qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Eric Blake <eblake@redhat.com>
To: qemu-devel@nongnu.org
Cc: Fam Zheng <fam@euphon.net>, Kevin Wolf <kwolf@redhat.com>,
	Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>,
	"open list:Block I/O path" <qemu-block@nongnu.org>,
	Max Reitz <mreitz@redhat.com>,
	Stefan Hajnoczi <stefanha@redhat.com>
Subject: [PULL 03/20] util/iov: make qemu_iovec_init_extended() honest
Date: Tue,  2 Feb 2021 16:45:12 -0600	[thread overview]
Message-ID: <20210202224529.642055-4-eblake@redhat.com> (raw)
In-Reply-To: <20210202224529.642055-1-eblake@redhat.com>

From: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>

Actually, we can't extend the io vector in all cases. Handle possible
MAX_IOV and size_t overflows.

For now add assertion to callers (actually they rely on success anyway)
and fix them in the following patch.

Add also some additional good assertions to qemu_iovec_init_slice()
while being here.

Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Message-Id: <20201211183934.169161-3-vsementsov@virtuozzo.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Eric Blake <eblake@redhat.com>
---
 include/qemu/iov.h |  2 +-
 block/io.c         | 10 +++++++---
 util/iov.c         | 25 +++++++++++++++++++++++--
 3 files changed, 31 insertions(+), 6 deletions(-)

diff --git a/include/qemu/iov.h b/include/qemu/iov.h
index b6b283a5e5c1..93307466809b 100644
--- a/include/qemu/iov.h
+++ b/include/qemu/iov.h
@@ -222,7 +222,7 @@ static inline void *qemu_iovec_buf(QEMUIOVector *qiov)

 void qemu_iovec_init(QEMUIOVector *qiov, int alloc_hint);
 void qemu_iovec_init_external(QEMUIOVector *qiov, struct iovec *iov, int niov);
-void qemu_iovec_init_extended(
+int qemu_iovec_init_extended(
         QEMUIOVector *qiov,
         void *head_buf, size_t head_len,
         QEMUIOVector *mid_qiov, size_t mid_offset, size_t mid_len,
diff --git a/block/io.c b/block/io.c
index 23abdae79468..ab953bd58f48 100644
--- a/block/io.c
+++ b/block/io.c
@@ -1680,13 +1680,17 @@ static bool bdrv_pad_request(BlockDriverState *bs,
                              int64_t *offset, unsigned int *bytes,
                              BdrvRequestPadding *pad)
 {
+    int ret;
+
     if (!bdrv_init_padding(bs, *offset, *bytes, pad)) {
         return false;
     }

-    qemu_iovec_init_extended(&pad->local_qiov, pad->buf, pad->head,
-                             *qiov, *qiov_offset, *bytes,
-                             pad->buf + pad->buf_len - pad->tail, pad->tail);
+    ret = qemu_iovec_init_extended(&pad->local_qiov, pad->buf, pad->head,
+                                   *qiov, *qiov_offset, *bytes,
+                                   pad->buf + pad->buf_len - pad->tail,
+                                   pad->tail);
+    assert(ret == 0);
     *bytes += pad->head + pad->tail;
     *offset -= pad->head;
     *qiov = &pad->local_qiov;
diff --git a/util/iov.c b/util/iov.c
index f3a9e92a378f..58c7b3eeee5f 100644
--- a/util/iov.c
+++ b/util/iov.c
@@ -415,7 +415,7 @@ int qemu_iovec_subvec_niov(QEMUIOVector *qiov, size_t offset, size_t len)
  * Compile new iovec, combining @head_buf buffer, sub-qiov of @mid_qiov,
  * and @tail_buf buffer into new qiov.
  */
-void qemu_iovec_init_extended(
+int qemu_iovec_init_extended(
         QEMUIOVector *qiov,
         void *head_buf, size_t head_len,
         QEMUIOVector *mid_qiov, size_t mid_offset, size_t mid_len,
@@ -425,12 +425,24 @@ void qemu_iovec_init_extended(
     int total_niov, mid_niov = 0;
     struct iovec *p, *mid_iov = NULL;

+    assert(mid_qiov->niov <= IOV_MAX);
+
+    if (SIZE_MAX - head_len < mid_len ||
+        SIZE_MAX - head_len - mid_len < tail_len)
+    {
+        return -EINVAL;
+    }
+
     if (mid_len) {
         mid_iov = qiov_slice(mid_qiov, mid_offset, mid_len,
                              &mid_head, &mid_tail, &mid_niov);
     }

     total_niov = !!head_len + mid_niov + !!tail_len;
+    if (total_niov > IOV_MAX) {
+        return -EINVAL;
+    }
+
     if (total_niov == 1) {
         qemu_iovec_init_buf(qiov, NULL, 0);
         p = &qiov->local_iov;
@@ -459,6 +471,8 @@ void qemu_iovec_init_extended(
         p->iov_base = tail_buf;
         p->iov_len = tail_len;
     }
+
+    return 0;
 }

