- Oct 21, 2020
-
-
Ville Syrjälä authored
Restore RPS for ILK-M. We lost it when an extra HAS_RPS() check appeared in intel_rps_enable(). Unfortunaltey this just makes the performance worse on my ILK because intel_ips insists on limiting the GPU freq to the minimum. If we don't do the RPS init then intel_ips will not limit the frequency for whatever reason. Either it can't get at some required information and thus makes wrong decisions, or we mess up some weights/etc. and cause it to make the wrong decisions when RPS init has been done, or the entire thing is just wrong. Would require a bunch of reverse engineering to figure out what's going on. Cc: stable@vger.kernel.org Cc: Chris Wilson <chris@chris-wilson.co.uk> Fixes: 9c878557 ("drm/i915/gt: Use the RPM config register to determine clk frequencies") Signed-off-by:
Ville Syrjälä <ville.syrjala@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20201021131443.25616-1-ville.syrjala@linux.intel.com Reviewed-by:
Chris Wilson <chris@chris-wilson.co.uk>
-
- Oct 15, 2020
-
-
Michel Thierry authored
While we do lack the faster shared LLC, we should still have support for snooping over PCIe. Signed-off-by:
Michel Thierry <michel.thierry@intel.com> Signed-off-by:
Matthew Auld <matthew.auld@intel.com> Reviewed-by:
Lucas De Marchi <lucas.demarchi@intel.com> Signed-off-by:
-
Stuart Summers authored
DG1 shares some workarounds with TGL and RKL and also has some additional workarounds of its own. v2: Correct location of Wa_1408615072 (JohnH). v3: Apply WAs 1606700617, 18011464164 and 22010931296 to DG1 (José) v4 (Anusha) - Add Wa_22010271021 - s/Wa_14010096844/Wa_1409836686 v5: - Extend Wa_14010919138 to all revs (Matt Atwood) - Power gate media is global gen12 design. (Rodrigo) - Rebase (Lucas) v6: use REG_BIT() to fix checkpatch warning (Lucas) BSpec: 53508 Cc: Matt Atwood <matthew.s.atwood@intel.com> Cc: Matt Roper <matthew.d.roper@intel.com> Cc: Radhakrishna Sripada <radhakrishna.sripada@intel.com> Cc: José Roberto de Souza <jose.souza@intel.com> Signed-off-by:
Stuart Summers <stuart.summers@intel.com> Signed-off-by:
Anusha Srivatsa <anusha.srivatsa@intel.com> Signed-off-by:
Rodrigo Vivi <rodrigo.vivi@intel.com> Reviewed-by:
-
- Oct 14, 2020
-
-
Tejas Upadhyay authored
Recently we came across requirement to identify EHL and JSL platform to program them differently. Thus Split the basic platform definition, macros, and PCI IDs to differentiate between EHL and JSL platforms. Also, IS_ELKHARTLAKE is replaced with IS_JSL_EHL everywhere. Changes since V1 : - Rebased to avoid merge conflicts - Added missed check for jasperlake in intel_uc_fw.c Cc : Matt Roper <matthew.d.roper@intel.com> Cc : Ville Syrjälä <ville.syrjala@linux.intel.com> Signed-off-by:
Tejas Upadhyay <tejaskumarx.surendrakumar.upadhyay@intel.com> Reviewed-by:
Matt Roper <matthew.d.roper@intel.com> Signed-off-by:
Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20201013192948.63470-1-tejaskumarx.surendrakumar.upadhyay@intel.com
-
- Oct 09, 2020
-
-
Ville Syrjälä authored
Implement the pci .shutdown() hook in order to quiesce the hardware prior to reboot. The main purpose here is to turn all displays off. Some displays/other drivers tend to get confused if the state after reboot isn't exactly as they expected. One specific example was the Dell UP2414Q in MST mode. It would require me to pull the power cord after a reboot or else it would just not come back to life. Sadly I don't have that at hand anymore so not sure if it's still misbehaving without the graceful shutdown, or if we managed to fix something else since I last tested it. For good measure we do a gem suspend as well, so that we match the suspend flow more closely. Also stopping all DMA and whatnot is probably a good idea for kexec. I would expect that some kind of GT reset happens on normal reboot so probably not totally necessary there. v2: Use the pci .shutdown() hook instead of a reboot notifier (Lukas) Do the gem suspend for kexec (Chris) Cc: Lukas Wunner <lukas@wunner.de> Cc: Chris Wilson <chris@chris-wilson.co.uk> Signed-off-by:Jani Nikula <jani.nikula@intel.com> Reviewed-by:
