Skip to contentSkip to conversation

nvmet: handle rw's limited retry flag

PATCH Guixin Liu 1 participant 2 messages Nov 26, 2024 – Dec 6, 2024 1 message not in the archive NVMe

lore
Share & tools
lore

A message that is not in the archive (sent off-list or not yet received): [email protected]

Guixin Liu

author
#1 lore raw

cc linux nvme mailing list, sorry for my oversight.

Best Regards,

Guixin Liu

Quoted 38 lines · Guixin Liu

In some scenarios, some multipath software setup places the REQ_FAILFAST_DEV flag on I/O to prevent retries and immediately switch to other paths for issuing I/O commands. This will reflect on the NVMe read and write commands with the limited retry flag.

However, the current NVMe target side does not handle the limited retry flag, and the target's underlying driver still retries the I/O. This will result in the I/O not being quickly switched to other paths, ultimately leading to increased I/O latency.

When the nvme target receive an rw command with limited retry flag, handle it in block backend by setting the REQ_FAILFAST_DEV flag to bio.

---
Changes from v1 to v2:
- Add details to commit body(Christoph).
- Add the rb tag from Chaitanya, thanks.

  drivers/nvme/target/io-cmd-bdev.c | 3 +++
  1 file changed, 3 insertions(+)
drivers/nvme/target/io-cmd-bdev.c +3 -0
index 0bda83d0fc3e..6380b60fd490 100644
@@ -272,6 +272,9 @@ static void nvmet_bdev_execute_rw(struct nvmet_req *req)
  		iter_flags = SG_MITER_FROM_SG;
  	}
 
+	if (req->cmd->rw.control & NVME_RW_LR)
+		opf |= REQ_FAILFAST_DEV;
+
  	if (is_pci_p2pdma_page(sg_page(req->sg)))
  		opf |= REQ_NOMERGE;
 
author
#2 lore raw
In reply to a message not in the archive

Hi Keith,

  Could you please apply this patch? thanks.

Best Regards,

Guixin Liu

Quoted 38 lines · Guixin Liu

In some scenarios, some multipath software setup places the REQ_FAILFAST_DEV flag on I/O to prevent retries and immediately switch to other paths for issuing I/O commands. This will reflect on the NVMe read and write commands with the limited retry flag.

However, the current NVMe target side does not handle the limited retry flag, and the target's underlying driver still retries the I/O. This will result in the I/O not being quickly switched to other paths, ultimately leading to increased I/O latency.

When the nvme target receive an rw command with limited retry flag, handle it in block backend by setting the REQ_FAILFAST_DEV flag to bio.

---
Changes from v1 to v2:
- Add details to commit body(Christoph).
- Add the rb tag from Chaitanya, thanks.

  drivers/nvme/target/io-cmd-bdev.c | 3 +++
  1 file changed, 3 insertions(+)
drivers/nvme/target/io-cmd-bdev.c +3 -0
index 0bda83d0fc3e..6380b60fd490 100644
@@ -272,6 +272,9 @@ static void nvmet_bdev_execute_rw(struct nvmet_req *req)
  		iter_flags = SG_MITER_FROM_SG;
  	}
 
+	if (req->cmd->rw.control & NVME_RW_LR)
+		opf |= REQ_FAILFAST_DEV;
+
  	if (is_pci_p2pdma_page(sg_page(req->sg)))
  		opf |= REQ_NOMERGE;