Userspace wants to know the size of CPU visible portion of device
local-memory, and on small BAR devices the probed_size is no longer
enough. In Vulkan, for example, it would like to know the size in bytes
for CPU visible VkMemoryHeap. We already track the io_size for each
region, so plumb that through to the region query.
v2: Drop the ( -1 = unknown ) stuff, which is confusing since nothing
can currently ever return such a value.
Testcase: igt@i915_query@query-regions-sanity-check
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Cc: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Cc: Jon Bloomfield <jon.bloomfield@intel.com>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: Jordan Justen <jordan.l.justen@intel.com>
Cc: Kenneth Graunke <kenneth@whitecape.org>
Cc: Akeem G Abodunrin <akeem.g.abodunrin@intel.com>
Acked-by: Nirmoy Das <nirmoy.das@intel.com>
Reviewed-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220629174350.384910-2-matthew.auld@intel.com
UAPI Changes:
- Expose per tile media freq factor in sysfs (Ashutosh Dixit, Dale B Stimson)
- Document memory residency and Flat-CCS capability of obj (Ramalingam C)
- Disable GETPARAM lookups of I915_PARAM_[SUB]SLICE_MASK on Xe_HP+ (Matt Roper)
Cross-subsystem Changes:
- Rename intel-gtt symbols (Lucas De Marchi)
Core Changes:
Driver Changes:
- Support programming the EU priority in the GuC descriptor (DG2) (Matthew Brost)
- DG2 HuC loading support (Daniele Ceraolo Spurio)
- Fix build error without CONFIG_PM (YueHaibing)
- Enable THP on Icelake and beyond (Tvrtko Ursulin)
- Only setup private tmpfs mount when needed and fix logging (Tvrtko Ursulin)
- Make __guc_reset_context aware of guilty engines (Umesh Nerlige Ramappa)
- DG2 small bar memory probing fixes (Nirmoy Das)
- Remove unnecessary GuC err capture noise (Alan Previn)
- Fix i915_gem_object_ggtt_pin_ww regression on old platforms (Maarten Lankhorst)
- Fix undefined behavior in GuC backend due to shift overflowing the constant (Borislav Petkov)
- New DG2 workarounds (Swathi Dhanavanthri, Anshuman Gupta)
- Report no hwconfig support on ADL-N (Balasubramani Vivekanandan)
- Fix error_state_read ptr + offset use (Alan Previn)
- Expose per tile media freq factor in sysfs (Ashutosh Dixit, Dale B Stimson)
- Fix memory leaks in per-gt sysfs (Ashutosh Dixit)
- Fix dma_resv fence handling in multi-batch execbuf (Nirmoy Das)
- Add extra registers to GPU error dump on Gen11+ (Stuart Summers)
- More PVC+DG2 workarounds (Matt Roper)
- Improve user experience and driver robustness under SIGINT or similar (Tvrtko Ursulin)
- Don't show engine classes not present (Tvrtko Ursulin)
- Improve on suspend / resume time with VT-d enabled (Thomas Hellström)
- Add missing else (katrinzhou)
- Don't leak lmem mapping in vma_evict (Juha-Pekka Heikkila)
- Add smem fallback allocation for dpt (Juha-Pekka Heikkila)
- Tweak the ordering in cpu_write_needs_clflush (Matthew Auld)
- Do not access rq->engine without a reference (Niranjana Vishwanathapura)
- Revert "drm/i915: Hold reference to intel_context over life of i915_request" (Niranjana Vishwanathapura)
- Don't update engine busyness stats too frequently (Alan Previn)
- Add additional steps for Wa_22011802037 for execlist backend (Umesh Nerlige Ramappa)
- Fix a lockdep warning at error capture (Nirmoy Das)
- Ponte Vecchio prep work and new blitter engines (Matt Roper, John Harrison, Lucas De Marchi)
- Read correct RP_STATE_CAP register (PVC) (Matt Roper)
- Define MOCS table for PVC (Ayaz A Siddiqui)
- Driver refactor and support Ponte Vecchio forcewake handling (Matt Roper)
- Remove additional 3D flags from PIPE_CONTROL (Ponte Vecchio) (Stuart Summers)
- XEHPSDV and PVC do not use HuC (Daniele Ceraolo Spurio)
- Extract stepping information from PCI revid (Ponte Vecchio) (Matt Roper)
- Add initial PVC workarounds (Stuart Summers)
- SSEU handling driver refactor and Ponte Vecchio support (Matt Roper)
- GuC depriv applies to PVC (Matt Roper)
- Add register steering (Ponte Vecchio) (Matt Roper)
- Add recommended MMIO setting (Ponte Vecchio) (Matt Roper)
- Move multicast register handling to a dedicated file (Matt Roper)
- Cleanup interface for MCR operations (Matt Roper)
- Extend i915_vma_pin_iomap() (CQ Tang)
- Re-do the intel-gtt split (Lucas De Marchi)
- Correct duplicated/misplaced GT register definitions (Matt Roper)
- Prefer "XEHP_" prefix for registers (Matt Roper)
- Don't use DRM_DEBUG_WARN_ON for unexpected l3bank/mslice config (Tvrtko Ursulin)
- Don't use DRM_DEBUG_WARN_ON for ring unexpectedly not idle (Tvrtko Ursulin)
- Make drop_pages() return bool (Lucas De Marchi)
- Fix CFI violation with show_dynamic_id() (Nathan Chancellor)
- Use i915_probe_error instead of drm_error in GuC code (Vinay Belgaumkar)
- Fix use of static in macro mismatch (Andi Shyti)
- Update tiled blits selftest (Bommu Krishnaiah)
- Future-proof platform checks (Matt Roper)
- Only include what's needed (Jani Nikula)
- remove accidental static from a local variable (Jani Nikula)
- Add global forcewake request to drpc (Vinay Belgaumkar)
- Fix spelling typo in comment (pengfuyuan)
- Increase timeout for live_parallel_switch selftest (Akeem G Abodunrin)
- Use non-blocking H2G for waitboost (Vinay Belgaumkar)
Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/YrwtLM081SQUG1Dc@tursulin-desk
The only difference between the ADL S and P GuC FWs is the HWConfig
support. ADL-N does not support HWConfig, so we should use the same
binary as ADL-S, otherwise the GuC might attempt to fetch a config
table that does not exist. ADL-N is internally identified as an ADL-P,
so we need to special-case it in the FW selection code.
Fixes: 7e28d0b267 ("drm/i915/adl-n: Enable ADL-N platform")
Cc: John Harrison <John.C.Harrison@Intel.com>
Cc: Tejas Upadhyay <tejas.upadhyay@intel.com>
Cc: Anusha Srivatsa <anusha.srivatsa@intel.com>
Cc: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Reviewed-by: Matt Roper <matthew.d.roper@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220621233005.3952293-1-daniele.ceraolospurio@intel.com
Display is turned off by i915_drm_suspend() during the suspend
procedure, removing the last reference of some gem objects that were
used by display.
The issue is that those objects are only actually freed when
mm.free_work executed and that can happen very late in the suspend
process causing issues.
So here draining all freed objects released by display fixing suspend
issues.
Signed-off-by: José Roberto de Souza <jose.souza@intel.com>
Reviewed-by: Matt Atwood <matthew.s.atwood@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220629134721.48375-1-jose.souza@intel.com
We've just set up device info in i915_driver_create() so we can use
DISPLAY_VER() intead of looking at match_info directly.
Semantically we want to check the display version instead of the
graphics version, and for the earlier platforms they are always the
same.
v2: Use DISPLAY_VER() instead of GRAPHICS_VER() (Ville)
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220628141005.226252-1-jani.nikula@intel.com
The ddc pin and aux channel sanitization may disable DVI/HDMI and DP,
respectively, of ports parsed earlier, in "last one wins" fashion. With
parsing and printing interleaved, we'll end up logging support first and
disabling later anyway.
Now that we've split ddi port info parsing and printing, take it further
by doing the printing in a separate loop, fixing the logging.
Note that this also changes the logging order from VBT child device
order to port number order.
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220621123732.1118437-1-jani.nikula@intel.com
This test will validate we can achieve actual frequency of RP0. Pcode
grants frequencies based on what GuC is requesting. However, thermal
throttling can limit what is being granted. Add a test to request for
max, but don't fail the test if RP0 is not granted due to throttle
reasons.
Also optimize the selftest by using a common run_test function to avoid
code duplication. Rename the "clamp" tests to vary_max_freq and
vary_min_freq.
v2: Fix compile warning
v3: Review comments (Ashutosh). Added a FIXME for the media RP0 case.
v4: Checkpatch (strict) fixes, remove FIXME and other comments (Ashutosh)
Fixes commit 8ee2c22782 ("drm/i915/guc/slpc: Add SLPC selftest")
Cc: Ashutosh Dixit <ashutosh.dixit@intel.com>
Signed-off-by: Vinay Belgaumkar <vinay.belgaumkar@intel.com>
Reviewed-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220627230346.27720-1-vinay.belgaumkar@intel.com
For some platfroms we use stop_machine version of
gen8_ggtt_insert_page/gen8_ggtt_insert_entries to avoid a
concurrent GGTT access bug but this causes a circular locking
dependency warning:
Possible unsafe locking scenario:
CPU0 CPU1
---- ----
lock(&ggtt->error_mutex);
lock(dma_fence_map);
lock(&ggtt->error_mutex);
lock(cpu_hotplug_lock);
Fix this by calling gen8_ggtt_insert_page/gen8_ggtt_insert_entries
directly at error capture which is concurrent GGTT access safe because
reset path make sure of that.
v2: Fix rebase conflict and added a comment.
Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/5595
Reviewed-by: Gwan-gyeong Mun <gwan-gyeong.mun@intel.com>
Suggested-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Nirmoy Das <nirmoy.das@intel.com>
Reviewed-by: Andrzej Hajda <andrzej.hajda@intel.com>
Signed-off-by: Ramalingam C <ramalingam.c@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220624110821.29190-1-nirmoy.das@intel.com
If drm_connector_init fails, intel_connector_free will be called to take
care of proper free. So it is necessary to drop the refcount of port
before intel_connector_free.
Fixes: 091a4f9194 ("drm/i915: Handle drm-layer errors in intel_dp_add_mst_connector")
Signed-off-by: Hangyu Hua <hbh25y@gmail.com>
Reviewed-by: José Roberto de Souza <jose.souza@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220624130406.17996-1-jose.souza@intel.com
Signed-off-by: José Roberto de Souza <jose.souza@intel.com>
This function is supposed to return zero or negative error codes but it
accidentally returns true on failure.
Fixes: 92a020747d ("drm/i915: Split shared dpll .get_dplls() into compute and get phases")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/YrVcHArNMTGCSy+9@kili
Reviewed-by: José Roberto de Souza <jose.souza@intel.com>
SLPC min/max frequency updates require H2G calls. We are seeing
timeouts when GuC channel is backed up and it is unable to respond
in a timely fashion causing warnings and affecting CI.
This is seen when waitboosting happens during a stress test.
this patch updates the waitboost path to use a non-blocking
H2G call instead, which returns as soon as the message is
successfully transmitted.
v2: Use drm_notice to report any errors that might occur while
sending the waitboost H2G request (Tvrtko)
v3: Add drm_notice inside force_min_freq (Ashutosh)
Cc: Ashutosh Dixit <ashutosh.dixit@intel.com>
Signed-off-by: Vinay Belgaumkar <vinay.belgaumkar@intel.com>
Reviewed-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220623003225.23301-1-vinay.belgaumkar@intel.com
For execlists backend, current implementation of Wa_22011802037 is to
stop the CS before doing a reset of the engine. This WA was further
extended to wait for any pending MI FORCE WAKEUPs before issuing a
reset. Add the extended steps in the execlist path of reset.
In addition, extend the WA to gen11.
v2: (Tvrtko)
- Clarify comments, commit message, fix typos
- Use IS_GRAPHICS_VER for gen 11/12 checks
v3: (Daneile)
- Drop changes to intel_ring_submission since WA does not apply to it
- Log an error if MSG IDLE is not defined for an engine
Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Fixes: f6aa0d713c ("drm/i915: Add Wa_22011802037 force cs halt")
Acked-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220621192105.2100585-1-umesh.nerlige.ramappa@intel.com
Using two different types of workoads, it was observed that
guc_update_engine_gt_clks was being called too frequently and/or
causing a CPU-to-lmem bandwidth hit over PCIE. Details on
the workloads and numbers are in the notes below.
Background: At the moment, guc_update_engine_gt_clks can be invoked
via one of 3 ways. #1 and #2 are infrequent under normal operating
conditions:
1.When a predefined "ping_delay" timer expires so that GuC-
busyness can sample the GTPM clock counter to ensure it
doesn't miss a wrap-around of the 32-bits of the HW counter.
(The ping_delay is calculated based on 1/8th the time taken
for the counter go from 0x0 to 0xffffffff based on the
GT frequency. This comes to about once every 28 seconds at a
GT frequency of 19.2Mhz).
2.In preparation for a gt reset.
3.In response to __gt_park events (as the gt power management
puts the gt into a lower power state when there is no work
being done).
Root-cause: For both the workloads described farther below, it was
observed that when user space calls IOCTLs that unparks the
gt momentarily and repeats such calls many times in quick succession,
it triggers calling guc_update_engine_gt_clks as many times. However,
the primary purpose of guc_update_engine_gt_clks is to ensure we don't
miss the wraparound while the counter is ticking. Thus, the solution
is to ensure we skip that check if gt_park is calling this function
earlier than necessary.
Solution: Snapshot jiffies when we do actually update the busyness
stats. Then get the new jiffies every time intel_guc_busyness_park
is called and bail if we are being called too soon. Use half of the
ping_delay as a safe threshold.
NOTE1: Workload1: IGTs' gem_create was modified to create a file handle,
allocate memory with sizes that range from a min of 4K to the max supported
(in power of two step-sizes). Its maps, modifies and reads back the
memory. Allocations and modification is repeated until total memory
allocation reaches the max. Then the file handle is closed. With this
workload, guc_update_engine_gt_clks was called over 188 thousand times
in the span of 15 seconds while this test ran three times. With this patch,
the number of calls reduced to 14.
NOTE2: Workload2: 30 transcode sessions are created in quick succession.
While these sessions are created, pcm-iio tool was used to measure I/O
read operation bandwidth consumption sampled at 100 milisecond intervals
over the course of 20 seconds. The total bandwidth consumed over 20 seconds
without this patch was measured at average at 311KBps per sample. With this
patch, the number went down to about 175Kbps which is about a 43% savings.
Signed-off-by: Alan Previn <alan.previn.teres.alexis@intel.com>
Reviewed-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Acked-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220623023157.211650-2-alan.previn.teres.alexis@intel.com
This reverts commit 1e98d8c52e.
The problem with this patch is that it makes i915_request to hold a
reference to intel_context, which in turn holds a reference on the VM.
This strong back referencing can lead to reference loops which leads
to resource leak.
An example is the upcoming VM_BIND work which requires VM to hold
a reference to some shared VM specific BO. But this BO's dma-resv
fences holds reference to the i915_request thus leading to reference
loop.
v2:
Do not use reserved requests for virtual engines
Signed-off-by: Niranjana Vishwanathapura <niranjana.vishwanathapura@intel.com>
Signed-off-by: Ramalingam C <ramalingam.c@intel.com>
Suggested-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Mathew Brost <matthew.brost@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220614184348.23746-3-ramalingam.c@intel.com
In i915_fence_get_driver_name(), user may not hold a
reference to rq->engine. Hence do not access it. Instead,
store required device private pointer in 'rq->i915' and use it.
Signed-off-by: Niranjana Vishwanathapura <niranjana.vishwanathapura@intel.com>
Suggested-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Ramalingam C <ramalingam.c@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220614184348.23746-2-ramalingam.c@intel.com
Stop hand rolling _MMIO_TRANS2() and just use the real thing.
Note that this register isn't even used atm, hence why this
builds despite the _PIPE2() macro not actually existing.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220623130900.26078-8-ville.syrjala@linux.intel.com
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Eliminate the PIPECONF RMWs from .comit_commit() so
that we can finally declare the whole vblank evade part
(and the noarm() part) of the pipe commit free of register
reads. Or at least I hope that's the last read...
Only the i9xx/ilk codepaths need this for now, but let's
add the same thing for hsw+ just in case we want to start
calling that during fastsets at some point (eg. to change
dithering settings/etc.).
Should open up the way to start experimenting with
different DSB usage approaches for pipe commits.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220413192607.27533-1-ville.syrjala@linux.intel.com
Reviewed-by: Uma Shankar <uma.shankar@intel.com>
For imported dma-buf objects we leave the object as cache_coherent = 0
across all platforms, which is reasonable given that have no clue what
the memory underneath is, and its not like the driver can ever manually
clflush the pages anyway (like with i915_gem_clflush_object) for such
objects. However on discrete we choose to treat cache_dirty = true as a
programmer error, leading to a warning. The simplest fix looks to be to
just change the ordering in cpu_write_needs_clflush to prevent ever
setting cache_dirty for dma-buf objects on discrete.
Fixes: d028a7690d ("drm/i915/dmabuf: Fix prime_mmap to work when using LMEM")
Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/5266
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Cc: Gwan-gyeong Mun <gwan-gyeong.mun@intel.com>
Reviewed-by: Gwan-gyeong Mun <gwan-gyeong.mun@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220622155919.355081-1-matthew.auld@intel.com
(cherry picked from commit 563aaf4a92)
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Currently i915 disables d3cold for i915 pci dev.
This blocks D3 for i915 gfx pci upstream bridge (VSP).
Let's disable d3cold at gfx root port to make sure that
i915 gfx VSP can transition to D3 to save some power.
We don't need to disable/enable d3cold in rpm, s2idle
suspend/resume handlers. Disabling/Enabling d3cold at
gfx root port in probe/remove phase is sufficient.
