git.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Taylor Blau <me@ttaylorr.com>
To: Patrick Steinhardt <ps@pks.im>
Cc: git@vger.kernel.org, Jeff King <peff@peff.net>,
	Junio C Hamano <gitster@pobox.com>,
	Kyle Lippincott <spectral@google.com>
Subject: Re: [PATCH] pack-bitmap.c: avoid uninitialized `pack_int_id` during reuse
Date: Mon, 10 Jun 2024 10:57:54 -0400	[thread overview]
Message-ID: <ZmcUclCErqwyad9D@nand.local> (raw)
In-Reply-To: <ZmaVYnmgyAr0vapK@tanuki>

On Mon, Jun 10, 2024 at 07:55:46AM +0200, Patrick Steinhardt wrote:
> On Sun, Jun 09, 2024 at 11:27:35AM -0400, Taylor Blau wrote:
> > In 795006fff4 (pack-bitmap: gracefully handle missing BTMP chunks,
> > 2024-04-15), we refactored the reuse_partial_packfile_from_bitmap()
> > function and stopped assigning the pack_int_id field when reusing only
> > the MIDX's preferred pack. This results in an uninitialized read down in
> > try_partial_reuse() like so:
>
> I feel like I'm blind, but I cannot see how the patch changed what we do
> with `pack_int_id`. It's not mentioned a single time in the diff, so how
> did it have the effect of not setting it anymore?

It's because prior to 795006fff4, we handled reusing a single pack from
a MIDX differently than in the post-image of that commit. Prior to
795006fff4, the loop looked like:

    if (bitmap_is_midx(bitmap_git)) {
            for (i = 0; i < bitmap_git->midx->num_packs; i++) {
                    struct bitmapped_pack pack;
                    if (nth_bitmapped_pack(r, bitmap_git->midx, &pack, i) < 0) {
                          /* ... */
                          return;
                    }
                    if (!pack.bitmap_nr)
                          continue;
                    if (!multi_pack_reuse && pack.bitmap_pos)
                          continue;

                    ALLOC_GROW(packs, packs_nr + 1, packs_alloc);
                    memcpy(&packs[packs_nr++], &pack, sizeof(pack));
            }
    }

Since nth_bitmapped_pack() fills out the pack_int_id field, we got it
automatically since we just memcpy()'d the result of
nth_bitmapped_pack() into our array.

In the single pack bitmap case, we don't need to initialize the
pack_int_id field because we never read it, hence the lack of MSan
failures in that mode.

But since 795006fff4 combined these two single pack cases (that is,
single-pack bitmaps, and reusing only a single pack out of a MIDX
bitmap) into one, 795006fff4 neglected to initialize the pack_int_id
field, causing this issue.

Thanks,
Taylor

  reply	other threads:[~2024-06-10 14:57 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-06-09 15:27 [PATCH] pack-bitmap.c: avoid uninitialized `pack_int_id` during reuse Taylor Blau
2024-06-10  5:55 ` Patrick Steinhardt
2024-06-10 14:57   ` Taylor Blau [this message]
2024-06-11  8:12     ` Patrick Steinhardt
2024-06-10 20:10 ` [PATCH v2 0/3] midx: various brown paper bag fixes Taylor Blau
2024-06-10 20:10   ` [PATCH v2 1/3] midx-write.c: do not read existing MIDX with `packs_to_include` Taylor Blau
2024-06-10 20:10   ` [PATCH v2 2/3] pack-bitmap.c: avoid uninitialized `pack_int_id` during reuse Taylor Blau
2024-06-11  9:11     ` Jeff King
2024-06-11 17:03       ` Junio C Hamano
2024-06-10 20:10   ` [PATCH v2 3/3] pack-revindex.c: guard against out-of-bounds pack lookups Taylor Blau
2024-06-11 17:28 ` [PATCH v2 0/3] midx: various brown paper bag fixes Taylor Blau
2024-06-11 17:28   ` [PATCH v2 1/3] midx-write.c: do not read existing MIDX with `packs_to_include` Taylor Blau
2024-06-11 17:28   ` [PATCH v2 2/3] pack-bitmap.c: avoid uninitialized `pack_int_id` during reuse Taylor Blau
2024-06-11 17:28   ` [PATCH v2 3/3] pack-revindex.c: guard against out-of-bounds pack lookups Taylor Blau
2024-06-11 17:31   ` [PATCH v2 0/3] midx: various brown paper bag fixes Taylor Blau

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=ZmcUclCErqwyad9D@nand.local \
    --to=me@ttaylorr.com \
    --cc=git@vger.kernel.org \
    --cc=gitster@pobox.com \
    --cc=peff@peff.net \
    --cc=ps@pks.im \
    --cc=spectral@google.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).