* src/copy.c (copy_reg): When cloning only skip the data copying
* tests/cp/reflink-perm: New test to check times and modes copied
* tests/Makefile.am: Reference the new test
* NEWS: Mention the fix
* doc/coreutils.texi (cp invocation): Document the new
"auto" and "always" options to --reflink.
* src/copy.c (copy_reg): Fall back to a standard copy
when reflink() is not supported and --reflink=auto specified.
* src/copy.h [struct cp_options] (reflink): Change type s/bool/enum/.
* src/cp.c (usage): Describe the --reflink={always,auto} options
and expand a little on what --reflink does.
(main): parse the new parameters to --reflink and allow all
--sparse options with --reflink=auto.
* src/install.c (cp_option_init): Init the enum instead of bool.
* src/mv.c (cp_option_init): Likewise.
* tests/cp/reflink-auto: A new test for falling back to normal copy.
* tests/Makefile.am: Reference the new test.
* NEWS: Mention the new feature.
* tests/cp/cp-mv-enotsup-xattr: Upon a set-up failiure, rather than
failing the test with a "framework failure" diagnostic, just skip it.
Russell Whitaker reported that this test failed on slackware.
* src/copy.c (copy_internal): On systems without utimensat don't
use utimens on a symlink, as that would dereference the symlink.
* tests/cp/abuse: To work around possible attribute preservation
failures breaking the test, use cp -dR rather than cp -a.
* NEWS: Mention it.
* doc/coreutils.texi (cp invocation): Describe it.
* src/copy.h (struct cp_options) [reflink]: New member.
* src/copy.c (usage): Describe it.
(copy_reg): If reflink is true try to clone the file.
(main): Check for --reflink.
(cp_option_init): Initialize the new member.
* src/install.c (cp_option_init): Initialize the new member.
* src/mv.c (cp_option_init): Likewise.
* tests/cp/sparse: Add a new test case.
* src/copy.c (utimensat_if_possible): New function.
(copy_internal): Remove variable, "preserve_metadata".
Replace with "dest_is_symlink". That covers all cases but one:
the one in which cp --link has created hard links to non-directories.
In that case, there is no need to update attributes of the links.
Use utimensat_if_possible, to preserve timestamps of symlinks.
* NEWS (New features): Mention this.
* tests/Makefile.am (TESTS): Add cp/preserve-slink-time.
* tests/cp/preserve-slink-time: New file.
* m4/jm-macros.m4 (coreutils_MACROS): Test for utimensat.
Reported in http://bugzilla.redhat.com/230866
Remove the optimization that avoided up to 50% of cp's read syscalls.
Do not assume that a short read on a regular file indicates EOF.
When reading from a file in /proc on linux [at least 2.6.9 - 2.6.29]
into a 4k-byte buffer or larger, a short read does not
always indicate EOF. For example, "cp /proc/slabinfo /tmp"
copies only 4068 of the total 7493 bytes. This optimization
(25719a3315, Improve performance a bit
by optimizing away; 2005-11-24) appears to have been worth less than
a 2% speed-up (and usually much less), so the impact of removing it
is negligible.
* src/copy.c (copy_reg): Don't exit the loop early.
* tests/cp/proc-short-read: New test, lightly based on a suggestion
from Mike Frysinger, to exercise this fix.
* tests/Makefile.am (TESTS): Add cp/proc-short-read.
* NEWS (Improve robustness): Mention this change.
* tests/test-lib.sh (skip_if_setgid_): New function.
* tests/setgid-check: Remove file.
* tests/Makefile.am (EXTRA_DIST): Remove setgid-check.
* tests/chmod/c-option: Use the new function rather than sourcing
the separate file.
* tests/cp/cp-parents: Likewise.
* tests/install/install-C: Likewise.
* tests/mkdir/parents: Likewise.
* tests/mkdir/perm: Likewise.
Running "mkdir dir; cp -rl dir dir" would create dir/dir/dir/...
rather than diagnosing the "copy-into-self" failure.
The easy fix would have been to revert this part of the change
[3ece0355 2008-11-09 cp: use far less memory in some cases]
that introduced the bug:
- remember_copied (dst_name, dst_sb.st_ino, dst_sb.st_dev);
+ if (!x->hard_link)
+ remember_copied (dst_name, dst_sb.st_ino, dst_sb.st_dev);
However, that would have induced the failure of the new cp/link-heap
test, due to the added memory pressure of recording 10k dev/ino pairs.
