drm/xe/pxp: Fail the load if PXP fails to initialize
The PXP implementation mimics the i915 approach of allowing the load
to continue even if PXP init has failed. On Xe however we're taking an
harder stance on boot error and only allowing the load to complete if
everything is working, so update the code to fail if anything goes wrong
during PXP init.
While at it, update the return code in case of PXP not supported to be 0
instead of EOPNOTSUPP, to follow the standard of functions called by
xe_device_probe where every non-zero value means failure.
Since we have a preallocated pool of relay transactions, which
should cover all our normal relay use cases, we may use the
GFP_NOWAIT flag when allocating new outgoing transactions.
max_remote_tiles is more related to the platform than the GT IP. Thus
move it to platform descriptor from graphics descriptor. Note that the
FIXME is no more required, thus it can be dropped.
v2: Rebase
v3: Change the position of comment (MattR)
The HW suspend flow kills all PXP HWDRM sessions, so we need to mark all
the queues and BOs as invalid and do a full termination when PXP is next
used.
v2: rebase
v3: rebase on new status flow, defer termination to next PXP use as it
makes things much easier and allows us to use the same function for all
types of suspend.
v4: fix the documentation of the suspend function (John)
drm/xe/pxp/uapi: Add API to mark a BO as using PXP
The driver needs to know if a BO is encrypted with PXP to enable the
display decryption at flip time.
Furthermore, we want to keep track of the status of the encryption and
reject any operation that involves a BO that is encrypted using an old
key. There are two points in time where such checks can kick in:
1 - at VM bind time, all operations except for unmapping will be
rejected if the key used to encrypt the BO is no longer valid. This
check is opt-in via a new VM_BIND flag, to avoid a scenario where a
malicious app purposely shares an invalid BO with a non-PXP aware
app (such as a compositor). If the VM_BIND was failed, the
compositor would be unable to display anything at all. Allowing the
bind to go through means that output still works, it just displays
garbage data within the bounds of the illegal BO.
2 - at job submission time, if the queue is marked as using PXP, all
objects bound to the VM will be checked and the submission will be
rejected if any of them was encrypted with a key that is no longer
valid.
Note that there is no risk of leaking the encrypted data if a user does
not opt-in to those checks; the only consequence is that the user will
not realize that the encryption key is changed and that the data is no
longer valid.
v2: Better commnnts and descriptions (John), rebase
v3: Properly return the result of key_assign up the stack, do not use
xe_bo in display headers (Jani)
Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com> Cc: Matthew Brost <matthew.brost@intel.com> Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com> Cc: John Harrison <John.C.Harrison@Intel.com> Cc: Jani Nikula <jani.nikula@intel.com> Reviewed-by: John Harrison <John.C.Harrison@Intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20250129174140.948829-11-daniele.ceraolospurio@intel.com
PXP prerequisites (SW proxy and HuC auth via GSC) are completed
asynchronously from driver load, which means that userspace can start
submitting before we're ready to start a PXP session. Therefore, we need
a query that userspace can use to check not only if PXP is supported but
also to wait until the prerequisites are done.
v2: Improve doc, do not report TYPE_NONE as supported (José)
v3: Better comments, remove unneeded copy_from_user (John)
drm/xe/pxp/uapi: Add userspace and LRC support for PXP-using queues
Userspace is required to mark a queue as using PXP to guarantee that the
PXP instructions will work. In addition to managing the PXP sessions,
when a PXP queue is created the driver will set the relevant bits in
its context control register.
On submission of a valid PXP queue, the driver will validate all
encrypted objects mapped to the VM to ensured they were encrypted with
the current key.
v2: Remove pxp_types include outside of PXP code (Jani), better comments
and code cleanup (John)
v3: split the internal PXP management to a separate patch for ease of
review. re-order ioctl checks to always return -EINVAL if parameters are
invalid, rebase on msix changes.
drm/xe/pxp: Add PXP queue tracking and session start
We expect every queue that uses PXP to be marked as doing so, to allow
the driver to correctly manage the encryption status. The API for doing
this from userspace is coming in the next patch, while this patch
implement the management side of things. When a PXP queue is created,
the driver will do the following:
- Start the default PXP session if it is not already running;
- assign an rpm ref to the queue to keep for its lifetime (this is
required because PXP HWDRM sessions are killed by the HW suspend flow).
Since PXP start and termination can race each other, this patch also
introduces locking and a state machine to keep track of the pending
operations. Note that since we'll need to take the lock from the
suspend/resume paths as well, we can't do submissions while holding it,
which means we need a slightly more complicated state machine to keep
track of intermediate steps.
v4: new patch in the series, split from the following interface patch to
keep review manageable. Lock and status rework to not do submissions
under lock.
drm/xe/pxp: Add GSC session initialization support
A session is initialized (i.e. started) by sending a message to the GSC.
The initialization will be triggered when a user opts-in to using PXP;
the interface for that is coming in a follow-up patch in the series.
v2: clean up error messages, use new ARB define (John)
When something happen to the session, the HW generates a termination
interrupt. In reply to this, the driver is required to submit an inline
session termination via the VCS, trigger the global termination and
notify the GSC FW that the session is now invalid.
v2: rename ARB define to make it cleaner to move it to uapi (John)
v3: fix parameter name in documentation
After a session is terminated, we need to inform the GSC so that it can
clean up its side of the allocation. This is done by sending an
invalidation command with the session ID.
