Skip to contentSkip to conversation

nvme-pci: disable write zeros command for Intel 600P/P3100

PATCH Applied David Milburn 3 participants 5 messages Sep 10, 2020 – Sep 15, 2020 NVMe

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 nvme-pci: disable write zeros command for Intel 600P/P3100

The write zeros command does not work with 4k range.

bash-4.4# ./blkdiscard /dev/nvme0n1p2
bash-4.4# strace -efallocate xfs_io -c "fzero 536895488 2048" /dev/nvme0n1p2 fallocate(3, FALLOC_FL_ZERO_RANGE, 536895488, 2048) = 0
+++ exited with 0 +++
bash-4.4# dd bs=1 if=/dev/nvme0n1p2 skip=536895488 count=512 | hexdump -C 00000000 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................| *
00000200

bash-4.4# ./blkdiscard /dev/nvme0n1p2
bash-4.4# strace -efallocate xfs_io -c "fzero 536895488 4096" /dev/nvme0n1p2 fallocate(3, FALLOC_FL_ZERO_RANGE, 536895488, 4096) = 0
+++ exited with 0 +++
bash-4.4# dd bs=1 if=/dev/nvme0n1p2 skip=536895488 count=512 | hexdump -C

00000000  5c 61 5c b0 96 21 1b 5e  85 0c 07 32 9c 8c eb 3c  |\a\..!.^...2...<|
00000010  4a a2 06 ca 67 15 2d 8e  29 8d a8 a0 7e 46 8c 62  |J...g.-.)...~F.b|
00000020  bb 4c 6c c1 6b f5 ae a5  e4 a9 bc 93 4f 60 ff 7a  |.Ll.k.......O`.z|
---
 drivers/nvme/host/pci.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/nvme/host/pci.c +2 -1
index 5e07d5628864..d31e298669a9 100644
@@ -3153,7 +3153,8 @@ static const struct pci_device_id nvme_id_table[] = {
 	{ PCI_VDEVICE(INTEL, 0xf1a5),	/* Intel 600P/P3100 */
 		.driver_data = NVME_QUIRK_NO_DEEPEST_PS |
 				NVME_QUIRK_MEDIUM_PRIO_SQ |
-				NVME_QUIRK_NO_TEMP_THRESH_CHANGE },
+				NVME_QUIRK_NO_TEMP_THRESH_CHANGE |
+				NVME_QUIRK_DISABLE_WRITE_ZEROES, },
 	{ PCI_VDEVICE(INTEL, 0xf1a6),	/* Intel 760p/Pro 7600p */
 		.driver_data = NVME_QUIRK_IGNORE_DEV_SUBNQN, },
 	{ PCI_VDEVICE(INTEL, 0x5845),	/* Qemu emulated controller */
Signature · 7 lines
2.18.1


_______________________________________________
Linux-nvme mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-nvme
Tested-by on v1 1/1
#1 lore raw
Quoted 1 line · David Milburn

The write zeros command does not work with 4k range.

... or larger, it seems.

mkfs.xfs started using FALLOC_FL_ZERO_RANGE to zero the log, and this resulted in a corrupted/un-parseable log & un-mountable filesystem after mkfs on a 600P drive. (this behavior was introduced in xfsprogs v5.5.0)

I just happened to notice that the smaller size seemed to work.

The particular model & firmware I had trouble with was:

# cat /sys/block/nvme0n1/device/model 
INTEL SSDPEKKW512G7

I don't have access to the machine right now, but I /think/ the firmware sysfs file said "PSF122C"

This patch does make it work for me, so if you like you can add:

Thanks,
-Eric

Quoted 31 lines

bash-4.4# ./blkdiscard /dev/nvme0n1p2
bash-4.4# strace -efallocate xfs_io -c "fzero 536895488 2048" /dev/nvme0n1p2 fallocate(3, FALLOC_FL_ZERO_RANGE, 536895488, 2048) = 0
+++ exited with 0 +++
bash-4.4# dd bs=1 if=/dev/nvme0n1p2 skip=536895488 count=512 | hexdump -C 00000000 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................| *
00000200

bash-4.4# ./blkdiscard /dev/nvme0n1p2
bash-4.4# strace -efallocate xfs_io -c "fzero 536895488 4096" /dev/nvme0n1p2 fallocate(3, FALLOC_FL_ZERO_RANGE, 536895488, 4096) = 0
+++ exited with 0 +++
bash-4.4# dd bs=1 if=/dev/nvme0n1p2 skip=536895488 count=512 | hexdump -C

