2020-08-08 23:51:10

by Sasha Levin

[permalink] [raw]
Subject: [PATCH AUTOSEL 5.8 70/72] nvme-multipath: do not fall back to __nvme_find_path() for non-optimized paths

From: Hannes Reinecke <[email protected]>

[ Upstream commit fbd6a42d8932e172921c7de10468a2e12c34846b ]

When nvme_round_robin_path() finds a valid namespace we should be using it;
falling back to __nvme_find_path() for non-optimized paths will cause the
result from nvme_round_robin_path() to be ignored for non-optimized paths.

Fixes: 75c10e732724 ("nvme-multipath: round-robin I/O policy")
Signed-off-by: Martin Wilck <[email protected]>
Signed-off-by: Hannes Reinecke <[email protected]>
Reviewed-by: Sagi Grimberg <[email protected]>
Signed-off-by: Christoph Hellwig <[email protected]>
Signed-off-by: Sasha Levin <[email protected]>
---
drivers/nvme/host/multipath.c | 11 +++++++----
1 file changed, 7 insertions(+), 4 deletions(-)

diff --git a/drivers/nvme/host/multipath.c b/drivers/nvme/host/multipath.c
index fe8f7f123fac7..57d51148e71b6 100644
--- a/drivers/nvme/host/multipath.c
+++ b/drivers/nvme/host/multipath.c
@@ -272,10 +272,13 @@ inline struct nvme_ns *nvme_find_path(struct nvme_ns_head *head)
struct nvme_ns *ns;

ns = srcu_dereference(head->current_path[node], &head->srcu);
- if (READ_ONCE(head->subsys->iopolicy) == NVME_IOPOLICY_RR && ns)
- ns = nvme_round_robin_path(head, node, ns);
- if (unlikely(!ns || !nvme_path_is_optimized(ns)))
- ns = __nvme_find_path(head, node);
+ if (unlikely(!ns))
+ return __nvme_find_path(head, node);
+
+ if (READ_ONCE(head->subsys->iopolicy) == NVME_IOPOLICY_RR)
+ return nvme_round_robin_path(head, node, ns);
+ if (unlikely(!nvme_path_is_optimized(ns)))
+ return __nvme_find_path(head, node);
return ns;
}

--
2.25.1


2020-08-10 15:38:51

by Martin Wilck

[permalink] [raw]
Subject: Re: [PATCH AUTOSEL 5.8 70/72] nvme-multipath: do not fall back to __nvme_find_path() for non-optimized paths

On Sat, 2020-08-08 at 19:35 -0400, Sasha Levin wrote:
> From: Hannes Reinecke <[email protected]>
>
> [ Upstream commit fbd6a42d8932e172921c7de10468a2e12c34846b ]
>
> When nvme_round_robin_path() finds a valid namespace we should be
> using it;
> falling back to __nvme_find_path() for non-optimized paths will cause
> the
> result from nvme_round_robin_path() to be ignored for non-optimized
> paths.
>
> Fixes: 75c10e732724 ("nvme-multipath: round-robin I/O policy")
> Signed-off-by: Martin Wilck <[email protected]>
> Signed-off-by: Hannes Reinecke <[email protected]>
> Reviewed-by: Sagi Grimberg <[email protected]>
> Signed-off-by: Christoph Hellwig <[email protected]>
> Signed-off-by: Sasha Levin <[email protected]>
> ---
> drivers/nvme/host/multipath.c | 11 +++++++----
> 1 file changed, 7 insertions(+), 4 deletions(-)

Hello Sasha,

this patch needs a fix that I recently submitted to linux-nvme.
The same holds for the respective 5.7 and 5.4 AUTOSEL patches.

http://lists.infradead.org/pipermail/linux-nvme/2020-August/018570.html

Regards,
Martin


2020-08-16 14:33:12

by Sasha Levin

[permalink] [raw]
Subject: Re: [PATCH AUTOSEL 5.8 70/72] nvme-multipath: do not fall back to __nvme_find_path() for non-optimized paths

On Mon, Aug 10, 2020 at 05:37:54PM +0200, Martin Wilck wrote:
>On Sat, 2020-08-08 at 19:35 -0400, Sasha Levin wrote:
>> From: Hannes Reinecke <[email protected]>
>>
>> [ Upstream commit fbd6a42d8932e172921c7de10468a2e12c34846b ]
>>
>> When nvme_round_robin_path() finds a valid namespace we should be
>> using it;
>> falling back to __nvme_find_path() for non-optimized paths will cause
>> the
>> result from nvme_round_robin_path() to be ignored for non-optimized
>> paths.
>>
>> Fixes: 75c10e732724 ("nvme-multipath: round-robin I/O policy")
>> Signed-off-by: Martin Wilck <[email protected]>
>> Signed-off-by: Hannes Reinecke <[email protected]>
>> Reviewed-by: Sagi Grimberg <[email protected]>
>> Signed-off-by: Christoph Hellwig <[email protected]>
>> Signed-off-by: Sasha Levin <[email protected]>
>> ---
>> drivers/nvme/host/multipath.c | 11 +++++++----
>> 1 file changed, 7 insertions(+), 4 deletions(-)
>
>Hello Sasha,
>
>this patch needs a fix that I recently submitted to linux-nvme.
>The same holds for the respective 5.7 and 5.4 AUTOSEL patches.
>
>http://lists.infradead.org/pipermail/linux-nvme/2020-August/018570.html

I'll grab it too, thanks!

--
Thanks,
Sasha