]> www.infradead.org Git - users/sagi/nvme-cli.git/log
users/sagi/nvme-cli.git
5 months agodoc: add reachability-associations-log command
Tokunori Ikegami [Thu, 13 Feb 2025 14:51:26 +0000 (23:51 +0900)]
doc: add reachability-associations-log command

Added the nvme-reachability-associations-log.txt file.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
5 months agonvme: add reachability-associations-log command
Tokunori Ikegami [Tue, 11 Feb 2025 15:15:37 +0000 (00:15 +0900)]
nvme: add reachability-associations-log command

Since added the NVMe 2.1 log page.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
5 months agonvme-wrap: do_admin_op() get_log_reachability_groups parameters order
Tokunori Ikegami [Wed, 12 Feb 2025 17:11:36 +0000 (02:11 +0900)]
nvme-wrap: do_admin_op() get_log_reachability_groups parameters order

Since the len parameter should be moved before the log parameter.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
5 months agobuild: bump libnvme wrap
Tokunori Ikegami [Wed, 12 Feb 2025 17:02:08 +0000 (02:02 +0900)]
build: bump libnvme wrap

Include nvme_reachability_association_desc member variable name fix.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
5 months agonvme: make -v output consistent with -H output for nvme get-feature
Martin George [Tue, 11 Feb 2025 16:23:20 +0000 (21:53 +0530)]
nvme: make -v output consistent with -H output for nvme get-feature

Current -v verbose output of the nvme get-feature command is a no-op.
So update it to make it consistent with the -H human-readable output
for this same command.

Signed-off-by: Martin George <marting@netapp.com>
5 months agonvme: make -v output consistent with -H output for nvme get-property
Martin George [Tue, 11 Feb 2025 16:07:24 +0000 (21:37 +0530)]
nvme: make -v output consistent with -H output for nvme get-property

Current -v verbose output of the nvme get-property command is a
no-op. So update it to make it consistent with the -H human-readable
output for this same command.

Signed-off-by: Martin George <marting@netapp.com>
5 months agonvme: fix reachability-groups-log command perror message
Tokunori Ikegami [Tue, 11 Feb 2025 15:17:42 +0000 (00:17 +0900)]
nvme: fix reachability-groups-log command perror message

Since copied the command implementation but missed to change.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
5 months agocompletions: add reachability-groups-log command
Tokunori Ikegami [Tue, 11 Feb 2025 14:17:45 +0000 (23:17 +0900)]
completions: add reachability-groups-log command

Both bash and zsh completions updated for the command.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
5 months agodoc: add reachability-groups-log command
Tokunori Ikegami [Tue, 11 Feb 2025 14:09:08 +0000 (23:09 +0900)]
doc: add reachability-groups-log command

Added the nvme-reachability-groups-log.txt file.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
5 months agoCI: add missing bind-mount option for /dev
Dennis Maisenbacher [Tue, 11 Feb 2025 11:09:01 +0000 (12:09 +0100)]
CI: add missing bind-mount option for /dev

The failing run-nightly-tests GitHub action is caused by a missing
bind-mount option of the /dev directory. Without the 'shared'
bind-propagation the device tree is not reliably updated within the
container. This causes newly created namespaces not to show up.

Signed-off-by: Dennis Maisenbacher <dennis.maisenbacher@wdc.com>
5 months agoRevert "argconfig: add argconfig_get_value()"
Daniel Wagner [Fri, 7 Feb 2025 13:55:28 +0000 (14:55 +0100)]
Revert "argconfig: add argconfig_get_value()"

This reverts commit cccefc48a9aca3c0b3f0f2a2a0fe5d01ad9eb59a.

Causes regression in nightly CI runs. The problem couldn't be figured out
thus revert it.

Signed-off-by: Daniel Wagner <wagi@kernel.org>
5 months agoRevert "nvme: use blkdev for direct if namespace-id specified"
Daniel Wagner [Fri, 7 Feb 2025 13:54:52 +0000 (14:54 +0100)]
Revert "nvme: use blkdev for direct if namespace-id specified"

This reverts commit 565b30e285934502e9fcaaf7d9b8bd81e5011d0f.

Causes regression in nightly CI runs. The problem couldn't be figured out
thus revert it.