The invalidation will be triggered in response to a termination,
interrupt, whose handling is coming in the next patch in the series.
The key termination is done with a specific submission to the VCS
engine. This flow will be triggered in response to a termination
interrupt, whose handling is coming in a follow-up patch in the series.
v2: clean up defines and command emission code. (John)
PXP requires submissions to the HW for the following operations
1) Key invalidation, done via the VCS engine
2) Communication with the GSC FW for session management, done via the
GSCCS.
Key invalidation submissions are serialized (only 1 termination can be
serviced at a given time) and done via GGTT, so we can allocate a simple
BO and a kernel queue for it.
Submissions for session management are tied to a PXP client (identified
by a unique host_session_id); from the GSC POV this is a user-accessible
construct, so all related submission must be done via PPGTT. The driver
does not currently support PPGTT submission from within the kernel, so
to add this support, the following changes have been included:
- a new type of kernel-owned VM (marked as GSC), required to ensure we
don't use fault mode on the engine and to mark the different lock
usage with lockdep.
- a new function to map a BO into a VM from within the kernel.
v2: improve comments and function name, remove unneeded include (John)
v3: fix variable/function names in documentation
As the first step towards adding PXP support, hook in the PXP init
function, allocate the PXP structure and initialize the KCR register to
allow PXP HWDRM sessions.
v2: remove unneeded includes, free PXP memory on error (John)
Lucas De Marchi [Fri, 31 Jan 2025 22:39:08 +0000 (14:39 -0800)]
drm/xe: Remove xe_dummy_exit()
Since commit 014125c64d09 ("drm/xe: Support 'nomodeset' kernel
command-line option") the dummy exit is not needed anymore since the
caller check for a NULL pointer. Drop it.
Riana Tauro [Fri, 31 Jan 2025 08:05:27 +0000 (13:35 +0530)]
drm/xe: Skip survivability mode for VF
Follow the probe flow in case of VF and do not enter survivability mode
in case of pcode init failure.
Fixes: 5e940312a2ac ("drm/xe: Add functions and sysfs for boot survivability") Suggested-by: Satyanarayana K V P <satyanarayana.k.v.p@intel.com> Signed-off-by: Riana Tauro <riana.tauro@intel.com> Reviewed-by: Satyanarayana K V P <satyanarayana.k.v.p@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20250131080527.2256475-1-riana.tauro@intel.com Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
drm/xe/display: Use a single early init call for display
Now that interrupts are disabled for xe_display_init_noaccel,
both xe_display_init_noirq and xe_display_init_noaccel run in the same
context.
This means that we can get rid of the 3 different init calls. Without
interrupts, nothing is touching display up to this point.
Unify those 3 early display calls into a single xe_display_init_early(),
this makes the init sequence cleaner, and display less tangled during
init.
We're changing the driver to have no interrupts during early init for
Xe, so we poll the PIPE_FRMSTMSMP counter instead.
Interrupts cannot be enabled during FB readout because memirq's requires
an allocation. This would overwrite the FB we want to read out.
While it might be possible to also run do the same in i915 and run
it without interrupts, the platforms i915 supports had a less clear
distinction between display and graphics. For this reason I choose
only to touch Xe for now.
Jakub Kolakowski [Tue, 28 Jan 2025 11:03:00 +0000 (11:03 +0000)]
drm/xe/pf: Add runtime registers for graphics gen >= 30
Add missing runtime registers for graphics versions of 3000 or higher.
This is required for Xe3 where additionally we have
MIRROR_L3BANK_ENABLE register.
Signed-off-by: Jakub Kolakowski <jakub1.kolakowski@intel.com> Suggested-by: Piotr Piórkowski <piotr.piorkowski@intel.com> Cc: Adam Miszczak <adam.miszczak@linux.intel.com> Cc: Jakub Kolakowski <jakub1.kolakowski@intel.com> Cc: Lukasz Laguna <lukasz.laguna@intel.com> Cc: Marcin Bernatowicz <marcin.bernatowicz@linux.intel.com> Cc: Michal Wajdeczko <michal.wajdeczko@intel.com> Cc: Piotr Piorkowski <piotr.piorkowski@intel.com> Cc: Satyanarayana K V P <satyanarayana.k.v.p@intel.com> Reviewed-by: Satyanarayana K V P <satyanarayana.k.v.p@intel.com> Tested-by: Marcin Bernatowicz <marcin.bernatowicz@linux.intel.com> Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20250128110300.2840596-2-jakub1.kolakowski@intel.com
Michal Wajdeczko [Wed, 29 Jan 2025 19:59:47 +0000 (20:59 +0100)]
drm/xe/pf: Reset GuC VF config when unprovisioning critical resource
GuC firmware counts received VF configuration KLVs and may start
validation of the complete VF config even if some resources where
unprovisioned in the meantime, leading to unexpected errors like:
$ echo 1 | sudo tee /sys/kernel/debug/dri/0000:00:02.0/gt0/vf1/contexts_quota
$ echo 0 | sudo tee /sys/kernel/debug/dri/0000:00:02.0/gt0/vf1/contexts_quota
$ echo 1 | sudo tee /sys/kernel/debug/dri/0000:00:02.0/gt0/vf1/doorbells_quota
$ echo 0 | sudo tee /sys/kernel/debug/dri/0000:00:02.0/gt0/vf1/doorbells_quota
$ echo 1 | sudo tee /sys/kernel/debug/dri/0000:00:02.0/gt0/vf1/ggtt_quota
tee: '/sys/kernel/debug/dri/0000:00:02.0/gt0/vf1/ggtt_quota': Input/output error
To mitigate this problem trigger explicit VF config reset after
unprovisioning any of the critical resources (GGTT, context or
doorbell IDs) that GuC is monitoring.