-
- Aug 17, 2020
-
-
Matt Roper authored
If HTI (also sometimes called HDPORT) is enabled at startup, it may be using some of the PHYs and DPLLs making them unavailable for general usage. Let's read out the HDPORT_STATE register and avoid making use of resources that HTI is already using. v2: - Fix minor checkpatch warnings v3: - Just readout HDPORT_STATE register once during init and then parse it later as needed. - Add a 'has_hti' device info flag to track whether we should readout HDPORT_STATE or not. We can skip the platform/flag tests later since the hti_state in dev_priv will remain 0 for platforms it does not apply to. - Move PLL masking into icl_get_combo_phy_dpll() since at the moment RKL is the only platform that has HTI. (Jose) Bspec: 49189 Bspec: 53707 Cc: Lucas De Marchi <lucas.demarchi@intel.com> Cc: José Roberto de Souza <jose.souza@intel.com> Signed-off-by:
José Roberto de Souza <jose.souza@intel.com> Signed-off-by:
-
- Jul 14, 2020
-
-
Abdiel Janulgue authored
Bspec: 33617, 33617 v2: s/intel_dg1_info/dg1_info/ as done for other platforms before and try to shut up compiler about ununsed variable that we know shouldn't be used (Lucas) v3: replace explicit attribute with __maybe_unused (Lucas) Cc: José Roberto de Souza <jose.souza@intel.com> Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com> Cc: Stuart Summers <stuart.summers@intel.com> Cc: Vanshidhar Konda <vanshidhar.r.konda@intel.com> Cc: Lucas De Marchi <lucas.demarchi@intel.com> Cc: Aravind Iddamsetty <aravind.iddamsetty@intel.com> Cc: Matt Roper <matthew.d.roper@intel.com> Signed-off-by:Abdiel Janulgue <abdiel.janulgue@linux.intel.com> Signed-off-by:
-
- Jul 08, 2020
-
-
Daniele Ceraolo Spurio authored
Since the engines belong to the GT, move the runtime-updated list of available engines to the intel_gt struct. The original mask has been renamed to indicate it contains the maximum engine list that can be found on a matching device. In preparation for other info being moved to the gt in follow up patches (sseu), introduce an intel_gt_info structure to group all gt-related runtime info. v2: s/max_engine_mask/platform_engine_mask (tvrtko), fix selftest Signed-off-by:
Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com> Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com> Cc: Andi Shyti <andi.shyti@intel.com> Cc: Venkata Sandeep Dhanalakota <venkata.s.dhanalakota@intel.com> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> #v1 Signed-off-by:
-
- Jul 03, 2020
-
-
Ville Syrjälä authored
Unlike all the other pre-snb desktop platforms i865 actually supports FBC. Let's enable it. Quote from the spec: "DevSDG provides the same Run-Length Encoded Frame Buffer Compression (RLEFBC) function as exists in DevMGM." As i865 only has the one pipe we want to skip massaging the plane<->pipe assignment aimed at getting FBC+LVDS working on the mobile platforms. Signed-off-by:
-
- Jun 09, 2020
-
-
Matt Roper authored
