verify: fix a bug with verify_async
There's a race between marking an io_u as deferred, completing it,
and checking of that flag. It cannot be done reliably for async
verify threads. So change the mechanism to have the verify_async
part pull the io_u completely, so we don't have to check for a flag
in it after we have run ->end_io().
This fixes a bug with verify_async, where fio would crash with
this message:
fio: io_u.c:1315: __get_io_u: Assertion `io_u->flags & IO_U_F_FREE' failed
This race has always existed, but was made considerably worse with
this commit:
commit 2ae0b204743d6b4048c6fffd46c6280a70f2ecd1
Author: Jens Axboe <axboe@kernel.dk>
Date: Tue May 28 14:16:55 2013 +0200
Replace list based free/busy/requeue list with FIFO + ring
Cache friendliness of the list is pretty low. This has
provably lower overhead.
since we moved from a single list holding the io, to a separate
verify list and io_u queues. The above bug is happening because
the io_u ends up on both the freelist and the pending verify list,
causing mayhem.
Reported-by: scameron@beardog.cce.hp.com
Signed-off-by: Jens Axboe <axboe@fb.com>
diff --git a/verify.c b/verify.c
index 7c99e15..217b686 100644
--- a/verify.c
+++ b/verify.c
@@ -656,19 +656,21 @@
/*
* Push IO verification to a separate thread
*/
-int verify_io_u_async(struct thread_data *td, struct io_u *io_u)
+int verify_io_u_async(struct thread_data *td, struct io_u **io_u_ptr)
{
- if (io_u->file)
- put_file_log(td, io_u->file);
+ struct io_u *io_u = *io_u_ptr;
pthread_mutex_lock(&td->io_u_lock);
+ if (io_u->file)
+ put_file_log(td, io_u->file);
+
if (io_u->flags & IO_U_F_IN_CUR_DEPTH) {
td->cur_depth--;
io_u->flags &= ~IO_U_F_IN_CUR_DEPTH;
}
flist_add_tail(&io_u->verify_list, &td->verify_list);
- io_u->flags |= IO_U_F_FREE_DEF;
+ *io_u_ptr = NULL;
pthread_mutex_unlock(&td->io_u_lock);
pthread_cond_signal(&td->verify_cond);
@@ -747,9 +749,10 @@
return EILSEQ;
}
-int verify_io_u(struct thread_data *td, struct io_u *io_u)
+int verify_io_u(struct thread_data *td, struct io_u **io_u_ptr)
{
struct verify_header *hdr;
+ struct io_u *io_u = *io_u_ptr;
unsigned int header_size, hdr_inc, hdr_num = 0;
void *p;
int ret;
@@ -1188,9 +1191,11 @@
while (!flist_empty(&list)) {
io_u = flist_first_entry(&list, struct io_u, verify_list);
- flist_del(&io_u->verify_list);
+ flist_del_init(&io_u->verify_list);
- ret = verify_io_u(td, io_u);
+ io_u->flags |= IO_U_F_NO_FILE_PUT;
+ ret = verify_io_u(td, &io_u);
+
put_io_u(td, io_u);
if (!ret)
continue;