linux-2.6-microblaze.git
9 months agodrm/xe/hw_engine_group: Ensure safe transition between execution modes
Francois Dugast [Fri, 9 Aug 2024 15:51:33 +0000 (17:51 +0200)]
drm/xe/hw_engine_group: Ensure safe transition between execution modes

Provide a way to safely transition execution modes of the hw engine group
ahead of the actual execution. When necessary, either wait for running
jobs to complete or preempt them, thus ensuring mutual exclusion between
execution modes.

Unlike a mutex, the rw_semaphore used in this context allows multiple
submissions in the same mode.

v2: Use lockdep_assert_held_write, add annotations (Matt Brost)

v3: Fix kernel doc, remove redundant code (Matt Brost)

v4: Now that xe_hw_engine_group_suspend_faulting_lr_jobs can fail,
    propagate the error to the caller (Matt Brost)

Signed-off-by: Francois Dugast <francois.dugast@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240809155156.1955925-9-francois.dugast@intel.com
9 months agodrm/xe/hw_engine_group: Add helper to wait for dma fence jobs
Francois Dugast [Fri, 9 Aug 2024 15:51:32 +0000 (17:51 +0200)]
drm/xe/hw_engine_group: Add helper to wait for dma fence jobs

This is a required feature for faulting long running jobs not to be
submitted while dma fence jobs are running on the hw engine group.

v2: Switch to lockdep_assert_held_write in worker, get a proper reference
    for the last fence (Matt Brost)

v3: Directly call dma_fence_put with the fence ref (Matt Brost)

Signed-off-by: Francois Dugast <francois.dugast@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240809155156.1955925-8-francois.dugast@intel.com
9 months agodrm/xe/exec_queue: Prepare last fence for hw engine group resume context
Francois Dugast [Fri, 9 Aug 2024 15:51:31 +0000 (17:51 +0200)]
drm/xe/exec_queue: Prepare last fence for hw engine group resume context

Ensure we can safely take a ref of the exec queue's last fence from the
context of resuming jobs from the hw engine group. The locking requirements
differ from the general case, hence the introduction of this new function.

v2: Add kernel doc, rework the code to prevent code duplication

v3: Fix kernel doc, remove now unnecessary lockdep variants (Matt Brost)

v4: Remove new put function (Matt Brost)

Signed-off-by: Francois Dugast <francois.dugast@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240809155156.1955925-7-francois.dugast@intel.com
9 months agodrm/xe/exec_queue: Remove duplicated code
Francois Dugast [Fri, 9 Aug 2024 15:51:30 +0000 (17:51 +0200)]
drm/xe/exec_queue: Remove duplicated code

This code section is the same as the body of
xe_exec_queue_last_fence_put_unlocked() so call the function instead and
remove duplicated code to make maintenance easier.

Signed-off-by: Francois Dugast <francois.dugast@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240809155156.1955925-6-francois.dugast@intel.com
9 months agodrm/xe/hw_engine_group: Add helper to suspend faulting LR jobs
Francois Dugast [Fri, 9 Aug 2024 15:51:29 +0000 (17:51 +0200)]
drm/xe/hw_engine_group: Add helper to suspend faulting LR jobs

This is a required feature for dma fence jobs to preempt faulting long
running jobs in order to ensure mutual exclusion on a given hw engine
group.

v2: Pipeline calls to suspend(q) and suspend_wait(q) to improve
    efficiency, switch to lockdep_assert_held_write (Matt Brost)

v3: Return error on suspend_wait failure to propagate on the call stack
    up to IOCTL (Matt Brost)

Signed-off-by: Francois Dugast <francois.dugast@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240809155156.1955925-5-francois.dugast@intel.com
9 months ago'drm/xe/hw_engine_group: Register hw engine group's exec queues
Francois Dugast [Fri, 9 Aug 2024 15:51:28 +0000 (17:51 +0200)]
'drm/xe/hw_engine_group: Register hw engine group's exec queues

Add helpers to safely add and delete the exec queues attached to a hw
engine group, and make use them at the time of creation and destruction of
the exec queues. Keeping track of them is required to control the
execution mode of the hw engine group.

v2: Improve error handling and robustness, suspend exec queues created in
    fault mode if group in dma-fence mode, init queue link (Matt Brost)

v3: Delete queue from hw engine group when it is destroyed by the user,
    also clean up at the time of closing the file in case the user did
    not destroy the queue

v4: Use correct list when checking if empty, do not add the queue if VM
    is in xe_vm_in_preempt_fence_mode (Matt Brost)

v5: Remove unrelated newline, add checks and asserts for group, unwind on
    suspend failure (Matt Brost)

Signed-off-by: Francois Dugast <francois.dugast@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240809155156.1955925-4-francois.dugast@intel.com
9 months agodrm/xe/guc_submit: Make suspend_wait interruptible
Francois Dugast [Fri, 9 Aug 2024 15:51:27 +0000 (17:51 +0200)]
drm/xe/guc_submit: Make suspend_wait interruptible

Rely on wait_event_interruptible_timeout() to put the process to sleep
with TASK_INTERRUPTIBLE. It allows using this function in interruptible
context.

v2: Propagate error on wait_event_interruptible_timeout (Matt Brost)

Signed-off-by: Francois Dugast <francois.dugast@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240809155156.1955925-3-francois.dugast@intel.com
9 months agodrm/xe/hw_engine_group: Introduce xe_hw_engine_group
Francois Dugast [Fri, 9 Aug 2024 15:51:26 +0000 (17:51 +0200)]
drm/xe/hw_engine_group: Introduce xe_hw_engine_group

A xe_hw_engine_group is a group of hw engines. Two hw engines belong to
the same xe_hw_engine_group if one hw engine cannot make progress while
the other is stuck on a page fault.

Typically, hw engines of the same group share some resources such as EUs,
but this really depends on the hardware configuration of the platforms.
The simple engines partitioning proposed here might be too conservative
but is intended to work for existing platforms. It can be optimized later
if more sets of independent engines are identified.

The hw engine groups are intended to be used in the context of faulting
long-running jobs submissions.

v2: Move to own files, improve error handling (Matt Brost)

v3: Fix build issue reported by CI, improve commit message (Matt Roper)

v4: Fix kernel doc

v5: Add switch case for XE_ENGINE_CLASS_OTHER

Signed-off-by: Francois Dugast <francois.dugast@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240809155156.1955925-2-francois.dugast@intel.com
9 months agodrm/xe: Use reserved copy engine for user binds on faulting devices
Matthew Brost [Fri, 16 Aug 2024 03:40:33 +0000 (20:40 -0700)]
drm/xe: Use reserved copy engine for user binds on faulting devices

User binds map to engines with can fault, faults depend on user binds
completion, thus we can deadlock. Avoid this by using reserved copy
engine for user binds on faulting devices.

While we are here, normalize bind queue creation with a helper.

v2:
 - Pass in extensions to bind queue creation (CI)
v3:
 - s/resevered/reserved (Lucas)
 - Fix NULL hwe check (Jonathan)

Fixes: dd08ebf6c352 ("drm/xe: Introduce a new DRM driver for Intel GPUs")
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240816034033.53837-1-matthew.brost@intel.com
9 months agodrm/xe/mcr: Try to derive dss_per_grp from hwconfig attributes
Matt Roper [Thu, 15 Aug 2024 17:26:04 +0000 (10:26 -0700)]
drm/xe/mcr: Try to derive dss_per_grp from hwconfig attributes

When steering MCR register ranges of type "DSS," the group_id and
instance_id values are calculated by dividing the DSS pool according to
the size of a gslice or cslice, depending on the platform.  These values
haven't changed much on past platforms, so we've been able to hardcode
the proper divisor so far.  However the layout may not be so fixed on
future platforms so the proper, future-proof way to determine this is by
using some of the attributes from the GuC's hwconfig table.  The
hwconfig has two attributes reflecting the architectural maximum slice
and subslice counts (i.e., before any fusing is considered) that can be
used for the purposes of calculating MCR steering targets.

If the hwconfig is lacking the necessary values (which should only be
possible on older platforms before these attributes were added), we can
still fall back to the old hardcoded values.  Going forward the hwconfig
is expected to always provide the information we need on newer
platforms, and any failure to do so will be considered a bug in the
firmware that will prevent us from switching to the buggy firmware
release.

It's worth noting that over time GuC's hwconfig has provided a couple
different keys with similar-sounding descriptions.  For our purposes
here, we only trust the newer key "70" which has supplanted the
similarly-named key "2" that existed on older platforms.

Bspec: 73210
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240815172602.2729146-4-matthew.d.roper@intel.com
9 months agodrm/xe: Add debugfs to dump GuC's hwconfig
Matt Roper [Thu, 15 Aug 2024 17:26:03 +0000 (10:26 -0700)]
drm/xe: Add debugfs to dump GuC's hwconfig

Although the query uapi is the official way to get at the GuC's hwconfig
table contents, it's still useful to have a quick debugfs interface to
dump the table in a human-readable format while debugging the driver.

Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Reviewed-by: Jagmeet Randhawa <jagmeet.randhawa@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240815172602.2729146-3-matthew.d.roper@intel.com
9 months agoMerge drm/drm-next into drm-xe-next
Lucas De Marchi [Fri, 16 Aug 2024 17:05:54 +0000 (10:05 -0700)]
Merge drm/drm-next into drm-xe-next

Get drm-xe-next on v6.11-rc2 and synchronized with drm-intel-next for
the display side. This resolves the current conflict for the
enable_display module parameter and allows further pending refactors.

Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
10 months agodrm/xe: Use for_each_remote_tile rather than manual check
Matthew Brost [Fri, 16 Aug 2024 04:02:08 +0000 (21:02 -0700)]
drm/xe: Use for_each_remote_tile rather than manual check

Replace for_each_tile plus a check against primary tile with
for_each_remote_tile in tiles_fini. The latter macro does this for us.

Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Reviewed-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240816040208.62695-1-matthew.brost@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
10 months agodrm/xe/uc: Use devm to register cleanup that includes exec_queues
Daniele Ceraolo Spurio [Thu, 15 Aug 2024 23:05:40 +0000 (16:05 -0700)]
drm/xe/uc: Use devm to register cleanup that includes exec_queues

Exec_queue cleanup requires HW access, so we need to use devm instead of
drmm for it.

Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: John Harrison <John.C.Harrison@Intel.com>
Cc: Alan Previn <alan.previn.teres.alexis@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240815230541.3828206-2-lucas.demarchi@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
10 months agodrm/xe/uc: Use managed bo for HuC and GSC objects
Daniele Ceraolo Spurio [Thu, 15 Aug 2024 23:05:39 +0000 (16:05 -0700)]
drm/xe/uc: Use managed bo for HuC and GSC objects

Drmm actions are not the right ones to clean up BOs and we should use
devm instead. However, we can also instead just allocate the objects
using the managed_bo function, which will internally register the
correct cleanup call and therefore allows us to simplify the code.

While at it, switch to drmm_kzalloc for the GSC proxy allocation to
further simplify the cleanup.

Cc: John Harrison <John.C.Harrison@Intel.com>
Cc: Alan Previn <alan.previn.teres.alexis@intel.com>
Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240815230541.3828206-1-lucas.demarchi@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
10 months agoMerge tag 'drm-intel-next-2024-08-13' of https://gitlab.freedesktop.org/drm/i915...
Dave Airlie [Fri, 16 Aug 2024 02:56:37 +0000 (12:56 +1000)]
Merge tag 'drm-intel-next-2024-08-13' of https://gitlab.freedesktop.org/drm/i915/kernel into drm-next

- Type-C programming fix for MTL+ (Gustavo)
 - Fix display clock workaround (Mitul)
 - Fix DP LTTPR detection (Imre)
 - Calculate vblank delay more accurately (Ville)
 - Make vrr_{enabling,disabling}() usable outside intel_display.c (Ville)
 - FBC clean-up (Ville)
 - DP link-training fixes and clean-up (Imre)
 - Make I2C terminology more inclusive (Easwar)
 - Make read-only array bw_gbps static const (Colin)
 - HDCP fixes and improvements (Suraj)
 - DP VSC SDP fixes and clean-ups (Suraj, Mitul)
 - Fix opregion leak in Xe code (Lucas)
 - Fix possible int overflow in skl_ddi_calculate_wrpll (Nikita)]
 - General display clean-ups and conversion towards intel_display (Jani)
 - On DP MST, Enable LT fallback for UHBR<->non-UHBR rates (Imre)
 - Add VRR condition for DPKGC Enablement (Suraj)
 - Use backlight power constants (Zimmermann)
 - Correct dual pps handling for MTL_PCH+ (Dnyaneshwar)
 - Dump DSC HW state (Imre)
 - Replace double blank with single blank after comma (Andi)
 - Read display register timeout on BMG (Mitul)

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/ZruWsyTv3nzdArDk@intel.com
10 months agodrm/xe: Fix tile fini sequence
Matthew Brost [Fri, 9 Aug 2024 23:28:30 +0000 (16:28 -0700)]
drm/xe: Fix tile fini sequence

Only set tile->mmio.regs to NULL if not the root tile in tile_fini. The
root tile mmio regs is setup ealier in MMIO init thus it should be set
to NULL in mmio_fini.

Fixes: dd08ebf6c352 ("drm/xe: Introduce a new DRM driver for Intel GPUs")
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Reviewed-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240809232830.3302251-1-matthew.brost@intel.com
10 months agodrm/xe: use devm instead of drmm for managed bo
Daniele Ceraolo Spurio [Fri, 9 Aug 2024 23:12:35 +0000 (16:12 -0700)]
drm/xe: use devm instead of drmm for managed bo

The BO cleanup touches the GGTT and therefore requires the HW to be
available, so we need to use devm instead of drmm.

Closes: https://gitlab.freedesktop.org/drm/xe/kernel/-/issues/1160
Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: Matthew Auld <matthew.auld@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240809231237.1503796-2-daniele.ceraolospurio@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
10 months agodrm/xe: Make exec_queue_kill safe to call twice
Daniele Ceraolo Spurio [Wed, 14 Aug 2024 20:56:54 +0000 (13:56 -0700)]
drm/xe: Make exec_queue_kill safe to call twice

An upcoming PXP patch will kill queues at runtime when a PXP
invalidation event occurs, so we need exec_queue_kill to be safe to call
multiple times.

v2: Add documentation (Matt B)

Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240814205654.1716586-1-daniele.ceraolospurio@intel.com
10 months agodrm/xe: fix engine_class bounds check again
Matthew Auld [Mon, 12 Aug 2024 14:13:32 +0000 (15:13 +0100)]
drm/xe: fix engine_class bounds check again

This was fixed in commit b7dce525c4fc ("drm/xe/queue: fix engine_class
bounds check"), but then re-introduced in commit 6f20fc09936e ("drm/xe:
Move and export xe_hw_engine lookup.") which should only be simple code
movement of the existing function.

Fixes: 6f20fc09936e ("drm/xe: Move and export xe_hw_engine lookup.")
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Dominik Grzegorzek <dominik.grzegorzek@intel.com>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240812141331.729843-2-matthew.auld@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
10 months agodrm/xe: Define STATELESS_COMPRESSION_CTRL as mcr register
Tejas Upadhyay [Wed, 14 Aug 2024 09:56:14 +0000 (15:26 +0530)]
drm/xe: Define STATELESS_COMPRESSION_CTRL as mcr register

Register STATELESS_COMPRESSION_CTRL should be considered
mcr register which should write to all slices as per
documentation.

Bspec: 71185
Fixes: ecabb5e6ce54 ("drm/xe/xe2: Add performance turning changes")
Signed-off-by: Tejas Upadhyay <tejas.upadhyay@intel.com>
Reviewed-by: Shekhar Chauhan <shekhar.chauhan@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240814095614.909774-4-tejas.upadhyay@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
10 months agodrm/xe: Write all slices if its mcr register
Tejas Upadhyay [Wed, 14 Aug 2024 09:56:13 +0000 (15:26 +0530)]
drm/xe: Write all slices if its mcr register

Register GAMREQSTRM_CTRL should be considered mcr register
which should write to all slices as per documentation.

Bspec: 71185
Fixes: 01570b446939 ("drm/xe/bmg: implement Wa_16023588340")
Reviewed-by: Matt Roper <matthew.d.roper@intel.com>
Signed-off-by: Tejas Upadhyay <tejas.upadhyay@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240814095614.909774-3-tejas.upadhyay@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
10 months agodrm/xe: Move enable host l2 VRAM post MCR init
Tejas Upadhyay [Wed, 14 Aug 2024 09:56:12 +0000 (15:26 +0530)]
drm/xe: Move enable host l2 VRAM post MCR init

xe_gt_enable_host_l2_vram() is reading the XE2_GAMREQSTRM_CTRL register
that is currently missing the MCR annotation. However, just adding the
annotation doesn't work as this function is called before MCR handling
is initialized in xe_gt_mcr_init().

xe_gt_enable_host_l2_vram() is used to implement WA 16023588340 that
needs to be done as early as possible during initialization in order to
be effective since the MMIO writes impact it. In the failure scenario,
driver would simply not be able to bind successfully.

Moving xe_gt_enable_host_l2_vram() later, after MCR initialization is
done, only incurs a few additional HW accesses, particularly when
loading GuC for hwconfig. Binding/unbinding the driver 100 times in BMG
still works so it should be ok to start handling the WA a little bit
later. This is sufficient to allow adding the MCR annotation to
XE2_GAMREQSTRM_CTRL.

Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Signed-off-by: Tejas Upadhyay <tejas.upadhyay@intel.com>
Reviewed-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240814095614.909774-2-tejas.upadhyay@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
10 months agodrm/xe: Rename enable_display module param
Lucas De Marchi [Tue, 13 Aug 2024 14:19:31 +0000 (07:19 -0700)]
drm/xe: Rename enable_display module param

The different approach used by xe regarding the initialization of
display HW has been proved a great addition for early driver bring up:
core xe can be tested without having all the bits sorted out on the
display side.

On the other hand, the approach exposed by i915-display is to *actively*
disable the display by programming it if needed, i.e. if it was left
enabled by firmware. It also has its use to make sure the HW is actually
disabled and not wasting power.

However having both the way it is in xe doesn't expose a good interface
wrt module params. From modinfo:

disable_display:Disable display (default: false) (bool)
enable_display:Enable display (bool)

Rename enable_display to probe_display to try to convey the message that
the HW is being touched and improve the module param description. To
avoid confusion, the enable_display is renamed everywhere, not only in
the module param. New description for the parameters:

disable_display:Disable display (default: false) (bool)
probe_display:Probe display HW, otherwise it's left untouched (default: true) (bool)

Reviewed-by: Matt Roper <matthew.d.roper@intel.com>
Acked-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240813141931.3141395-1-lucas.demarchi@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
10 months agodrm/xe: Remove unused xe parameter
Himal Prasad Ghimiray [Tue, 13 Aug 2024 10:44:19 +0000 (16:14 +0530)]
drm/xe: Remove unused xe parameter

Remove the xe parameter from the pde_encode_pat_index and
pte_encode_pat_index functions, as it is no longer used.

Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Reviewed-by: Tejas Upadhyay <tejas.upadhyay@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240813104419.2958046-1-himal.prasad.ghimiray@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
10 months agodrm/xe: Name and document Wa_14019789679
Matt Roper [Mon, 12 Aug 2024 18:10:43 +0000 (11:10 -0700)]
drm/xe: Name and document Wa_14019789679

Early in the development of Xe we identified an issue with SVG state
handling on DG2 and MTL (and later on Xe2 as well).  In
commit 72ac304769dd ("drm/xe: Emit SVG state on RCS during driver load
on DG2 and MTL") and commit fb24b858a20d ("drm/xe/xe2: Update SVG state
handling") we implemented our own workaround to prevent SVG state from
leaking from context A to context B in cases where context B never
issues a specific state setting.

