summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2017-12-14drm/amd/display: Print DC_VER at DC initHarry Wentland
This has proven helpful on other OSes to give a quick state of the DC driver when a bug is reported. Signed-off-by: Harry Wentland <harry.wentland@amd.com> Reviewed-by: Tony Cheng <Tony.Cheng@amd.com> Acked-by: Harry Wentland <harry.wentland@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-12-14drm/amd/display: really fix time out in init sequenceTony Cheng
REG_UPDATE_2 return the reg value it write out through MMIO we need to do a REG_READ to confirm the value is written out Signed-off-by: Tony Cheng <tony.cheng@amd.com> Reviewed-by: Tony Cheng <Tony.Cheng@amd.com> Acked-by: Harry Wentland <harry.wentland@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-12-14drm/amd/display: Add disclaimer to BW and DML code provided by HWHarry Wentland
This code can sometimes look troubling but we trust it as it comes from HW teams with a guarantee of correctness. Add a note to these files to explain this. v2: thing -> things Signed-off-by: Harry Wentland <harry.wentland@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-12-14drm/amd/display: Use macro for isnan checkHarry Wentland
In code provided by HW teams we do a NaN check on floats by comparing the number against itself. This confuses most people including myself. Macro it out to make it self-explanatory. Don't do a NaN check for int. v2: parantheses around 'number' expression Signed-off-by: Harry Wentland <harry.wentland@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-12-14drm/amd/display: dal 3.1.24Tony Cheng
Signed-off-by: Tony Cheng <tony.cheng@amd.com> Reviewed-by: Tony Cheng <Tony.Cheng@amd.com> Acked-by: Harry Wentland <harry.wentland@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-12-14drm/amd/display: Set mpcc_disconnect_pending during MPC resetTony Cheng
This prevents an issue where the MPCC will not go to idle due to us not waiting for it to become idle during disable plane. Signed-off-by: Andrew Jiang <Andrew.Jiang@amd.com> Signed-off-by: Tony Cheng <tony.cheng@amd.com> Reviewed-by: Tony Cheng <Tony.Cheng@amd.com> Acked-by: Harry Wentland <harry.wentland@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-12-14drm/amd/display: Refine update flags usage in update_dchubp_dppAndrew Jiang
- Only update DPP clock if it's a full update. - Program viewport on position change. This caused SLS regressions. Signed-off-by: Andrew Jiang <Andrew.Jiang@amd.com> Reviewed-by: Tony Cheng <Tony.Cheng@amd.com> Acked-by: Harry Wentland <harry.wentland@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-12-14drm/amd/display: Cache cursor position.Yongqiang Sun
When programming cursor position after front end programmed, if position is already set previously, it doesn't make sense to program position with all 0. Cache position and use the parameter after front end programming. Signed-off-by: Yongqiang Sun <yongqiang.sun@amd.com> Reviewed-by: Tony Cheng <Tony.Cheng@amd.com> Acked-by: Harry Wentland <harry.wentland@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-12-14drm/amd/display: Program cursor regs after context swapped.Yongqiang Sun
Cursor is abnormal after pipe_ctx is changed in context. Cause: cursor attributes is programmed right after front end programming, but it use old pipe_ctx to program which is not updated yet. Solution: Program cursor regs after context swapped. Signed-off-by: Yongqiang Sun <yongqiang.sun@amd.com> Reviewed-by: Tony Cheng <Tony.Cheng@amd.com> Acked-by: Harry Wentland <harry.wentland@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-12-14drm/amd/display: fix recout_skip calculation when rotating 180 or 270Yongqiang Sun
Fixed fliped landscape and fliped portrait hard hang. Signed-off-by: Yongqiang Sun <yongqiang.sun@amd.com> Reviewed-by: Tony Cheng <Tony.Cheng@amd.com> Acked-by: Harry Wentland <harry.wentland@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-12-14drm/amd/display: Add is_tiling_rotated flag to plane_stateEric Murphy-Zaremba
