linux-stable.git/drivers/nvme, branch v6.3.7

nvme-pci: Add quirk for Teamgroup MP33 SSD

2023-06-09T08:48:10+00:00

[ Upstream commit 0649728123cf6a5518e154b4e1735fc85ea4f55c ]

Add a quirk for Teamgroup MP33 that reports duplicate ids for disk.

Signed-off-by: Daniel Smith 
[kch: patch formatting]
Signed-off-by: Chaitanya Kulkarni 
Tested-by: Daniel Smith 
Signed-off-by: Keith Busch 
Signed-off-by: Sasha Levin

nvme: do not let the user delete a ctrl before a complete initialization

2023-06-09T08:48:09+00:00

[ Upstream commit 2eb94dd56a4a4e3fe286def3e2ba207804a37345 ]

If a userspace application performes a "delete_controller" command
early during the ctrl initialization, the delete operation
may race against the init code and the kernel will crash.

nvme nvme5: Connect command failed: host path error
nvme nvme5: failed to connect queue: 0 ret=880
PF: supervisor write access in kernel mode
PF: error_code(0x0002) - not-present page
 blk_mq_quiesce_queue+0x18/0x90
 nvme_tcp_delete_ctrl+0x24/0x40 [nvme_tcp]
 nvme_do_delete_ctrl+0x7f/0x8b [nvme_core]
 nvme_sysfs_delete.cold+0x8/0xd [nvme_core]
 kernfs_fop_write_iter+0x124/0x1b0
 new_sync_write+0xff/0x190
 vfs_write+0x1ef/0x280

Fix the crash by checking the NVME_CTRL_STARTED_ONCE bit;
if it's not set it means that the nvme controller is still
in the process of getting initialized and the kernel
will return an -EBUSY error to userspace.
Set the NVME_CTRL_STARTED_ONCE later in the nvme_start_ctrl()
function, after the controller start operation is completed.

Signed-off-by: Maurizio Lombardi 
Reviewed-by: Sagi Grimberg 
Reviewed-by: Christoph Hellwig 
Signed-off-by: Keith Busch 
Signed-off-by: Sasha Levin

nvme-multipath: don't call blk_mark_disk_dead in nvme_mpath_remove_disk

2023-06-09T08:48:09+00:00

[ Upstream commit 1743e5f6000901a11f4e1cd741bfa9136f3ec9b1 ]

nvme_mpath_remove_disk is called after del_gendisk, at which point a
blk_mark_disk_dead call doesn't make any sense.

Signed-off-by: Christoph Hellwig 
Reviewed-by: Sagi Grimberg 
Signed-off-by: Keith Busch 
Signed-off-by: Sasha Levin

nvme-pci: clamp max_hw_sectors based on DMA optimized limitation

2023-06-09T08:48:02+00:00

[ Upstream commit 3710e2b056cb92ad816e4d79fa54a6a5b6ad8cbd ]

When running the fio test on a 448-core AMD server + a NVME disk,
a soft lockup or a hard lockup call trace is shown:

[soft lockup]
watchdog: BUG: soft lockup - CPU#126 stuck for 23s! [swapper/126:0]
RIP: 0010:_raw_spin_unlock_irqrestore+0x21/0x50
...
Call Trace:
 
 fq_flush_timeout+0x7d/0xd0
 ? __pfx_fq_flush_timeout+0x10/0x10
 call_timer_fn+0x2e/0x150
 run_timer_softirq+0x48a/0x560
 ? __pfx_fq_flush_timeout+0x10/0x10
 ? clockevents_program_event+0xaf/0x130
 __do_softirq+0xf1/0x335
 irq_exit_rcu+0x9f/0xd0
 sysvec_apic_timer_interrupt+0xb4/0xd0
 
 
 asm_sysvec_apic_timer_interrupt+0x1f/0x30
...

