)]}' { "log": [ { "commit": "a838834b2f7cbc09b6319a1fc332c03e4d665b20", "tree": "548865f29964a44d9f62b51f88586832e58aba08", "parents": [ "42a21826dc54583cdb79cc8477732e911ac9c376" ], "author": { "name": "Alex Deucher", "email": "alexander.deucher@amd.com", "time": "Tue Jul 30 16:43:55 2013 -0400" }, "committer": { "name": "Alex Deucher", "email": "alexander.deucher@amd.com", "time": "Tue Jul 30 17:24:13 2013 -0400" }, "message": "drm: fix 64 bit drm fixed point helpers\n\nSign bit wasn\u0027t handled properly and a small typo.\n\nThanks to Christian for helping me sort this out.\n\nSigned-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n" }, { "commit": "6d35dea107834eb549c1fba28fea6ec39c81d0ba", "tree": "bbada84a6400a79ec71d47df6264c55e1485c7fa", "parents": [ "78467dc5f70fb9bee4a32c0c3714c99b0b5465c7" ], "author": { "name": "Joonyoung Shim", "email": "jy0922.shim@samsung.com", "time": "Fri Jun 28 14:24:55 2013 +0900" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Jul 05 15:44:54 2013 +1000" }, "message": "drm/cma: remove GEM CMA specific dma_buf functionality\n\nWe can use prime helpers instead.\n\nSigned-off-by: Joonyoung Shim \u003cjy0922.shim@samsung.com\u003e\nAcked-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "78467dc5f70fb9bee4a32c0c3714c99b0b5465c7", "tree": "fe5b399e17578e2a819083df5d7e0fee11df4027", "parents": [ "7c397cd97b8f46659698396b420bd48c3e6703e6" ], "author": { "name": "Joonyoung Shim", "email": "jy0922.shim@samsung.com", "time": "Fri Jun 28 14:24:54 2013 +0900" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Jul 05 15:44:49 2013 +1000" }, "message": "drm/cma: add low-level hook functions to use prime helpers\n\nInstead of using the dma_buf functionality for GEM CMA, we can use prime\nhelpers if we can provide low-level hook functions for GEM CMA.\n\nSigned-off-by: Joonyoung Shim \u003cjy0922.shim@samsung.com\u003e\nAcked-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "7c397cd97b8f46659698396b420bd48c3e6703e6", "tree": "9125dd69b151ae14c47fa4f3b9e11c27b626821a", "parents": [ "f9d8a1294d37449f3f1b842ffb275e2ca41f5cf4" ], "author": { "name": "Joonyoung Shim", "email": "jy0922.shim@samsung.com", "time": "Fri Jun 28 14:24:53 2013 +0900" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Jul 05 15:44:44 2013 +1000" }, "message": "drm: add mmap function to prime helpers\n\nThis adds to call low-level mmap() from prime helpers.\n\nSigned-off-by: Joonyoung Shim \u003cjy0922.shim@samsung.com\u003e\nAcked-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "69163ea82732894e8c1e17df4010372ed078efdd", "tree": "b0565b782a3ecaa67b64c99a3eaa6b1a8c6685fd", "parents": [ "04274cd058193cefc22f1ebda1e634d96cb59fa5" ], "author": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Mon Jul 01 22:05:53 2013 +0200" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Jul 04 10:57:40 2013 +1000" }, "message": "drm/mm: kill color_search_free/get_block\n\ndrm/i915 is the only user of the color allocation handling and\nswitched to insert_node a while ago. So we can ditch this.\n\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nReviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "77ef8bbc87be7ad10b410247efc6d0f10676b401", "tree": "9cf69a4bbf55b95eeb25a1c4b8a700db138486c7", "parents": [ "6ef92fbea2b5680204da5b8796e8972109d01bd3" ], "author": { "name": "David Herrmann", "email": "dh.herrmann@gmail.com", "time": "Mon Jul 01 20:32:58 2013 +0200" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Tue Jul 02 13:34:41 2013 +1000" }, "message": "drm: make drm_mm_init() return void\n\nThere is no reason to return \"int\" as this function never fails.\nFurthermore, several drivers (ast, sis) already depend on this.\n\nSigned-off-by: David Herrmann \u003cdh.herrmann@gmail.com\u003e\nReviewed-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "d482e5fa299c2cfbb4700143dd766273730e2357", "tree": "817263cf3963a143ffa63396c4ac4ee32bed10a6", "parents": [ "421cda3e324fce40e3f7abdf3d31cae1d0deddcd" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Jun 28 20:31:34 2013 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Jun 28 20:31:34 2013 +1000" }, "message": "Revert \"drm: kms_helper: don\u0027t lose hotplug event\"\n\nThis reverts commit 160954b7bca43da7cd3cfbce310e6df919a8216e.\n\nThis was rearming the workqueue with a 0 timeout, causing\na WARN_ON, and possible loop.\n\nDaniel writes:\n\"I\u0027ve looked a bit into this and I think we need to have a separate\nwork struct for recovering these lost hotplug events since the\ncontinuous self-rearming case is a real risk (e.g. if a connector\nflip-flops all the time). At least I don\u0027t see a sane way to block out\nre-arming with the current code in a simple way. So reverting the\noffender seems like the right thing and I\u0027ll go back to the drawing\nboard for 3.12.\"\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "8f262540e61c7caaff791bde8336196ac7aca77a", "tree": "f258135d3155156a99773441e56b23ccfb766049", "parents": [ "8bd4ce56783da6dc96484462ddb113417e52150c" ], "author": { "name": "Maarten Lankhorst", "email": "m.b.lankhorst@gmail.com", "time": "Thu Jun 27 13:48:28 2013 +0200" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Jun 28 12:04:15 2013 +1000" }, "message": "drm/ttm: get rid of ttm_bo_is_reserved\n\nSigned-off-by: Maarten Lankhorst \u003cmaarten.lankhorst@canonical.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "3482032457f50cae196f6397ebec7f5f2ad3cf7d", "tree": "7f41fe11224f91d1fb1b90edb0ccec56970cc921", "parents": [ "c43f9b16991950c00621641ef2c5cd4a3af2a052" ], "author": { "name": "Maarten Lankhorst", "email": "m.b.lankhorst@gmail.com", "time": "Thu Jun 27 13:48:24 2013 +0200" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Jun 28 12:04:09 2013 +1000" }, "message": "drm/ttm: inline ttm_bo_reserve and related calls\n\nMakes lockdep a lot more useful.\n\nSigned-off-by: Maarten Lankhorst \u003cmaarten.lankhorst@canonical.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "5e338405119a80aa59e811626739122d1c15045d", "tree": "3474b9ab52408c78480a92a9d0c33626c61d7473", "parents": [ "b580c9e2b7ba5030a795aa2fb73b796523d65a78" ], "author": { "name": "Maarten Lankhorst", "email": "m.b.lankhorst@gmail.com", "time": "Thu Jun 27 13:48:19 2013 +0200" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Jun 28 12:04:01 2013 +1000" }, "message": "drm/ttm: convert to the reservation api\n\nNow that the code is compatible in semantics, flip the switch.\nUse ww_mutex instead of the homegrown implementation.\n\nww_mutex uses -EDEADLK to signal that the caller has to back off,\nand -EALREADY to indicate this buffer is already held by the caller.\n\nttm used -EAGAIN and -EDEADLK for those, respectively. So some changes\nwere needed to handle this correctly.\n\nSigned-off-by: Maarten Lankhorst \u003cmaarten.lankhorst@canonical.com\u003e\nReviewed-by: Jerome Glisse \u003cjglisse@redhat.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "ecff665f5e3f1c6909353e00b9420e45ae23d995", "tree": "8fed7d4570ec707427e954c6d2695d1549e08364", "parents": [ "786d7257e537da0674c02e16e3b30a44665d1cee" ], "author": { "name": "Maarten Lankhorst", "email": "m.b.lankhorst@gmail.com", "time": "Thu Jun 27 13:48:17 2013 +0200" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Jun 28 12:02:20 2013 +1000" }, "message": "drm/ttm: make ttm reservation calls behave like reservation calls\n\nThis commit converts the source of the val_seq counter to\nthe ww_mutex api. The reservation objects are converted later,\nbecause there is still a lockdep splat in nouveau that has to\nresolved first.\n\nSigned-off-by: Maarten Lankhorst \u003cmaarten.lankhorst@canonical.com\u003e\nReviewed-by: Jerome Glisse \u003cjglisse@redhat.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "5b0207bb59b26a8cb5b53f13c99c67eb0d2a7abb", "tree": "114adf60884b7a69ed2812c7b06f41d5c228981a", "parents": [ "28419261b09aa3a5118647b1ed93809ca97c5354", "a144acbcfbfea44a80afc8f880a7ad72bf01c819" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Jun 28 10:53:23 2013 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Jun 28 10:53:23 2013 +1000" }, "message": "Merge branch \u0027drm-next-3.11\u0027 of git://people.freedesktop.org/~agd5f/linux into drm-next\n\nThis is the pull request for radeon for 3.11. Highlights include:\n\n- Support for CIK (Sea Islands) asics: 3D, compute, UVD\n- DPM (Dynamic Power Management) support for 6xx-SI\n- ASPM support for 6xx-SI\n- Assorted bug fixes\n\n* \u0027drm-next-3.11\u0027 of git://people.freedesktop.org/~agd5f/linux: (168 commits)\n drm/radeon/SI: fix TDP adjustment in set_power_state\n drm/radeon/NI: fix TDP adjustment in set_power_state\n drm/radeon: fix endian issues in atombios dpm code\n drm/radeon/dpm: fix UVD clock setting on SI\n drm/radeon/dpm: fix UVD clock setting on cayman\n drm/radeon/dpm: add support for setting UVD clock on rv6xx\n drm/radeon/dpm: add support for setting UVD clock on rs780\n drm/radeon: fix typo in ni_print_power_state\n drm/radeon: fix typo in cik_select_se_sh()\n drm/radeon/si: fix typo in function name\n drm/radeon/dpm: fix typo in setting uvd clock\n drm/radeon/dpm: add dpm_set_power_state failure output (si)\n add dpm_set_power_state failure output (7xx-ni)\n drm/radeon/dpm: add dpm_set_power_state failure output (7xx-ni)\n drm/radeon/dpm: add dpm_enable failure output (si)\n drm/radeon/dpm: add dpm_enable failure output (7xx-ni)\n drm/radeon/kms: add dpm support for SI (v7)\n drm/radeon: switch SI to use radeon_ucode.h\n drm/radeon: add SI to r600_is_internal_thermal_sensor()\n drm/radeon/dpm/rs780: properly catch errors in dpm setup\n ...\n" }, { "commit": "28419261b09aa3a5118647b1ed93809ca97c5354", "tree": "5098381ee695009fce1fa4a25ba34d487eb4f35e", "parents": [ "4a009085978de90db40f9f38bcfad501f86ca959", "854c94a7854a4fabdd7db451cf1774e6dcba6bab" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Jun 28 09:50:34 2013 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Jun 28 09:50:34 2013 +1000" }, "message": "Merge tag \u0027drm-intel-next-2013-06-18\u0027 of git://people.freedesktop.org/~danvet/drm-intel into drm-next\n\nLast 3.11 feature pull. I have a few odds bits and pieces and fixes in my\nqueue, I\u0027ll sort them out later on to see what\u0027s for 3.11-fixes and what\u0027s\nfor 3.12. But nothing to hold this here up imo.\n\nHighlights:\n- more hangcheck work from Mika and Chris to prepare for arb robustness\n- trickle feed fixes from Ville\n- first parts of the shared pch pll rework, with some basic hw state\n readout and cross-checking (this shuts up the confused pch pll refcount\n WARN that Linus just recently forwarded)\n- Haswell audio power well support from Wang Xingchao (alsa bits acked by\n Takashi)\n- some cleanups and asserts sprinkling around the plane/gamma enabling\n sequence from Ville\n- more gtt refactoring from Ben\n- clear up the adjusted-\u003emode vs. pixel clock vs. port clock confusion\n- 30bpp support, this time for real hopefully\n\n* tag \u0027drm-intel-next-2013-06-18\u0027 of git://people.freedesktop.org/~danvet/drm-intel: (97 commits)\n drm/i915: remove a superflous semi-colon\n drm/i915: Kill useless \"Enable panel fitter\" comments\n drm/i915: Remove extra \"ring\" from error message\n drm/i915: simplify the reduced clock handling for pch plls\n drm/i915: stop killing pfit on i9xx\n drm/i915: explicitly set up PIPECONF (and gamma table) on haswell\n drm/i915: set up PIPECONF explicitly for i9xx/vlv platforms\n drm/i915: set up PIPECONF explicitly on ilk-ivb\n drm/i915: find guilty batch buffer on ring resets\n drm/i915: store ring hangcheck action\n drm/i915: add batch bo to i915_add_request()\n drm/i915: change i915_add_request to macro\n drm/i915: add i915_gem_context_get_hang_stats()\n drm/i915: add struct i915_ctx_hang_stats\n drm/i915: Try harder to disable trickle feed on VLV\n drm/i915: fix up pch pll enabling for pixel multipliers\n drm/i915: hw state readout and cross-checking for shared dplls\n drm/i915: WARN on lack of shared dpll\n drm/i915: split up intel_modeset_check_state\n drm/i915: extract readout_hw_state from setup_hw_state\n ...\n\nConflicts:\n\tdrivers/gpu/drm/i915/intel_display.c\n\tdrivers/gpu/drm/i915/intel_fb.c\n\tdrivers/gpu/drm/i915/intel_sdvo.c\n" }, { "commit": "210a0b9e212370ed8c2784c2115e7ff4bb1259bd", "tree": "35efeaaa2b97e7f4e0a9fe3475967bd859cf55d6", "parents": [ "9ed36f750534e2c6533fcbf32df89cf20cf87e91" ], "author": { "name": "Alex Deucher", "email": "alexander.deucher@amd.com", "time": "Fri Mar 22 10:35:50 2013 -0400" }, "committer": { "name": "Alex Deucher", "email": "alexander.deucher@amd.com", "time": "Thu Jun 27 19:16:37 2013 -0400" }, "message": "drm: add some additional fixed point helpers (v3)\n\nRequired for certain driver calculations. Code\nwas written by Christian König and ported to the\ndrm by me.\n\nv2: fix 64 bit divides\nv3: fix 64 bit for real (math64.h)\n\nSigned-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n" }, { "commit": "4c813d4d759c0e6b83bfd73795e9526493556dc2", "tree": "5cab2327d0db876f86fc72cb4acac23d58107cba", "parents": [ "c19b3e238d7573cbe0bb60f4578b7d1de4a13746" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Jun 20 11:48:52 2013 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Jun 28 09:13:39 2013 +1000" }, "message": "drm: add hotspot support for cursors.\n\nSo it looks like for virtual hw cursors on QXL we need to inform\nthe \"hw\" device what the cursor hotspot parameters are. This\nmakes sense if you think the host has to draw the cursor and interpret\nclicks from it. However the current modesetting interface doesn\u0027t support\npassing the hotspot information from userspace.\n\nThis implements a new cursor ioctl, that takes the hotspot info as well,\nuserspace can try calling the new interface and if it gets -ENOSYS it means\nits on an older kernel and can just fallback.