Signed-off-by: Eric Murphy-Zaremba <Eric.Murphy-zaremba@amd.com> Reviewed-by: Aric Cyr <Aric.Cyr@amd.com> Acked-by: Harry Wentland <harry.wentland@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-12-14drm/amd/display: OPP DPG test patternEric Bernstein
Create opp_set_test_pattern function with similar interface and implementation as timing generator test pattern. Signed-off-by: Eric Bernstein <eric.bernstein@amd.com> Reviewed-by: Tony Cheng <Tony.Cheng@amd.com> Acked-by: Harry Wentland <harry.wentland@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-12-14drm/amd/display: Build unity lut for shaperVitaly Prosyak
Add color module to diagnostic compilation Signed-off-by: Vitaly Prosyak <vitaly.prosyak@amd.com> Reviewed-by: Charlene Liu <Charlene.Liu@amd.com> Acked-by: Harry Wentland <harry.wentland@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-12-14drm/amd/display: Reset MPCC muxes during initEric Bernstein
During HW initialization, instead of assuming or detecting the existing MPCC mux configuration and then removing existing planes, reset all the MPCC muxes. Signed-off-by: Eric Bernstein <eric.bernstein@amd.com> Reviewed-by: Tony Cheng <Tony.Cheng@amd.com> Acked-by: Harry Wentland <harry.wentland@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-12-14drm/amd/display: CNVC pseudocode review follow upYue Hin Lau
Signed-off-by: Yue Hin Lau <Yuehin.Lau@amd.com> Reviewed-by: Eric Bernstein <Eric.Bernstein@amd.com> Acked-by: Harry Wentland <harry.wentland@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-12-14drm/amd/display: Disable plane right after disconnectedYongqiang Sun
HDR display playing video underflow is observed when switching to full screen due to program a lower watermark right after unlock otg. Instead of disable plane in next flip coming, if there is a plane disconnected, after otg unlock wait for mpcc idle and disable the plane, then program watermark. So there is enough warter mark to make sure current frame data pass through. Signed-off-by: Yongqiang Sun <yongqiang.sun@amd.com> Reviewed-by: Tony Cheng <Tony.Cheng@amd.com> Acked-by: Harry Wentland <harry.wentland@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-12-14drm/amd/display: dal 3.1.23Tony Cheng
Signed-off-by: Tony Cheng <tony.cheng@amd.com> Reviewed-by: Tony Cheng <Tony.Cheng@amd.com> Acked-by: Harry Wentland <harry.wentland@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-12-14drm/amd/display: Clean up os_types.h a bitHarry Wentland
Signed-off-by: Harry Wentland <harry.wentland@amd.com> Reviewed-by: Tony Cheng <Tony.Cheng@amd.com> Acked-by: Harry Wentland <harry.wentland@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-12-14drm/amd/display: Fix access of wrong array element TF format conversionHarry Wentland
Found by smatch: drivers/gpu/drm/amd/amdgpu/../display/dc/dce110/dce110_hw_sequencer.c:357 convert_to_custom_float() error: buffer overflow 'arr_points' 2 <= 2 drivers/gpu/drm/amd/amdgpu/../display/dc/dce110/dce110_hw_sequencer.c:358 convert_to_custom_float() warn: buffer overflow 'arr_points' 2 <= 2 Regression: drm/amd/display: Remove extra arr_points element Signed-off-by: Harry Wentland <harry.wentland@amd.com> Reviewed-by: Tony Cheng <Tony.Cheng@amd.com> Acked-by: Harry Wentland <harry.wentland@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-12-14drm/amd/display: dal 3.1.22Tony Cheng
Signed-off-by: Tony Cheng <tony.cheng@amd.com> Reviewed-by: Tony Cheng <Tony.Cheng@amd.com> Acked-by: Harry Wentland <harry.wentland@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-12-14drm/amd/display: Add dcc_change surface update flagAndrew Jiang
Program the DCC registers when dcc_change is true. Signed-off-by: Andrew Jiang <Andrew.Jiang@amd.com> Reviewed-by: Tony Cheng <Tony.Cheng@amd.com> Acked-by: Harry Wentland <harry.wentland@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-12-14drm/amd/display: Implement work around for optc underflow.Yongqiang Sun