Rocket Lake uses the same 'abox0' mechanism to handle pixel data transfers from memory that gen11 platforms used, rather than the abox1/abox2 interfaces used by TGL/DG1. For the most part this is a hardware implementation detail that's transparent to driver software, but we do have to program a couple of tuning registers (MBUS_ABOX_CTL and BW_BUDDY registers) according to which ABOX instances are used by a platform. Let's track the platform's ABOX usage in the device info structure and use that to determine which instances of these registers to program. As an exception to this rule is that even though TGL/DG1 use ABOX1+ABOX2 for data transfers, we're still directed to program the ABOX_CTL register for ABOX0; so we'll handle that as a special case. v2: - Store the mask of platform-specific abox registers in the device info structure. - Add a TLB_REQ_TIMER() helper macro. (Aditya) v3: - Squash ABOX and BW_BUDDY patches together and use a single mask for both of them, plus a special-case for programming the ABOX0 instance on all gen12. (Ville) Bspec: 50096 Bspec: 49218 Cc: Ville Syrjälä <ville.syrjala@linux.intel.com> Cc: Aditya Swarup <aditya.swarup@intel.com> Signed-off-by:
-
- Jun 04, 2020
-
-
José Roberto de Souza authored
RKL doesn't have PSR2 HW tracking, it was replaced by software/manual tracking. The driver is required to track the areas that needs update and program hardware to send selective updates. So until the software tracking is implemented, PSR2 needs to be disabled for platforms without PSR2 HW tracking. BSpec: 50422 BSpec: 50424 Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Signed-off-by:
-
Matt Roper authored
Although we properly captured RKL's three pipes in the device info structure, we forgot to make the corresponding update to the transcoder mask. Set this field so that our transcoder loops will operate properly. Fixes: 123f62de ("drm/i915/rkl: Add RKL platform info and PCI ids") Signed-off-by:
-
- Jun 02, 2020
-
-
Chris Wilson authored
Cometlake is a small refresh of Coffeelake, but since we have found out a difference in the plaforms, we need to identify them as separate platforms. Since we previously took Coffeelake/Cometlake as identical, update all IS_COFFEELAKE() to also include IS_COMETLAKE(). Signed-off-by:
-
- May 20, 2020
-
-
Matt Roper authored
Introduce the basic platform definition, macros, and PCI IDs. Bspec: 44501 Cc: Lucas De Marchi <lucas.demarchi@intel.com> Cc: Caz Yokoyama <caz.yokoyama@intel.com> Cc: Aditya Swarup <aditya.swarup@intel.com> Signed-off-by:
Caz Yokoyama <caz.yokoyama@intel.com> Reviewed-by:
-
- May 11, 2020
-
-
Chris Wilson authored
This reverts commit 0b718ba1. There are still some residual issues with asynchronous binding and execution, but since commit 92581f9f ("drm/i915: Immediately execute the fenced work") we prefer not to use asynchronous binds, and the remaining issues do not seem restricted to Cherryview [at least the ones seen over a few dozen CI runs, less frequent issues are sure to be discovered!] These issues seem to be mitigated, if not eliminated entirely, by the previous commit 84eac0c6 ("drm/i915/gt: Force pte cacheline to main memory"). Signed-off-by:
Mika Kuoppala <mika.kuoppala@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20200510102431.21959-3-chris@chris-wilson.co.uk
-
- Apr 28, 2020
-
-
Daniel Vetter authored
Luckily we're already well set up in the main driver, with drm_dev_put() being the last thing in both the unload error case and the pci remove function. Acked-by:
Jani Nikula <jani.nikula@linux.intel.com> Signed-off-by:
Daniel Vetter <daniel.vetter@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20200415074034.175360-39-daniel.vetter@ffwll.ch
-
- Apr 18, 2020
-
-
Michael J. Ruhl authored