\n\nReviewed-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "26e2235d5f0d3d30c6d9e7cabd37ce38da35ce5d", "tree": "fe21e41bbeffb8291f4c88fba9ed0fa78cb4a363", "parents": [ "516184bd01b56aad8bb53bb6ffe51f55cb8e3112" ], "author": { "name": "Alex Deucher", "email": "alexander.deucher@amd.com", "time": "Thu Jan 24 12:30:24 2013 -0500" }, "committer": { "name": "Alex Deucher", "email": "alexander.deucher@amd.com", "time": "Thu Jun 27 10:49:14 2013 -0400" }, "message": "drm/radeon: add current KB pci ids\n\nSigned-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n" }, { "commit": "516184bd01b56aad8bb53bb6ffe51f55cb8e3112", "tree": "e2a2cdb7119c368fcc4fc7984a2646f3b9614df1", "parents": [ "39aee490288908a0fe50f09de8b13e8423ed7b21" ], "author": { "name": "Alex Deucher", "email": "alexander.deucher@amd.com", "time": "Fri Jun 07 11:52:42 2013 -0400" }, "committer": { "name": "Alex Deucher", "email": "alexander.deucher@amd.com", "time": "Thu Jun 27 10:49:13 2013 -0400" }, "message": "drm/radeon: add current Bonaire PCI ids\n\nSigned-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n" }, { "commit": "160954b7bca43da7cd3cfbce310e6df919a8216e", "tree": "c10b295d21363961641e452a92bb8179e05106f0", "parents": [ "2b54f78190a2683f48f1295a4c28e98df042546a" ], "author": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Thu Jun 06 00:17:25 2013 +0200" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Jun 27 20:34:16 2013 +1000" }, "message": "drm: kms_helper: don\u0027t lose hotplug event\n\nThere\u0027s a race window (small for hpd, 10s large for polled outputs)\nwhere userspace could sneak in with an unrelated connnector probe\nioctl call and eat the hotplug event (since neither the hpd nor the\npoll code see a state change).\n\nTo avoid this, check whether the connector state changes in all other\n-\u003edetect calls (in the current helper code that\u0027s only probe_single)\nand if that\u0027s the case, fire off a hotplug event. Note that we can\u0027t\ndirectly call the hotplug event handler, since that expects that no\nlocks are held (due to reentrancy with the fb code to update the kms\nconsole).\n\nAlso, this requires that drivers using the probe_single helper\nfunction set up the poll work. All current drivers do that already,\nand with the reworked hpd handling there\u0027ll be no downside to\nunconditionally setting up the poll work any more.\n\nReviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "5cef29aa5227e6347145940a7bccde92fd9a1afa", "tree": "43b7f4fce89084737d5509f03aeb1dd1ae3eeb59", "parents": [ "cc85e1217f598f342b69dc44710d7a7355513a1b" ], "author": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Sat Jun 15 00:13:16 2013 +0200" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Tue Jun 25 13:04:11 2013 +1000" }, "message": "drm: fix fb leak in setcrtc\n\nDrivers are allowed (actually have to) disable unrelated crtcs in\ntheir -\u003eset_config callback (when we steal all the connectors from\nthat crtc). If they do that they\u0027ll clear crtc-\u003efb to NULL.\n\nWhich results in a refcount leak, since the drm core is keeping track\nof that reference.\n\nTo fix this track the old fb of all crtcs and adjust references for\nall of them. Of course, since we only hold an additional reference for\nthe fb for the current crtc we need to increase refcounts before we\ndrop the old one.\n\nThis approach has the benefit that it inches us a bit closer to an\natomic modeset world, where we want to update the config of all crtcs\nin one step.\n\nThis regression has been introduce in the framebuffer refcount\nconversion, specifically in\n\ncommit b0d1232589df5575c5971224ac4cb30e7e525884\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Tue Dec 11 01:07:12 2012 +0100\n\n drm: refcounting for crtc framebuffers\n\nReported-by: Russell King \u003clinux@arm.linux.org.uk\u003e\nCc: Russell King \u003clinux@arm.linux.org.uk\u003e\nCc: stable@vger.kernel.org\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "778ad903f951a45a309a5a70dbdcf38eccabeaf0", "tree": "820d3bd434493b7a055798eeb30e2b879dd42c96", "parents": [ "b72447cdf1298b46af87d754bfb5d3db6eb7dbfe" ], "author": { "name": "Ville Syrjälä", "email": "ville.syrjala@linux.intel.com", "time": "Mon Jun 03 16:11:42 2013 +0300" }, "committer": { "name": "Dave Airlie", "email": "airlied@gmail.com", "time": "Mon Jun 17 19:42:46 2013 +1000" }, "message": "drm: Remove some unused stuff from drm_plane\n\nThere\u0027s a bunch of unused members inside drm_plane, bloating the size of\nthe structure needlessly. Eliminate them.\n\nv2: Remove all of it from kernel-doc too\n\nReviewed-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\nReviewed-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nSigned-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@gmail.com\u003e\n" }, { "commit": "9125e6186822b2698da17690416cd1b55c030115", "tree": "33c993cb831125d9c59dda1b011737cfb2be54ef", "parents": [ "e6e792092e816bea0797995c886fb057c91d4546" ], "author": { "name": "Ville Syrjälä", "email": "ville.syrjala@linux.intel.com", "time": "Mon Jun 03 16:10:40 2013 +0300" }, "committer": { "name": "Dave Airlie", "email": "airlied@gmail.com", "time": "Mon Jun 17 18:32:54 2013 +1000" }, "message": "drm: Add drm_plane_force_disable()\n\ndrm_plane_force_disable() will forcibly disable the plane even if user\nhad previously requested the plane to be enabled.\n\nThis can be used to force planes to be off when restoring the fbdev\nmode.\n\nThe code was simply pulled from drm_framebuffer_remove(), which now\ncalls the new function as well.\n\nv2: Check plane-\u003efb in drm_plane_force_disable(), drop bogus comment\n about disabling crtc\n\nSigned-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@gmail.com\u003e\n" }, { "commit": "e6dfcc5303d5d31cb36e36405acd766c8ed2c923", "tree": "986beb9aba4a62ea1449dde210005e82084e8c5e", "parents": [ "9bc3cd5673d84d29272fa7181a4dfca83cbb48c1", "92d44621ad2d083bc03920c904ca0a5eb10d9ded" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Tue Jun 11 08:38:56 2013 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Tue Jun 11 08:38:56 2013 +1000" }, "message": "Merge tag \u0027drm-intel-next-2013-06-01\u0027 of git://people.freedesktop.org/~danvet/drm-intel into drm-next\n\nDaniel writes:\nAnother round of drm-intel-next for 3.11. Highlights:\n- Haswell IPS support (Paulo Zanoni)\n- VECS support on Haswell (Ben Widawsky, Xiang Haihao, ...)\n- Haswell watermark fixes (Paulo Zanoni)\n- \"Make the gun bigger again\" multithread fence fix from Chris.\n- i915_error_state finnally no longer fails with -ENOMEM! Big thanks to\n Mika for tackling this.\n- vlv sideband locking fixes from Jani\n- Hangcheck prep work for arb_robustness support (Mika\u0026Chris)\n- edp vs cpu port confusion clean-up from Imre\n- pile of smaller fixes and cleanups all over.\n\n* tag \u0027drm-intel-next-2013-06-01\u0027 of git://people.freedesktop.org/~danvet/drm-intel: (70 commits)\n drm/i915: add i915_ips_status debugfs entry\n drm/i915: add enable_ips module option\n drm/i915: implement IPS feature\n drm/i915: fix up the edp power well check\n drm/i915: add I915_PARAM_HAS_VEBOX to i915_getparam\n drm/i915: add I915_EXEC_VEBOX to i915_gem_do_execbuffer()\n drm/i915: add VEBOX into debugfs\n drm/i915: Enable vebox interrupts\n drm/i915: vebox interrupt get/put\n drm/i915: consolidate interrupt naming scheme\n drm/i915: Convert irq_refounct to struct\n drm/i915: make PM interrupt writes non-destructive\n drm/i915: Add PM regs to pre/post install\n drm/i915: Create an ivybridge_irq_preinstall\n drm/i915: Create a more generic pm handler for hsw+\n drm/i915: add support for 5/6 data buffer partitioning on Haswell\n drm/i915: properly set HSW WM_LP watermarks\n drm/i915: properly set HSW WM_PIPE registers\n drm/i915: fix pch_nop support\n drm/i915: Vebox ringbuffer init\n ...\n" }, { "commit": "edee06b6f6b55c81c4cb78d0776d8ca01b05380d", "tree": "b2ed77452d8375d53a57012c0ff16affb7aef717", "parents": [ "d20d3174806ef6589cb912a488657d21fcd7ece2", "71d7282a0f1abb488e5be4d154893579624bc683" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Tue Jun 11 08:14:21 2013 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Tue Jun 11 08:14:21 2013 +1000" }, "message": "Merge branch \u0027drm/next\u0027 of git://linuxtv.org/pinchartl/fbdev into drm-next\n\nGEM CMA PRIME support from Laurent.\n\n* \u0027drm/next\u0027 of git://linuxtv.org/pinchartl/fbdev:\n drm: GEM CMA: Add DRM PRIME support\n drm: GEM CMA: Split object mapping into GEM mapping and CMA mapping\n drm: GEM CMA: Split object creation into object alloc and DMA memory alloc\n drm/omap: Use drm_gem_mmap_obj() to implement dma-buf mmap\n drm/gem: Split drm_gem_mmap() into object search and object mapping\n" }, { "commit": "d20d3174806ef6589cb912a488657d21fcd7ece2", "tree": "61b40825fc9a2eacd6e2fe94ff4b3c3cc509d077", "parents": [ "4ee62c7669be1a6f1dd407e5ba7e38c0e2204e92" ], "author": { "name": "Ville Syrjälä", "email": "ville.syrjala@linux.intel.com", "time": "Fri Jun 07 15:43:07 2013 +0000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Tue Jun 11 08:13:56 2013 +1000" }, "message": "drm: Constify the pretty-print functions\n\nThe structures and strings involved with various pretty-print functions\naren\u0027t meant to be modified, so make them all const. The exception is\ndrm_connector_enum_list which does get modified in drm_connector_init().\n\nWhile at it move the drm_get_connector_status_name() prototype from\ndrmP.h to drm_crtc.h where it belongs.\n\nSigned-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nReviewed-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "6ba6d03e69125ef42a63e90d45e49c659ea3c34f", "tree": "5a1d0fbe6fc7a61e51781ea732ce6ac049dbd1da", "parents": [ "ca793f75d91d61fd0009c4179eb88aac54a74239" ], "author": { "name": "Ville Syrjälä", "email": "ville.syrjala@linux.intel.com", "time": "Mon Jun 10 11:15:10 2013 +0300" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Tue Jun 11 08:13:54 2013 +1000" }, "message": "drm: Print pretty names for pixel formats\n\nRather than just printing the pixel format as a hex number, decode the\nfourcc into human readable form, and also decode the LE vs. BE flag.\n\nKeep printing the raw hex number too in case it contains non-printable\ncharacters.\n\nSome examples what the new drm_get_format_name() produces:\nDRM_FORMAT_XRGB8888: \"XR24 little-endian (0x34325258)\"\nDRM_FORMAT_YUYV: \"YUYV little-endian (0x56595559)\"\nDRM_FORMAT_RGB565|DRM_FORMAT_BIG_ENDIAN: \"RG16 big-endian (0xb6314752)\"\nUnprintable characters: \"D??? big-endian (0xff7f0244)\"\n\nv2: Fix patch author\n\nSigned-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "102d6dba306c825cd5c310f73868b130931f47aa", "tree": "75346aaa4f291511fa64a94132813678f9cefd09", "parents": [ "328a4719b6a0930721b5f8d5c69993d3b6e3913f" ], "author": { "name": "Maarten Lankhorst", "email": "maarten.lankhorst@canonical.com", "time": "Tue Apr 09 09:18:44 2013 +0200" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Tue Jun 11 07:51:11 2013 +1000" }, "message": "drm: add unpin function to prime helpers\n\nPrevents buffers from being pinned forever.\n\nSigned-off-by: Maarten Lankhorst \u003cmaarten.lankhorst@canonical.com\u003e\nReviewed-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "71d7282a0f1abb488e5be4d154893579624bc683", "tree": "73b0d225ea22cec962cd543b1fc2ed22e3ebf7e9", "parents": [ "ebaf9e033e6dc9b584176c1731f4e07360d4d231" ], "author": { "name": "Laurent Pinchart", "email": "laurent.pinchart+renesas@ideasonboard.com", "time": "Sun Feb 17 01:57:30 2013 +0100" }, "committer": { "name": "Laurent Pinchart", "email": "laurent.pinchart+renesas@ideasonboard.com", "time": "Sat Jun 08 09:14:05 2013 +0200" }, "message": "drm: GEM CMA: Add DRM PRIME support\n\nSigned-off-by: Laurent Pinchart \u003claurent.pinchart+renesas@ideasonboard.com\u003e\nReviewed-by: Rob Clark \u003crobdclark@gmail.com\u003e\n" }, { "commit": "1c5aafa6eee2d5712f774676d407e5ab6dae9a1b", "tree": "5cb8d947cca86c336977ad86a5e0f76eac0e0533", "parents": [ "328a4719b6a0930721b5f8d5c69993d3b6e3913f" ], "author": { "name": "Laurent Pinchart", "email": "laurent.pinchart+renesas@ideasonboard.com", "time": "Tue Apr 16 14:14:52 2013 +0200" }, "committer": { "name": "Laurent Pinchart", "email": "laurent.pinchart+renesas@ideasonboard.com", "time": "Sat Jun 08 09:14:03 2013 +0200" }, "message": "drm/gem: Split drm_gem_mmap() into object search and object mapping\n\nThe drm_gem_mmap() function first finds the GEM object to be mapped\nbased on the fake mmap offset and then maps the object. Split the object\nmapping code into a standalone drm_gem_mmap_obj() function that can be\nused to implement dma-buf mmap() operations.\n\nSigned-off-by: Laurent Pinchart \u003claurent.pinchart+renesas@ideasonboard.com\u003e\nReviewed-by: Rob Clark \u003crobdclark@gmail.com\u003e\n" }, { "commit": "a38911a3fede294e2adfd2deea8104dfbbd760c5", "tree": "9262c62e16108554fb626cb04189d17e449bb938", "parents": [ "99a2008d0b32d72dfc2a54e7be1eb698dd2e3bd6" ], "author": { "name": "Wang Xingchao", "email": "xingchao.wang@linux.intel.com", "time": "Thu May 30 22:07:11 2013 +0800" }, "committer": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Thu Jun 06 17:32:16 2013 +0200" }, "message": "i915/drm: Add private api for power well usage\n\nHaswell Display audio depends on power well in graphic side, it should\nrequest power well before use it and release power well after use.\nI915 will not shutdown power well if it detects audio is using.\nThis patch protects display audio crash for Intel Haswell C3 stepping board.\n\nSigned-off-by: Wang Xingchao \u003cxingchao.wang@linux.intel.com\u003e\nReviewed-by: Takashi Iwai \u003ctiwai@suse.de\u003e\nReviewed-by: Damien Lespiau \u003cdamien.lespiau@intel.com\u003e\nReviewed-by: Jesse Barnes \u003cjbarnes@virtuousgeek.org\u003e\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n" }, { "commit": "03dae7c567d24c49e826a033df45802ac9d1d6c8", "tree": "d9eb736873638681e390e0648da9ea21442563da", "parents": [ "63e28a7a5ffce59b645ca9cbcc01e1e8be56bd75" ], "author": { "name": "Andy Lutomirski", "email": "luto@amacapital.net", "time": "Mon May 13 23:58:47 2013 +0000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri May 31 13:37:37 2013 +1000" }, "message": "drm: Remove mtrr_add and mtrr_del fallback hack for non-MTRR systems\n\nThere are no users left in drivers/gpu.\n\nReviewed-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nSigned-off-by: Andy Lutomirski \u003cluto@amacapital.net\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "247d36d75128ba1f63702e0e6185d9a7a23ee5cb", "tree": "947293f7e2ce7a9d65f5c7325a9cc4fab05dba32", "parents": [ "d0d98eedee2178c803dd824bb09f52b0e2ac1811" ], "author": { "name": "Andy Lutomirski", "email": "luto@amacapital.net", "time": "Mon May 13 23:58:41 2013 +0000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri May 31 13:02:54 2013 +1000" }, "message": "drm (ast, cirrus, mgag200, nouveau, savage, vmwgfx): Remove drm_mtrr_{add, del}\n\nThis replaces drm_mtrr_{add,del} with arch_phys_wc_{add,del}. The\ninterface is simplified (because the base and size parameters to\ndrm_mtrr_del never did anything), and it no longer adds MTRRs on\nsystems that don\u0027t need them.