Work around for a hw bug causing optc underflow if blank data double buffer disable and remove mpcc. Checking optc status after otg unlock, after wait mpcc idle check status again, if optc underflow just happens after wait mpcc idle, clear underflow status and enable blank data double buffer. Signed-off-by: Yongqiang Sun <yongqiang.sun@amd.com> Reviewed-by: Tony Cheng <Tony.Cheng@amd.com> Acked-by: Harry Wentland <harry.wentland@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-12-14drm/amd/display: Change optimized_required logicAndrew Jiang
Rather than setting it every time there's a full update with surface count > 0, set it when we need to do plane_atomic_disconnect. Also make sure that we unset the flag in plane_atomic_disable, so that in the event we run through a sequence where we do disconnect followed by an immediate disable, we do not do unnecessarily request a passive flip to do the optimization. Signed-off-by: Andrew Jiang <Andrew.Jiang@amd.com> Reviewed-by: Tony Cheng <Tony.Cheng@amd.com> Acked-by: Harry Wentland <harry.wentland@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-12-14drm/amd/display: Add optimized_required flagHarry Wentland
Signed-off-by: Harry Wentland <harry.wentland@amd.com> Reviewed-by: Leo (Sunpeng) Li <sunpeng.li@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-12-14drm/amd/display: Use same wait mpcc idle function.Yongqiang Sun
There is already wait mpcc idle function. It is better to use the same function for all wait mpcc idle. Signed-off-by: Yongqiang Sun <yongqiang.sun@amd.com> Reviewed-by: Tony Cheng <Tony.Cheng@amd.com> Acked-by: Harry Wentland <harry.wentland@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-12-14drm/amd/display: MPC updatesEric Bernstein
Fix update_mpcc logic to only call assert_mpcc_idle_before_connect if mpcc is not already being used (and required removal). Update set_out_rate_control to include optional flow control parameter. In init_mpcc_from_hw check for case where bot_sel is same as mpcc_id. Signed-off-by: Eric Bernstein <eric.bernstein@amd.com> Reviewed-by: Tony Cheng <Tony.Cheng@amd.com> Acked-by: Harry Wentland <harry.wentland@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-12-14drm/amd/display: Fixed read wrong reg to get bot_sel.Yongqiang Sun
Signed-off-by: Yongqiang Sun <yongqiang.sun@amd.com> Reviewed-by: Tony Cheng <Tony.Cheng@amd.com> Acked-by: Harry Wentland <harry.wentland@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-12-14drm/amd/display: Fix potential mem leak in DC constructHarry Wentland
Found by smatch: drivers/gpu/drm/amd/amdgpu/../display/dc/core/dc.c:506 construct() warn: possible memory leak of 'dc_ctx' drivers/gpu/drm/amd/amdgpu/../display/dc/core/dc.c:506 construct() warn: possible memory leak of 'dc_vbios' drivers/gpu/drm/amd/amdgpu/../display/dc/core/dc.c:506 construct() warn: possible memory leak of 'dcn_ip' drivers/gpu/drm/amd/amdgpu/../display/dc/core/dc.c:506 construct() warn: possible memory leak of 'dcn_soc' Signed-off-by: Harry Wentland <harry.wentland@amd.com> Reviewed-by: Tony Cheng <Tony.Cheng@amd.com> Acked-by: Harry Wentland <harry.wentland@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-12-14drm/amd/display: Remove redundant checks in set_default_colorsHarry Wentland
pipe_ctx->stream and pipe_ctx->plane_state are never NULL Found by smatch: drivers/gpu/drm/amd/amdgpu/../display/dc/dce110/dce110_hw_sequencer.c:2111 set_default_colors() error: we previously assumed 'pipe_ctx->stream' could be null (see line 2101) Signed-off-by: Harry Wentland <harry.wentland@amd.com> Reviewed-by: Tony Cheng <Tony.Cheng@amd.com> Acked-by: Harry Wentland <harry.wentland@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-12-14drm/amd/display: Remove plane_res.mi check in dce110_apply_ctx_for_surfaceHarry Wentland