Signed-off-by: Daniel Wagner <wagi@kernel.org>
5 months agoRevert "nvme: add flush command chardev handle error message"
Daniel Wagner [Fri, 7 Feb 2025 13:53:29 +0000 (14:53 +0100)]
Revert "nvme: add flush command chardev handle error message"

This reverts commit 946029cf58d2b039a20c0c96dba7e6ef636070f9.

Causes regression in nightly CI runs. The problem couldn't be figured out
thus revert it.

Signed-off-by: Daniel Wagner <wagi@kernel.org>
5 months agonvme: add reachability-groups-log command
Tokunori Ikegami [Mon, 10 Feb 2025 14:56:07 +0000 (23:56 +0900)]
nvme: add reachability-groups-log command

Since added the NVMe 2.1 log page.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
5 months agonvme: do not output debug infos on info level
Daniel Wagner [Mon, 10 Feb 2025 13:19:16 +0000 (14:19 +0100)]
nvme: do not output debug infos on info level

Do not clutter the info message level with debug information.

Signed-off-by: Daniel Wagner <wagi@kernel.org>
5 months agobuild: checkpatch limit line length to 100 chars
Daniel Wagner [Mon, 10 Feb 2025 13:04:18 +0000 (14:04 +0100)]
build: checkpatch limit line length to 100 chars

The default is 100 chars.

Signed-off-by: Daniel Wagner <wagi@kernel.org>
5 months agonvme-print-stdout: add details in list-ns verbose output
Martin George [Sun, 9 Feb 2025 18:42:28 +0000 (00:12 +0530)]
nvme-print-stdout: add details in list-ns verbose output

Add identifier & NSID tags in the verbose output of the list-ns
command for better clarity. And while at it, print a common
"NVMe Namespace List" tag at the top too. For e.g.

nvme list-ns /dev/nvme1n1 -v
dev: /dev/nvme1n1, fd: 3
NVME Namespace List:
Identifier    0: NSID 0x1
Identifier    1: NSID 0x2
...

Signed-off-by: Martin George <marting@netapp.com>
5 months agonvme-print-stdout: add descriptions for nsze, ncap & nuse
Martin George [Sun, 9 Feb 2025 18:29:52 +0000 (23:59 +0530)]
nvme-print-stdout: add descriptions for nsze, ncap & nuse

Add descriptions for the nsze, ncap & nuse fields in the verbose
output of the id-ns command for better clarity. For e.g.

nvme id-ns /dev/nvme1n1 -v
dev: /dev/nvme1n1, fd: 3
NVME Identify Namespace 1:
nsze    : 0x280000      Total size in logical blocks
ncap    : 0x280000      Maximum size in logical blocks
nuse    : 0x13b3e       Current size in logical blocks
...

Signed-off-by: Martin George <marting@netapp.com>
5 months agocompletions: add and fix io-mgmt-recv and io-mgmt-send commands
Tokunori Ikegami [Sun, 9 Feb 2025 05:45:26 +0000 (14:45 +0900)]
completions: add and fix io-mgmt-recv and io-mgmt-send commands

Also fixed get-reg and set-reg commands completions.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
5 months agocompletions: add dispersed-ns-participating-nss-log command
Tokunori Ikegami [Sun, 9 Feb 2025 03:02:17 +0000 (12:02 +0900)]
completions: add dispersed-ns-participating-nss-log command

Both bash and zsh completions updated for the command.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
5 months agodoc: add dispersed-ns-participating-nss-log command
Tokunori Ikegami [Sun, 9 Feb 2025 02:33:07 +0000 (11:33 +0900)]
doc: add dispersed-ns-participating-nss-log command

Added the nvme-dispersed-ns-participating-nss-log.txt file.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
5 months agonvme: add dispersed-ns-participating-nss-log command
Tokunori Ikegami [Sun, 2 Feb 2025 11:20:46 +0000 (20:20 +0900)]
nvme: add dispersed-ns-participating-nss-log command

Since added the NVMe 2.1 log page.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
5 months agocompletions: add changed-alloc-ns-list-log command
Tokunori Ikegami [Fri, 7 Feb 2025 15:10:15 +0000 (00:10 +0900)]
completions: add changed-alloc-ns-list-log command

Both bash and zsh completions updated for the command.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
5 months agocompletions: fix mgmt-addr-list-log command
Tokunori Ikegami [Sun, 26 Jan 2025 01:29:09 +0000 (10:29 +0900)]
completions: fix mgmt-addr-list-log command

