]> sourceware.org Git - lvm2.git/log
lvm2.git
4 years agotests: aux better handle invalid table
Zdenek Kabelac [Fri, 18 Sep 2020 12:23:20 +0000 (14:23 +0200)]
tests: aux better handle invalid table

In cases where internal functions like zero_dev, delay_dev pass-in
invalid parameter so resulting table can't work, resume at least
previous table line before failing out - so the cleaning process
later on is not stuck waiting on a suspended device.

4 years agotests: also use sed to shorten log output
Zdenek Kabelac [Thu, 17 Sep 2020 22:31:59 +0000 (00:31 +0200)]
tests: also use sed to shorten log output

4 years agotests: resolve missing removal of loopdevice on error path
Zdenek Kabelac [Thu, 17 Sep 2020 22:31:11 +0000 (00:31 +0200)]
tests: resolve missing removal of loopdevice on error path

In case of test failure, loop device leaked and occupied space forever.

4 years agotests: bigger data still needed for 0.7.0
Zdenek Kabelac [Thu, 17 Sep 2020 21:26:03 +0000 (23:26 +0200)]
tests: bigger data still needed for 0.7.0

4 years agotests: lower memory usage
Zdenek Kabelac [Thu, 17 Sep 2020 21:25:23 +0000 (23:25 +0200)]
tests: lower memory usage

Reduce memory needed by test at one time.

4 years agoconfigure: enable integrity by default
David Teigland [Wed, 16 Sep 2020 20:14:51 +0000 (15:14 -0500)]
configure: enable integrity by default

4 years agohints: enhance debug messages
David Teigland [Wed, 16 Sep 2020 19:46:50 +0000 (14:46 -0500)]
hints: enhance debug messages

4 years agolabel: cleanup set_byte error exit
David Teigland [Wed, 16 Sep 2020 18:54:16 +0000 (13:54 -0500)]
label: cleanup set_byte error exit

4 years agoRevert "label: use formaters FMTu64 and FMTsize_t"
David Teigland [Wed, 16 Sep 2020 18:47:06 +0000 (13:47 -0500)]
Revert "label: use formaters FMTu64 and FMTsize_t"

This reverts commit d0ccb2521ba4f34455078f87f46cc41f5761ffc0.

4 years agotests: smaller delay and lowered version
Zdenek Kabelac [Wed, 16 Sep 2020 11:21:16 +0000 (13:21 +0200)]
tests: smaller delay and lowered version

See if this will still work. Some boxes are delayed too much.
Also try to check for raid extend progress from version 1.13.

4 years agotests: fix bash regex syntax
Zdenek Kabelac [Wed, 16 Sep 2020 11:12:17 +0000 (13:12 +0200)]
tests: fix bash regex syntax

Typo before last commit.

4 years agotests: switch for checking version of installed tools
Zdenek Kabelac [Mon, 14 Sep 2020 19:25:04 +0000 (21:25 +0200)]
tests: switch for checking version of installed tools

It looks like older tools were compacting metadata more.

4 years agotests: skip kernel for this test
Zdenek Kabelac [Mon, 14 Sep 2020 19:23:16 +0000 (21:23 +0200)]
tests: skip kernel for this test

Kills this kernel ATM

4 years agolabel: use formaters FMTu64 and FMTsize_t
Zdenek Kabelac [Tue, 15 Sep 2020 19:47:39 +0000 (21:47 +0200)]
label: use formaters FMTu64 and FMTsize_t

Produces code without casts to differntly signed types
and also shortens and enhances readbility.

4 years agowipe: dev_set_bytes resolves zeroing
Zdenek Kabelac [Tue, 15 Sep 2020 18:35:48 +0000 (20:35 +0200)]
wipe: dev_set_bytes resolves zeroing

Since dev_write_zeros() is just subset of dev_set_bytes()
use it directly and simplify code.

4 years agowipe: convert zero_value to uint8_t
Zdenek Kabelac [Tue, 15 Sep 2020 14:40:53 +0000 (16:40 +0200)]
wipe: convert zero_value to uint8_t

We always write this value as byte.