plane_res.mi (memory interface) can never be NULL for DCE110 Found by smatch: drivers/gpu/drm/amd/amdgpu/../display/dc/dce110/dce110_hw_sequencer.c:2881 dce110_apply_ctx_for_surface() error: we previously assumed 'pipe_ctx->plane_res.mi' could be null (see line 2873) Signed-off-by: Harry Wentland <harry.wentland@amd.com> Reviewed-by: Tony Cheng <Tony.Cheng@amd.com> Acked-by: Harry Wentland <harry.wentland@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-12-14drm/amd/display: Remove PSR functions in LinuxHarry Wentland
NULL check issue found by smatch: drivers/gpu/drm/amd/amdgpu/../display/dc/core/dc_link.c:1976 dc_link_setup_psr() warn: variable dereferenced before check 'link' (see line 1970) We don't use these functions so might as well remove them. Signed-off-by: Harry Wentland <harry.wentland@amd.com> Reviewed-by: Tony Cheng <Tony.Cheng@amd.com> Acked-by: Harry Wentland <harry.wentland@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-12-14drm/amd/display: Integrating MPC pseudocodeEric Bernstein
Integrating MPC pseudocode to support new blending cases with secondary MPCC list. This includes a design change to MPC data structures and interfaces. Signed-off-by: Eric Bernstein <eric.bernstein@amd.com> Reviewed-by: Tony Cheng <Tony.Cheng@amd.com> Acked-by: Harry Wentland <harry.wentland@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-12-14drm/amd/display: dal 3.1.21Tony Cheng
Signed-off-by: Tony Cheng <tony.cheng@amd.com> Reviewed-by: Tony Cheng <Tony.Cheng@amd.com> Acked-by: Harry Wentland <harry.wentland@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-12-14drm/amd/display: try to find matching audio inst for enc inst firstCharlene Liu
[Description] in eDP+ HDMI/DP clone or extended configuration, audio inst changed from inst 1 to inst0. No failure related this though, just playback device endpoint inst changed. Also remove one addition register read. Signed-off-by: Charlene Liu <charlene.liu@amd.com> Reviewed-by: Tony Cheng <Tony.Cheng@amd.com> Acked-by: Harry Wentland <harry.wentland@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-12-14drm/amd/display: fix seq issue: turn on clock before programming afmt.Charlene Liu
Signed-off-by: Charlene Liu <charlene.liu@amd.com> Reviewed-by: Krunoslav Kovac <Krunoslav.Kovac@amd.com> Acked-by: Harry Wentland <harry.wentland@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-12-14drm/amd/display: Remove unnecessary wait mpcc idle.Yongqiang Sun
Before power gate plane, mpcc idle wait is processed, no need to wait another time. Signed-off-by: Yongqiang Sun <yongqiang.sun@amd.com> Reviewed-by: Tony Cheng <Tony.Cheng@amd.com> Acked-by: Harry Wentland <harry.wentland@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-12-14drm/amd/pp: reset dpm level when adjust power stateRex Zhu
Acked-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Rex Zhu <Rex.Zhu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-12-14drm/ttm: init locked again to prevent incorrect unlockRoger He
Signed-off-by: Roger He <Hongbo.He@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2017-12-14arm64/sve: Report SVE to userspace via CPUID only if supportedDave Martin
Currently, the SVE field in ID_AA64PFR0_EL1 is visible unconditionally to userspace via the CPU ID register emulation, irrespective of the kernel config. This means that if a kernel configured with CONFIG_ARM64_SVE=n is run on SVE-capable hardware, userspace will see SVE reported as present in the ID regs even though the kernel forbids execution of SVE instructions. This patch makes the exposure of the SVE field in ID_AA64PFR0_EL1 conditional on CONFIG_ARM64_SVE=y. Since future architecture features are likely to encounter a similar requirement, this patch adds a suitable helper macros for use when declaring config-conditional ID register fields. Fixes: 43994d824e84 ("arm64/sve: Detect SVE and activate runtime support") Reviewed-by: Suzuki K Poulose <suzuki.poulose@arm.com> Reported-by: Mark Rutland <mark.rutland@arm.com> Signed-off-by: Dave Martin <Dave.Martin@arm.com> Cc: Suzuki Poulose <suzuki.poulose@arm.com> Signed-off-by: Will Deacon <will.deacon@arm.com>
2017-12-14drm/i915: Show engine state when hangcheck detects a stallChris Wilson