\n\nReviewed-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nSigned-off-by: Andy Lutomirski \u003cluto@amacapital.net\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "e81f3d81e282a156b47c1c2c09a1976e34073060", "tree": "89c9b5ce24c9ac1a031ada1b01392134b4211e18", "parents": [ "970fa986fadb1165cf38b45b70e98302a3bee497", "e1b73cba13a0cc68dd4f746eced15bd6bb24cda4" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri May 31 12:56:05 2013 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri May 31 12:56:05 2013 +1000" }, "message": "Merge tag \u0027drm-intel-next-2013-05-20-merged\u0027 of git://people.freedesktop.org/~danvet/drm-intel into drm-next\n\nDaniel writes:\nHighlights (copy-pasted from my testing cycle mails):\n- fbc support for Haswell (Rodrigo)\n- streamlined workaround comments, including an igt tool to grep for\n them (Damien)\n- sdvo and TV out cleanups, including a fixup for sdvo multifunction devices\n- refactor our eDP mess a bit (Imre)\n- don\u0027t register the hdmi connector on haswell when desktop eDP is present\n- vlv support is no longer preliminary!\n- more vlv fixes from Jesse for stolen and dpll handling\n- more flexible power well checking infrastructure from Paulo\n- a few gtt patches from Ben\n- a bit of OCD cleanups for transcoder #defines and an assorted pile\n of smaller things.\n- fixes for the gmch modeset sequence\n- a bit of OCD around plane/pipe usage (Ville)\n- vlv turbo support (Jesse)\n- tons of vlv modeset fixes (Jesse et al.)\n- vlv pte write fixes (Kenneth Graunke)\n- hpd filtering to avoid costly probes on unaffected outputs (Egbert Eich)\n- intel dev_info cleanups and refactorings (Damien)\n- vlv rc6 support (Jesse)\n- random pile of fixes around non-24bpp modes handling\n- asle/opregion cleanups and locking fixes (Jani)\n- dp dpll refactoring\n- improvements for reduced_clock computation on g4x/ilk+\n- pfit state refactored to use pipe_config (Jesse)\n- lots more computed modeset state moved to pipe_config, including readout\n and cross-check support\n- fdi auto-dithering for ivb B/C links, using the neat pipe_config\n improvements\n- drm_rect helpers plus sprite clipping fixes (Ville)\n- hw context refcounting (Mika + Ben)\n\n* tag \u0027drm-intel-next-2013-05-20-merged\u0027 of git://people.freedesktop.org/~danvet/drm-intel: (155 commits)\n drm/i915: add support for dvo Chrontel 7010B\n drm/i915: Use pipe config state to control gmch pfit enable/disable\n drm/i915: Use pipe_config state to disable ilk+ pfit\n drm/i915: panel fitter hw state readout\u0026check support\n drm/i915: implement WADPOClockGatingDisable for LPT\n drm/i915: Add missing platform tags to FBC workaround comments\n drm/i915: rip out an unused lvds_reg variable\n drm/i915: Compute WR PLL dividers dynamically\n drm/i915: HSW FBC WaFbcDisableDpfcClockGating\n drm/i915: HSW FBC WaFbcAsynchFlipDisableFbcQueue\n drm/i915: Enable FBC at Haswell.\n drm/i915: IVB FBC WaFbcDisableDpfcClockGating\n drm/i915: IVB FBC WaFbcAsynchFlipDisableFbcQueue\n drm/i915: Add support for FBC on Ivybridge.\n drm/i915: Organize VBT stuff inside drm_i915_private\n drm/i915: make SDVO TV-out work for multifunction devices\n drm/i915: rip out now unused is_foo tracking from crtc code\n drm/i915: rip out TV-out lore ...\n drm/i915: drop TVclock special casing on ilk+\n drm/i915: move sdvo TV clock computation to intel_sdvo.c\n ...\n" }, { "commit": "039735369c8fb105d0a090c949b7f894425121d8", "tree": "610d3316c98abdb35bf3e0113784e3364793fa55", "parents": [ "edbe1581c5f94f7fba39cd9a5b2facd624aab661" ], "author": { "name": "Ville Syrjälä", "email": "ville.syrjala@linux.intel.com", "time": "Wed May 08 17:16:45 2013 +0300" }, "committer": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Thu May 23 12:51:32 2013 +0200" }, "message": "drm: Fix drm_rect documentation\n\nThe \u0027struct\u0027 keyword was missing so struct drm_rect documentation never\nended up in the generated docs.\n\nAlso move the drm_rect documentations to a new section alognside the\nvarious helper functions and add a short description about the intended\npurpose of drm_rect.\n\nv2: Move to new section and add general description\n\nSigned-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n" }, { "commit": "0ddaa974258a4cde9a06c079bfd7863644e10d31", "tree": "d0b70cb3f82399cb0c48e3c74d3b76639c624290", "parents": [ "007b703b3ed74e9af9c0576e7698ccda0170d370", "cc3e06a57d4314ca0582fcf7d6b56dea5ca11f77" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue May 21 08:50:57 2013 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue May 21 08:50:57 2013 -0700" }, "message": "Merge branch \u0027drm-radeon-sun-hainan\u0027 of git://people.freedesktop.org/~airlied/linux\n\nPull radeon sun/hainan support from Dave Airlie:\n \"Since I know its outside the merge window, but since this is new hw I\n thought I\u0027d try and provoke the new hw exception, it just fills in the\n blanks in the driver for the new AMD sun and hainan chipsets.\"\n\n* \u0027drm-radeon-sun-hainan\u0027 of git://people.freedesktop.org/~airlied/linux:\n drm/radeon: add Hainan pci ids\n drm/radeon: add golden register settings for Hainan (v2)\n drm/radeon: sun/hainan chips do not have UVD (v2)\n drm/radeon: track which asics have UVD\n drm/radeon: radeon-asic updates for Hainan\n drm/radeon: fill in ucode loading support for Hainan\n drm/radeon: don\u0027t touch DCE or VGA regs on Hainan (v3)\n drm/radeon: fill in GPU init for Hainan (v2)\n drm/radeon: add chip family for Hainan\n" }, { "commit": "e1b73cba13a0cc68dd4f746eced15bd6bb24cda4", "tree": "b1c9e10730724024a700031ad56c20419dabb500", "parents": [ "98304ad186296dc1e655399e28d5973c21db6a73", "c7788792a5e7b0d5d7f96d0766b4cb6112d47d75" ], "author": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Tue May 21 09:52:16 2013 +0200" }, "committer": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Tue May 21 09:52:16 2013 +0200" }, "message": "Merge tag \u0027v3.10-rc2\u0027 into drm-intel-next-queued\n\nBackmerge Linux 3.10-rc2 since the various (rather trivial) conflicts\ngrew a bit out of hand. intel_dp.c has the only real functional\nconflict since the logic changed while dev_priv-\u003eedp.bpp was moved\naround.\n\nAlso squash in a whitespace fixup from Ben Widawsky for\ni915_gem_gtt.c, git seems to do something pretty strange in there\n(which I don\u0027t fully understand tbh).\n\nConflicts:\n\tdrivers/gpu/drm/i915/i915_reg.h\n\tdrivers/gpu/drm/i915/intel_dp.c\n\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n" }, { "commit": "2a0f90551a76a626fd6a606877c8635f52f066e6", "tree": "9e873aa6e1fdd4dbb72211ba259462c1262352f9", "parents": [ "fffbdda4eee69f99b8c798d8eaca91c7e0513f08" ], "author": { "name": "Alex Deucher", "email": "alexander.deucher@amd.com", "time": "Fri Jul 27 18:48:52 2012 -0400" }, "committer": { "name": "Alex Deucher", "email": "alexander.deucher@amd.com", "time": "Mon May 20 12:09:39 2013 -0400" }, "message": "drm/radeon: add Hainan pci ids\n\nSigned-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\nReviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n" }, { "commit": "ec50f2a97a4a7098a81b40030e0bfe28bdc43740", "tree": "3b02c10ed8578fafd2d07156ecd7896c923aee01", "parents": [ "d5fe85af85367d5892e4308f596de4e2a5fb9017", "e9ced8e040ebe40e9953db90acbe7d0b58702ebb" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 16 19:01:46 2013 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 16 19:01:46 2013 -0700" }, "message": "Merge branch \u0027drm-next\u0027 of git://people.freedesktop.org/~airlied/linux\n\nPull drm fixes from Dave Airlie:\n \"Fix for radeon nomodeset regression, old radeon interface cliprects\n fix, 2 qxl crasher fixes, and a couple of minor cleanups.\n\n I may have a new AMD hw support branch next week, its one of those\n doesn\u0027t affect anything existing just adds new support, I\u0027ll see how\n it shapes up and I might ask you to take it, just thought I\u0027d warn in\n advance.\"\n\n* \u0027drm-next\u0027 of git://people.freedesktop.org/~airlied/linux:\n drm/radeon: restore nomodeset operation (v2)\n qxl: fix bug with object eviction and update area\n drm/qxl: drop active_user_framebuffer as its unneeded\n qxl: drop unused variable.\n drm/qxl: fix ioport interactions for kernel submitted commands.\n drm: remove unused wrapper macros\n drm/radeon: check incoming cliprects pointer\n" }, { "commit": "fea0f9ff56258734ff1c49008c36ef6904b8c541", "tree": "0c0c383ce253177180e2ade24894ea900cfd5102", "parents": [ "aef2ea912e0ee39fbca74fed0ab8e612ac2c3ba2", "9f1d036648c1c5ed81b0e98d7a06d55df972701e" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon May 13 07:59:59 2013 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon May 13 07:59:59 2013 -0700" }, "message": "Merge branch \u0027drm-next\u0027 of git://people.freedesktop.org/~airlied/linux\n\nPull drm fixes from Dave Airlie:\n \"Just a few straggling fixes I hoovered up, and an intel fixes pull\n from Daniel which fixes some regressions, and some mgag200 fixes from\n Matrox.\"\n\n* \u0027drm-next\u0027 of git://people.freedesktop.org/~airlied/linux:\n drm/mgag200: Fix framebuffer base address programming\n drm/mgag200: Convert counter delays to jiffies\n drm/mgag200: Fix writes into MGA1064_PIX_CLK_CTL register\n drm/mgag200: Don\u0027t change unrelated registers during modeset\n drm: Only print a debug message when the polled connector has changed\n drm: Make the HPD status updates debug logs more readable\n drm: Use names of ioctls in debug traces\n drm: Remove pointless \u0027-\u0027 characters from drm_fb_helper documentation\n drm: Add kernel-doc for drm_fb_helper_funcs-\u003einitial_config\n drm: refactor call to request_module\n drm: Don\u0027t prune modes loudly when a connector is disconnected\n drm: Add missing break in the command line mode parsing code\n drm/i915: clear the stolen fb before resuming\n Revert \"drm/i915: Calculate correct stolen size for GEN7+\"\n drm/i915: hsw: fix link training for eDP on port-A\n Revert \"drm/i915: revert eDP bpp clamping code changes\"\n drm: don\u0027t check modeset locks in panic handler\n drm/i915: Fix pipe enabled mask for pipe C in WM calculations\n drm/mm: fix dump table BUG\n drm/i915: Always normalize return timeout for wait_timeout_ioctl\n" }, { "commit": "95643359f8b31e74b35901f4e36cd069cd67fd48", "tree": "4ecb510ed4b7d0d5f82d521e57900e8000620063", "parents": [ "fefaedcfb82d2e57c2320acf60604ab03b750cc0" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon May 13 15:23:49 2013 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon May 13 15:23:49 2013 +1000" }, "message": "drm: remove unused wrapper macros\n\nWe don\u0027t use these anymore so nuke them.\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "b9434d0f162c351e02e0e8f66ca6b75a67595537", "tree": "607eb8c2b8946b891e96c8699fbab9e3b6e1c556", "parents": [ "7b97936f298af7991ca4717a4c23610d9ecd8927" ], "author": { "name": "Chris Cummins", "email": "christopher.e.cummins@intel.com", "time": "Thu May 09 14:20:40 2013 +0100" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri May 10 14:46:50 2013 +1000" }, "message": "drm: Use names of ioctls in debug traces\n\nThe intention here is to make the output of dmesg with full verbosity a\nbit easier for a human to parse. This commit transforms:\n\n[drm:drm_ioctl], pid\u003d699, cmd\u003d0x6458, nr\u003d0x58, dev 0xe200, auth\u003d1\n[drm:drm_ioctl], pid\u003d699, cmd\u003d0xc010645b, nr\u003d0x5b, dev 0xe200, auth\u003d1\n[drm:drm_ioctl], pid\u003d699, cmd\u003d0xc0106461, nr\u003d0x61, dev 0xe200, auth\u003d1\n[drm:drm_ioctl], pid\u003d699, cmd\u003d0xc01c64ae, nr\u003d0xae, dev 0xe200, auth\u003d1\n[drm:drm_mode_addfb], [FB:32]\n[drm:drm_ioctl], pid\u003d699, cmd\u003d0xc0106464, nr\u003d0x64, dev 0xe200, auth\u003d1\n[drm:drm_vm_open_locked], 0x7fd9302fe000,0x00a00000\n[drm:drm_ioctl], pid\u003d699, cmd\u003d0x400c645f, nr\u003d0x5f, dev 0xe200, auth\u003d1\n[drm:drm_ioctl], pid\u003d699, cmd\u003d0xc00464af, nr\u003d0xaf, dev 0xe200, auth\u003d1\n[drm:intel_crtc_set_config], [CRTC:3] [NOFB]\n\ninto:\n\n[drm:drm_ioctl], pid\u003d699, dev\u003d0xe200, auth\u003d1, I915_GEM_THROTTLE\n[drm:drm_ioctl], pid\u003d699, dev\u003d0xe200, auth\u003d1, I915_GEM_CREATE\n[drm:drm_ioctl], pid\u003d699, dev\u003d0xe200, auth\u003d1, I915_GEM_SET_TILING\n[drm:drm_ioctl], pid\u003d699, dev\u003d0xe200, auth\u003d1, IOCTL_MODE_ADDFB\n[drm:drm_mode_addfb], [FB:32]\n[drm:drm_ioctl], pid\u003d699, dev\u003d0xe200, auth\u003d1, I915_GEM_MMAP_GTT\n[drm:drm_vm_open_locked], 0x7fd9302fe000,0x00a00000\n[drm:drm_ioctl], pid\u003d699, dev\u003d0xe200, auth\u003d1, I915_GEM_SET_DOMAIN\n[drm:drm_ioctl], pid\u003d699, dev\u003d0xe200, auth\u003d1, DRM_IOCTL_MODE_RMFB\n[drm:intel_crtc_set_config], [CRTC:3] [NOFB]\n\nv2: drm_ioctls is now a constant (Ville Syrjälä)\n\nSigned-off-by: Chris Cummins \u003cchristopher.e.cummins@intel.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "7b97936f298af7991ca4717a4c23610d9ecd8927", "tree": "3083e748a78f415e343dbec792a304ce82d163ff", "parents": [ "54afc1214133deddd49162b101971af5711f1b3a" ], "author": { "name": "Ville Syrjälä", "email": "ville.syrjala@linux.intel.com", "time": "Wed May 08 16:38:34 2013 +0300" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri May 10 14:46:11 2013 +1000" }, "message": "drm: Remove pointless \u0027-\u0027 characters from drm_fb_helper documentation\n\nSigned-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "54afc1214133deddd49162b101971af5711f1b3a", "tree": "2c3633d1d3b79887ed61d51a2edefa1a52e4d3a9", "parents": [ "d1fd3ddc469b576e36184f2bd02c0301d80118d3" ], "author": { "name": "Ville Syrjälä", "email": "ville.syrjala@linux.intel.com", "time": "Wed May 08 16:38:33 2013 +0300" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri May 10 14:46:07 2013 +1000" }, "message": "drm: Add kernel-doc for drm_fb_helper_funcs-\u003einitial_config\n\nSigned-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "20a2078ce7705a6e0722ef5184336eb8657a58d8", "tree": "5b927c96516380aa0ecd68d8a609f7cd72120ad5", "parents": [ "0279b3c0ada1d78882f24acf94ac4595bd657a89", "307b9c022720f9de90d58e51743e01e9a42aec59" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 02 19:40:34 2013 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 02 19:40:34 2013 -0700" }, "message": "Merge branch \u0027drm-next\u0027 of git://people.freedesktop.org/~airlied/linux\n\nPull drm updates from Dave Airlie:\n \"This is the main drm pull request for 3.10.\n\n Wierd bits:\n - OMAP drm changes required OMAP dss changes, in drivers/video, so I\n took them in here.\n - one more fbcon fix for font handover\n - VT switch avoidance in pm code\n - scatterlist helpers for gpu drivers - have acks from akpm\n\n Highlights:\n - qxl kms driver - driver for the spice qxl virtual GPU\n\n Nouveau:\n - fermi/kepler VRAM compression\n - GK110/nvf0 modesetting support.