4 years agowipe: zeroing of 8 sectors is granted
Zdenek Kabelac [Tue, 15 Sep 2020 14:38:50 +0000 (16:38 +0200)]
wipe: zeroing of 8 sectors is granted

With do_zero min is always 8 sectors, so use 0 as default.

4 years agolabel: deduplicate dev_set_bytes
Zdenek Kabelac [Tue, 15 Sep 2020 14:29:12 +0000 (16:29 +0200)]
label: deduplicate dev_set_bytes

As dev_write_zeros() is same as dev_set_bytes() reused the code
directly.

4 years agolabel: code deduplication
Zdenek Kabelac [Tue, 15 Sep 2020 14:20:16 +0000 (16:20 +0200)]
label: code deduplication

4 years agohints: enhance debug with log_sys_debug
Zdenek Kabelac [Tue, 15 Sep 2020 19:33:30 +0000 (21:33 +0200)]
hints: enhance debug with log_sys_debug

4 years agolvcreate: change error message
Zdenek Kabelac [Tue, 15 Sep 2020 20:45:49 +0000 (22:45 +0200)]
lvcreate: change error message

Provide more useful error message.

4 years agolvcreate: add extra synchronization at error path
Zdenek Kabelac [Fri, 4 Sep 2020 16:08:59 +0000 (18:08 +0200)]
lvcreate: add extra synchronization at error path

Put explict udev synchronization before we try to deactive devices.

4 years agotests: protect this test for another kernel
Zdenek Kabelac [Sat, 12 Sep 2020 17:08:24 +0000 (19:08 +0200)]
tests: protect this test for another kernel

Thisi 3.10.0-862 kernel also dies with this test.

4 years agotests: tune usage of smaller metadata
Zdenek Kabelac [Sat, 12 Sep 2020 17:06:05 +0000 (19:06 +0200)]
tests: tune usage of smaller metadata

While the previous commit c9b40083fc34b5e2a1bfc7b251b38c0b8737483f
decresed version to 1.19 for using bigger datasets,  it's not
been quite right - so from our bb machine it looks like
bigger metadata consumption started with 1.19 and kernel 4.18
(fc27)

4 years agotests: raise needed target version
Zdenek Kabelac [Sat, 12 Sep 2020 17:09:41 +0000 (19:09 +0200)]
tests: raise needed target version

Require higher version to avoid early bugs.

4 years agotests: improve cache abort test
Zdenek Kabelac [Sun, 13 Sep 2020 09:33:13 +0000 (11:33 +0200)]
tests: improve cache abort test

Use bigger volume and slowdown writing to cache device.
This allows more simple to reach 'dirty' state.
Also document exactly 1 SIGINT has to fire aborting of flushing.

4 years agotests: check in_sync prints also dm status
Zdenek Kabelac [Sat, 12 Sep 2020 17:04:56 +0000 (19:04 +0200)]
tests: check in_sync prints also dm status

It's more useful to see how the progress of status checking is moving.

4 years agocache: simplier signal handling
Zdenek Kabelac [Sun, 13 Sep 2020 09:33:39 +0000 (11:33 +0200)]
cache: simplier signal handling

Use just single sigint_allow()/restore() within flushing loop
and void one extra signal manipulation.

4 years agoheaders: remove unused headers
Zdenek Kabelac [Sun, 13 Sep 2020 09:32:29 +0000 (11:32 +0200)]
headers: remove unused headers

4 years agoconfigure: just upper case start of sentence
Zdenek Kabelac [Sat, 12 Sep 2020 17:04:22 +0000 (19:04 +0200)]
configure: just upper case start of sentence

4 years agolocking: restore blocking signal for VG_GLOBAL lck
Zdenek Kabelac [Sun, 13 Sep 2020 19:10:17 +0000 (21:10 +0200)]
locking: restore blocking signal for VG_GLOBAL lck

During removal of a lot of locking code the signal blocking got lost
and signal processing got broken leading to unpredictable
behavior of i.e. activation code the can get interrupted in the
middle of DM table processing.

lvm2 code always expects signals are blocked while lock is held
unless it is explictelly placed into section of:
sigint_allow();....;sigint_restore();
For checking catched interrupt there is sigint_catched();

4 years agotests: avoid using string
Zdenek Kabelac [Sat, 12 Sep 2020 10:25:58 +0000 (12:25 +0200)]
tests: avoid using string

String 'TEST WARNING' may not be present in the test script itself.
Add '\ ' to avoid 'grep' matching test as the test with warning.