Since missed to add the command into the command list.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
5 months agodoc: add changed-alloc-ns-list-log command
Tokunori Ikegami [Sun, 26 Jan 2025 00:59:16 +0000 (09:59 +0900)]
doc: add changed-alloc-ns-list-log command

Added the nvme-changed-alloc-ns-list-log.txt file.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
5 months agonvme: add NVMe 2.1 changed-alloc-ns-list-log command
Tokunori Ikegami [Mon, 27 Jan 2025 14:22:35 +0000 (23:22 +0900)]
nvme: add NVMe 2.1 changed-alloc-ns-list-log command

The command functions combined with the changed-ns-list-log command.
The output almost same but different as lists attached or allocated NSs.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
5 months agonvme-wrap: handle get log rotational media info correctly
Tokunori Ikegami [Mon, 10 Feb 2025 12:28:42 +0000 (13:28 +0100)]
nvme-wrap: handle get log rotational media info correctly

Return error for the missing get log rotational media function
for non MI devices.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
5 months agobuild: sync checkpatch config file
Daniel Wagner [Mon, 10 Feb 2025 12:44:01 +0000 (13:44 +0100)]
build: sync checkpatch config file

Sync nvme-cli checkpatch configuration with the one in libnvme.
There is no point in having different configurations for this.

Signed-off-by: Daniel Wagner <wagi@kernel.org>
5 months agobuild: checkpatch ignore new/del/mov files
Daniel Wagner [Mon, 10 Feb 2025 12:37:50 +0000 (13:37 +0100)]
build: checkpatch ignore new/del/mov files

Do not warn about missing MAINTAINER file updates.

Signed-off-by: Daniel Wagner <wagi@kernel.org>
5 months agonvme-wrap: use rotational media information log libnvme-mi API
Tokunori Ikegami [Sun, 2 Feb 2025 13:39:06 +0000 (22:39 +0900)]
nvme-wrap: use rotational media information log libnvme-mi API

Since the libnvme-mi missed to add but will be added then use it.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
5 months agobuild: bump libnvme wrap
Tokunori Ikegami [Fri, 7 Feb 2025 15:01:14 +0000 (00:01 +0900)]
build: bump libnvme wrap

Include libnvme-mi NVMe 2.1 log page APIs.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
5 months agoocp: fix to free hwcomp log desc memory allocated
Tokunori Ikegami [Wed, 5 Feb 2025 15:30:12 +0000 (00:30 +0900)]
ocp: fix to free hwcomp log desc memory allocated

Since the desc pointer used _cleanup_free_ only set NULL value.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
[wagi: free buffer on error in get_hwcomp_log_data]
Signed-off-by: Daniel Wagner <wagi@kernel.org>
5 months agonvme: change attach-ns to error out if wrong device given
Tokunori Ikegami [Mon, 3 Feb 2025 15:38:14 +0000 (00:38 +0900)]
nvme: change attach-ns to error out if wrong device given

Also updating the documentation.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
5 months agonvme-print: print new id-ns fields added in NVM Command Set Spec 1.1
Francis Pravin [Tue, 21 Jan 2025 08:30:08 +0000 (14:00 +0530)]
nvme-print: print new id-ns fields added in NVM Command Set Spec 1.1

Print the new fields of Identify Namespace Data Structure added in NVM
Command Set Specification 1.1.

Signed-off-by: Francis Pravin <francis.p@samsung.com>
5 months agobuild: bump libnvme wrap
Francis Pravin [Tue, 21 Jan 2025 06:07:12 +0000 (11:37 +0530)]
build: bump libnvme wrap

dfc5cdf - types: Update id-ns fields based on NVM Command Set Spec 1.1
d8d9542 - types: Update nvm-id-ns fields based on NVM Command Set Spec 1.1

Signed-off-by: Francis Pravin <francis.p@samsung.com>
5 months agoplugins/ocp: fix telemetry parser buffer overflow
Daniel Wagner [Fri, 7 Feb 2025 14:52:41 +0000 (15:52 +0100)]
plugins/ocp: fix telemetry parser buffer overflow

The fixed 256-byte description_str will overflow for any reasonably
sized data_size >= 128.

Max data_size is for OCP VU Event Data is 0xFF Dwords, so the buffer
should be at least 1020 bytes + 1 for null. 1024 seems like a nice
number for OCP.