DMA_MASK bit values are different for different generations. This will become more difficult to manage over time with the open coded usage of different versions of the device. Fix by: disallow setting of dma mask in AGP path (< GEN(5) for i915, add dma_mask_size to the device info configuration, updating open code call sequence to the latest interface, refactoring into a common function for setting the dma segment and mask info Reviewed-by:
Michael J. Ruhl <michael.j.ruhl@intel.com> cc: Brian Welty <brian.welty@intel.com> cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com> Signed-off-by:
-
- Apr 03, 2020
-
-
Ville Syrjälä authored
We have a bunch of code that would like to know which CPU transcoders are actually present in the hardware. Rather than use various ad-hoc methods let's just include a full bitmask in the device info, alongside pipe_mask. v2: Rebase Signed-off-by:
-
- Mar 13, 2020
-
-
José Roberto de Souza authored
We have a few TGL machines in our CI and it is mostly green with failures in tests that will not impact future Linux installations. Also there is no warnings, errors, flickering or any visual defects while doing ordinary tasks like browsing and editing documents in a dual monitor setup. As a reminder i915.require_force_probe was created to protect future Linux installation's iso images that might contain a kernel from the enabling time of the new platform. Without this protection most of linux installation was recommending nomodeset option during installation that was getting stick there after installation. Reference: https://intel-gfx-ci.01.org/tree/drm-tip/fi-tgl-u.html Reference: https://intel-gfx-ci.01.org/tree/drm-tip/shard-tglb.html Cc: James Ausmus <james.ausmus@intel.com> Cc: Jani Saarinen <jani.saarinen@intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Jani Nikula <jani.nikula@intel.com> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Signed-off-by:
-
- Feb 27, 2020
-
-
Jani Nikula authored
The #include has been splattered all over the place, but there are precious few places, all .c files, that actually need it. v2: remove leftover double newlines Reviewed-by:
-
- Feb 26, 2020
-
-
Chris Wilson authored
Full-ppgtt on gen7 is proving to be highly unstable and not robust. Closes: https://gitlab.freedesktop.org/drm/intel/issues/694 Fixes: 3cd6e886 ("drm/i915/gen7: Re-enable full-ppgtt for ivb & hsw") Signed-off-by:
-
- Feb 24, 2020
-
-
Chris Wilson authored
Full-ppgtt on gen7 is proving to be highly unstable and not robust. Closes: https://gitlab.freedesktop.org/drm/intel/issues/694 Fixes: 3cd6e886 ("drm/i915/gen7: Re-enable full-ppgtt for ivb & hsw") Signed-off-by:
-
- Feb 07, 2020
-
-
Chris Wilson authored
Do not try and dereference the i915 private before it has been allocated and attached to the drvdata! Fixes: 7daac72e ("drm/i915/pci: conversion to drm_device based logging macros.") Signed-off-by:
-
- Feb 05, 2020
-
-
Stanislav Lisovskiy authored
Start manipulating DBuf slices as a mask, but not as a total number, as current approach doesn't give us full control on all combinations of slices, which we might need(like enabling S2 only can't enabled by setting enabled_slices=1). Removed wrong code from intel_get_ddb_size as it doesn't match to BSpec. For now still just use DBuf slice until proper algorithm is implemented. Other minor code refactoring to get prepared for major DBuf assignment changes landed: - As now enabled slices contain a mask we still need some value which should reflect how much DBuf slices are supported by the platform, now device info contains num_supported_dbuf_slices. - Removed unneeded assertion as we are now manipulating slices in a more proper way. v2: Start using enabled_slices in dev_priv v3: "enabled_slices" is