4 years agotests: support for 16T is needed
Zdenek Kabelac [Sat, 12 Sep 2020 07:47:38 +0000 (09:47 +0200)]
tests: support for 16T is needed

Likely 32bit machines can't pass here.

4 years agotests: change skip to die for upstream crash
Zdenek Kabelac [Sat, 12 Sep 2020 07:46:43 +0000 (09:46 +0200)]
tests: change skip to die for upstream crash

So the failing test is not lost from sight.

4 years agotests: skip test on failing kernel
Zdenek Kabelac [Sat, 12 Sep 2020 07:46:21 +0000 (09:46 +0200)]
tests: skip test on failing kernel

4 years agocov: drop checking for EWOULDBLOCK
Zdenek Kabelac [Sat, 12 Sep 2020 11:15:12 +0000 (13:15 +0200)]
cov: drop checking for EWOULDBLOCK

Reduce cov warning and remove this really ancient define
as lvm2 was never compilable on such platform.

4 years agocov: check sscanf result
Zdenek Kabelac [Sat, 12 Sep 2020 11:03:07 +0000 (13:03 +0200)]
cov: check sscanf result

4 years agocov: drop model for origin_from_cow
Zdenek Kabelac [Sat, 12 Sep 2020 07:45:52 +0000 (09:45 +0200)]
cov: drop model for origin_from_cow

4 years agolvconvert: check if LV has cow type
Zdenek Kabelac [Sat, 12 Sep 2020 10:08:15 +0000 (12:08 +0200)]
lvconvert: check if LV has cow type

Cow may not be a COW type, the return value of origin_from_cow(cow) may be NULL.

Reported-by: Wu Guanghao <wuguanghao3@huawei.com>
Reported-by: Zhiqiang Liu <liuzhiqiang26@huawei.com>
4 years agorevert "lvconvert: check if LV has snapshot type"
Zdenek Kabelac [Sat, 12 Sep 2020 10:07:12 +0000 (12:07 +0200)]
revert "lvconvert: check if LV has snapshot type"

This reverts commit 7db124774a2b79105e7c9c6b42df88219fc070ca.
Actually we need to check for lv_is_cow().

4 years agotests: check for boundary allocation sizes
Zdenek Kabelac [Fri, 11 Sep 2020 19:15:42 +0000 (21:15 +0200)]
tests: check for boundary allocation sizes

4 years agotests: reduce disk space usage by pvck-dump
Zdenek Kabelac [Fri, 11 Sep 2020 12:44:05 +0000 (14:44 +0200)]
tests: reduce disk space usage by pvck-dump

Lower disk usage for 'dd'.

4 years agotests: lower at_least to version 1.19
Zdenek Kabelac [Fri, 11 Sep 2020 11:35:30 +0000 (13:35 +0200)]
tests: lower at_least to version 1.19

With this version already can be seen different metadata usage on
kernel side, so lower the target version.

4 years agotests: have_cache function checks for cache-pool
Zdenek Kabelac [Fri, 11 Sep 2020 09:59:55 +0000 (11:59 +0200)]
tests: have_cache function checks for cache-pool

Check for cache-pool segment as plain cache can match writecache.

4 years agoalloc: improve estimation of sufficient_pes_free
Zdenek Kabelac [Fri, 11 Sep 2020 18:57:20 +0000 (20:57 +0200)]
alloc: improve estimation of sufficient_pes_free

Metadata size was calculated correctly only for raids.

Fixes problem for crash during lvcreate when thin-pool was created
on a VG where remaining free space had the size to only fit a single
metadata LV and not also its _pmspare.

Lvcreate crashed with this assert message:

lvcreate: metadata/pv_map.c:198: consume_pv_area: Assertion `to_go <= pva->count' failed.
Aborted (core dumped)

TODO: there is probably to large overload of several alloc_handle
variables.

