Skip to contentSkip to conversation

LTP Release plans

LTP Cyril Hrubis 5 participants 12 messages Dec 13, 2019 – Jan 16, 2020

lore
Share & tools
lore
#0 lore raw

Hi!
As usuall we are supposed to produce a release in January. I'm on vacation starting next week till the new year, so we should start to prepare for the release once I'm back on the 2. January 2020.

Ideally we should release around 20. January, which leaves us at least two weeks for the last minute fixes and testing but feel free to start with the pre-release testing sooner, that will help us to find bugs sooner and we will have more time for fixes.

As usuall if there are any patches that should included in the release, just reply to this mail and I will try to review them once I'm back.

And lastly but not least Merry Christmass && Happy New Year to these who celebrate them.

Signature · 2 lines
Cyril Hrubis
[email protected]
#1 lore raw
Quoted 2 lines · Cyril Hrubis

As usuall if there are any patches that should included in the release, just reply to this mail and I will try to review them once I'm back.

I'd like to include the fix for SCTP test_1_to_1_initmsg_connect (mostly a simple API port with minor additions).

I also need some feedback for my proposal to change the test scenario in fallocate05 (from anyone). I'll need a few days to submit a new patch so the sooner the better. I'd really like to get the final fallocate05 patch into the new release.

Signature · 7 lines
Martin Doucha   [email protected]
QA Engineer for Software Maintenance
SUSE LINUX, s.r.o.
CORSO IIa
Krizikova 148/34
186 00 Prague 8
Czech Republic
#2 lore raw

Hi!

Quoted 5 lines
Quoted 2 lines

As usuall if there are any patches that should included in the release, just reply to this mail and I will try to review them once I'm back.

I'd like to include the fix for SCTP test_1_to_1_initmsg_connect (mostly a simple API port with minor additions).

This has been handled already.

Quoted 4 lines

I also need some feedback for my proposal to change the test scenario in fallocate05 (from anyone). I'll need a few days to submit a new patch so the sooner the better. I'd really like to get the final fallocate05 patch into the new release.

Acked by Jan, I will have a look ASAP.

Signature · 2 lines
Cyril Hrubis
[email protected]
#3 lore raw
In reply to Cyril Hrubis

Hi Cyril

Quoted 5 lines

As usuall if there are any patches that should included in the release, just reply to this mail and I will try to review them once I'm back.

And lastly but not least Merry Christmass && Happy New Year to these who celebrate them.

I think capset/capget cleanup should be include this release. Also, I plan to clean up cap_bound filecap testcases. If I can figure out cap_ambient intention, I will write cases to test it.

Kind Regrads
Yang Xu

#4 lore raw

Hi!

Quoted 8 lines
Quoted 5 lines

As usuall if there are any patches that should included in the release, just reply to this mail and I will try to review them once I'm back.

And lastly but not least Merry Christmass && Happy New Year to these who celebrate them.

I think capset/capget cleanup should be include this release. Also, I plan to clean up cap_bound filecap testcases. If I can figure out cap_ambient intention, I will write cases to test it.

I will have a look at the capset/capget patches if they are clean and safe enough for the release ASAP.

Signature · 2 lines
Cyril Hrubis
[email protected]
#5 lore raw
In reply to Cyril Hrubis

Hi!
First of all Happy New Year!

I'm back at patch review and we are getting close to a pre-release testing freeze.

As I stated before I will have a look at he fallocate05 test and the capset/capget patches. Is there anything else that should be part of the release?

Looking at the patch queue we should probably include:

tst_dev_bytes_written: parsing /proc/diskstats instead of sys file syscalls/userfaultfd01: add hint about unprivileged_userfaultfd

Anything else I should look at?

Signature · 2 lines
Cyril Hrubis
[email protected]
#6 lore raw

Hi Cyril

Quoted 10 lines

Hi!
First of all Happy New Year!

I'm back at patch review and we are getting close to a pre-release testing freeze.

As I stated before I will have a look at he fallocate05 test and the capset/capget patches. Is there anything else that should be part of the release?

Quoted 6 lines

Looking at the patch queue we should probably include:

tst_dev_bytes_written: parsing /proc/diskstats instead of sys file syscalls/userfaultfd01: add hint about unprivileged_userfaultfd

Anything else I should look at?

Yes. I also sent a pipe test[1]to ltp(it has undefined F_GETPIPE_SZ error on old kernel [2]and I will fix it in v2), as below:
syscalls/pipe12: add new test for piep when write bytes > PIPE_BUF

Also, Cypresslin has a pull[3] about syscalls/statx05: add mkfs.ext4 package version check.

[1]https://patchwork.ozlabs.org/patch/1215360/ [2]https://travis-ci.org/xuyang0410/ltp/jobs/629597516?utm_medium=notification&utm_source=github_status [3]https://github.com/linux-test-project/ltp/pull/626

Best Regards
Yang Xu

Quoted 1 line
#7 lore raw
Quoted 19 lines · Yang Xu

Hi Cyril

Quoted 16 lines

Hi!
First of all Happy New Year!

I'm back at patch review and we are getting close to a pre-release testing freeze.

As I stated before I will have a look at he fallocate05 test and the capset/capget patches. Is there anything else that should be part of the release?

Looking at the patch queue we should probably include:

tst_dev_bytes_written: parsing /proc/diskstats instead of sys file syscalls/userfaultfd01: add hint about unprivileged_userfaultfd

Anything else I should look at?