Reported-by: Nate Thornton <n.thornton@samsung.com>
Signed-off-by: Daniel Wagner <wagi@kernel.org>
5 months agosed: add '--read-only' to lock/unlock commands
Greg Joyce [Fri, 31 Jan 2025 20:26:05 +0000 (14:26 -0600)]
sed: add '--read-only' to lock/unlock commands

This change allows a user to specify that a drive is to be locked
read-only rather than just read-write. Also a drive that is locked
read-write can be "unlocked" to locked read-only.

The flag '--read-only' is used so the locking type can be specified.

Signed-off-by: Greg Joyce <gjoyce@linux.ibm.com>
5 months agoudev-rules-ontap: switch to queue-depth iopolicy
Martin George [Fri, 31 Jan 2025 16:30:29 +0000 (22:00 +0530)]
udev-rules-ontap: switch to queue-depth iopolicy

The queue_depth based iopolicy suits ONTAP NVMe controllers better.
So switch to the same.

Signed-off-by: Martin George <marting@netapp.com>
5 months agodoc: add rotational-media-info-log command
Tokunori Ikegami [Thu, 30 Jan 2025 15:05:33 +0000 (00:05 +0900)]
doc: add rotational-media-info-log command

Added the nvme-rotational-media-info-log.txt file.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
5 months agocompletions: add rotational-media-info-log command
Tokunori Ikegami [Thu, 30 Jan 2025 14:40:38 +0000 (23:40 +0900)]
completions: add rotational-media-info-log command

Both bash and zsh completions updated for the command.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
5 months agotests: monitor /dev for new nvme block device
Daniel Wagner [Wed, 29 Jan 2025 16:19:49 +0000 (17:19 +0100)]
tests: monitor /dev for new nvme block device

The attach operation takes a bit of time and is asynchronous, thus we
should monitor if the block device appears for a period, before
returning an error.

Signed-off-by: Daniel Wagner <wagi@kernel.org>
5 months agoplugins/solidigm: Telemetry null check json object
hsdhillo [Thu, 2 Jan 2025 19:02:17 +0000 (11:02 -0800)]
plugins/solidigm: Telemetry null check json object

ensure object is not null

Signed-off-by: hsdhillo <Hardeep.Dhillon@solidigm.com>
5 months agoplugins/solidigm: Replace snprintf with asprintf
hsdhillo [Tue, 31 Dec 2024 16:29:53 +0000 (08:29 -0800)]
plugins/solidigm: Replace snprintf with asprintf

Replace snprintf with better memory management option

Signed-off-by: hsdhillo <Hardeep.Dhillon@solidigm.com>
5 months agoRevert "tests: add blkdev checking after NS creation"
Tokunori Ikegami [Tue, 28 Jan 2025 15:00:54 +0000 (00:00 +0900)]
Revert "tests: add blkdev checking after NS creation"

This reverts commit c276f3ddd780c55458e509b74d80633728828950.

The namespace will be created but not yet attached, thus the block
device is not created either.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
[wagi: added commit message]
Signed-off-by: Daniel Wagner <wagi@kernel.org>
5 months agofabrics: make 'disconnect' accepting only -n or -d
Hannes Reinecke [Mon, 27 Jan 2025 11:20:46 +0000 (12:20 +0100)]
fabrics: make 'disconnect' accepting only -n or -d

There is limited usability in specifying both -n and -d, so make
both exclusive.

Signed-off-by: Hannes Reinecke <hare@suse.de>
5 months agonvme: add rotational-media-info-log command
Tokunori Ikegami [Sun, 26 Jan 2025 14:36:58 +0000 (23:36 +0900)]
nvme: add rotational-media-info-log command

Since added the NVMe 2.1 log page.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
5 months agotests: add verbose options for NS management commands
Tokunori Ikegami [Thu, 23 Jan 2025 15:45:54 +0000 (00:45 +0900)]
tests: add verbose options for NS management commands

This is for debugging output if test failed.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
5 months agotests: add blkdev checking after NS creation
Tokunori Ikegami [Thu, 23 Jan 2025 15:03:45 +0000 (00:03 +0900)]
tests: add blkdev checking after NS creation

To make sure the blkdev as same with NS attachment.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
5 months agotests: correct to check blkdev for NS attached
Tokunori Ikegami [Thu, 23 Jan 2025 15:40:49 +0000 (00:40 +0900)]
tests: correct to check blkdev for NS attached

Since currently only default NS blkdev checked for all NSs.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
5 months agoocp-telemetry-decode: fix ascii string lookup
Paul Wood [Mon, 23 Dec 2024 19:57:11 +0000 (11:57 -0800)]
ocp-telemetry-decode: fix ascii string lookup

Lookup ascii string in VU Event Identifier String Table for vendor
unique event classes. Fix typo.

