nvme: core: shorten duration of multipath namespace rescan
PATCH v2 1/1 nvme: core: shorten duration of multipath namespace rescan
For multipath devices, nvme_update_ns_info() needs to freeze both the queue of the path and the queue of the multipath device. For both operations, it waits for one RCU grace period to pass, ~25ms on my test system. By calling blk_freeze_queue_start() for the multipath queue early, we avoid waiting twice; tests using ftrace have shown that the second blk_mq_freeze_queue_wait() call finishes in just a few microseconds. The path queue is unfrozen before calling blk_mq_freeze_queue_wait() on the multipath queue, so that possibly outstanding IO in the multipath queue can be flushed.
I tested this using the "controller rescan under I/O load" test I submitted recently [1].
[1] https://lore.kernel.org/linux-nvme/[email protected]/T/#u
---
v2: (all changes suggested by Sagi Grimberg)
- patch subject changed from "nvme: core: freeze multipath queue early in nvme_update_ns_info()" to "nvme: core: shorten duration of multipath namespace rescan"
- inserted comment explaining why blk_freeze_queue_start() is called early
- wait for queue to be frozen even if ret != 0
- make code structure more obvious vs. freeze_start / freeze_wait / unfreeze
Hannes and Daniel had already added Reviewed-by: tags to the v1 patch, but I didn't add them above, because the patch looks quite different now.
--- drivers/nvme/host/core.c | 21 ++++++++++++++++----- 1 file changed, 16 insertions(+), 5 deletions(-)
drivers/nvme/host/core.c +16 -5
@@ -2217,6 +2217,16 @@ static int nvme_update_ns_info(struct nvme_ns *ns, struct nvme_ns_info *info)bool unsupported = false;int ret; +/* +* The controller queue is going to be frozen in +* nvme_update_ns_info_{generic,block}(). Every freeze implies waiting +* for an RCU grace period to pass. For multipath devices, we +* need to freeze the multipath queue, too. Start freezing the +* multipath queue now, lest we need to wait for two grace periods. +*/ +if (nvme_ns_head_multipath (ns->head)) +blk_freeze_queue_start (ns->head->disk->queue); +switch (info->ids.csi) {case NVME_CSI_ZNS :if (!IS_ENABLED (CONFIG_BLK_DEV_ZONED )) {
@@ -2250,11 +2260,13 @@ static int nvme_update_ns_info(struct nvme_ns *ns, struct nvme_ns_info *info) ret =0 ; } -if (!ret &&nvme_ns_head_multipath (ns->head)) { +if (!nvme_ns_head_multipath (ns->head)) +return ret; + +blk_mq_freeze_queue_wait (ns->head->disk->queue); +if (!ret) {struct queue_limits *ns_lim = &ns->disk->queue->limits;struct queue_limits lim; - -blk_mq_freeze_queue (ns->head->disk->queue);/* * queue_limits mixes values that are the hardware limitations * for bio splitting with what is the device configuration.
@@ -2286,9 +2298,8 @@ static int nvme_update_ns_info(struct nvme_ns *ns, struct nvme_ns_info *info)set_capacity_and_notify (ns->head->disk,get_capacity (ns->disk));set_disk_ro (ns->head->disk,nvme_ns_is_readonly (ns, info));nvme_mpath_revalidate_paths (ns); - -blk_mq_unfreeze_queue (ns->head->disk->queue); } +blk_mq_unfreeze_queue (ns->head->disk->queue);return ret; }
Signature · 1 line
2.46.0
Sagi Grimberg
maintainerQuoted 16 lines · Martin Wilck
For multipath devices,
nvme_update_ns_info()needs to freeze both the queue of the path and the queue of the multipath device. For both operations, it waits for one RCU grace period to pass, ~25ms on my test system. By callingblk_freeze_queue_start()for the multipath queue early, we avoid waiting twice; tests using ftrace have shown that the secondblk_mq_freeze_queue_wait()call finishes in just a few microseconds. The path queue is unfrozen before callingblk_mq_freeze_queue_wait()on the multipath queue, so that possibly outstanding IO in the multipath queue can be flushed.I tested this using the "controller rescan under I/O load" test I submitted recently [1].
[1] https://lore.kernel.org/linux-nvme/[email protected]/T/#u
Signed-off-by Martin Wilck <[email protected]>
Quoted 8 lines
---
v2: (all changes suggested by Sagi Grimberg)
- patch subject changed from "nvme: core: freeze multipath queue early innvme_update_ns_info()" to "nvme: core: shorten duration of multipath namespace rescan"
- inserted comment explaining whyblk_freeze_queue_start()is called early
- wait for queue to be frozen even if ret != 0
- make code structure more obvious vs.freeze_start/freeze_wait/ unfreeze
This looks a lot better, thanks.
Quoted 47 lines
Hannes and Daniel had already added Reviewed-by: tags to the v1 patch, but I didn't add them above, because the patch looks quite different now.
