From ab03a61c66149327e022bdafa5843c6f82be267e Mon Sep 17 00:00:00 2001 From: Uday Shankar Date: Thu, 29 May 2025 17:47:10 -0600 Subject: ublk: have a per-io daemon instead of a per-queue daemon Currently, ublk_drv associates to each hardware queue (hctx) a unique task (called the queue's ubq_daemon) which is allowed to issue COMMIT_AND_FETCH commands against the hctx. If any other task attempts to do so, the command fails immediately with EINVAL. When considered together with the block layer architecture, the result is that for each CPU C on the system, there is a unique ublk server thread which is allowed to handle I/O submitted on CPU C. This can lead to suboptimal performance under imbalanced load generation. For an extreme example, suppose all the load is generated on CPUs mapping to a single ublk server thread. Then that thread may be fully utilized and become the bottleneck in the system, while other ublk server threads are totally idle. This issue can also be addressed directly in the ublk server without kernel support by having threads dequeue I/Os and pass them around to ensure even load. But this solution requires inter-thread communication at least twice for each I/O (submission and completion), which is generally a bad pattern for performance. The problem gets even worse with zero copy, as more inter-thread communication would be required to have the buffer register/unregister calls to come from the correct thread. Therefore, address this issue in ublk_drv by allowing each I/O to have its own daemon task. Two I/Os in the same queue are now allowed to be serviced by different daemon tasks - this was not possible before. Imbalanced load can then be balanced across all ublk server threads by having the ublk server threads issue FETCH_REQs in a round-robin manner. As a small toy example, consider a system with a single ublk device having 2 queues, each of depth 4. A ublk server having 4 threads could issue its FETCH_REQs against this device as follows (where each entry is the qid,tag pair that the FETCH_REQ targets): ublk server thread: T0 T1 T2 T3 0,0 0,1 0,2 0,3 1,3 1,0 1,1 1,2 This setup allows for load that is concentrated on one hctx/ublk_queue to be spread out across all ublk server threads, alleviating the issue described above. Add the new UBLK_F_PER_IO_DAEMON feature to ublk_drv, which ublk servers can use to essentially test for the presence of this change and tailor their behavior accordingly. Signed-off-by: Uday Shankar Reviewed-by: Caleb Sander Mateos Reviewed-by: Ming Lei Reviewed-by: Jens Axboe Link: https://lore.kernel.org/r/20250529-ublk_task_per_io-v8-1-e9d3b119336a@purestorage.com Signed-off-by: Jens Axboe --- include/uapi/linux/ublk_cmd.h | 9 +++++++++ 1 file changed, 9 insertions(+) (limited to 'include') diff --git a/include/uapi/linux/ublk_cmd.h b/include/uapi/linux/ublk_cmd.h index 56c7e3fc666f..77d9d6af46da 100644 --- a/include/uapi/linux/ublk_cmd.h +++ b/include/uapi/linux/ublk_cmd.h @@ -272,6 +272,15 @@ */ #define UBLK_F_QUIESCE (1ULL << 12) +/* + * If this feature is set, ublk_drv supports each (qid,tag) pair having + * its own independent daemon task that is responsible for handling it. + * If it is not set, daemons are per-queue instead, so for two pairs + * (qid1,tag1) and (qid2,tag2), if qid1 == qid2, then the same task must + * be responsible for handling (qid1,tag1) and (qid2,tag2). + */ +#define UBLK_F_PER_IO_DAEMON (1ULL << 13) + /* device state */ #define UBLK_S_DEV_DEAD 0 #define UBLK_S_DEV_LIVE 1 -- cgit v1.2.3-59-g8ed1b From 10f4a7cd724e34b7a6ff96e57ac49dc0cadececc Mon Sep 17 00:00:00 2001 From: Keith Busch Date: Tue, 20 May 2025 13:20:37 -0700 Subject: nvme: fix command limits status code The command specific status code, 0x183, was introduced in the NVMe 2.0 specification defined to "Command Size Limits Exceeded" and only ever applied to DSM and Copy commands. Fix the name and, remove the incorrect translation to error codes and special treatment in the target code