On Fri, Jun 09, 2017 at 10:41:13AM +0100, Stefan Hajnoczi wrote:
On Fri, Jun 09, 2017 at 02:22:55AM +0300, Manos Pitsidianakis wrote:diff --git a/hw/ide/core.c b/hw/ide/core.c index 0b48b64..1a74339 100644 --- a/hw/ide/core.c +++ b/hw/ide/core.c @@ -416,16 +416,16 @@ static void ide_issue_trim_cb(void *opaque, int ret) /* 6-byte LBA + 2-byte range per entry */ uint64_t entry = le64_to_cpu(buffer[i]); uint64_t sector = entry & 0x0000ffffffffffffULL; - uint16_t count = entry >> 48; + uint16_t bytes = entry >> 48;- if (count == 0) { + if (bytes == 0) { continue; } /* Got an entry! Submit and exit. */ iocb->aiocb = blk_aio_pdiscard(iocb->blk, sector << BDRV_SECTOR_BITS, - count << BDRV_SECTOR_BITS, + bytes << BDRV_SECTOR_BITS, ide_issue_trim_cb, opaque); return; }This change looks suspicious: bytes << BDRV_SECTOR_BITS Therefore 'bytes' is actually in units of sectors! Please drop this hunk.
:) Rookie mistake, thanks. (At some other places there's count >> BDRV_SECTOR_BITS, hope this change makes it clear) Will send a v3.
signature.asc
Description: PGP signature