Reported-by: Wu Guanghao<wuguanghao3@huawei.com>
Reported-by: Zhiqiang Liu <liuzhiqiang26@huawei.com>
4 years agolvconvert: check if LV has snapshot type
Wu Guanghao [Fri, 11 Sep 2020 09:42:16 +0000 (17:42 +0800)]
lvconvert: check if LV has snapshot type

Cow may not be a snapshot type, the return value of origin_from_cow(cow) may be NULL

Signed-off-by: Wu Guanghao <wuguanghao3@huawei.com>
Signed-off-by: Zhiqiang Liu <liuzhiqiang26@huawei.com>
4 years agolvconvert_poll: ensure LV has snapshot type
Wu Guanghao [Fri, 11 Sep 2020 09:41:01 +0000 (17:41 +0800)]
lvconvert_poll: ensure LV has snapshot type

LV may not be a snapshot type, the return value of find_snapshot(lv) may be NULL.
Here, we will call stack if LV is not a snapshot type.

Signed-off-by: Wu Guanghao <wuguanghao3@huawei.com>
Signed-off-by: Zhiqiang Liu <liuzhiqiang26@huawei.com>
4 years agopvmove: check return value of top_level_lv_name()
Wu Guanghao [Fri, 11 Sep 2020 09:40:09 +0000 (17:40 +0800)]
pvmove: check return value of top_level_lv_name()

The return value of top_level_lv_name() may be NULL, so we should
check return value of top_level_lv_name before calling
strcmp(lv->name, top_level_lv_name(vg, lv_name)).

Signed-off-by: Wu Guanghao <wuguanghao3@huawei.com>
Signed-off-by: Zhiqiang Liu <liuzhiqiang26@huawei.com>
4 years agogitignore: ignore all cscope generated files
Zhao Heming [Thu, 10 Sep 2020 15:38:07 +0000 (23:38 +0800)]
gitignore: ignore all cscope generated files

When using cscope to read code, it will generate below 3 files for speedup
cross-refer: cscope.files, cscope.in.out, cscope.po.out

The .gitignore only contains "/cscope.out". It a little bit messy when
executing 'git status', and other git commands.
This patch add all cscope generated files in .gitignore.

Signed-off-by: Zhao Heming <heming.zhao@suse.com>
4 years agothin: compensate metadata size by extra percent
Zdenek Kabelac [Fri, 11 Sep 2020 11:31:55 +0000 (13:31 +0200)]
thin: compensate metadata size by extra percent

When using --use-policy for automatic extension of thin-pool,
the extension of thin-pool's metadata itself can actually take
some extra space.
Since I'm not aware of exact compensation formula, add just
1% extra to calculated amount and hope it fits.

Wanted target is to always have usable thin-pool that fits
bellow pool_metadata_min_threshold().

4 years agolv_manip: even better rounding
Zdenek Kabelac [Fri, 11 Sep 2020 11:31:29 +0000 (13:31 +0200)]
lv_manip: even better rounding

4 years agotests: aux collects stack trace when stuck is assumed
Zdenek Kabelac [Thu, 10 Sep 2020 15:05:02 +0000 (17:05 +0200)]
tests: aux collects stack trace when stuck is assumed

Automatically collect traces in this case.

4 years agotests: adding "" around DM_DEV_DIR
Zdenek Kabelac [Thu, 10 Sep 2020 14:47:52 +0000 (16:47 +0200)]
tests: adding "" around DM_DEV_DIR

4 years agotests: disable dbustest with valgrind testing
Zdenek Kabelac [Thu, 10 Sep 2020 14:39:07 +0000 (16:39 +0200)]
tests: disable dbustest with valgrind testing

4 years agocleanup: comment typo
Zdenek Kabelac [Thu, 10 Sep 2020 20:38:13 +0000 (22:38 +0200)]
cleanup: comment typo

4 years agodebug: drop debug trace from regular path
Zdenek Kabelac [Thu, 10 Sep 2020 21:33:46 +0000 (23:33 +0200)]
debug: drop debug trace from regular path

Since we query on regular code these:
  lv_raid_has_integrity()
  lv_has_integrity_recalculate_metadata()
without prior checking for lv_is_raid() - these 'return 0' should
not use <stacktrace> as they are expected.

