FIND, 'find . -name CVS -type d |') or die $!;
-while (<FIND>) {
-    chomp;
-    s#^\./##;
+import os, argparse
 
-    my $entries = "$_/Entries";
-    s/CVS$/.cvsinclude/;
-    my $filter = $_;
+INC_NAME = '.cvsinclude'
 
-    open(ENTRIES, $entries) or die "Unable to open $entries: $!\n";
-    my @includes;
-    while (<ENTRIES>) {
-	push(@includes, $1) if m#/(.+?)/#;
-    }
-    close ENTRIES;
-    if (@includes) {
-	open(FILTER, ">$filter") or die "Unable to write $filter: $!\n";
-	print FILTER map "+ /$_\n", @includes;
-	close FILTER;
-	print "Updated $filter\n";
-    } elsif (-f $filter) {
-	unlink($filter);
-	print "Removed $filter\n";
-    }
-}
-close FIND;
+def main():
+    if args.dir:
+        os.chdir(args.dir)
+
+    cvs_includes = set()
+    for root, dirs, files in os.walk('.'):
+        if INC_NAME in files:
+            cvs_includes.add((root + '/' + INC_NAME)[2:])
+        if root.endswith('/CVS') and 'Entries' in files:
+            entries = root[2:] + '/Entries'
+            includes = [ ]
+            with open(entries) as fh:
+                for line in fh:
+                    if line.startswith(('/', 'D/')):
+                        includes.append(line.split('/', 2)[1])
+            if includes:
+                inc = root[2:-3] + INC_NAME
+                cvs_includes.discard(inc)
+                try:
+                    with open(inc) as fh:
+                        old_txt = fh.read()
+                except OSError:
+                    old_txt = ''
+                txt = ''.join(f"+ /{x}\n" for x in includes)
+                if txt == old_txt:
+                    print("Unchanged", inc)
+                else:
+                    print("Updating", inc)
+                    with open(inc, 'w') as fh:
+                        fh.write(txt)
+        dirs.sort()
+
+    for inc in sorted(cvs_includes):
+        print("Removing", inc)
+        os.unlink(inc)
+
+
+if __name__ == '__main__':
+    parser = argparse.ArgumentParser(description=f"Transform CVS/Entries into {INC_NAME} files.", add_help=False)
+    parser.add_argument("--help", "-h", action="help", help="Output this help message and exit.")
+    parser.add_argument("dir", nargs='?', help="The top CVS dir. Defaults to the current directory.")
+    args = parser.parse_args()
+    main()
+
+# vim: sw=4 et

support/files-to-excludes

@@ -1,27 +1,37 @@
-#!/usr/bin/env perl
-# This script takes an input of filenames and outputs a set of
-# include/exclude directives that can be used by rsync to copy
-# just the indicated files using an --exclude-from=FILE option.
-use strict;
+#!/usr/bin/env python3
+# This script takes an input of filenames and outputs a set of include/exclude
+# directives that can be used by rsync to copy just the indicated files using
+# an --exclude-from=FILE or -f'. FILE' option. To be able to delete files on
+# the receiving side, either use --delete-excluded or change the exclude (-)
+# rules to hide filter rules (H) that only affect the sending side.
 
-my %hash;
+import os, fileinput, argparse
 
-while (<>) {
-    chomp;
-    s#^/+##;
-    my $path = '/';
-    while (m#([^/]+/)/*#g) {
-	$path .= $1;
-	print "+ $path\n" unless $hash{$path}++;
-    }
-    if (m#([^/]+)$#) {
-	print "+ $path$1\n";
-    } else {
-	delete $hash{$path};
-    }
-}
+def main():
+    paths = set()
+    for line in fileinput.input(args.files):
+        dirs = line.strip().lstrip('/').split('/')
+        if not dirs:
+            continue
+        for j in range(1, len(dirs)):
+            if dirs[j] == '':
+                continue
+            path = '/' + '/'.join(dirs[:j]) + '/'
+            if path not in paths:
+                print('+', path)
+                paths.add(path)
+        print('+', '/' + '/'.join(dirs))
 
-foreach (sort keys %hash) {
-    print "- $_*\n";
-}
-print "- /*\n";
+    for path in sorted(paths):
+        print('-', path + '*')
+    print('-', '/*')
+
+
+if __name__ == '__main__':
+    parser = argparse.ArgumentParser(description="Transform a list of files into a set of include/exclude rules.", add_help=False)
+    parser.add_argument("--help", "-h", action="help", help="Output this help message and exit.")
+    parser.add_argument("files", metavar="FILE", default='-', nargs='*', help="The file(s) that hold the pathnames to translate. Defaults to stdin.")
+    args = parser.parse_args()
+    main()
+
+# vim: sw=4 et

support/git-set-file-times

@@ -27,17 +27,18 @@ def main():
 
     if not args.tree:
         # All modified files keep their current mtime.
-        proc = subprocess.Popen(git + 'ls-files -m -z'.split(), stdout=subprocess.PIPE, encoding='utf-8')
+        proc = subprocess.Popen(git + 'status -z --no-renames'.split(), stdout=subprocess.PIPE, encoding='utf-8')
         out = proc.communicate()[0]
         for fn in out.split('\0'):
-            if fn == '':
+            if fn == '' or (fn[0] != 'M' and fn[1] != 'M'):
                 continue
+            fn = fn[3:]
             if args.list:
                 mtime = os.lstat(fn).st_mtime
                 print_line(fn, mtime, mtime)
             ls.discard(fn)
 