now "enabled_dbuf_slices_mask", as this now sits in dev_priv independently. v4: - Fixed debug print formatting to hex(Matt Roper) - Optimized dbuf slice updates to be used only if slice union is different from current conf(Matt Roper) - Fixed some functions to be static(Matt Roper) - Created a parameterized version for DBUF_CTL to simplify DBuf programming cycle(Matt Roper) - Removed unrequred field from GEN10_FEATURES(Matt Roper) v5: - Removed redundant programming dbuf slices helper(Ville Syrjälä) - Started to use parameterized loop for hw readout to get slices (Ville Syrjälä) - Added back assertion checking amount of DBUF slices enabled after DC states 5/6 transition, also added new assertion as starting from ICL DMC seems to restore the last DBuf power state set, rather than power up all dbuf slices as assertion was previously expecting(Ville Syrjälä) v6: - Now using enum for DBuf slices in this patch (Ville Syrjälä) - Removed gen11_assert_dbuf_enabled and put gen9_assert_dbuf_enabled back, as we really need to have a single unified assert here however currently enabling always slice 1 is enforced by BSpec, so we will have to OR enabled slices mask with 1 in order to be consistent with BSpec, that way we can unify that assertion and against the actual state from the driver, but not some hardcoded value.(concluded with Ville) - Remove parameterized DBUF_CTL version, to extract it to another patch.(Ville Syrjälä) v7: - Removed unneeded hardcoded return value for older gens from intel_enabled_dbuf_slices_mask - this now is handled in a unified manner since device info anyway returns max dbuf slices as 1 for older platforms(Matthew Roper) - Now using INTEL_INFO(dev_priv)->num_supported_dbuf_slices instead of intel_dbuf_max_slices function as it is trivial(Matthew Roper) v8: - Fixed icl_dbuf_disable to disable all dbufs still(Ville Syrjälä) v9: - Renamed _DBUF_CTL_S to DBUF_CTL_S(Ville Syrjälä) - Now using power_domain mutex to protect from race condition, which can occur because intel_dbuf_slices_update might be running in parallel to gen9_dc_off_power_well_enable being called from intel_dp_detect for instance, which causes assertion triggered by race condition, as gen9_assert_dbuf_enabled might preempt this when registers were already updated, while dev_priv was not. Reviewed-by:Stanislav Lisovskiy <stanislav.lisovskiy@intel.com> Signed-off-by:
-
- Feb 04, 2020
-
-
Wambui Karuga authored
Manual conversion of instances of printk based drm logging macros to the struct drm_device based logging macros in i915/i915_pci.c. Signed-off-by:
Wambui Karuga <wambui.karugax@gmail.com> Signed-off-by:
-
- Jan 24, 2020
-
-
Chris Wilson authored
We disabled rps while it appeared to be a contributing factor to system instablity, as that is now resolved, re-enable RPS and see how we fare. Signed-off-by:
-
- Jan 22, 2020
-
-
Jani Nikula authored
It's been a long enough transition period since the DRM_I915_FORCE_PROBE config and i915.force_probe module parameter were introduced in commit 7ef5ef5c ("drm/i915: add force_probe module parameter to replace alpha_support"). Remove alpha support. Cc: Arkadiusz Hiler <arkadiusz.hiler@intel.com> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Tomi Sarvela <tomi.p.sarvela@intel.com> Reviewed-by:
Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Acked-by:
-
- Dec 28, 2019
-
-
Lucas De Marchi authored
Drop the intel prefix since all these structs are static and prefer using the 3-letter prefix for each platform. v2: also remove gen from the device info (Ville) Signed-off-by:
-
- Dec 13, 2019
-
-
Venkata Sandeep Dhanalakota authored