\n\n Tegra:\n - host1x core merged with 2D engine support\n\n i915:\n - vt switchless resume\n - more valleyview support\n - vblank fixes\n - modesetting pipe config rework\n\n radeon:\n - UVD engine support\n - SI chip tiling support\n - GPU registers initialisation from golden values.\n\n exynos:\n - device tree changes\n - fimc block support\n\n Otherwise:\n - bunches of fixes all over the place.\"\n\n* \u0027drm-next\u0027 of git://people.freedesktop.org/~airlied/linux: (513 commits)\n qxl: update to new idr interfaces.\n drm/nouveau: fix build with nv50-\u003envc0\n drm/radeon: fix handling of v6 power tables\n drm/radeon: clarify family checks in pm table parsing\n drm/radeon: consolidate UVD clock programming\n drm/radeon: fix UPLL_REF_DIV_MASK definition\n radeon: add bo tracking debugfs\n drm/radeon: add new richland pci ids\n drm/radeon: add some new SI PCI ids\n drm/radeon: fix scratch reg handling for UVD fence\n drm/radeon: allocate SA bo in the requested domain\n drm/radeon: fix possible segfault when parsing pm tables\n drm/radeon: fix endian bugs in atom_allocate_fb_scratch()\n OMAPDSS: TFP410: return EPROBE_DEFER if the i2c adapter not found\n OMAPDSS: VENC: Add error handling for venc_probe_pdata\n OMAPDSS: HDMI: Add error handling for hdmi_probe_pdata\n OMAPDSS: RFBI: Add error handling for rfbi_probe_pdata\n OMAPDSS: DSI: Add error handling for dsi_probe_pdata\n OMAPDSS: SDI: Add error handling for sdi_probe_pdata\n OMAPDSS: DPI: Add error handling for dpi_probe_pdata\n ...\n" }, { "commit": "62d1f92e06aef9665d71ca7e986b3047ecf0b3c7", "tree": "d6c4e05d6196451f6f8540f1956c1e5274298ac0", "parents": [ "18932a28419596bc9403770f5d8a108c5433fe59" ], "author": { "name": "Alex Deucher", "email": "alexander.deucher@amd.com", "time": "Thu Apr 25 14:06:05 2013 -0400" }, "committer": { "name": "Alex Deucher", "email": "alexander.deucher@amd.com", "time": "Thu May 02 10:01:49 2013 -0400" }, "message": "drm/radeon: add new richland pci ids\n\nSigned-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\nCc: stable@vger.kernel.org\n" }, { "commit": "18932a28419596bc9403770f5d8a108c5433fe59", "tree": "8a38e47d6b12979e7856fc2b9aaaed2434131b06", "parents": [ "581bc3a9f628dad6221536d9f37af0472c66bb92" ], "author": { "name": "Alex Deucher", "email": "alexander.deucher@amd.com", "time": "Thu Apr 25 13:55:15 2013 -0400" }, "committer": { "name": "Alex Deucher", "email": "alexander.deucher@amd.com", "time": "Thu May 02 10:01:48 2013 -0400" }, "message": "drm/radeon: add some new SI PCI ids\n\nSigned-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\nCc: stable@vger.kernel.org\n" }, { "commit": "b63e6aa5028dfde3b360945564290de28b47c2d7", "tree": "1761e4b0f59254bb30602d9690c2431c5a7f5b36", "parents": [ "ce089b5472f7d0321bcb2cbc22d85bac15e4778b" ], "author": { "name": "David Howells", "email": "dhowells@redhat.com", "time": "Fri Apr 12 15:34:31 2013 +0100" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Wed May 01 17:29:44 2013 -0400" }, "message": "drm: proc: Use minor-\u003eindex to label things, not PDE-\u003ename\n\nUse minor-\u003eindex to label things, not the name field from the proc_dir_entry\nof the /proc/dwm/\u003cminor\u003e/ directory.\n\nAlso, use \"%u\" not \"%d\" to render the value and use a 12-byte buffer in which\nto render the integer, not a 16-byte buffer. The longest string an unsigned\nint can give you is 10 chars (4294967295) plus a NUL, so round up to 12 as the\nstack is likely to be 4- or 8-byte aligned.\n\nSigned-off-by: David Howells \u003cdhowells@redhat.com\u003e\ncc: dri-devel@lists.freedesktop.org\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "ce089b5472f7d0321bcb2cbc22d85bac15e4778b", "tree": "43af07022796d28494460ffb5d796a811bf0a245", "parents": [ "0c2f343825192044ffaf542854d478e00f7f0a59" ], "author": { "name": "David Howells", "email": "dhowells@redhat.com", "time": "Fri Apr 12 15:23:25 2013 +0100" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Wed May 01 17:29:43 2013 -0400" }, "message": "drm: Constify drm_proc_list[]\n\nConstify drm_proc_list[] and related pointers.\n\nSigned-off-by: David Howells \u003cdhowells@redhat.com\u003e\ncc: dri-devel@lists.freedesktop.org\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "219b47339ced80ca580bb6ce7d1636166984afa7", "tree": "1646e5e3b5998a5b13f71c8f06b7a28a0163fbe7", "parents": [ "c55b6b3da25aa3af36ec51a13a4ed15fef0d7a73" ], "author": { "name": "Dave Airlie", "email": "airlied@gmail.com", "time": "Mon Apr 22 09:54:36 2013 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Wed May 01 09:30:15 2013 +1000" }, "message": "drm/prime: keep a reference from the handle to exported dma-buf (v6)\n\nCurrently we have a problem with this:\n1. i915: create gem object\n2. i915: export gem object to prime\n3. radeon: import gem object\n4. close prime fd\n5. radeon: unref object\n6. i915: unref object\n\ni915 has an imported object reference in its file priv, that isn\u0027t\ncleaned up properly until fd close. The reference gets added at step 2,\nbut at step 6 we don\u0027t have enough info to clean it up.\n\nThe solution is to take a reference on the dma-buf when we export it,\nand drop the reference when the gem handle goes away.\n\nSo when we export a dma_buf from a gem object, we keep track of it\nwith the handle, we take a reference to the dma_buf. When we close\nthe handle (i.e. userspace is finished with the buffer), we drop\nthe reference to the dma_buf, and it gets collected.\n\nThis patch isn\u0027t meant to fix any other problem or bikesheds, and it doesn\u0027t\nfix any races with other scenarios.\n\nv1.1: move export symbol line back up.\n\nv2: okay I had to do a bit more, as the first patch showed a leak\non one of my tests, that I found using the dma-buf debugfs support,\nthe problem case is exporting a buffer twice with the same handle,\nwe\u0027d add another export handle for it unnecessarily, however\nwe now fail if we try to export the same object with a different gem handle,\nhowever I\u0027m not sure if that is a case I want to support, and I\u0027ve\ngotten the code to WARN_ON if we hit something like that.\n\nv2.1: rebase this patch, write better commit msg.\nv3: cleanup error handling, track import vs export in linked list,\nthese two patches were separate previously, but seem to work better\nlike this.\nv4: danvet is correct, this code is no longer useful, since the buffer\nbetter exist, so remove it.\nv5: always take a reference to the dma buf object, import or export.\n(Imre Deak contributed this originally)\nv6: square the circle, remove import vs export tracking now\nthat there is no difference\n\nReviewed-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nCc: stable@vger.kernel.org\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "0894c96bff762d0474a8722bba3d420f643db359", "tree": "c29be8c57296ccd5d3be5f06788236412607e859", "parents": [ "e7272df342ba337e87e210470bb93d97d192f2e0" ], "author": { "name": "Ville Syrjälä", "email": "ville.syrjala@linux.intel.com", "time": "Wed Apr 24 18:52:37 2013 +0300" }, "committer": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Tue Apr 30 22:20:05 2013 +0200" }, "message": "drm: Add drm_rect_equals()\n\ndrm_rect_equals() tells whether two drm_rects are equal.\n\nReviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nSigned-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nAcked-by: Dave Airlie \u003cairlied@gmail.com\u003e\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n" }, { "commit": "e7272df342ba337e87e210470bb93d97d192f2e0", "tree": "19fd9d002d434a0517ef84c8f778ccbb4a5b3872", "parents": [ "4954c4282f6b945f1dd5716f92b594a07fa4ffe3" ], "author": { "name": "Ville Syrjälä", "email": "ville.syrjala@linux.intel.com", "time": "Wed Apr 24 18:52:36 2013 +0300" }, "committer": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Tue Apr 30 22:20:00 2013 +0200" }, "message": "drm: Add drm_rect_debug_print()\n\nAdd a debug function to print the rectangle in a human readable format.\n\nv2: Renamed drm_region to drm_rect, the function from drm_region_debug\n to drm_rect_debug_print(), and use %+d instead of +%d in the format.\nv3: Use %d format for width/height in the non fixed point case as well\n\nReviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nSigned-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nAcked-by: Dave Airlie \u003cairlied@gmail.com\u003e\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n" }, { "commit": "4954c4282f6b945f1dd5716f92b594a07fa4ffe3", "tree": "3c72d5bd093e1c066cd1ef8c0f9608231ad4d44f", "parents": [ "3512f976d252bd5d07d04e9e157f0cd210c959a0" ], "author": { "name": "Ville Syrjälä", "email": "ville.syrjala@linux.intel.com", "time": "Wed Apr 24 18:52:35 2013 +0300" }, "committer": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Tue Apr 30 22:19:39 2013 +0200" }, "message": "drm: Add drm_rect_calc_{hscale, vscale}() utility functions\n\nThese functions calculate the scaling factor based on the source and\ndestination rectangles.\n\nThere are two version of the functions, the strict ones that will\nreturn an error if the min/max scaling factor is exceeded, and the\nrelaxed versions that will adjust the src/dst rectangles in order to\nkeep the scaling factor withing the limits.\n\nv2: Return error instead of adjusting regions, refactor common parts\n into one function, and split into strict and relaxed versions.\nv3: Renamed drm_region to drm_rect, add \"_rect_\" to the function\n names.\nv4: Fix \"calculcate\" typos\n\nReviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nSigned-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nAcked-by: Dave Airlie \u003cairlied@gmail.com\u003e\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n" }, { "commit": "3512f976d252bd5d07d04e9e157f0cd210c959a0", "tree": "5541eb3fee53099e1ea416c959cbe1890ba93aa2", "parents": [ "5a80c45c5297a025c2615624042ea8b6840a5376" ], "author": { "name": "Ville Syrjälä", "email": "ville.syrjala@linux.intel.com", "time": "Wed Apr 24 18:52:34 2013 +0300" }, "committer": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Tue Apr 30 22:19:08 2013 +0200" }, "message": "drm: Add struct drm_rect and assorted utility functions\n\nstruct drm_rect represents a simple rectangle. The utility\nfunctions are there to help driver writers.\n\nv2: Moved the region stuff into its own file, made the smaller funcs\n static inline, used 64bit maths in the scaled clipping function to\n avoid overflows (instead it will saturate to INT_MIN or INT_MAX).\nv3: Renamed drm_region to drm_rect, drm_region_clip to\n drm_rect_intersect, and drm_region_subsample to drm_rect_downscale.\nv4: Renamed some function parameters, improve kernel-doc comments a bit,\n and actually generate documentation for drm_rect.[ch].\nv5: s/RETUTRNS/RETURNS/\n\nReviewed-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\nReviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nSigned-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nAcked-by: Dave Airlie \u003cairlied@gmail.com\u003e\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n" }, { "commit": "c55b6b3da25aa3af36ec51a13a4ed15fef0d7a73", "tree": "240e093507676748e5cd1b52433d7e10523b5636", "parents": [ "ea9cbb063ce7509d98febdc756e77d6b69a10b56" ], "author": { "name": "Ville Syrjälä", "email": "ville.syrjala@linux.intel.com", "time": "Fri Apr 26 17:40:28 2013 +0300" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Tue Apr 30 10:03:07 2013 +1000" }, "message": "drm: Kill user_modes list and the associated ioctls\n\nThere is no way to use modes added to the user_modes list. We never\nlook at the contents of said list in the kernel, and the only operations\nuserspace can do are attach and detach. So the only \"benefit\" of this\ninterface is wasting kernel memory.\n\nFortunately it seems no real user space application ever used these\nioctls. So just kill them.\n\nAlso remove the prototypes for the non-existing drm_mode_addmode_ioctl()\nand drm_mode_rmmode_ioctl() functions.\n\nv2: Use drm_noop instead of completely removing the ioctls\n\nSigned-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "ea9cbb063ce7509d98febdc756e77d6b69a10b56", "tree": "06cef9ab6d943315d63f8122278409c7eaa57146", "parents": [ "7d05336b0c54cd5921b748e306ee0513b620dedf" ], "author": { "name": "Ville Syrjälä", "email": "ville.syrjala@linux.intel.com", "time": "Thu Apr 25 20:09:20 2013 +0300" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Tue Apr 30 10:02:25 2013 +1000" }, "message": "drm: Silence some sparse warnings\n\ndrivers/gpu/drm/drm_pci.c:155:5: warning: symbol \u0027drm_pci_set_busid\u0027 was not declared. Should it be static?\ndrivers/gpu/drm/drm_pci.c:197:5: warning: symbol \u0027drm_pci_set_unique\u0027 was not declared. Should it be static?\ndrivers/gpu/drm/drm_pci.c:269:5: warning: symbol \u0027drm_pci_agp_init\u0027 was not declared. Should it be static?\n\ndrivers/gpu/drm/drm_crtc.c:181:1: warning: symbol \u0027drm_get_dirty_info_name\u0027 was not declared. Should it be static?\ndrivers/gpu/drm/drm_crtc.c:1123:5: warning: symbol \u0027drm_mode_group_init\u0027 was not declared. Should it be static?\n\ndrivers/gpu/drm/drm_modes.c:918:6: warning: symbol \u0027drm_mode_validate_clocks\u0027 was not declared. Should it be static?\n\nSigned-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "8cc3f23caf4066880f54f636beb3e30b4d383eca", "tree": "2c66a755d1f2c17d6c774aec84b277f632fc82a8", "parents": [ "67af116e4ac0ebbb3d7849a1d3e2344e289dd76d" ], "author": { "name": "Ville Syrjälä", "email": "ville.syrjala@linux.intel.com", "time": "Wed Apr 24 19:07:16 2013 +0300" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Apr 26 10:25:09 2013 +1000" }, "message": "drm: Add drm_mode_equal_no_clocks()\n\ndrm_mode_equal_no_clocks() is like drm_mode_equal() except it doesn\u0027t\ncompare the clock or vrefresh values. drm_mode_equal() is now\nimplemented by first doing the clock checks, and then calling\ndrm_mode_equal_no_clocks().\n\nv2: Add missing EXPORT_SYMBOL()\n\nReviewed-by: Paulo Zanoni \u003cpaulo.r.zanoni@intel.com\u003e\nSigned-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "67af116e4ac0ebbb3d7849a1d3e2344e289dd76d", "tree": "2eff3e3c37c65493f6fd0c2b3e3eee70d3086a45", "parents": [ "17f0efc4f463e5b74026863521cbd47fb93f5686" ], "author": { "name": "Ville Syrjälä", "email": "ville.syrjala@linux.intel.com", "time": "Wed Apr 24 19:07:15 2013 +0300" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Apr 26 10:25:06 2013 +1000" }, "message": "drm: Remove explicit vrefresh initialization from DRM_MODE()\n\nNo need to zero initialize .vrefresh in DRM_MODE() since it\u0027s using\ndesgignated initializers.