-    cmd = git + 'log -r --name-only --no-color --pretty=raw --no-renames -z'.split()
+    cmd = git + 'log -r --name-only --format=%x00commit%x20%H%n%x00commit_time%x20%ct%n --no-renames -z'.split()
     if args.tree:
         cmd.append(args.tree)
     cmd += ['--'] + args.files
@@ -45,7 +46,7 @@ def main():
     proc = subprocess.Popen(cmd, stdout=subprocess.PIPE, encoding='utf-8')
     for line in proc.stdout:
         line = line.strip()
-        m = re.match(r'^committer .*? (\d+) [-+]\d+$', line)
+        m = re.match(r'^\0commit_time (\d+)$', line)
         if m:
             commit_time = int(m[1])
         elif NULL_COMMIT_RE.search(line):

support/idmap

@@ -0,0 +1,45 @@
+#!/usr/bin/env python3
+# This helper script makes it easy to use a passwd or group file to map values
+# in a LOCAL transfer.  For instance, if you mount a backup that does not have
+# the same passwd setup as the local machine, you can do a copy to/from the
+# backup area as follows and get the differing ID values mapped just like a
+# remote transfer to/from the backed-up machine would do:
+#
+# rsync -av --usermap=`idmap --to /mnt/backup/etc/passwd` \
+#           --groupmap=`idmap --to /mnt/backup/etc/group` \
+#           /some/src/ /mnt/backup/some/dest/
+#
+# rsync -av --usermap=`idmap --from /mnt/backup/etc/passwd` \
+#           --groupmap=`idmap --from /mnt/backup/etc/group` \
+#           /mnt/backup/some/src/ /some/dest/
+
+import re, fileinput, argparse
+
+NAME_ID_RE = re.compile(r'^(\w+):[^:]+:(\d+)')
+
+def main():
+    maps = [ ]
+    for line in fileinput.input(args.files):
+        m = NAME_ID_RE.match(line)
+        if not m:
+            continue
+        if args.to:
+            pair = (m[1], m[2])
+        else:
+            pair = (m[2], m[1])
+        maps.append(':'.join(pair))
+    print(','.join(maps))
+
+
+if __name__ == '__main__':
+    parser = argparse.ArgumentParser(description="Output usermap or groupmap args for rsync.", add_help=False)
+    action = parser.add_argument_group()
+    action = parser.add_mutually_exclusive_group(required=True)
+    action.add_argument("--from", action="store_true", help="Output the map for use on the sending side.")
+    action.add_argument("--to", action="store_true", help="Output the map for use on the receiving side.")
+    parser.add_argument("--help", "-h", action="help", help="Output this help message and exit.")
+    parser.add_argument("files", metavar="FILE", default='-', nargs='*', help="The file(s) that hold the name & id pairs. Defaults to stdin.")
+    args = parser.parse_args()
+    main()
+
+# vim: sw=4 et

support/json-rsync-version

@@ -0,0 +1,93 @@
+#!/usr/bin/env python3
+
+import sys, argparse, subprocess, json
+
+TWEAK_NAME = {
+        'asm': 'asm_roll',
+        'ASM': 'asm_roll',
+        'hardlink_special': 'hardlink_specials',
+        'protect_args': 'secluded_args',
+        'protected_args': 'secluded_args',
+        'SIMD': 'SIMD_roll',
+    }
+
+MOVE_OPTIM = set('asm_roll SIMD_roll'.split())
+
+def main():
+    if not args.rsync or args.rsync == '-':
+        ver_out = sys.stdin.read().strip()
+    else:
+        ver_out = subprocess.check_output([args.rsync, '--version', '--version'], encoding='utf-8').strip()
+    if ver_out.startswith('{'):
+        print(ver_out)
+        return
+    info = { }
+    misplaced_optims = { }
+    for line in ver_out.splitlines():
+        if line.startswith('rsync '):
+            prog, vstr, ver, pstr, vstr2, proto = line.split()
+            info['program'] = prog
+            if ver.startswith('v'):
+                ver = ver[1:]
+            info[vstr] = ver
+            if '.' not in proto:
+                proto += '.0'
+            else:
+                proto = proto.replace('.PR', '.')
+            info[pstr] = proto
+        elif line.startswith('Copyright '):
+            info['copyright'] = line[10:]
+        elif line.startswith('Web site: '):
+            info['url'] = line[10:]
+        elif line.startswith('  '):
+            if not saw_comma and ',' in line:
+                saw_comma = True
+                info[sect_name] = { }
+            if saw_comma:
+                for x in line.strip(' ,').split(', '):
+                    if ' ' in x:
+                        val, var = x.split(' ', 1)
+                        if val == 'no':
+                            val = False
+                        elif val.endswith('-bit'):
+                            var = var[:-1] + '_bits'
+                            val = int(val.split('-')[0])
+                    else:
+                        var = x
+                        val = True
+                    var = var.replace(' ', '_').replace('-', '_')
+                    if var in TWEAK_NAME:
+                        var = TWEAK_NAME[var]
+                    if sect_name[0] != 'o' and var in MOVE_OPTIM:
+                        misplaced_optims[var] = val
+                    else:
+                        info[sect_name][var] = val
+            else:
+                info[sect_name] += [ x for x in line.split() if not x.startswith('(') ]
+        elif line == '':
+            break
+        else:
+            sect_name = line.strip(' :').replace(' ', '_').lower()
+            info[sect_name] = [ ]
+            saw_comma = False
+    for chk in 'capabilities optimizations'.split():
+        if chk not in info:
+            info[chk] = { }
+    if misplaced_optims:
+        info['optimizations'].update(misplaced_optims)
+    for chk in 'checksum_list compress_list daemon_auth_list'.split():
+        if chk not in info:
+            info[chk] = [ ]
+    info['license'] = 'GPLv3' if ver[0] == '3' else 'GPLv2'
+    info['caveat'] = 'rsync comes with ABSOLUTELY NO WARRANTY'
+    print(json.dumps(info))
+
+
+if __name__ == '__main__':
+    parser = argparse.ArgumentParser(description="Output rsync's version data in JSON format, even if the rsync doesn't support a native json-output method.", add_help=False)
+    parser.add_argument('rsync', nargs='?', help="Specify an rsync command to run. Otherwise stdin is consumed.")
+    parser.add_argument("--help", "-h", action="help", help="Output this help message and exit.")
+    args = parser.parse_args()
+    main()
+
+# vim: sw=4 et

support/lsh

@@ -15,6 +15,8 @@ GetOptions(
     'b|c|D|e|F|i|L|m|O|o|p|R|S|w=s' => sub { }, # Ignore
     'no-cd' => \( my $no_chdir ),
     'sudo' => \( my $use_sudo ),
+    'rrsync=s' => \( my $rrsync_dir ),
+    'rropts=s' => \( my $rrsync_opts ),
 ) or &usage;
 &usage unless @ARGV > 1;
 
