* [PATCH] nvme: fix possible initialization failure for P4500/P4600 SSDs
@ 2021-06-17 16:05 Zheng Wu
2021-12-10 17:31 ` Keith Busch
0 siblings, 1 reply; 5+ messages in thread
From: Zheng Wu @ 2021-06-17 16:05 UTC (permalink / raw)
To: linux-nvme; +Cc: Wu Zheng
From: Wu Zheng <wu.zheng@intel.com>
Subnqn isn't supported in P4500/P4600 SSDs.
It's a fixed unavailable value in each P4500/P4600 SSD.
Multiple SSDs need to return the different subnqn when SSDs are initialization.
To P4500/P4600 SSDs, in nvme_init_subnqn, id->subnqn is not null and
it's from P4500/P4600. However, multiple P4500/P4600 SSDs return the same subnqn.
It causes the possible initialization failure.
To fix this, the flag of NVME_QUIRK_IGNORE_DEV_SUBNQN need to be added to
driver_data of P4500/P4600 for nvme_id_table.
Signed-off-by: Zheng Wu <wu.zheng@intel.com>
---
drivers/nvme/host/pci.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c
index ca2ee80..953ea3d 100644
--- a/drivers/nvme/host/pci.c
+++ b/drivers/nvme/host/pci.c
@@ -3326,7 +3326,8 @@ static const struct pci_device_id nvme_id_table[] = {
NVME_QUIRK_DEALLOCATE_ZEROES, },
{ PCI_VDEVICE(INTEL, 0x0a54), /* Intel P4500/P4600 */
.driver_data = NVME_QUIRK_STRIPE_SIZE |
- NVME_QUIRK_DEALLOCATE_ZEROES, },
+ NVME_QUIRK_DEALLOCATE_ZEROES |
+ NVME_QUIRK_IGNORE_DEV_SUBNQN, },
{ PCI_VDEVICE(INTEL, 0x0a55), /* Dell Express Flash P4600 */
.driver_data = NVME_QUIRK_STRIPE_SIZE |
NVME_QUIRK_DEALLOCATE_ZEROES, },
--
2.7.4
^ permalink raw reply related [flat|nested] 5+ messages in thread* Re: [PATCH] nvme: fix possible initialization failure for P4500/P4600 SSDs
2021-06-17 16:05 [PATCH] nvme: fix possible initialization failure for P4500/P4600 SSDs Zheng Wu
@ 2021-12-10 17:31 ` Keith Busch
0 siblings, 0 replies; 5+ messages in thread
From: Keith Busch @ 2021-12-10 17:31 UTC (permalink / raw)
To: Zheng Wu; +Cc: linux-nvme
On Thu, Jun 17, 2021 at 12:05:52PM -0400, Zheng Wu wrote:
> From: Wu Zheng <wu.zheng@intel.com>
>
> Subnqn isn't supported in P4500/P4600 SSDs.
> It's a fixed unavailable value in each P4500/P4600 SSD.
According to ark.intel.com, P4500/4600 were released nearly 4 years ago.
My understanding was these were a popular product. Why are we hearing
about this issue only now? Is this issue specific to a particular
firmware version, or has Linux really been incompatible with these this
entire time?
> @@ -3326,7 +3326,8 @@ static const struct pci_device_id nvme_id_table[] = {
> NVME_QUIRK_DEALLOCATE_ZEROES, },
> { PCI_VDEVICE(INTEL, 0x0a54), /* Intel P4500/P4600 */
> .driver_data = NVME_QUIRK_STRIPE_SIZE |
> - NVME_QUIRK_DEALLOCATE_ZEROES, },
> + NVME_QUIRK_DEALLOCATE_ZEROES |
> + NVME_QUIRK_IGNORE_DEV_SUBNQN, },
> { PCI_VDEVICE(INTEL, 0x0a55), /* Dell Express Flash P4600 */
> .driver_data = NVME_QUIRK_STRIPE_SIZE |
> NVME_QUIRK_DEALLOCATE_ZEROES, },
So the generic P4600 is broken, but Dell's is not?
^ permalink raw reply [flat|nested] 5+ messages in thread
* [PATCH] nvme: fix possible initialization failure for P4500/P4600 SSDs
@ 2021-06-21 23:07 Zheng Wu
2022-01-10 16:03 ` Keith Busch
2022-01-26 16:14 ` Christoph Hellwig
0 siblings, 2 replies; 5+ messages in thread
From: Zheng Wu @ 2021-06-21 23:07 UTC (permalink / raw)
To: linux-nvme; +Cc: Wu Zheng, Ye Jinhe
From: Wu Zheng <wu.zheng@intel.com>
Subnqn isn't supported in P4500/P4600 SSDs.