4 years agolv_manip: relocate check to proper function
Zdenek Kabelac [Thu, 10 Sep 2020 15:39:43 +0000 (17:39 +0200)]
lv_manip: relocate check to proper function

4 years agolvextend: improve percentage estimation
Zdenek Kabelac [Thu, 10 Sep 2020 20:38:34 +0000 (22:38 +0200)]
lvextend: improve percentage estimation

Correcting rounding rules for percentage evaluation.

Validate supported range of percentage.
(although ranges are already validated earlier on code path)

4 years agoconfigure: compile with vdo and writecache by default
Zdenek Kabelac [Wed, 9 Sep 2020 13:56:06 +0000 (15:56 +0200)]
configure: compile with vdo and writecache by default

Enable compilation of vdo and writecache support as internaly
supported segment types by default.

For disabling use:

--with-vdo=none
--with-writcache=none

4 years agointegrity: fix segfault reporting integrity for other lvs
David Teigland [Wed, 9 Sep 2020 15:22:07 +0000 (10:22 -0500)]
integrity: fix segfault reporting integrity for other lvs

4 years agoman: correcting vdo issues
Zdenek Kabelac [Wed, 9 Sep 2020 13:14:22 +0000 (15:14 +0200)]
man: correcting vdo issues

Fixing reported bugs within provided examples - so examples
can be used via cut&paste.

4 years agoman: correctly use configured directories
Zdenek Kabelac [Wed, 9 Sep 2020 11:22:37 +0000 (13:22 +0200)]
man: correctly use configured directories

4 years agoRevert "raid: add _rimage and _rmeta as origin_only"
Zdenek Kabelac [Tue, 8 Sep 2020 22:58:52 +0000 (00:58 +0200)]
Revert "raid: add _rimage and _rmeta as origin_only"

This reverts commit 3388e194892e9ab7a36b72e512796c2d218dae6e.
More thinking needed.

4 years agoRevert "raid: do not enforce flushing of raids when it is not required"
Zdenek Kabelac [Tue, 8 Sep 2020 22:58:32 +0000 (00:58 +0200)]
Revert "raid: do not enforce flushing of raids when it is not required"

This reverts commit ce5ea07411d57a2bd276d8d29e46ce30b957f657.
More thinking needed.

4 years agotests: tune extend test
Zdenek Kabelac [Mon, 7 Sep 2020 22:22:16 +0000 (00:22 +0200)]
tests: tune extend test

For proper checking of extension progress require version 1.15

It looks with older versoin extension happens during very slow
resume within lvm command - although speed is still somewhat slow
with latest version.

4 years agotests: check for writecache being compiled in
Zdenek Kabelac [Fri, 4 Sep 2020 19:08:32 +0000 (21:08 +0200)]
tests: check for writecache being compiled in

4 years agotests: check for cache_version that supports v2
Zdenek Kabelac [Fri, 4 Sep 2020 18:35:55 +0000 (20:35 +0200)]
tests: check for cache_version that supports v2

4 years agoraid: do not enforce flushing of raids when it is not required
Zdenek Kabelac [Tue, 8 Sep 2020 19:09:06 +0000 (21:09 +0200)]
raid: do not enforce flushing of raids when it is not required

This is probably somewhat experimantal patch - but when i.e. raid device
is just extend, there should not be a technical need for flush,
unless the target would stricly need it.  It should allow faster
processing of lvm command not being blocked by possibly longer flush.

4 years agoraid: add _rimage and _rmeta as origin_only
Zdenek Kabelac [Tue, 8 Sep 2020 19:06:23 +0000 (21:06 +0200)]
raid: add _rimage and _rmeta as origin_only

Since we do not support rimage & rmeta for snapshots - we can
avoid quering for -cow devices and add them as origin_only -
since their snapshots (-cow) could have never existed.
This redumes several ioctl operation during table preloading.