And besides, I liked that improvement and wanted to keep it.
Now that it's obvious recording the just-created-directory dev/ino
needn't depend on the setting of hard_link, I realized it is necessary
to record the pair only for the first directory created for each
source command-line argument.
I made that change, then noticed the new test, cp -rl a d d, would pass
when run once, yet output the into-self diagnostic twice. Also note
the side effect: it creates d/a and d/d. However, running that same
command a second time, now with the modified directory, would fail.
That turned out to be due to the fact that although the first into-self
failure was detected in copy_dir, that function would continue copying
other entries regardless -- and that would make it fail (eventually)
with the unwanted recursion.
* src/copy.c (copy_internal): This function needed an indicator of
whether, for a give command line argument, it had already created its
first directory. If so, no more need to record dev/ino pairs. If this
is the first, then do record its pair. Hence, the new parameter.
(copy_dir, copy): Update callers.
(copy_dir): Upon any into-self failure, break out of the loop.
* tests/cp/into-self: Test for the above.
Reported by Mikael Magnusson.
* copy.c (copy_reg): Reduce SELinux context diagnostics for 'cp -a'.
(copy_internal): Likewise
* copy.h (cp_options): Add boolean reduce_diagnostics.
* cp.c (usage): Say that --archive (-a) behaves like -dR --preserve=all.
(cp_option_init): Initialize added reduce_diagnostics.
(main): Add reduce_diagnostics for the -a option, and preserve SELinux
context, if possible.
* mv.c (cp_options_init): Initialize new cp_options booleans.
* install.c (cp_option_init): Likewise.
* NEWS: Mention those behaviour changes.
* doc/coreutils.texi: Document --preserve=context, document that
diagnostics are not shown for failures of non-mandatory attributes
(just SELinux at the moment).
* tests/cp/cp-a-selinux: Check not only failures, but succesful use
of preserving SELinux context in cp.
* src/cp.c (usage): Show new option -n in --help.
(main): Handle new option -n.
* src/mv.c (usage): Show new option -n in --help.
(main): Handle new option -n.
* doc/coreutils.texi: Document new cp/mv option -n.
* tests/cp/cp-i: Add tests for -f, -i and -n options.
* tests/mv/mv-n: New test for mv -n.
* tests/Makefile.am: Add test mv/mv-n to the list.
* NEWS: Mention the change.
cp --link was "remembering" many name,dev,inode triples unnecessarily.
cp was doing the same, even without --link, for every directory in the
source hierarchy, while it can do its job with entries merely for the
command-line arguments. Prompted by a report from Patrick Shoenfeld.
Details <http://thread.gmane.org/gmane.comp.gnu.coreutils.bugs/15081>.
* src/copy.c (copy_internal): Refrain from remembering
name,dev,inode for most files, when invoked via cp --link.
Record an infloop-avoidance triple for each directory specified
on the command line, not for each directory in the source tree.
Don't record a dir-triple when x->hard_link is set.
* NEWS (Buf fixes): Mention it.
* tests/cp/link-heap: New file. Test for cp's lowered memory usage.
* tests/Makefile.am (TESTS): Add link-heap.
* tests/test-lib.sh (Exit): New function by Ralf Wildenhues in automake
http://git.sv.gnu.org/gitweb/?p=automake.git;a=commitdiff;h=20594c08f63
* tests/**: Convert all uses:
This restrictive change converted the vast majority:
git grep -l '^(exit \$fail); exit \$fail$' \
| xargs perl -pi -e 's/'^\(exit \$fail\); exit \$fail$/Exit \$fail/'
And this did the rest, plus a few undesirable ones, so I manually
backed out the changes to ChangeLog-* and build-aux/check.mk:
git grep -l -E '\(exit [^)]+\); exit ' \
| xargs perl -pi -e 's/\(exit (.+?)\); exit \1/Exit $1/'
* tests/sparse-file: Remove file Move contents into ...
* tests/test-lib.sh (require_sparse_support_): ...here. New function.
* tests/cp/sparse: Use the function, not the file.
* tests/du/8gb: Likewise.
* tests/Makefile.am (EXTRA_DIST): Remove sparse-file.