The hardware teams have now created official workaround Wa_14019789679
to cover this issue.  The workaround description only requires emitting
3DSTATE_MESH_CONTROL, since they believe that's the only SVG instruction
that would potentially remain unset by a context B, but still cause
notable issues if unwanted values were inherited from context A.
However since we already have a more extensive implementation that emits
the entire SVG state and prevents _any_ SVG state from unintentionally
leaking, we'll stick with our existing implementation just to be safe.

Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240812181042.2013508-2-matthew.d.roper@intel.com
10 months agodrm/xe: add kdev_to_xe_device() helper and use it
Jani Nikula [Mon, 12 Aug 2024 10:30:21 +0000 (13:30 +0300)]
drm/xe: add kdev_to_xe_device() helper and use it

There are enough users for kernel device to xe device conversion, add a
helper for it.

Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/38c80846e70c7e410850530426384e17cff9d031.1723458544.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
10 months agodrm/xe: use pdev_to_xe_device() instead of pci_get_drvdata() directly
Jani Nikula [Mon, 12 Aug 2024 10:30:20 +0000 (13:30 +0300)]
drm/xe: use pdev_to_xe_device() instead of pci_get_drvdata() directly

We have a helper for converting pci device to xe device, use it.

Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/1b87c2e56200e001ce3a5d2f4a93eb26b294df32.1723458544.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
10 months agodrm/xe/tests: remove unused leftover xe_call_for_each_device()
Jani Nikula [Mon, 12 Aug 2024 10:30:19 +0000 (13:30 +0300)]
drm/xe/tests: remove unused leftover xe_call_for_each_device()

xe_call_for_each_device() has been unused since commit 57ecead343e7
("drm/xe/tests: Convert xe_mocs live tests"). Remove it and the related
dev_to_xe_device_fn() and struct kunit_test_data.

Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/fa3bb23d005313c9797f557e1211fde09fcb59cc.1723458544.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
10 months agodrm/xe/migrate: Parameterize ccs and bo data clear in xe_migrate_clear()
Nirmoy Das [Fri, 9 Aug 2024 22:03:47 +0000 (00:03 +0200)]
drm/xe/migrate: Parameterize ccs and bo data clear in xe_migrate_clear()

Parameterize clearing ccs and bo data in xe_migrate_clear() which  higher
layers can utilize. This patch will be used later on when doing bo data
clear for igfx as well.

v2: Replace multiple params with flags in xe_migrate_clear (Matt B)
v3: s/CLEAR_BO_DATA_FLAG_*/XE_MIGRATE_CLEAR_FLAG_* and move to
    xe_migrate.h. other nits(Matt B)

Cc: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Cc: Matthew Auld <matthew.auld@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Signed-off-by: Akshata Jahagirdar <akshata.jahagirdar@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240809220347.25330-1-nirmoy.das@intel.com
Signed-off-by: Nirmoy Das <nirmoy.das@intel.com>
10 months agodrm/i915: use pdev_to_i915() instead of pci_get_drvdata() directly
Jani Nikula [Mon, 12 Aug 2024 10:34:15 +0000 (13:34 +0300)]
drm/i915: use pdev_to_i915() instead of pci_get_drvdata() directly

We have a helper for converting pci device to i915 device, use it.

v2: Also convert i915_pci_probe() (Gustavo)

Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240812103415.1540096-1-jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
10 months agodrm/xe/xe2hpg: Add Wa_14021821874
Tejas Upadhyay [Mon, 12 Aug 2024 13:41:17 +0000 (19:11 +0530)]
drm/xe/xe2hpg: Add Wa_14021821874

Wa_14021821874 applies to xe2_hpg

V2(Himal):
  - Use space after define

Cc: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Signed-off-by: Tejas Upadhyay <tejas.upadhyay@intel.com>
Reviewed-by: Matt Roper <matthew.d.roper@intel.com>
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240812134117.813670-1-tejas.upadhyay@intel.com
10 months agodrm/xe: Add stats for tlb invalidation count
Nirmoy Das [Sat, 10 Aug 2024 19:15:19 +0000 (21:15 +0200)]
drm/xe: Add stats for tlb invalidation count

Add stats for tlb invalidation count which can be viewed with per GT
stat debugfs file.

Example output:
cat /sys/kernel/debug/dri/0/gt0/stats
tlb_inval_count: 22

v2: fix #include order(Tejas)

Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Sai Gowtham Ch <sai.gowtham.ch@intel.com>
Reviewed-by: Tejas Upadhyay <tejas.upadhyay@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240810191522.18616-2-nirmoy.das@intel.com
Signed-off-by: Nirmoy Das <nirmoy.das@intel.com>
10 months agodrm/xe/gt: Add APIs for printing stats over debugfs
Nirmoy Das [Sat, 10 Aug 2024 19:15:18 +0000 (21:15 +0200)]
drm/xe/gt: Add APIs for printing stats over debugfs

Add skeleton APIs for recording and printing various stats over
debugfs. This currently only added counter types stats which is backed
by atomic_t and wrapped with CONFIG_DRM_XE_STATS so this can be disabled
on production system.

v4: Rebase and other minor fixes (Matt)
v3: s/CONFIG_DRM_XE_STATS/CONFIG_DEBUG_FS(Lucas)
v2: add missing docs
    Add boundary checks for stats id and other improvements (Michal)
    Fix build when CONFIG_DRM_XE_STATS is disabled(Matt)

Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Sai Gowtham Ch <sai.gowtham.ch@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240810191522.18616-1-nirmoy.das@intel.com
Signed-off-by: Nirmoy Das <nirmoy.das@intel.com>
10 months agodrm/i915/bios: convert to struct intel_display
Jani Nikula [Fri, 9 Aug 2024 14:27:06 +0000 (17:27 +0300)]
drm/i915/bios: convert to struct intel_display

Going forward, struct intel_display shall replace struct
drm_i915_private as the main display device data pointer type. Convert
intel_bios.[ch] to struct intel_display.

Do one drive-by conversion of unnecessary hex usage to decimal.

Reviewed-by: Imre Deak <imre.deak@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/0d0261a53aff5f141b16b482222a5ffce78e176e.1723213547.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
10 months agodrm/i915/opregion: convert to struct intel_display
Jani Nikula [Fri, 9 Aug 2024 14:27:05 +0000 (17:27 +0300)]
drm/i915/opregion: convert to struct intel_display

Going forward, struct intel_display shall replace struct
drm_i915_private as the main display device data pointer type. Convert
intel_opregion.[ch] to struct intel_display.

v2:
- Fix declarations for !CONFIG_ACPI (Imre, kernel test robot)
- Pass encoder/connector directly to intel_display() (Imre)

Reviewed-by: Imre Deak <imre.deak@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/aef94503909bbbf95f0244dc382a4d4cd050b903.1723213547.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
10 months agodrm/i915/opregion: unify intel_encoder/intel_connector naming
Jani Nikula [Fri, 9 Aug 2024 14:27:04 +0000 (17:27 +0300)]
drm/i915/opregion: unify intel_encoder/intel_connector naming

Prefer the short encoder/connector names for struct
intel_encoder/intel_connector variables and parameters.

Reviewed-by: Imre Deak <imre.deak@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/60c67da6b7282ab521366524109ade0470408cf8.1723213547.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
10 months agodrm/i915/acpi: convert to struct intel_display
Jani Nikula [Fri, 9 Aug 2024 14:27:03 +0000 (17:27 +0300)]
drm/i915/acpi: convert to struct intel_display

Going forward, struct intel_display shall replace struct
drm_i915_private as the main display device data pointer type. Convert
intel_acpi.[ch] to struct intel_display.

Reviewed-by: Imre Deak <imre.deak@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/465436a3442807b49609fc55c9f652a29f96fd02.1723213547.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
10 months agodrm/i915/bmg: Read display register timeout
Mitul Golani [Wed, 7 Aug 2024 14:21:05 +0000 (19:51 +0530)]
drm/i915/bmg: Read display register timeout

Log the address of the register that caused the timeout
interrupt by reading RMTIMEOUTREG_CAPTURE

--v2:
- Update RMTIMEOUTREG_CAPTURE naming (Suraj)

--v3:
- XeLpdp naming convention.
- Use if condition instead of else if

Signed-off-by: Mitul Golani <mitulkumar.ajitkumar.golani@intel.com>
Reviewed-by: Suraj Kandpal <suraj.kandpal@intel.com>
Signed-off-by: Suraj Kandpal <suraj.kandpal@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240807142106.1270213-1-mitulkumar.ajitkumar.golani@intel.com
10 months agodrm/xe: Allow to compile out debugfs
Lucas De Marchi [Thu, 8 Aug 2024 17:11:21 +0000 (10:11 -0700)]
drm/xe: Allow to compile out debugfs

Use a dummy xe_debugfs_register() if debugfs is not enabled and move all
debugfs-related files under `ifeq ($(CONFIG_DEBUG_FS),y)` in the
Makefile. This is similar to what was done for display in
commit 439987f6f471 ("drm/xe: don't build debugfs files when
CONFIG_DEBUG_FS=n").

This removes the following warning while loading xe with
CONFIG_DEUBG_FS=n:

xe 0000:03:00.0: [drm] Create GT directory failed

Reviewed-by: Nirmoy Das <nirmoy.das@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240808171121.2484237-1-lucas.demarchi@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
10 months agodrm/xe: Allow suspend / resume to be safely called multiple times
Matthew Brost [Fri, 9 Aug 2024 19:19:29 +0000 (12:19 -0700)]
drm/xe: Allow suspend / resume to be safely called multiple times

Switching modes between LR and dma-fence can result in multiple calls to
suspend / resume. Make these calls safe while still enforcing call
order.

Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240809191929.3138956-6-matthew.brost@intel.com
10 months agodrm/xe: Only enable scheduling upon resume if needed
Matthew Brost [Fri, 9 Aug 2024 19:19:28 +0000 (12:19 -0700)]
drm/xe: Only enable scheduling upon resume if needed

No need to enable scheduling in already enabled.

Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240809191929.3138956-5-matthew.brost@intel.com
10 months agodrm/xe: Add xe_sched_add_msg_locked helper
Matthew Brost [Fri, 9 Aug 2024 19:19:27 +0000 (12:19 -0700)]
drm/xe: Add xe_sched_add_msg_locked helper

Will help by allowing callers to own message locking.

Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240809191929.3138956-4-matthew.brost@intel.com
10 months agodrm/xe: Reinit msg link when processing a message
Matthew Brost [Fri, 9 Aug 2024 19:19:26 +0000 (12:19 -0700)]
drm/xe: Reinit msg link when processing a message

Will help to avoid adding a static message twice.

Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240809191929.3138956-3-matthew.brost@intel.com
10 months agodrm/xe: Add xe_sched_msg_lock/unlock helper
Matthew Brost [Fri, 9 Aug 2024 19:19:25 +0000 (12:19 -0700)]
drm/xe: Add xe_sched_msg_lock/unlock helper

Will help callers to own locking when adding messages to scheduler.

Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240809191929.3138956-2-matthew.brost@intel.com
10 months agodrm/xe: Move VM dma-resv lock from xe_exec_queue_create to __xe_exec_queue_init
Matthew Brost [Wed, 24 Jul 2024 15:28:31 +0000 (08:28 -0700)]
drm/xe: Move VM dma-resv lock from xe_exec_queue_create to __xe_exec_queue_init

The critical section which requires the VM dma-resv is the call
xe_lrc_create in __xe_exec_queue_init. Move this lock to
__xe_exec_queue_init holding it just around xe_lrc_create. Not only is
good practice, this also fixes a locking double of the VM dma-resv in
the error paths of __xe_exec_queue_init as xe_lrc_put tries to acquire
this too resulting in a deadlock.

Fixes: dd08ebf6c352 ("drm/xe: Introduce a new DRM driver for Intel GPUs")
Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240724152831.1848325-1-matthew.brost@intel.com
10 months agoMerge tag 'drm-misc-next-2024-08-09' of https://gitlab.freedesktop.org/drm/misc/kerne...
Daniel Vetter [Fri, 9 Aug 2024 08:41:58 +0000 (10:41 +0200)]
Merge tag 'drm-misc-next-2024-08-09' of https://gitlab.freedesktop.org/drm/misc/kernel into drm-next

drm-misc-next for v6.12:

UAPI Changes:

- remove Power Saving Policy property

Core Changes:

- update connector documentation

CI:
- add tests for mediatek, meson, rockchip

Driver Changes:

amdgpu:
- revert support for Power Saving Policy property

bridge:
- lt9611uxc: require DRM_BRIDGE_ATTACH_NO_CONNECTOR

mgag200:
- transparently support BMC outputs

omapdrm:
- use common helper for_each_endpoint_of_node()

panel:
- panel-edp: fix name for HKC MB116AN01

vkms:
- clean up endianess warnings

Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
From: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://patchwork.freedesktop.org/patch/msgid/20240809071241.GA222501@localhost.localdomain
10 months agodrm/xe/pf: Fix VF config validation on multi-GT platforms
Michal Wajdeczko [Tue, 6 Aug 2024 18:05:16 +0000 (20:05 +0200)]
drm/xe/pf: Fix VF config validation on multi-GT platforms

When validating VF config on the media GT, we may wrongly report
that VF is already partially configured on it, as we consider GGTT
and LMEM provisioning done on the primary GT (since both GGTT and
LMEM are tile-level resources, not a GT-level).

This will cause skipping a VF auto-provisioning on the media-GT and
in result will block a VF from successfully initialize that GT.

Fix that by considering GGTT and LMEM configurations only when
checking if a VF provisioning is complete, and omit GGTT and LMEM
when reporting empty/partial provisioning.

Fixes: 234670cea9a2 ("drm/xe/pf: Skip fair VFs provisioning if already provisioned")
Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Piotr Piórkowski <piotr.piorkowski@intel.com>
Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240806180516.618-1-michal.wajdeczko@intel.com
10 months agodrm/mgag200: Remove BMC output
Thomas Zimmermann [Mon, 5 Aug 2024 13:06:01 +0000 (15:06 +0200)]
drm/mgag200: Remove BMC output

Mgag200's BMC connector tracks the status of an underlying physical
connector and updates the BMC status accordingly. This functionality
works around GNOME's settings app, which cannot handle multiple
outputs on the same CRTC.

The workaround is now obsolete as the VGA-BMC connector handles BMC
support internally. Hence, remove the driver's code and the BMC output
entirely.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Jocelyn Falempe <jfalempe@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240805130622.63458-6-tzimmermann@suse.de
10 months agodrm/mgag200: vga-bmc: Control BMC scanout from encoder
Thomas Zimmermann [Mon, 5 Aug 2024 13:06:00 +0000 (15:06 +0200)]
drm/mgag200: vga-bmc: Control BMC scanout from encoder

Move calls to stop and start BMC scanout from CRTC helpers to the
VGA-BMC encoder's atomic_disable and atomic_enable. Makes the BMC
scanout transparent to the CRTC.

DRM's atomic helpers call an encoder's atomic_disable and atomic_enable
helpers for all enabled encoders. The BMC stops scanning out the VGA
signal if modeset disables the VGA encoder, and starts scanning out
if the modeset enables the VGA encoder.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Jocelyn Falempe <jfalempe@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240805130622.63458-5-tzimmermann@suse.de
10 months agodrm/mgag200: vga-bmc: Control CRTC VIDRST flag from encoder
Thomas Zimmermann [Mon, 5 Aug 2024 13:05:59 +0000 (15:05 +0200)]
drm/mgag200: vga-bmc: Control CRTC VIDRST flag from encoder

Control the VIDRST pin from the VGA-BMC encoder's atomic_check and
remove the respective code from CRTC. Makes the VIDRST functionality
fully composable.

The VIDRST pin allows an external clock source to control the SYNC
signals of the Matrox chip. The functionality is part of the CRTC,
but depends on the presence of the clock source. This is the case for
some BMCs, so control the pin from the VGA-BMC output.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Jocelyn Falempe <jfalempe@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240805130622.63458-4-tzimmermann@suse.de
10 months agodrm/mgag200: vga-bmc: Transparently handle BMC
Thomas Zimmermann [Mon, 5 Aug 2024 13:05:58 +0000 (15:05 +0200)]
drm/mgag200: vga-bmc: Transparently handle BMC

The VGA-BMC connector selects the VGA output if a display has been
attached to the physical connector. Otherwise it selects the BMC
output. In any case, the connector status is set to 'detected', so
that the userspace compositor displays to it.

Depending on the setting, the connector's display modes either come
from the VGA monitor's EDID or from an internal list of BMC-compatible
modes.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Jocelyn Falempe <jfalempe@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240805130622.63458-3-tzimmermann@suse.de
10 months agodrm/mgag200: Add VGA-BMC output
Thomas Zimmermann [Mon, 5 Aug 2024 13:05:57 +0000 (15:05 +0200)]
drm/mgag200: Add VGA-BMC output

Duplicate VGA output to VGA-BMC output and update all code for Matrox
server chips. The new output represents a VGA output that has a BMC
attached to it. No functional changes so far.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Jocelyn Falempe <jfalempe@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240805130622.63458-2-tzimmermann@suse.de
10 months agodrm/xe: fix WA 14018094691
Daniele Ceraolo Spurio [Wed, 7 Aug 2024 23:53:32 +0000 (16:53 -0700)]
drm/xe: fix WA 14018094691

This WA is applied while initializing the media GT, but it a primary
GT WA (because it modifies a register on the primary GT), so the XE_WA
macro is returning false even when the WA should be applied.
Fix this by using the primary GT in the macro.

Note that this WA only applies to PXP and we don't yet support that in
Xe, so there are no negative effects to this bug, which is why we didn't
see any errors in testing.

v2: use the primary GT in the macro instead of marking the WA as
platform-wide (Lucas, Matt).

Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Matt Roper <matthew.d.roper@intel.com>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Reviewed-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240807235333.1370915-1-daniele.ceraolospurio@intel.com
10 months agodrm/xe/guc: Enable w/a 14022293748 and 22019794406
Julia Filipchuk [Mon, 5 Aug 2024 20:54:35 +0000 (13:54 -0700)]
drm/xe/guc: Enable w/a 14022293748 and 22019794406

Enable workarounds for HW bug where render engine reset fails. Given
that we're bumping the minimum required GuC version to 70.29, we're
guaranteed to always have support for this KLV in the GuC.

v2: Enable KLV correctly for either workaround (Lucas)
v4: Add check for minimum supported GuC firmware version. Enable w/a for
hw version 20.01 too. (Daniele)
v5 (Daniele): remove now unneeded fw type and version checks (JohnH)

Signed-off-by: Julia Filipchuk <julia.filipchuk@intel.com>
Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240805205435.921921-1-daniele.ceraolospurio@intel.com
10 months agoMerge tag 'drm-misc-next-2024-08-01' of https://gitlab.freedesktop.org/drm/misc/kerne...
Daniel Vetter [Thu, 8 Aug 2024 16:58:45 +0000 (18:58 +0200)]
Merge tag 'drm-misc-next-2024-08-01' of https://gitlab.freedesktop.org/drm/misc/kernel into drm-next

drm-misc-next for v6.12:

UAPI Changes:

virtio:
- Define DRM capset

Cross-subsystem Changes:

dma-buf:
- heaps: Clean up documentation

printk:
- Pass description to kmsg_dump()

Core Changes:

CI:
- Update IGT tests
- Point upstream repo to GitLab instance

modesetting:
- Introduce Power Saving Policy property for connectors
- Add might_fault() to drm_modeset_lock priming
- Add dynamic per-crtc vblank configuration support

panic:
- Avoid build-time interference with framebuffer console

docs:
- Document Colorspace property

scheduler:
- Remove full_recover from drm_sched_start

TTM:
- Make LRU walk restartable after dropping locks
- Allow direct reclaim to allocate local memory

