Re: [PATCH v2 2/2] virtio-blk: Convert QEMUBH callback to "bitops.h" API

["Philippe Mathieu-Daudé" <philmd@redhat.com>]

On 5/10/21 3:25 PM, Stefan Hajnoczi wrote:
> On Fri, May 07, 2021 at 07:06:34PM +0200, Philippe Mathieu-Daudé wrote:
>> By directly using find_first_bit() and find_next_bit from the
>> "bitops.h" API to iterate over the bitmap, we can remove the
>> bitmap[] variable-length array copy on the stack and the complex
>> manual bit testing/clearing logic.
>>
>> Suggested-by: Stefan Hajnoczi <stefanha@redhat.com>
>> Suggested-by: Richard Henderson <richard.henderson@linaro.org>
>> Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
>> ---
>>  hw/block/dataplane/virtio-blk.c | 19 ++++---------------
>>  1 file changed, 4 insertions(+), 15 deletions(-)
>>
>> diff --git a/hw/block/dataplane/virtio-blk.c b/hw/block/dataplane/virtio-blk.c
>> index e9050c8987e..a7b5bda06fc 100644
>> --- a/hw/block/dataplane/virtio-blk.c
>> +++ b/hw/block/dataplane/virtio-blk.c
>> @@ -60,23 +60,12 @@ static void notify_guest_bh(void *opaque)
>>  {
>>      VirtIOBlockDataPlane *s = opaque;
>>      unsigned nvqs = s->conf->num_queues;
>> -    unsigned long bitmap[BITS_TO_LONGS(nvqs)];
>> -    unsigned j;
>>  
>> -    memcpy(bitmap, s->batch_notify_vqs, sizeof(bitmap));
>> -    memset(s->batch_notify_vqs, 0, sizeof(bitmap));
>> +    for (unsigned long i = find_first_bit(s->batch_notify_vqs, nvqs);
>> +             i < nvqs; i = find_next_bit(s->batch_notify_vqs, nvqs, i)) {
>> +        VirtQueue *vq = virtio_get_queue(s->vdev, i);
>>  
>> -    for (j = 0; j < nvqs; j += BITS_PER_LONG) {
>> -        unsigned long bits = bitmap[j / BITS_PER_LONG];
>> -
>> -        while (bits != 0) {
>> -            unsigned i = j + ctzl(bits);
>> -            VirtQueue *vq = virtio_get_queue(s->vdev, i);
>> -
>> -            virtio_notify_irqfd(s->vdev, vq);
>> -
>> -            bits &= bits - 1; /* clear right-most bit */
>> -        }
>> +        virtio_notify_irqfd(s->vdev, vq);
>>      }
>>  }
> 
> Bits in s->batch_notify_vqs[] must be cleared. Otherwise we may raise
> spurious irqs next time this function is called. The memset() can be
> moved after the loop.

Doh... good catch! I missed it when removing the previous
test_and_clear_bit() call (from v1).