All of lore.kernel.org
 help / color / mirror / Atom feed
From: Eric Biggers <ebiggers@kernel.org>
To: linux-xfs@vger.kernel.org
Cc: fstests@vger.kernel.org
Subject: [xfsprogs PATCH] xfs_io: fix mread with length 1 mod page size
Date: Tue, 11 Jun 2024 11:29:28 -0700	[thread overview]
Message-ID: <20240611182928.12813-1-ebiggers@kernel.org> (raw)

From: Eric Biggers <ebiggers@google.com>

Fix a weird bug in mread where if you passed it a length that was 1
modulo the page size, for example

        xfs_io -r file -c "mmap -r 0 8192" -c "mread -v 0 4097"

... it never reset its pointer into the buffer into which it copies the
data from the memory map.  This caused an out-of-bounds write, which
depending on the length passed could be very large and reliably
segfault.  Also nothing was printed, despite the use of -v option.

(I don't know if this case gets reached by any existing xfstest, but
presumably not.  I noticed it while working on a patch to an xfstest.)

Signed-off-by: Eric Biggers <ebiggers@google.com>
---
 io/mmap.c | 20 ++++++--------------
 1 file changed, 6 insertions(+), 14 deletions(-)

diff --git a/io/mmap.c b/io/mmap.c
index 85087f57..4c03e3d5 100644
--- a/io/mmap.c
+++ b/io/mmap.c
@@ -469,38 +469,30 @@ mread_f(
 	dumplen = length % pagesize;
 	if (!dumplen)
 		dumplen = pagesize;
 
 	if (rflag) {
-		for (tmp = length - 1, c = 0; tmp >= 0; tmp--, c = 1) {
-			*bp = *(((char *)mapping->addr) + dumpoffset + tmp);
-			cnt++;
-			if (c && cnt == dumplen) {
+		for (tmp = length - 1; tmp >= 0; tmp--) {
+			bp[cnt++] = ((char *)mapping->addr)[dumpoffset + tmp];
+			if (cnt == dumplen) {
 				if (dump) {
 					dump_buffer(printoffset, dumplen);
 					printoffset += dumplen;
 				}
-				bp = (char *)io_buffer;
 				dumplen = pagesize;
 				cnt = 0;
-			} else {
-				bp++;
 			}
 		}
 	} else {
-		for (tmp = 0, c = 0; tmp < length; tmp++, c = 1) {
-			*bp = *(((char *)mapping->addr) + dumpoffset + tmp);
-			cnt++;
-			if (c && cnt == dumplen) {
+		for (tmp = 0; tmp < length; tmp++) {
+			bp[cnt++] = ((char *)mapping->addr)[dumpoffset + tmp];
+			if (cnt == dumplen) {
 				if (dump)
 					dump_buffer(printoffset + tmp -
 						(dumplen - 1), dumplen);
-				bp = (char *)io_buffer;
 				dumplen = pagesize;
 				cnt = 0;
-			} else {
-				bp++;
 			}
 		}
 	}
 	return 0;
 }

base-commit: df4bd2d27189a98711fd35965c18bee25a25a9ea
-- 
2.45.1


             reply	other threads:[~2024-06-11 18:30 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-06-11 18:29 Eric Biggers [this message]
2024-06-12 18:01 ` [xfsprogs PATCH] xfs_io: fix mread with length 1 mod page size Darrick J. Wong

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=20240611182928.12813-1-ebiggers@kernel.org \
    --to=ebiggers@kernel.org \
    --cc=fstests@vger.kernel.org \
    --cc=linux-xfs@vger.kernel.org \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.