linux-stable.git/drivers/nvme, branch v6.1.3

nvmet: don't defer passthrough commands with trivial effects to the workqueue

2023-01-04T10:28:52+00:00

[ Upstream commit 2a459f6933e1c459bffb7cc73fd6c900edc714bd ]

Mask out the "Command Supported" and "Logical Block Content Change" bits
and only defer execution of commands that have non-trivial effects to
the workqueue for synchronous execution.  This allows to execute admin
commands asynchronously on controllers that provide a Command Supported
and Effects log page, and will keep allowing to execute Write commands
asynchronously once command effects on I/O commands are taken into
account.

Fixes: c1fef73f793b ("nvmet: add passthru code to process commands")
Signed-off-by: Christoph Hellwig 
Reviewed-by: Keith Busch 
Reviewed-by: Sagi Grimberg 
Reviewed-by: Kanchan Joshi 
Signed-off-by: Sasha Levin

nvme-pci: fix page size checks

2023-01-04T10:28:49+00:00

[ Upstream commit 841734234a28fd5cd0889b84bd4d93a0988fa11e ]

The size allocated out of the dma pool is at most NVME_CTRL_PAGE_SIZE,
which may be smaller than the PAGE_SIZE.

Fixes: c61b82c7b7134 ("nvme-pci: fix PRP pool size")
Signed-off-by: Keith Busch 
Signed-off-by: Christoph Hellwig 
Signed-off-by: Sasha Levin

nvme-pci: fix mempool alloc size

2023-01-04T10:28:49+00:00

[ Upstream commit c89a529e823d51dd23c7ec0c047c7a454a428541 ]

Convert the max size to bytes to match the units of the divisor that
calculates the worst-case number of PRP entries.

The result is used to determine how many PRP Lists are required. The
code was previously rounding this to 1 list, but we can require 2 in the
worst case. In that scenario, the driver would corrupt memory beyond the
size provided by the mempool.

While unlikely to occur (you'd need a 4MB in exactly 127 phys segments
on a queue that doesn't support SGLs), this memory corruption has been
observed by kfence.

Cc: Jens Axboe 
Fixes: 943e942e6266f ("nvme-pci: limit max IO size and segments to avoid high order allocations")
Signed-off-by: Keith Busch 
Reviewed-by: Jens Axboe 
Reviewed-by: Kanchan Joshi 
Reviewed-by: Chaitanya Kulkarni 
Signed-off-by: Christoph Hellwig 
Signed-off-by: Sasha Levin

nvme-pci: fix doorbell buffer value endianness

2023-01-04T10:28:48+00:00

[ Upstream commit b5f96cb719d8ba220b565ddd3ba4ac0d8bcfb130 ]

When using shadow doorbells, the event index and the doorbell values are
written to host memory. Prior to this patch, the values written would
erroneously be written in host endianness. This causes trouble on
big-endian platforms. Fix this by adding missing endian conversions.

This issue was noticed by Guenter while testing various big-endian
platforms under QEMU[1]. A similar fix required for hw/nvme in QEMU is
up for review as well[2].

  [1]: https://lore.kernel.org/qemu-devel/20221209110022.GA3396194@roeck-us.net/
  [2]: https://lore.kernel.org/qemu-devel/20221212114409.34972-4-its@irrelevant.dk/

Fixes: f9f38e33389c ("nvme: improve performance for virtual NVMe devices")
Reported-by: Guenter Roeck 
Signed-off-by: Klaus Jensen 
Signed-off-by: Christoph Hellwig 
Signed-off-by: Sasha Levin

nvme-auth: don't override ctrl keys before validation

2022-12-31T12:33:02+00:00

[ Upstream commit 01604350e14560d4d69323eb1ba12a257a643ea8 ]

Replace ctrl ctrl_key/host_key only after nvme_auth_generate_key is successful.
Also, this fixes a bug where the keys are leaked.

Reviewed-by: Hannes Reinecke 
Signed-off-by: Sagi Grimberg 
Reviewed-by: Chaitanya Kulkarni 
Signed-off-by: Christoph Hellwig 
Signed-off-by: Sasha Levin

nvme: pass nr_maps explicitly to nvme_alloc_io_tag_set

2022-12-31T12:32:25+00:00

[ Upstream commit dcef77274ae52136925287b6b59d5c6e6a4adfb9 ]

Don't look at ctrl->ops as only RDMA and TCP actually support multiple
maps.

Fixes: 6dfba1c09c10 ("nvme-fc: use the tagset alloc/free helpers")
Fixes: ceee1953f923 ("nvme-loop: use the tagset alloc/free helpers")
Signed-off-by: Christoph Hellwig 
Reviewed-by: Sagi Grimberg 
Reviewed-by: Chaitanya Kulkarni 
Signed-off-by: Sasha Levin

nvme: return err on nvme_init_non_mdts_limits fail

2022-12-31T12:32:11+00:00

[ Upstream commit bcaf434b8f04e1ee82a8b1e1bce0de99fbff67fa ]

In nvme_init_non_mdts_limits function we were returning 0 when kzalloc
failed; it now returns -ENOMEM.

Fixes: 5befc7c26e5a ("nvme: implement non-mdts command limits")
Signed-off-by: Joel Granados 
Reviewed-by: Chaitanya Kulkarni 
Signed-off-by: Christoph Hellwig 
Signed-off-by: Sasha Levin

nvmet: only allocate a single slab for bvecs

2022-12-31T12:32:11+00:00

[ Upstream commit fa8f9ac42350edd3ce82d0d148a60f0fa088f995 ]

There is no need to have a separate slab cache for each namespace,
and having separate ones creates duplicate debugs file names as well.

Fixes: d5eff33ee6f8 ("nvmet: add simple file backed ns support")
Signed-off-by: Christoph Hellwig 
Reviewed-by: Keith Busch 
Reviewed-by: Sagi Grimberg 
Reviewed-by: Chaitanya Kulkarni 
Reviewed-by: Martin K. Petersen 
Signed-off-by: Sasha Levin

Merge tag 'block-6.1-2022-12-08' of git://git.kernel.dk/linux

2022-12-08T23:53:39+00:00

Pull block fix from Jens Axboe:
 "A small fix for initializing the NVMe quirks before initializing the
  subsystem"

* tag 'block-6.1-2022-12-08' of git://git.kernel.dk/linux:
  nvme initialize core quirks before calling nvme_init_subsystem

nvme initialize core quirks before calling nvme_init_subsystem

2022-12-06T08:05:59+00:00

A device might have a core quirk for NVME_QUIRK_IGNORE_DEV_SUBNQN
(such as Samsung X5) but it would still give a:

    "missing or invalid SUBNQN field"

warning as core quirks are filled after calling nvme_init_subnqn.  Fill
ctrl->quirks from struct core_quirks before calling nvme_init_subsystem
to fix this.

Tested on a Samsung X5.

Fixes: ab9e00cc72fa ("nvme: track subsystems")
Signed-off-by: Pankaj Raghav 
Signed-off-by: Christoph Hellwig