Driver Changes:

amdgpu:
- Support Power Saving Policy connector property

ast:
- astdp: Support AST2600 with VGA; Clean up HPD

bridge:
- Silence error message on -EPROBE_DEFER
- analogix: Clean aup
- bridge-connector: Fix double free
- lt6505: Disable interrupt when powered off
- tc358767: Make default DP port preemphasis configurable

gma500:
- Update i2c terminology

ivpu:
- Add MODULE_FIRMWARE()

lcdif:
- Fix pixel clock

loongson:
- Use GEM refcount over TTM's

mgag200:
- Improve BMC handling
- Support VBLANK intterupts

nouveau:
- Refactor and clean up internals
- Use GEM refcount over TTM's

panel:
- Shutdown fixes plus documentation
- Refactor several drivers for better code sharing
- boe-th101mb31ig002: Support for starry-er88577 MIPI-DSI panel plus
  DT; Fix porch parameter
- edp: Support AOU B116XTN02.3, AUO B116XAN06.1, AOU B116XAT04.1,
  BOE NV140WUM-N41, BOE NV133WUM-N63, BOE NV116WHM-A4D, CMN N116BCA-EA2,
  CMN N116BCP-EA2, CSW MNB601LS1-4
- himax-hx8394: Support Microchip AC40T08A MIPI Display panel plus DT
- ilitek-ili9806e: Support Densitron DMT028VGHMCMI-1D TFT plus DT
- jd9365da: Support Melfas lmfbx101117480 MIPI-DSI panel plus DT; Refactor
  for code sharing

sti:
- Fix module owner

stm:
- Avoid UAF wih managed plane and CRTC helpers
- Fix module owner
- Fix error handling in probe
- Depend on COMMON_CLK
- ltdc: Fix transparency after disabling plane; Remove unused interrupt

tegra:
- Call drm_atomic_helper_shutdown()

v3d:
- Clean up perfmon

vkms:
- Clean up

Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
From: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://patchwork.freedesktop.org/patch/msgid/20240801121406.GA102996@linux.fritz.box
10 months agodrm/i915: remove __i915_printk()
Jani Nikula [Tue, 6 Aug 2024 13:38:32 +0000 (16:38 +0300)]
drm/i915: remove __i915_printk()

With the previous cleanups, the last remaining user of __i915_printk()
is i915_probe_error(). Switch that to use drm_dbg() and drm_err()
instead, dropping the request to report bugs in the few remaining
specific cases.

It's not common for drivers to log bug filing requests to begin with,
but these cases are in init, which is most likely to be tested in CI and
least likely to be hit by end users anyway.

Acked-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/be9baeab281f75999e96cc7ad1c06c6680494bc1.1722951405.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
10 months agodrm/i915: remove i915_report_error()
Jani Nikula [Tue, 6 Aug 2024 13:38:31 +0000 (16:38 +0300)]
drm/i915: remove i915_report_error()

i915_report_error() presently acts as a wrapper for __i915_printk(). In
practice, it would be better to use drm level error reporting wherever
possible, so replace all uses of i915_report_error() with the equivalent
drm_err() call. These cases are not worth having a dedicated wrapper to
also print bug reporting info. Replacing the calls leaves
i915_report_error() with no users, so remove it.

Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/19eab020c57c0fa45acacf4e4a8077e57cd4d561.1722951405.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
10 months agodrm/i915: remove a few __i915_printk() uses
Jani Nikula [Tue, 6 Aug 2024 13:38:30 +0000 (16:38 +0300)]
drm/i915: remove a few __i915_printk() uses

__i915_printk() does nothing special for notice/info levels. Just use
the regular drm_notice() and drm_info() calls.

Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/82857a0c04d3c11ca6758f05c13a3cec4f1a2f01.1722951405.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
10 months agodrm/i915: Replace double blank with single blank after comma
Andi Shyti [Wed, 7 Aug 2024 13:05:16 +0000 (14:05 +0100)]
drm/i915: Replace double blank with single blank after comma

Do not use double blanks, ",  " in function parameters where it's
not required by any alignment purpose. Replase it with a single
blank, ", ".

Signed-off-by: Andi Shyti <andi.shyti@linux.intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240807130516.491053-3-andi.shyti@linux.intel.com
10 months agodrm/vkms: Fix cpu_to_le16()/le16_to_cpu() warnings
José Expósito [Tue, 16 Jul 2024 16:17:26 +0000 (18:17 +0200)]
drm/vkms: Fix cpu_to_le16()/le16_to_cpu() warnings

Building with Sparse enabled prints this warning for cpu_to_le16()
calls:

    warning: incorrect type in assignment (different base types)
        expected unsigned short [usertype]
        got restricted __le16 [usertype]

And this warning for le16_to_cpu() calls:

    warning: cast to restricted __le16

Declare the target buffer as __le16 to fix both warnings.

Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Louis Chauvet <louis.chauvet@bootlin.com>
Acked-by: Maíra Canal <mcanal@igalia.com>
Signed-off-by: José Expósito <jose.exposito89@gmail.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240716161725.41408-2-jose.exposito89@gmail.com
10 months agodrm/xe: Fix access_ok check in user_fence_create
Nirmoy Das [Tue, 6 Aug 2024 11:07:22 +0000 (13:07 +0200)]
drm/xe: Fix access_ok check in user_fence_create

Check size of the data not size of the pointer.

Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202407300421.IBkAja96-lkp@intel.com/
Fixes: 0fde907da2d5 ("drm/xe: Validate user fence during creation")
Cc: Matthew Auld <matthew.auld@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Reviewed-by: Tejas Upadhyay <tejas.upadhyay@intel.com>
Reviewed-by: Apoorva Singh <apoorva.singh@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240806110722.28661-1-nirmoy.das@intel.com
Signed-off-by: Nirmoy Das <nirmoy.das@intel.com>
10 months agogpu: drm: use for_each_endpoint_of_node()
Kuninori Morimoto [Tue, 30 Jul 2024 00:34:40 +0000 (00:34 +0000)]
gpu: drm: use for_each_endpoint_of_node()

We already have for_each_endpoint_of_node(), don't use
of_graph_get_next_endpoint() directly. Replace it.

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Acked-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen+renesas@ideasonboard.com>
Link: https://patchwork.freedesktop.org/patch/msgid/87jzh3lnts.wl-kuninori.morimoto.gx@renesas.com
10 months agodrm/xe: Add kernel doc for xe_hw_engine_lookup
Mika Kuoppala [Tue, 6 Aug 2024 15:30:09 +0000 (18:30 +0300)]
drm/xe: Add kernel doc for xe_hw_engine_lookup

kerneldoc was missing from earlier commit where
we exported xe_hw_engine_lookup. Add it.

Cc: Dominik Grzegorzek <dominik.grzegorzek@intel.com>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: "Thomas Hellström" <thomas.hellstrom@linux.intel.com>
Cc: intel-xe@lists.freedesktop.org
Signed-off-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240806153009.1081382-2-mika.kuoppala@linux.intel.com
10 months agodrm/xe: Export xe_hw_engine's mmio accessors
Dominik Grzegorzek [Tue, 6 Aug 2024 15:30:08 +0000 (18:30 +0300)]
drm/xe: Export xe_hw_engine's mmio accessors

Export hw engine's mmio accessors. This is in preparation
to use these from eudebug code.

v2: s/hw_engine_mmio/xe_hw_engine_mmio (Matthew)
v3: kernel doc (Matthew)

Cc: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Dominik Grzegorzek <dominik.grzegorzek@intel.com>
Signed-off-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240806153009.1081382-1-mika.kuoppala@linux.intel.com
10 months agodrm/xe/guc: Define GuC version v70.29.2 for BMG
Julia Filipchuk [Fri, 2 Aug 2024 22:21:28 +0000 (15:21 -0700)]
drm/xe/guc: Define GuC version v70.29.2 for BMG

UAPI version 1.13.4

Signed-off-by: Julia Filipchuk <julia.filipchuk@intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240802222129.3976212-4-John.C.Harrison@Intel.com
10 months agodrm/xe/guc: Bump minimum required GuC version to v70.29.2
Julia Filipchuk [Fri, 2 Aug 2024 22:21:27 +0000 (15:21 -0700)]
drm/xe/guc: Bump minimum required GuC version to v70.29.2

The VF API version for this release is 1.13.4.

Bumping the minimum required GuC version just before force-probe
removal allows us to set a baseline for what API features are expected
to be available. I.e., at this point there is no need for any version
checking in the code before using a feature. Of course, if/when the
API is extended in future GuC releases, those new features will need
API version checks in the code.

Bump the recommended GuC versions to match.

Also add numerical comparison helpers to simplify the version number
checks.

v2: Reword commit message and make comparison helpers GuC specific -
review feedback from Daniele, done by JohnH

Signed-off-by: Julia Filipchuk <julia.filipchuk@intel.com>
Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240802222129.3976212-3-John.C.Harrison@Intel.com
10 months agodrm/xe/xe2: Add performance turning changes
Shekhar Chauhan [Mon, 5 Aug 2024 05:37:10 +0000 (11:07 +0530)]
drm/xe/xe2: Add performance turning changes

Update performance tuning according to the hardware spec.

Bspec: 72161
Signed-off-by: Shekhar Chauhan <shekhar.chauhan@intel.com>
Reviewed-by: Sai Teja Pottumuttu <sai.teja.pottumuttu@intel.com>
Reviewed-by: Akshata Jahagirdar <akshata.jahagirdar@intel.com>
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240805053710.877119-1-shekhar.chauhan@intel.com
10 months agodrm/ci: rockchip: add tests for rockchip display driver
Vignesh Raman [Tue, 30 Jul 2024 02:15:41 +0000 (07:45 +0530)]
drm/ci: rockchip: add tests for rockchip display driver

For rockchip rk3288 and rk3399, the display driver is rockchip
and gpu driver is panfrost. Currently, in drm-ci for rockchip
rk3288 and rk3399, only the gpu driver is tested. Refactor
the existing rockchip jobs to test both display and gpu driver
and update xfails.

Since the correct driver name is passed from the job to test gpu
and display driver, remove the check to set IGT_FORCE_DRIVER
based on driver name for rockchip jobs.