@@ -67,22 +69,40 @@ unless ($no_chdir) {
     chdir $home_dir or die "Unable to chdir to $home_dir: $!\n";
 }
 
-push @cmd, '/bin/sh', '-c', "@ARGV";
+if ($rrsync_dir) {
+    $ENV{SSH_ORIGINAL_COMMAND} = join(' ', @ARGV);
+    push @cmd, 'rrsync';
+    if ($rrsync_opts) {
+	foreach my $opt (split(/[ ,]+/, $rrsync_opts)) {
+	    $opt = "-$opt" unless $opt =~ /^-/;
+	    push @cmd, $opt;
+	}
+    }
+    push @cmd, $rrsync_dir;
+} else {
+    push @cmd, '/bin/sh', '-c', "@ARGV";
+}
 exec @cmd;
 die "Failed to exec: $!\n";
 
 sub usage
 {
     die <<EOT;
-Usage: lsh [-l USER] [--sudo] [--no-cd] localhost COMMAND [...]
+Usage: lsh [OPTIONS] localhost|lh COMMAND [...]
 
 This is a "local shell" command that works like a remote shell but only for the
 local host.  This is useful for rsync testing or for running a local copy where
 the sender and the receiver need to use different options (e.g. --fake-super).
-If the -l option is used, we try to become the USER, either directly (when
-root) or by using "sudo -H -u USER" (requires --sudo option).
 
-Note that if you pass hostname "lh" instead of "localhost" that the --no-cd
-option is implied.  The default is to "cd \$HOME" to simulate ssh behavior.
+Options:
+
+-l USER        Choose the USER that lsh tries to become.
+--no-cd        Skip the chdir \$HOME (the default with hostname "lh")
+--sudo         Use sudo -H -l USER to become root or the specified USER.
+--rrsync=DIR   Test rrsync restricted copying without using ssh.
+--rropts=STR   The string "munge,no-del,no-lock" would pass 3 options to
+               rrsync (must be combined with --rrsync=DIR).
+
+The script also ignores a bunch of single-letter ssh options.
 EOT
 }

support/lsh.sh

@@ -4,7 +4,8 @@
 # for testing or for running a local copy where the sender and the
 # receiver needs to use different options (e.g. --fake-super).  If
 # we get a -l USER option, we try to use "sudo -u USER" to run the
-# command.
+# command.  Supports only the hostnames "localhost" and "lh", with
+# the latter implying the --no-cd option.
 
 user=''
 do_cd=y # Default path is user's home dir (just like ssh) unless host is "lh".

support/mapfrom

@@ -1,15 +0,0 @@
-#!/usr/bin/env perl
-# This helper script makes it easy to use a passwd or group file to map
-# values in a LOCAL transfer.  For instance, if you mount a backup that
-# does not have the same passwd setup as the local machine, you can do
-# a copy FROM the backup area as follows and get the differing ID values
-# mapped just like a remote transfer FROM the backed-up machine would do:
-#
-# rsync -av --usermap=`mapfrom /mnt/backup/etc/passwd` \
-#           --groupmap=`mapfrom /mnt/backup/etc/group` \
-#           /mnt/backup/some/src/ /some/dest/
-
-while (<>) {
-    push @_, "$2:$1" if /^(\w+):[^:]+:(\d+)/;
-}
-print join(',', @_), "\n";

support/mapto

@@ -1,15 +0,0 @@
-#!/usr/bin/env perl
-# This helper script makes it easy to use a passwd or group file to map
-# values in a LOCAL transfer.  For instance, if you mount a backup that
-# does not have the same passwd setup as the local machine, you can do
-# a copy TO the backup area as follows and get the differing ID values
-# mapped just like a remote transfer TO the backed-up machine would do:
-#
-# rsync -av --usermap=`mapto /mnt/backup/etc/passwd` \
-#           --groupmap=`mapto /mnt/backup/etc/group` \
-#           /some/src/ /mnt/backup/some/dest/
-
-while (<>) {
-    push @_, "$1:$2" if /^(\w+):[^:]+:(\d+)/;
-}
-print join(',', @_), "\n";

support/mnt-excl

@@ -1,4 +1,4 @@
-#!/usr/bin/env perl
+#!/usr/bin/env python3
 # This script takes a command-line arg of a source directory
 # that will be passed to rsync, and generates a set of excludes
 # that will exclude all mount points from the list.  This is
@@ -27,23 +27,33 @@
 # awk '{print $2}' /proc/mounts | grep -v '^/$' | \
 #   rsync -avf 'merge,/- -' /dir host:/dest/
 
-use strict;
-use warnings;
-use Cwd 'abs_path';
+import os, argparse
 
-my $file = '/proc/mounts';
-my $dir = shift || '/';
-my $trailing_slash = $dir =~ m{./$} ? '/' : '';
-$dir = abs_path($dir) . $trailing_slash;
-$dir =~ s{([^/]*)$}{};
-my $trailing = $1;
-$trailing = '' if $trailing eq '.' || !-d "$dir$trailing";
-$trailing .= '/' if $trailing ne '';
+MNT_FILE = '/proc/mounts';
 
-open(IN, $file) or die "Unable to open $file: $!\n";
-while (<IN>) {
-    $_ = (split)[1];
-    next unless s{^\Q$dir$trailing\E}{}o && $_ ne '';
-    print "- /$trailing$_\n";
-}
-close IN;
+def main():
+    trailing_slash = '/' if args.path.endswith(('/', '/.')) and args.path != '/' else ''
+    args.path = os.path.realpath(args.path) + trailing_slash
+    parent_dir = os.path.dirname(args.path)
+    trailing = os.path.basename(args.path)
+    if not os.path.isdir(args.path):
+        trailing = ''
+    elif trailing != '':
+        trailing += '/'
+    want_path = os.path.join(parent_dir, trailing)
+    wp_len = len(want_path)
+
+    with open(MNT_FILE) as fh:
+        for line in fh:
+            mnt_path = line.split()[1]
+            if mnt_path.startswith(want_path) and mnt_path != want_path:
+                print(f"- /{trailing}{mnt_path[wp_len:]}")
+
+if __name__ == '__main__':
+    parser = argparse.ArgumentParser(description="Output mount points as rsync excludes.", add_help=False)
+    parser.add_argument("--help", "-h", action="help", help="Output this help message and exit.")
+    parser.add_argument('path', metavar='PATH', nargs='?', default='/', help="Limit output to those within the PATH hierarchy.")
+    args = parser.parse_args()
+    main()
+
+# vim: sw=4 et

support/munge-symlinks

@@ -1,60 +1,71 @@
-#!/usr/bin/env perl
+#!/usr/bin/env python3
 # This script will either prefix all symlink values with the string
 # "/rsyncd-munged/" or remove that prefix.
 
