GH-118943: Fix another race condition when generating jit_stencils.h - #120690
Conversation
…ils.h Another process might have already moved jit_stencils.h.new
…7nn.rst Co-authored-by: Kirill Podoprigora <kirill.bast9@mail.ru>
|
We run regen-all first, then all. Are you interested in full logs? |
|
Ah, I think I see the issue! I think we should remove |
|
This was not merged in time for 3.13.0b3. How can I move this forward? |
|
@brandtbucher Could you please merge this? Is this waiting for something else? |
|
@Yhg1s Hey Thomas. We have carried this patch in Fedora since 3.13.0b2. Is there any chance we could get this merged to at least 3.13.0rc2? |
I've pinged Brandt in the private coredev chat on Discord to get his attention. |
brandtbucher
left a comment
There was a problem hiding this comment.
Sorry for letting this slip! Looks good to me. One thing, though: can you also remove regen-jit from the regen-all target in Makefile.pre.in?
@Yhg1s I think this is reasonable to backport to the next RC. It's just minor fixes for race conditions in the JIT build (thanks @hroncok for your work to make it available on Fedora).
|
Sure, this is fine to get in 3.13.0rc2. |
I'm happy to do that in a separate PR. |
|
See #122602 |
|
Thanks for the second approval @brandtbucher. Could you please merge this as well? |
|
Thanks @hroncok for the PR, and @brandtbucher for merging it 🌮🎉.. I'm working now to backport this PR to: 3.13. |
|
GH-122709 is a backport of this pull request to the 3.13 branch. |
…0690) Co-authored-by: Kirill Podoprigora <kirill.bast9@mail.ru>
…0690) Co-authored-by: Kirill Podoprigora <kirill.bast9@mail.ru>

Another process might have already moved jit_stencils.h.new
jit.cmay be built with an incompletejit_stencils.h#118943