On Fri, Feb 21, 2020 at 09:39:51AM +0000, Stefan Hajnoczi wrote:
> The ctx->first_bh list contains all created BHs, including those that
> are not scheduled.  The list is iterated by the event loop and therefore
> has O(n) time complexity with respected to the number of created BHs.
> 
> Rewrite BHs so that only scheduled or deleted BHs are enqueued.
> Only BHs that actually require action will be iterated.
> 
> One semantic change is required: qemu_bh_delete() enqueues the BH and
> therefore invokes aio_notify().  The
> tests/test-aio.c:test_source_bh_delete_from_cb() test case assumed that
> g_main_context_iteration(NULL, false) returns false after
> qemu_bh_delete() but it now returns true for one iteration.  Fix up the
> test case.
> 
> This patch makes aio_compute_timeout() and aio_bh_poll() drop from a CPU
> profile reported by perf-top(1).  Previously they combined to 9% CPU
> utilization when AioContext polling is commented out and the guest has 2
> virtio-blk,num-queues=1 and 99 virtio-blk,num-queues=32 devices.
> 
> Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
> ---
> v3:
>  * Use QSLIST_FOREACH_RCU() and QSLIST_FIRST_RCU() [Paolo]
> v2:
>  * Use QSLIST for BHs and QSIMPLEQ for BHListSlices [Paolo]
>    (Note that I replaced bh = atomic_rcu_read(&first_bh) with
>     QSLIST_FOREACH(&bh_list) so there is no memory ordering but I think
>     this is safe.)
>  * Comment clarifications [Paolo]
> 
> Based-on: 20200220103828.24525-1-pbonzini@redhat.com
>           ("[PATCH] rcu_queue: add QSLIST functions")
> ---
>  include/block/aio.h |  20 +++-
>  tests/test-aio.c    |   3 +-
>  util/async.c        | 237 ++++++++++++++++++++++++++------------------
>  3 files changed, 158 insertions(+), 102 deletions(-)

Thanks, applied to my block tree:
https://github.com/stefanha/qemu/commits/block

Stefan