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

[PATCH 09/11] multi-pack-index: verify object offsets

From
Derrick Stolee via GitGitGadget <gitgitgadget@gmail.com>
Date
Sep 5, 2018, 14:46 UTC
Message-ID
<e1498aea45eb46feb3664413f49b70bdf048d284.1536158789.git.gitgitgadget@gmail.com>
In-Reply-To
<pull.34.git.gitgitgadget@gmail.com>
From: Derrick Stolee <dstolee@microsoft.com>

The 'git multi-pack-index verify' command must verify the object offsets stored in the multi-pack-index are correct. There are two ways the offset chunk can be incorrect: the pack-int-id and the object offset.

Replace the BUG() statement with a die() statement, now that we may hit a bad pack-int-id during a 'verify' command on a corrupt multi-pack-index, and it is covered by a test.

Signed-off-by: Derrick Stolee <dstolee@microsoft.com>
---
 midx.c                      | 28 +++++++++++++++++++++++++++-
 t/t5319-multi-pack-index.sh | 27 +++++++++++++++++++++++++++
 2 files changed, 54 insertions(+), 1 deletion(-)
diff --git a/midx.c b/midx.c
index 80094c02a7..87708dc367 100644
--- a/midx.c
+++ b/midx.c
@@ -197,7 +197,7 @@ int prepare_midx_pack(struct multi_pack_index *m, uint32_t pack_int_id)
 	struct strbuf pack_name = STRBUF_INIT;
 
 	if (pack_int_id >= m->num_packs)
-		BUG("bad pack-int-id");
+		die(_("bad pack-int-id"));
 
 	if (m->packs[pack_int_id])
 		return 0;
@@ -970,5 +970,31 @@ int verify_midx_file(const char *object_dir)
 				    i, oid_to_hex(&oid1), oid_to_hex(&oid2), i + 1);
 	}
 
+	for (i = 0; i < m->num_objects; i++) {
+		struct object_id oid;
+		struct pack_entry e;
+		off_t m_offset, p_offset;
+
+		nth_midxed_object_oid(&oid, m, i);
+		if (!fill_midx_entry(&oid, &e, m)) {
+			midx_report(_("failed to load pack entry for oid[%d] = %s"),
+				    i, oid_to_hex(&oid));
+			continue;
+		}
+
+		if (open_pack_index(e.p)) {
+			midx_report(_("failed to load pack-index for packfile %s"),
+				    e.p->pack_name);
+			break;
+		}
+
+		m_offset = e.offset;
+		p_offset = find_pack_entry_one(oid.hash, e.p);
+
+		if (m_offset != p_offset)
+			midx_report(_("incorrect object offset for oid[%d] = %s: %"PRIx64" != %"PRIx64),
+				    i, oid_to_hex(&oid), m_offset, p_offset);
+	}
+
 	return verify_midx_error;
 }
diff --git a/t/t5319-multi-pack-index.sh b/t/t5319-multi-pack-index.sh
index 072b1d1a74..b6c34631d3 100755
--- a/t/t5319-multi-pack-index.sh
+++ b/t/t5319-multi-pack-index.sh
@@ -176,6 +176,7 @@ test_expect_success 'verify bad signature' '
 '
 
 HASH_LEN=20
+NUM_OBJECTS=74
 MIDX_BYTE_VERSION=4
 MIDX_BYTE_OID_VERSION=5
 MIDX_BYTE_CHUNK_COUNT=6
@@ -192,6 +193,10 @@ MIDX_OID_FANOUT_WIDTH=4
 MIDX_BYTE_OID_FANOUT_ORDER=$((MIDX_OFFSET_OID_FANOUT + 250 * $MIDX_OID_FANOUT_WIDTH + 1))
 MIDX_OFFSET_OID_LOOKUP=$(($MIDX_OFFSET_OID_FANOUT + 256 * $MIDX_OID_FANOUT_WIDTH))
 MIDX_BYTE_OID_LOOKUP=$(($MIDX_OFFSET_OID_LOOKUP + 16 * $HASH_LEN))
+MIDX_OFFSET_OBJECT_OFFSETS=$(($MIDX_OFFSET_OID_LOOKUP + $NUM_OBJECTS * $HASH_LEN))
+MIDX_OFFSET_WIDTH=8
+MIDX_BYTE_PACK_INT_ID=$(($MIDX_OFFSET_OBJECT_OFFSETS + 16 * $MIDX_OFFSET_WIDTH + 2))
+MIDX_BYTE_OFFSET=$(($MIDX_OFFSET_OBJECT_OFFSETS + 16 * $MIDX_OFFSET_WIDTH + 6))
 
 test_expect_success 'verify bad version' '
 	corrupt_midx_and_verify $MIDX_BYTE_VERSION "\00" $objdir \
@@ -243,6 +248,16 @@ test_expect_success 'verify oid lookup out of order' '
 		"oid lookup out of order"
 '
 