4 years agolv_manip: add synchronization points
Zdenek Kabelac [Tue, 8 Sep 2020 13:20:50 +0000 (15:20 +0200)]
lv_manip: add synchronization points

4 years agolvconvert: flip return value of _raid_split_image_conversion
Zdenek Kabelac [Fri, 4 Sep 2020 18:55:02 +0000 (20:55 +0200)]
lvconvert: flip return value of _raid_split_image_conversion

Use '0' for error and '1' as success.
Also drop INTERNAL_ERROR from path - as this error
is ATM used for invalid devices.
(i.e. test lvconvert-raid1-split-trackchanges.sh)

4 years agotests: fix pvck repair in hints.sh
David Teigland [Fri, 4 Sep 2020 16:22:42 +0000 (11:22 -0500)]
tests: fix pvck repair in hints.sh

4 years agotests: use delayzero_dev
Zdenek Kabelac [Fri, 4 Sep 2020 16:05:01 +0000 (18:05 +0200)]
tests: use delayzero_dev

Speed-up a bit the first synchronization with just 50ms write delay,
but later set also delay on read to slowdown lvextend.

FIXME: there are still things to look at:

0 229376 raid raid1 2 AA 229376/229376 idle 0 0
0 229376 raid raid1 2 AA 0/229376 frozen 0 0 -
0 262144 raid raid1 2 AA 229376/262144 repair 0 0 -
0 262144 raid raid1 2 AA 229376/262144 repair 0 0 -
0 262144 raid raid1 2 AA 245888/262144 repair 0 0 -

4 years agotests: add aux delayzero_dev support
Zdenek Kabelac [Fri, 4 Sep 2020 11:50:05 +0000 (13:50 +0200)]
tests: add aux delayzero_dev  support

Just like we have 'writeerror_dev' supporting creation of device
which 'readable' segment and segments where write will fail we
have now support for delay zero mappings.

This is useful if we want to 'fake' large writing areas where we do
not really care about the actual 'disk' content - since we test
operation logic and it doesn't matter we read and write zeroes.
With combination with 'delay' target we can create specific mappings
and avoid using large memory areas of ramdisk.

4 years agowritecache: allow pvmove on origin
David Teigland [Wed, 2 Sep 2020 19:45:52 +0000 (14:45 -0500)]
writecache: allow pvmove on origin

The removed check didn't actually prevent pvmoving the origin,
which was possible by naming the wcorig lv, or naming no lv.

4 years agopvck: repair should clear hints
David Teigland [Wed, 2 Sep 2020 19:21:17 +0000 (14:21 -0500)]
pvck: repair should clear hints

repairing a pv can cause the hint file to become incorrect

4 years agohints: remove warning when clearing hint file
David Teigland [Wed, 2 Sep 2020 19:06:46 +0000 (14:06 -0500)]
hints: remove warning when clearing hint file

When the hint file cannot be accessed, silently
ignore hints, like other instances do.

4 years agointegrity: improve lv type checks
David Teigland [Wed, 2 Sep 2020 17:40:45 +0000 (12:40 -0500)]
integrity: improve lv type checks

4 years agointegrity: fix segfault for lv with no seg
David Teigland [Wed, 2 Sep 2020 14:15:58 +0000 (09:15 -0500)]
integrity: fix segfault for lv with no seg

in lv_raid_has_integrity

4 years agotests: add new integrity reporting fields
David Teigland [Tue, 1 Sep 2020 22:07:36 +0000 (17:07 -0500)]
tests: add new integrity reporting fields

4 years agointegrity: report mismatches
David Teigland [Tue, 1 Sep 2020 21:15:31 +0000 (16:15 -0500)]
integrity: report mismatches

with lvs -o integritymismatches

reported for integrity images, which may report
different values

4 years agointegrity: report raidintegritymode randintegrityblocksize
David Teigland [Tue, 1 Sep 2020 19:39:34 +0000 (14:39 -0500)]
integrity: report raidintegritymode randintegrityblocksize

reported for the raid lv and the integrity images

4 years agointegrity: always default to journal mode
David Teigland [Tue, 1 Sep 2020 17:53:00 +0000 (12:53 -0500)]
integrity: always default to journal mode

lvconvert was defaulting to bitmap mode,
and lvcreate was defaulting to journal mode.