-use strict;
-use Getopt::Long;
+import os, sys, argparse
 
-my $SYMLINK_PREFIX = '/rsyncd-munged/';
+SYMLINK_PREFIX = '/rsyncd-munged/'
+PREFIX_LEN = len(SYMLINK_PREFIX)
 
-my $munge_opt;
+def main():
+    for arg in args.names:
+        if os.path.islink(arg):
+            process_one_arg(arg)
+        elif os.path.isdir(arg):
+            for fn in find_symlinks(arg):
+                process_one_arg(fn)
+        else:
+            print("Arg is not a symlink or a dir:", arg, file=sys.stderr)
 
-&GetOptions(
-    'munge' => sub { $munge_opt = 1 },
-    'unmunge' => sub { $munge_opt = 0 },
-    'all' => \( my $all_opt ),
-    'help|h' => \( my $help_opt ),
-) or &usage;
 
-&usage if $help_opt || !defined $munge_opt;
+def find_symlinks(path):
+    for entry in os.scandir(path):
+        if entry.is_symlink():
+            yield entry.path
+        elif entry.is_dir(follow_symlinks=False):
+            yield from find_symlinks(entry.path)
 
-my $munged_re = $all_opt ? qr/^($SYMLINK_PREFIX)+(?=.)/ : qr/^$SYMLINK_PREFIX(?=.)/;
 
-push(@ARGV, '.') unless @ARGV;
+def process_one_arg(fn):
+    lnk = os.readlink(fn)
+    if args.unmunge:
+        if not lnk.startswith(SYMLINK_PREFIX):
+            return
+        lnk = lnk[PREFIX_LEN:]
+        while args.all and lnk.startswith(SYMLINK_PREFIX):
+            lnk = lnk[PREFIX_LEN:]
+    else:
+        if not args.all and lnk.startswith(SYMLINK_PREFIX):
+            return
+        lnk = SYMLINK_PREFIX + lnk
 
-open(PIPE, '-|', 'find', @ARGV, '-type', 'l') or die $!;
+    try:
+        os.unlink(fn)
+    except OSError as e:
+        print("Unable to unlink symlink:", str(e), file=sys.stderr)
+        return
+    try:
+        os.symlink(lnk, fn)
+    except OSError as e:
+        print("Unable to recreate symlink", fn, '->', lnk + ':', str(e), file=sys.stderr)
+        return
+    print(fn, '->', lnk)
 
-while (<PIPE>) {
-    chomp;
-    my $lnk = readlink($_) or next;
-    if ($munge_opt) {
-	next if !$all_opt && $lnk =~ /$munged_re/;
-	$lnk =~ s/^/$SYMLINK_PREFIX/;
-    } else {
-	next unless $lnk =~ s/$munged_re//;
-    }
-    if (!unlink($_)) {
-	warn "Unable to unlink symlink: $_ ($!)\n";
-    } elsif (!symlink($lnk, $_)) {
-	warn "Unable to recreate symlink: $_ -> $lnk ($!)\n";
-    } else {
-	print "$_ -> $lnk\n";
-    }
-}
 
-close PIPE;
-exit;
+if __name__ == '__main__':
+    our_desc = """\
+Adds or removes the %s prefix to/from the start of each symlink's value.
+When given the name of a directory, affects all the symlinks in that directory hierarchy.
+""" % SYMLINK_PREFIX
+    epilog = 'See the "munge symlinks" option in the rsyncd.conf manpage for more details.'
+    parser = argparse.ArgumentParser(description=our_desc, epilog=epilog, add_help=False)
+    uniq_group = parser.add_mutually_exclusive_group()
+    uniq_group.add_argument('--munge', action='store_true', help="Add the prefix to symlinks (the default).")
+    uniq_group.add_argument('--unmunge', action='store_true', help="Remove the prefix from symlinks.")
+    parser.add_argument('--all', action='store_true', help="Always adds the prefix when munging (even if already munged) or removes multiple instances of the prefix when unmunging.")
+    parser.add_argument('--help', '-h', action='help', help="Output this help message and exit.")
+    parser.add_argument('names', metavar='NAME', nargs='+', help="One or more directories and/or symlinks to process.")
+    args = parser.parse_args()
+    main()
 
-sub usage
-{
-    die <<EOT;
-Usage: munge-symlinks --munge|--unmunge [--all] [DIR|SYMLINK...]
-
---munge      Add the $SYMLINK_PREFIX prefix to symlinks if not already
-             present, or always when combined with --all.
---unmunge    Remove one $SYMLINK_PREFIX prefix from symlinks or all
-             such prefixes with --all.
-
-See the "munge symlinks" option in the rsyncd.conf manpage for more details.
-EOT
-}
+# vim: sw=4 et

support/nameconvert

@@ -14,8 +14,8 @@
 # An unknown ID_NUM or NAME results in an empty return value.
 #
 # This is used by an rsync daemon when configured with the "name converter" and
-# "use chroot = true".  While this converter uses real user & group lookups you
-# could change it to use any mapping idiom you'd like.
+# (often) "use chroot = true".  While this converter uses real user & group
+# lookups you could change it to use any mapping idiom you'd like.
 
 import sys, argparse, pwd, grp
 

support/rrsync

@@ -1,252 +1,383 @@
-#!/usr/bin/env perl
-# Name: /usr/local/bin/rrsync (should also have a symlink in /usr/bin)
-# Purpose: Restricts rsync to subdirectory declared in .ssh/authorized_keys
-# Author: Joe Smith <js-cgi@inwap.com> 30-Sep-2004
-# Modified by: Wayne Davison <wayne@opencoder.net>
-use strict;
+#!/usr/bin/env python3
 
-use Socket;
-use Cwd 'abs_path';
-use File::Glob ':glob';
+# Restricts rsync to subdirectory declared in .ssh/authorized_keys.  See
+# the rrsync man page for details of how to make use of this script.
 
