Re: [PATCH] t/perf/run: fix bin-wrappers computation

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Tue, Sep 21, 2021 at 03:46:12PM +0000, Derrick Stolee via GitGitGadget wrote:
> This is critical to successfully computing performance of commands that
> execute subcommands. The bin-wrappers ensure that the --exec-path is set
> correctly.

Just sanity-checking everything you said: with this bug, we'll set
mydir_abs_wrappers to "/bin-wrappers", then realize that directory
doesn't exist, and fall back to GIT_TEST_INSTALLED=$mydir_abs. Putting
'set -x' at the top of t/perf/run, we can see the relevant section:

  + mydir=build/73cd7d9420bb7d75207e8149521db375c789a81c
  + cd build/73cd7d9420bb7d75207e8149521db375c789a81c
  + pwd
  + mydir_abs=/home/ttaylorr/src/git/t/perf/build/73cd7d9420bb7d75207e8149521db375c789a81c
  + mydir_abs_wrappers=/bin-wrappers
  + test -d /bin-wrappers
  + GIT_TEST_INSTALLED=/home/ttaylorr/src/git/t/perf/build/73cd7d9420bb7d75207e8149521db375c789a81c
  + export GIT_TEST_INSTALLED

OK. But the real problem is in t/test-lib.sh where we read
"$GIT_TEST_INSTALLED". There we ask for the `--exec-path`, which appears
to be wrong, at least in my setup. Printing out the $GIT_EXEC_PATH and
$GIT_TEST_INSTALLED, I get:

  /home/ttaylorr/local/git/ds.sparse-checkout/libexec/git-core,
  /home/ttaylorr/src/git/t/perf/build/73cd7d9420bb7d75207e8149521db375c789a81c

where the former is the branch I happen to have checked out, and the
latter is the revision that I asked to run performance tests on via
t/perf/run.

So I think we'll run the right top-level Git command since the latter
path ends up first in our $PATH, but the exec path is definitely wrong.

> diff --git a/t/perf/run b/t/perf/run
> index d19dec258a2..55219aa4056 100755
> --- a/t/perf/run
> +++ b/t/perf/run
> @@ -74,7 +74,7 @@ set_git_test_installed () {
>  	mydir=$1
>
>  	mydir_abs=$(cd $mydir && pwd)
> -	mydir_abs_wrappers="$mydir_abs_wrappers/bin-wrappers"
> +	mydir_abs_wrappers="$mydir_abs/bin-wrappers"

This fix looks obviously right to me (and gives me the expected output
from above when I apply it locally).

Thanks,
Taylor



[Index of Archives]     [Linux Kernel Development]     [Gcc Help]     [IETF Annouce]     [DCCP]     [Netdev]     [Networking]     [Security]     [V4L]     [Bugtraq]     [Yosemite]     [MIPS Linux]     [ARM Linux]     [Linux Security]     [Linux RAID]     [Linux SCSI]     [Fedora Users]

  Powered by Linux