\n\nThis will also avoid some duplicate initialization warnings later.\n\nReviewed-by: Paulo Zanoni \u003cpaulo.r.zanoni@intel.com\u003e\nSigned-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "fe214163cc6e6b33253b7ac14bc3dd616e76b513", "tree": "32c17795c76da99c458c61f71127d4897f82deb3", "parents": [ "205996c0776a229052c33688ee58a605a021740d" ], "author": { "name": "Rafał Miłecki", "email": "zajec5@gmail.com", "time": "Fri Apr 19 19:01:25 2013 +0200" }, "committer": { "name": "Alex Deucher", "email": "alexander.deucher@amd.com", "time": "Tue Apr 23 18:03:58 2013 -0400" }, "message": "drm: add drm_edid_to_eld helper extracting SADs from EDID (v2)\n\nSome devices (ATI/AMD cards) don\u0027t support passing ELD struct to the\nhardware but just require filling specific registers and then the\nhardware/firmware does the rest. In such cases we need to read the info\nfrom SAD blocks and put them in the correct registers.\n\nagd5f: note that the returned pointer needs to be kfreed as per\nChristian\u0027s suggestion.\n\nv2: fix warning\n\nSigned-off-by: Rafał Miłecki \u003czajec5@gmail.com\u003e\nReviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\nSigned-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n" }, { "commit": "28184f22178ec9cfadd72e37e99921fdefa95f89", "tree": "57f43b9701292c06d5356240408ead77b83f9c0f", "parents": [ "9131d3d87bf7c8cfc9bdc3d3d927ec3967f0e9a6", "bae3699182027525d92b97d904578a533264b242" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Tue Apr 16 13:03:47 2013 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Tue Apr 16 13:03:47 2013 +1000" }, "message": "Merge tag \u0027drm-intel-next-2013-04-06\u0027 of git://people.freedesktop.org/~danvet/drm-intel into drm-next\n\nDaniel writes:\nSince I expect Linus to open the merge window in about a week I guess this\nis the last i915 feature pull for 3.10. Highlights:\nUpdated testing tree for -next. Highlights:\n- Corner case fixes discovered with static analyzers (Damien).\n- More fixes to combat unclaimed register errors on Haswell (Paulo).\n- Some small fixes to the gpu turbo code (Rodrigo+Ben), Ben has more\n fixes for overclocking support pending.\n- More prep work for fastboot from Chris.\n- VT-switchless suspend/resume from Jesse.\n- The prep work of Egbert Eich\u0027s hpd irq storm handling. Hopefully we can\n squeeze in the actual storm handling code for 3.10 ...\n- More convenience helpers for Imre\u0027s sg iterator. Core parts acked by\n Andrew Morton.\n- A bit of backlight code cleanup from Jani.\n- Fixed ilk gpu reset (Jesse).\n- Reduced color range handling fixes for VLV (Ville).\n\nThe big item here is though the introduction of pipe_config to properly\npre-compute the desired modeset state before touching the hw. Together\nwith some very basic support to read out the current config from the hw\nand compare the state with the sw tracking. This is all prep work for more\nreliable fastboot, atomic modesets and other cool features. Stuff\nconverted to the new world includes:\n- Most simple pipe attributes (reduce color range, pixel multiplier).\n- Pipe bpp/dither handling.\n- Some convenience flags like -\u003ehas_pch_encoder to simplify the code flow.\n- (Almost) DP clock handling, had to be reverted since part of a prep\n patch was lost in rebasing ...\nExpect a lot of patches for this throughout 3.11, there\u0027s tons of work\ntill we have all state properly tracked for fastbooting to woExpect a lot\nof patches for this throughout 3.11, there\u0027s tons of work till we have all\nstate properly tracked for fastbooting to work.\n\nFor 3.10 I have a bunch of fixes queued up and I plan to send them all out\nat the end of this week. I need to shuffle patches in my -next queue a bit\nso that we don\u0027t but feature-y stuff in there, too. The main thing I\u0027d\nlike to sneak in is Egbert\u0027s hpd irq storm handling, which should be\npretty low-risk since all the infrastructure work has landed already. I\nalso have the oops fix pending, but that only mustered review before the\nw/e and giving how hairy that part of our modeset code is, I want to give\nit some more testing before forwarding.\n\nNote: annarchy.fd.o seems to run out of disk space, so couldn\u0027t push the\nusual for-airlied branch. Tag should work though.\n\nNote 2: I\u0027ve had to do a backmerge since conflicts grew too ugly, but the\nupstream -rc I\u0027ve backmerged is already in your drm-next.\n\n* tag \u0027drm-intel-next-2013-04-06\u0027 of git://people.freedesktop.org/~danvet/drm-intel: (75 commits)\n drm/i915: info level for simulated gpu hang dmesg notice\n drm/i915: revert eDP bpp clamping code changes\n Revert \"drm/i915: fix DP get_hw_state return value\"\n drm/i915: Don\u0027t use the HDMI port color range bit on Valleyview\n drm/i915: Set PIPECONF color range bit on Valleyview\n drm/i915: extract i9xx_set_pipeconf\n drm/i915: Add no-lvds quirk for Fujitsu Esprimo Q900\n drm/i915: create pipe_config-\u003edpll for clock state\n drm/i915: hw readout support for -\u003ehas_pch_encoders\n drm/i915: add hw state readout/checking for pipe_config\n drm/i915: rip out superflous is_dp\u0026is_cpu_edp tracking\n drm/i915: remove leaky eDP functions\n drm/i915: track dp target_clock in pipe_config\n drm/i915: move dp_m_n computation to dp_encoder-\u003ecompute_config\n drm/i915: clear up the fdi/dp set_m_n confusion\n drm/i915: Fix sdvo connector get_hw_state function\n drm/i915: drop DPFLIPSTAT enables on VLV v3\n drm/i915: add Punit read/write routines for VLV v2\n drm/i915: panel power sequencing for VLV eDP v2\n drm/i915/dp: fix up VLV DP handling v2\n ...\n" }, { "commit": "715f59cc2391cb4e28a4d464a0564a60a00ecca4", "tree": "958f4cd1dbb9600a80bf5c5e1b00c7911f80e5ad", "parents": [ "8091ee5c092a1e7dab2cb49d02d5f3925ece3995" ], "author": { "name": "Christopher Harvey", "email": "charvey@matrox.com", "time": "Fri Apr 05 15:28:32 2013 +0000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Apr 12 14:07:20 2013 +1000" }, "message": "drm: Misc comment cleanup\n\nSigned-off-by: Christopher Harvey \u003ccharvey@matrox.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "afe6804c045fbd69a1b75c681107b5d6df9190de", "tree": "d102a41ad9bd8ace8edd9b6c2b380fca3d707499", "parents": [ "97d103311aab2c652ef37bc3cd157e701057e535" ], "author": { "name": "Dave Airlie", "email": "airlied@gmail.com", "time": "Tue Jan 22 13:56:04 2013 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Apr 12 13:50:46 2013 +1000" }, "message": "ttm: export functions to allow qxl do its own iomapping\n\nqxl wants to use io mapping like i915 gem does, for now\njust export the symbols so the driver can implement atomic\npage maps using io mapping.\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "97d103311aab2c652ef37bc3cd157e701057e535", "tree": "0be73f5193a1beb9a0938b3b6857f3c8cd9a651a", "parents": [ "399403c7cebb4dcd42491e511e1ec91442ecaf80" ], "author": { "name": "Alon Levy", "email": "alevy@redhat.com", "time": "Sun Sep 09 10:47:32 2012 +0300" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Apr 12 13:50:02 2013 +1000" }, "message": "include/drm/drm_fb_helper.h: remove left over function declaration\n\nSigned-off-by: Alon Levy \u003calevy@redhat.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "ecb135a1a1953d2895d149e78926be479fdc6f2c", "tree": "e40aa30eefa5cc42ed1169b9798810a710a1bd9a", "parents": [ "bd6946e87a98fea11907b2a47368e13044458a35", "07961ac7c0ee8b546658717034fe692fd12eefa9" ], "author": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Wed Apr 03 11:25:32 2013 +0200" }, "committer": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Wed Apr 03 11:28:48 2013 +0200" }, "message": "Merge tag \u0027v3.9-rc5\u0027 into drm-intel-next-queued\n\nBackmerge Linux 3.9-rc5 since I want to merge a few dp clock cleanups\nfor -next, but they will conflict all over the place with\n\ncommit 9d1a455b0ca1c2c956b4d9ab212864a8695270f1\nAuthor: Takashi Iwai \u003ctiwai@suse.de\u003e\nDate: Mon Mar 18 11:25:36 2013 +0100\n\n drm/i915: Use the fixed pixel clock for eDP in intel_dp_set_m_n()\n\nfrom -fixes.\n\nConflicts:\n\tdrivers/gpu/drm/i915/intel_dp.c: Simply adjacent lines changed.\n\tdrivers/gpu/drm/i915/intel_panel.c: A field rename in -next\n\tconflicts with a bugfix in -fixes. Take the version from\n\t-fixes and apply the rename.\n\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n" }, { "commit": "11e17a0873aba66b251e77ff246590b9390cdc68", "tree": "2f409245edf3d4e102ac9977c95e438c5582f2dc", "parents": [ "866d12b4ee5faefbc1a98b787ac4242a2f04b051" ], "author": { "name": "Jesse Barnes", "email": "jbarnes@virtuousgeek.org", "time": "Tue Feb 19 13:31:39 2013 -0800" }, "committer": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Tue Mar 26 21:32:53 2013 +0100" }, "message": "drm: add initial_config function to fb helper\n\nRather than building a config which may or may not work, let the driver\nbuild an initial fb config. This allows the driver to use the BIOS boot\nconfiguration for example, displaying kernel messages and the initial fb\nconsole on the same outputs the BIOS lit up at boot time. If that\nfails, the driver can still fall back the same way as the core.\n\nSigned-off-by: Jesse Barnes \u003cjbarnes@virtuousgeek.org\u003e\nSigned-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nCc: dri-devel@lists.freedesktop.org\nAcked-by: Dave Airlie \u003cairlied@gmail.com\u003e\nReviewed-by: Imre Deak \u003cimre.deak@intel.com\u003e\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n" }, { "commit": "b75bbaa038ffc426e88ea3df6c4ae11834fc3e4f", "tree": "edd59a061b1b0619dd74811e0a95010a142c2479", "parents": [ "e4d170633fde379f39a90f8a5e7eb619b5d1144d" ], "author": { "name": "Alex Deucher", "email": "alexander.deucher@amd.com", "time": "Fri Mar 08 13:36:54 2013 -0500" }, "committer": { "name": "Alex Deucher", "email": "alexander.deucher@amd.com", "time": "Fri Mar 15 18:47:19 2013 -0400" }, "message": "drm/radeon: add Richland pci ids\n\nReviewed-by: Jerome Glisse \u003cjglisse@redhat.com\u003e\nSigned-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\nCc: stable@vger.kernel.org\n" }, { "commit": "13bcf01b33e6e19a7fe7ff396f9ed02803e225ec", "tree": "a00c00316a83d6078e1b4dad1aee7c7baf8a331f", "parents": [ "ce495960ff33f96362cf81f0eb7c52d1a89f64be" ], "author": { "name": "Christopher Harvey", "email": "charvey@matrox.com", "time": "Thu Mar 07 10:42:25 2013 -0500" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Mar 08 08:32:23 2013 +1000" }, "message": "drm: Documentation typo fixes\n\nSigned-off-by: Christopher Harvey \u003ccharvey@matrox.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "10a8512008655d5ce62f8c56323a6b5bd221c920", "tree": "a61d19544363a8acbc0792e96e1fa5adac6dff7a", "parents": [ "f142d3bd556c5e82e9bb3d33d07d6708702ea4ce" ], "author": { "name": "Thierry Reding", "email": "thierry.reding@avionic-design.de", "time": "Wed Nov 21 15:31:35 2012 +0100" }, "committer": { "name": "Thierry Reding", "email": "thierry.reding@avionic-design.de", "time": "Fri Feb 22 08:20:10 2013 +0100" }, "message": "drm: Add HDMI infoframe helpers\n\nAdd a generic helper to fill in an HDMI AVI infoframe with data\nextracted from a DRM display mode.\n\nSigned-off-by: Thierry Reding \u003cthierry.reding@avionic-design.de\u003e\nReviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n" }, { "commit": "595887eb90b7e74a3e276b8f17d35236d85b0d37", "tree": "0c06f44d572ef36abb9b69bc19ed2dc2dea2fa7c", "parents": [ "a6b21831772db4088bca770dc99697356e82a7e2" ], "author": { "name": "Thierry Reding", "email": "thierry.reding@avionic-design.de", "time": "Wed Nov 21 15:00:47 2012 +0100" }, "committer": { "name": "Thierry Reding", "email": "thierry.reding@avionic-design.de", "time": "Fri Feb 22 08:20:03 2013 +0100" }, "message": "drm: Add some missing forward declarations\n\nThe drm_file and drm_clip_rect structures are used throughout the file\nbut they are never declared nor pulled in through an include. Add\nforward declarations to make them available.\n\nSigned-off-by: Thierry Reding \u003cthierry.reding@avionic-design.de\u003e\nReviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n" }, { "commit": "18316c8c39a85c8b6e3db0a150b1bee5b6c4c053", "tree": "10fdeb50456771b21f5d8f9102b015c29e8b8a58", "parents": [ "c976cb37a95a8dcaf23d04d6487fbacc33d3c913" ], "author": { "name": "Thierry Reding", "email": "thierry.reding@avionic-design.de", "time": "Thu Dec 20 15:41:44 2012 +0100" }, "committer": { "name": "Thierry Reding", "email": "thierry.reding@avionic-design.de", "time": "Fri Feb 22 08:19:53 2013 +0100" }, "message": "drm: Remove duplicate drm_mode_cea_vic()\n\nThe same function had already been merged with a different name. Remove\nthe duplicate one but reuse some of its kerneldoc fragments for the\nexisting implementation.\n\nSigned-off-by: Thierry Reding \u003cthierry.reding@avionic-design.de\u003e\nReviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n" }, { "commit": "ca18e1426bb2db987b67030256477c9571aebd09", "tree": "72e7cbe5f00f90390d00f5d3d4c9bdec79e48bbe", "parents": [ "74e1697478ffdee0e12e48db024a9b3677fd8cee", "0d4bbaf9f3e5b9f52150ddc5a4ee8b0ab83a440b" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Feb 21 09:31:47 2013 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Feb 21 09:31:47 2013 +1000" }, "message": "Merge branch \u0027tilcdc-next\u0027 of git://people.freedesktop.org/~robclark/linux into drm-next\n\nKMS driver for TI LCD controller\n\n* \u0027tilcdc-next\u0027 of git://people.freedesktop.org/~robclark/linux:\n drm/tilcdc: add support for LCD panels (v5)\n drm/tilcdc: add encoder slave (v2)\n drm/i2c: nxp-tda998x (v3)\n drm/tilcdc: add TI LCD Controller DRM driver (v4)\n drm/nouveau: use i2c encoder helper wrappers\n drm: i2c encoder helper wrappers\n drm/cma: add debugfs helpers\n drm: small fix in drm_send_vblank_event()\n drm: Don\u0027t set the plane-\u003efb to NULL on successfull set_plane\n drm/cma-helper: fixup compilation\n\nConflicts:\n\tdrivers/gpu/drm/Kconfig\n\tdrivers/gpu/drm/Makefile\n\tdrivers/gpu/drm/drm_fb_cma_helper.c\n" }, { "commit": "b81e059ec5a7128622ab5d74d78e9b4f361b54ae", "tree": "65344a0adc351fd0a14a14d9fa50e93b69a5bb6e", "parents": [ "35f8badc1cf652381fa3f82c1fbea39f4dbe87fd", "210561ffd72d00eccf12c0131b8024d5436bae95" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Wed Feb 20 11:40:49 2013 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Wed Feb 20 11:41:26 2013 +1000" }, "message": "Merge branch \u0027drm-intel-fixes\u0027 of git://people.freedesktop.org/~danvet/drm-intel into drm-next\n\nSo here\u0027s my promised pile of fixes for 3.9. I\u0027ve dropped the core prep\npatches for vt-switchless suspend/resume as discussed on irc. Highlights:\n- Fix dmar on g4x. Not really gfx related, but I\u0027m fed up with getting\n blamed for dmar crapouts.