Acked-by: Helen Koike <helen.koike@collabora.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>
Signed-off-by: Vignesh Raman <vignesh.raman@collabora.com>
Signed-off-by: Helen Koike <helen.koike@collabora.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240730021545.912271-7-vignesh.raman@collabora.com
10 months agodrm/ci: meson: add tests for meson display driver
Vignesh Raman [Tue, 30 Jul 2024 02:15:40 +0000 (07:45 +0530)]
drm/ci: meson: add tests for meson display driver

For Amlogic Meson G12B (A311D) SOC the display driver is meson and
gpu driver is panfrost. Currently, in drm-ci for Meson G12B (A311D),
only the gpu driver is tested. Refactor the existing meson jobs
and add support in drm-ci to test both display and gpu driver for
Amlogic Meson G12B (A311D) and update xfails.

Since the correct driver name is passed from the job to test gpu
and display driver, remove the check to set IGT_FORCE_DRIVER
based on driver name for meson jobs.

Acked-by: Helen Koike <helen.koike@collabora.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>
Signed-off-by: Vignesh Raman <vignesh.raman@collabora.com>
Signed-off-by: Helen Koike <helen.koike@collabora.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240730021545.912271-6-vignesh.raman@collabora.com
10 months agodrm/ci: mediatek: add tests for powervr gpu driver
Vignesh Raman [Tue, 30 Jul 2024 02:15:39 +0000 (07:45 +0530)]
drm/ci: mediatek: add tests for powervr gpu driver

For mediatek mt8173, the display driver is mediatek, while the
gpu driver is powervr. Currently, in drm-ci for mt8173, only the
display driver is tested. Add support in drm-ci to test powervr
driver for mt8173. Powervr driver was merged in linux kernel,
but there's no mediatek support yet. So disable the powervr:mt8173
job which uses powervr driver.

Also update the MAINTAINERS file to include xfails for powervr driver.

Acked-by: Helen Koike <helen.koike@collabora.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>
Signed-off-by: Vignesh Raman <vignesh.raman@collabora.com>
Signed-off-by: Helen Koike <helen.koike@collabora.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240730021545.912271-5-vignesh.raman@collabora.com
10 months agodrm/ci: mediatek: add tests for mediatek display driver
Vignesh Raman [Tue, 30 Jul 2024 02:15:38 +0000 (07:45 +0530)]
drm/ci: mediatek: add tests for mediatek display driver

For mediatek mt8183, the display driver is mediatek, while the
gpu driver is panfrost. Currently, in drm-ci for mt8183, only
the gpu driver is tested. Refactor the existing mediatek jobs
and add support in drm-ci to test both display and gpu driver
for mt8183 and update xfails.

Since the correct driver name is passed from the job to test gpu
and display driver, remove the check to set IGT_FORCE_DRIVER
based on driver name for mediatek jobs.

Update the MAINTAINERS file to include xfails for panfrost driver.

Acked-by: Helen Koike <helen.koike@collabora.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>
Signed-off-by: Vignesh Raman <vignesh.raman@collabora.com>
Signed-off-by: Helen Koike <helen.koike@collabora.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240730021545.912271-4-vignesh.raman@collabora.com
10 months agodrm/ci: skip tools_test on non-intel platforms
Vignesh Raman [Tue, 30 Jul 2024 02:15:37 +0000 (07:45 +0530)]
drm/ci: skip tools_test on non-intel platforms

tools_test is a wrapper for running tools/intel_reg
and tools/intel_l3_parity. So skip these tests on
non-intel platforms and update expectaion files.

Acked-by: Helen Koike <helen.koike@collabora.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>
Signed-off-by: Vignesh Raman <vignesh.raman@collabora.com>
Signed-off-by: Helen Koike <helen.koike@collabora.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240730021545.912271-3-vignesh.raman@collabora.com
10 months agodrm/ci: arm64.config: Enable CONFIG_DRM_ANALOGIX_ANX7625
Vignesh Raman [Tue, 30 Jul 2024 02:15:36 +0000 (07:45 +0530)]
drm/ci: arm64.config: Enable CONFIG_DRM_ANALOGIX_ANX7625

Enable CONFIG_DRM_ANALOGIX_ANX7625 in the arm64 defconfig to get
display driver probed on the mt8183-kukui-jacuzzi-juniper machine.

arch/arm64/configs/defconfig has CONFIG_DRM_ANALOGIX_ANX7625=m,
but drm-ci don't have initrd with modules, so add
CONFIG_DRM_ANALOGIX_ANX7625=y in CI arm64 config.

Acked-by: Helen Koike <helen.koike@collabora.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>
Signed-off-by: Vignesh Raman <vignesh.raman@collabora.com>
Signed-off-by: Helen Koike <helen.koike@collabora.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240730021545.912271-2-vignesh.raman@collabora.com
10 months agodrm/bridge: lt9611uxc: drop support for !DRM_BRIDGE_ATTACH_NO_CONNECTOR
Dmitry Baryshkov [Mon, 1 Jul 2024 18:19:53 +0000 (21:19 +0300)]
drm/bridge: lt9611uxc: drop support for !DRM_BRIDGE_ATTACH_NO_CONNECTOR

Having no in-kernel devices that use !DRM_BRIDGE_ATTACH_NO_CONNECTOR
mode for the Lontium LT9611UXC bridge, drop the in-bridge implementation
of the drm_connector.

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Robert Foss <rfoss@kernel.org>
Signed-off-by: Robert Foss <rfoss@kernel.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20240701-lt9611uxc-next-bridge-v1-2-665bce5fdaaa@linaro.org
10 months agodrm/bridge: lt9611uxc: properly attach to a next bridge
Dmitry Baryshkov [Mon, 1 Jul 2024 18:19:52 +0000 (21:19 +0300)]
drm/bridge: lt9611uxc: properly attach to a next bridge

If there is a next bridge in the OF graph don't let it be ignored.
Attach the next bridge to the chain.

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Robert Foss <rfoss@kernel.org>
Signed-off-by: Robert Foss <rfoss@kernel.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20240701-lt9611uxc-next-bridge-v1-1-665bce5fdaaa@linaro.org
10 months agodrm/i915: Remove DSC register dump
Imre Deak [Mon, 5 Aug 2024 15:07:56 +0000 (18:07 +0300)]
drm/i915: Remove DSC register dump

The Display Engine's DSC register values are deducted from the DSC
configuration stored in intel_crtc_state::dsc. The latter one is
dumped in a human-readable format, so dumping the register values is
redundant, remove it.

Acked-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240805150802.3568970-8-imre.deak@intel.com
10 months agodrm/i915: Dump DSC state to dmesg and debugfs/i915_display_info
Imre Deak [Mon, 5 Aug 2024 15:07:55 +0000 (18:07 +0300)]
drm/i915: Dump DSC state to dmesg and debugfs/i915_display_info

Dump the DSC state to dmesg during HW readout and state computation as
well as the i915_display_info debugfs entry.

v2: Rebase on the s/DRM_X16/FXP_Q4 change.

Acked-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240805150802.3568970-7-imre.deak@intel.com
10 months agodrm/i915: Replace BPP_X16_FMT()/ARGS() with FXP_Q4_FMT()/ARGS()
Imre Deak [Mon, 5 Aug 2024 15:07:54 +0000 (18:07 +0300)]
drm/i915: Replace BPP_X16_FMT()/ARGS() with FXP_Q4_FMT()/ARGS()

Replace the BPP_X16_FMT()/ARGS() helpers defined by the driver with the
equivalent FXP_Q4_FMT()/ARGS() helpers defined by DRM core.

v2: Rebase on the s/DRM_X16/FXP_Q4 change.

Acked-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240805150802.3568970-6-imre.deak@intel.com
10 months agodrm/i915: Replace to_bpp_frac() with fxp_q4_to_frac()
Imre Deak [Mon, 5 Aug 2024 15:07:53 +0000 (18:07 +0300)]
drm/i915: Replace to_bpp_frac() with fxp_q4_to_frac()

Replace the to_bpp_frac() helper defined by the driver with the
equivalent fxp_q4_to_frac() helper defined by DRM core.

v2: Rebase on the s/drm_x16/fxp_q4 change.

Acked-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240805150802.3568970-5-imre.deak@intel.com
10 months agodrm/i915: Replace to_bpp_int_roundup() with fxp_q4_to_int_roundup()
Imre Deak [Mon, 5 Aug 2024 15:07:52 +0000 (18:07 +0300)]
drm/i915: Replace to_bpp_int_roundup() with fxp_q4_to_int_roundup()

Replace the to_bpp_int_roundup() helper defined by the driver with the
equivalent fxp_q4_to_int_roundup() helper defined by DRM core.

v2: Rebase on s/drm_x16/fxp_q4 change.

Acked-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240805150802.3568970-4-imre.deak@intel.com
10 months agodrm/i915: Replace to_bpp_int() with fxp_q4_to_int()
Imre Deak [Mon, 5 Aug 2024 15:07:51 +0000 (18:07 +0300)]
drm/i915: Replace to_bpp_int() with fxp_q4_to_int()

Replace the to_bpp_int() helper defined by the driver with the
equivalent fxp_q4_to_int() helper defined by DRM core.

v2: Rebase on the s/drm_x16/fxp_q4 change.

Acked-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240805150802.3568970-3-imre.deak@intel.com
10 months agodrm/i915: Replace to_bpp_x16() with fxp_q4_from_int()
Imre Deak [Mon, 5 Aug 2024 15:07:50 +0000 (18:07 +0300)]
drm/i915: Replace to_bpp_x16() with fxp_q4_from_int()

Replace the to_bpp_x16() helper defined by the driver with the
equivalent fxp_q4_from_int() helper defined by DRM core.

v2: Rebase on the s/drm_x16/fxp_q4 change.

Acked-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240805150802.3568970-2-imre.deak@intel.com
10 months agodrm/xe: Only check last fence on user binds
Matthew Brost [Mon, 5 Aug 2024 20:02:33 +0000 (13:02 -0700)]
drm/xe: Only check last fence on user binds

We only set the last fence on user binds, so no need to check last fence
kernel issued binds. Will avoid blowing up last fence lockdep asserts.