Fixes: 1a085e2341 ("drm/i915: Disable D3Cold in s2idle and runtime pm")
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Anshuman Gupta <anshuman.gupta@intel.com>
Reviewed-by: Badal Nilawar <badal.nilawar@intel.com>
Acked-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220616122249.5007-1-anshuman.gupta@intel.com
(cherry picked from commit 138c2fca6f)
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
We've been introducing new registers with a mix of "XEHP_"
(architecture) and "XEHPSDV_" (platform) prefixes. For consistency,
let's settle on "XEHP_" as the preferred form.
XEHPSDV_RP_STATE_CAP stays with its current name since that's truly a
platform-specific register and not something that applies to the Xe_HP
architecture as a whole.
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Caz Yokoyama <caz@caztech.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220624210328.308630-2-matthew.d.roper@intel.com
XEHPSDV_FLAT_CCS_BASE_ADDR, GEN8_L3_LRA_1_GPGPU, and MMCD_MISC_CTRL were
duplicated between i915_reg.h and intel_gt_regs.h. These are all GT
registers, so we should drop the copy from i915_reg.h.
XEHPSDV_TILE0_ADDR_RANGE was defined in i915_reg.h, but really belongs
in intel_gt_regs.h. Move it.
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220624210328.308630-1-matthew.d.roper@intel.com
For imported dma-buf objects we leave the object as cache_coherent = 0
across all platforms, which is reasonable given that have no clue what
the memory underneath is, and its not like the driver can ever manually
clflush the pages anyway (like with i915_gem_clflush_object) for such
objects. However on discrete we choose to treat cache_dirty = true as a
programmer error, leading to a warning. The simplest fix looks to be to
just change the ordering in cpu_write_needs_clflush to prevent ever
setting cache_dirty for dma-buf objects on discrete.
Fixes: d028a7690d ("drm/i915/dmabuf: Fix prime_mmap to work when using LMEM")
Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/5266
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Cc: Gwan-gyeong Mun <gwan-gyeong.mun@intel.com>
Reviewed-by: Gwan-gyeong Mun <gwan-gyeong.mun@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220622155919.355081-1-matthew.auld@intel.com
Currently i915 disables d3cold for i915 pci dev.
This blocks D3 for i915 gfx pci upstream bridge (VSP).
Let's disable d3cold at gfx root port to make sure that
i915 gfx VSP can transition to D3 to save some power.
We don't need to disable/enable d3cold in rpm, s2idle
suspend/resume handlers. Disabling/Enabling d3cold at
gfx root port in probe/remove phase is sufficient.
Fixes: 1a085e2341 ("drm/i915: Disable D3Cold in s2idle and runtime pm")
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Anshuman Gupta <anshuman.gupta@intel.com>
Reviewed-by: Badal Nilawar <badal.nilawar@intel.com>
Acked-by:: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220616122249.5007-1-anshuman.gupta@intel.com
When the last reference of a gem object is removed it is added to the
mm.free_list list and mm.free_work is queued to actually free the
object.
So gem objects that had their last reference removed by display during
drm_atomic_helper_shutdown() are added to mm.free_list what could
cause that mm.free_work is executed at the same time as
intel_runtime_pm_driver_release() causing raw-wakerefs warning.
So here only calling i915_gem_suspend() and by consequence
i915_gem_drain_freed_objects() only after display is down making
sure all display gem objecs are freed when
intel_runtime_pm_driver_release() is executed.
Signed-off-by: José Roberto de Souza <jose.souza@intel.com>
Reviewed-by: Matt Roper <matthew.d.roper@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220617190629.355356-1-jose.souza@intel.com
Re-do what was attempted in commit 7a5c922377 ("drm/i915/gt: Split
intel-gtt functions by arch"). The goal of that commit was to split the
handlers for older hardware that depend on intel-gtt.ko so i915 can
be built for non-x86 archs, after some more patches. Other archs do not
need intel-gtt.ko.
Main issue with the previous approach: it moved all the hooks, including
the gen8, which is used by all platforms gen8 and newer. Re-do the
split moving only the handlers for gen < 6, which are the only ones
calling out to the separate module.
While at it do some minor cleanups:
- Rename the prefix s/gen5_/gmch_/ to be more accurate what platforms
are covered by intel_ggtt_gmch.c
- Remove dead code for gen12 out of needs_idle_maps()
- Remove TODO comment leftover
- Re-order if/else ladder in ggtt_probe_hw() to keep newest platforms
first
v2: Add minor cleanups (Matt Roper)
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
Acked-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Matt Roper <matthew.d.roper@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220617230559.2109427-2-lucas.demarchi@intel.com
Exporting the symbols like intel_gtt_* creates some confusion inside
i915 that has symbols named similarly. In an attempt to isolate
platforms needing intel-gtt.ko, commit 7a5c922377 ("drm/i915/gt: Split
intel-gtt functions by arch") moved way too much
inside gt/intel_gt_gmch.c, even the functions that don't callout to this
module. Rename the symbols to make the separation clear.
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220617230559.2109427-1-lucas.demarchi@intel.com
In the future display might try call this with a normal smem object,
which doesn't require PIN_MAPPABLE underneath in order to CPU map the
pages (unlike stolen). Extend i915_vma_pin_iomap() to directly use
i915_gem_object_pin_map() for such cases.
This change was suggested by Chris P Wilson, that we pin
the smem with i915_gem_object_pin_map_unlocked().
v2 (jheikkil): Change i915_gem_object_pin_map_unlocked to
i915_gem_object_pin_map
Signed-off-by: CQ Tang <cq.tang@intel.com>
Signed-off-by: Juha-Pekka Heikkila <juhapekka.heikkila@gmail.com>
Cc: Chris Wilson <chris.p.wilson@intel.com>
Cc: Jari Tahvanainen <jari.tahvanainen@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
[mauld: tweak commit message, plus minor checkpatch fix]
Link: https://patchwork.freedesktop.org/patch/msgid/20220610121205.29645-2-juhapekka.heikkila@gmail.com
We have seen multiple RC6 issues where it is useful to know
which global forcewake bits are set. Add this to the 'drpc'
debugfs output.
v2: Review comments (Ashutosh)
Reviewed-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
Signed-off-by: Vinay Belgaumkar <vinay.belgaumkar@intel.com>
Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220617212032.34577-1-vinay.belgaumkar@intel.com
drm_crtc.h has no need for drm_blend.h, so don't include it.
Avoids useless rebuilds of the entire universe when
touching drm_blend.h.
Quite a few placs do currently depend on drm_blend.h without
actually including it directly. All of those need to be fixed
up.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220613200317.11305-4-ville.syrjala@linux.intel.com
Acked-by: Sam Ravnborg <sam@ravnborg.org>
Acked-by: Jani Nikula <jani.nikula@intel.com>
drm_crtc.h has no need for drm_frambuffer.h, so don't include it.
Avoids useless rebuilds of the entire universe when
touching drm_framebuffer.h.
Quite a few placs do currently depend on drm_framebuffer.h without
actually including it directly. All of those need to be fixed
up.
v2: Fix up msm some more
v2: Deal with ingenic and shmobile as well
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220614095449.29311-1-ville.syrjala@linux.intel.com
Acked-by: Sam Ravnborg <sam@ravnborg.org>
Acked-by: Jani Nikula <jani.nikula@intel.com>
drm_crtc.h has no need for drm_edid.h, so don't include it.
Avoids useless rebuilds of the entire universe when
touching drm_edid.h.
Quite a few placs do currently depend on drm_edid.h without
actually including it directly. All of those need to be fixed
up.
v2: Fix up i915 and msm some more
v3: Fix alphabetical ordering (Sam)
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220614090245.30283-1-ville.syrjala@linux.intel.com
Acked-by: Sam Ravnborg <sam@ravnborg.org>
Acked-by: Jani Nikula <jani.nikula@intel.com>
Each LFP may have different panel type which is stored in LFP data
data block. Based on the child device index respective panel-type/
panel-type2 field will be used.
v1: Initial rfc verion.
v2: Based on review comments from Jani,
- Used panel-type instead addition panel-index variable.
- DEVICE_HANDLE_* name changed and placed before DEVICE_TYPE_*
macro.
v3:
- passing intel_bios_encoder_data as argument of
intel_bios_init_panel(). Passing NULL to indicate encoder is not
initialized yet for dsi as current focus is to enable dual EDP. [Jani]
v4:
- encoder->devdata used which is initialized before from vbt
structure. [Jani]
Signed-off-by: Animesh Manna <animesh.manna@intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Acked-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220620065138.5126-1-animesh.manna@intel.com
This reverts commit 73867c8709 ("drm/i915/display: Remove check for
low voltage sku for max dp source rate"), which, on an i7-11850H iGPU
with a Thinkpad X1 Extreme Gen 4, attached to a LG LP160UQ1-SPB1
embedded panel, causes wild flickering glitching technicolor
pyrotechnics on resumption from suspend. The display shows strobing
colors in an utter disaster explosion of pantone, as though bombs were
dropped on the leprechauns at the base of the rainbow.
Rebooting the machine fixes the issue, presumably because the display is
initialized by firmware rather than by i915. Otherwise, the GPU appears
to work fine.