00000000  5c 61 5c b0 96 21 1b 5e  85 0c 07 32 9c 8c eb 3c  |\a\..!.^...2...<|
00000010  4a a2 06 ca 67 15 2d 8e  29 8d a8 a0 7e 46 8c 62  |J...g.-.)...~F.b|
00000020  bb 4c 6c c1 6b f5 ae a5  e4 a9 bc 93 4f 60 ff 7a  |.Ll.k.......O`.z|
---
 drivers/nvme/host/pci.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/nvme/host/pci.c +0 -0
index 5e07d5628864..d31e298669a9 100644

drivers/nvme/host/pci.c static const struct pci_device_id nvme_id_table[] = { · line 3160

Quoted 9 lines
drivers/nvme/host/pci.c +2 -1
@@ -3153,7 +3153,8 @@ static const struct pci_device_id nvme_id_table[] = {
 	{ PCI_VDEVICE(INTEL, 0xf1a5),	/* Intel 600P/P3100 */
 		.driver_data = NVME_QUIRK_NO_DEEPEST_PS |
 				NVME_QUIRK_MEDIUM_PRIO_SQ |
-				NVME_QUIRK_NO_TEMP_THRESH_CHANGE },
+				NVME_QUIRK_NO_TEMP_THRESH_CHANGE |
+				NVME_QUIRK_DISABLE_WRITE_ZEROES, },
 	{ PCI_VDEVICE(INTEL, 0xf1a6),	/* Intel 760p/Pro 7600p */
 		.driver_data = NVME_QUIRK_IGNORE_DEV_SUBNQN, },
 	{ PCI_VDEVICE(INTEL, 0x5845),	/* Qemu emulated controller */

_______________________________________________
Linux-nvme mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-nvme

maintainer Reviewed-by on v1 1/1
#2 lore raw
In reply to David Milburn

drivers/nvme/host/pci.c static const struct pci_device_id nvme_id_table[] = { · line 3157

Quoted 11 lines · David Milburn
drivers/nvme/host/pci.c +2 -1
index 5e07d5628864..d31e298669a9 100644
@@ -3153,7 +3153,8 @@ static const struct pci_device_id nvme_id_table[] = {
 	{ PCI_VDEVICE(INTEL, 0xf1a5),	/* Intel 600P/P3100 */
 		.driver_data = NVME_QUIRK_NO_DEEPEST_PS |
 				NVME_QUIRK_MEDIUM_PRIO_SQ |
-				NVME_QUIRK_NO_TEMP_THRESH_CHANGE },
+				NVME_QUIRK_NO_TEMP_THRESH_CHANGE |
+				NVME_QUIRK_DISABLE_WRITE_ZEROES, },

That's one amazingly buggy driver :) Wasn't it the one with the corruptions with 512 byte writes as well that got fixed by the firmware upgrade?

But yes, we'll have to work around it, so:

_______________________________________________
Linux-nvme mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-nvme

#3 lore raw
Quoted 16 lines · Christoph Hellwig

drivers/nvme/host/pci.c static const struct pci_device_id nvme_id_table[] = { · line 3157

Quoted 11 lines · David Milburn
drivers/nvme/host/pci.c +2 -1
index 5e07d5628864..d31e298669a9 100644
@@ -3153,7 +3153,8 @@ static const struct pci_device_id nvme_id_table[] = {
 	{ PCI_VDEVICE(INTEL, 0xf1a5),	/* Intel 600P/P3100 */
 		.driver_data = NVME_QUIRK_NO_DEEPEST_PS |
 				NVME_QUIRK_MEDIUM_PRIO_SQ |
-				NVME_QUIRK_NO_TEMP_THRESH_CHANGE },
+				NVME_QUIRK_NO_TEMP_THRESH_CHANGE |
+				NVME_QUIRK_DISABLE_WRITE_ZEROES, },

That's one amazingly buggy driver :) Wasn't it the one with the corruptions with 512 byte writes as well that got fixed by the firmware upgrade?

Yes

Quoted 3 lines

But yes, we'll have to work around it, so:

Thanks Christoph

-Eric

_______________________________________________
Linux-nvme mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-nvme