Signed-off-by: Paul Wood <p.wood@samsung.com>
6 months agonvmf-keys: udev rule should be using /etc/nvme
Chris Leech [Wed, 22 Jan 2025 17:29:31 +0000 (09:29 -0800)]
nvmf-keys: udev rule should be using /etc/nvme

Append nvme to @SYSCONFDIR@ in udev rule.

I'd think the TLS keyfile should be located at /etc/nvme/tls-keys and
not just /etc/tls-keys.

Signed-off-by: Chris Leech <cleech@redhat.com>
6 months agonvme: set eds to true if controller supports 128 bit hostid
Bryan Gurney [Tue, 21 Jan 2025 17:20:34 +0000 (12:20 -0500)]
nvme: set eds to true if controller supports 128 bit hostid

A controller that uses a 128-bit Host Identifier may result in the
"nvme resv-report" command failing with a "Host Identifier Inconsistent
Format" error that suggests the "simultaneous use of 64-bit and
128-bit Host Identifier values on different controllers".

This error can be avoided if the "--eds" option is used, to request
the extended data structure.  However, the controller's ctratt
value indicates whether the Host Identifier is 64 bits or 128 bits.
Therefore, check the ctratt flag, and set eds to true if the
controller indicates a 128-bit Host Identifier.

Signed-off-by: Bryan Gurney <bgurney@redhat.com>
6 months agonvme: add flush command chardev handle error message
Tokunori Ikegami [Sun, 19 Jan 2025 05:36:22 +0000 (14:36 +0900)]
nvme: add flush command chardev handle error message

Since IO commands rejected by the chardev if more than one namespace.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
6 months agonvme: use blkdev for direct if namespace-id specified
Tokunori Ikegami [Sun, 19 Jan 2025 01:08:06 +0000 (10:08 +0900)]
nvme: use blkdev for direct if namespace-id specified

Since IO commands rejected by the chardev if more than one namespace.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
6 months agoargconfig: add argconfig_get_value()
Tokunori Ikegami [Sat, 18 Jan 2025 16:30:37 +0000 (01:30 +0900)]
argconfig: add argconfig_get_value()

This is to get the option value by name.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
6 months agonvme: add NS management commands error message output
Tokunori Ikegami [Sat, 18 Jan 2025 08:15:32 +0000 (17:15 +0900)]
nvme: add NS management commands error message output

Print it if the commands failed and NS management not supported.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
6 months agobuild: remove thank you bot
Daniel Wagner [Mon, 20 Jan 2025 13:17:45 +0000 (14:17 +0100)]
build: remove thank you bot

Once again beaten by security.

Signed-off-by: Daniel Wagner <wagi@kernel.org>
6 months agobuild: fix REST API call and permissions for thanks you note
Daniel Wagner [Mon, 20 Jan 2025 09:33:36 +0000 (10:33 +0100)]
build: fix REST API call and permissions for thanks you note

The github action needs to use the correct REST API and the write
permission to be able to comment on the issue/pull request.

Signed-off-by: Daniel Wagner <wagi@kernel.org>
6 months agonvme-print-stdout: use NVME_CTRL_OACS definitions
Tokunori Ikegami [Sat, 18 Jan 2025 08:27:59 +0000 (17:27 +0900)]
nvme-print-stdout: use NVME_CTRL_OACS definitions

Still needed to add some remaining definitions in libnvme.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
6 months agobuild: automatically add a thank you note on PR merge
Daniel Wagner [Mon, 20 Jan 2025 09:05:24 +0000 (10:05 +0100)]
build: automatically add a thank you note on PR merge

Be friendly and add a thank you note to all merged PRs. Currently, I
type it always by hand.