Bisection traced it back to this commit, which makes sense given the
issues.
Note: This re-opens, and puts back to the drawing board,
https://gitlab.freedesktop.org/drm/intel/-/issues/5272 which was fixed
by the regressing commit.
Fixes: 73867c8709 ("drm/i915/display: Remove check for low voltage sku for max dp source rate")
Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/6205
Cc: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Cc: Imre Deak <imre.deak@intel.com>
Cc: Jani Nikula <jani.nikula@intel.com>
Cc: Uma Shankar <uma.shankar@intel.com>
Cc: Animesh Manna <animesh.manna@intel.com>
Cc: Jani Saarinen <jani.saarinen@intel.com>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220613102241.9236-1-Jason@zx2c4.com
(cherry picked from commit d592983508)
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
This reverts commit 73867c8709 ("drm/i915/display: Remove check for
low voltage sku for max dp source rate"), which, on an i7-11850H iGPU
with a Thinkpad X1 Extreme Gen 4, attached to a LG LP160UQ1-SPB1
embedded panel, causes wild flickering glitching technicolor
pyrotechnics on resumption from suspend. The display shows strobing
colors in an utter disaster explosion of pantone, as though bombs were
dropped on the leprechauns at the base of the rainbow.
Rebooting the machine fixes the issue, presumably because the display is
initialized by firmware rather than by i915. Otherwise, the GPU appears
to work fine.
Bisection traced it back to this commit, which makes sense given the
issues.
Note: This re-opens, and puts back to the drawing board,
https://gitlab.freedesktop.org/drm/intel/-/issues/5272 which was fixed
by the regressing commit.
Fixes: 73867c8709 ("drm/i915/display: Remove check for low voltage sku for max dp source rate")
Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/6205
Cc: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Cc: Imre Deak <imre.deak@intel.com>
Cc: Jani Nikula <jani.nikula@intel.com>
Cc: Uma Shankar <uma.shankar@intel.com>
Cc: Animesh Manna <animesh.manna@intel.com>
Cc: Jani Saarinen <jani.saarinen@intel.com>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220613102241.9236-1-Jason@zx2c4.com
Stop displaying engine classes with no engines - it is not a huge problem
if they are shown, since the values will correctly be all zeroes, but it
does count as misleading.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Fixes: 055634e4b6 ("drm/i915: Expose client engine utilisation via fdinfo")
Cc: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Reviewed-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220616140056.559074-1-tvrtko.ursulin@linux.intel.com
(cherry picked from commit 9f1b1d0b22)
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
When DMAR / VT-d is enabled, the display engine uses overfetching,
presumably to deal with the increased latency. To avoid display engine
errors and DMAR faults, as a workaround the GGTT is populated with scatch
PTEs when VT-d is enabled. However starting with gen10, Write-combined
writing of scratch PTES is no longer possible and as a result, populating
the full GGTT with scratch PTEs like on resume becomes very slow as
uncached access is needed.
Therefore, on integrated GPUs utilize the fact that the PTEs are stored in
stolen memory which retain content across S3 suspend. Don't clear the PTEs
on suspend and resume. This improves on resume time with around 100 ms.
While 100+ms might appear like a short time it's 10% to 20% of total resume
time and important in some applications.
One notable exception is Intel Rapid Start Technology which may cause
stolen memory to be lost across what the OS percieves as S3 suspend.
If IRST is enabled or if we can't detect whether IRST is enabled, retain
the old workaround, clearing and re-instating PTEs.
As an additional measure, if we detect that the last ggtt pte was lost
during suspend, print a warning and re-populate the GGTT ptes
On discrete GPUs, the display engine scans out from LMEM which isn't
subject to DMAR, and presumably the workaround is therefore not needed,
but that needs to be verified and disabling the workaround for dGPU,
if possible, will be deferred to a follow-up patch.
v2:
- Rely on retained ptes to also speed up suspend and resume re-binding.
- Re-build GGTT ptes if Intel rst is enabled.
v3:
- Re-build GGTT ptes also if we can't detect whether Intel rst is enabled,
and if the guard page PTE and end of GGTT was lost.
v4:
- Fix some kerneldoc issues (Matthew Auld), rebase.
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Acked-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220617152856.249295-1-thomas.hellstrom@linux.intel.com
Let's replace the assortment of intel_gt_* and intel_uncore_* functions
that operate on MCR registers with a cleaner set of interfaces:
* intel_gt_mcr_read -- unicast read from specific instance
* intel_gt_mcr_read_any[_fw] -- unicast read from any non-terminated
instance
* intel_gt_mcr_unicast_write -- unicast write to specific instance
* intel_gt_mcr_multicast_write[_fw] -- multicast write to all instances
We'll also replace the historic "slice" and "subslice" terminology with
"group" and "instance" to match the documentation for more recent
platforms; these days MCR steering applies to more types of replication
than just slice/subslice.
v2:
- Reference the new kerneldoc from i915.rst. (Jani)
- Tweak the wording of the documentation for a couple functions to
clarify the difference between "_fw" and non-"_fw" forms.
v3:
- s/read/write/ to fix copy-paste mistake in a couple comments.
(Harish)
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Acked-by: Jani Nikula <jani.nikula@linux.intel.com>
Reviewed-by: Harish Chegondi <harish.chegondi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220615001019.1821989-3-matthew.d.roper@intel.com
Handling of multicast/replicated registers is spread across intel_gt.c
and intel_uncore.c today. As multicast handling and the related
steering logic gets more complicated with the addition of new platforms
and new rules it makes sense to centralize it all in one place.
For now the existing functions have been moved to the new .c/.h as-is.
Function renames and updates to operate in a more consistent manner will
be done in subsequent patches.
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Acked-by: Jani Nikula <jani.nikula@linux.intel.com>
Reviewed-by: Harish Chegondi <harish.chegondi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220615001019.1821989-2-matthew.d.roper@intel.com
Add new file intel_modeset_verify.c for high level modeset verification
code to declutter intel_display.h. The new file is supposed to be about
crtc/encoder/connector verification; the state verification for very
specific functionality such as plls or wm should be placed next to the
code it verifies.
Fix some minor checkpatch issues while at it.
v2: Rebase
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/b9b47c14316a9edb772a8b8f934eabe7e928dd76.1655372759.git.jani.nikula@intel.com
Stop displaying engine classes with no engines - it is not a huge problem
if they are shown, since the values will correctly be all zeroes, but it
does count as misleading.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Fixes: 055634e4b6 ("drm/i915: Expose client engine utilisation via fdinfo")
Cc: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Reviewed-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220616140056.559074-1-tvrtko.ursulin@linux.intel.com
We have long standing customer complaints that pressing Ctrl-C (or to the
effect of) causes engine resets with otherwise well behaving programs.
Not only is logging engine resets during normal operation not desirable
since it creates support incidents, but more fundamentally we should avoid
going the engine reset path when we can since any engine reset introduces
a chance of harming an innocent context.
Reason for this undesirable behaviour is that the driver currently does
not distinguish between banned contexts and non-persistent contexts which
have been closed.
To fix this we add the distinction between the two reasons for revoking
contexts, which then allows the strict timeout only be applied to banned,
while innocent contexts (well behaving) can preempt cleanly and exit
without triggering the engine reset path.
Note that the added context exiting category applies both to closed non-
persistent context, and any exiting context when hangcheck has been
disabled by the user.
At the same time we rename the backend operation from 'ban' to 'revoke'
which more accurately describes the actual semantics. (There is no ban at
the backend level since banning is a concept driven by the scheduling
frontend. Backends are simply able to revoke a running context so that
is the more appropriate name chosen.)
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Andrzej Hajda <andrzej.hajda@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220527072452.2225610-1-tvrtko.ursulin@linux.intel.com
Older VBTs don't have all the stuff we've defined for the
LVDS options block (40). In particular we're currently parsing
the DPS panel type bits even though they may not exist, which
could mean we end up flagging the machine as supporting static
DRRS when the VBT declared no such thing.
We don't actually have a clear idea which VBT versions have
which bits so we rely on the block size instead.
Here's a quick list from my VBT stash:
mgm version 108 -> 4 bytes
alv version 120 -> 4 bytes
cst version 134 -> 14 bytes
pnv version 144 -> 14 bytes
cl version 142 -> 16 bytes
ctg version 155 -> 24 bytes
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220615151445.8531-2-ville.syrjala@linux.intel.com
Acked-by: Jani Nikula <jani.nikula@intel.com>
Parsing the panel_type is a bit special and should be done
before we parse anything else potentially panel-specific from
the VBT. So move it out from parse_panel_options(). It doesn't
neet to be there anyway since it'll do its own LVDS options
block lookup.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220615151445.8531-1-ville.syrjala@linux.intel.com
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
The WARN shouldn't have been added yet. For the moment the
clock that gets passed here is just what the user has requested
(via the modeline) and may not be exactly what iCLKIP can
generate.
Later on the plan is to change things so that we already get
passed the exact clock here, at which point the WARN should
be reintroduced.
Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/6246
Fixes: 97708335b0 ("drm/i915: Introduce struct iclkip_params")
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220616095530.15024-1-ville.syrjala@linux.intel.com
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Make sure FIFO underrun reporting is flagged as disabled very
early during the state readout so that we don't get any
spurious FIFO underruns reports from intel_crtc_disable_noatomic().
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220615174851.20658-2-ville.syrjala@linux.intel.com
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Connector polling is waking up the polled device. Polling
is unnecessary if our device is known to not have display.
Fix this and save some power by disabling starting connector
polling when we are having headless sku. Use information from
opregion.
v2: Move headless sku check into INTEL_DISPLAY_ENABLED macro
Signed-off-by: Jouni Högander <jouni.hogander@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/20220610085429.52935-4-jouni.hogander@intel.com
Currently we are starting connector polling if display is disabled
using disable_display module parameter. Polling is just returning
always "not connected" state. This can be optimized by not starting
polling at all.
Signed-off-by: Jouni Högander <jouni.hogander@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/20220610085429.52935-3-jouni.hogander@intel.com
Export headless sku bit (bit 13) from opregion->header->pcon as an
interface to check if our device is headless configuration.
This is mainly targeted for hybrid gfx systems. E.g. when display
is not supposed to be connected discrete graphics card it's
opregion can inform this is headless graphics card.
v3: Dummy version is now static inline function
v2: Check also opregion version
Bspec: 53441
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Jouni Högander <jouni.hogander@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220610085429.52935-2-jouni.hogander@intel.com
As with past platforms, the bspec's performance tuning guide provides
recommended MMIO settings. Although not technically "workarounds" we
apply these through the workaround framework to ensure that they're
re-applied at the proper times (e.g., on engine resets) and that any
conflicts with real workarounds are flagged.
Bspec: 72161
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: José Roberto de Souza <jose.souza@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220613165314.862029-1-matthew.d.roper@intel.com
If we're treating each bit in the EU fuse register as a single EU
instead of a pair of EUs, then that also cuts the number of potential
EUs per subslice in half.
Fixes: 5ac342ef84 ("drm/i915/pvc: Add SSEU changes")
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Anusha Srivatsa <anusha.srivatsa@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220610230801.459577-1-matthew.d.roper@intel.com
The VBIOS/GOP may not program the FDI M/n vs. dotclock entirely
consistently. Eg. on a SNB Thinkpad X220 LVDS I see dotclock of
69.286 MHz (the best the DPLL can do) vs. FDI M/N 69.3 MHz
(matches what the EDID actually declares).
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220503182242.18797-15-ville.syrjala@linux.intel.com
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Pull the various iCLKIP parameters into a struct. Later on
we'll reuse this during the state computation to determine
the exact dotclock the hardware will be generating for us.
v2: Don't lose the phaseinc calculation
v3: Drop the misplaced '#include <intel_pch_refclk.h>' from intel_crt.c (Jani)
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220504212109.26369-1-ville.syrjala@linux.intel.com
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Extract intel_crtc_dotclock() from ddi_dotclock_get(). We'll reuse
this during state computation in order to determine the actual final
dotclcok after the DPLL computation has been done (which may not give
us the exact same port_clock that we fed in).
v2: Add the prototype
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220504123350.13235-1-ville.syrjala@linux.intel.com
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
The arrays are static const, but the pointer shouldn't be static.
Fixes: 3d832f370d ("drm/i915/uc: Allow platforms to have GuC but not HuC")
Cc: John Harrison <John.C.Harrison@Intel.com>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220511094619.27889-1-jani.nikula@intel.com
(cherry picked from commit 5821a0bbb4)
_i915_vma_move_to_active() can receive > 1 fences for
multiple batch buffers submission. Because dma_resv_add_fence()
can only accept one fence at a time, change _i915_vma_move_to_active()
to be aware of multiple fences so that it can add individual
fences to the dma resv object.
v6: fix multi-line comment.
v5: remove double fence reservation for batch VMAs.
v4: Reserve fences for composite_fence on multi-batch contexts and
also reserve fence slots to composite_fence for each VMAs.
v3: dma_resv_reserve_fences is not cumulative so pass num_fences.
v2: make sure to reserve enough fence slots before adding.
Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/5614
Fixes: 544460c338 ("drm/i915: Multi-BB execbuf")
Cc: <stable@vger.kernel.org> # v5.16+
Signed-off-by: Nirmoy Das <nirmoy.das@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Reviewed-by: Andrzej Hajda <andrzej.hajda@intel.com>
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220525095955.15371-1-nirmoy.das@intel.com
(cherry picked from commit 420a07b841)
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
All kmalloc'd kobjects need a kobject_put() to free memory. For example in
previous code, kobj_gt_release() never gets called. The requirement of
kobject_put() now results in a slightly different code organization.
v2: s/gtn/gt/ (Andi)
Fixes: b770bcfae9 ("drm/i915/gt: create per-tile sysfs interface")
Signed-off-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Acked-by: Andrzej Hajda <andrzej.hajda@intel.com>
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/a6f6686517c85fba61a0c45097f5bb4fe7e257fb.1653484574.git.ashutosh.dixit@intel.com
(cherry picked from commit 69d6bf5c37)
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Commit 3cf0507625 ("drm/i915/bios: Split VBT data into per-panel vs.
global parts") cause PSR to be disabled when enable_psr has the
default value and there is at least one DP port that do not supports
PSR.
That was happening because intel_psr_init() is called for every DP
port and then enable_psr is globaly set to 0 based on the PSR support
of the DP port.
Here dropping the enable_psr overwritten and using the VBT PSR value
when enable_psr is set as default.
Fixes: 3cf0507625 ("drm/i915/bios: Split VBT data into per-panel vs. global parts")
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: Jani Nikula <jani.nikula@intel.com>
Cc: Jouni Högander <jouni.hogander@intel.com>
Cc: Mika Kahola <mika.kahola@intel.com>
Signed-off-by: José Roberto de Souza <jose.souza@intel.com>
Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220608203344.513082-1-jose.souza@intel.com
Ponte Vecchio no longer has MSLICE or LNCF steering, but the bspec does
document several new types of multicast register ranges. Fortunately,
most of the different MCR types all provide valid values at instance
(0,0) so there's no need to read fuse registers and calculate a
non-terminated instance. We'll lump all of those range types (BSLICE,
HALFBSLICE, TILEPSMI, CC, and L3BANK) into a single category called
"INSTANCE0" to keep things simple. We'll also perform explicit steering
for each of these multicast register types, even if the implicit
steering setup for COMPUTE/DSS ranges would have worked too; this is
based on guidance from our hardware architects who suggested that we
move away from implicit steering and start explicitly steer all MCR
register accesses on modern platforms (we'll work on transitioning
COMPUTE/DSS to explicit steering in the future).
Note that there's one additional MCR range type defined in the bspec
(SQIDI) that we don't handle here. Those ranges use a different
steering control register that we never touch; since instance 0 is also
always a valid setting there, we can just ignore those ranges.
Finally, we'll rename the HAS_MSLICES() macro to HAS_MSLICE_STEERING().
PVC hardware still has units referred to as mslices, but there's no
register steering based on mslice for this platform.
v2:
- Rebase on other recent changes
- Swap two table rows to keep table sorted & easy to read. (Harish)
Bspec: 67609
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Harish Chegondi <harish.chegondi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220608170700.4026648-1-matthew.d.roper@intel.com
Another mistake during the conversion to DSS bitmaps: after retrieving
the DSS ID intel_sseu_find_first_xehp_dss() we forgot to modulo it down
to obtain which ID within the current gslice it is.
Fixes: b87d390196 ("drm/i915/sseu: Disassociate internal subslice mask representation from uapi")
Cc: Balasubramani Vivekanandan <balasubramani.vivekanandan@intel.com>
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Acked-by: Balasubramani Vivekanandan <balasubramani.vivekanandan@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220607175716.3338661-1-matthew.d.roper@intel.com
A new PVC+DG2 workaround has appeared recently:
- Wa_16015675438
And a couple existing DG2 workarounds have been extended to PVC:
- Wa_14015795083
- Wa_18018781329
Note that Wa_16015675438 asks us to program a register that is in the
0x2xxx range typically associated with the RCS engine, even though PVC
does not have an RCS. By default the GuC will think we've made a
mistake and throw an exception when it sees this register on a CCS
engine's save/restore list, so we need to pass an extra GuC control flag
to tell it that this is expected and not a problem.
Signed-off-by: Anshuman Gupta <anshuman.gupta@intel.com>
Signed-off-by: Badal Nilawar <badal.nilawar@intel.com>
Signed-off-by: Prathap Kumar Valsan <prathap.kumar.valsan@intel.com>
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Anshuman Gupta <anshuman.gupta@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220608005108.3717895-1-matthew.d.roper@intel.com
The eDP BDB block has gained yet another max link rate field.
Let's parse it and consult it during the source rate filtering.
v2: *20 instead of *2 to get the correct units (Jani)
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220602205723.11341-1-ville.syrjala@linux.intel.com
We're not parsing the 5.4 Gbps value for the old eDP fast link
training link rate, nor are we parsing the new fast link training
link rate field. Remedy both.
