Karel Zak [Thu, 14 Apr 2016 12:26:54 +0000 (14:26 +0200)]
libmount: don't support /etc/mtab by default
The file mtab is evil and already unused by mainstream distributions.
Now libmount is able to detect mtab->/proc/mounts and use
/proc/self/mountinfo if necessary. This heuristic seems overkill in
many cases. It's also dangerous on systems where mountinfo is strongly
required (systemd based distros).
This patch #ifdefs mtab code and forces libmount to always use
/proc/self/mountinfo.
The new configure option --enable-libmount-support-mtab is necessary
to enable old behavior to support mtab.
Karel Zak [Wed, 13 Apr 2016 12:41:47 +0000 (14:41 +0200)]
wipefs: force GPT detection
The library libblkid (as well as fdisks) requires protective MBR when
probe for GPT by default. This is unnecessary for wipefs where we're
more promiscuous and we want to delete as much as possible. This patch
enables BLKID_PARTS_FORCE_GPT for wipefs.
Addresses: https://bugzilla.redhat.com/show_bug.cgi?id=1326474 Signed-off-by: Karel Zak <kzak@redhat.com>
Karel Zak [Thu, 24 Mar 2016 10:51:12 +0000 (11:51 +0100)]
libmount: try absolute target before canonicalize
The path canonicalization is expensive and in many cases unwanted due
to problems with readlink() on unreachable NFS and automounters.
This patch add a possibility to search also by $(CWD)/<path> if the
<path> is relative to reduce number of situation when we convert the
path to the canonical absolute path.
Karel Zak [Wed, 13 Apr 2016 09:52:43 +0000 (11:52 +0200)]
script: use empty-slave heuristic more carefully
script(1) waits for empty slave FD (shell stdin) before it writes to
master. This feature has been intorduiced by 54c6611d6f7b73609a5331f4d0bcf63c4af6429e
to avoid misbehavior when we need to send EOF to the shell.
Unfortunately, this feature has been used all time for all messages.
This is wrong because command in the session (or shell) may ignore
stdin at all and wait forever in busy loop is really bad idea. Test
case:
This patch forces script to use empty-stave detection only when we
need to write EOF. The busy loop has been modified to use nanosleep
and it does not wait forever...
Addresses: http://bugs.debian.org/820843 Signed-off-by: Karel Zak <kzak@redhat.com>
build-sys: add --disable-logger and --disable-lslogins
Now we are able to disable all programs which have systemd/journald
support. This feature is needed by openSUSE packagers who are building
util-linux in 2 stages to avoid build cycles.
Sami Kerola [Sun, 3 Apr 2016 08:35:28 +0000 (09:35 +0100)]
bash-completion: update chrt completion
Add couple missing options, and make the completion overall work better.
That said completion is still incomplete, pardon the pun. After user has
specified policy then giving a hint what priority needs to be specified is
theoretically possible, but such hint is not given. There does not seem to
be easy way to know when user wants stops specifying options and move to
defining priority in: chrt [options] [prio] [command|pid].
Karel Zak [Wed, 30 Mar 2016 12:53:33 +0000 (14:53 +0200)]
libblkid: revert mmap usage
The implementation has not been ready for I/O errors and it seems that
there is no elegant way how to resolve this issue. Linux returns
SIGBUS on mmap errors and play with signals (or longjumps) in shared
library is really bad idea.
It also seems that mmaped devices have some unexpected side-effects
with page-cache where for example dd returns old data for already
modified device etc.
Add some more tests:
* error handling "unknown arch" and "unknown command"
* "noop" test with host arch and no other options
* all options except --uname-2.6
* --uname-2.6 whithout any other options but handle fatal
glibc error "kernel too old" (with debug output)
* add a "real" --uname-2.6 test which validates uname(1)
output
Note the "kernel too old" cases are systems where glibc was
configured to support only kernels > 3.0. On some archs
(e.g. x86) --uname-2.6 still works with such glibc because
2.6.39 had all needed features on board (e.g. vdso).
See https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=806911
Some people reported segfaults (after execvp) but I can only
reproduce it on arm* and aarch64 qemu-user-space builds. We
don't need to fix our tests for such broken systems where
also many other tests fail currently.