Knowing the state of the engine when hangcheck thinks it is stalling is useful for both debugging hangcheck itself and the potential cause of an unwanted stall. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com> Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20171214122613.26134-1-chris@chris-wilson.co.uk
2017-12-14drm/i915: make CS frequency read support missing more obviousLionel Landwerlin
As suggested by Chris, we should make this more obvious for people working with newer generations. Suggested-by: Chris Wilson <chris@chris-wilson.co.uk> Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com> Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Link: https://patchwork.freedesktop.org/patch/msgid/20171213171154.6201-1-lionel.g.landwerlin@intel.com
2017-12-14drm/atomic-helper: Make zpos property kerneldoc less misleadingThierry Reding
kerneldoc for drm_plane_create_zpos_property() says that the DRM core will automatically calculate the normalized zpos values, but it doesn't actually do that anymore since commit 38d868e41c4b ("drm: Don't force all planes to be added to the state due to zpos"). Instead, drivers are supposed to explicitly call the drm_atomic_normalize_zpos() function. Change the kerneldoc comment to reflect that. v2: reference the commit that introduced the kerneldoc unclarity Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch> Signed-off-by: Thierry Reding <treding@nvidia.com> Link: https://patchwork.freedesktop.org/patch/msgid/20171120150959.26913-1-thierry.reding@gmail.com
2017-12-14arm64: fix CONFIG_DEBUG_WX address reportingMark Rutland
In ptdump_check_wx(), we pass walk_pgd() a start address of 0 (rather than VA_START) for the init_mm. This means that any reported W&X addresses are offset by VA_START, which is clearly wrong and can make them appear like userspace addresses. Fix this by telling the ptdump code that we're walking init_mm starting at VA_START. We don't need to update the addr_markers, since these are still valid bounds regardless. Cc: <stable@vger.kernel.org> Fixes: 1404d6f13e47 ("arm64: dump: Add checking for writable and exectuable pages") Signed-off-by: Mark Rutland <mark.rutland@arm.com> Cc: Kees Cook <keescook@chromium.org> Cc: Laura Abbott <labbott@redhat.com> Reported-by: Timur Tabi <timur@codeaurora.org> Signed-off-by: Will Deacon <will.deacon@arm.com>
2017-12-14ovl: fix overlay: warning prefixAmir Goldstein
Conform two stray warning messages to the standard overlayfs: prefix. Signed-off-by: Amir Goldstein <amir73il@gmail.com> Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2017-12-14drm/i915/guc: Extract doorbell verification into a functionMichał Winiarski
We have the selftest that's checking doorbell create/destroy, so there's no need to check all doorbells delaying the reset every time. We do want to have that extra sanity check at module load/unload though. Signed-off-by: Michał Winiarski <michal.winiarski@intel.com> Cc: Chris Wilson <chris@chris-wilson.co.uk> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Cc: Michal Wajdeczko <michal.wajdeczko@intel.com> Reviewed-by: Michal Wajdeczko <michal.wajdeczko@intel.com> Reviewed-by: Michel Thierry <michel.thierry@intel.com> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Link: https://patchwork.freedesktop.org/patch/msgid/20171213221352.7173-7-michal.winiarski@intel.com
2017-12-14drm/i915/guc: Extract clients allocation to submission_initMichał Winiarski
We can now move the clients allocation to submission_init path, rather than keeping the condition inside submission_enable called on every reset. Signed-off-by: Michał Winiarski <michal.winiarski@intel.com> Cc: Chris Wilson <chris@chris-wilson.co.uk> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Cc: Michal Wajdeczko <michal.wajdeczko@intel.com> Reviewed-by: Michel Thierry <michel.thierry@intel.com> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Link: https://patchwork.freedesktop.org/patch/msgid/20171213221352.7173-6-michal.winiarski@intel.com