* README (Running tests as root): Also set PATH in suggested "sudo"
command. This avoids failure of at least tests/cp/cp-a-selinux
when the default PATH does not contain /sbin.
* tests/cp/cp-a-selinux: Don't redirect stderr to /dev/null.
"mkfs" was failing due to /sbin not being in PATH.
* tests/cp/perm: Use stat to get the permission string, not ls.
This test was run only when RUN_VERY_EXPENSIVE_TESTS=yes was set
in the environment. It would fail on SELinux-enable systems
because ls-generated permission strings would not match, e.g.,
"test _-rw-r--r--+ = _-rw-r--r--" would fail.
* tests/test-lib.sh (working_umask_or_skip_): New function, from...
* tests/umask-check: ...here. Remove file.
* tests/Makefile.am (EXTRA_DIST): Remove umask-check.
* tests/mkdir/perm: Use the function rather than sourcing the file.
* tests/cp/cp-parents: Likewise.
* tests/cp/parent-perm: Likewise.
Bruno Haible reported that parent-perm was failing to run umask-check.
One side-effect of this change is that "make check" now works even if
you put "." early in your shell's search PATH (don't do that!).
Remove all test-related Makefile.am files, except those generated
by mk-script. Instead, tests/Makefile.am now lists not only the
tests directly under tests/, but also those in tests/*/ that are
not generated by mk-script, e.g., cp/abuse, cp/acl, mv/i-1, etc.
A lot of these changes are like this:
-. $srcdir/../lang-default
+. $top_srcdir/tests/lang-default
-. $srcdir/../test-lib.sh
+. $top_srcdir/tests/test-lib.sh
* configure.ac (AC_CONFIG_FILES): Remove corresponding Makefiles.
* tests/check.mk (vc_exe_in_TESTS): Relax syntax requirements.
* tests/rwx-to-mode: Remove file. Rewritten as...
* tests/test-lib.sh (rwx_to_mode_): ...this new function.
* tests/Makefile.am (EXTRA_DIST): Remove rwx-to-mode.
(SUBDIRS): Remove each dir with a removed Makefile.am.
(EXTRA_DIST): Add $(TESTS).
(TESTS): Add over 300 entries.
* src/copy.c (copy_internal): Revert change of 2005-03-01,
4303f04545
Prompted by Michael Stone, who pointed me to an old bug report from
Ian Jackson: http://thread.gmane.org/gmane.comp.gnu.coreutils.bugs/7504
* tests/cp/special-f: New file. Test for the above.
* tests/cp/Makefile.am (TESTS): Add special-f.
* NEWS: mention this bug fix.
When mcstransd is not running (i.e., after service mcstrans stop),
"make check" as root would provoke two test failures.
* tests/cp/cp-a-selinux: Use the context, root:object_r:tmp_t:s0,
that works both with and without mcstransd.
Thanks to Eric Paris for the tip and to Ondřej Vašík for alerting
me to the problem.
Reported by Robert Scheck in <http://bugzilla.redhat.com/436717>.
* tests/priv-check: Remove file.
* tests/Makefile.am (EXTRA_DIST): Remove priv-check.
* tests/test-lib.sh (require_root_): New function.
Use this function rather than sourcing the priv-check file.
* tests/sample-test: Use require_root_ rather than priv-check.
* tests/chown/basic: Likewise.
* tests/cp/cp-a-selinux: Likewise.
* tests/cp/preserve-gid: Likewise.
* tests/cp/special-bits: Likewise.
* tests/ls/nameless-uid: Likewise.
* tests/misc/chcon: Likewise.
* tests/mkdir/writable-under-readonly: Likewise.
* tests/mv/sticky-to-xpart: Likewise.
* tests/rm/fail-2eperm: Likewise.
* tests/rm/no-give-up: Likewise.
* tests/rm/one-file-system: Likewise.
* tests/tail-2/append-only: Likewise.
* tests/touch/now-owned-by-other: Likewise.
* tests/rm/fail-eperm: Use skip_if_root_ rather than priv-check.
* Makefile.maint (sc_root_tests): Reflect this change:
search for the new function name.
This avoids another spurious Mac OS 10.5.1 (Darwin 9.1) test failure.
* tests/cp/preserve-gid: Accept a group ID of 0, as well.
Reported by Elias Pipping.