Kent Overstreet [Tue, 15 Mar 2016 02:44:36 +0000 (18:44 -0800)]
libblkid: Update for newer bcache superblocks
Later version of bcache add different checksum types, and allow for superblocks
greater than 4k - skipping the checksum check (as in most other probes) is the
easiest solution.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
Werner Fink [Tue, 22 Mar 2016 09:38:59 +0000 (10:38 +0100)]
sulogin: Always make echo work after performing getpasswd even if root account is locked
If the root account is locked and no password was provided then the terminal
line is not set back to do echo of the input. This correct a small overlook
in commit 7ff1162e67164cb4ece19dd809c26272461aa254
Dongli Zhang [Tue, 22 Mar 2016 05:38:14 +0000 (13:38 +0800)]
lscpu: correct the Virtualization type on Xen DomU PV guest
Nowadays, most Intel CPUs have "cpuid faulting" available which could trap
the execution of "cpuid" instruction when CPL>0 with GP fault. Thus,
"cpuid" instruction could trap to Xen hypervisor on the paravirtualized PV
guest on most servers today, except on old CPUs prior to 2011. On CPU after
2011, Xen will put "XenVMMXenVMM" on both HVM and PV guests, which could
have lscpu command erroneously classify the guest as type "full". The
current lscpu command, which is based on "cpuid" instruction, still assumes
that it will not cause the trap to Xen hypervisor on Xen PV guest and uses
/proc/xen to identify whether it's running on PV DomU or not. To identify
this kind of information under the help of
/sys/hypervisor/properties/features would be more accurate for the CPU
nowadays. The bit 5 (XENFEAT_mmu_pt_update_preserve_ad) of the features
will be set only when it's running on Xen PV domain. The combo of bit 3 and
8 (XENFEAT_supervisor_mode_kernel and XENFEAT_hvm_callback_vector) will be
set simultaneously only when it's running on Xen PVH domain.
[kzak@redhat.com: - add path_exist()]
Signed-off-by: Dongli Zhang <dongli.zhang@oracle.com> Signed-off-by: Karel Zak <kzak@redhat.com>
Karel Zak [Mon, 21 Mar 2016 20:13:25 +0000 (21:13 +0100)]
docs: add note about CLONE_NEWCGROUP support
The patch has been planned for weeks and now the kernel part is
already in Linus' tree. It's a new feature, but it's probably better
to merge the userspace stuff now (v2.28 rc1) than wait next 6 months
for the next util-linux release.
Ruediger Meier [Mon, 21 Mar 2016 13:04:45 +0000 (14:04 +0100)]
ipcs: consolidate output header printing
Print a warning (instead of header) if --limits fails, like we did
it in past (2.20.1) and like we are still doing for --summary. Note
in past we were printing the same message like for --summary
"kernel not configured for ...", but actually this message is not
really correct.
This patch simply consolidates the current behavior. Probably we
should refactor it regarding warnings (stderr) and exit codes.
Ruediger Meier [Fri, 18 Mar 2016 11:49:17 +0000 (12:49 +0100)]
ipcs: make sure to parse whole lines for shm_data
We want to parse 16 columns _per_row_ without mixing them up. The
existing code is unsafe for more or less columns and could even
run into endless loops. This patch assures that we parse row-wise
and really skip lines with columns != 16.
Probably somehow we could have also done this with fscanf() only.
Using fgets() additionally makes the code more easy to read and
to improve later.
Karel Zak [Fri, 18 Mar 2016 13:28:29 +0000 (14:28 +0100)]
swapon: rewrite control struct usage
The old version has been pretty broken... the most important is to
keep swap options specified on command line as read-only template.
For example if we call "swapon --all" then we cannot modify the global
options for each fstab swap entry.
The another story has been control struct modification due to device
reinitialization etc.
This patch splits all to:
* struct swapon_control; top-level struct with command line options
* struct swap_device; this is device specific and never globally
maintained by swapon_control.
* struct swap_prop; used as global read-only template swap options
and per device swap options (when parse fstab).
Addresses: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=818252 Signed-off-by: Karel Zak <kzak@redhat.com>
Karel Zak [Wed, 16 Mar 2016 12:01:10 +0000 (13:01 +0100)]
tests: mark old bsd fdisk test as deprecated
It seems better to use hexdump rather than md5sum, but it means that
we have to gather hexdumps of the all possible BSD variants. For this
purpose will be introduced a new bsd fdisk test and to verify the
new hexdumps we can use this old test as both tests are exactly the
same.
projects / thirdparty / util-linux.git / log