2017-12-14drm/i915/guc: Extract doorbell creation from client allocationMichał Winiarski
Full GPU reset causes GuC to be reset. This means that every time we're doing a reset, we need to talk to GuC and tell it about doorbells. Let's separate the communication part (create_doorbell) from our internal bookkeeping (reserve_doorbell) so that we can cleanly separate the initialization done at module load from reinitialization done at reset in the following patch. While I'm here, let's also add a proper (although slightly asymetric) cleanup that doesn't try to communicate with GuC after it's already gone, getting rid of "expected" warnings caused by GuC action failures on module unload. Note that I've also removed one of the tests (bitmap out of sync), since it doesn't make much sense anymore - bitmaps are now not expected to change during the lifetime of a client. Signed-off-by: Michał Winiarski <michal.winiarski@intel.com> Cc: Chris Wilson <chris@chris-wilson.co.uk> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Cc: Michal Wajdeczko <michal.wajdeczko@intel.com> Cc: Michel Thierry <michel.thierry@intel.com> Reviewed-by: Michel Thierry <michel.thierry@intel.com> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Link: https://patchwork.freedesktop.org/patch/msgid/20171213221352.7173-5-michal.winiarski@intel.com
2017-12-14drm/i915/guc: Call invalidate after changing the vfuncMichał Winiarski
To make this operation a bit cleaner, we should make sure that the HW can catch up by calling the new implementation right away. Note that currently we're only touching the vfunc at module load time (before GuC is even loaded), so this shouldn't cause any functional changes. Suggested-by: Chris Wilson <chris@chris-wilson.co.uk> Signed-off-by: Michał Winiarski <michal.winiarski@intel.com> Cc: Chris Wilson <chris@chris-wilson.co.uk> Cc: Michal Wajdeczko <michal.wajdeczko@intel.com> Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Link: https://patchwork.freedesktop.org/patch/msgid/20171213221352.7173-4-michal.winiarski@intel.com
2017-12-14drm/i915/guc: Extract guc_init from guc_init_hwMichał Winiarski
After GPU reset, GuC HW needs to be reinitialized (with FW reload). Unfortunately, we're doing some extra work there (mostly allocating stuff), work that can be moved to guc_init and called once at driver load time. As a side effect we're no longer hitting an assert in i915_ggtt_enable_guc on suspend/resume. v2: Do not duplicate disable_communication / reset_guc_interrupts v3: Add proper teardown after rebase References: 04f7b24eccdf ("drm/i915/guc: Assert that we switch between known ggtt->invalidate functions") Signed-off-by: Michał Winiarski <michal.winiarski@intel.com> Cc: Chris Wilson <chris@chris-wilson.co.uk> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Cc: Michal Wajdeczko <michal.wajdeczko@intel.com> Cc: Sagar Arun Kamble <sagar.a.kamble@intel.com> Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Link: https://patchwork.freedesktop.org/patch/msgid/20171213221352.7173-3-michal.winiarski@intel.com
2017-12-14drm/i915/guc: Move GuC workqueue allocations outside of the mutexMichał Winiarski
This gets rid of the following lockdep splat: ====================================================== WARNING: possible circular locking dependency detected 4.15.0-rc2-CI-Patchwork_7428+ #1 Not tainted ------------------------------------------------------ debugfs_test/1351 is trying to acquire lock: (&dev->struct_mutex){+.+.}, at: [<000000009d90d1a3>] i915_mutex_lock_interruptible+0x47/0x130 [i915] but task is already holding lock: (&mm->mmap_sem){++++}, at: [<000000005df01c1e>] __do_page_fault+0x106/0x560 which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #6 (&mm->mmap_sem){++++}: __might_fault+0x63/0x90 _copy_to_user+0x1e/0x70 filldir+0x8c/0xf0 dcache_readdir+0xeb/0x160 iterate_dir+0xe6/0x150 SyS_getdents+0xa0/0x130 entry_SYSCALL_64_fastpath+0x1c/0x89 -> #5 (&sb->s_type->i_mutex_key#5){++++}: lockref_get+0x9/0x20 -> #4 ((completion)&req.done){+.