We use REQ_TYPE_BLOCK_PC for flush requests from file systems. scsi-ml's definition of REQ_TYPE_BLOCK_PC is that we don't retry PC requests even when we can (e.g. UNIT ATTENTION) and we send the response to the callers, and expect the callers to decide what they want. File systems (and the block layer) don't have the mechnism to retry requests so file systems hit avoidable I/O errors. We need a workaround such as the commit 77a4229719e511a0d38d9c355317ae1469adeb54 to retry BLOCK_PC flush requests. We will need the similar workaround for discard requests too since SCSI-ml handle them as BLOCK_PC internally. This uses REQ_TYPE_FS for flush requests from file systems instead of REQ_TYPE_BLOCK_PC. flush requests are retried when appropriate. This patchset can be applied to the block tree's for-2.6.36. It contains of some changes to scsi so this patchset can't be applied to scsi-misc. = block/blk-barrier.c | 4 ++-- drivers/scsi/scsi_error.c | 19 ++++--------------- drivers/scsi/sd.c | 19 +++++++++++++------ include/linux/bio.h | 2 ++ 4 files changed, 21 insertions(+), 23 deletions(-) -- To unsubscribe from this list: send the line "unsubscribe linux-scsi" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