4 years agotests: skip this test for current 5.8 5.9 kernels
Zdenek Kabelac [Tue, 1 Sep 2020 21:14:47 +0000 (23:14 +0200)]
tests: skip this test for current 5.8 5.9 kernels

Kernel is hitting not yet fixed kernel bug.
Skip the test to avoid killing testing machine.

4 years agotests: slightly faster
Zdenek Kabelac [Tue, 1 Sep 2020 21:05:11 +0000 (23:05 +0200)]
tests: slightly faster

Use lvm shell to agregrate lots of lvm commands
Reduce initial zeroing.

4 years agotests: set skipping autoactivation
Zdenek Kabelac [Tue, 1 Sep 2020 20:45:17 +0000 (22:45 +0200)]
tests: set skipping autoactivation

On test system with 'default' filter  (aka accept all) test
after enabling device can suffer from automatic system
activation - so for created LVs setup skipping this automatic
activation. This should prevent getting LVs into table
with pvscan service.

4 years agotests: skip test when gcore cannot catch securetest
Zdenek Kabelac [Tue, 1 Sep 2020 19:39:47 +0000 (21:39 +0200)]
tests: skip test when gcore cannot catch securetest

4 years agogcc: avoid shadowing of dev_name and pvs
Zdenek Kabelac [Tue, 1 Sep 2020 19:17:14 +0000 (21:17 +0200)]
gcc: avoid shadowing of dev_name and pvs

Since we declare dev_name in lib/device/device.h
and pvs in commands.h
rename local dev_name to device_name
and pvs to pvs_list to prevent shadowing warning.

m

4 years agogcc: hide warn about possible uninitialized use of dev_ret
Zdenek Kabelac [Tue, 1 Sep 2020 18:46:18 +0000 (20:46 +0200)]
gcc: hide warn about possible uninitialized use of dev_ret

Older gcc reports this fp problem.

4 years agopvck: add simple check for fwrite
Zdenek Kabelac [Tue, 1 Sep 2020 19:14:39 +0000 (21:14 +0200)]
pvck: add simple check for fwrite

Add at least very light check for result code of fwrite().

4 years agopvck: use array of bytes
Zdenek Kabelac [Tue, 1 Sep 2020 19:08:12 +0000 (21:08 +0200)]
pvck: use array of bytes

Fix typo in use array of pointers instead of array of bytes.
This fixes 'break strict-aliasing rules' warning printed with older gcc.

4 years agocleanup: better expressing passing key arg to _hash
Zdenek Kabelac [Sat, 29 Aug 2020 20:37:17 +0000 (22:37 +0200)]
cleanup: better expressing passing key arg to _hash

4 years agocleanup: add spaces between literals
Zdenek Kabelac [Sat, 29 Aug 2020 19:53:31 +0000 (21:53 +0200)]
cleanup: add spaces between literals

4 years agocleanup: user force_t enums instead of ints
Zdenek Kabelac [Sat, 29 Aug 2020 19:53:05 +0000 (21:53 +0200)]
cleanup: user force_t enums instead of ints

4 years agocleanup: matching declaration order
Zdenek Kabelac [Sat, 29 Aug 2020 19:37:39 +0000 (21:37 +0200)]
cleanup: matching declaration order

Cosmetic

4 years agocov: avoid duplicated assign
Zdenek Kabelac [Sat, 29 Aug 2020 18:40:20 +0000 (20:40 +0200)]
cov: avoid duplicated assign

4 years agocov: always initialized values
Zdenek Kabelac [Sat, 29 Aug 2020 18:27:32 +0000 (20:27 +0200)]
cov: always initialized values

Make sure values are initialized for all possible paths.

4 years agocov: explicitely ignore function result
Zdenek Kabelac [Tue, 1 Sep 2020 15:53:36 +0000 (17:53 +0200)]
cov: explicitely ignore function result

4 years agocov: always sure we end with '0'
Zdenek Kabelac [Sat, 29 Aug 2020 17:56:37 +0000 (19:56 +0200)]
cov: always sure we end with '0'

Use easier dm_strncpy().

4 years agogcc: keep using unsigned type
Zdenek Kabelac [Sat, 29 Aug 2020 20:24:34 +0000 (22:24 +0200)]
gcc: keep using unsigned type

This page took 0.064388 seconds and 5 git commands to generate.