Skip to content

file-posix: Clean up and fix zoned checks

Hanna Czenczek qemu-devel 1 version In review

Get series with b4
b4 am [email protected]

Versions newest first

  1. v1 Latest5 patches 7 repliesIn review

Patches in v1 No applies recorded yet

#Subject and filesSizeRepliesApplied
1/5file-posix: Clear bs->bl.zoned on errorblock/file-posix.c+12 -91 reply
2/5file-posix: Check bs->bl.zoned for zone infoblock/file-posix.c+7 -51 reply
3/5file-posix: Fix zone update in I/O error pathblock/file-posix.c+2 -11 reply
4/5file-posix: Simplify raw_co_prw's 'out' zone codeblock/file-posix.c+5 -131 reply
5/5tests/file-io-error: New testtests/qemu-iotests/tests/file-io-error tests/qemu-iotests/tests/file-io-error.out+130 -00 replies

Reviews trailers per patch

4 of 5 patches have a Reviewed-by or Acked-by.

PatchSam Li
1/5R-b
2/5R-b
3/5R-b
4/5R-b
5/5

Trailers given in replies are filled; those carried in the patches from earlier versions are outlined with dashes.

Read the v1 thread

Cover letter [PATCH 0/5] file-posix: Clean up and fix zoned checks

Hi,

As presented in [1] there is a bug in the zone code in raw_co_prw(), specifically we don’t check whether there actually is zone information before running code that assumes there is (and thus we run into a division by zero). This has now also been reported in [2].

I believe the solution [1] is incomplete, though, which is why I’m sending this separate series: I don’t think checking bs->wps and/or bs->bl.zone_size to determine whether there is zone information is right; for example, we do not have raw_refresh_zoned_limits() clear those values if on a refresh, zone information were to disappear.

It is also weird that we separate checking bs->wps and bs->bl.zone_size at all; raw_refresh_zoned_limits() seems to intend to ensure that either we have information with non-NULL bs->wps and non-zero bs->bl.zone_size, or we don’t.

I think we should have a single flag that tells whether we have valid information or not, and it looks to me like bs->bl.zoned != BLK_Z_NONE is the condition that fits best.

Patch 1 ensures that raw_refresh_zoned_limits() will set bs->bl.zoned to BLK_Z_NONE on error, so that we can actually be sure that this condition tells us whether we have valid information or not.

Patch 2 unifies all conditional checks for zone information to use bs->bl.zoned != BLK_Z_NONE.

Patch 3 is the I/O error path fix, which is not really different from [1].

Patch 4 does a bit of clean-up.

Patch 5 adds a regression test.

[1] https://lists.nongnu.org/archive/html/qemu-devel/2023-06/msg01742.html
[2] https://bugzilla.redhat.com/show_bug.cgi?id=2234374

Hanna Czenczek (5):
file-posix: Clear bs->bl.zoned on error
file-posix: Check bs->bl.zoned for zone info
file-posix: Fix zone update in I/O error path
file-posix: Simplify raw_co_prw's 'out' zone code tests/file-io-error: New test

 block/file-posix.c                         | 42 +++++----
 tests/qemu-iotests/tests/file-io-error     | 99 ++++++++++++++++++++++
 tests/qemu-iotests/tests/file-io-error.out | 31 +++++++
 3 files changed, 150 insertions(+), 22 deletions(-)
 create mode 100755 tests/qemu-iotests/tests/file-io-error
 create mode 100644 tests/qemu-iotests/tests/file-io-error.out
Signature · 1 line
2.41.0