Michal Wajdeczko [Wed, 29 Jan 2025 19:59:46 +0000 (20:59 +0100)]
drm/xe/pf: Don't send BEGIN_ID if VF has no context/doorbells
It turned out that GuC validates VF configuration immediately
after receiving "some" set of configuration KLVs and complains
if one of the critical, from GuC understanding, resource is left
unprovisioned, even if PF should be still allowed to make late VF
config adjustments, since VF was not yet started.
This issue was discovered after we decided to asynchronously
re-send configuration KLVs after GT reset/resume, as then fair
VF auto-provisioning could already allocate some of the resources,
which was a prerequiste for sending those config KLVs:
Francois Dugast [Wed, 29 Jan 2025 17:52:41 +0000 (18:52 +0100)]
drm/xe/gt_pagefault: Print engine class string
The engine class index which is printed here is an internal representation
for debugging. It is _not_ an index based on DRM_XE_ENGINE_CLASS_* values
provided in the uAPI. Add the string representation of the engine class to
the output in order to limit possible confusion by users when analyzing the
logs.
Lucas De Marchi [Tue, 28 Jan 2025 15:42:42 +0000 (07:42 -0800)]
drm/xe/guc: Fix size_t print format
Use %zx format to print size_t to remove the following warning when
building for i386:
>> drivers/gpu/drm/xe/xe_guc_ct.c:1727:43: warning: format specifies type 'unsigned long' but the argument has type 'size_t' (aka 'unsigned int') [-Wformat]
1727 | drm_printf(p, "[CTB].length: 0x%lx\n", snapshot->ctb_size);
| ~~~ ^~~~~~~~~~~~~~~~~~
| %zx
Cc: José Roberto de Souza <jose.souza@intel.com> Reported-by: kernel test robot <lkp@intel.com> Closes: https://lore.kernel.org/oe-kbuild-all/202501281627.H6nj184e-lkp@intel.com/ Fixes: cb1f868ca137 ("drm/xe: Make GUC binaries dump consistent with other binaries in devcoredump") Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20250128154242.3371687-1-lucas.demarchi@intel.com Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
DCC in LNL should be disabled. It was a mistake to decide
to go against GuC platform defaults in this case and this
could lead to regressions in some TDP limited scenarios
instead of helping.
Melissa Wen [Tue, 28 Jan 2025 00:41:10 +0000 (21:41 -0300)]
drm/amd/display: restore invalid MSA timing check for freesync
This restores the original behavior that gets min/max freq from EDID and
only set DP/eDP connector as freesync capable if "sink device is capable
of rendering incoming video stream without MSA timing parameters", i.e.,
`allow_invalid_MSA_timing_params` is true. The condition was mistakenly
removed by 0159f88a99c9 ("drm/amd/display: remove redundant freesync
parser for DP").
CC: Mario Limonciello <mario.limonciello@amd.com> CC: Alex Hung <alex.hung@amd.com> Link: https://gitlab.freedesktop.org/drm/amd/-/issues/3915 Fixes: 0159f88a99c9 ("drm/amd/display: remove redundant freesync parser for DP") Reviewed-by: Harry Wentland <harry.wentland@amd.com> Signed-off-by: Melissa Wen <mwen@igalia.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Cc: stable@vger.kernel.org
Prike Liang [Tue, 14 Jan 2025 03:20:17 +0000 (11:20 +0800)]
drm/amdkfd: only flush the validate MES contex
The following page fault was observed duringthe KFD process release.
In this particular error case, the HIP test (./MemcpyPerformance -h)
does not require the queue. As a result, the process_context_addr was
not assigned when the KFD process was released, ultimately leading to
this page fault during the execution of the function
kfd_process_dequeue_from_all_devices().
Signed-off-by: Prike Liang <Prike.Liang@amd.com> Reviewed-by: Jonathan Kim <jonathan.kim@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Cc: stable@vger.kernel.org
Jay Cornwall [Thu, 16 Jan 2025 20:36:39 +0000 (14:36 -0600)]
drm/amdkfd: Block per-queue reset when halt_if_hws_hang=1
The purpose of halt_if_hws_hang is to preserve GPU state for driver
debugging when queue preemption fails. Issuing per-queue reset may
kill wavefronts which caused the preemption failure.
Signed-off-by: Jay Cornwall <jay.cornwall@amd.com> Reviewed-by: Jonathan Kim <Jonathan.Kim@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Cc: stable@vger.kernel.org # 6.12.x
Riana Tauro [Tue, 28 Jan 2025 09:56:31 +0000 (15:26 +0530)]
drm/xe: Enable Boot Survivability mode
Enable boot survivability mode if pcode initialization fails and
if boot status indicates a failure. In this mode, drm card is not
exposed and driver probe returns success after loading the bare minimum
to allow firmware to be flashed via mei.
Riana Tauro [Tue, 28 Jan 2025 09:56:30 +0000 (15:26 +0530)]
drm/xe: Add functions and sysfs for boot survivability
Boot Survivability is a software based workflow for recovering a system
in a failed boot state. Here system recoverability is concerned with
recovering the firmware responsible for boot.