We do not require to register the sysctl paths per instance, so making registration global. v2: make sysctl path register and unregister function driver specific (Tvrtko and Lucas). Cc: Sudeep Dutt <sudeep.dutt@intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Daniel Vetter <daniel.vetter@ffwll.ch> Cc: Chris Wilson <chris@chris-wilson.co.uk> Cc: Jani Nikula <jani.nikula@intel.com> Signed-off-by:Venkata Sandeep Dhanalakota <venkata.s.dhanalakota@intel.com> Reviewed-by:
-
- Dec 12, 2019
-
-
Lee Shawn C authored
U series device need different DDI buffer setup for eDP and DP. If driver did not recognize ULT id proerply. The setting for H and S series would be used. Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Jani Nikula <jani.nikula@intel.com> Cc: Anusha Srivatsa <anusha.srivatsa@intel.com> Cc: Cooper Chiou <cooper.chiou@intel.com> Signed-off-by:
Lee Shawn C <shawn.c.lee@intel.com> Signed-off-by:
-
- Dec 03, 2019
-
-
Chris Wilson authored
And Haswell still occasionally forgets it is meant to be using a new page directory, so repeat ourselves a little louder. <7> [509.919864] heartbeat rcs0 heartbeat {prio:-2147483645} not ticking <7> [509.919895] heartbeat Awake? 8 <7> [509.919903] heartbeat Barriers?: no <7> [509.919912] heartbeat Heartbeat: 3008 ms ago <7> [509.919930] heartbeat Reset count: 0 (global 0) <7> [509.919937] heartbeat Requests: <7> [509.921008] heartbeat active a7eb:56e1* @ 5847ms: <7> [509.921157] heartbeat ring->start: 0x00001000 <7> [509.921164] heartbeat ring->head: 0x00001610 <7> [509.921170] heartbeat ring->tail: 0x000023d8 <7> [509.921176] heartbeat ring->emit: 0x000023d8 <7> [509.921182] heartbeat ring->space: 0x00002570 <7> [509.921189] heartbeat ring->hwsp: 0x7fffe100 <7> [509.921197] heartbeat [head 1628, postfix 1738, tail 1750, batch 0xffffffff_ffffffff]: <7> [509.921289] heartbeat [0000] 7a000002 00100002 00000000 00000000 7a000002 01154c1e 7ffff080 00000000 <7> [509.921299] heartbeat [0020] 11000001 00002220 ffffffff 12400001 00002220 7ffff000 00000000 11000001 <7> [509.921308] heartbeat [0040] 00002228 6e900000 7a000002 00100002 00000000 00000000 7a000002 01154c1e <7> [509.921317] heartbeat [0060] 7ffff080 00000000 12400001 00002228 7ffff000 00000000 7a000002 00100002 <7> [509.921326] heartbeat [0080] 00000000 00000000 7a000002 01154c1e 7ffff080 00000000 7a000002 001010a1 <7> [509.921335] heartbeat [00a0] 7ffff080 00000000 04000000 11000005 00022050 00010001 00012050 00010001 <7> [509.921345] heartbeat [00c0] 0001a050 00010001 00000000 0c000000 459a110c 00000000 11000005 00022050 <7> [509.921354] heartbeat [00e0] 00010000 00012050 00010000 0001a050 00010000 12400001 0001a050 7ffff000 <7> [509.921363] heartbeat [0100] 00000000 04000001 18802100 00000000 7a000002 011050a1 7fffe100 000056e1 <7> [509.921370] heartbeat [0120] 01000000 00000000 <7> [509.921538] heartbeat MMIO base: 0x00002000 <7> [509.921682] heartbeat CCID: 0x3fa0110d <7> [509.922342] heartbeat RING_START: 0x00001000 <7> [509.922353] heartbeat RING_HEAD: 0x00001628 <7> [509.922366] heartbeat RING_TAIL: 0x000023d8 <7> [509.922381] heartbeat RING_CTL: 0x00003001 <7> [509.922396] heartbeat RING_MODE: 0x00004000 <7> [509.922408] heartbeat RING_IMR: ffffffde <7> [509.922421] heartbeat ACTHD: 0x00000000_30e01628 <7> [509.922434] heartbeat BBADDR: 0x00000000_00004004 <7> [509.922446] heartbeat DMA_FADDR: 0x00000000_00002800 <7> [509.922458] heartbeat IPEIR: 0x00000000 <7> [509.922470] heartbeat IPEHR: 0x780c0000 <7> [509.922642] heartbeat PP_DIR_BASE: 0x6e700000 <7> [509.922652] heartbeat PP_DIR_BASE_READ: 0x00000000 <7> [509.922662] heartbeat PP_DIR_DCLV: 0xffffffff <7> [509.922678] heartbeat E