Obvisouly, fq_flush_timeout spends over 20 seconds. Here is ftrace log:

               |  fq_flush_timeout() {
               |    fq_ring_free() {
               |      put_pages_list() {
   0.170 us    |        free_unref_page_list();
   0.810 us    |      }
               |      free_iova_fast() {
               |        free_iova() {
 * 85622.66 us |          _raw_spin_lock_irqsave();
   2.860 us    |          remove_iova();
   0.600 us    |          _raw_spin_unlock_irqrestore();
   0.470 us    |          lock_info_report();
   2.420 us    |          free_iova_mem.part.0();
 * 85638.27 us |        }
 * 85638.84 us |      }
               |      put_pages_list() {
   0.230 us    |        free_unref_page_list();
   0.470 us    |      }
   ...            ...
 $ 31017069 us |  }

Most of cores are under lock contention for acquiring iova_rbtree_lock due
to the iova flush queue mechanism.

[hard lockup]
NMI watchdog: Watchdog detected hard LOCKUP on cpu 351
RIP: 0010:native_queued_spin_lock_slowpath+0x2d8/0x330

Call Trace:
 
 _raw_spin_lock_irqsave+0x4f/0x60
 free_iova+0x27/0xd0
 free_iova_fast+0x4d/0x1d0
 fq_ring_free+0x9b/0x150
 iommu_dma_free_iova+0xb4/0x2e0
 __iommu_dma_unmap+0x10b/0x140
 iommu_dma_unmap_sg+0x90/0x110
 dma_unmap_sg_attrs+0x4a/0x50
 nvme_unmap_data+0x5d/0x120 [nvme]
 nvme_pci_complete_batch+0x77/0xc0 [nvme]
 nvme_irq+0x2ee/0x350 [nvme]
 ? __pfx_nvme_pci_complete_batch+0x10/0x10 [nvme]
 __handle_irq_event_percpu+0x53/0x1a0
 handle_irq_event_percpu+0x19/0x60
 handle_irq_event+0x3d/0x60
 handle_edge_irq+0xb3/0x210
 __common_interrupt+0x7f/0x150
 common_interrupt+0xc5/0xf0
 
 
 asm_common_interrupt+0x2b/0x40
...

ftrace shows fq_ring_free spends over 10 seconds [1]. Again, most of
cores are under lock contention for acquiring iova_rbtree_lock due
to the iova flush queue mechanism.

[Root Cause]
The root cause is that the max_hw_sectors_kb of nvme disk (mdts=10)
is 4096kb, which streaming DMA mappings cannot benefit from the
scalable IOVA mechanism introduced by the commit 9257b4a206fc
("iommu/iova: introduce per-cpu caching to iova allocation") if
the length is greater than 128kb.

To fix the lock contention issue, clamp max_hw_sectors based on
DMA optimized limitation in order to leverage scalable IOVA mechanism.

Note: The issue does not happen with another NVME disk (mdts = 5
and max_hw_sectors_kb = 128)

[1] https://gist.github.com/AdrianHuang/bf8ec7338204837631fbdaed25d19cc4

Suggested-by: Keith Busch 
Reported-and-tested-by: Jiwei Sun 
Signed-off-by: Adrian Huang 
Reviewed-by: Keith Busch 
Signed-off-by: Christoph Hellwig 
Signed-off-by: Sasha Levin

nvme-pci: add quirk for missing secondary temperature thresholds

2023-06-09T08:48:02+00:00

[ Upstream commit bd375feeaf3408ed00e08c3bc918d6be15f691ad ]

On Kingston KC3000 and Kingston FURY Renegade (both have the same PCI
IDs) accessing temp3_{min,max} fails with an invalid field error (note
that there is no problem setting the thresholds for temp1).

This contradicts the NVM Express Base Specification 2.0b, page 292:

  The over temperature threshold and under temperature threshold
  features shall be implemented for all implemented temperature sensors
  (i.e., all Temperature Sensor fields that report a non-zero value).

Define NVME_QUIRK_NO_SECONDARY_TEMP_THRESH that disables the thresholds
for all but the composite temperature and set it for this device.

Signed-off-by: Hristo Venev 
Reviewed-by: Guenter Roeck 
Signed-off-by: Christoph Hellwig 
Signed-off-by: Sasha Levin

nvme-pci: add NVME_QUIRK_BOGUS_NID for HS-SSD-FUTURE 2048G

2023-06-09T08:48:02+00:00

[ Upstream commit 1616d6c3717bae9041a4240d381ec56ccdaafedc ]

Add a quirk to fix HS-SSD-FUTURE 2048G SSD drives reporting duplicate
nsids.

