Re: tb/plug-pack-bitmap-leaks (was: What's cooking in git.git (Oct 2021, #06; Mon, 25))

[Taylor Blau <me@ttaylorr.com>]

On Tue, Oct 26, 2021 at 02:13:44PM +0200, Ævar Arnfjörð Bjarmason wrote:
>
> On Mon, Oct 25 2021, Junio C Hamano wrote:
>
> > * tb/plug-pack-bitmap-leaks (2021-10-21) 9 commits
> >  - pack-bitmap.c: more aggressively free in free_bitmap_index()
> >  - pack-bitmap.c: don't leak type-level bitmaps
> >  - pack-bitmap.c: avoid leaking via midx_bitmap_filename()
> >  - builtin/multi-pack-index.c: don't leak concatenated options
> >  - builtin/repack.c: avoid leaking child arguments
> >  - builtin/pack-objects.c: don't leak memory via arguments
> >  - t/helper/test-read-midx.c: free MIDX within read_midx_file()
> >  - midx.c: don't leak MIDX from verify_midx_file
> >  - midx.c: clean up chunkfile after reading the MIDX
> >
> >  Leakfix.
> >
> >  Will merge to 'next'?
>
> These patches all look good to me.
>
> I see you peeled off 10/11 and 11/11 from Taylor's submitted
> patches. The 10/11 re-submitted a patch that's in my
> ab/only-single-progress-at-once, and I really preferred 11/11 not going
> in, and instead suggested [1].
>
> But since you've peeled off those two (I wouldn't have 10/11 at all) I
> think this is definitely ready for 'next'.
>
> 1. https://lore.kernel.org/git/patch-1.1-9190f3c128f-20211022T102725Z-avarab@gmail.com/

I sent a small updated version to fix a couple of things that I noticed
during review here:

    https://lore.kernel.org/git/cover.1635282024.git.me@ttaylorr.com/T/#t

Using either version is fine, of course, but the one above may be a
little nicer. (Apologies for the out-of-thread v2, I only noticed that I
hadn't set `--in-reply-to` until after I had sent out the cover letter).

Thanks,
Taylor