Also we'll now use the actual link rate instead of the DPCD BW
register value.
Note that we're not even using this information for anything
currently, so should perhaps just nuke it all unless someone
is planning on implementing fast link training finally...
v2: Stop using the DPCD BW values (Jani)
*20 instead of *2 to get the rate in correct units (Jani)
Cc: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220602205649.11283-1-ville.syrjala@linux.intel.com
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
We'll need to know the VBT panel_type before we can determine the
maximum link rate for eDP. To that end move
intel_dp_set_source_rates() & co. to be called after the per-panel
VBT parsing has been done.
intel_dp_mst_encoder_init() depends on the source rates so we'll
have to do it a bit later as well.
v2: Fix the intel_dp_mst_encoder_init() oops
Reviewed-by: Jani Nikula <jani.nikula@intel.com> #v1
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220603165841.15481-1-ville.syrjala@linux.intel.com
The arrays are static const, but the pointer shouldn't be static.
Fixes: 3d832f370d ("drm/i915/uc: Allow platforms to have GuC but not HuC")
Cc: John Harrison <John.C.Harrison@Intel.com>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220511094619.27889-1-jani.nikula@intel.com
Fix:
drivers/gpu/drm/i915/pxp/intel_pxp_debugfs.c:61:6: warning: symbol
'intel_pxp_debugfs_register' was not declared. Should it be static?
Sort and remove the redundant pxp prefixes from the includes while at
it.
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220506120405.2582372-1-jani.nikula@intel.com
When converting our DSS masks to bitmaps, we fumbled the condition used
to check whether any DSS are present in the first gslice. Since
intel_sseu_find_first_xehp_dss() returns a 0-based number, we need a >=
condition rather than >.
Fixes: b87d390196 ("drm/i915/sseu: Disassociate internal subslice mask representation from uapi")
Reported-by: Balasubramani Vivekanandan <balasubramani.vivekanandan@intel.com>
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Acked-by: Balasubramani Vivekanandan <balasubramani.vivekanandan@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220607154724.3155521-1-matthew.d.roper@intel.com
The intel_dp_dsc_get_output_bpp() function outputs two lines of
unconditional logs, which was okay when it was called only once. But
now, we also call this function from intel_dp_mode_valid(), which is
in turn called for every mode we need to validate. This causes a lot
of useless noise.
Remove the unconditional prints to avoid spamming the logs. Also
remove one more print that is not unconditional, but is related.
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220607074433.1202917-1-luca@coelho.fi
This series includes the following patchsets:
- bitmap: optimize bitmap_weight() usage(w/o bitmap_weight_cmp), from me;
- lib/bitmap.c make bitmap_print_bitmask_to_buf parseable, from Mauro
Carvalho Chehab;
- include/linux/find: Fix documentation, from Anna-Maria Behnsen;
- bitmap: fix conversion from/to fix-sized arrays, from me;
- bitmap: Fix return values to be unsigned, from Kees Cook.
It has been in linux-next for at least a week with no problems.
-----BEGIN PGP SIGNATURE-----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=U7DN
-----END PGP SIGNATURE-----
Merge tag 'bitmap-for-5.19-rc1' of https://github.com/norov/linux
Pull bitmap updates from Yury Norov:
- bitmap: optimize bitmap_weight() usage, from me
- lib/bitmap.c make bitmap_print_bitmask_to_buf parseable, from Mauro
Carvalho Chehab
- include/linux/find: Fix documentation, from Anna-Maria Behnsen
- bitmap: fix conversion from/to fix-sized arrays, from me
- bitmap: Fix return values to be unsigned, from Kees Cook
It has been in linux-next for at least a week with no problems.
* tag 'bitmap-for-5.19-rc1' of https://github.com/norov/linux: (31 commits)
nodemask: Fix return values to be unsigned
bitmap: Fix return values to be unsigned
KVM: x86: hyper-v: replace bitmap_weight() with hweight64()
KVM: x86: hyper-v: fix type of valid_bank_mask
ia64: cleanup remove_siblinginfo()
drm/amd/pm: use bitmap_{from,to}_arr32 where appropriate
KVM: s390: replace bitmap_copy with bitmap_{from,to}_arr64 where appropriate
lib/bitmap: add test for bitmap_{from,to}_arr64
lib: add bitmap_{from,to}_arr64
lib/bitmap: extend comment for bitmap_(from,to)_arr32()
include/linux/find: Fix documentation
lib/bitmap.c make bitmap_print_bitmask_to_buf parseable
MAINTAINERS: add cpumask and nodemask files to BITMAP_API
arch/x86: replace nodes_weight with nodes_empty where appropriate
mm/vmstat: replace cpumask_weight with cpumask_empty where appropriate
clocksource: replace cpumask_weight with cpumask_empty in clocksource.c
genirq/affinity: replace cpumask_weight with cpumask_empty where appropriate
irq: mips: replace cpumask_weight with cpumask_empty where appropriate
drm/i915/pmu: replace cpumask_weight with cpumask_empty where appropriate
arch/x86: replace cpumask_weight with cpumask_empty where appropriate
...
i915_pmu_cpu_online() calls cpumask_weight() to check if any bit of a
given cpumask is set. We can do it more efficiently with cpumask_empty()
because cpumask_empty() stops traversing the cpumask as soon as it finds
first set bit, while cpumask_weight() counts all bits unconditionally.
Signed-off-by: Yury Norov <yury.norov@gmail.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Split out audio registers to a header of its own to reduce the size of
i915_reg.h.
TODO: Remove direct audio register access from intel_ddi.c. However,
unification of audio get config is cumbersome due to the audio enable
bit being in the DP or HDMI registers on older platforms.
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Matt Roper <matthew.d.roper@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220602094542.1386151-1-jani.nikula@intel.com
Our internal teams have identified a few additional engine registers
that are worth inspecting in error state dumps during development &
debug. Let's capture and print them as part of our error dump.
For simplicity we'll just dump these registers on gen11 and beyond.
Most of these registers have existed since earlier platforms (e.g., gen6
or gen7) but were initially introduced only for a subset of the
platforms' engines; gen11 seems to be where they became available on all
engines.
Signed-off-by: Stuart Summers <stuart.summers@intel.com>
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: José Roberto de Souza <jose.souza@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220601210646.615946-1-matthew.d.roper@intel.com
PVC splits the mask of enabled DSS over two registers. It also changes
the meaning of the EU fuse register such that each bit represents a
single EU rather than a pair of EUs.
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Acked-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Balasubramani Vivekanandan <balasubramani.vivekanandan@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220601150725.521468-7-matthew.d.roper@intel.com
As with EU masks, it's easier to store subslice/DSS masks internally in
a format that's more natural for the driver to work with, and then only
covert into the u8[] uapi form when the query ioctl is invoked. Since
the hardware design changed significantly with Xe_HP, we'll use a union
to choose between the old "hsw-style" subslice masks or the newer xehp
mask. HSW-style masks will be stored in an array of u8's, indexed by
slice (there's never more than 6 subslices per slice on older
platforms). For Xe_HP and beyond where slices no longer exist, we only
need a single bitmask. However we already know that this mask is
eventually going to grow too large for a simple u64 to hold, so we'll
represent it in a manner that can be operated on by the utilities in
linux/bitmap.h.
v2:
- Fix typo: BIT(s) -> BIT(ss) in gen9_sseu_device_status()
v3:
- Eliminate sseu->ss_stride and just calculate the stride while
specifically handling uapi. (Tvrtko)
- Use BITMAP_BITS() macro to refer to size of masks rather than
passing I915_MAX_SS_FUSE_BITS directly. (Tvrtko)
- Report compute/geometry DSS masks separately when dumping Xe_HP SSEU
info. (Tvrtko)
- Restore dropped range checks to intel_sseu_has_subslice(). (Tvrtko)
v4:
- Make the bitmap size macro check the size of the .xehp field rather
than the containing union. (Tvrtko)
- Don't add GEM_BUG_ON() intel_sseu_has_subslice()'s check for whether
slice or subslice ID exceed sseu->max_[sub]slices; various loops
in the driver are expected to exceed these, so we should just
silently return 'false.'
v5:
- Move XEHP_BITMAP_BITS() to the header so that we can also replace a
usage of I915_MAX_SS_FUSE_BITS in one of the inline functions.
(Bala)
- Change the local variable in intel_slicemask_from_xehp_dssmask() from
u16 to 'unsigned long' to make it a bit more future-proof.
Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Cc: Balasubramani Vivekanandan <balasubramani.vivekanandan@intel.com>
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Acked-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Balasubramani Vivekanandan <balasubramani.vivekanandan@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220601150725.521468-6-matthew.d.roper@intel.com
Storing the EU mask internally in the same format the I915_QUERY
topology queries use makes the final copy_to_user() a bit simpler, but
makes the rest of the driver's SSEU more complicated and harder to
follow. Let's switch to an internal representation that's more natural:
Xe_HP platforms will be a simple array of u16 masks, whereas pre-Xe_HP
platforms will be a two-dimensional array, indexed by [slice][subslice].
