From: Fiona Ebner <f.ebner@proxmox.com>

Allow overlapping request by removing the assert that made it
impossible. There are only two callers:

1. block_copy_task_create()

It already asserts the very same condition before calling
reqlist_init_req().

2. cbw_snapshot_read_lock()

There is no need to have read requests be non-overlapping in
copy-before-write when used for snapshot-access. In fact, there was no
protection against two callers of cbw_snapshot_read_lock() calling
reqlist_init_req() with overlapping ranges and this could lead to an
assertion failure [1].

In particular, with the reproducer script below [0], two
cbw_co_snapshot_block_status() callers could race, with the second
calling reqlist_init_req() before the first one finishes and removes
its conflicting request.

[0]:

> #!/bin/bash -e
> dd if=/dev/urandom of=/tmp/disk.raw bs=1M count=1024
> ./qemu-img create /tmp/fleecing.raw -f raw 1G
> (
> ./qemu-system-x86_64 --qmp stdio \
> --blockdev raw,node-name=node0,file.driver=file,file.filename=/tmp/disk.raw \
> --blockdev raw,node-name=node1,file.driver=file,file.filename=/tmp/fleecing.raw \
> <<EOF
> {"execute": "qmp_capabilities"}
> {"execute": "blockdev-add", "arguments": { "driver": "copy-before-write", "file": "node0", "target": "node1", "node-name": "node3" } }
> {"execute": "blockdev-add", "arguments": { "driver": "snapshot-access", "file": "node3", "node-name": "snap0" } }
> {"execute": "nbd-server-start", "arguments": {"addr": { "type": "unix", "data": { "path": "/tmp/nbd.socket" } } } }
> {"execute": "block-export-add", "arguments": {"id": "exp0", "node-name": "snap0", "type": "nbd", "name": "exp0"}}
> EOF
> ) &
> sleep 5
> while true; do
> ./qemu-nbd -d /dev/nbd0
> ./qemu-nbd -c /dev/nbd0 nbd:unix:/tmp/nbd.socket:exportname=exp0 -f raw -r
> nbdinfo --map 'nbd+unix:///exp0?socket=/tmp/nbd.socket'
> done

[1]:

> #5  0x000071e5f0088eb2 in __GI___assert_fail (...) at ./assert/assert.c:101
> #6  0x0000615285438017 in reqlist_init_req (...) at ../block/reqlist.c:23
> #7  0x00006152853e2d98 in cbw_snapshot_read_lock (...) at ../block/copy-before-write.c:237
> #8  0x00006152853e3068 in cbw_co_snapshot_block_status (...) at ../block/copy-before-write.c:304
> #9  0x00006152853f4d22 in bdrv_co_snapshot_block_status (...) at ../block/io.c:3726
> #10 0x000061528543a63e in snapshot_access_co_block_status (...) at ../block/snapshot-access.c:48
> #11 0x00006152853f1a0a in bdrv_co_do_block_status (...) at ../block/io.c:2474
> #12 0x00006152853f2016 in bdrv_co_common_block_status_above (...) at ../block/io.c:2652
> #13 0x00006152853f22cf in bdrv_co_block_status_above (...) at ../block/io.c:2732
> #14 0x00006152853d9a86 in blk_co_block_status_above (...) at ../block/block-backend.c:1473
> #15 0x000061528538da6c in blockstatus_to_extents (...) at ../nbd/server.c:2374
> #16 0x000061528538deb1 in nbd_co_send_block_status (...) at ../nbd/server.c:2481
> #17 0x000061528538f424 in nbd_handle_request (...) at ../nbd/server.c:2978
> #18 0x000061528538f906 in nbd_trip (...) at ../nbd/server.c:3121
> #19 0x00006152855a7caf in coroutine_trampoline (...) at ../util/coroutine-ucontext.c:175

