git/list[1] front-page[2] threads[3] people[4] search[5] about
 

Re: [PATCH v2 2/2] ci: align job counts across CI providers

From
Tamir Duberstein <tamird@gmail.com>
Date
Sep 28, 2026, 10:16 UTC
Message-ID
<CAJ-ks9mWU2cFWLfioSSM_6Ct1ydtxn3i4Dh-L537M5EDiMW3vQ@mail.gmail.com>
In-Reply-To
<aroK7KcRabARSqd8@pks.im>
On Mon, Sep 28, 2026 at 2:36 AM Patrick Steinhardt <ps@pks.im> wrote:
Show 11 quoted lines
>
> On Fri, Sep 25, 2026 at 12:35:39PM -0400, Tamir Duberstein wrote:
> > GitHub Actions sets JOBS to ten regardless of runner size, while
> > GitLab CI uses the detected CPU count. Use the CPU count for Make and
> > prove on both providers, selecting JOBS after the operating system
> > is identified.
>
> Again, it should be noted here what the effect of this is. In other
> words, does GitHub slow down as a result? You already showed numbers
> during the discussion on v1 of this series, and these numbers should
> probably be included in this message, too.

Agreed, but in this case there was no reliable performance change across 10 runs; I could include that.

Show 14 quoted lines
>
> > Use nproc on Linux and NUMBER_OF_PROCESSORS on Windows. On macOS, use
> > sysctl to avoid requiring nproc before the dependency installer has run;
> > GitHub macOS images need not provide GNU coreutils.
>
> Huh... "need not" feels somewhat weird as phrasing. I guess it's rather
> "does not", and consequently we have to adapt? I think instead of
> describing what you do, I'd directly pinpoint what matters:
>
>   Note that we continue to use the same logic to detect the number of
>   processors on both Linux and Windows. But on macOS, we cannot continue
>   to use nproc(1) because the image used by GitHub does not provide that
>   tool. Use sysctl instead, which is available on both GitLab and
>   GitHub

Agreed. .

>
> Thanks!
>
> Patrick
Previous: Patrick SteinhardtNext: Patrick Steinhardt
Message 17 of 25 in “ci: reduce pressure from large test fixtures”
  1. 0/2 ci: reduce pressure from large test fixturesTamir Duberstein, Sep 23, 2026
  2. 1/2 t4205: compare huge output without diffTamir Duberstein, Sep 23, 2026
  3. Patrick SteinhardtSep 24, 2026
  4. Tamir DubersteinSep 24, 2026
  5. Jeff KingSep 25, 2026
  6. Junio C HamanoSep 25, 2026
  7. Patrick SteinhardtSep 28, 2026
  8. 2/2 ci: match Linux jobs to available CPUsTamir Duberstein, Sep 23, 2026
  9. Patrick SteinhardtSep 24, 2026
  10. Tamir DubersteinSep 25, 2026
  11. Tamir DubersteinSep 23, 2026
  12. 0/2 ci: use cmp and align job-count selectionTamir Duberstein, Sep 25, 2026
  13. 1/2 t4205: compare huge output without diffTamir Duberstein, Sep 25, 2026
  14. Patrick SteinhardtSep 28, 2026
  15. 2/2 ci: align job counts across CI providersTamir Duberstein, Sep 25, 2026
  16. Patrick SteinhardtSep 28, 2026
  17. Tamir DubersteinSep 28, 2026
  18. Patrick SteinhardtSep 28, 2026
  19. 0/2 ci: use cmp and align job-count selectionTamir Duberstein, Sep 30, 2026
  20. 1/2 t4205: compare huge output without diffTamir Duberstein, Sep 30, 2026
  21. 2/2 ci: use twice the CPU count on both providersTamir Duberstein, Sep 30, 2026
  22. Patrick SteinhardtSep 30, 2026
  23. Patrick SteinhardtSep 30, 2026
  24. Tamir DubersteinSep 30, 2026
  25. Junio C HamanoSep 30, 2026

Read the whole thread, see it on lore, or plain text.

$ cat FOOTERMessages come from the public archive at lore.kernel.org/git, fetched every hour. The front page is chosen and written each morning by an AI editor and can be wrong; the threads themselves are the record. About and API. For agents: an MCP server at https://gitlist.dev/mcp, and any thread, story or person page as Markdown by adding .md to its URL (or sending Accept: text/markdown). Details in /llms.txt.