This is implemented by loading the driver with bare minimum (no drm card)
to allow the firmware to be flashed through mei-gsc and collect telemetry.
The driver's probe flow is modified such that it enters survivability mode
when pcode initialization is incomplete and boot status denotes a failure.
In this mode, drm card is not exposed and presence of survivability_mode
entry in PCI sysfs is used to indicate survivability mode and
provide additional information required for debug
This patch adds initialization functions and exposes admin
readable sysfs entries
The new sysfs will have the below layout
/sys/bus/.../bdf
├── survivability_mode
v2: reorder headers
fix doc
remove survivability info and use mode to display information
use separate function for logging survivability information
for critical error (Rodrigo)
v3: use for loop
use dev logs instead of drm
use helper function for aux history(Rodrigo)
remove unnecessary error check of greater than max_scratch
as we are reading only 3 bit
v4: fix checkpatch warnings
fix space (Rodrigo)
rename register
drm/xe: Make GUC binaries dump consistent with other binaries in devcoredump
All other(hwsp, hwctx and vmas) binaries follow this format:
[name].length: 0x1000
[name].data: xxxxxxx
[name].error: errno
The error one is just in case by some reason it was not able to
capture the binary.
So this GuC binaries should follow the same patern.
v2:
- renamed GUC binary to LOG
Cc: John Harrison <John.C.Harrison@Intel.com> Cc: Lucas De Marchi <lucas.demarchi@intel.com> Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com> Signed-off-by: José Roberto de Souza <jose.souza@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20250123202307.95103-3-jose.souza@intel.com Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
Lucas De Marchi [Thu, 23 Jan 2025 20:22:03 +0000 (12:22 -0800)]
drm/xe: Fix and re-enable xe_print_blob_ascii85()
Commit 70fb86a85dc9 ("drm/xe: Revert some changes that break a mesa
debug tool") partially reverted some changes to workaround breakage
caused to mesa tools. However, in doing so it also broke fetching the
GuC log via debugfs since xe_print_blob_ascii85() simply bails out.
The fix is to avoid the extra newlines: the devcoredump interface is
line-oriented and adding random newlines in the middle breaks it. If a
tool is able to parse it by looking at the data and checking for chars
that are out of the ascii85 space, it can still do so. A format change
that breaks the line-oriented output on devcoredump however needs better
coordination with existing tools.
v2: Add suffix description comment
v3: Reword explanation of xe_print_blob_ascii85() calling drm_puts()
in a loop
Reviewed-by: José Roberto de Souza <jose.souza@intel.com> Cc: John Harrison <John.C.Harrison@Intel.com> Cc: Julia Filipchuk <julia.filipchuk@intel.com> Cc: José Roberto de Souza <jose.souza@intel.com> Cc: stable@vger.kernel.org Fixes: 70fb86a85dc9 ("drm/xe: Revert some changes that break a mesa debug tool") Fixes: ec1455ce7e35 ("drm/xe/devcoredump: Add ASCII85 dump helper function") Link: https://patchwork.freedesktop.org/patch/msgid/20250123202307.95103-2-jose.souza@intel.com Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
Lucas De Marchi [Thu, 23 Jan 2025 05:11:11 +0000 (21:11 -0800)]
drm/xe/devcoredump: Move exec queue snapshot to Contexts section
Having the exec queue snapshot inside a "GuC CT" section was always
wrong. Commit c28fd6c358db ("drm/xe/devcoredump: Improve section
headings and add tile info") tried to fix that bug, but with that also
broke the mesa tool that parses the devcoredump, hence it was reverted
in commit 70fb86a85dc9 ("drm/xe: Revert some changes that break a mesa
debug tool").
With the mesa tool also fixed, this can propagate as a fix on both
kernel and userspace side to avoid unnecessary headache for a debug
feature.
Cc: John Harrison <John.C.Harrison@Intel.com> Cc: Julia Filipchuk <julia.filipchuk@intel.com> Cc: José Roberto de Souza <jose.souza@intel.com> Cc: stable@vger.kernel.org Fixes: 70fb86a85dc9 ("drm/xe: Revert some changes that break a mesa debug tool") Reviewed-by: José Roberto de Souza <jose.souza@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20250123051112.1938193-2-lucas.demarchi@intel.com Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
John Harrison [Sat, 18 Jan 2025 00:54:03 +0000 (16:54 -0800)]
drm/xe: Upgrade complaint about missing slice info
The steering code needs to know slice/subslice counts and this
information should be retrieved from the hwconfig table. However,
earlier platforms don't have it, hence the KMD has a fallback path.
Newer platforms really should have the entries and if they are missing
that is a bug that needs to be fixed in the table.
So update the complaint to be an error on newer platforms and remove
it completely for older ones that we know are bad (but are not POR for
the Xe driver anyway). Also, re-word the message a little to make it
clearer what the issue is.
Michal Wajdeczko [Sat, 25 Jan 2025 21:55:05 +0000 (22:55 +0100)]
drm/xe/pf: Move VFs reprovisioning to worker
Since the GuC is reset during GT reset, we need to re-send the
entire SR-IOV provisioning configuration to the GuC. But since
this whole configuration is protected by the PF master mutex and
we can't avoid making allocations under this mutex (like during
LMEM provisioning), we can't do this reprovisioning from gt-reset
path if we want to be reclaim-safe. Move VFs reprovisioning to a
async worker that we will start from the gt-reset path.
Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com> Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com> Cc: Matthew Brost <matthew.brost@intel.com> Reviewed-by: Michał Winiarski <michal.winiarski@intel.com> Reviewed-by: Stuart Summers <stuart.summers@intel.com> Reviewed-by: Matthew Brost <matthew.brost@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20250125215505.720-1-michal.wajdeczko@intel.com
Michal Wajdeczko [Fri, 24 Jan 2025 18:52:47 +0000 (19:52 +0100)]
drm/xe/pf: Use GuC Buffer Cache during policy provisioning
Start using GuC buffer cache for the SRIOV policy configuration
actions. This is a required step before we could declare SRIOV
PF as being a reclaim safe.
Vinay Belgaumkar [Fri, 24 Jan 2025 05:04:11 +0000 (21:04 -0800)]
drm/xe/pmu: Add GT C6 events
Provide a PMU interface for GT C6 residency counters. The interface is
similar to the one available for i915, but gt is passed in the config
when creating the event.
Sample usage and output:
$ perf list | grep gt-c6
xe_0000_00_02.0/gt-c6-residency/ [Kernel PMU event]
==> /sys/bus/event_source/devices/xe_0000_00_02.0/events/gt-c6-residency.unit <==
ms
$ perf stat -e xe_0000_00_02.0/gt-c6-residency,gt=0/ -I1000
# time counts unit events
1.001196056 1,001 ms xe_0000_00_02.0/gt-c6-residency,gt=0/
2.005216219 1,003 ms xe_0000_00_02.0/gt-c6-residency,gt=0/
Lucas De Marchi [Fri, 24 Jan 2025 05:04:10 +0000 (21:04 -0800)]
drm/xe/pmu: Add attribute skeleton
Add the generic support for defining new attributes. This only adds
the macros and common infra for the event counters, but no counters
yet. This is going to be added as follow up changes.
Lucas De Marchi [Fri, 24 Jan 2025 05:04:09 +0000 (21:04 -0800)]
drm/xe/pmu: Get/put runtime pm on event init
When the event is created, make sure runtime pm is taken and later put:
in order to read an event counter the GPU needs to remain accessible and
doing a get/put during perf's read is not possible it's holding a
raw_spinlock.
Lucas De Marchi [Fri, 24 Jan 2025 05:04:07 +0000 (21:04 -0800)]
drm/xe/pmu: Assert max gt
XE_PMU_MAX_GT needs to be used due to a circular dependency, but we
should make sure it doesn't go out of sync with XE_PMU_MAX_GT. Add a
compile check for that.
Simona Vetter [Fri, 24 Jan 2025 16:06:06 +0000 (17:06 +0100)]
Merge tag 'drm-misc-next-fixes-2025-01-24' of https://gitlab.freedesktop.org/drm/misc/kernel into drm-next
drm-misc-next-fixes for v6.14-rc1:
- Fix a serious regression from commit e4b5ccd392b9 ("drm/v3d: Ensure
job pointer is set to NULL after job completion")
- dmem cgroup Kconfig fix (acked by Tejun)
- virtio: uaf in dma_buf free path
- xlnx: kerneldoc
Aric Cyr [Tue, 10 Dec 2024 23:38:15 +0000 (18:38 -0500)]
drm/amd/display: Optimize cursor position updates
[why]
Updating the cursor enablement register can be a slow operation and accumulates
when high polling rate cursors cause frequent updates asynchronously to the
cursor position.
[how]
Since the cursor enable bit is cached there is no need to update the
enablement register if there is no change to it. This removes the
read-modify-write from the cursor position programming path in HUBP and
DPP, leaving only the register writes.
Cc: Mario Limonciello <mario.limonciello@amd.com> Cc: Alex Deucher <alexander.deucher@amd.com> Cc: stable@vger.kernel.org Reviewed-by: Sung Lee <sung.lee@amd.com> Signed-off-by: Aric Cyr <Aric.Cyr@amd.com> Signed-off-by: Wayne Lin <wayne.lin@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Aric Cyr [Thu, 9 Jan 2025 20:03:48 +0000 (15:03 -0500)]
drm/amd/display: Add hubp cache reset when powergating
[Why]
When HUBP is power gated, the SW state can get out of sync with the
hardware state causing cursor to not be programmed correctly.
[How]
Similar to DPP, add a HUBP reset function which is called wherever
HUBP is initialized or powergated. This function will clear the cursor
position and attribute cache allowing for proper programming when the
HUBP is brought back up.
Cc: Mario Limonciello <mario.limonciello@amd.com> Cc: Alex Deucher <alexander.deucher@amd.com> Cc: stable@vger.kernel.org Reviewed-by: Sung Lee <sung.lee@amd.com> Signed-off-by: Aric Cyr <Aric.Cyr@amd.com> Signed-off-by: Wayne Lin <wayne.lin@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Shaoyun Liu [Tue, 14 Jan 2025 16:57:41 +0000 (11:57 -0500)]
drm/amd/amdgpu: Enable scratch data dump for mes 12
MES internal will check CP_MES_MSCRATCH_LO/HI register to set scratch
data location during ucode start, driver side need to start the MES
one by one with different setting for each pipe
Signed-off-by: Shaoyun Liu <shaoyun.liu@amd.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Effectively amdgpu.gttsize gets set to ~1/2 of RAM, but that's controlled
by what the TTM page limit is set to. Clarify the kdoc.
Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Mario Limonciello <mario.limonciello@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amd/amdgpu: Prevent null pointer dereference in GPU bandwidth calculation
If the parent is NULL, adev->pdev is used to retrieve the PCIe speed and
width, ensuring that the function can still determine these
capabilities from the device itself.