Link: https://bugzilla.kernel.org/show_bug.cgi?id=217384
Reported-by: Andrey God 
Signed-off-by: Sagi Grimberg 
Signed-off-by: Christoph Hellwig 
Signed-off-by: Sasha Levin

nvme: fix the name of Zone Append for verbose logging

2023-06-09T08:47:54+00:00

[ Upstream commit 856303797724d28f1d65b702f0eadcee1ea7abf5 ]

No Management involved in Zone Appened.

Fixes: bd83fe6f2cd2 ("nvme: add verbose error logging")
Signed-off-by: Christoph Hellwig 
Reviewed-by: Alan Adamson 
Signed-off-by: Keith Busch 
Signed-off-by: Sasha Levin

nvme-fcloop: fix "inconsistent {IN-HARDIRQ-W} -> {HARDIRQ-ON-W} usage"

2023-05-11T14:17:17+00:00

[ Upstream commit 4f86a6ff6fbd891232dda3ca97fd1b9630b59809 ]

fcloop_fcp_op() could be called from flush request's ->end_io(flush_end_io) in
which the spinlock of fq->mq_flush_lock is grabbed with irq saved/disabled.

So fcloop_fcp_op() can't call spin_unlock_irq(&tfcp_req->reqlock) simply
which enables irq unconditionally.

Fixes the warning by switching to spin_lock_irqsave()/spin_unlock_irqrestore()

Fixes: c38dbbfab1bc ("nvme-fcloop: fix inconsistent lock state warnings")
Reported-by: Yi Zhang 
Signed-off-by: Ming Lei 
Reviewed-by: Ewan D. Milne 
Tested-by: Yi Zhang 
Signed-off-by: Christoph Hellwig 
Signed-off-by: Sasha Levin

nvme: fix async event trace event

2023-05-11T14:17:17+00:00

[ Upstream commit 6622b76fe922b94189499a90ccdb714a4a8d0773 ]

Mixing AER Event Type and Event Info has masking clashes. Just print the
event type, but also include the event info of the AER result in the
trace.

Fixes: 09bd1ff4b15143b ("nvme-core: add async event trace helper")
Reported-by: Nate Thornton 
Reviewed-by: Sagi Grimberg 
Reviewed-by: Minwoo Im 
Reviewed-by: Chaitanya Kulkarni 
Signed-off-by: Keith Busch 
Signed-off-by: Christoph Hellwig 
Signed-off-by: Sasha Levin

nvmet: fix I/O Command Set specific Identify Controller

2023-05-11T14:17:17+00:00

[ Upstream commit a5a6ab0950b46ab1ef4a5c83c80234018b81b38a ]

For an identify command with cns set to NVME_ID_CNS_CS_CTRL, the NVMe
2.0 specification states that:

If the I/O Command Set specified by the CSI field does not have an
Identify Controller data structure, then the controller shall return
a zero filled data structure. If the host requests a data structure for
an I/O Command Set that the controller does not support, the controller
shall abort the command with a status code of Invalid Field in Command.

However, the current implementation of this identify command in
nvmet_execute_identify() only handles the ZNS command set, returning an
error for the NVM command set, which is not compliant with the
specifications as we do support this command set.

Fix this by:
1) Renaming nvmet_execute_identify_cns_cs_ctrl() to
   nvmet_execute_identify_ctrl_zns() to continue handling the
   ZNS command set as is.
2) Introduce a nvmet_execute_identify_ctrl_ns() helper to handle the
   NVM command set, returning a zero filled nvme_id_ctrl_nvm data
   structure.
3) Modify nvmet_execute_identify() to call these helpers based on
   the csi specified, returning an error for unsupported command sets.

Fixes: aaf2e048af27 ("nvmet: add ZBD over ZNS backend support")
Signed-off-by: Damien Le Moal 
Reviewed-by: Chaitanya Kulkarni 
Tested-by: Chaitanya Kulkarni 
Signed-off-by: Christoph Hellwig 
Signed-off-by: Sasha Levin