\n- Disable wc ptes updates on ilk when dmar is enabled (Chris). So again,\n dmar, but this time gfx related :(\n- Reduced range support for hsw, using the pipe CSC (Ville).\n- Fixup pll limits for gen3/4 (Patrick Jakobsson). The sdvo patch is\n already confirmed to fix 2 bug reports, so added cc: stable on that one.\n- Regression fix for 8bit fb console (Ville).\n- Preserve lane reversal bits on DDI/FDI ports (Damien).\n- Page flip vs. gpu hang fixes (Ville). Unfortuntely not quite all of\n them, need to decide what to do with the currently still in-flight ones.\n- Panel fitter regression fix from Mika Kuoppala (was accidentally left on\n on some pipes with the new modset code since 3.7). This also improves\n the modeset sequence and might help a few other unrelated issues with\n lvds.\n- Write backlight regs even harder ... another installement in our eternal\n fight against the BIOS and backlights.\n- Fixup lid notifier vs. suspend/resume races (Zhang Rui). Prep work for\n new ACPI stuff, but closing the race itself seems worthwile on its own.\n- A few other small fixes and tiny cleanups all over.\n\nLots of the patches are cc: stable since I\u0027ve stalled on a few\nnot-so-important fixes for 3.8 due to the grumpy noise Linus made.\n\n* \u0027drm-intel-fixes\u0027 of git://people.freedesktop.org/~danvet/drm-intel: (33 commits)\n intel/iommu: force writebuffer-flush quirk on Gen 4 Chipsets\n drm/i915: Disable WC PTE updates to w/a buggy IOMMU on ILK\n drm/i915: Implement pipe CSC based limited range RGB output\n drm/i915: inverted brightness quirk for Acer Aspire 4736Z\n drm/i915: Print the hw context status is debugfs\n drm/i915: Use HAS_L3_GPU_CACHE in i915_gem_l3_remap\n drm/i915: Fix PIPE_CONTROL DW/QW write through global GTT on IVB+\n drm/i915: Set i9xx sdvo clock limits according to specifications\n drm/i915: Set i9xx lvds clock limits according to specifications\n drm/i915: Preserve the DDI link reversal configuration\n drm/i915: Preserve the FDI line reversal override bit on CPT\n drm/i915: add missing \\n to UTS_RELEASE in the error_state\n drm: Use C8 instead of RGB332 when determining the format from depth/bpp\n drm: Fill depth/bits_per_pixel for C8 format\n drm/i915: don\u0027t clflush gem objects in stolen memory\n drm/i915: Don\u0027t wait for page flips if there was GPU reset\n drm/i915: Kill obj-\u003epending_flip\n drm/i915: Fix a typo in a intel_modeset_stage_output_state() comment\n drm/i915: remove bogus mutex_unlock from error-path\n drm/i915: Print the pipe control page GTT address\n ...\n" }, { "commit": "a7c47d6dc4a201b811e847e5449c8cffdc556deb", "tree": "e05562b96177a385386df338e53e31eaf47b5d80", "parents": [ "6f646095ec2f5d38a6429dd896cb2d5cbbb776dc" ], "author": { "name": "Rob Clark", "email": "robdclark@gmail.com", "time": "Tue Jan 08 17:50:48 2013 -0600" }, "committer": { "name": "Rob Clark", "email": "robdclark@gmail.com", "time": "Sun Feb 17 17:55:42 2013 -0500" }, "message": "drm: i2c encoder helper wrappers\n\nSimplify life for drivers using an encoder-slave, so that they can make\ntheir drm_encoder_helper_funcs const, rather than needing to dynamically\nallocate and populate them.\n\nSigned-off-by: Rob Clark \u003crobdclark@gmail.com\u003e\n" }, { "commit": "6f646095ec2f5d38a6429dd896cb2d5cbbb776dc", "tree": "41444767a379449c4e57f60dbee6d3bf2c670904", "parents": [ "21a245d2d62ef617978316203af032d499805cd2" ], "author": { "name": "Rob Clark", "email": "robdclark@gmail.com", "time": "Mon Dec 10 10:46:43 2012 -0600" }, "committer": { "name": "Rob Clark", "email": "robdclark@gmail.com", "time": "Sun Feb 17 17:55:42 2013 -0500" }, "message": "drm/cma: add debugfs helpers\n\nAdd helper to display fb\u0027s which can be used directly in drm_info_list:\n\nstatic struct drm_info_list foo_debugfs_list[] \u003d {\n\t\t...\n\t\t{ \"fb\", drm_fb_cma_debugfs_show, 0 },\n};\n\nto display information about CMA fb objects, as well as a\ndrm_gem_cma_describe() which can be used if the driver bothers to keep\na list of CMA GEM objects.\n\nSigned-off-by: Rob Clark \u003crobdclark@gmail.com\u003e\n" }, { "commit": "41907ddc1b71aaa4ef5290f46f0ec49d581d6aac", "tree": "5cb9829dfc0753d61713246f2cd60955827b19e8", "parents": [ "d46da4377689bd938795e53c4e2fb54dbcaeea44" ], "author": { "name": "Ben Widawsky", "email": "ben@bwidawsk.net", "time": "Fri Feb 08 11:32:47 2013 -0800" }, "committer": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Fri Feb 15 10:30:38 2013 +0100" }, "message": "drm/i915: Fix gen2 mappable calculations\n\nWhen I refactored the code initially, I forgot that gen2 uses a\ndifferent bar for the CPU mappable aperture. The agp-less code knows\nnothing of generations less than 5, so we have to expand the gtt_probe\nfunction to include the mappable base and end.\n\nIt was originally broken by me:\ncommit baa09f5fd8a6d033ec075355dda99a65b7f6a0f3\nAuthor: Ben Widawsky \u003cben@bwidawsk.net\u003e\nDate: Thu Jan 24 13:49:57 2013 -0800\n\n drm/i915: Add probe and remove to the gtt ops\n\nReported-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nSigned-off-by: Ben Widawsky \u003cben@bwidawsk.net\u003e\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n" }, { "commit": "a065b46a01b25d7d364e01e75f7ec2bd9ed5d9cb", "tree": "a23152729fa5c8917a8b0b527a725febaee54da3", "parents": [ "1b1d5397058f06bc5bd87d43ed93f34b28546ea4" ], "author": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Tue Feb 12 00:17:09 2013 +0100" }, "committer": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Thu Feb 14 00:08:41 2013 +0100" }, "message": "drm/fb-helper: remove unused members of struct drm_fb_helper\n\nSpotted by Rob Clark.\n\nReviewed-by: Rob Clark \u003crobdclark@gmail.com\u003e\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n" }, { "commit": "207fd32970b1def91b11ae28f6bebffc792db714", "tree": "75dd7d58e32b54395be26aba4381012251f65f55", "parents": [ "cd5428a5447cc6ca77ec6547d6f86834b205eac7" ], "author": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Sun Jan 20 22:13:14 2013 +0100" }, "committer": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Thu Feb 14 00:08:32 2013 +0100" }, "message": "drm/fb-helper: improve kerneldoc\n\nNow that the fbdev helper interface for drivers is trimmed down,\nupdate the kerneldoc for all the remaining exported functions.\n\nI\u0027ve tried to beat the DocBook a bit by reordering the function\nreferences a bit into a more sensible ordering. But that didn\u0027t work\nout at all. Hence just extend the in-code DOC: section a bit.\n\nAlso remove the LOCKING: sections - especially for the setup functions\nthey\u0027re totally bogus. But that\u0027s not a documentation problem, but\nsimply an artifact of the current rather hazardous locking around drm\ninit and even more so around fbdev setup ...\n\nv2: Some further improvements:\n- Also add documentation for drm_fb_helper_single_add_all_connectors,\n Dave Airlie didn\u0027t want me to kill this one from the fb helper\n interface.\n- Update docs for drm_fb_helper_fill_var/fix - they should be used\n from the driver\u0027s -\u003efb_probe callback to setup the fbdev info\n structure.\n- Clarify what the -\u003efb_probe callback should all do - it needs to\n setup both the fbdev info and allocate the drm framebuffer used as\n backing storage.\n- Add basic documentaation for the drm_fb_helper_funcs driver callback\n vfunc.\n\nv3: Implement clarifications Laurent Pinchart suggested in his review.\n\nv4: Fix another mispelling Laurent spotted.\n\nCc: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\nAcked-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n" }, { "commit": "de1ace5b56d11db26abe69fd19754f91c326763f", "tree": "0eef3e87a7c474c15283c2f955a779478c7f17a1", "parents": [ "43c8a849a1e92fba6ea4493ed950d2a2e31ac87c" ], "author": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Sun Jan 20 21:50:49 2013 +0100" }, "committer": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Thu Feb 14 00:07:52 2013 +0100" }, "message": "drm/fb-helper: unexport drm_fb_helper_single_fb_probe\n\nNot called by anyone, and really, shouldn\u0027t be. Drivers are supposed\neither drm_fb_helper_initial_config or drm_fb_helper_hotplug_event.\nOriginally this was done differently, but is now consolidated in the\nhelper functions and no longer done by drivers directly.\n\nReviewed-by: Rob Clark \u003crobdclark@gmail.com\u003e\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n" }, { "commit": "d21bf469d5301d025cd82997bb1529bcdc7086af", "tree": "f819eb34fd3f32ae79cca5421e6c688ed3d6f928", "parents": [ "6aed8ec3f76a22217c9ae183d32b1aa990bed069" ], "author": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Sun Jan 20 18:09:52 2013 +0100" }, "committer": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Thu Feb 14 00:07:50 2013 +0100" }, "message": "drm/fb-helper: kill drm_fb_helper_restore\n\nIt\u0027s only used internally for the sysrq and panic handlers provided by\nthe drm fb helper implementation. Hence just inline it, kill the\nexport and remove the confusing kerneldoc. Driver\u0027s are supposed to\ncall drm_fb_helper_restore_fbdev_mode on lastclose.\n\nNote that locking is totally fubar - the sysrq case doesn\u0027t take any\nlocks at all. The panic handler probably shouldn\u0027t take any locks\nsince it\u0027ll only make things worse. Otoh it\u0027s probably better to\nswitch things over to the atomic modeset callbacks (and disable the\npanic handler for those drivers which don\u0027t implement it).\n\nBut that\u0027s both better done in separate patches.\n\nReviewed-by: Rob Clark \u003crobdclark@gmail.com\u003e\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n" }, { "commit": "6aed8ec3f76a22217c9ae183d32b1aa990bed069", "tree": "429627e96391fd615275c1150d9482651696e623", "parents": [ "b9e5071386007729110e86fd2c55c687085624e3" ], "author": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Sun Jan 20 17:32:21 2013 +0100" }, "committer": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Thu Feb 14 00:07:50 2013 +0100" }, "message": "drm: review locking for drm_fb_helper_restore_fbdev_mode\n\n... it\u0027s required. Fix up exynos and the cma helper, and add a\ncorresponding WARN_ON to drm_fb_helper_restore_fbdev_mode.\n\nNote that tegra calls the fbdev cma helper restore function also from\nit\u0027s driver-load callback. Which is a bit against current practice,\nsince usually the call is only from -\u003elastclose, and initial setup is\ndone by drm_fb_helper_initial_config.\n\nAlso add the relevant drm DocBook entry.\n\nv2: Add promised WARN to restore_fbdev_mode.\n\nReviewed-by: Rob Clark \u003crobdclark@gmail.com\u003e\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n" }, { "commit": "b9e5071386007729110e86fd2c55c687085624e3", "tree": "4720ecd2ee9be3f8c46e71d75119219db1692507", "parents": [ "85a7ce67f3ebfd5975ffd1febcabfe4999ca911d", "cc4c0c4de3c775be22072ec3251f2e581b63d9a0" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Feb 08 14:02:32 2013 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Feb 08 14:02:32 2013 +1000" }, "message": "Merge branch \u0027for-airlied\u0027 of git://people.freedesktop.org/~mlankhorst/linux into drm-next\n\nTTM reservations changes, preparing for new reservation mutex system.\n\n* \u0027for-airlied\u0027 of git://people.freedesktop.org/~mlankhorst/linux:\n drm/ttm: unexport ttm_bo_wait_unreserved\n drm/nouveau: use ttm_bo_reserve_slowpath in validate_init, v2\n drm/ttm: use ttm_bo_reserve_slowpath_nolru in ttm_eu_reserve_buffers, v2\n drm/ttm: add ttm_bo_reserve_slowpath\n drm/ttm: cleanup ttm_eu_reserve_buffers handling\n drm/ttm: remove lru_lock around ttm_bo_reserve\n drm/nouveau: increase reservation sequence every retry\n drm/vmwgfx: always use ttm_bo_is_reserved\n" }, { "commit": "89177644a7b6306e6084a89eab7e290f4bfef397", "tree": "e8bac93d626a14df07b2d904797c5d0e86f6251a", "parents": [ "6504d0d9900a2c05ea1fbab2ec008bf442993d94" ], "author": { "name": "Aaron Plattner", "email": "aplattner@nvidia.com", "time": "Tue Jan 15 20:47:42 2013 +0000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Feb 08 13:39:08 2013 +1000" }, "message": "drm: add prime helpers\n\nInstead of reimplementing all of the dma_buf functionality in every driver,\ncreate helpers drm_prime_import and drm_prime_export that implement them in\nterms of new, lower-level hook functions:\n\n gem_prime_pin: callback when a buffer is created, used to pin buffers into GTT\n gem_prime_get_sg_table: convert a drm_gem_object to an sg_table for export\n gem_prime_import_sg_table: convert an sg_table into a drm_gem_object\n gem_prime_vmap, gem_prime_vunmap: map and unmap an object\n\nThese hooks are optional; drivers can opt in by using drm_gem_prime_import and\ndrm_gem_prime_export as the .gem_prime_import and .gem_prime_export fields of\nstruct drm_driver.\n\nv2:\n- Drop .begin_cpu_access. None of the drivers this code replaces implemented\n it. Having it here was a leftover from when I was trying to include i915 in\n this rework.\n- Use mutex_lock instead of mutex_lock_interruptible, as these three drivers\n did. This patch series shouldn\u0027t change that behavior.\n- Rename helpers to gem_prime_get_sg_table and gem_prime_import_sg_table.\n Rename struct sg_table* variables to \u0027sgt\u0027 for clarity.\n- Update drm.tmpl for these new hooks.\n\nv3:\n- Pass the vaddr down to the driver. This lets drivers that just call vunmap on\n the pointer avoid having to store the pointer in their GEM private structures.\n- Move documentation into a /** DOC */ comment in drm_prime.c and include it in\n drm.tmpl with a !P line. I tried to use !F lines to include documentation of\n the individual functions from drmP.h, but the docproc / kernel-doc scripts\n barf on that file, so hopefully this is good enough for now.\n- apply refcount fix from commit be8a42ae60addd8b6092535c11b42d099d6470ec\n (\"drm/prime: drop reference on imported dma-buf come from gem\")\n\nSigned-off-by: Aaron Plattner \u003caplattner@nvidia.com\u003e\nCc: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nCc: David Airlie \u003cairlied@linux.ie\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "73ccd6962fff19e53a7d4baaa17cf9311284ac9c", "tree": "3eb6fe56281e4de883587d71eabb2ffd53fc7a38", "parents": [ "ed914f69f8f979ea2b664abc4f1437235cf3db35", "0e3d50bfcbd338254795a700dcff429a96cba1a6" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Feb 08 13:27:28 2013 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Feb 08 13:27:28 2013 +1000" }, "message": "Merge branch \u0027drm-next-3.9\u0027 of git://people.freedesktop.org/~agd5f/linux into drm-next\n\nAlex writes:\n- CS ioctl cleanup and unification. Unification of a lot of functionality\nthat was duplicated across multiple generates of hardware.