Cc: Francois Dugast <francois.dugast@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240805200233.3050325-1-matthew.brost@intel.com
10 months agodrm/xe/observation: Drop empty sysctl table entry
Ashutosh Dixit [Mon, 5 Aug 2024 06:20:57 +0000 (23:20 -0700)]
drm/xe/observation: Drop empty sysctl table entry

An empty sysctl table entry was inadvertently left behind for observation
sysctl. The breaks on 6.11 with the following errors:

[  219.654850] sysctl table check failed: dev/xe/(null) procname is null
[  219.654862] sysctl table check failed: dev/xe/(null) No proc_handler

Drop the empty entry.

Fixes: 8169b2097d88 ("drm/xe/uapi: Rename xe perf layer as xe observation layer")
Closes: https://gitlab.freedesktop.org/drm/xe/kernel/-/issues/2419
Signed-off-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
Reviewed-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240805062057.3547560-1-ashutosh.dixit@intel.com
10 months agodrm/connector: Document destroy hook in drmm init functions
José Expósito [Sun, 4 Aug 2024 17:05:52 +0000 (19:05 +0200)]
drm/connector: Document destroy hook in drmm init functions

Document that the drm_connector_funcs.destroy hook must be NULL in
drmm_connector_init() and drmm_connector_hdmi_init().

Signed-off-by: José Expósito <jose.exposito89@gmail.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: https://patchwork.freedesktop.org/patch/msgid/20240804170551.33971-2-jose.exposito89@gmail.com
10 months agodrm/i915/display: correct dual pps handling for MTL_PCH+
Dnyaneshwar Bhadane [Thu, 1 Aug 2024 11:11:41 +0000 (16:41 +0530)]
drm/i915/display: correct dual pps handling for MTL_PCH+

On the PCH side the second PPS was introduced in ICP+.Add condition
On MTL_PCH and greater platform also having the second PPS.

Note that DG1/2 south block only has the single PPS, so need
to exclude the fake DG1/2 PCHs

Closes: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/11488
Fixes: 93cbc1accbce ("drm/i915/mtl: Add fake PCH for Meteor Lake")
Cc: <stable@vger.kernel.org> # v6.9+
Signed-off-by: Dnyaneshwar Bhadane <dnyaneshwar.bhadane@intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240801111141.574854-1-dnyaneshwar.bhadane@intel.com
10 months agoMerge tag 'drm-xe-next-2024-07-30' of https://gitlab.freedesktop.org/drm/xe/kernel...
Dave Airlie [Mon, 5 Aug 2024 08:47:30 +0000 (18:47 +1000)]
Merge tag 'drm-xe-next-2024-07-30' of https://gitlab.freedesktop.org/drm/xe/kernel into drm-next

drm-xe-next for 6.12

UAPI Changes:
- Rename xe perf layer as xe observation layer, but was
  also made available via fixes to previous verison (Ashutosh)
- Use write-back caching mode for system memory on DGFX,
  but was also mad available via fixes to previous version (Thomas)
- Expose SIMD16 EU mask in topology query for userspace to know
  the type of EU, as available in PVC, Lunar Lake and Battlemage
  (Lucas)
- Return ENOBUFS instead of ENOMEM in vm_bind if failure is tied
  to an array of binds (Matthew Brost)

Driver Changes:
- Log cleanup moving messages to debug priority (Michal Wajdeczko)
- Add timeout to fences to adhere to dma_buf rules (Matthew Brost)
- Rename old engine nomenclature to exec_queue (Matthew Brost)
- Convert multiple bind ops to 1 job (Matthew Brost)
- Add error injection for vm bind to help testing error path
  (Matthew Brost)
- Fix error handling in page table to propagate correctly
  to userspace (Matthew Brost)
- Re-organize and cleanup SR-IOV related registers (Michal Wajdeczko)
- Make the device write barrier compatible with VF (Michal Wajdeczko)
- New display workarounds for Battlemage (Matthew  Auld)
- New media workarounds for Lunar Lake and Battlemage (Ngai-Mint Kwan)
- New graphics workarounds for Lunar Lake (Bommu Krishnaiah)
- Tracepoint updates (Matthew Brost, Nirmoy Das)
- Cleanup the header generation for OOB workarounds (Lucas De Marchi)
- Fix leaking HDCP-related object (Nirmoy Das)
- Serialize L2 flushes to avoid races (Tejas Upadhyay)
- Log pid and comm on job timeout (José Roberto de Souza)
- Simplify boilerplate code for live kunit (Michal Wajdeczko)
- Improve kunit skips for live kunit (Michal Wajdeczko)
- Fix xe_sync cleanup when handling xe_exec ioctl (Ashutosh Dixit)
- Limit fair VF LMEM provisioning (Michal Wajdeczko)
- New workaround to fence mmio writes in Lunar Lake (Tejas Upadhyay)
- Warn on writes inaccessible register in VF (Michal Wajdeczko)
- Fix register lookup in VF (Michal Wajdeczko)
- Add GSC support for Battlemage (Alexander Usyskin)
- Fix wedging only the GT in which timeout occurred (Matthew Brost)
- Block device suspend when wedging (Matthew Brost)
- Handle compression and migration changes for Battlemage
  (Akshata Jahagirdar)
- Limit access of stolen memory for Lunar Lake (Uma Shankar)
- Fail invalid addresses during user fence creation (Matthew Brost)
- Refcount xe_file to safely and accurately store fdinfo stats
  (Umesh Nerlige Ramappa)
- Cleanup and fix PM reference for TLB invalidation code
  (Matthew Brost)
- Fix PM reference handling when communicating with GuC (Matthew Brost)
- Add new BO flag for 2 MiB alignement and use in VF (Michal Wajdeczko)
- Simplify MMIO setup for multi-tile platforms (Lucas De Marchi)
- Add check for uninitialized access to OOB workarounds
  (Lucas De Marchi)
- New GSC and HuC firmware blobs for Lunar Lake and Battlemage
  (Daniele Ceraolo Spurio)
- Unify mmio wait logic (Gustavo Sousa)
- Fix off-by-one when processing RTP rules (Lucas De Marchi)
- Future-proof migrate logic with compressed PAT flag (Matt Roper)
- Add WA kunit tests for Battlemage (Lucas De Marchi)
- Test active tracking for workaorunds with kunit (Lucas De Marchi)
- Add kunit tests for RTP with no actions (Lucas De Marchi)
- Unify parse of OR rules in RTP (Lucas De Marchi)
- Add performance tuning for Battlemage (Sai Teja Pottumuttu)
- Make bit masks unsigned (Geert Uytterhoeven)

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/k7xuktfav4zmtxxjr77glu2hszypvzgmzghoumh757nqfnk7kn@ccfi4ts3ytbk
10 months agoLinux 6.11-rc2
Linus Torvalds [Sun, 4 Aug 2024 20:50:53 +0000 (13:50 -0700)]
Linux 6.11-rc2

10 months agoprofiling: remove profile=sleep support
Tetsuo Handa [Sun, 4 Aug 2024 09:48:10 +0000 (18:48 +0900)]
profiling: remove profile=sleep support

The kernel sleep profile is no longer working due to a recursive locking
bug introduced by commit 42a20f86dc19 ("sched: Add wrapper for get_wchan()
to keep task blocked")

Booting with the 'profile=sleep' kernel command line option added or
executing

  # echo -n sleep > /sys/kernel/profiling

after boot causes the system to lock up.

Lockdep reports

  kthreadd/3 is trying to acquire lock:
  ffff93ac82e08d58 (&p->pi_lock){....}-{2:2}, at: get_wchan+0x32/0x70

  but task is already holding lock:
  ffff93ac82e08d58 (&p->pi_lock){....}-{2:2}, at: try_to_wake_up+0x53/0x370

with the call trace being

   lock_acquire+0xc8/0x2f0
   get_wchan+0x32/0x70
   __update_stats_enqueue_sleeper+0x151/0x430
   enqueue_entity+0x4b0/0x520
   enqueue_task_fair+0x92/0x6b0
   ttwu_do_activate+0x73/0x140
   try_to_wake_up+0x213/0x370
   swake_up_locked+0x20/0x50
   complete+0x2f/0x40
   kthread+0xfb/0x180

However, since nobody noticed this regression for more than two years,
let's remove 'profile=sleep' support based on the assumption that nobody
needs this functionality.

Fixes: 42a20f86dc19 ("sched: Add wrapper for get_wchan() to keep task blocked")
Cc: stable@vger.kernel.org # v5.16+
Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
10 months agoMerge tag 'x86-urgent-2024-08-04' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Sun, 4 Aug 2024 15:57:08 +0000 (08:57 -0700)]
Merge tag 'x86-urgent-2024-08-04' of git://git./linux/kernel/git/tip/tip

Pull x86 fixes from Thomas Gleixner:

 - Prevent a deadlock on cpu_hotplug_lock in the aperf/mperf driver.

   A recent change in the ACPI code which consolidated code pathes moved
   the invocation of init_freq_invariance_cppc() to be moved to a CPU
   hotplug handler. The first invocation on AMD CPUs ends up enabling a
   static branch which dead locks because the static branch enable tries
   to acquire cpu_hotplug_lock but that lock is already held write by
   the hotplug machinery.

   Use static_branch_enable_cpuslocked() instead and take the hotplug
   lock read for the Intel code path which is invoked from the
   architecture code outside of the CPU hotplug operations.

 - Fix the number of reserved bits in the sev_config structure bit field
   so that the bitfield does not exceed 64 bit.

 - Add missing Zen5 model numbers

 - Fix the alignment assumptions of pti_clone_pgtable() and
   clone_entry_text() on 32-bit:

   The code assumes PMD aligned code sections, but on 32-bit the kernel
   entry text is not PMD aligned. So depending on the code size and
   location, which is configuration and compiler dependent, entry text
   can cross a PMD boundary. As the start is not PMD aligned adding PMD
   size to the start address is larger than the end address which
   results in partially mapped entry code for user space. That causes
   endless recursion on the first entry from userspace (usually #PF).

