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

Re: [PATCH v2] transport-helper: enforce atomic in push_refs_with_push

From
Jonathan Nieder <jrnieder@gmail.com>
Date
Jul 19, 2019, 01:31 UTC
Message-ID
<20190719013128.GB251102@google.com>
In-Reply-To
<20190718152234.GI20404@szeder.dev>
SZEDER Gábor wrote:
Show 12 quoted lines
> I expected that this will eventually happen after Travis CI's default
> Linux image recently changed from Ubuntu 14.04 to 16.04; explanation
> in the commit message below.
>
> With that patch issues like this could be caught earlier, while they
> are only in 'pu' but not yet in 'next'.  But do we really want to do
> that, is that the right tradeoff?  Dunno...  Adding a dedicated CI job
> just to check that there are no 'for' loop initial declarations seems
> kind of excessive, even if it only builds but doesn't run the test
> suite.  And I don't know whether there are any other undesired ("too
> new") constructs that GCC 4.8 would catch but later compilers quietly
> accept.

This makes sense to me. Not really for the 'for' loop declaration aspect: for that, I'd want some more specialized tool that allows turning on such a check specifically. But more because Ubuntu trusty is still a platform that some people use (though hopefully not for long), so it's helpful as a representative old platform to see if we break the build on it.

[...]
> [2] On Travis CI 'make test' alone would take about 9 minutes in this
>     new job (without running httpd, Subversion, and P4 tests).  For
>     comparison, starting the job and building Git with GCC 4.8 takes
>     only about 2 minutes.

Nice. In an ideal world there would be some kind of "make fasttest" that runs some fast subset of tests. But this seems pretty safe.

Show 5 quoted lines
> Signed-off-by: SZEDER Gábor <szeder.dev@gmail.com>
> ---
>  .travis.yml               |  4 ++++
>  ci/run-build-and-tests.sh | 17 +++++++++++++----
>  2 files changed, 17 insertions(+), 4 deletions(-)
Thanks.
[...]
Show 30 quoted lines
> --- a/ci/run-build-and-tests.sh
> +++ b/ci/run-build-and-tests.sh
> @@ -11,9 +11,9 @@ windows*) cmd //c mklink //j t\\.prove "$(cygpath -aw "$cache_dir/.prove")";;
>  esac
>  
>  make
> -make test
> -if test "$jobname" = "linux-gcc"
> -then
> +case "$jobname" in
> +linux-gcc)
> +	make test
>  	export GIT_TEST_SPLIT_INDEX=yes
>  	export GIT_TEST_FULL_IN_PACK_ARRAY=true
>  	export GIT_TEST_OE_SIZE=10
> @@ -21,7 +21,16 @@ then
>  	export GIT_TEST_COMMIT_GRAPH=1
>  	export GIT_TEST_MULTI_PACK_INDEX=1
>  	make test
> -fi
> +	;;
> +linux-gcc-4.8)
> +	# Don't run the tests; we only care about whether Git can be
> +	# built with GCC 4.8, as it errors out on some undesired (C99)
> +	# constructs that newer compilers seem to quietly accept.
> +	;;
> +*)
> +	make test
> +	;;
> +esac
Does what it says on the tin.
Reviewed-by: Jonathan Nieder <jrnieder@gmail.com>
Previous: Eric SunshineNext: Carlo Arenas
Message 41 of 45 in “transport-helper: enforce atomic in push_refs_with_push”
  1. transport-helper: enforce atomic in push_refs_with_pushEmily Shaffer, Jul 2, 2019
  2. Johannes SchindelinJul 2, 2019
  3. Junio C HamanoJul 2, 2019
  4. Emily ShafferJul 3, 2019
  5. Emily ShafferJul 3, 2019
  6. Johannes SchindelinJul 3, 2019
  7. Emily ShafferJul 3, 2019
  8. Johannes SchindelinJul 4, 2019
  9. Emily ShafferJul 9, 2019
  10. Junio C HamanoJul 2, 2019
  11. Junio C HamanoJul 2, 2019
  12. Emily ShafferJul 3, 2019
  13. Junio C HamanoJul 2, 2019
  14. Emily ShafferJul 3, 2019
  15. SZEDER GáborJul 3, 2019
  16. Junio C HamanoJul 3, 2019
  17. Emily ShafferJul 3, 2019
  18. transport-helper: enforce atomic in push_refs_with_pushEmily Shaffer, Jul 9, 2019
  19. Junio C HamanoJul 10, 2019
  20. Junio C HamanoJul 10, 2019
  21. Emily ShafferJul 11, 2019
  22. Emily ShafferJul 11, 2019
  23. Junio C HamanoJul 11, 2019
  24. transport-helper: enforce atomic in push_refs_with_pushEmily Shaffer, Jul 11, 2019
  25. Junio C HamanoJul 12, 2019
  26. Carlo ArenasJul 16, 2019
  27. Junio C HamanoJul 16, 2019
  28. CodingGuidelines: spell out post-C89 rulesJunio C Hamano, Jul 16, 2019
  29. Jonathan NiederJul 17, 2019
  30. Junio C HamanoJul 17, 2019
  31. Jonathan NiederJul 19, 2019
  32. Bryan TurnerJul 17, 2019
  33. Junio C HamanoJul 17, 2019
  34. Carlo ArenasJul 16, 2019
  35. transport-helper: avoid var decl in for () loop controlJunio C Hamano, Jul 16, 2019
  36. Jonathan NiederJul 17, 2019
  37. SZEDER GáborJul 18, 2019
  38. Junio C HamanoJul 18, 2019
  39. SZEDER GáborJul 18, 2019
  40. Eric SunshineJul 18, 2019
  41. Jonathan NiederJul 19, 2019
  42. Carlo ArenasJul 19, 2019
  43. Junio C HamanoJul 19, 2019
  44. SZEDER GáborJul 27, 2019
  45. Junio C HamanoJul 27, 2019

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.