\n- Add support for Oland GPUs\n- Deprecate UMS support. Mesa and the ddx dropped support for UMS and\napparently very few people still use it since the UMS CS ioctl was broken\nfor several kernels and no one reported it. It was fixed in 3.8/stable.\n- Rework GPU reset. Use the status registers to determine what blocks\nto reset. This better matches the recommended reset programming model.\nThis also allows us to properly reset blocks besides GFX and DMA.\n- Switch the VM set page code to use an IB rather than the ring. This\nfixes overflow issues when doing large page table updates using a small\nring like DMA.\n- Several small cleanups and bug fixes.\n\n* \u0027drm-next-3.9\u0027 of git://people.freedesktop.org/~agd5f/linux: (38 commits)\n drm/radeon/dce6: fix display powergating\n drm/radeon: add Oland pci ids\n drm/radeon: radeon-asic updates for Oland\n drm/radeon: add ucode loading support for Oland\n drm/radeon: fill in gpu init for Oland\n drm/radeon: add Oland chip family\n drm/radeon: switch back to using the DMA ring for VM PT updates\n drm/radeon: use IBs for VM page table updates v2\n drm/radeon: don\u0027t reset the MC on IGPs/APUs\n drm/radeon: use the reset mask to determine if rings are hung\n drm/radeon: halt engines before disabling MC (si)\n drm/radeon: halt engines before disabling MC (cayman/TN)\n drm/radeon: halt engines before disabling MC (evergreen)\n drm/radeon: halt engines before disabling MC (6xx/7xx)\n drm/radeon: use status regs to determine what to reset (si)\n drm/radeon: use status regs to determine what to reset (cayman)\n drm/radeon: use status regs to determine what to reset (evergreen)\n drm/radeon: use status regs to determine what to reset (6xx/7xx)\n drm/radeon: rework GPU reset on cayman/TN\n drm/radeon: rework GPU reset on cayman/TN\n ...\n" }, { "commit": "ed914f69f8f979ea2b664abc4f1437235cf3db35", "tree": "43d839d79d02f4a9415199a2db7a4b6cfeb281eb", "parents": [ "bb0f78dd7ded88082b2430c43c65bc821c4ea360", "edb37a95c58147f89713e6c5cd220fa8fdfb4833" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Feb 08 12:34:44 2013 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Feb 08 12:34:44 2013 +1000" }, "message": "Merge tag \u0027of_videomode_helper\u0027 of git://git.pengutronix.de/git/str/linux into drm-next\n\nvideomode helpers for of + devicetree stuff, required for new kms drivers\n\n(not the fbdev maintainer).\n\n* tag \u0027of_videomode_helper\u0027 of git://git.pengutronix.de/git/str/linux:\n drm_modes: add of_videomode helpers\n drm_modes: add videomode helpers\n fbmon: add of_videomode helpers\n fbmon: add videomode helpers\n video: add of helper for display timings/videomode\n video: add display_timing and videomode\n viafb: rename display_timing to via_display_timing\n" }, { "commit": "cd17ef4114ad5c514b17e6a0bb02a309ab90b692", "tree": "9c162eaa96931597b83e165702e3483ba5c6bb1e", "parents": [ "67c964000236497e00c646472cd6b70b5c5109c8", "7d37beaaf3dbc6ff16f4d32a4dd6f8c557c6ab50" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Feb 08 11:08:10 2013 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Feb 08 11:08:10 2013 +1000" }, "message": "Merge tag \u0027drm-intel-next-2013-02-01\u0027 of git://people.freedesktop.org/~danvet/drm-intel into drm-next\n\nDaniel writes:\n\"Probably the last feature pull for 3.9, there\u0027s some fixes outstanding\nthought that I\u0027d like to sneak in. And maybe 3.8 takes a bit longer ...\nAnyway, highlights of this pull:\n- Kill the horrible IS_DISPLAYREG hack to handle the mmio offset movements\n on vlv, big thanks to Ville.\n- Dynamic power well support for Haswell, shaves away a bit when only\n using the eDP port on pipe A (Paulo). Plus unclaimed register fixes\n uncovered by this.\n- Clarifications of the gpu hang/reset state transitions, hopefully fixing\n a few spurious -EIO deaths in userspace.\n- Haswell ELD fixes.\n- Some more (pp)gtt cleanups from Ben.\n- A few smaller things all over.\n\nPlus all the stuff from the previous rather small pull request:\n- Broadcast RBG improvements and reduced color range fixes from Ville.\n- Ben is on a \"kill legacy gtt code for good\" spree, first pile of patches\n included.\n- No-relocs and bo lut improvements for faster execbuf from Chris.\n- Some refactorings from Imre.\"\n\n* tag \u0027drm-intel-next-2013-02-01\u0027 of git://people.freedesktop.org/~danvet/drm-intel: (101 commits)\n GPU/i915: Fix acpi_bus_get_device() check in drivers/gpu/drm/i915/intel_opregion.c\n drm/i915: Set the SR01 \"screen off\" bit in i915_redisable_vga() too\n drm/i915: Kill IS_DISPLAYREG()\n drm/i915: Introduce i915_vgacntrl_reg()\n drm/i915: gen6_gmch_remove can be static\n drm/i915: dynamic Haswell display power well support\n drm/i915: check the power down well on assert_pipe()\n drm/i915: don\u0027t send DP \"idle\" pattern before \"normal\" on HSW PORT_A\n drm/i915: don\u0027t run hsw power well code on !hsw\n drm/i915: kill cargo-culted locking from power well code\n drm/i915: Only run idle processing from i915_gem_retire_requests_worker\n drm/i915: Fix CAGF for HSW\n drm/i915: Reclaim GTT space for failed PPGTT\n drm/i915: remove intel_gtt structure\n drm/i915: Add probe and remove to the gtt ops\n drm/i915: extract hw ppgtt setup/cleanup code\n drm/i915: pte_encode is gen6+\n drm/i915: vfuncs for ppgtt\n drm/i915: vfuncs for gtt_clear_range/insert_entries\n drm/i915: Error state should print /sys/kernel/debug\n ...\n" }, { "commit": "7271c7f999609787c4617c51750d39e447859e7d", "tree": "b2ed6d707b1686aef519c5bb1a40e0811ae11284", "parents": [ "e737a14cd1f30a0bca753ae32ebada10adf93c3b" ], "author": { "name": "Alex Deucher", "email": "alexander.deucher@amd.com", "time": "Thu Jul 26 18:48:01 2012 -0400" }, "committer": { "name": "Alex Deucher", "email": "alexander.deucher@amd.com", "time": "Fri Feb 01 19:35:02 2013 -0500" }, "message": "drm/radeon: add Oland pci ids\n\nSigned-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n" }, { "commit": "a54c0c279f3864171fe53c66e769d5a137c5c651", "tree": "08d17841b72e39afdc4ce601f46770824c12fe8a", "parents": [ "baa09f5fd8a6d033ec075355dda99a65b7f6a0f3" ], "author": { "name": "Ben Widawsky", "email": "ben@bwidawsk.net", "time": "Thu Jan 24 14:45:00 2013 -0800" }, "committer": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Thu Jan 31 11:50:07 2013 +0100" }, "message": "drm/i915: remove intel_gtt structure\n\nWith the probe call in our dispatch table, we can now cut away the\nlast three remaining members in the intel_gtt shared struct and so\nremove it completely.\n\nv2: Rebased on top of Daniel\u0027s series\n\nSigned-off-by: Ben Widawsky \u003cben@bwidawsk.net\u003e\nReviewed-by: Damien Lespiau \u003cdamien.lespiau@intel.com\u003e\n[danvet: bikeshed commit message a bit.]\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n" }, { "commit": "edb37a95c58147f89713e6c5cd220fa8fdfb4833", "tree": "f4f344a417e57d115e80f2bffc05d2b81c08986c", "parents": [ "ebc64e453857e93b230881f48126257be9aa8830" ], "author": { "name": "Steffen Trumtrar", "email": "s.trumtrar@pengutronix.de", "time": "Sun Oct 28 18:28:06 2012 +0100" }, "committer": { "name": "Steffen Trumtrar", "email": "s.trumtrar@pengutronix.de", "time": "Thu Jan 24 09:04:14 2013 +0100" }, "message": "drm_modes: add of_videomode helpers\n\nAdd helper to get drm_display_mode from devicetree.\n\nSigned-off-by: Steffen Trumtrar \u003cs.trumtrar@pengutronix.de\u003e\nReviewed-by: Thierry Reding \u003cthierry.reding@avionic-design.de\u003e\nAcked-by: Thierry Reding \u003cthierry.reding@avionic-design.de\u003e\nTested-by: Thierry Reding \u003cthierry.reding@avionic-design.de\u003e\nTested-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\nReviewed-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\nAcked-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\nTested-by: Afzal Mohammed \u003cAfzal@ti.com\u003e\nTested-by: Rob Clark \u003crobclark@gmail.com\u003e\nTested-by: Leela Krishna Amudala \u003cleelakrishna.a@gmail.com\u003e\n" }, { "commit": "ebc64e453857e93b230881f48126257be9aa8830", "tree": "8ccb80725d3c273f5579bc1539ca841acd35381f", "parents": [ "790890444f591145d3a00623af461c9006e17d51" ], "author": { "name": "Steffen Trumtrar", "email": "s.trumtrar@pengutronix.de", "time": "Wed Nov 14 11:22:52 2012 +0100" }, "committer": { "name": "Steffen Trumtrar", "email": "s.trumtrar@pengutronix.de", "time": "Thu Jan 24 09:04:09 2013 +0100" }, "message": "drm_modes: add videomode helpers\n\nAdd conversion from videomode to drm_display_mode\n\nSigned-off-by: Steffen Trumtrar \u003cs.trumtrar@pengutronix.de\u003e\nReviewed-by: Thierry Reding \u003cthierry.reding@avionic-design.de\u003e\nAcked-by: Thierry Reding \u003cthierry.reding@avionic-design.de\u003e\nTested-by: Thierry Reding \u003cthierry.reding@avionic-design.de\u003e\nTested-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\nReviewed-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\nAcked-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\nTested-by: Afzal Mohammed \u003cAfzal@ti.com\u003e\nTested-by: Rob Clark \u003crobclark@gmail.com\u003e\nTested-by: Leela Krishna Amudala \u003cleelakrishna.a@gmail.com\u003e\n" }, { "commit": "735dc0d1e29329ff34ec97f66e130cce481c9607", "tree": "cf946856ff1defac833e601a3e4a4d8e841ee73e", "parents": [ "bac4b7c3b5c0660c08dc4949fe40e08e20364ee3", "20c60c35de3285222b3476c3445c66bedf0c449c" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Jan 21 07:44:58 2013 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Jan 21 07:44:58 2013 +1000" }, "message": "Merge branch \u0027drm-kms-locking\u0027 of git://people.freedesktop.org/~danvet/drm-intel into drm-next\n\nThe aim of this locking rework is that ioctls which a compositor should be\nmight call for every frame (set_cursor, page_flip, addfb, rmfb and\ngetfb/create_handle) should not be able to block on kms background\nactivities like output detection. And since each EDID read takes about\n25ms (in the best case), that always means we\u0027ll drop at least one frame.\n\nThe solution is to add per-crtc locking for these ioctls, and restrict\nbackground activities to only use the global lock. Change-the-world type\nof events (modeset, dpms, ...) need to grab all locks.\n\nTwo tricky parts arose in the conversion:\n- A lot of current code assumes that a kms fb object can\u0027t disappear while\n holding the global lock, since the current code serializes fb\n destruction with it. Hence proper lifetime management using the already\n created refcounting for fbs need to be instantiated for all ioctls and\n interfaces/users.\n\n- The rmfb ioctl removes the to-be-deleted fb from all active users. But\n unconditionally taking the global kms lock to do so introduces an\n unacceptable potential stall point. And obviously changing the userspace\n abi isn\u0027t on the table, either. Hence this conversion opportunistically\n checks whether the rmfb ioctl holds the very last reference, which\n guarantees that the fb isn\u0027t in active use on any crtc or plane (thanks\n to the conversion to the new lifetime rules using proper refcounting).\n Only if this is not the case will the code go through the slowpath and\n grab all modeset locks. Sane compositors will never hit this path and so\n avoid the stall, but userspace relying on these semantics will also not\n break.\n\nAll these cases are exercised by the newly added subtests for the i-g-t\nkms_flip, tested on a machine where a full detect cycle takes around 100\nms. It works, and no frames are dropped any more with these patches\napplied. kms_flip also contains a special case to exercise the\nabove-describe rmfb slowpath.\n\n* \u0027drm-kms-locking\u0027 of git://people.freedesktop.org/~danvet/drm-intel: (335 commits)\n drm/fb_helper: check whether fbcon is bound\n drm/doc: updates for new framebuffer lifetime rules\n drm: don\u0027t hold crtc mutexes for connector -\u003edetect callbacks\n drm: only grab the crtc lock for pageflips\n drm: optimize drm_framebuffer_remove\n drm/vmwgfx: add proper framebuffer refcounting\n drm/i915: dump refcount into framebuffer debugfs file\n drm: refcounting for crtc framebuffers\n drm: refcounting for sprite framebuffers\n drm: fb refcounting for dirtyfb_ioctl\n drm: don\u0027t take modeset locks in getfb ioctl\n drm: push modeset_lock_all into -\u003efb_create driver callbacks\n drm: nest modeset locks within fpriv-\u003efbs_lock\n drm: reference framebuffers which are on the idr\n drm: revamp framebuffer cleanup interfaces\n drm: create drm_framebuffer_lookup\n drm: revamp locking around fb creation/destruction\n drm: only take the crtc lock for -\u003ecursor_move\n drm: only take the crtc lock for -\u003ecursor_set\n drm: add per-crtc locks\n ...\n" }, { "commit": "362063619cf67c2c2fc2eb90951b2623cbb69a7c", "tree": "58af436089e986b742d4ec91e5ea57e4a4b2aeb5", "parents": [ "786b99ed13223d8ac58a937dd348aead45eb8191" ], "author": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Mon Dec 10 20:42:17 2012 +0100" }, "committer": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Sun Jan 20 22:17:00 2013 +0100" }, "message": "drm: revamp framebuffer cleanup interfaces\n\nWe have two classes of framebuffer\n- Created by the driver (atm only for fbdev), and the driver holds\n onto the last reference count until destruction.\n- Created by userspace and associated with a given fd. These\n framebuffers will be reaped when their assoiciated fb is closed.\n\nNow these two cases are set up differently, the framebuffers are on\ndifferent lists and hence destruction needs to clean up different\nthings. Also, for userspace framebuffers we remove them from any\ncurrent usage, whereas for internal framebuffers it is assumed that\nthe driver has done this already.\n\nLong story short, we need two different ways to cleanup such drivers.\nThree functions are involved in total:\n- drm_framebuffer_remove: Convenience function which removes the fb\n from all active usage and then drops the passed-in reference.\n- drm_framebuffer_unregister_private: Will remove driver-private\n framebuffers from relevant lists and drop the corresponding\n references. Should be called for driver-private framebuffers before\n dropping the last reference (or like for a lot of the drivers where\n the fbdev is embedded someplace else, before doing the cleanup\n manually).