diff options
author | Quinn Tran <quinn.tran@cavium.com> | 2016-12-24 03:06:14 +0100 |
---|---|---|
committer | Bart Van Assche <bart.vanassche@sandisk.com> | 2017-01-17 20:26:58 +0100 |
commit | 300af14bdb28157090f0c6f89d244fda940082da (patch) | |
tree | 75162cff9c3283438299e5fff5461bfcc25fe997 | |
parent | qla2xxx: Fix erroneous invalid handle message (diff) | |
download | linux-300af14bdb28157090f0c6f89d244fda940082da.tar.xz linux-300af14bdb28157090f0c6f89d244fda940082da.zip |
qla2xxx: Disable out-of-order processing by default in firmware
Out of order(OOO) processing requires initiator, switch
and target to support OOO. In today's environment, none
of the switches support OOO. OOO requires extra buffer
space which affect performance. By turning ON this feature
in QLogic's FW, it delays error recovery because dropped
frame is treated as out of order frame. We're turning OFF
this option of speed up error recovery.
Signed-off-by: Quinn Tran <quinn.tran@cavium.com>
Signed-off-by: Himanshu Madhani <himanshu.madhani@cavium.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
[ bvanassche: Fixed spelling in patch description ]
Signed-off-by: Bart Van Assche <bart.vanassche@sandisk.com>
-rw-r--r-- | drivers/scsi/qla2xxx/qla_target.c | 6 |
1 files changed, 0 insertions, 6 deletions
diff --git a/drivers/scsi/qla2xxx/qla_target.c b/drivers/scsi/qla2xxx/qla_target.c index 73ec6b764951..e4fda84b959e 100644 --- a/drivers/scsi/qla2xxx/qla_target.c +++ b/drivers/scsi/qla2xxx/qla_target.c @@ -6581,9 +6581,6 @@ qlt_24xx_config_nvram_stage1(struct scsi_qla_host *vha, struct nvram_24xx *nv) return; } - /* out-of-order frames reassembly */ - nv->firmware_options_3 |= BIT_6|BIT_9; - if (ha->tgt.enable_class_2) { if (vha->flags.init_done) fc_host_supported_classes(vha->host) = @@ -6685,9 +6682,6 @@ qlt_81xx_config_nvram_stage1(struct scsi_qla_host *vha, struct nvram_81xx *nv) return; } - /* out-of-order frames reassembly */ - nv->firmware_options_3 |= BIT_6|BIT_9; - if (ha->tgt.enable_class_2) { if (vha->flags.init_done) fc_host_supported_classes(vha->host) = |