Fixes the below:
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c:6193 amdgpu_device_gpu_bandwidth()
error: we previously assumed 'parent' could be null (see line 6180)
Fixes: 757e8b951ce2 ("drm/amdgpu: cache gpu pcie link width") Cc: Christian König <christian.koenig@amd.com> Cc: Alex Deucher <alexander.deucher@amd.com> Reported-by: Dan Carpenter <dan.carpenter@linaro.org> Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com> Suggested-by: Lijo Lazar <lijo.lazar@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amd/display: Fix error pointers in amdgpu_dm_crtc_mem_type_changed
The function amdgpu_dm_crtc_mem_type_changed was dereferencing pointers
returned by drm_atomic_get_plane_state without checking for errors. This
could lead to undefined behavior if the function returns an error pointer.
This commit adds checks using IS_ERR to ensure that new_plane_state and
old_plane_state are valid before dereferencing them.
Fixes the below:
drivers/gpu/drm/amd/amdgpu/../display/amdgpu_dm/amdgpu_dm.c:11486 amdgpu_dm_crtc_mem_type_changed()
error: 'new_plane_state' dereferencing possible ERR_PTR()
Fixes: 4caacd1671b7 ("drm/amd/display: Do not elevate mem_type change to full update") Cc: Leo Li <sunpeng.li@amd.com> Cc: Tom Chung <chiahsuan.chung@amd.com> Cc: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com> Cc: Roman Li <roman.li@amd.com> Cc: Alex Hung <alex.hung@amd.com> Cc: Aurabindo Pillai <aurabindo.pillai@amd.com> Cc: Harry Wentland <harry.wentland@amd.com> Cc: Hamza Mahfooz <hamza.mahfooz@amd.com> Reported-by: Dan Carpenter <dan.carpenter@linaro.org> Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com> Reviewed-by: Roman Li <roman.li@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Lin.Cao [Tue, 14 Jan 2025 09:42:01 +0000 (17:42 +0800)]
drm/amdgpu: fix ring timeout issue in gfx10 sr-iov environment
commit 26c95e838e63 ("drm/amdgpu: set the VM pointer to NULL in
amdgpu_job_prepare") set job->vm as NULL if there is no fence. It will
cause emit switch buffer be skippen if job->vm set as NULL.
Check job rather than vm could solve this problem.
Fixes: 26c95e838e63 ("drm/amdgpu: set the VM pointer to NULL in amdgpu_job_prepare") Signed-off-by: Lin.Cao <lincao12@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Lijo Lazar [Fri, 17 Jan 2025 14:42:42 +0000 (20:12 +0530)]
drm/amd/pm: Fix smu v13.0.6 caps initialization
Fix the initialization and usage of SMU v13.0.6 capability values. Use
caps_set/clear functions to set/clear capability.
Also, fix SET_UCLK_MAX capability on APUs, it is supported on APUs.
Fixes: e9b86b841baf ("drm/amd/pm: Add capability flags for SMU v13.0.6") Signed-off-by: Lijo Lazar <lijo.lazar@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Yang Wang <kevinyang.wang@amd.com> Reviewed-by: Asad Kamal <asad.kamal@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amd/pm: Refactor SMU 13.0.6 SDMA reset firmware version checks
This patch refactors the firmware version checks in `smu_v13_0_6_reset_sdma`
to support multiple SMU programs with different firmware version thresholds.
V2: return -EOPNOTSUPP for unspported pmfw
Suggested-by: Lazar Lijo <Lijo.Lazar@amd.com> Signed-off-by: Jesse Zhang <jesse.zhang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Lijo Lazar [Thu, 16 Jan 2025 12:26:12 +0000 (17:56 +0530)]
drm/amd/pm: Add capability flags for SMU v13.0.6
Add capability flags for SMU v13.0.6 variants. Initialize the flags
based on firmware support. As there are multiple IP versions maintained,
it is more manageable with one time initialization of caps flags based
on IP version and firmware feature support.
Signed-off-by: Lijo Lazar <lijo.lazar@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Asad Kamal <asad.kamal@amd.com> Reviewed-by: Yang Wang <kevinyang.wang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Fixes: 7594874227e1 ("drm/amd/display: add CEC notifier to amdgpu driver") Reported-by: Stephen Rothwell <sfr@canb.auug.org.au> Reviewed-by: Harry Wentland <harry.wentland@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Cc: Kun Liu <Kun.Liu2@amd.com>
Mark static functions called by dml_prefetch_check() and
dml_core_mode_support() noinline_for_stack to avoid them become huge
functions and thus exceed the frame size limit.
A way to reproduce:
$ git checkout next-20250107
$ mkdir build_dir
$ export PATH=/tmp/llvm-19.1.6-x86_64/bin:$PATH
$ make LLVM=1 O=build_dir allmodconfig
$ make LLVM=1 O=build_dir drivers/gpu/drm/ -j
The way how it chose static functions to mark:
[0] Unset CONFIG_WERROR in build_dir/.config.
To get display_mode_core.o without errors.
[1] Get a function list called by dml_prefetch_check().
$ sed -n '6268,6711p' drivers/gpu/drm/amd/display/dc/dml2/display_mode_core.c \
| sed -n -r 's/.*\W(\w+)\(.*/\1/p' | sort -u >/tmp/syms
[2] Get the non-inline function list.