-# You may configure these values to your liking.  See also the section
-# of options if you want to disable any options that rsync accepts.
-use constant RSYNC => '/usr/bin/rsync';
-use constant LOGFILE => 'rrsync.log';
+# NOTE: install python3 braceexpand to support brace expansion in the args!
 
-my $Usage = <<EOM;
-Use 'command="$0 [-ro|-wo] SUBDIR"'
-	in front of lines in $ENV{HOME}/.ssh/authorized_keys
-EOM
+# Originally a perl script by: Joe Smith <js-cgi@inwap.com> 30-Sep-2004
+# Python version by: Wayne Davison <wayne@opencoder.net>
 
-# Handle the -ro and -wo options.
-our $only = '';
-while (@ARGV && $ARGV[0] =~ /^-([rw])o$/) {
-    my $r_or_w = $1;
-    if ($only && $only ne $r_or_w) {
-	die "$0: the -ro and -wo options conflict.\n";
-    }
-    $only = $r_or_w;
-    shift;
-}
+# You may configure these 2 values to your liking.  See also the section of
+# short & long options if you want to disable any options that rsync accepts.
+RSYNC = '/usr/bin/rsync'
+LOGFILE = 'rrsync.log' # NOTE: the file must exist for a line to be appended!
 
-our $subdir = shift;
-die "$0: No subdirectory specified\n$Usage" unless defined $subdir;
-$subdir = abs_path($subdir);
-die "$0: Restricted directory does not exist!\n" if $subdir ne '/' && !-d $subdir;
+# The following options are mainly the options that a client rsync can send
+# to the server, and usually just in the one option format that the stock
+# rsync produces. However, there are some additional convenience options
+# added as well, and thus a few options are present in both the short and
+# long lists (such as --group, --owner, and --perms).
 
-# The client uses "rsync -av -e ssh src/ server:dir/", and sshd on the server
-# executes this program when .ssh/authorized_keys has 'command="..."'.
-# For example:
-# command="rrsync logs/client" ssh-rsa AAAAB3NzaC1yc2EAAAABIwAAAIEAzGhEeNlPr...
-# command="rrsync -ro results" ssh-rsa AAAAB3NzaC1yc2EAAAABIwAAAIEAmkHG1WCjC...
-#
-# Format of the environment variables set by sshd:
-# SSH_ORIGINAL_COMMAND=rsync --server          -vlogDtpr --partial . ARG # push
-# SSH_ORIGINAL_COMMAND=rsync --server --sender -vlogDtpr --partial . ARGS # pull
-# SSH_CONNECTION=client_addr client_port server_port
+# NOTE when disabling: check for both a short & long version of the option!
 
-my $command = $ENV{SSH_ORIGINAL_COMMAND};
-die "$0: Not invoked via sshd\n$Usage"	unless defined $command;
-die "$0: SSH_ORIGINAL_COMMAND='$command' is not rsync\n" unless $command =~ s/^rsync\s+//;
-die "$0: --server option is not first\n" unless $command =~ /^--server\s/;
-our $am_sender = $command =~ /^--server\s+--sender\s/; # Restrictive on purpose!
-die "$0 sending to read-only server not allowed\n" if $only eq 'r' && !$am_sender;
-die "$0 reading from write-only server not allowed\n" if $only eq 'w' && $am_sender;
-
-### START of options data produced by the cull_options script. ###
-
-# These options are the only options that rsync might send to the server,
-# and only in the option format that the stock rsync produces.
+### START of options data produced by the cull-options script. ###
 
 # To disable a short-named option, add its letter to this string:
-our $short_disabled = 's';
+short_disabled = 's'
 
-our $short_no_arg = 'ACDEHIJKLORSUWXbcdgklmnopqrstuvxyz'; # DO NOT REMOVE ANY
-our $short_with_num = '@B'; # DO NOT REMOVE ANY
+# These are also disabled when the restricted dir is not "/":
+short_disabled_subdir = 'KLk'
+
+# These are all possible short options that we will accept (when not disabled above):
+short_no_arg = 'ACDEHIJKLNORSUWXbcdgklmnopqrstuvxyz' # DO NOT REMOVE ANY
+short_with_num = '@B' # DO NOT REMOVE ANY
 