Signed-off-by: Daniel Wagner <wagi@kernel.org>
6 months agonvme-print-json: add get feature pls mode print
Tokunori Ikegami [Sun, 12 Jan 2025 11:04:43 +0000 (20:04 +0900)]
nvme-print-json: add get feature pls mode print

This change required libnvme FEAT_PLS_MODE definition added.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
6 months agonvme-print-stdout: add get feature pls mode print
Tokunori Ikegami [Sun, 12 Jan 2025 08:00:08 +0000 (17:00 +0900)]
nvme-print-stdout: add get feature pls mode print

This change required libnvme FEAT_PLS_MODE definition added.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
6 months agodoc: add mgmt-addr-list-log command
Tokunori Ikegami [Mon, 13 Jan 2025 16:58:59 +0000 (01:58 +0900)]
doc: add mgmt-addr-list-log command

Added the nvme-mgmt-addr-list-log.txt file.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
6 months agocompletions: add mgmt-addr-list-log command
Tokunori Ikegami [Mon, 13 Jan 2025 16:57:24 +0000 (01:57 +0900)]
completions: add mgmt-addr-list-log command

Both bash and zsh completions updated for the command.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
6 months agocompletions: fix ocp hardware-component-log zsh completion
Tokunori Ikegami [Mon, 13 Jan 2025 16:52:56 +0000 (01:52 +0900)]
completions: fix ocp hardware-component-log zsh completion

Incorrectly the short option -v duplicated.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
6 months agodoc: fix ocp-hardware-component-log documentation errors
Tokunori Ikegami [Mon, 13 Jan 2025 16:50:40 +0000 (01:50 +0900)]
doc: fix ocp-hardware-component-log documentation errors

Since described as set-error-injection incorrectly.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
6 months agonvme-print-binary: add binary_mgmt_addr_list_log()
Tokunori Ikegami [Mon, 13 Jan 2025 15:59:55 +0000 (00:59 +0900)]
nvme-print-binary: add binary_mgmt_addr_list_log()

Since added the NVMe 2.1 log page.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
6 months agonvme-print-json: add json_mgmt_addr_list_log()
Tokunori Ikegami [Mon, 13 Jan 2025 07:01:42 +0000 (16:01 +0900)]
nvme-print-json: add json_mgmt_addr_list_log()

Since added the NVMe 2.1 log page.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
6 months agonvme: add mgmt-addr-list-log command
Tokunori Ikegami [Mon, 13 Jan 2025 05:28:06 +0000 (14:28 +0900)]
nvme: add mgmt-addr-list-log command

Since added the NVMe 2.1 log page.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
6 months agobuild: bump libnvme wrap
Tokunori Ikegami [Mon, 13 Jan 2025 16:12:31 +0000 (01:12 +0900)]
build: bump libnvme wrap

Include libnvme nvme_mi_admin_get_log_mgmt_addr_list().

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
6 months agonvme-print-json: add NVMe 2.1 identify power state descriptor fields
Tokunori Ikegami [Sun, 12 Jan 2025 05:04:51 +0000 (14:04 +0900)]
nvme-print-json: add NVMe 2.1 identify power state descriptor fields

Since power loss signaling function introduced by NVMe 2.1.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
6 months agonvme-print-stdout: add NVMe 2.1 identify power state descriptor fields
Tokunori Ikegami [Sat, 11 Jan 2025 15:21:58 +0000 (00:21 +0900)]
nvme-print-stdout: add NVMe 2.1 identify power state descriptor fields

Since power loss signaling function introduced by NVMe 2.1.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
6 months agotypes: move STR_LEN definition from json.h
Tokunori Ikegami [Sat, 11 Jan 2025 15:19:34 +0000 (00:19 +0900)]
types: move STR_LEN definition from json.h

Since to use the definition in nvme-print-stdout.c also.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
6 months agonvme-print: add NVMe 2.1 admin commands strings
Tokunori Ikegami [Thu, 9 Jan 2025 16:20:30 +0000 (01:20 +0900)]
nvme-print: add NVMe 2.1 admin commands strings

Also added some missed admin commands strings.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
6 months agonvme-print-stdout: add human readable output for DSTO field
Francis Pravin [Thu, 9 Jan 2025 13:23:59 +0000 (18:53 +0530)]
nvme-print-stdout: add human readable output for DSTO field

Add human readable output for DSTO field of Identify Controller Data
Structure.

Signed-off-by: Francis Pravin <francis.p@samsung.com>
6 months agonvme: add the new Self-test Code of Device Self-test command
Francis Pravin [Thu, 9 Jan 2025 12:54:01 +0000 (18:24 +0530)]
nvme: add the new Self-test Code of Device Self-test command

Add the new Self-test Code (SCT) Host-Initiated Refresh operation support
to Device Self-test command. Also, add it in the Device Self-test Log Page.
TP4058 - Environmental Extremes Management.