Cc: qemu-stable@nongnu.org
Suggested-by: Vladimir Sementsov-Ogievskiy <vsementsov@yandex-team.ru>
Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
Message-Id: <20240712140716.517911-1-f.ebner@proxmox.com>
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@yandex-team.ru>
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@yandex-team.ru>
---
 block/copy-before-write.c | 3 ++-
 block/reqlist.c           | 2 --
 2 files changed, 2 insertions(+), 3 deletions(-)

diff --git a/block/copy-before-write.c b/block/copy-before-write.c
index e835987e52..81afeff1c7 100644
--- a/block/copy-before-write.c
+++ b/block/copy-before-write.c
@@ -66,7 +66,8 @@ typedef struct BDRVCopyBeforeWriteState {
 
     /*
      * @frozen_read_reqs: current read requests for fleecing user in bs->file
-     * node. These areas must not be rewritten by guest.
+     * node. These areas must not be rewritten by guest. There can be multiple
+     * overlapping read requests.
      */
     BlockReqList frozen_read_reqs;
 
diff --git a/block/reqlist.c b/block/reqlist.c
index 08cb57cfa4..098e807378 100644
--- a/block/reqlist.c
+++ b/block/reqlist.c
@@ -20,8 +20,6 @@
 void reqlist_init_req(BlockReqList *reqs, BlockReq *req, int64_t offset,
                       int64_t bytes)
 {
-    assert(!reqlist_find_conflict(reqs, offset, bytes));
-
     *req = (BlockReq) {
         .offset = offset,
         .bytes = bytes,
-- 
2.34.1

30.09.2024 11:43, Vladimir Sementsov-Ogievskiy wrote:
> From: Fiona Ebner <f.ebner@proxmox.com>
> 
> Allow overlapping request by removing the assert that made it
> impossible. There are only two callers:
> 
> 1. block_copy_task_create()
> 
> It already asserts the very same condition before calling
> reqlist_init_req().
> 
> 2. cbw_snapshot_read_lock()
> 
> There is no need to have read requests be non-overlapping in
> copy-before-write when used for snapshot-access. In fact, there was no
> protection against two callers of cbw_snapshot_read_lock() calling
> reqlist_init_req() with overlapping ranges and this could lead to an
> assertion failure [1].
> 
> In particular, with the reproducer script below [0], two
> cbw_co_snapshot_block_status() callers could race, with the second
> calling reqlist_init_req() before the first one finishes and removes
> its conflicting request.

Hm.  This one applies to 7.2 too (current oldest stable series), with
the description above matching what the code is doing.

I picked it up for up to 7.2.  Please let me know if this shouldn't be
done :)

Thanks,

/mjt

On 01.10.24 19:28, Michael Tokarev wrote:
> 30.09.2024 11:43, Vladimir Sementsov-Ogievskiy wrote:
>> From: Fiona Ebner <f.ebner@proxmox.com>
>>
>> Allow overlapping request by removing the assert that made it
>> impossible. There are only two callers:
>>
>> 1. block_copy_task_create()
>>
>> It already asserts the very same condition before calling
>> reqlist_init_req().
>>
>> 2. cbw_snapshot_read_lock()
>>
>> There is no need to have read requests be non-overlapping in
>> copy-before-write when used for snapshot-access. In fact, there was no
>> protection against two callers of cbw_snapshot_read_lock() calling
>> reqlist_init_req() with overlapping ranges and this could lead to an
>> assertion failure [1].
>>
>> In particular, with the reproducer script below [0], two
>> cbw_co_snapshot_block_status() callers could race, with the second
>> calling reqlist_init_req() before the first one finishes and removes
>> its conflicting request.
> 
> Hm.  This one applies to 7.2 too (current oldest stable series), with
> the description above matching what the code is doing.
> 
> I picked it up for up to 7.2.  Please let me know if this shouldn't be
> done :)
> 

I don't see any problems) Still, that's not a guarantee that we don't have them. At least, we definitely lack a test for this case.


-- 
Best regards,
Vladimir