It's a fixed unavailable value in each P4500/P4600 SSD.
Multiple SSDs need to return the different subnqn when SSDs are initialization.
To P4500/P4600 SSDs, in nvme_init_subnqn, id->subnqn is not null and
it's from P4500/P4600. However, multiple P4500/P4600 SSDs return the same subnqn.
It causes the possible initialization failure.
To fix this, the flag of NVME_QUIRK_IGNORE_DEV_SUBNQN need to be added to
driver_data of P4500/P4600 for nvme_id_table.
Signed-off-by: Zheng Wu <wu.zheng@intel.com>
Signed-off-by: Ye Jinhe <jinhe.ye@intel.com>
---
drivers/nvme/host/pci.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c
index ca2ee80..953ea3d 100644
--- a/drivers/nvme/host/pci.c
+++ b/drivers/nvme/host/pci.c
@@ -3326,7 +3326,8 @@ static const struct pci_device_id nvme_id_table[] = {
NVME_QUIRK_DEALLOCATE_ZEROES, },
{ PCI_VDEVICE(INTEL, 0x0a54), /* Intel P4500/P4600 */
.driver_data = NVME_QUIRK_STRIPE_SIZE |
- NVME_QUIRK_DEALLOCATE_ZEROES, },
+ NVME_QUIRK_DEALLOCATE_ZEROES |
+ NVME_QUIRK_IGNORE_DEV_SUBNQN, },
{ PCI_VDEVICE(INTEL, 0x0a55), /* Dell Express Flash P4600 */
.driver_data = NVME_QUIRK_STRIPE_SIZE |
NVME_QUIRK_DEALLOCATE_ZEROES, },
--
2.7.4
^ permalink raw reply related [flat|nested] 5+ messages in thread* Re: [PATCH] nvme: fix possible initialization failure for P4500/P4600 SSDs
2021-06-21 23:07 Zheng Wu
@ 2022-01-10 16:03 ` Keith Busch
2022-01-26 16:14 ` Christoph Hellwig
1 sibling, 0 replies; 5+ messages in thread
From: Keith Busch @ 2022-01-10 16:03 UTC (permalink / raw)
To: Zheng Wu; +Cc: linux-nvme, Ye Jinhe
On Mon, Jun 21, 2021 at 07:07:01PM -0400, Zheng Wu wrote:
> From: Wu Zheng <wu.zheng@intel.com>
>
> Subnqn isn't supported in P4500/P4600 SSDs.
> It's a fixed unavailable value in each P4500/P4600 SSD.
>
> Multiple SSDs need to return the different subnqn when SSDs are initialization.
> To P4500/P4600 SSDs, in nvme_init_subnqn, id->subnqn is not null and
> it's from P4500/P4600. However, multiple P4500/P4600 SSDs return the same subnqn.
> It causes the possible initialization failure.
>
> To fix this, the flag of NVME_QUIRK_IGNORE_DEV_SUBNQN need to be added to
> driver_data of P4500/P4600 for nvme_id_table.
>
> Signed-off-by: Zheng Wu <wu.zheng@intel.com>
> Signed-off-by: Ye Jinhe <jinhe.ye@intel.com>
Looks fine.
Reviewed-by: Keith Busch <kbusch@kernel.org>
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH] nvme: fix possible initialization failure for P4500/P4600 SSDs
2021-06-21 23:07 Zheng Wu
2022-01-10 16:03 ` Keith Busch
@ 2022-01-26 16:14 ` Christoph Hellwig
1 sibling, 0 replies; 5+ messages in thread
From: Christoph Hellwig @ 2022-01-26 16:14 UTC (permalink / raw)
To: Zheng Wu; +Cc: linux-nvme, Ye Jinhe
Thanks,
applied to nvme-5.17 after fixing up the commit message.
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2022-01-26 17:12 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-06-17 16:05 [PATCH] nvme: fix possible initialization failure for P4500/P4600 SSDs Zheng Wu
2021-12-10 17:31 ` Keith Busch
2021-06-21 23:07 Zheng Wu
2022-01-10 16:03 ` Keith Busch
2022-01-26 16:14 ` Christoph Hellwig
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.