Also. In December-end, pengfei sent a kconfig patch set to support CONFIG_X86_INTEL_UMIP|CONFIG_X86_UMIP=y. CONFIG_X86_INTEL_UMIP=y for umip kconfig before and v5.4 mainline kernel. CONFIG_X86_UMIP=y for umip kconfig after v5.5 mainline kernel. I think we can include this patchset before release if it is ok.

url:
http://lists.linux.it/pipermail/ltp/2019-December/014841.html

Quoted 18 lines

Yes. I also sent a pipe test[1]to ltp(it has undefined F_GETPIPE_SZ error on old kernel [2]and I will fix it in v2), as below:
syscalls/pipe12: add new test for piep when write bytes > PIPE_BUF

Also, Cypresslin has a pull[3] about syscalls/statx05: add mkfs.ext4 package version check.

[1]https://patchwork.ozlabs.org/patch/1215360/ [2]https://travis-ci.org/xuyang0410/ltp/jobs/629597516?utm_medium=notification&utm_source=github_status

[3]https://github.com/linux-test-project/ltp/pull/626

Best Regards
Yang Xu

Quoted 1 line
#8 lore raw
In reply to Yang Xu

Hi!

Quoted 3 lines

Yes. I also sent a pipe test[1]to ltp(it has undefined F_GETPIPE_SZ error on old kernel [2]and I will fix it in v2), as below:
syscalls/pipe12: add new test for piep when write bytes > PIPE_BUF

I will have a look.

Quoted 2 lines

Also, Cypresslin has a pull[3] about syscalls/statx05: add mkfs.ext4 package version check.

The patch works only on debian, and I do not see a v2 there, so this has to wait.

Signature · 2 lines
Cyril Hrubis
[email protected]
#9 lore raw
In reply to Cyril Hrubis

Hi Cyril,

Quoted 1 line

Anything else I should look at?

I'd like to have TI-RPC related patch-set

* RFC,1/1] rpc-tirpc: Remove authdes related tests
https://patchwork.ozlabs.org/patch/1216506/
Acked by Li, waiting for Steve (libtirpc maintainer) to be acked.

And if you have time then this one (not that important, fixing
https://github.com/linux-test-project/ltp/pull/628
and adding libntirpc support):
* TI-RPC: enhancements (pkg-config, libntirpc, travis)
https://patchwork.ozlabs.org/project/ltp/list/?series=151849
(first 2 commits are relevant, the others are travis)

Thanks!

Kind regards,
Petr

#10 lore raw
In reply to Cyril Hrubis

Hi!
I've just finished patchwork queue review and I think that we are fine and that we should freeze the git for anything but bugfixes at this point.

I will do a bit more pre-release testing, however the results so far looks good to me.

This is also last call to test the latest git head, if you haven't done so plese do so now.

Ideally I would like to stick to the plan and do the release on Monday
20. unless there is a major breakage.

Signature · 2 lines
Cyril Hrubis
[email protected]
#11 lore raw

Hi,

Quoted 8 lines · Cyril Hrubis

Hi!
I've just finished patchwork queue review and I think that we are fine and that we should freeze the git for anything but bugfixes at this point.

I will do a bit more pre-release testing, however the results so far looks good to me.

FYI:

I did a round of testing against RHEL6/7/8 and upstream kernel, the test result looks good except only below tiny issues:

- RHEL6 x86_64 (it is a kernel bug that in old distro) sysctl02 1 TFAIL: /proc/sys/fs/file-max overflows and is set to 0 sysctl02 2 TINFO: trying to set fs.file-max=18446744073709551615 sysctl02 2 TFAIL: /proc/sys/fs/file-max overflows and is set to 18446744073709551615
sysctl02 3 TINFO: trying to set fs.file-max=9223372036854775808 sysctl02 3 TFAIL: /proc/sys/fs/file-max overflows and is set to 9223372036854775808

fallocate04, fallocate05, fsetxattr01, fdatasync03, fgetxattr01 have the same issue as:
tst_test.c:1278: INFO: Testing on vfat
tst_mkfs.c:90: INFO: Formatting /dev/loop0 with vfat opts='' extra opts='' mkfs.vfat: unable to open /dev/loop0
tst_mkfs.c:101: BROK: mkfs.vfat:1: tst_test.c failed with 831

- RHEL7.8 x86_64
copy_file_range02 failed as a known kernel bug

- RHEL7.8 ppc64le (I have confirmed the latest rhel7 kernel has fixed this bug)
timer_delete01.c:47: FAIL: Aborting test - timer_create(CLOCK_BOOTTIME_ALARM) failed: ??? (524) timer_delete01.c:36: INFO: Testing CLOCK_REALTIME_ALARM timer_delete01.c:47: FAIL: Aborting test - timer_create(CLOCK_REALTIME_ALARM) failed: ??? (524) timer_delete01.c:36: INFO: Testing CLOCK_TAI timer_delete01.c:54: PASS: Timer deleted successfully!

- Upstream kernel-5.5.0-rc5 x86_64/aarch64
system hang at fs_fill, it more like the unstable kernel issue but not LTP

Signature · 5 lines
Regards,
Li Wang
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linux.it/pipermail/ltp/attachments/20200116/650951c5/attachment.htm>