Signed-off-by: Francis Pravin <francis.p@samsung.com>
6 months agonvme-print-stdout: split PEL functions
Tokunori Ikegami [Fri, 3 Jan 2025 10:43:59 +0000 (19:43 +0900)]
nvme-print-stdout: split PEL functions

This is to reduce the PEL long function implementation.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
6 months agonvme-print-stdout: split PEL set feature event function
Tokunori Ikegami [Wed, 8 Jan 2025 15:47:32 +0000 (00:47 +0900)]
nvme-print-stdout: split PEL set feature event function

This is to reduce the PEL long function implementation.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
6 months agonvme-print-stdout: use libnvme FDP events feature definitions
Tokunori Ikegami [Fri, 3 Jan 2025 07:01:20 +0000 (16:01 +0900)]
nvme-print-stdout: use libnvme FDP events feature definitions

This is to reduce to use the bitwise operators.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
6 months agonvme-print-stdout: fix to print PEL set feature event FDP events
Tokunori Ikegami [Thu, 2 Jan 2025 12:39:51 +0000 (21:39 +0900)]
nvme-print-stdout: fix to print PEL set feature event FDP events

Since previously only FDP event type 0 output incorrectly.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
6 months agonvme-print: add NVMe 2.1 get and set features FIDs string
Tokunori Ikegami [Wed, 8 Jan 2025 14:56:03 +0000 (23:56 +0900)]
nvme-print: add NVMe 2.1 get and set features FIDs string

Fixed also the indentation error.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
6 months agobuild: bump libnvme wrap
Tokunori Ikegami [Wed, 8 Jan 2025 15:19:33 +0000 (00:19 +0900)]
build: bump libnvme wrap

Include libnvme NVMe 2.1 features FID definition addition.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
6 months agoocp-nvme: Add missing data media_dies_offline to smart_extended_log.
Minsik Jeon [Wed, 8 Jan 2025 09:52:44 +0000 (18:52 +0900)]
ocp-nvme: Add missing data media_dies_offline to smart_extended_log.

related commit : https://github.com/linux-nvme/nvme-cli/commit/d4b5f77dc862530aed8523da3662ecab3850f0f4

Signed-off-by: Minsik Jeon <hmi.jeon@samsung.com>
6 months agoocp-nvme: Update ocp v2.6 Smart Information Extended
Minsik Jeon [Wed, 8 Jan 2025 05:37:50 +0000 (14:37 +0900)]
ocp-nvme: Update ocp v2.6 Smart Information Extended

Log Identifier C0h.
Spec Documents:
https://www.opencompute.org/documents/datacenter-nvme-ssd-specification-v2-6-2-pdf

Signed-off-by: Minsik Jeon <hmi.jeon@samsung.com>
Co-authored-by: Steven Seungcheol Lee <sc108.lee@samsung.com>
6 months agoplugins/ocp: Align ocp_smart_extended_log struct.
Minsik Jeon [Wed, 11 Dec 2024 05:39:37 +0000 (14:39 +0900)]
plugins/ocp: Align ocp_smart_extended_log struct.

If the dssd point, minor version are declared as __le16, the alignment
will be broken. Remove __packed keyword.
Change parameter void to ocp_smart_extended_log.

Reported-by: Steven Seungcheol Lee <sc108.lee@samsung.com>
Signed-off-by: Minsik Jeon <hmi.jeon@samsung.com>
6 months agoplugins/ocp: Use structure for ocp smart log
Steven Seungcheol Lee [Tue, 7 Jan 2025 09:01:39 +0000 (18:01 +0900)]
plugins/ocp: Use structure for ocp smart log

printed log tested (same), except below
Remove wrong duplicated prints from stdout
  NVMe base errata version
  NVMe command set errata version

Signed-off-by: Steven Seungcheol Lee <sc108.lee@samsung.com>
6 months agoocp: Update ocp plugin version
jeff-lien-wdc [Tue, 7 Jan 2025 19:10:39 +0000 (13:10 -0600)]
ocp: Update ocp plugin version

Set plugin version to 2.11.0

Signed-off-by: jeff-lien-wdc <jeff.lien@sandisk.com>
6 months agoocp: Fixes for OCP 2.5 Telemetry Stats Parsing
jeff-lien-wdc [Fri, 20 Dec 2024 21:34:54 +0000 (15:34 -0600)]
ocp: Fixes for OCP 2.5 Telemetry Stats Parsing

The OCP 2.5 Telemetry log statistics parsing code was not
checking for the end of the stats data and therefore getting
stuck in an infinite while loop.
Added code to parse out the statistic specific data for the
Max Die, Max NAND Channel, and Min NAND Channel Bad Block.