+test_expect_success 'verify incorrect pack-int-id' '
+	corrupt_midx_and_verify $MIDX_BYTE_PACK_INT_ID "\07" $objdir \
+		"bad pack-int-id"
+'
+
+test_expect_success 'verify incorrect offset' '
+	corrupt_midx_and_verify $MIDX_BYTE_OFFSET "\07" $objdir \
+		"incorrect object offset"
+'
+
 test_expect_success 'repack removes multi-pack-index' '
 	test_path_is_file $objdir/pack/multi-pack-index &&
 	git repack -adf &&
@@ -310,4 +325,16 @@ test_expect_success 'verify multi-pack-index with 64-bit offsets' '
 	git multi-pack-index verify --object-dir=objects64
 '
 
+NUM_OBJECTS=63
+MIDX_OFFSET_OID_FANOUT=$((MIDX_OFFSET_PACKNAMES + 54))
+MIDX_OFFSET_OID_LOOKUP=$((MIDX_OFFSET_OID_FANOUT + 256 * $MIDX_OID_FANOUT_WIDTH))
+MIDX_OFFSET_OBJECT_OFFSETS=$(($MIDX_OFFSET_OID_LOOKUP + $NUM_OBJECTS * $HASH_LEN))
+MIDX_OFFSET_LARGE_OFFSETS=$(($MIDX_OFFSET_OBJECT_OFFSETS + $NUM_OBJECTS * $MIDX_OFFSET_WIDTH))
+MIDX_BYTE_LARGE_OFFSET=$(($MIDX_OFFSET_LARGE_OFFSETS + 3))
+
+test_expect_success 'verify incorrect 64-bit offset' '
+	corrupt_midx_and_verify $MIDX_BYTE_LARGE_OFFSET "\07" objects64 \
+		"incorrect object offset"
+'
+
 test_done
-- 
gitgitgadget
Previous: Derrick Stolee via GitGitGadgetNext: Eric Sunshine
Message 17 of 34 in “Add 'git multi-pack-index verify' command”
  1. 00/11 Add 'git multi-pack-index verify' commandDerrick Stolee via GitGitGadget, Sep 5, 2018
  2. 01/11 multi-pack-index: add 'verify' verbDerrick Stolee via GitGitGadget, Sep 5, 2018
  3. Eric SunshineSep 5, 2018
  4. Derrick StoleeSep 5, 2018
  5. 02/11 multi-pack-index: verify bad headerDerrick Stolee via GitGitGadget, Sep 5, 2018
  6. Eric SunshineSep 7, 2018
  7. 04/11 multi-pack-index: verify packname orderDerrick Stolee via GitGitGadget, Sep 5, 2018
  8. Stefan BellerSep 5, 2018
  9. Derrick StoleeSep 5, 2018
  10. Stefan BellerSep 5, 2018
  11. Derrick StoleeSep 5, 2018
  12. 05/11 multi-pack-index: verify missing packDerrick Stolee via GitGitGadget, Sep 5, 2018
  13. 06/11 multi-pack-index: verify oid fanout orderDerrick Stolee via GitGitGadget, Sep 5, 2018
  14. 07/11 multi-pack-index: verify oid lookup orderDerrick Stolee via GitGitGadget, Sep 5, 2018
  15. 08/11 multi-pack-index: fix 32-bit vs 64-bit size checkDerrick Stolee via GitGitGadget, Sep 5, 2018
  16. 03/11 multi-pack-index: verify corrupt chunk lookup tableDerrick Stolee via GitGitGadget, Sep 5, 2018
  17. 09/11 multi-pack-index: verify object offsetsDerrick Stolee via GitGitGadget, Sep 5, 2018
  18. Eric SunshineSep 7, 2018
  19. Derrick StoleeSep 7, 2018
  20. 10/11 multi-pack-index: report progress during 'verify'Derrick Stolee via GitGitGadget, Sep 5, 2018
  21. 11/11 fsck: verify multi-pack-indexDerrick Stolee via GitGitGadget, Sep 5, 2018
  22. 00/11 Add 'git multi-pack-index verify' commandDerrick Stolee via GitGitGadget, Sep 13, 2018
  23. 02/11 multi-pack-index: verify bad headerDerrick Stolee via GitGitGadget, Sep 13, 2018
  24. 03/11 multi-pack-index: verify corrupt chunk lookup tableDerrick Stolee via GitGitGadget, Sep 13, 2018
  25. 04/11 multi-pack-index: verify packname orderDerrick Stolee via GitGitGadget, Sep 13, 2018
  26. 05/11 multi-pack-index: verify missing packDerrick Stolee via GitGitGadget, Sep 13, 2018
  27. 06/11 multi-pack-index: verify oid fanout orderDerrick Stolee via GitGitGadget, Sep 13, 2018
  28. 07/11 multi-pack-index: verify oid lookup orderDerrick Stolee via GitGitGadget, Sep 13, 2018
  29. 08/11 multi-pack-index: fix 32-bit vs 64-bit size checkDerrick Stolee via GitGitGadget, Sep 13, 2018
  30. 09/11 multi-pack-index: verify object offsetsDerrick Stolee via GitGitGadget, Sep 13, 2018
  31. 10/11 multi-pack-index: report progress during 'verify'Derrick Stolee via GitGitGadget, Sep 13, 2018
  32. Ævar Arnfjörð BjarmasonSep 13, 2018
  33. 11/11 fsck: verify multi-pack-indexDerrick Stolee via GitGitGadget, Sep 13, 2018
  34. 01/11 multi-pack-index: add 'verify' verbDerrick Stolee via GitGitGadget, Sep 13, 2018

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.