 # To disable a long-named option, change its value to a -1.  The values mean:
 # 0 = the option has no arg; 1 = the arg doesn't need any checking; 2 = only
 # check the arg when receiving; and 3 = always check the arg.
-our %long_opt = (
-  'append' => 0,
-  'backup-dir' => 2,
-  'block-size' => 1,
-  'bwlimit' => 1,
-  'checksum-choice' => 1,
-  'checksum-seed' => 1,
-  'compare-dest' => 2,
-  'compress-choice' => 1,
-  'compress-level' => 1,
-  'copy-dest' => 2,
-  'copy-unsafe-links' => 0,
-  'daemon' => -1,
-  'debug' => 1,
-  'delay-updates' => 0,
-  'delete' => 0,
-  'delete-after' => 0,
-  'delete-before' => 0,
-  'delete-delay' => 0,
-  'delete-during' => 0,
-  'delete-excluded' => 0,
-  'delete-missing-args' => 0,
-  'existing' => 0,
-  'fake-super' => 0,
-  'files-from' => 3,
-  'force' => 0,
-  'from0' => 0,
-  'fuzzy' => 0,
-  'group' => 0,
-  'groupmap' => 1,
-  'hard-links' => 0,
-  'iconv' => 1,
-  'ignore-errors' => 0,
-  'ignore-existing' => 0,
-  'ignore-missing-args' => 0,
-  'ignore-times' => 0,
-  'info' => 1,
-  'inplace' => 0,
-  'link-dest' => 2,
-  'links' => 0,
-  'list-only' => 0,
-  'log-file' => $only eq 'r' ? -1 : 3,
-  'log-format' => 1,
-  'max-alloc' => 1,
-  'max-delete' => 1,
-  'max-size' => 1,
-  'min-size' => 1,
-  'modify-window' => 1,
-  'new-compress' => 0,
-  'no-implied-dirs' => 0,
-  'no-r' => 0,
-  'no-relative' => 0,
-  'no-specials' => 0,
-  'numeric-ids' => 0,
-  'old-compress' => 0,
-  'one-file-system' => 0,
-  'only-write-batch' => 1,
-  'open-noatime' => 0,
-  'owner' => 0,
-  'partial' => 0,
-  'partial-dir' => 2,
-  'perms' => 0,
-  'preallocate' => 0,
-  'recursive' => 0,
-  'remove-sent-files' => $only eq 'r' ? -1 : 0,
-  'remove-source-files' => $only eq 'r' ? -1 : 0,
-  'safe-links' => 0,
-  'sender' => $only eq 'w' ? -1 : 0,
-  'server' => 0,
-  'size-only' => 0,
-  'skip-compress' => 1,
-  'specials' => 0,
-  'stats' => 0,
-  'suffix' => 1,
-  'super' => 0,
-  'temp-dir' => 2,
-  'timeout' => 1,
-  'times' => 0,
-  'use-qsort' => 0,
-  'usermap' => 1,
-  'write-devices' => -1,
-);
-
-### END of options data produced by the cull_options script. ###
-
-if ($short_disabled ne '') {
-    $short_no_arg =~ s/[$short_disabled]//go;
-    $short_with_num =~ s/[$short_disabled]//go;
-}
-$short_no_arg = "[$short_no_arg]" if length($short_no_arg) > 1;
-$short_with_num = "[$short_with_num]" if length($short_with_num) > 1;
-
-my $write_log = -f LOGFILE && open(LOG, '>>', LOGFILE);
-
-chdir($subdir) or die "$0: Unable to chdir to restricted dir: $!\n";
-
-my(@opts, @args);
-my $in_options = 1;
-my $last_opt = '';
-my $check_type;
-while ($command =~ /((?:[^\s\\]+|\\.[^\s\\]*)+)/g) {
-  $_ = $1;
-  if ($check_type) {
-    push(@opts, check_arg($last_opt, $_, $check_type));
-    $check_type = 0;
-  } elsif ($in_options) {
-    push(@opts, $_);
-    if ($_ eq '.') {
-      $in_options = 0;
-    } else {
-      die "$0: invalid option: '-'\n" if $_ eq '-';
-      next if /^-$short_no_arg*(e\d*\.\w*)?$/o || /^-$short_with_num\d+$/o;
-
-      my($opt,$arg) = /^--([^=]+)(?:=(.*))?$/;
-      my $disabled;
-      if (defined $opt) {
-	my $ct = $long_opt{$opt};
-	last unless defined $ct;
-	next if $ct == 0;
-	if ($ct > 0) {
-	  if (!defined $arg) {
-	    $check_type = $ct;
-	    $last_opt = $opt;
-	    next;
-	  }
-	  $arg = check_arg($opt, $arg, $ct);
-	  $opts[-1] =~ s/=.*/=$arg/;
-	  next;
-	}
-	$disabled = 1;
-	$opt = "--$opt";
-      } elsif ($short_disabled ne '') {
-	$disabled = /^-$short_no_arg*([$short_disabled])/o;
-	$opt = "-$1";
-      }
-
-      last unless $disabled; # Generate generic failure
-      die "$0: option $opt has been disabled on this server.\n";
-    }
-  } else {
-    if ($subdir ne '/') {
-      # Validate args to ensure they don't try to leave our restricted dir.
-      s{//+}{/}g;
-      s{^/}{};
-      s{^$}{.};
-    }
-    push(@args, bsd_glob($_, GLOB_LIMIT|GLOB_NOCHECK|GLOB_BRACE|GLOB_QUOTE));
-  }
-}
-die "$0: invalid rsync-command syntax or options\n" if $in_options;
-
-if ($subdir ne '/') {
-    die "$0: do not use .. in any path!\n" if grep m{(^|/)\.\.(/|$)}, @args;
+long_opts = {
+  'append': 0,
+  'backup-dir': 2,
+  'block-size': 1,
+  'bwlimit': 1,
+  'checksum-choice': 1,
+  'checksum-seed': 1,
+  'compare-dest': 2,
+  'compress-choice': 1,
+  'compress-level': 1,
+  'copy-dest': 2,
+  'copy-devices': -1,
+  'copy-unsafe-links': 0,
+  'daemon': -1,
+  'debug': 1,
+  'delay-updates': 0,
+  'delete': 0,
+  'delete-after': 0,
+  'delete-before': 0,
+  'delete-delay': 0,
+  'delete-during': 0,
+  'delete-excluded': 0,
+  'delete-missing-args': 0,
+  'existing': 0,
+  'fake-super': 0,
+  'files-from': 3,
+  'force': 0,
+  'from0': 0,
+  'fsync': 0,
+  'fuzzy': 0,
+  'group': 0,
+  'groupmap': 1,
+  'hard-links': 0,
+  'iconv': 1,
+  'ignore-errors': 0,
+  'ignore-existing': 0,
+  'ignore-missing-args': 0,
+  'ignore-times': 0,
+  'info': 1,
+  'inplace': 0,
+  'link-dest': 2,
+  'links': 0,
+  'list-only': 0,
+  'log-file': 3,
+  'log-format': 1,
+  'max-alloc': 1,
+  'max-delete': 1,
+  'max-size': 1,
+  'min-size': 1,
+  'mkpath': 0,
+  'modify-window': 1,
+  'msgs2stderr': 0,
+  'munge-links': 0,
+  'new-compress': 0,
+  'no-W': 0,
+  'no-implied-dirs': 0,
+  'no-msgs2stderr': 0,
+  'no-munge-links': -1,
+  'no-r': 0,
+  'no-relative': 0,
+  'no-specials': 0,
+  'numeric-ids': 0,
+  'old-compress': 0,
+  'one-file-system': 0,
+  'only-write-batch': 1,
+  'open-noatime': 0,
+  'owner': 0,
+  'partial': 0,
+  'partial-dir': 2,
+  'perms': 0,
+  'preallocate': 0,
+  'recursive': 0,
+  'remove-sent-files': 0,
+  'remove-source-files': 0,
+  'safe-links': 0,
+  'sender': 0,
+  'server': 0,
+  'size-only': 0,
+  'skip-compress': 1,
+  'specials': 0,
+  'stats': 0,
+  'stderr': 1,
+  'suffix': 1,
+  'super': 0,
+  'temp-dir': 2,
+  'timeout': 1,
+  'times': 0,
+  'use-qsort': 0,
+  'usermap': 1,
+  'write-devices': -1,
 }
 