We'll convert to the uapi format only when the query uapi is called.
v2:
- Drop has_common_ss_eumask. We waste some space repeating identical
EU masks for every single DSS, but the code is simpler without it.
(Tvrtko)
v3:
- Mask down EUs passed to sseu_set_eus at the callsite rather than
inside the function. (Tvrtko)
- Eliminate sseu->eu_stride and calculate it when needed. (Tvrtko)
Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Acked-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Balasubramani Vivekanandan <balasubramani.vivekanandan@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220601150725.521468-5-matthew.d.roper@intel.com
Although gen11 and gen12 architectures supported the concept of multiple
slices, in practice all the platforms that were actually designed only
had a single slice (i.e., note the parameters to 'intel_sseu_set_info'
that we pass for each platform). We can simplify the code slightly by
dropping the multi-slice logic from gen11+ platforms.
v2:
- Promote drm_dbg to drm_WARN_ON if the slice fuse register reports
unexpected fusing. (Tvrtko)
Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Acked-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Balasubramani Vivekanandan <balasubramani.vivekanandan@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220601150725.521468-4-matthew.d.roper@intel.com
Slice/subslice/EU information should be obtained via the topology
queries provided by the I915_QUERY interface; let's turn off support for
the old GETPARAM lookups on Xe_HP and beyond where we can't return
meaningful values.
The slice mask lookup is meaningless since Xe_HP doesn't support
traditional slices (and we make no attempt to return the various new
units like gslices, cslices, mslices, etc.) here.
The subslice mask lookup is even more problematic; given the distinct
masks for geometry vs compute purposes, the combined mask returned here
is likely not what userspace would want to act upon anyway. The value
is also limited to 32-bits by the nature of the GETPARAM ioctl which is
sufficient for the initial Xe_HP platforms, but is unable to convey the
larger masks that will be needed on other upcoming platforms. Finally,
the value returned here becomes even less meaningful when used on
multi-tile platforms where each tile will have its own masks.
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Acked-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Acked-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com> # mesa
Reviewed-by: Balasubramani Vivekanandan <balasubramani.vivekanandan@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220601150725.521468-3-matthew.d.roper@intel.com
Xe_HP has enough fundamental differences from previous platforms that it
makes sense to use a separate SSEU init function to keep things
straightforward and easy to understand. We'll also add a has_xehp_dss
flag to the SSEU structure that will be used by other upcoming changes.
v2:
- Add has_xehp_dss flag
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Acked-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Balasubramani Vivekanandan <balasubramani.vivekanandan@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220601150725.521468-2-matthew.d.roper@intel.com
- Improvements to mlx5 vfio-pci variant driver, including support
for parallel migration per PF (Yishai Hadas)
- Remove redundant iommu_present() check (Robin Murphy)
- Ongoing refactoring to consolidate the VFIO driver facing API
to use vfio_device (Jason Gunthorpe)
- Use drvdata to store vfio_device among all vfio-pci and variant
drivers (Jason Gunthorpe)
- Remove redundant code now that IOMMU core manages group DMA
ownership (Jason Gunthorpe)
- Remove vfio_group from external API handling struct file ownership
(Jason Gunthorpe)
- Correct typo in uapi comments (Thomas Huth)
- Fix coccicheck detected deadlock (Wan Jiabing)
- Use rwsem to remove races and simplify code around container and
kvm association to groups (Jason Gunthorpe)
- Harden access to devices in low power states and use runtime PM to
enable d3cold support for unused devices (Abhishek Sahu)
- Fix dma_owner handling of fake IOMMU groups (Jason Gunthorpe)
- Set driver_managed_dma on vfio-pci variant drivers (Jason Gunthorpe)
- Pass KVM pointer directly rather than via notifier (Matthew Rosato)
-----BEGIN PGP SIGNATURE-----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=etEI
-----END PGP SIGNATURE-----
Merge tag 'vfio-v5.19-rc1' of https://github.com/awilliam/linux-vfio
Pull vfio updates from Alex Williamson:
- Improvements to mlx5 vfio-pci variant driver, including support for
parallel migration per PF (Yishai Hadas)
- Remove redundant iommu_present() check (Robin Murphy)
- Ongoing refactoring to consolidate the VFIO driver facing API to use
vfio_device (Jason Gunthorpe)
- Use drvdata to store vfio_device among all vfio-pci and variant
drivers (Jason Gunthorpe)
- Remove redundant code now that IOMMU core manages group DMA ownership
(Jason Gunthorpe)
- Remove vfio_group from external API handling struct file ownership
(Jason Gunthorpe)
- Correct typo in uapi comments (Thomas Huth)
- Fix coccicheck detected deadlock (Wan Jiabing)
- Use rwsem to remove races and simplify code around container and kvm
association to groups (Jason Gunthorpe)
- Harden access to devices in low power states and use runtime PM to
enable d3cold support for unused devices (Abhishek Sahu)
- Fix dma_owner handling of fake IOMMU groups (Jason Gunthorpe)
- Set driver_managed_dma on vfio-pci variant drivers (Jason Gunthorpe)
- Pass KVM pointer directly rather than via notifier (Matthew Rosato)
* tag 'vfio-v5.19-rc1' of https://github.com/awilliam/linux-vfio: (38 commits)
vfio: remove VFIO_GROUP_NOTIFY_SET_KVM
vfio/pci: Add driver_managed_dma to the new vfio_pci drivers
vfio: Do not manipulate iommu dma_owner for fake iommu groups
vfio/pci: Move the unused device into low power state with runtime PM
vfio/pci: Virtualize PME related registers bits and initialize to zero
vfio/pci: Change the PF power state to D0 before enabling VFs
vfio/pci: Invalidate mmaps and block the access in D3hot power state
vfio: Change struct vfio_group::container_users to a non-atomic int
vfio: Simplify the life cycle of the group FD
vfio: Fully lock struct vfio_group::container
vfio: Split up vfio_group_get_device_fd()
vfio: Change struct vfio_group::opened from an atomic to bool
vfio: Add missing locking for struct vfio_group::kvm
kvm/vfio: Fix potential deadlock problem in vfio
include/uapi/linux/vfio.h: Fix trivial typo - _IORW should be _IOWR instead
vfio/pci: Use the struct file as the handle not the vfio_group
kvm/vfio: Remove vfio_group from kvm
vfio: Change vfio_group_set_kvm() to vfio_file_set_kvm()
vfio: Change vfio_external_check_extension() to vfio_file_enforced_coherent()
vfio: Remove vfio_external_group_match_file()
...
Some machines declare DRRS type = seamless, DRRS = no, DMRRS = yes.
I *think* DMRRS stands for "dynamcic media refresh rate", and
I suspect the way it's meant to work is that it lets the driver
switch refresh rates to match the frame rate for media playback.
Obviously for us all that kind of policy stuff is entirely up to
userspace, so the only thing we may do is make the extra refresh
rate(s) available.
So let's treat this case as just static DRRS for now. In the
future We might want to differentiate the "seamless w/ downclocking"
vs. "seamless w/o downclocking" cases so that we could do seamless
refresh rate changes for systems that only claim to support DMRRS.
Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/125
Acked-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220531191844.11313-7-ville.syrjala@linux.intel.com
It seem that when dealing with VRR capable eDP panels we need
to accept fixed modes with variable vblank length (which is what
VRR varies dynamically). Typically the preferred mode seems to be
a non-VRR more (lowish dotclock/refresh rate + short vblank).
We also have examples where it looks like even the hblank length
is a bit different between the preferred mode vs. VRR mode(s).
So let's just accept anything that has matching hdisp+vdisp+flags.
v2: Document that is_alt_drrs_mode() is a subset of is_alt_vrr_mode() (Jani)
Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/125
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220531191844.11313-4-ville.syrjala@linux.intel.com
Avoid bringing the entire machine down even if there's a bug that
shouldn't happen, but won't corrupt the system either. Log them loudly
and limp on.
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Andrzej Hajda <andrzej.hajda@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220531162527.1062319-1-jani.nikula@intel.com
For PVC, the base die and compute tile have separate stepping values
that we need to track; we'll use the existing graphics_step field to
represent the compute tile stepping and add a new 'basedie_step' field.
Unlike past platforms, steppings for these components are represented by
specific bitfields within the PCI revision ID, and we shouldn't make
assumptions about the non-CT, non-BD bits staying 0. Let's update our
stepping code accordingly.
Bspec: 44484
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Matt Atwood <matthew.s.atwood@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220527163348.1936146-2-matthew.d.roper@intel.com
Don't see a real reson not to check hw.active and hw.enable in
intel_pipe_config_compare(). We do have some checks for them
at a higher level, but I think better check them also in
intel_pipe_config_compare() in case something else doesn't
do a thorough enough job.
Also shuffle the mst_master_transcoder check next to the
cpu_transcoder check for a bit of consistency.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220503182242.18797-21-ville.syrjala@linux.intel.com
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Use the "[CRTC:%d:%s]'/etc. format for some of the modeset debugs
so we know more about what has happened during the modeset state
computation.
