[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[PATCH 4/6] test-bdrv-drain.c: adapt test to the coming subtree drains
From: |
Emanuele Giuseppe Esposito |
Subject: |
[PATCH 4/6] test-bdrv-drain.c: adapt test to the coming subtree drains |
Date: |
Tue, 8 Feb 2022 10:36:53 -0500 |
There will be 2 problems in this test when we will add
subtree drains in bdrv_replace_child_noperm:
- First, the test is inconsistent about taking the AioContext lock when
calling bdrv_replace_child_noperm. bdrv_replace_child_noperm is reached
in two places: from blk_insert_bs directly, and via block_job_create.
Only the second does it with the AioContext lock taken, and there seems
to be no reason why the lock is needed.
Move aio_context_acquire further down, to just protect block_job_add_bdrv()
- Second, test_detach_indirect is only interested in observing the first
call to .drained_begin. In the original test, there was only a single
subtree drain; however, with additional drains introduced in
bdrv_replace_child_noperm(), the test callback would be called too early
and/or multiple times.
Override the callback only when we actually want to use it, and put back
the original after it's been invoked.
Signed-off-by: Emanuele Giuseppe Esposito <eesposit@redhat.com>
---
tests/unit/test-bdrv-drain.c | 17 ++++++++++++-----
1 file changed, 12 insertions(+), 5 deletions(-)
diff --git a/tests/unit/test-bdrv-drain.c b/tests/unit/test-bdrv-drain.c
index 4924ceb562..c52ba2db4e 100644
--- a/tests/unit/test-bdrv-drain.c
+++ b/tests/unit/test-bdrv-drain.c
@@ -912,12 +912,12 @@ static void test_blockjob_common_drain_node(enum
drain_type drain_type,
blk_insert_bs(blk_target, target, &error_abort);
blk_set_allow_aio_context_change(blk_target, true);
- aio_context_acquire(ctx);
tjob = block_job_create("job0", &test_job_driver, NULL, src,
0, BLK_PERM_ALL,
0, 0, NULL, NULL, &error_abort);
tjob->bs = src;
job = &tjob->common;
+ aio_context_acquire(ctx);
block_job_add_bdrv(job, "target", target, 0, BLK_PERM_ALL, &error_abort);
switch (result) {
@@ -1342,15 +1342,18 @@ static void detach_by_parent_aio_cb(void *opaque, int
ret)
}
}
+static BdrvChildClass detach_by_driver_cb_class;
+
static void detach_by_driver_cb_drained_begin(BdrvChild *child)
{
+ /* restore .drained_begin cb, we don't need it anymore. */
+ detach_by_driver_cb_class.drained_begin = child_of_bds.drained_begin;
+
aio_bh_schedule_oneshot(qemu_get_current_aio_context(),
detach_indirect_bh, &detach_by_parent_data);
child_of_bds.drained_begin(child);
}
-static BdrvChildClass detach_by_driver_cb_class;
-
/*
* Initial graph:
*
@@ -1382,8 +1385,6 @@ static void test_detach_indirect(bool by_parent_cb)
if (!by_parent_cb) {
detach_by_driver_cb_class = child_of_bds;
- detach_by_driver_cb_class.drained_begin =
- detach_by_driver_cb_drained_begin;
}
/* Create all involved nodes */
@@ -1441,6 +1442,12 @@ static void test_detach_indirect(bool by_parent_cb)
acb = blk_aio_preadv(blk, 0, &qiov, 0, detach_by_parent_aio_cb, NULL);
g_assert(acb != NULL);
+ if (!by_parent_cb) {
+ /* set .drained_begin cb to run only in the following drain. */
+ detach_by_driver_cb_class.drained_begin =
+ detach_by_driver_cb_drained_begin;
+ }
+
/* Drain and check the expected result */
bdrv_subtree_drained_begin(parent_b);
--
2.31.1
- [PATCH 1/6] block/io.c: fix bdrv_child_cb_drained_begin invocations from a coroutine, (continued)
[PATCH 2/6] block.c: bdrv_replace_child_noperm: first remove the child, and then call ->detach(), Emanuele Giuseppe Esposito, 2022/02/08
[PATCH 4/6] test-bdrv-drain.c: adapt test to the coming subtree drains,
Emanuele Giuseppe Esposito <=
[PATCH 6/6] jobs: ensure sleep in job_sleep_ns is fully performed, Emanuele Giuseppe Esposito, 2022/02/08
[PATCH 3/6] block.c: bdrv_replace_child_noperm: first call ->attach(), and then add child, Emanuele Giuseppe Esposito, 2022/02/08
[PATCH 5/6] test-bdrv-drain.c: remove test_detach_by_parent_cb(), Emanuele Giuseppe Esposito, 2022/02/08