+.}: wait_for_common+0x54/0x210 devtmpfs_create_node+0x130/0x150 device_add+0x5ad/0x5e0 device_create_groups_vargs+0xd4/0xe0 device_create+0x35/0x40 msr_device_create+0x22/0x40 cpuhp_invoke_callback+0xc5/0xbf0 cpuhp_thread_fun+0x167/0x210 smpboot_thread_fn+0x17f/0x270 kthread+0x173/0x1b0 ret_from_fork+0x24/0x30 -> #3 (cpuhp_state-up){+.+.}: cpuhp_issue_call+0x132/0x1c0 __cpuhp_setup_state_cpuslocked+0x12f/0x2a0 __cpuhp_setup_state+0x3a/0x50 page_writeback_init+0x3a/0x5c start_kernel+0x393/0x3e2 secondary_startup_64+0xa5/0xb0 -> #2 (cpuhp_state_mutex){+.+.}: __mutex_lock+0x81/0x9b0 __cpuhp_setup_state_cpuslocked+0x4b/0x2a0 __cpuhp_setup_state+0x3a/0x50 page_alloc_init+0x1f/0x26 start_kernel+0x139/0x3e2 secondary_startup_64+0xa5/0xb0 -> #1 (cpu_hotplug_lock.rw_sem){++++}: cpus_read_lock+0x34/0xa0 apply_workqueue_attrs+0xd/0x40 __alloc_workqueue_key+0x2c7/0x4e1 intel_guc_submission_init+0x10c/0x650 [i915] intel_uc_init_hw+0x29e/0x460 [i915] i915_gem_init_hw+0xca/0x290 [i915] i915_gem_init+0x115/0x3a0 [i915] i915_driver_load+0x9a8/0x16c0 [i915] i915_pci_probe+0x2e/0x90 [i915] pci_device_probe+0x9c/0x120 driver_probe_device+0x2a3/0x480 __driver_attach+0xd9/0xe0 bus_for_each_dev+0x57/0x90 bus_add_driver+0x168/0x260 driver_register+0x52/0xc0 do_one_initcall+0x39/0x150 do_init_module+0x56/0x1ef load_module+0x231c/0x2d70 SyS_finit_module+0xa5/0xe0 entry_SYSCALL_64_fastpath+0x1c/0x89 -> #0 (&dev->struct_mutex){+.+.}: lock_acquire+0xaf/0x200 __mutex_lock+0x81/0x9b0 i915_mutex_lock_interruptible+0x47/0x130 [i915] i915_gem_fault+0x201/0x760 [i915] __do_fault+0x15/0x70 __handle_mm_fault+0x85b/0xe40 handle_mm_fault+0x14f/0x2f0 __do_page_fault+0x2d1/0x560 page_fault+0x22/0x30 other info that might help us debug this: Chain exists of: &dev->struct_mutex --> &sb->s_type->i_mutex_key#5 --> &mm->mmap_sem Possible unsafe locking scenario: CPU0 CPU1 ---- ---- lock(&mm->mmap_sem); lock(&sb->s_type->i_mutex_key#5); lock(&mm->mmap_sem); lock(&dev->struct_mutex); *** DEADLOCK *** 1 lock held by debugfs_test/1351: #0: (&mm->mmap_sem){++++}, at: [<000000005df01c1e>] __do_page_fault+0x106/0x560 stack backtrace: CPU: 2 PID: 1351 Comm: debugfs_test Not tainted 4.15.0-rc2-CI-Patchwork_7428+ #1 Hardware name: /NUC6i5SYB, BIOS SYSKLi35.86A.0057.2017.0119.1758 01/19/2017 Call Trace: dump_stack+0x5f/0x86 print_circular_bug+0x230/0x3b0 check_prev_add+0x439/0x7b0 ? lockdep_init_map_crosslock+0x20/0x20 ? unwind_get_return_address+0x16/0x30 ? __lock_acquire+0x1385/0x15a0 __lock_acquire+0x1385/0x15a0 lock_acquire+0xaf/0x200 ? i915_mutex_lock_interruptible+0x47/0x130 [i915] __mutex_lock+0x81/0x9b0 ? i915_mutex_lock_interruptible+0x47/0x130 [i915] ? i915_mutex_lock_interruptible+0x47/0x130 [i915] ? i915_mutex_lock_interruptible+0x47/0x130 [i915] i915_mutex_lock_interruptible+0x47/0x130 [i915] ? __pm_runtime_resume+0x4f/0x80 i915_gem_fault+0x201/0x760 [i915] __do_fault+0x15/0x70 __handle_mm_fault+0x85b/0xe40 handle_mm_fault+0x14f/0x2f0 __do_page_fault+0x2d1/0x560 page_fault+0x22/0x30 RIP: 0033:0x7f98d6f49116 RSP: 002b:00007ffd6ffc3278 EFLAGS: 00010283 RAX: 00007f98d39a2bc0 RBX: 0000000000000000 RCX: 0000000000001680 RDX: 0000000000001680 RSI: 00007ffd6ffc3400 RDI: 00007f98d39a2bc0 RBP: 00007ffd6ffc33a0 R08: 0000000000000000 R09: 00000000000005a0 R10: 000055e847c2a830 R11: 0000000000000002 R12: 0000000000000001 R13: 000055e847c1d040 R14: 00007ffd6ffc3400 R15: 00007f98d6752ba0 v2: Init preempt_work unconditionally (Chris) v3: Mention that we need the enable_guc=1 for lockdep splat (Chris) Testcase: igt/debugfs_test/read_all_entries # with i915.enable_guc=1 Signed-off-by: Michał Winiarski <michal.winiarski@intel.com> Cc: Chris Wilson <chris@chris-wilson.co.uk> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Cc: Michal Wajdeczko <michal.wajdeczko@intel.com> Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Link: https://patchwork.freedesktop.org/patch/msgid/20171213221352.7173-2-michal.winiarski@intel.com
generated by cgit (git 2.34.1) at 2025-09-28 15:37:52 +0000