for it. Fixes: 3b7c33b28a44d4 ("nvme.h: add Write Zeroes definitions") Cc: Chaitanya Kulkarni Reviewed-by: Chaitanya Kulkarni Signed-off-by: Christoph Hellwig --- drivers/nvme/host/constants.c | 2 +- drivers/nvme/host/core.c | 1 - drivers/nvme/host/pr.c | 2 -- drivers/nvme/target/core.c | 9 +-------- drivers/nvme/target/io-cmd-bdev.c | 9 +-------- include/linux/nvme.h | 2 +- 6 files changed, 4 insertions(+), 21 deletions(-) (limited to 'include') diff --git a/drivers/nvme/host/constants.c b/drivers/nvme/host/constants.c index 2b9e6cfaf2a8..1a0058be5821 100644 --- a/drivers/nvme/host/constants.c +++ b/drivers/nvme/host/constants.c @@ -145,7 +145,7 @@ static const char * const nvme_statuses[] = { [NVME_SC_BAD_ATTRIBUTES] = "Conflicting Attributes", [NVME_SC_INVALID_PI] = "Invalid Protection Information", [NVME_SC_READ_ONLY] = "Attempted Write to Read Only Range", - [NVME_SC_ONCS_NOT_SUPPORTED] = "ONCS Not Supported", + [NVME_SC_CMD_SIZE_LIM_EXCEEDED ] = "Command Size Limits Exceeded", [NVME_SC_ZONE_BOUNDARY_ERROR] = "Zoned Boundary Error", [NVME_SC_ZONE_FULL] = "Zone Is Full", [NVME_SC_ZONE_READ_ONLY] = "Zone Is Read Only", diff --git a/drivers/nvme/host/core.c b/drivers/nvme/host/core.c index f69a232a000a..7ba35b573335 100644 --- a/drivers/nvme/host/core.c +++ b/drivers/nvme/host/core.c @@ -290,7 +290,6 @@ static blk_status_t nvme_error_status(u16 status) case NVME_SC_NS_NOT_READY: return BLK_STS_TARGET; case NVME_SC_BAD_ATTRIBUTES: - case NVME_SC_ONCS_NOT_SUPPORTED: case NVME_SC_INVALID_OPCODE: case NVME_SC_INVALID_FIELD: case NVME_SC_INVALID_NS: diff --git a/drivers/nvme/host/pr.c b/drivers/nvme/host/pr.c index cf2d2c5039dd..ca6a74607b13 100644 --- a/drivers/nvme/host/pr.c +++ b/drivers/nvme/host/pr.c @@ -82,8 +82,6 @@ static int nvme_status_to_pr_err(int status) return PR_STS_SUCCESS; case NVME_SC_RESERVATION_CONFLICT: return PR_STS_RESERVATION_CONFLICT; - case NVME_SC_ONCS_NOT_SUPPORTED: - return -EOPNOTSUPP; case NVME_SC_BAD_ATTRIBUTES: case NVME_SC_INVALID_OPCODE: case NVME_SC_INVALID_FIELD: diff --git a/drivers/nvme/target/core.c b/drivers/nvme/target/core.c index db7b17d1094e..e824f149746a 100644 --- a/drivers/nvme/target/core.c +++ b/drivers/nvme/target/core.c @@ -62,14 +62,7 @@ inline u16 errno_to_nvme_status(struct nvmet_req *req, int errno) return NVME_SC_LBA_RANGE | NVME_STATUS_DNR; case -EOPNOTSUPP: req->error_loc = offsetof(struct nvme_common_command, opcode); - switch (req->cmd->common.opcode) { - case nvme_cmd_dsm: - case nvme_cmd_write_zeroes: - return NVME_SC_ONCS_NOT_SUPPORTED | NVME_STATUS_DNR; - default: - return NVME_SC_INVALID_OPCODE | NVME_STATUS_DNR; - } - break; + return NVME_SC_INVALID_OPCODE | NVME_STATUS_DNR; case -ENODATA: req->error_loc = offsetof(struct nvme_rw_command, nsid); return NVME_SC_ACCESS_DENIED; diff --git a/drivers/nvme/target/io-cmd-bdev.c b/drivers/nvme/target/io-cmd-bdev.c index 83be0657e6df..1cfa13d029bf 100644 --- a/drivers/nvme/target/io-cmd-bdev.c +++ b/drivers/nvme/target/io-cmd-bdev.c @@ -145,15 +145,8 @@ u16 blk_to_nvme_status(struct nvmet_req *req, blk_status_t blk_sts) req->error_loc = offsetof(struct nvme_rw_command, slba); break; case BLK_STS_NOTSUPP: + status = NVME_SC_INVALID_OPCODE | NVME_STATUS_DNR; req->error_loc = offsetof(struct nvme_common_command, opcode); - switch (req->cmd->common.opcode) { - case nvme_cmd_dsm: - case nvme_cmd_write_zeroes: - status = NVME_SC_ONCS_NOT_SUPPORTED | NVME_STATUS_DNR; - break; - default: - status = NVME_SC_INVALID_OPCODE | NVME_STATUS_DNR; - } break; case BLK_STS_MEDIUM: status = NVME_SC_ACCESS_DENIED; diff --git a/include/linux/nvme.h b/include/linux/nvme.h index 51308f65b72f..b65a1b9f2116 100644 --- a/include/linux/nvme.h +++ b/include/linux/nvme.h @@ -2171,7 +2171,7 @@ enum { NVME_SC_BAD_ATTRIBUTES = 0x180, NVME_SC_INVALID_PI = 0x181, NVME_SC_READ_ONLY = 0x182, - NVME_SC_ONCS_NOT_SUPPORTED = 0x183, + NVME_SC_CMD_SIZE_LIM_EXCEEDED = 0x183, /* * I/O Command Set Specific - Fabrics commands: -- cgit v1.2.3-59-g8ed1b