Objdump won't show the symbols if they are inline functions.
Jay Cornwall [Wed, 8 Jan 2025 03:59:06 +0000 (21:59 -0600)]
drm/amdkfd: Clear MODE.VSKIP in gfx9 trap handler
If user shader issues S_SETVSKIP then this state will persist when
executing the trap handler, causing vector instructions to be
skipped.
VSKIP state is already saved/restored through the MODE register.
Signed-off-by: Jay Cornwall <jay.cornwall@amd.com> Reviewed-by: Lancelot Six <lancelot.six@amd.com> Suggested-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Lijo Lazar [Wed, 1 Jan 2025 04:31:50 +0000 (10:01 +0530)]
drm/amdgpu: Add handler for SDMA context empty
Context empty interrupt is enabled for SDMA 4.4.2. Add a handler for
context empty interrupt so that it is disposed of fast, and not
propagated to KFD layer.
Signed-off-by: Lijo Lazar <lijo.lazar@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Suggested-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Aurabindo Pillai [Mon, 13 Jan 2025 22:05:16 +0000 (17:05 -0500)]
drm/amd: Add debug option to disable subvp
Some monitors flicker when subvp is enabled which maybe related to
an uncommon timing they use. To isolate such issues, add a debug
option to help isolate this the issue for debugging.
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Suggested-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Jay Cornwall [Wed, 8 Jan 2025 03:26:32 +0000 (21:26 -0600)]
drm/amdkfd: Sync trap handler binary with source
Source and binary have become mismatched during branch activity.
Signed-off-by: Jay Cornwall <jay.cornwall@amd.com> Reviewed-by: Lancelot Six <lancelot.six@amd.com> Suggested-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Emily Deng [Tue, 31 Dec 2024 08:13:14 +0000 (16:13 +0800)]
drm/amdkfd: Fix partial migrate issue
For partial migrate from ram to vram, the migrate->cpages is not
equal to migrate->npages, should use migrate->npages to check all needed
migrate pages which could be copied or not.
And only need to set those pages could be migrated to migrate->dst[i], or
the migrate_vma_pages will migrate the wrong pages based on the migrate->dst[i].
v2:
Add mpages to break the loop earlier.
v3:
Uses MIGRATE_PFN_MIGRATE to identify whether page could be migrated.
v4:
Correct the error part.
Signed-off-by: Emily Deng <Emily.Deng@amd.com> Reviewed-by: Philip Yang<Philip.Yang@amd.com> Suggested-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Ashutosh Dixit [Wed, 15 Jan 2025 22:20:29 +0000 (14:20 -0800)]
drm/xe/oa: Set stream->pollin in xe_oa_buffer_check_unlocked
We rely on stream->pollin to decide whether or not to block during
poll/read calls. However, currently there are blocking read code paths
which don't even set stream->pollin. The best place to consistently set
stream->pollin for all code paths is therefore to set it in
xe_oa_buffer_check_unlocked.
Fixes: e936f885f1e9 ("drm/xe/oa/uapi: Expose OA stream fd") Signed-off-by: Ashutosh Dixit <ashutosh.dixit@intel.com> Acked-by: Rodrigo Vivi <rodrigo.vivi@intel.com> Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com> Reviewed-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20250115222029.3002103-1-ashutosh.dixit@intel.com
Maíra Canal [Thu, 23 Jan 2025 01:24:03 +0000 (22:24 -0300)]
drm/v3d: Assign job pointer to NULL before signaling the fence
In commit e4b5ccd392b9 ("drm/v3d: Ensure job pointer is set to NULL
after job completion"), we introduced a change to assign the job pointer
to NULL after completing a job, indicating job completion.
However, this approach created a race condition between the DRM
scheduler workqueue and the IRQ execution thread. As soon as the fence is
signaled in the IRQ execution thread, a new job starts to be executed.
This results in a race condition where the IRQ execution thread sets the
job pointer to NULL simultaneously as the `run_job()` function assigns
a new job to the pointer.
This race condition can lead to a NULL pointer dereference if the IRQ
execution thread sets the job pointer to NULL after `run_job()` assigns
it to the new job. When the new job completes and the GPU emits an
interrupt, `v3d_irq()` is triggered, potentially causing a crash.
Fix the crash by assigning the job pointer to NULL before signaling the
fence. This ensures that the job pointer is cleared before any new job
starts execution, preventing the race condition and the NULL pointer
dereference crash.
Cc: stable@vger.kernel.org Fixes: e4b5ccd392b9 ("drm/v3d: Ensure job pointer is set to NULL after job completion") Signed-off-by: Maíra Canal <mcanal@igalia.com> Reviewed-by: Jose Maria Casanova Crespo <jmcasanova@igalia.com> Reviewed-by: Iago Toral Quiroga <itoral@igalia.com> Tested-by: Phil Elwell <phil@raspberrypi.com> Link: https://patchwork.freedesktop.org/patch/msgid/20250123012403.20447-1-mcanal@igalia.com
Merge remote-tracking branch 'drm/drm-next' into drm-misc-next-fixes
A regression was caused by commit e4b5ccd392b9 ("drm/v3d: Ensure job pointer is set to NULL
after job completion"), but this commit is not yet in next-fixes,
fast-forward it.
Simona Vetter [Thu, 23 Jan 2025 13:39:49 +0000 (14:39 +0100)]
Merge v6.13 into drm-next
A regression was caused by commit e4b5ccd392b9 ("drm/v3d: Ensure job
pointer is set to NULL after job completion"), but this commit is not
yet in next-fixes, fast-forward it.