--- drivers/nvme/host/core.c | 21 ++++++++++++++++----- 1 file changed, 16 insertions(+), 5 deletions(-)drivers/nvme/host/core.c +15
-3@@ -2217,6 +2217,16 @@ static int nvme_update_ns_info(struct nvme_ns *ns, struct nvme_ns_info *info)bool unsupported = false;int ret; +/* +* The controller queue is going to be frozen in +* nvme_update_ns_info_{generic,block}(). Every freeze implies waiting +* for an RCU grace period to pass. For multipath devices, we +* need to freeze the multipath queue, too. Start freezing the +* multipath queue now, lest we need to wait for two grace periods. +*/ +if (nvme_ns_head_multipath (ns->head)) +blk_freeze_queue_start (ns->head->disk->queue); +switch (info->ids.csi) {case NVME_CSI_ZNS :if (!IS_ENABLED (CONFIG_BLK_DEV_ZONED )) {@@ -2250,11 +2260,13 @@ static int nvme_update_ns_info(struct nvme_ns *ns, struct nvme_ns_info *info) ret =0 ; } -if (!ret &&nvme_ns_head_multipath (ns->head)) { +if (!nvme_ns_head_multipath (ns->head)) +return ret; + +blk_mq_freeze_queue_wait (ns->head->disk->queue); +if (!ret) {struct queue_limits *ns_lim = &ns->disk->queue->limits;struct queue_limits lim; - -blk_mq_freeze_queue (ns->head->disk->queue);/* * queue_limits mixes values that are the hardware limitations * for bio splitting with what is the device configuration.
drivers/nvme/host/core.c static int nvme_update_ns_info(struct nvme_ns *ns, struct nvme_ns_info *info) · line 2302
Quoted 7 lines
drivers/nvme/host/core.c +1
-2@@ -2286,9 +2298,8 @@ static int nvme_update_ns_info(struct nvme_ns *ns, struct nvme_ns_info *info)set_capacity_and_notify (ns->head->disk,get_capacity (ns->disk));set_disk_ro (ns->head->disk,nvme_ns_is_readonly (ns, info));nvme_mpath_revalidate_paths (ns); - -blk_mq_unfreeze_queue (ns->head->disk->queue); } +blk_mq_unfreeze_queue (ns->head->disk->queue);
I'd even make it nicer with adding an out label and reverse the ret condition polarity.
--blk_mq_freeze_queue_wait(ns->head->disk->queue); if (ret) goto out; ... ... out: blk_mq_unfreeze_queue(ns->head->disk->queue); return ret;
--
Sagi Grimberg
maintainerQuoted 5 lines
Signature · 4 lines
-- blk_mq_freeze_queue_wait(ns->head->disk->queue); if (ret) goto out; ... ... out: blk_mq_unfreeze_queue(ns->head->disk->queue); return ret; --
Not sure what happened here....
Quoted 3 lines · Sagi Grimberg
I'd even make it nicer with adding an out label and reverse the ret condition polarity.
Ok, if you prefer that ... I thought you might like the if block better, as the goto variant comes with lots of whitespace changes. I'll send a v3.
Thanks
Martin
Quoted 25 lines · Martin Wilck
For multipath devices,
nvme_update_ns_info()needs to freeze both the queue of the path and the queue of the multipath device. For both operations, it waits for one RCU grace period to pass, ~25ms on my test system. By callingblk_freeze_queue_start()for the multipath queue early, we avoid waiting twice; tests using ftrace have shown that the secondblk_mq_freeze_queue_wait()call finishes in just a few microseconds. The path queue is unfrozen before callingblk_mq_freeze_queue_wait()on the multipath queue, so that possibly outstanding IO in the multipath queue can be flushed.I tested this using the "controller rescan under I/O load" test I submitted recently [1].
[1] https://lore.kernel.org/linux-nvme/[email protected]/T/#u
Signed-off-by Martin Wilck <[email protected]>---
v2: (all changes suggested by Sagi Grimberg)
- patch subject changed from "nvme: core: freeze multipath queue early innvme_update_ns_info()" to "nvme: core: shorten duration of multipath namespace rescan"
- inserted comment explaining whyblk_freeze_queue_start()is called early
- wait for queue to be frozen even if ret != 0
- make code structure more obvious vs.freeze_start/freeze_wait/ unfreeze
Quoted 7 lines
Hannes and Daniel had already added Reviewed-by: tags to the v1 patch, but I didn't add them above, because the patch looks quite different now.
--- drivers/nvme/host/core.c | 21 ++++++++++++++++----- 1 file changed, 16 insertions(+), 5 deletions(-)
Cheers,
Hannes
Signature · 4 lines
Dr. Hannes Reinecke Kernel Storage Architect [email protected] +49 911 74053 688 SUSE Software Solutions GmbH, Frankenstr. 146, 90461 Nürnberg HRB 36809 (AG Nürnberg), GF: I. Totev, A. McDonald, W. Knoblich