Jamie Lokier <jamie@xxxxxxxxxxxxx> writes:
> David Woodhouse wrote:
>> > Currently it seems we schedule data direction as READ (no flags).
>> > Although I'm not sure, we might want to schedule it as WRITE? (I guess
>> > WRITE can be allowed delay more than READ...)
>>
>> I'm not sure it matters much -- this is mostly going to be used on
>> devices which aren't affected by I/O scheduling at all and should
>> probably be using the no-op scheduler.
[...]
> you wouldn't want them to affect latency-sensitive READ.
Yes. My concern is like it. If it's READ and it's passed to device
immediately like sync, I guess there is impact to userland throughput.
[...]
@@ -1489,6 +1489,8 @@ void submit_bio(int rw, struct bio *bio)
{
int count = bio_sectors(bio);
+ if (rw & (1 << BIO_RW_DISCARD))
+ rw |= WRITE;
bio->bi_rw |= rw;
Or another options is caller should do? e.g. (WRITE | (1 << BIO_RW_DISCARD))
--
OGAWA Hirofumi <hirofumi@xxxxxxxxxxxxxxxxxx>
--
To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