a7eb:56e1* @ 5849ms: <7> [509.922689] heartbeat E a7eb:56e2- @ 5849ms: <7> [509.922698] heartbeat E a7eb:56e3 @ 5848ms: <7> [509.922707] heartbeat E a7eb:56e4 @ 5848ms: <7> [509.922715] heartbeat E a7eb:56e5 @ 5847ms: <7> [509.922724] heartbeat E a7eb:56e6 @ 5846ms: <7> [509.922735] heartbeat E a7eb:56e7 @ 5846ms: <7> [509.922744] heartbeat ...skipping 4 executing requests... <7> [509.922754] heartbeat E a7eb:56ec @ 3010ms: <7> [509.922796] heartbeat HWSP: <7> [509.922807] heartbeat [0000] 00000001 00000000 00000000 00000000 00000000 00000000 00000000 00000000 <7> [509.922817] heartbeat [0020] 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 <7> [509.922826] heartbeat * <7> [509.922836] heartbeat [0100] 000056e0 00000000 00000000 00000000 00000000 00000000 00000000 00000000 <7> [509.922845] heartbeat [0120] 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 <7> [509.922851] heartbeat * <7> [509.922870] heartbeat Idle? no <7> [509.922878] heartbeat Signals: <7> [509.923000] heartbeat [a7eb:56e2] @ 5850ms Here, we have a failed context restore after the PD switch, but note that the PP_DIR_BASE register does not match the LRI in the ring. Bump it to 8^W 4 loops, and with that Baytrail starts passing the sanity checks. Signed-off-by:
-
- Nov 30, 2019
-
-
Chris Wilson authored
After much hair pulling, resort to preallocating the ppGTT entries on init to circumvent the apparent lack of PD invalidate following the write to PP_DCLV upon switching mm between contexts (and here the same context after binding new objects). However, the details of that PP_DCLV invalidate are still unknown, and it appears we need to reload the mm twice to cover over a timing issue. Worrying. Fixes: 3dc007fe ("drm/i915/gtt: Downgrade gen7 (ivb, byt, hsw) back to aliasing-ppgtt") Signed-off-by:
-
- Nov 11, 2019
-
-
Chris Wilson authored
Some basic information that is useful to know, such as how many cycles is a MI_NOOP. v2: Keep volatile pages pinned at all times! (Matthew) Signed-off-by:
-
- Oct 29, 2019
-
-
José Roberto de Souza authored
DSC could be fused off, so not all GEN10+ platforms will support it. Cc: Manasi Navare <manasi.d.navare@intel.com> Cc: Martin Peres <martin.peres@linux.intel.com> Reviewed-by:
Ramalingam C <ramalingam.c@intel.com> Reviewed-by:
Manasi Navare <manasi.d.navare@intel.com> Signed-off-by:
-
José Roberto de Souza authored
HDCP could be fused off, so not all GEN9+ platforms will support it. Cc: Ville Syrjälä <ville.syrjala@linux.intel.com> Cc: Martin Peres <martin.peres@linux.intel.com> Reviewed-by:
-
- Oct 25, 2019
-
-
Stuart Summers authored
Add a new macro for GEN12 platforms to be grouped under dgfx feature set. Signed-off-by:
-
- Oct 18, 2019
-
-
Matthew Auld authored
Convert stolen memory over to a region object. Still leaves open the question with what to do with pre-allocated objects... Signed-off-by:
-
Matthew Auld authored
Convert shmem to an intel_memory_region. Signed-off-by:
-
- Oct 02, 2019
-
-
Jani Nikula authored
Split out the code related to vga client and vgaarb all over the place into new intel_vga.[ch]. No functional changes. Cc: Ville Syrjälä <ville.syrjala@linux.intel.com> Cc: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by:
-
- Sep 24, 2019
-
-
Chris Wilson authored
If we disable rps, it appears the Tigerlake is stable enough to run multiple engines simultaneously in CI. As disabling rps should only cause the execution to be slow, whereas many features depend on the different engines, we would prefer to have the engines enabled while the machine hangs are being debugged. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=111714 Signed-off-by:
-