-@args = ( '.' ) if !@args;
+### END of options data produced by the cull-options script. ###
 
-if ($write_log) {
-  my ($mm,$hh) = (localtime)[1,2];
-  my $host = $ENV{SSH_CONNECTION} || 'unknown';
-  $host =~ s/ .*//; # Keep only the client's IP addr
-  $host =~ s/^::ffff://;
-  $host = gethostbyaddr(inet_aton($host),AF_INET) || $host;
-  printf LOG "%02d:%02d %-13s [%s]\n", $hh, $mm, $host, "@opts @args";
-  close LOG;
-}
+import os, sys, re, argparse, glob, socket, time, subprocess
+from argparse import RawTextHelpFormatter
 
-# Note: This assumes that the rsync protocol will not be maliciously hijacked.
-exec(RSYNC, @opts, '--', @args) or die "exec(rsync @opts -- @args) failed: $? $!";
+try:
+    from braceexpand import braceexpand
+except:
+    braceexpand = lambda x: [ DE_BACKSLASH_RE.sub(r'\1', x) ]
 
-sub check_arg
-{
-  my($opt, $arg, $type) = @_;
-  $arg =~ s/\\(.)/$1/g;
-  if ($subdir ne '/' && ($type == 3 || ($type == 2 && !$am_sender))) {
-    $arg =~ s{//}{/}g;
-    die "Do not use .. in --$opt; anchor the path at the root of your restricted dir.\n"
-      if $arg =~ m{(^|/)\.\.(/|$)};
-    $arg =~ s{^/}{$subdir/};
-  }
-  $arg;
-}
+HAS_DOT_DOT_RE = re.compile(r'(^|/)\.\.(/|$)')
+LONG_OPT_RE = re.compile(r'^--([^=]+)(?:=(.*))?$')
+DE_BACKSLASH_RE = re.compile(r'\\(.)')
+
+def main():
+    if not os.path.isdir(args.dir):
+        die("Restricted directory does not exist!")
+
+    # The format of the environment variables set by sshd:
+    #   SSH_ORIGINAL_COMMAND:
+    #     rsync --server          -vlogDtpre.iLsfxCIvu --etc . ARG  # push
+    #     rsync --server --sender -vlogDtpre.iLsfxCIvu --etc . ARGS # pull
+    #   SSH_CONNECTION (client_ip client_port server_ip server_port):
+    #     192.168.1.100 64106 192.168.1.2 22
+
+    command = os.environ.get('SSH_ORIGINAL_COMMAND', None)
+    if not command:
+        die("Not invoked via sshd")
+    command = command.split(' ', 2)
+    if command[0:1] != ['rsync']:
+        die("SSH_ORIGINAL_COMMAND does not run rsync")
+    if command[1:2] != ['--server']:
+        die("--server option is not the first arg")
+    command = '' if len(command) < 3 else command[2]
+
+    global am_sender
+    am_sender = command.startswith("--sender ") # Restrictive on purpose!
+    if args.ro and not am_sender:
+        die("sending to read-only server is not allowed")
+    if args.wo and am_sender:
+        die("reading from write-only server is not allowed")
+
+    if args.wo or not am_sender:
+        long_opts['sender'] = -1
+    if args.no_del:
+        for opt in long_opts:
+            if opt.startswith(('remove', 'delete')):
+                long_opts[opt] = -1
+    if args.ro:
+        long_opts['log-file'] = -1
+
+    if args.dir != '/':
+        global short_disabled
+        short_disabled += short_disabled_subdir
+
+    short_no_arg_re = short_no_arg
+    short_with_num_re = short_with_num
+    if short_disabled:
+        for ltr in short_disabled:
+            short_no_arg_re = short_no_arg_re.replace(ltr, '')
+            short_with_num_re = short_with_num_re.replace(ltr, '')
+        short_disabled_re = re.compile(r'^-[%s]*([%s])' % (short_no_arg_re, short_disabled))
+    short_no_arg_re = re.compile(r'^-(?=.)[%s]*(e\d*\.\w*)?$' % short_no_arg_re)
+    short_with_num_re = re.compile(r'^-[%s]\d+$' % short_with_num_re)
+
+    log_fh = open(LOGFILE, 'a') if os.path.isfile(LOGFILE) else None
+
+    try:
+        os.chdir(args.dir)
+    except OSError as e:
+        die('unable to chdir to restricted dir:', str(e))
+
+    rsync_opts = [ '--server' ]
+    rsync_args = [ ]
+    saw_the_dot_arg = False
+    last_opt = check_type = None
+
+    for arg in re.findall(r'(?:[^\s\\]+|\\.[^\s\\]*)+', command):
+        if check_type:
+            rsync_opts.append(validated_arg(last_opt, arg, check_type))
+            check_type = None
+        elif saw_the_dot_arg:
+            # NOTE: an arg that starts with a '-' is safe due to our use of "--" in the cmd tuple.
+            try:
+                b_e = braceexpand(arg) # Also removes backslashes
+            except: # Handle errors such as unbalanced braces by just de-backslashing the arg:
+                b_e = [ DE_BACKSLASH_RE.sub(r'\1', arg) ]
+            for xarg in b_e:
+                rsync_args += validated_arg('arg', xarg, wild=True)
+        else: # parsing the option args
+            if arg == '.':
+                saw_the_dot_arg = True
+                continue
+            rsync_opts.append(arg)
+            if short_no_arg_re.match(arg) or short_with_num_re.match(arg):
+                continue
+            disabled = False
+            m = LONG_OPT_RE.match(arg)
+            if m:
+                opt = m.group(1)
+                opt_arg = m.group(2)
+                ct = long_opts.get(opt, None)
+                if ct is None:
+                    break # Generate generic failure due to unfinished arg parsing
+                if ct == 0:
+                    continue
+                opt = '--' + opt
+                if ct > 0:
+                    if opt_arg is not None:
+                        rsync_opts[-1] = opt + '=' + validated_arg(opt, opt_arg, ct)
+                    else:
+                        check_type = ct
+                        last_opt = opt