 /*
@@ -492,7 +506,14 @@ bool qemu_iovec_is_zero(QEMUIOVector *qiov, size_t offset, size_t bytes)
 void qemu_iovec_init_slice(QEMUIOVector *qiov, QEMUIOVector *source,
                            size_t offset, size_t len)
 {
-    qemu_iovec_init_extended(qiov, NULL, 0, source, offset, len, NULL, 0);
+    int ret;
+
+    assert(source->size >= len);
+    assert(source->size - len >= offset);
+
+    /* We shrink the request, so we can't overflow neither size_t nor MAX_IOV */
+    ret = qemu_iovec_init_extended(qiov, NULL, 0, source, offset, len, NULL, 0);
+    assert(ret == 0);
 }

 void qemu_iovec_destroy(QEMUIOVector *qiov)
-- 
2.30.0



  parent reply	other threads:[~2021-02-02 22:52 UTC|newest]

Thread overview: 24+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-02-02 22:45 [PULL 00/20] NBD patches for 2021-02-02 Eric Blake
2021-02-02 22:45 ` [PULL 01/20] iotests: Fix expected whitespace for 185 Eric Blake
2021-02-02 22:45 ` [PULL 02/20] block: refactor bdrv_check_request: add errp Eric Blake
2021-02-02 22:45 ` Eric Blake [this message]
2021-02-02 22:45 ` [PULL 04/20] block: fix theoretical overflow in bdrv_init_padding() Eric Blake
2021-02-02 22:45 ` [PULL 05/20] block/io: refactor bdrv_pad_request(): move bdrv_pad_request() up Eric Blake
2021-02-02 22:45 ` [PULL 06/20] block/io: bdrv_pad_request(): support qemu_iovec_init_extended failure Eric Blake
2021-02-02 22:45 ` [PULL 07/20] block/throttle-groups: throttle_group_co_io_limits_intercept(): 64bit bytes Eric Blake
2021-02-02 22:45 ` [PULL 08/20] block/io: improve bdrv_check_request: check qiov too Eric Blake
2021-02-02 22:45 ` [PULL 09/20] block: use int64_t as bytes type in tracked requests Eric Blake
2021-02-02 22:45 ` [PULL 10/20] block/io: use int64_t bytes in driver wrappers Eric Blake
2021-02-02 22:45 ` [PULL 11/20] block/io: support int64_t bytes in bdrv_co_do_pwrite_zeroes() Eric Blake
2021-02-02 22:45 ` [PULL 12/20] block/io: support int64_t bytes in bdrv_aligned_pwritev() Eric Blake
2021-02-02 22:45 ` [PULL 13/20] block/io: support int64_t bytes in bdrv_co_do_copy_on_readv() Eric Blake
2021-02-02 22:45 ` [PULL 14/20] block/io: support int64_t bytes in bdrv_aligned_preadv() Eric Blake
2021-02-02 22:45 ` [PULL 15/20] block/io: support int64_t bytes in bdrv_co_p{read, write}v_part() Eric Blake
2021-02-02 22:45 ` [PULL 16/20] block/io: support int64_t bytes in read/write wrappers Eric Blake
2021-02-02 22:45 ` [PULL 17/20] block/io: use int64_t bytes in copy_range Eric Blake
2021-02-02 22:45 ` [PULL 18/20] block/nbd: only detach existing iochannel from aio_context Eric Blake
2021-02-02 22:45 ` [PULL 19/20] block/nbd: only enter connection coroutine if it's present Eric Blake
2021-02-02 22:45 ` [PULL 20/20] nbd: make nbd_read* return -EIO on error Eric Blake
2021-02-02 23:12 ` [PULL 00/20] NBD patches for 2021-02-02 no-reply
2021-02-03 13:25   ` Eric Blake
2021-02-03 13:27 ` Eric Blake

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=20210202224529.642055-4-eblake@redhat.com \
    --to=eblake@redhat.com \
    --cc=fam@euphon.net \
    --cc=kwolf@redhat.com \
    --cc=mreitz@redhat.com \
    --cc=qemu-block@nongnu.org \
    --cc=qemu-devel@nongnu.org \
    --cc=stefanha@redhat.com \
    --cc=vsementsov@virtuozzo.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).