Re: [PATCH 4/4] async: use explicit memory barriers and relaxed accesses

[Paolo Bonzini]

On 07/04/20 11:09, Stefan Hajnoczi wrote:
>>      if (blocking) {
>> -        atomic_add(&ctx->notify_me, 2);
>> +        atomic_set(&ctx->notify_me, atomic_read(&ctx->notify_me) + 2);
> Non-atomic "atomic" code looks suspicious and warrants a comment
> mentioning that this is only executed from one thread.  This applies to
> the other instances in this patch too.

Yes, that's the patch that is missing from this series, which
strengthens the assertion to ensure that we're in the home thread.

Paolo

signature.asc
Description: OpenPGP digital signature