Removed unneeded brackets
Fixed lines exceeding 100 chars

Signed-off-by: jeff-lien-wdc <jeff.lien@sandisk.com>
6 months agomake: fix make rpm target
Sagi Grimberg [Tue, 7 Jan 2025 11:27:11 +0000 (13:27 +0200)]
make: fix make rpm target

Not specifying 70-nvmf-keys.rules as a known spec file results in this
error from make rpm:
error: Installed (but unpackaged) file(s) found:
   /usr/local/%{_udevrulesdir}/70-nvmf-keys.rules

Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
Link: https://lore.kernel.org/r/20250107112711.192068-1-sagi@grimberg.me
Signed-off-by: Daniel Wagner <wagi@kernel.org>
6 months agonvme-print: add NVMe 2.1 get log page LIDs string
Tokunori Ikegami [Sat, 4 Jan 2025 03:12:52 +0000 (12:12 +0900)]
nvme-print: add NVMe 2.1 get log page LIDs string

Added also Phy Rx EOM log page missed.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
6 months agoocp: change hwcomp log size variable name to log_bytes
Tokunori Ikegami [Fri, 3 Jan 2025 15:01:18 +0000 (00:01 +0900)]
ocp: change hwcomp log size variable name to log_bytes

Since the size value calculated to bytes from dwords for version 1.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
6 months agoocp: fix to check hwcomp log size if valid
Tokunori Ikegami [Sun, 29 Dec 2024 12:01:05 +0000 (21:01 +0900)]
ocp: fix to check hwcomp log size if valid

Since the log size value needed to reduce the descriptor length.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
6 months agoocp: add ocp dssd log id and feature id definitions
Tokunori Ikegami [Fri, 3 Jan 2025 14:58:41 +0000 (23:58 +0900)]
ocp: add ocp dssd log id and feature id definitions

This is to reduce the hardcoded fid and lid values.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
6 months agoutil: introduce int56_to_long() type conversion helper
Tokunori Ikegami [Fri, 3 Jan 2025 13:15:54 +0000 (22:15 +0900)]
util: introduce int56_to_long() type conversion helper

This is to remove unnecessary mask with 0x00FFFFFFFFFFFFFF.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
6 months agoocp-print: use int48_to_long() to print 48 bits values
Tokunori Ikegami [Fri, 3 Jan 2025 13:14:10 +0000 (22:14 +0900)]
ocp-print: use int48_to_long() to print 48 bits values

This is to reduce unnecessary mask with 0x0000FFFFFFFFFFFF.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
6 months agoocp: remove unnecessary 64 bits mask
Tokunori Ikegami [Fri, 3 Jan 2025 11:12:20 +0000 (20:12 +0900)]
ocp: remove unnecessary 64 bits mask

Since the 64 bits variables not necessary to mask by 64 bits.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
6 months agoocp: use struct nvme_timestamp type for FAHE-5 timestamp field
Tokunori Ikegami [Fri, 3 Jan 2025 11:07:44 +0000 (20:07 +0900)]
ocp: use struct nvme_timestamp type for FAHE-5 timestamp field

Since the field defined by the NVMe timestamp feature.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
6 months agonvme-print: move obj_add_* definitions into util/json.h
Tokunori Ikegami [Fri, 3 Jan 2025 12:04:43 +0000 (21:04 +0900)]
nvme-print: move obj_add_* definitions into util/json.h

Also use the same prefix json_object_add_*.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
6 months agonvme: fix to use NVME_VAL() for CQ entry status field filter
Tokunori Ikegami [Thu, 2 Jan 2025 13:41:41 +0000 (22:41 +0900)]
nvme: fix to use NVME_VAL() for CQ entry status field filter

Previously used incorrectly NVME_GET() but not shifted the masked values.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>
6 months agonvme-print-json: update id-ctrl print code to use libnvme definition
Tokunori Ikegami [Tue, 31 Dec 2024 15:16:44 +0000 (00:16 +0900)]
nvme-print-json: update id-ctrl print code to use libnvme definition

Use libnvme register definitions instead of the shift operator.

Signed-off-by: Tokunori Ikegami <ikegami.t@gmail.com>