   Cure this by aligning the start address in the addition so it ends up
   at the next PMD start address.

   clone_entry_text() enforces PMD mapping, but on 32-bit the tail might
   eventually be PTE mapped, which causes a map fail because the PMD for
   the tail is not a large page mapping. Use PTI_LEVEL_KERNEL_IMAGE for
   the clone() invocation which resolves to PTE on 32-bit and PMD on
   64-bit.

 - Zero the 8-byte case for get_user() on range check failure on 32-bit

   The recend consolidation of the 8-byte get_user() case broke the
   zeroing in the failure case again. Establish it by clearing ECX
   before the range check and not afterwards as that obvioulsy can't be
   reached when the range check fails

* tag 'x86-urgent-2024-08-04' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/uaccess: Zero the 8-byte get_range case on failure on 32-bit
  x86/mm: Fix pti_clone_entry_text() for i386
  x86/mm: Fix pti_clone_pgtable() alignment assumption
  x86/setup: Parse the builtin command line before merging
  x86/CPU/AMD: Add models 0x60-0x6f to the Zen5 range
  x86/sev: Fix __reserved field in sev_config
  x86/aperfmperf: Fix deadlock on cpu_hotplug_lock

10 months agoMerge tag 'timers-urgent-2024-08-04' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sun, 4 Aug 2024 15:50:16 +0000 (08:50 -0700)]
Merge tag 'timers-urgent-2024-08-04' of git://git./linux/kernel/git/tip/tip

Pull timer fixes from Thomas Gleixner:
 "Two fixes for the timer/clocksource code:

   - The recent fix to make the take over of the broadcast timer more
     reliable retrieves a per CPU pointer in preemptible context.

     This went unnoticed in testing as some compilers hoist the access
     into the non-preemotible section where the pointer is actually
     used, but obviously compilers can rightfully invoke it where the
     code put it.

     Move it into the non-preemptible section right to the actual usage
     side to cure it.

   - The clocksource watchdog is supposed to emit a warning when the
     retry count is greater than one and the number of retries reaches
     the limit.

     The condition is backwards and warns always when the count is
     greater than one. Fixup the condition to prevent spamming dmesg"

* tag 'timers-urgent-2024-08-04' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  clocksource: Fix brown-bag boolean thinko in cs_watchdog_read()
  tick/broadcast: Move per CPU pointer access into the atomic section

10 months agoMerge tag 'sched-urgent-2024-08-04' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sun, 4 Aug 2024 15:46:14 +0000 (08:46 -0700)]
Merge tag 'sched-urgent-2024-08-04' of git://git./linux/kernel/git/tip/tip

Pull scheduler fixes from Thomas Gleixner:

 - When stime is larger than rtime due to accounting imprecision, then
   utime = rtime - stime becomes negative. As this is unsigned math, the
   result becomes a huge positive number.

   Cure it by resetting stime to rtime in that case, so utime becomes 0.

 - Restore consistent state when sched_cpu_deactivate() fails.

   When offlining a CPU fails in sched_cpu_deactivate() after the SMT
   present counter has been decremented, then the function aborts but
   fails to increment the SMT present counter and leaves it imbalanced.
   Consecutive operations cause it to underflow. Add the missing fixup
   for the error path.

   For SMT accounting the runqueue needs to marked online again in the
   error exit path to restore consistent state.

* tag 'sched-urgent-2024-08-04' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  sched/core: Fix unbalance set_rq_online/offline() in sched_cpu_deactivate()
  sched/core: Introduce sched_set_rq_on/offline() helper
  sched/smt: Fix unbalance sched_smt_present dec/inc
  sched/smt: Introduce sched_smt_present_inc/dec() helper
  sched/cputime: Fix mul_u64_u64_div_u64() precision for cputime

10 months agoMerge tag 'perf-urgent-2024-08-04' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Sun, 4 Aug 2024 15:42:18 +0000 (08:42 -0700)]
Merge tag 'perf-urgent-2024-08-04' of git://git./linux/kernel/git/tip/tip

Pull x86 perf fixes from Thomas Gleixner:

 - Move the smp_processor_id() invocation back into the non-preemtible
   region, so that the result is valid to use

 - Add the missing package C2 residency counters for Sierra Forest CPUs
   to make the newly added support actually useful

* tag 'perf-urgent-2024-08-04' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  perf/x86: Fix smp_processor_id()-in-preemptible warnings
  perf/x86/intel/cstate: Add pkg C2 residency counter for Sierra Forest

10 months agoMerge tag 'irq-urgent-2024-08-04' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Sun, 4 Aug 2024 15:36:57 +0000 (08:36 -0700)]
Merge tag 'irq-urgent-2024-08-04' of git://git./linux/kernel/git/tip/tip

Pull irq fixes from Thomas Gleixner:
 "A couple of fixes for interrupt chip drivers:

   - Make sure to skip the clear register space in the MBIGEN driver
     when calculating the node register index. Otherwise the clear
     register is clobbered and the wrong node registers are accessed.

   - Fix a signed/unsigned confusion in the loongarch CPU driver which
     converts an error code to a huge "valid" interrupt number.

   - Convert the mesion GPIO interrupt controller lock to a raw spinlock
     so it works on RT.

   - Add a missing static to a internal function in the pic32 EVIC
     driver"

* tag 'irq-urgent-2024-08-04' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  irqchip/mbigen: Fix mbigen node address layout
  irqchip/meson-gpio: Convert meson_gpio_irq_controller::lock to 'raw_spinlock_t'
  irqchip/irq-pic32-evic: Add missing 'static' to internal function
  irqchip/loongarch-cpu: Fix return value of lpic_gsi_to_irq()

10 months agoMerge tag 'locking-urgent-2024-08-04' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sun, 4 Aug 2024 15:32:31 +0000 (08:32 -0700)]
Merge tag 'locking-urgent-2024-08-04' of git://git./linux/kernel/git/tip/tip

Pull locking fixes from Thomas Gleixner:
 "Two fixes for locking and jump labels:

   - Ensure that the atomic_cmpxchg() conditions are correct and
     evaluating to true on any non-zero value except 1. The missing
     check of the return value leads to inconsisted state of the jump
     label counter.

   - Add a missing type conversion in the paravirt spinlock code which
     makes loongson build again"

* tag 'locking-urgent-2024-08-04' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  jump_label: Fix the fix, brown paper bags galore
  locking/pvqspinlock: Correct the type of "old" variable in pv_kick_node()

10 months agoarm: dts: arm: versatile-ab: Fix duplicate clock node name
Rob Herring (Arm) [Tue, 30 Jul 2024 21:00:30 +0000 (15:00 -0600)]
arm: dts: arm: versatile-ab: Fix duplicate clock node name

Commit 04f08ef291d4 ("arm/arm64: dts: arm: Use generic clock and
regulator nodenames") renamed nodes and created 2 "clock-24000000" nodes
(at different paths).

The kernel can't handle these duplicate names even though they are at
different paths.  Fix this by renaming one of the nodes to "clock-pclk".

This name is aligned with other Arm boards (those didn't have a known
frequency to use in the node name).

Fixes: 04f08ef291d4 ("arm/arm64: dts: arm: Use generic clock and regulator nodenames")
Reported-by: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: Rob Herring (Arm) <robh@kernel.org>
Tested-by: Guenter Roeck <linux@roeck-us.net>
Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Tested-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
10 months agoMerge tag '6.11-rc1-smb-client-fixes' of git://git.samba.org/sfrench/cifs-2.6
Linus Torvalds [Sun, 4 Aug 2024 15:18:40 +0000 (08:18 -0700)]
Merge tag '6.11-rc1-smb-client-fixes' of git://git.samba.org/sfrench/cifs-2.6

Pull smb client fixes from Steve French:

 - two reparse point fixes

 - minor cleanup

 - additional trace point (to help debug a recent problem)

* tag '6.11-rc1-smb-client-fixes' of git://git.samba.org/sfrench/cifs-2.6:
  cifs: update internal version number
  smb: client: fix FSCTL_GET_REPARSE_POINT against NetApp
  smb3: add dynamic tracepoints for shutdown ioctl
  cifs: Remove cifs_aio_ctx
  smb: client: handle lack of FSCTL_GET_REPARSE_POINT support

10 months agoMerge tag 'media/v6.11-2' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab...
Linus Torvalds [Sun, 4 Aug 2024 15:12:33 +0000 (08:12 -0700)]
Merge tag 'media/v6.11-2' of git://git./linux/kernel/git/mchehab/linux-media

Pull media fixes from Mauro Carvalho Chehab:

 - two Kconfig fixes

 - one fix for the UVC driver addressing probing time detection of a UVC
   custom controls

 - one fix related to PDF generation

* tag 'media/v6.11-2' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-media:
  media: v4l: Fix missing tabular column hint for Y14P format
  media: intel/ipu6: select AUXILIARY_BUS in Kconfig
  media: ipu-bridge: fix ipu6 Kconfig dependencies
  media: uvcvideo: Fix custom control mapping probing

10 months agoMerge tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi
Linus Torvalds [Sat, 3 Aug 2024 22:12:56 +0000 (15:12 -0700)]
Merge tag 'scsi-fixes' of git://git./linux/kernel/git/jejb/scsi

Pull SCSI fixes from James Bottomley:
 "One core change that reverts the double message print patch in sd.c
  (it was causing regressions on embedded systems).

  The rest are driver fixes in ufs, mpt3sas and mpi3mr"

* tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:
  scsi: ufs: exynos: Don't resume FMP when crypto support is disabled
  scsi: mpt3sas: Avoid IOMMU page faults on REPORT ZONES
  scsi: mpi3mr: Avoid IOMMU page faults on REPORT ZONES
  scsi: ufs: core: Do not set link to OFF state while waking up from hibernation
  scsi: Revert "scsi: sd: Do not repeat the starting disk message"
  scsi: ufs: core: Fix deadlock during RTC update
  scsi: ufs: core: Bypass quick recovery if force reset is needed
  scsi: ufs: core: Check LSDBS cap when !mcq