nvmet-fc: remove target cpu scheduling flag

Remove NVMET_FCTGTFEAT_NEEDS_CMD_CPUSCHED. It's unnecessary.

Signed-off-by: James Smart <james.smart@broadcom.com>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
This commit is contained in:
James Smart
2017-04-25 16:23:09 -07:00
committed by Jens Axboe
parent 2952a879ba
commit 4b8ba5fa52
4 changed files with 3 additions and 15 deletions

View File

@@ -642,15 +642,7 @@ enum {
* sequence in one LLDD operation. Errors during Data
* sequence transmit must not allow RSP sequence to be sent.
*/
NVMET_FCTGTFEAT_NEEDS_CMD_CPUSCHED = (1 << 1),
/* Bit 1: When 0, the LLDD will deliver FCP CMD
* on the CPU it should be affinitized to. Thus work will
* be scheduled on the cpu received on. When 1, the LLDD
* may not deliver the CMD on the CPU it should be worked
* on. The transport should pick a cpu to schedule the work
* on.
*/
NVMET_FCTGTFEAT_CMD_IN_ISR = (1 << 2),
NVMET_FCTGTFEAT_CMD_IN_ISR = (1 << 1),
/* Bit 2: When 0, the LLDD is calling the cmd rcv handler
* in a non-isr context, allowing the transport to finish
* op completion in the calling context. When 1, the LLDD
@@ -658,7 +650,7 @@ enum {
* requiring the transport to transition to a workqueue
* for op completion.
*/
NVMET_FCTGTFEAT_OPDONE_IN_ISR = (1 << 3),
NVMET_FCTGTFEAT_OPDONE_IN_ISR = (1 << 2),
/* Bit 3: When 0, the LLDD is calling the op done handler
* in a non-isr context, allowing the transport to finish
* op completion in the calling context. When 1, the LLDD