Re: [PATCH v6 7/8] blockdev: Return bs to the proper context on snapshot abort

[Kevin Wolf]

Am 08.01.2020 um 15:31 hat Sergio Lopez geschrieben:
> external_snapshot_abort() calls to bdrv_set_backing_hd(), which
> returns state->old_bs to the main AioContext, as it's intended to be
> used then the BDS is going to be released. As that's not the case when
> aborting an external snapshot, return it to the AioContext it was
> before the call.
> 
> This issue can be triggered by issuing a transaction with two actions,
> a proper blockdev-snapshot-sync and a bogus one, so the second will
> trigger a transaction abort. This results in a crash with an stack
> trace like this one:
> 
>  #0  0x00007fa1048b28df in __GI_raise (sig=sig@entry=6) at 
> ../sysdeps/unix/sysv/linux/raise.c:50
>  #1  0x00007fa10489ccf5 in __GI_abort () at abort.c:79
>  #2  0x00007fa10489cbc9 in __assert_fail_base
>      (fmt=0x7fa104a03300 "%s%s%s:%u: %s%sAssertion `%s' failed.\n%n", 
> assertion=0x5572240b44d8 "bdrv_get_aio_context(old_bs) == 
> bdrv_get_aio_context(new_bs)", file=0x557224014d30 "block.c", line=2240, 
> function=<optimized out>) at assert.c:92
>  #3  0x00007fa1048aae96 in __GI___assert_fail
>      (assertion=assertion@entry=0x5572240b44d8 "bdrv_get_aio_context(old_bs) 
> == bdrv_get_aio_context(new_bs)", file=file@entry=0x557224014d30 "block.c", 
> line=line@entry=2240, function=function@entry=0x5572240b5d60 
> <__PRETTY_FUNCTION__.31620> "bdrv_replace_child_noperm") at assert.c:101
>  #4  0x0000557223e631f8 in bdrv_replace_child_noperm (child=0x557225b9c980, 
> new_bs=new_bs@entry=0x557225c42e40) at block.c:2240
>  #5  0x0000557223e68be7 in bdrv_replace_node (from=0x557226951a60, 
> to=0x557225c42e40, errp=0x5572247d6138 <error_abort>) at block.c:4196
>  #6  0x0000557223d069c4 in external_snapshot_abort (common=0x557225d7e170) at 
> blockdev.c:1731
>  #7  0x0000557223d069c4 in external_snapshot_abort (common=0x557225d7e170) at 
> blockdev.c:1717
>  #8  0x0000557223d09013 in qmp_transaction (dev_list=<optimized out>, 
> has_props=<optimized out>, props=0x557225cc7d70, 
> errp=errp@entry=0x7ffe704c0c98) at blockdev.c:2360
>  #9  0x0000557223e32085 in qmp_marshal_transaction (args=<optimized out>, 
> ret=<optimized out>, errp=0x7ffe704c0d08) at 
> qapi/qapi-commands-transaction.c:44
>  #10 0x0000557223ee798c in do_qmp_dispatch (errp=0x7ffe704c0d00, 
> allow_oob=<optimized out>, request=<optimized out>, cmds=0x5572247d3cc0 
> <qmp_commands>) at qapi/qmp-dispatch.c:132
>  #11 0x0000557223ee798c in qmp_dispatch (cmds=0x5572247d3cc0 <qmp_commands>, 
> request=<optimized out>, allow_oob=<optimized out>) at qapi/qmp-dispatch.c:175
>  #12 0x0000557223e06141 in monitor_qmp_dispatch (mon=0x557225c69ff0, 
> req=<optimized out>) at monitor/qmp.c:120
>  #13 0x0000557223e0678a in monitor_qmp_bh_dispatcher (data=<optimized out>) 
> at monitor/qmp.c:209
>  #14 0x0000557223f2f366 in aio_bh_call (bh=0x557225b9dc60) at util/async.c:117
>  #15 0x0000557223f2f366 in aio_bh_poll (ctx=ctx@entry=0x557225b9c840) at 
> util/async.c:117
>  #16 0x0000557223f32754 in aio_dispatch (ctx=0x557225b9c840) at 
> util/aio-posix.c:459
>  #17 0x0000557223f2f242 in aio_ctx_dispatch (source=<optimized out>, 
> callback=<optimized out>, user_data=<optimized out>) at util/async.c:260
>  #18 0x00007fa10913467d in g_main_dispatch (context=0x557225c28e80) at 
> gmain.c:3176
>  #19 0x00007fa10913467d in g_main_context_dispatch 
> (context=context@entry=0x557225c28e80) at gmain.c:3829
>  #20 0x0000557223f31808 in glib_pollfds_poll () at util/main-loop.c:219
>  #21 0x0000557223f31808 in os_host_main_loop_wait (timeout=<optimized out>) 
> at util/main-loop.c:242
>  #22 0x0000557223f31808 in main_loop_wait (nonblocking=<optimized out>) at 
> util/main-loop.c:518
>  #23 0x0000557223d13201 in main_loop () at vl.c:1828
>  #24 0x0000557223bbfb82 in main (argc=<optimized out>, argv=<optimized out>, 
> envp=<optimized out>) at vl.c:4504
> 
> RHBZ: https://bugzilla.redhat.com/show_bug.cgi?id=1779036
> Signed-off-by: Sergio Lopez <address@hidden>
> ---
>  blockdev.c | 21 +++++++++++++++++++++
>  1 file changed, 21 insertions(+)
> 
> diff --git a/blockdev.c b/blockdev.c
> index 292961a88d..cfed87f434 100644
> --- a/blockdev.c
> +++ b/blockdev.c
> @@ -1731,6 +1731,8 @@ static void external_snapshot_abort(BlkActionState 
> *common)
>      if (state->new_bs) {
>          if (state->overlay_appended) {
>              AioContext *aio_context;
> +            AioContext *tmp_context;
> +            int ret;
>  
>              aio_context = bdrv_get_aio_context(state->old_bs);
>              aio_context_acquire(aio_context);
> @@ -1738,6 +1740,25 @@ static void external_snapshot_abort(BlkActionState 
> *common)
>              bdrv_ref(state->old_bs);   /* we can't let bdrv_set_backind_hd()
>                                            close state->old_bs; we need it */
>              bdrv_set_backing_hd(state->new_bs, NULL, &error_abort);
> +
> +            /*
> +             * The call to bdrv_set_backing_hd() above returns state->old_bs 
> to
> +             * the main AioContext. As we're still going to be using it, 
> return
> +             * it to the AioContext it was before.
> +             */
> +            tmp_context = bdrv_get_aio_context(state->old_bs);
> +            if (aio_context != tmp_context) {
> +                aio_context_release(aio_context);
> +                aio_context_acquire(tmp_context);
> +
> +                ret = bdrv_try_set_aio_context(state->old_bs,
> +                                               aio_context, NULL);
> +                assert(ret == 0);
> +
> +                aio_context_release(tmp_context);
> +                aio_context_acquire(aio_context);
> +            }
> +
>              bdrv_replace_node(state->new_bs, state->old_bs, &error_abort);
>              bdrv_unref(state->old_bs); /* bdrv_replace_node() ref'ed old_bs 
> */

Arguably, bdrv_replace_node() should share the AioContext-switching
logic with bdrv_root_attach_child() so that the general case is solved.

I guess this patch is simple enough and solves a relevant special case,
so no objection. But it might not be the final fix.

Kevin