qemu-kvm/kvm-file-posix-Check-bs-bl.zoned-for-zone-info.patch
Miroslav Rezanina 5f95659303 * Mon Oct 16 2023 Miroslav Rezanina <mrezanin@redhat.com> - 8.1.0-3
- kvm-migration-Fix-race-that-dest-preempt-thread-close-to.patch [RHEL-11219]
- kvm-migration-Fix-possible-race-when-setting-rp_state.er.patch [RHEL-11219]
- kvm-migration-Fix-possible-races-when-shutting-down-the-.patch [RHEL-11219]
- kvm-migration-Fix-possible-race-when-shutting-down-to_ds.patch [RHEL-11219]
- kvm-migration-Remove-redundant-cleanup-of-postcopy_qemuf.patch [RHEL-11219]
- kvm-migration-Consolidate-return-path-closing-code.patch [RHEL-11219]
- kvm-migration-Replace-the-return-path-retry-logic.patch [RHEL-11219]
- kvm-migration-Move-return-path-cleanup-to-main-migration.patch [RHEL-11219]
- kvm-file-posix-Clear-bs-bl.zoned-on-error.patch [RHEL-7360]
- kvm-file-posix-Check-bs-bl.zoned-for-zone-info.patch [RHEL-7360]
- kvm-file-posix-Fix-zone-update-in-I-O-error-path.patch [RHEL-7360]
- kvm-file-posix-Simplify-raw_co_prw-s-out-zone-code.patch [RHEL-7360]
- kvm-tests-file-io-error-New-test.patch [RHEL-7360]
- Resolves: RHEL-11219
  (migration tests failing for RHEL 9.4 sometimes)
- Resolves: RHEL-7360
  (Qemu Core Dumped When Writing Larger Size Than The Size of A Data Disk)
2023-10-16 03:40:21 -04:00

74 lines
2.9 KiB
Diff

From 42dac3d063e0f44e92211a4fbdaf1f0d64843562 Mon Sep 17 00:00:00 2001
From: Hanna Czenczek <hreitz@redhat.com>
Date: Thu, 24 Aug 2023 17:53:41 +0200
Subject: [PATCH 10/13] file-posix: Check bs->bl.zoned for zone info
RH-Author: Hanna Czenczek <hreitz@redhat.com>
RH-MergeRequest: 202: file-posix: Fix zone update in I/O error path
RH-Jira: RHEL-7360
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Commit: [2/5] d853b88d6b71b6f4fe1bb70ae2e3b6108cbdb8dc (hreitz/qemu-kvm-c-9-s)
Instead of checking bs->wps or bs->bl.zone_size for whether zone
information is present, check bs->bl.zoned. That is the flag that
raw_refresh_zoned_limits() reliably sets to indicate zone support. If
it is set to something other than BLK_Z_NONE, other values and objects
like bs->wps and bs->bl.zone_size must be non-null/zero and valid; if it
is not, we cannot rely on their validity.
Signed-off-by: Hanna Czenczek <hreitz@redhat.com>
Message-Id: <20230824155345.109765-3-hreitz@redhat.com>
Reviewed-by: Sam Li <faithilikerun@gmail.com>
(cherry picked from commit 4b5d80f3d02096a9bb1f651f6b3401ba40877159)
Downstream: Fixed typo in the last hunk ("bs->blk" changed to
"bs->blk"). This code is removed in the commit after the
next, still, should be fixed now that I noticed.
Signed-off-by: Hanna Czenczek <hreitz@redhat.com>
---
block/file-posix.c | 12 +++++++-----
1 file changed, 7 insertions(+), 5 deletions(-)
diff --git a/block/file-posix.c b/block/file-posix.c
index 2b88b9eefa..9f14850e24 100644
--- a/block/file-posix.c
+++ b/block/file-posix.c
@@ -2455,9 +2455,10 @@ static int coroutine_fn raw_co_prw(BlockDriverState *bs, uint64_t offset,
if (fd_open(bs) < 0)
return -EIO;
#if defined(CONFIG_BLKZONED)
- if ((type & (QEMU_AIO_WRITE | QEMU_AIO_ZONE_APPEND)) && bs->wps) {
+ if ((type & (QEMU_AIO_WRITE | QEMU_AIO_ZONE_APPEND)) &&
+ bs->bl.zoned != BLK_Z_NONE) {
qemu_co_mutex_lock(&bs->wps->colock);
- if (type & QEMU_AIO_ZONE_APPEND && bs->bl.zone_size) {
+ if (type & QEMU_AIO_ZONE_APPEND) {
int index = offset / bs->bl.zone_size;
offset = bs->wps->wp[index];
}
@@ -2508,8 +2509,8 @@ out:
{
BlockZoneWps *wps = bs->wps;
if (ret == 0) {
- if ((type & (QEMU_AIO_WRITE | QEMU_AIO_ZONE_APPEND))
- && wps && bs->bl.zone_size) {
+ if ((type & (QEMU_AIO_WRITE | QEMU_AIO_ZONE_APPEND)) &&
+ bs->bl.zoned != BLK_Z_NONE) {
uint64_t *wp = &wps->wp[offset / bs->bl.zone_size];
if (!BDRV_ZT_IS_CONV(*wp)) {
if (type & QEMU_AIO_ZONE_APPEND) {
@@ -2529,7 +2530,8 @@ out:
}
}
- if ((type & (QEMU_AIO_WRITE | QEMU_AIO_ZONE_APPEND)) && wps) {
+ if ((type & (QEMU_AIO_WRITE | QEMU_AIO_ZONE_APPEND)) &&
+ bs->bl.zoned != BLK_Z_NONE) {
qemu_co_mutex_unlock(&wps->colock);
}
}
--
2.39.3