+                    continue
+                disabled = True
+            elif short_disabled:
+                m = short_disabled_re.match(arg)
+                if m:
+                    disabled = True
+                    opt = '-' + m.group(1)
+
+            if disabled:
+                die("option", opt, "has been disabled on this server.")
+            break # Generate a generic failure
+
+    if not saw_the_dot_arg:
+        die("invalid rsync-command syntax or options")
+
+    if args.munge:
+        rsync_opts.append('--munge-links')
+    
+    if args.no_overwrite:
+      rsync_opts.append('--ignore-existing')
+
+    if not rsync_args:
+        rsync_args = [ '.' ]
+
+    cmd = (RSYNC, *rsync_opts, '--', '.', *rsync_args)
+
+    if log_fh:
+        now = time.localtime()
+        host = os.environ.get('SSH_CONNECTION', 'unknown').split()[0] # Drop everything after the IP addr
+        if host.startswith('::ffff:'):
+            host = host[7:]
+        try:
+            host = socket.gethostbyaddr(socket.inet_aton(host))
+        except:
+            pass
+        log_fh.write("%02d:%02d:%02d %-16s %s\n" % (now.tm_hour, now.tm_min, now.tm_sec, host, str(cmd)))
+        log_fh.close()
+
+    # NOTE: This assumes that the rsync protocol will not be maliciously hijacked.
+    if args.no_lock:
+        os.execlp(RSYNC, *cmd)
+        die("execlp(", RSYNC, *cmd, ')  failed')
+    child = subprocess.run(cmd)
+    if child.returncode != 0:
+        sys.exit(child.returncode)
+
+
+def validated_arg(opt, arg, typ=3, wild=False):
+    if opt != 'arg': # arg values already have their backslashes removed.
+        arg = DE_BACKSLASH_RE.sub(r'\1', arg)
+
+    orig_arg = arg
+    if arg.startswith('./'):
+        arg = arg[1:]
+    arg = arg.replace('//', '/')
+    if args.dir != '/':
+        if HAS_DOT_DOT_RE.search(arg):
+            die("do not use .. in", opt, "(anchor the path at the root of your restricted dir)")
+        if arg.startswith('/'):
+            arg = args.dir + arg
+
+    if wild:
+        got = glob.glob(arg)
+        if not got:
+            got = [ arg ]
+    else:
+        got = [ arg ]
+
+    ret = [ ]
+    for arg in got:
+        if args.dir != '/' and arg != '.' and (typ == 3 or (typ == 2 and not am_sender)):
+            arg_has_trailing_slash = arg.endswith('/')
+            if arg_has_trailing_slash:
+                arg = arg[:-1]
+            else:
+                arg_has_trailing_slash_dot = arg.endswith('/.')
+                if arg_has_trailing_slash_dot:
+                    arg = arg[:-2]
+            real_arg = os.path.realpath(arg)
+            if arg != real_arg and not real_arg.startswith(args.dir_slash):
+                die('unsafe arg:', orig_arg, [arg, real_arg])
+            if arg_has_trailing_slash:
+                arg += '/'
+            elif arg_has_trailing_slash_dot:
+                arg += '/.'
+            if opt == 'arg' and arg.startswith(args.dir_slash):
+                arg = arg[args.dir_slash_len:]
+                if arg == '':
+                    arg = '.'
+        ret.append(arg)
+
+    return ret if wild else ret[0]
+
+
+def lock_or_die(dirname):
+    import fcntl
+    global lock_handle
+    lock_handle = os.open(dirname, os.O_RDONLY)
+    try:
+        fcntl.flock(lock_handle, fcntl.LOCK_EX | fcntl.LOCK_NB)
+    except:
+        die('Another instance of rrsync is already accessing this directory.')
+
+
+def die(*msg):
+    print(sys.argv[0], 'error:', *msg, file=sys.stderr)
+    if sys.stdin.isatty():
+        arg_parser.print_help(sys.stderr)
+    sys.exit(1)
+
+
+# This class displays the --help to the user on argparse error IFF they're running it interactively.
+class OurArgParser(argparse.ArgumentParser):
+    def error(self, msg):
+        die(msg)
+
+
+if __name__ == '__main__':
+    our_desc = """Use "man rrsync" to learn how to restrict ssh users to using a restricted rsync command."""
+    arg_parser = OurArgParser(description=our_desc, add_help=False)
+    only_group = arg_parser.add_mutually_exclusive_group()
+    only_group.add_argument('-ro', action='store_true', help="Allow only reading from the DIR. Implies -no-del and -no-lock.")
+    only_group.add_argument('-wo', action='store_true', help="Allow only writing to the DIR.")
+    arg_parser.add_argument('-munge', action='store_true', help="Enable rsync's --munge-links on the server side.")
+    arg_parser.add_argument('-no-del', action='store_true', help="Disable rsync's --delete* and --remove* options.")
+    arg_parser.add_argument('-no-lock', action='store_true', help="Avoid the single-run (per-user) lock check.")
+    arg_parser.add_argument('-no-overwrite', action='store_true', help="Prevent overwriting existing files by enforcing --ignore-existing")
+    arg_parser.add_argument('-help', '-h', action='help', help="Output this help message and exit.")
+    arg_parser.add_argument('dir', metavar='DIR', help="The restricted directory to use.")
+    args = arg_parser.parse_args()
+    args.dir = os.path.realpath(args.dir)
+    args.dir_slash = args.dir + '/'
+    args.dir_slash_len = len(args.dir_slash)
+    if args.ro:
+        args.no_del = True
+    elif not args.no_lock:
+        lock_or_die(args.dir)
+    main()
+
+# vim: sw=4 et