gh-131093: Enforce -std=c11 for gcc at configure time rather than make time#133608
gh-131093: Enforce -std=c11 for gcc at configure time rather than make time#133608CS-cwhite wants to merge 1 commit intopython:mainfrom
Conversation
|
The following commit authors need to sign the Contributor License Agreement: |
|
I previously signed the CLA on behalf of my organization, at the direction of the organization's legal department. Is there a way to tell CLA-bot to recognize that? |
|
cc @ambv re CLA |
AA-Turner
added
the
🔨 test-with-buildbots
|
🤖 New build scheduled with the buildbot fleet by @AA-Turner for commit b4fbd36 🤖 Results will be shown at: https://buildbot.python.org/all/#/grid?branch=refs%2Fpull%2F133608%2Fmerge If you want to schedule another build, you need to add the 🔨 test-with-buildbots label again. |
bedevere-bot
removed
the
🔨 test-with-buildbots
|
This is my first cpython PR, so I don't know --- do I have any action items from the build failures? E.g., Check if generated files are up to date appears to be checking C code, and this PR doesn't touch any C code. The failures don't appear to be related to this PR's changes, but I'm happy to work on fixes if they are! |
|
We had some teething issues with the bump to 3.15, so I've updated the branch to check the failures aren't related. In general, you may find the Python Developer's Guide useful. A |
I think redoing the commits with whatever email you signed with and then force-pushing should work. |
…an make time When compiling under gcc, configure.ac currently assumes -std=c11 rather than checking it. This causes configure to succeed but make to fail on old compilers. Fix this by adding a check for -std=c11 in configure.ac.
CS-cwhite
force-pushed
the
bugfix/SDX-12236-configure-std-c11
branch
from
f14b4f8 to
bc5c9ed
Compare
|
@ZeroIntensity Thanks! I checked, and the email is the same. However, the author name was different (it was "Chris White" in the commit but "Christopher White" on the CLA). I amended the commit to show the same name as the CLA in case that made a difference. |
|
I've asked internally for advice. A |
|
@CS-cwhite, the advice I've had is that you should click the button to sign the CLA again. Sorry for the delay and the trouble this has caused. There's some more background/colour in #120171 (comment) A |
| @@ -0,0 +1,3 @@ | |||
| On GCC-compatible compilers, check at configure time that ``-std=c11`` | |||
| works. This way you don't hit an "invalid option" error at make time on old | |||
There was a problem hiding this comment.
| works. This way you don't hit an "invalid option" error at make time on old | |
| works. This way you don't hit an "invalid option" error at build time on old |
|
@AA-Turner Unfortunately, corporate legal says I am not permitted to sign the individual CLA, and that the organizational CLA must control my contributions. |
|
I'll ask for advice. |
|
@AA-Turner Wondering if you've heard anything back. Thanks! |
|
Yes, sorry; the best person to help/contact is Łukasz Langa (@ambv). A |
|
@ambv Any thoughts? Thanks! |
|
@ambv @AA-Turner bump! Any word on getting the signed organizational CLA to cover this PR? Thanks for your help with this! |
|
From this webpage, the relevant email appears to be A |
|
I have emailed that address. Fingers crossed! |
|
@ambv Would you be willing to ask someone on the Legal mailing list if they received my email of 2025-06-17? I haven't heard back, but it's possible someone responded and the email got caught in a spam filter. Thanks for considering this request! |
|
I just sent another note to |
|
@ambv could you ask the PSF legal folks if they ever received my request for approval of this PR under the organizational CLA? I still haven't heard back, but I don't know if my messages to |
|
Any news? Thanks! |
|
According to the comment that Adam linked (#120171 (comment)), we basically cannot accept a CLA without an individual signature. If CrowdStrike has an issue with you doing that, is it possible for you to open a new PR (possibly under a new GitHub account) with an email completely dissociated from CrowdStrike? |
|
This PR is stale because it has been open for 30 days with no activity. |
6 participants
When compiling under gcc, configure.ac currently assumes
-std=c11rather than checking it. This causes configure to succeed but make to fail on old compilers. Fix this by adding a check for-std=c11in configure.ac.The changes to
configureare the result of runningTools/build/regen-configure.sh.Manual test on gcc 4.4.7:
Before:
After:
Fixes #131093.