Note that this recreates Linus merge in 96c84703f1cf ("Merge tag
'drm-next-2025-01-17' of https://gitlab.freedesktop.org/drm/kernel")
because I didn't want to backmerge a random point in the merge window.
drm/bochs: Do not put DRM device in PCI remove callback
Removing the bochs PCI device should mark the DRM device as unplugged
without removing it. Hence clear the respective call to drm_dev_put()
from bochs_pci_remove().
Fixes a double unref in devm_drm_dev_init_release(). An example error
message is shown below:
Arnd Bergmann [Wed, 22 Jan 2025 09:02:03 +0000 (10:02 +0100)]
drm: select DRM_KMS_HELPER from DRM_GEM_SHMEM_HELPER
In the combination of DRM_KMS_HELPER=m, DRM_GEM_SHMEM_HELPER=y, DRM_FBDEV_EMULATION=y,
The shmem code fails to link against the KMS helpers:
x86_64-linux-ld: vmlinux.o: in function `drm_fbdev_shmem_driver_fbdev_probe':
(.text+0xeec601): undefined reference to `drm_fb_helper_alloc_info'
x86_64-linux-ld: (.text+0xeec633): undefined reference to `drm_fb_helper_fill_info'
x86_64-linux-ld: vmlinux.o: in function `drm_fbdev_shmem_get_page':
drm_fbdev_shmem.c:(.text+0xeec7d2): undefined reference to `drm_gem_fb_get_obj'
x86_64-linux-ld: vmlinux.o: in function `drm_fbdev_shmem_fb_mmap':
drm_fbdev_shmem.c:(.text+0xeec9f6): undefined reference to `drm_gem_fb_get_obj'
x86_64-linux-ld: vmlinux.o: in function `drm_fbdev_shmem_defio_imageblit':
(.rodata+0x5b2288): undefined reference to `drm_fb_helper_check_var'
x86_64-linux-ld: (.rodata+0x5b2290): undefined reference to `drm_fb_helper_set_par'
This can happen for a number of device drivers that select DRM_GEM_SHMEM_HELPER
without also selecting DRM_KMS_HELPER. To work around this, add another select
that forces DRM_KMS_HELPER to be built-in rather than a loadable module, but
only if FBDEV emulation is also enabled. DRM_TTM_HELPER and DRM_GEM_DMA_HELPER
look like they have the same problem in theory even if there is no possible
configuration that shows it. For consistency, do the same change to those.
Closes: https://lore.kernel.org/all/20250121-greedy-flounder-of-abundance-4d2ee8-mkl@pengutronix.de Reported-by: Marc Kleine-Budde <mkl@pengutronix.de> Tested-by: Marc Kleine-Budde <mkl@pengutronix.de> Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de> Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de> Link: https://patchwork.freedesktop.org/patch/msgid/20250122090211.3161186-1-arnd@kernel.org
Lucas De Marchi [Fri, 17 Jan 2025 16:45:29 +0000 (08:45 -0800)]
MAINTAINERS: Also exclude xe for drm-misc
When the xe driver was added, it didn't extend the exclude entries for
drm-misc, as done in commit 5a44d50f0072 ("MAINTAINERS: Update drm-misc
entry to match all drivers"). Exclude it like is done for i915 and other
drivers with dedicated maintainers.
Michal Wajdeczko [Mon, 20 Jan 2025 23:24:43 +0000 (00:24 +0100)]
drm/xe/pf: Fix migration initialization
The migration support only needs to be initialized once, but it
was incorrectly called from the xe_gt_sriov_pf_init_hw(), which
is part of the reset flow and may be called multiple times.
Ashutosh Dixit [Fri, 17 Jan 2025 03:21:55 +0000 (19:21 -0800)]
drm/xe/oa: Preserve oa_ctrl unused bits
UMD's have interest in setting unused bits of the oa_ctrl register "out of
band" for certain experiments. To facilitate this, don't clobber previous
oa_ctrl unused bits, i.e. rmw the values rather than simply write them.
Rodrigo Vivi [Fri, 17 Jan 2025 19:38:27 +0000 (14:38 -0500)]
drm/xe/uapi: Fix documentation indentation
Fix these issues:
Documentation/gpu/driver-uapi:29: include/uapi/drm/xe_drm.h:817: WARNING:
+Bullet list ends without a blank line; unexpected unindent.
Documentation/gpu/driver-uapi:29: include/uapi/drm/xe_drm.h:835: WARNING:
+Definition list ends without a blank line; unexpected unindent.
Linus Torvalds [Sun, 19 Jan 2025 17:33:40 +0000 (09:33 -0800)]
Merge tag 'x86_urgent_for_v6.13' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull x86 fixes from Borislav Petkov:
- Mark serialize() noinstr so that it can be used from instrumentation-
free code
- Make sure FRED's RSP0 MSR is synchronized with its corresponding
per-CPU value in order to avoid double faults in hotplug scenarios
- Disable EXECMEM_ROX on x86 for now because it didn't receive proper
x86 maintainers review, went in and broke a bunch of things
* tag 'x86_urgent_for_v6.13' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/asm: Make serialize() always_inline
x86/fred: Fix the FRED RSP0 MSR out of sync with its per-CPU cache
x86: Disable EXECMEM_ROX support
projects / thirdparty / linux.git / log
