Issue 1790 in webm: Seemingly wrong changes shown in https://build.webmproject.org/jenkins/view/libvpx%20nightly/job/libvpx__unit_tests-valgrind/1662/

381 views
Skip to first unread message

w… via monorail

unread,
Feb 10, 2023, 7:42:39 PM2/10/23
to webm-d...@webmproject.org
Status: Accepted
Owner: ----
Labels: Type-Task Pri-2 Jenkins-Nightly-Failure

New issue 1790 by w...@google.com: Seemingly wrong changes shown in https://build.webmproject.org/jenkins/view/libvpx%20nightly/job/libvpx__unit_tests-valgrind/1662/
https://bugs.chromium.org/p/webm/issues/detail?id=1790

The changes in https://build.webmproject.org/jenkins/view/libvpx%20nightly/job/libvpx__unit_tests-valgrind/1662/ are:

1. Update CHANGELOG (details)
2. Fix per frame qp for temporal layers (details)
3. Update AUTHORS .mailmap and version (details)
3. Fix unsigned integer overflow in sse computation (details)

This list seems wrong. James, do you know why?

The changes in the earlier libvpx__unit_tests-valgrind pages also seem wrong:
https://build.webmproject.org/jenkins/view/libvpx%20nightly/job/libvpx__unit_tests-valgrind/1661/
https://build.webmproject.org/jenkins/view/libvpx%20nightly/job/libvpx__unit_tests-valgrind/1660/
https://build.webmproject.org/jenkins/view/libvpx%20nightly/job/libvpx__unit_tests-valgrind/1659/

--
You received this message because:
1. The project was configured to send all issue notifications to this address

You may adjust your notification preferences at:
https://bugs.chromium.org/hosting/settings

w… via monorail

unread,
Feb 10, 2023, 7:57:58 PM2/10/23
to webm-d...@webmproject.org

w… via monorail

unread,
Feb 10, 2023, 7:58:16 PM2/10/23
to webm-d...@webmproject.org

jz… via monorail

unread,
Feb 10, 2023, 9:53:53 PM2/10/23
to webm-d...@webmproject.org


> 1. Update CHANGELOG (details)
> 2. Fix per frame qp for temporal layers (details)
> 3. Update AUTHORS .mailmap and version (details)
> 3. Fix unsigned integer overflow in sse computation (details)
>
> This list seems wrong. James, do you know why?

That's the right parent of the merge of v1.13.0:

$ git log --oneline -4 f903d99650eb3f7a294738259ece879e0528f752^2^2
d6eb9696a (tag: v1.13.0, origin/ugly-duckling) Fix unsigned integer overflow in sse computation
b5a2b3a92 (tag: v1.13.0-rc1) Update AUTHORS .mailmap and version
aa5b62236 Fix per frame qp for temporal layers
3f109f786 Update CHANGELOG


> I also noticed that libvpx__nightly_compile always reports "No changes". For example,

I don't have an answer for that yet.

jz… via monorail

unread,
Feb 10, 2023, 10:13:46 PM2/10/23
to webm-d...@webmproject.org


>> I also noticed that libvpx__nightly_compile always reports "No changes". For example,
>
> I don't have an answer for that yet.

Because it does a shallow clone. Since this is a nightly, this minor efficiency
isn't strictly necessary. I'll change it to do a full clone and we can check
the results.

jz… via monorail

unread,
Feb 13, 2023, 10:01:49 PM2/13/23
to webm-d...@webmproject.org
Updates:
Status: Done

Comment #5 on issue 1790 by jz...@google.com: Seemingly wrong changes shown in https://build.webmproject.org/jenkins/view/libvpx%20nightly/job/libvpx__unit_tests-valgrind/1662/
https://bugs.chromium.org/p/webm/issues/detail?id=1790#c5

Now that a full clone is being done, the changes included in the run are reported.
Reply all
Reply to author
Forward
0 new messages