Also tweak the connector bpp debug message a bit to make it less
confusing.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220503182242.18797-10-ville.syrjala@linux.intel.com
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Use the state+crtc calling convention for intel_modeset_pipe_config()
and othere related functions. Many of these need the full atomic state
anyway so passing it all the way through is just nicer than having to
worry about whether it can actually be extracted from eg. the crtc
state passed in.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220503182242.18797-8-ville.syrjala@linux.intel.com
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Deduplicate the drm_rect comparisons.
We also drop the redundant pch_pfit.enabled check since the
pch_pfit.dst rectanble will be zeroed anyway when the pfit
is not enabled.
v2: Document why we drop the enabled check (Jani)
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220503182242.18797-7-ville.syrjala@linux.intel.com
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
The debugs in lower level DPLL code don't really provide any
useful extra information AFAICS. Better just streamline the
code and just put the necessary debugs (to identify at which
step the modeset failed) into the higher level code. In
addition we'll get the full state dump as well, which should
hopefully have enough information to figure out what went wrong.
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220503182242.18797-4-ville.syrjala@linux.intel.com
Split the DPLL state computation into a separate function
from the current .get_dplls() which currently serves a dual duty
by also reserving the shared DPLLs.
v2: s/false/-EINVAL/ (Jani)
Cc: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220503182242.18797-2-ville.syrjala@linux.intel.com
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Commit 30e114ef4b ("drm/i915/tc: Check for DP-alt, legacy sinks before
taking PHY ownership") defaults any disconnected Type-C ports to TBT-alt
mode which presents a problem (which could most likely result in a system
hang) when userspace forces a modeset on a Type-C port that is wired for
legacy HDMI. The following warning is seen when Weston forces a modeset
on a disconnected legacy Type-C port (HDMI) on a TGL based Gigabyte system:
(https://www.gigabyte.com/Mini-PcBarebone/GB-BSi3-1115G4-rev-10#ov)
Missing case (clock == 173000)
WARNING: CPU: 1 PID: 438 at drivers/gpu/drm/i915/display/intel_ddi.c:245
icl_ddi_tc_enable_clock.cold+0x16a/0x1cf [i915]
CPU: 1 PID: 438 Comm: kworker/u8:3 Tainted: G U W E
5.18.0-rc5-drm-tip+ #20
Hardware name: GIGABYTE GB-BSi3-1115G4/GB-BSi3-1115G4, BIOS F9
10/16/2021
Workqueue: i915_modeset intel_atomic_commit_work [i915]
RIP: 0010:icl_ddi_tc_enable_clock.cold+0x16a/0x1cf [i915]
Code: 74 6c 7f 10 81 fd d0 78 02 00 74 6d 81 fd b0 1e 04 00 74 70 48 63
d5 48 c7 c6 c0 7b ab c0 48 c7 c7 20 75 ab c0 e8 b8 b5 c1 f0 <0f> 0b 45
31 ed e9 fb fe ff ff 49 63 d5
48 c7 c6 80 7b ab c0 48 c7
RSP: 0018:ffff8882522c78f0 EFLAGS: 00010282
RAX: 0000000000000000 RBX: 0000000000000003 RCX: 0000000000000000
RDX: 0000000000000027 RSI: 0000000000000004 RDI: ffffed104a458f10
RBP: 0000000000011558 R08: ffffffffb078de4e R09: ffff888269ca748b
R10: ffffed104d394e91 R11: 0000000000000000 R12: ffff888255a318f8
R13: 0000000000000002 R14: ffff888255a30000 R15: ffff88823ef00348
FS: 0000000000000000(0000) GS:ffff888269c80000(0000)
knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007fd7afa42000 CR3: 0000000255c02004 CR4: 00000000007706e0
PKRU: 55555554
Call Trace:
<TASK>
intel_ddi_pre_enable.cold+0x96/0x5bf [i915]
intel_encoders_pre_enable+0x10e/0x140 [i915]
hsw_crtc_enable+0x207/0x99d [i915]
? ilk_crtc_enable.cold+0x2a/0x2a [i915]
? prepare_to_wait_exclusive+0x120/0x120
intel_enable_crtc+0x9a/0xf0 [i915]
skl_commit_modeset_enables+0x466/0x820 [i915]
? intel_commit_modeset_enables+0xd0/0xd0 [i915]
? intel_mbus_dbox_update+0x1ed/0x250 [i915]
intel_atomic_commit_tail+0xf2d/0x3040 [i915]
_raw_spin_lock_irqsave+0x87/0xe0
_raw_read_unlock_irqrestore+0x40/0x40
__update_load_avg_cfs_rq+0x70/0x5c0
__i915_sw_fence_complete+0x85/0x3b0 [i915]
? intel_get_crtc_new_encoder+0x190/0x190 [i915]
? sysvec_irq_work+0x13/0x90
? asm_sysvec_irq_work+0x12/0x20
? _raw_spin_lock_irq+0x82/0xd0
? read_word_at_a_time+0xe/0x20
? process_one_work+0x393/0x690
process_one_work+0x393/0x690
worker_thread+0x2b7/0x620
? process_one_work+0x690/0x690
kthread+0x15a/0x190
? kthread_complete_and_exit+0x20/0x20
ret_from_fork+0x1f/0x30
Continuing with the modeset without setting the DDI clock results in
more warnings and eventually a system hang. This does not seem to
happen with disconnected legacy or DP-alt DP ports because the clock
rate defaults to 162000 (which is a valid TBT clock) during the link
training process. Therefore, to fix this issue, this patch avoids
setting disconnected Type-C legacy ports to TBT-alt mode which prevents
the selection of TBT PLL when a modeset is forced.
v2: (Imre)
- Retain the check for legacy hotplug live status to account for
incorrect VBTs.
Cc: Imre Deak <imre.deak@intel.com>
Cc: José Roberto de Souza <jose.souza@intel.com>
Signed-off-by: Vivek Kasireddy <vivek.kasireddy@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220526001939.4031112-2-vivek.kasireddy@intel.com
Apparently when the VBT panel_type==0xff we should trawl through
the PNPID table and check for a match against the EDID. If a
match is found the index gives us the panel_type.
Tried to match the Windows behaviour here with first looking
for an exact match, and if one isn't found we fall back to
looking for a match w/o the mfg year/week.
v2: Rebase due to vlv_dsi changes
v3: Adjust to .get_panel_type() vfunc
Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/5545
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220510104242.6099-14-ville.syrjala@linux.intel.com
Move the panel specific VBT parsing to happen during the
output probing stage. Needs to be done because the VBT
parsing will need to look at the EDID to determine
the correct panel_type on some machines.
We split the parsed VBT data (i915->vbt) along the same
boundary. For the moment we just hoist all the panel
specific stuff into connector->panel.vbt since that seems
like the most convenient place for eg. the backlight code.
Note that we simply drop the drrs type check from
intel_drrs_frontbuffer_update() since that operates on the whole
device rather than a specific connector/encoder. But the check
was just a micro optimization so removing it doesn't actually
mattter for correctness.
TODO: Lot's of cleanup to be done in the future. Eg. most of
the DSI stuff could probably be eliminated entirely and just
parsed on demand during DSI init.
v2: Note the intel_drrs_frontbuffer_update() change
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220510104242.6099-13-ville.syrjala@linux.intel.com
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Parsing the panel specific data (anything that depends on panel_type)
from VBT is currently happening too early. Split the whole thing
into global vs. panel specific parts so that we can start doing
the panel specific parsing at a later time.
v2: Clarify that this is about panel_type (Jani)
Split out the leak checks (Jani)
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220510104242.6099-12-ville.syrjala@linux.intel.com
We use the "driver features" block for two different kinds
of data: global data, and per panel data. Split the function
into two parts along that line so that we can start doing the
parsing in two different locations.
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220510104242.6099-11-ville.syrjala@linux.intel.com
Disable the delayed VDD off work during the eDP probe.
If we never turn off the VDD then we can't violate the
panel's power sequencing delays despite not having read
them out yet from the VBT.
This is mostly a belt+suspenders type of thing since the
the timeout we'd use for the delayed work should be long
enough that this won't normally happen. But I don't really
like relying on timeouts for correctless so might as well
make sure.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220510104242.6099-10-ville.syrjala@linux.intel.com
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
During the eDP probe we may not yet know the panel_type used
to index the VBT panel tables. So the initial eDP probe will have
to be done without that, and thus we won't yet have the PPS delays
from the VBT. Once the VBT has been fully parse we should reinit
the PPS delays to make sure it's fully accounted for.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220510104242.6099-9-ville.syrjala@linux.intel.com
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
In order to do the panel VBT parsing after the EDID read
(needed to determine panel_type from PNPID) we need to stash
away the original BIOS programmed PPS delays so that we
can consult them again when we reinit the PPS delays after
the VBT parsing has been done.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220510104242.6099-7-ville.syrjala@linux.intel.com
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Skip QUIRK_INCREASE_T12_DELAY and the t11_t12 adjustment of the
VBT PPS delays if we've not yet initialized them. Will be important
later when the PPS delay init can happen before VBT parsing.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220510104242.6099-6-ville.syrjala@linux.intel.com
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
