qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Alexander Ivanov <alexander.ivanov@virtuozzo.com>
To: qemu-devel@nongnu.org
Cc: qemu-block@nongnu.org, den@virtuozzo.com, stefanha@redhat.com,
	vsementsov@yandex-team.ru, kwolf@redhat.com, hreitz@redhat.com
Subject: [PATCH v8 09/10] parallels: Add data_off check
Date: Tue, 18 Jul 2023 12:44:31 +0200	[thread overview]
Message-ID: <20230718104432.1130469-10-alexander.ivanov@virtuozzo.com> (raw)
In-Reply-To: <20230718104432.1130469-1-alexander.ivanov@virtuozzo.com>

data_off field of the parallels image header can be corrupted. Check if
this field greater than the header + BAT size and less than file size.

Signed-off-by: Alexander Ivanov <alexander.ivanov@virtuozzo.com>
---
 block/parallels.c | 80 +++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 80 insertions(+)

diff --git a/block/parallels.c b/block/parallels.c
index 3414807089..103acb7b40 100644
--- a/block/parallels.c
+++ b/block/parallels.c
@@ -450,6 +450,81 @@ static void parallels_check_unclean(BlockDriverState *bs,
     }
 }
 
+/*
+ * Returns true if data_off is correct, otherwise false. In both cases
+ * correct_offset is set to the proper value.
+ */
+static bool parallels_test_data_off(BDRVParallelsState *s,
+                                    int64_t file_nb_sectors,
+                                    uint32_t *correct_offset)
+{
+    uint32_t data_off, min_off;
+    bool old_magic;
+
+    /*
+     * There are two slightly different image formats: with "WithoutFreeSpace"
+     * or "WithouFreSpacExt" magic words. Call the first one as "old magic".
+     * In such images data_off field can be zero. In this case the offset is
+     * calculated as the end of BAT table plus some padding to ensure sector
+     * size alignment.
+     */
+    old_magic = !memcmp(s->header->magic, HEADER_MAGIC, 16);
+
+    min_off = DIV_ROUND_UP(bat_entry_off(s->bat_size), BDRV_SECTOR_SIZE);
+    if (!old_magic) {
+        min_off = ROUND_UP(min_off, s->cluster_size / BDRV_SECTOR_SIZE);
+    }
+
+    if (correct_offset) {
+        *correct_offset = min_off;
+    }
+
+    data_off = le32_to_cpu(s->header->data_off);
+    if (data_off == 0 && old_magic) {
+        return true;
+    }
+
+    if (data_off < min_off || data_off > file_nb_sectors) {
+        return false;
+    }
+
+    if (correct_offset) {
+        *correct_offset = data_off;
+    }
+
+    return true;
+}
+
+static int coroutine_fn GRAPH_RDLOCK
+parallels_check_data_off(BlockDriverState *bs, BdrvCheckResult *res,
+                         BdrvCheckMode fix)
+{
+    BDRVParallelsState *s = bs->opaque;
+    int64_t file_size;
+    uint32_t data_off;
+
+    file_size = bdrv_co_nb_sectors(bs->file->bs);
+    if (file_size < 0) {
+        res->check_errors++;
+        return file_size;
+    }
+
+    if (parallels_test_data_off(s, file_size, &data_off)) {
+        return 0;
+    }
+
+    res->corruptions++;
+    if (fix & BDRV_FIX_ERRORS) {
+        s->header->data_off = cpu_to_le32(data_off);
+        res->corruptions_fixed++;
+    }
+
+    fprintf(stderr, "%s data_off field has incorrect value\n",
+            fix & BDRV_FIX_ERRORS ? "Repairing" : "ERROR");
+
+    return 0;
+}
+
 static int coroutine_fn GRAPH_RDLOCK
 parallels_check_outside_image(BlockDriverState *bs, BdrvCheckResult *res,
                               BdrvCheckMode fix)
@@ -713,6 +788,11 @@ parallels_co_check(BlockDriverState *bs, BdrvCheckResult *res,
     WITH_QEMU_LOCK_GUARD(&s->lock) {
         parallels_check_unclean(bs, res, fix);
 
+        ret = parallels_check_data_off(bs, res, fix);
+        if (ret < 0) {
+            return ret;
+        }
+
         ret = parallels_check_outside_image(bs, res, fix);
         if (ret < 0) {
             return ret;
-- 
2.34.1



  parent reply	other threads:[~2023-07-18 10:47 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-07-18 10:44 [PATCH v8 00/10] parallels: Add duplication check, repair at open, fix bugs Alexander Ivanov
2023-07-18 10:44 ` [PATCH v8 01/10] parallels: Fix comments formatting inside parallels driver Alexander Ivanov
2023-07-26  9:41   ` Denis V. Lunev
2023-07-18 10:44 ` [PATCH v8 02/10] parallels: Incorrect data end calculation in parallels_open() Alexander Ivanov
2023-07-18 10:44 ` [PATCH v8 03/10] parallels: Check if data_end greater than the file size Alexander Ivanov
2023-07-18 10:44 ` [PATCH v8 04/10] parallels: Add "explicit" argument to parallels_check_leak() Alexander Ivanov
2023-07-18 10:44 ` [PATCH v8 05/10] parallels: Add data_start field to BDRVParallelsState Alexander Ivanov
2023-07-18 10:44 ` [PATCH v8 06/10] parallels: Add checking and repairing duplicate offsets in BAT Alexander Ivanov
2023-07-18 10:44 ` [PATCH v8 07/10] parallels: Image repairing in parallels_open() Alexander Ivanov
2023-07-18 10:44 ` [PATCH v8 08/10] parallels: Use bdrv_co_getlength() in parallels_check_outside_image() Alexander Ivanov
2023-07-18 10:44 ` Alexander Ivanov [this message]
2023-07-18 10:44 ` [PATCH v8 10/10] parallels: Add data_off repairing to parallels_open() Alexander Ivanov

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=20230718104432.1130469-10-alexander.ivanov@virtuozzo.com \
    --to=alexander.ivanov@virtuozzo.com \
    --cc=den@virtuozzo.com \
    --cc=hreitz@redhat.com \
    --cc=kwolf@redhat.com \
    --cc=qemu-block@nongnu.org \
    --cc=qemu-devel@nongnu.org \
    --cc=stefanha@redhat.com \
    --cc=vsementsov@yandex-team.ru \
    /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).