\n- drm_framebuffer_cleanup: Final cleanup for both classes of fbs,\n should be called by the driver\u0027s -\u003edestroy callback once the last\n reference is gone.\n\nThis patch just rolls out the new interfaces and updates all drivers\n(by adding calls to drm_framebuffer_unregister_private at all the\nright places)- no functional changes yet. Follow-on patches will move\ndrm core code around and update the lifetime management for\nframebuffers, so that we are no longer required to keep framebuffers\nalive by locking mode_config.mutex.\n\nI\u0027ve also updated the kerneldoc already.\n\nvmwgfx seems to again be a bit special, at least I haven\u0027t figured out\nhow the fbdev support in that driver works. It smells like it\u0027s\nexternal though.\n\nv2: The i915 driver creates another private framebuffer in the\nload-detect code. Adjust its cleanup code, too.\n\nReviewed-by: Rob Clark \u003crob@ti.com\u003e\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n" }, { "commit": "786b99ed13223d8ac58a937dd348aead45eb8191", "tree": "d2fed735d5be17d37d02a2a30260c766c079cf43", "parents": [ "4b096ac10da0b63f09bd123b86fed8deb80646ce" ], "author": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Sun Dec 02 21:53:40 2012 +0100" }, "committer": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Sun Jan 20 22:16:59 2013 +0100" }, "message": "drm: create drm_framebuffer_lookup\n\nAnd replace all fb lookups with it. Also add a WARN to\ndrm_mode_object_find since that is now no longer the blessed interface\nto look up an fb. And add kerneldoc to both functions.\n\nThis only updates all callsites, but immediately drops the acquired\nrefence again. Hence all callers still rely on the fact that a mode fb\ncan\u0027t disappear while they\u0027re holding the struct mutex. Subsequent\npatches will instate proper use of refcounts, and then rework the rmfb\nand unref code to no longer serialize fb destruction with the\nmode_config lock. We don\u0027t want that since otherwise a compositor\nmight end up stalling for a few frames in rmfb.\n\nv2: Don\u0027t use kref_get_unless_zero - Greg KH doesn\u0027t like that kind of\ninterface.\n\nReviewed-by: Rob Clark \u003crob@ti.com\u003e\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n" }, { "commit": "4b096ac10da0b63f09bd123b86fed8deb80646ce", "tree": "3d252a1b2ed182beb769796fb99eb68693176bfa", "parents": [ "dac35663cef4ca7f572d430bb54b14be8f03cb10" ], "author": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Mon Dec 10 21:19:18 2012 +0100" }, "committer": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Sun Jan 20 22:16:58 2013 +0100" }, "message": "drm: revamp locking around fb creation/destruction\n\nWell, at least step 1. The goal here is that framebuffer objects can\nsurvive outside of the mode_config lock, with just a reference held\nas protection. The first step to get there is to introduce a special\nfb_lock which protects fb lookup, creation and destruction, to make\nthem appear atomic.\n\nThis new fb_lock can nest within the mode_config lock. But the idea is\n(once the reference counting part is completed) that we only quickly\ntake that fb_lock to lookup a framebuffer and grab a reference,\nwithout any other locks involved.\n\nvmwgfx is the only driver which does framebuffer lookups itself, also\nwrap those calls to drm_mode_object_find with the new lock.\n\nAlso protect the fb_list walking in i915 and omapdrm with the new lock.\n\nAs a slight complication there\u0027s also the list of user-created fbs\nattached to the file private. The problem now is that at fclose() time\nwe need to walk that list, eventually do a modeset call to remove the\nfb from active usage (and are required to be able to take the\nmode_config lock), but in the end we need to grab the new fb_lock to\nremove the fb from the list. The easiest solution is to add another\nmutex to protect this per-file list.\n\nCurrently that new fbs_lock nests within the modeset locks and so\nappears redudant. But later patches will switch around this sequence\nso that taking the modeset locks in the fb destruction path is\noptional in the fastpath. Ultimately the goal is that addfb and rmfb\ndo not require the mode_config lock, since otherwise they have the\npotential to introduce stalls in the pageflip sequence of a compositor\n(if the compositor e.g. switches to a fullscreen client or if it\nenables a plane). But that requires a few more steps and hoops to jump\nthrough.\n\nNote that framebuffer creation/destruction is now double-protected -\nonce by the fb_lock and in parts by the idr_lock. The later would be\nunnecessariy if framebuffers would have their own idr allocator. But\nthat\u0027s material for another patch (series).\n\nv2: Properly initialize the fb-\u003efilp_head list in _init, otherwise the\nnewly added WARN to check whether the fb isn\u0027t on a fpriv list any\nmore will fail for driver-private objects.\n\nv3: Fixup two error-case unlock bugs spotted by Richard Wilbur.\n\nReviewed-by: Rob Clark \u003crob@ti.com\u003e\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n" }, { "commit": "29494c174dc4793ebd236aa522a2a1ed73b7180e", "tree": "a9ae7208969e1c632b0a8bf92fb92876d8f3f232", "parents": [ "d5d2636ed7990b93c7216f6a4d323f6b0eee08af" ], "author": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Sun Dec 02 02:18:25 2012 +0100" }, "committer": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Sun Jan 20 22:16:54 2013 +0100" }, "message": "drm: add per-crtc locks\n\n*drumroll*\n\nThe basic idea is to protect per-crtc state which can change without\ntouching the output configuration with separate mutexes, i.e. all the\ninput side state to a crtc like framebuffers, cursor settings or plane\nconfiguration. Holding such a crtc lock gives a read-lock on all the\nother crtc state which can be changed by e.g. a modeset.\n\nAll non-crtc state is still protected by the mode_config mutex.\nCallers that need to change modeset state of a crtc (e.g. dpms or\nset_mode) need to grab both the mode_config lock and nested within any\ncrtc locks.\n\nNote that since there can only ever be one holder of the mode_config\nlock we can grab the subordinate crtc locks in any order (if we need\nto grab more than one of them). Lockdep can handle such nesting with\nthe mutex_lock_nest_lock call correctly.\n\nWith this functions that only touch connectors/encoders but not crtcs\nonly need to take the mode_config lock. The biggest such case is the\noutput probing, which means that we can now pageflip and move cursors\nwhile the output probe code is reading an edid.\n\nMost cases neatly fall into the three buckets:\n- Only touches connectors and similar output state and so only needs\n the mode_config lock.\n- Touches the global configuration and so needs all locks.\n- Only touches the crtc input side and so only needs the crtc lock.\n\nBut a few cases that need special consideration:\n\n- Load detection which requires a crtc. The mode_config lock already\n prevents a modeset change, so we can use any unused crtc as we like\n to do load detection. The only thing to consider is that such\n temporary state changes don\u0027t leak out to userspace through ioctls\n that only take the crtc look (like a pageflip). Hence the load\n detect code needs to grab the crtc of any output pipes it touches\n (but only if it touches state used by the pageflip or cursor\n ioctls).\n\n- Atomic pageflip when moving planes. The first case is sane hw, where\n planes have a fixed association with crtcs - nothing needs to be\n done there. More insane^Wflexible hw needs to have plane-\u003ecrtc\n mapping which is separately protect with a lock that nests within\n the crtc lock. If the plane is unused we can just assign it to the\n current crtc and continue. But if a plane is already in use by\n another crtc we can\u0027t just reassign it.\n\n Two solution present themselves: Either go back to a slow-path which\n takes all modeset locks, potentially incure quite a hefty delay. Or\n simply disallowing such changes in one atomic pageflip - in general\n the vblanks of two crtcs are not synced, so there\u0027s no sane way to\n atomically flip such plane changes accross more than one crtc. I\u0027d\n heavily favour the later approach, going as far as mandating it as\n part of the ABI of such a new a nuclear pageflip.\n\n And if we _really_ want such semantics, we can always get them by\n introducing another pageflip mutex between the mode_config.mutex and\n the individual crtc locks. Pageflips crossing more than one crtc\n would then need to take that lock first, to lock out concurrent\n multi-crtc pageflips.\n\n- Optimized global modeset operations: We could just take the\n mode_config lock and then lazily lock all crtc which are affected by\n a modeset operation. This has the advantage that pageflip could\n continue unhampered on unaffected crtc. But if e.g. global resources\n like plls need to be reassigned and so affect unrelated crtcs we can\n still do that - nested locking works in any order.\n\nThis patch just adds the locks and takes them in drm_modeset_lock_all,\nno real locking changes yet.\n\nv2: Need to initialize the new lock in crtc_init and lock it righ\naway, for otherwise the modeset_unlock_all below will try to unlock a\nnot-locked mutex.\n\nReviewed-by: Rob Clark \u003crob@ti.com\u003e\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n" }, { "commit": "848499032504b1defdebdcf930aa70bd01a45ac1", "tree": "54e84b73e68ce68ece6ef9adfa888afbb6ec8ed9", "parents": [ "2d13b6796e420ed00389b7399a5d5ac7b1fed436" ], "author": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Sun Dec 02 00:28:11 2012 +0100" }, "committer": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Sun Jan 20 22:16:38 2013 +0100" }, "message": "drm: add drm_modeset_lock|unlock_all\n\nThis is the first step towards introducing the new modeset locking\nscheme. The plan is to put helper functions into place at all the\nright places step-by-step, so that the final patch to switch on the\nnew locking scheme doesn\u0027t need to touch every single driver.\n\nThis helper here will serve as the shotgun solutions for all places\nwhere a more fine-grained locking isn\u0027t (yet) implemented.\n\nv2: Fixup kerneldoc for unlock_all.\n\nReviewed-by: Rob Clark \u003crob@ti.com\u003e\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n" }, { "commit": "2d13b6796e420ed00389b7399a5d5ac7b1fed436", "tree": "8889af1340e267331fd7969d7f949ac9d6661019", "parents": [ "af26ef3b3978349cfbd864163a6ebb4906b733b5" ], "author": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Tue Dec 11 13:47:23 2012 +0100" }, "committer": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Sun Jan 20 15:57:58 2013 +0100" }, "message": "drm: encapsulate crtc-\u003eset_config calls\n\nWith refcounting we need to adjust framebuffer refcounts at each\ncallsite - much easier to do if they all call the same little helper\nfunction.\n\nReviewed-by: Rob Clark \u003crob@ti.com\u003e\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n" }, { "commit": "8faf6b18a2a6bece008de1e6bb80f0c608e58483", "tree": "e30089b80dc9706a4842c9d53094a4f9ecadb2e4", "parents": [ "7d1f9aeff1ee4a20b1aeb377dd0f579fe9647619" ], "author": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Sat Dec 01 23:43:11 2012 +0100" }, "committer": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Sun Jan 20 15:29:04 2013 +0100" }, "message": "drm: review locking rules in drm_crtc.c\n\n- config_cleanup was confused: It claimed that callers need to hold\n the modeset lock, but the connector|encoder_cleanup helpers grabbed\n that themselves (note that crtc_cleanup did _not_ grab the modeset\n lock). Which resulted in all drivers _not_ hodling the lock. Since\n this is for single-threaded cleanup code, drop the requirement from\n docs and also drop the lock_grabbing from all _cleanup functions.\n\n- Kill the LOCKING section in the doctype, since clearly we\u0027re not\n good enough to keep them up-to-date. And misleading locking\n documentation is worse than useless (see e.g. the comment in the\n vmgfx driver about the cleanup mess). And since for most functions\n the very first line either grabs the lock or has a WARN_ON(!locked)\n the documentation doesn\u0027t really add anything.\n\n- Instead put in some effort into explaining the only two special\n cases a bit better: config_init and config_cleanup are both called\n from single-threaded setup/teardown code, so don\u0027t do any locking.\n It\u0027s the driver\u0027s job though to enforce this.\n\n- Where lacking, add a WARN_ON(!is_locked). Not many places though,\n since locking around fbdev setup/teardown is through-roughly screwed\n up, and so will break almost every single WARN annotation I\u0027ve tried\n to add.\n\n- Add a drm_modeset_is_locked helper - the Grate Modset Locking Rework\n will use the compiler to assist in the big reorg by renaming the\n mode lock, so start encapsulating things. Unfortunately this ended\n up in the \"wrong\" header file since it needs the definition of\n struct drm_device.\n\nv2: Drop most WARNS again - we hit them all over the place, mostly in\nthe setup and teardown sequences. And trying to fix it up leads to\nnice deadlocks, since the locking in the setup code is really\ninconsistent.\n\nReviewed-by: Rob Clark \u003crob@ti.com\u003e\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n" }, { "commit": "e5c653777986b40e2986d2c918847fddbcba3a34", "tree": "3d1ec01de02c7770bbee08e2daf0b0edf67f35e6", "parents": [ "8d2e630899165d413ae8a2adc36846ac0b71bada" ], "author": { "name": "Ben Widawsky", "email": "ben@bwidawsk.net", "time": "Fri Jan 18 12:30:34 2013 -0800" }, "committer": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Sun Jan 20 13:11:12 2013 +0100" }, "message": "agp/intel: Add gma_bus_addr\n\nIt is no longer used in the i915 code, so isolate it from the shared\nstruct.\n\nThis was originally part of:\ncommit 0e275518f325418d559c05327775bff894b237f7\nAuthor: Ben Widawsky \u003cben@bwidawsk.net\u003e\nDate: Mon Jan 14 13:35:33 2013 -0800\n\n agp/intel: decouple more of the agp-i915 sharing\n\n Reviewed-by: Rodrigo Vivi \u003crodrigo.vivi@gmail.com\u003e\n Signed-off-by: Ben Widawsky \u003cben@bwidawsk.net\u003e\n\nThat commit had some other hunks which can\u0027t be used due to issues\nDaniel found in previous commits.\n\nSigned-off-by: Ben Widawsky \u003cben@bwidawsk.net\u003e\n[danvet: drop squash notice from the commit since it\u0027s imo ok to keep\nthis one separate.]\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n" }, { "commit": "8d2e630899165d413ae8a2adc36846ac0b71bada", "tree": "f2df0bedcdedca4a63ecb40402680d6099f40fa3", "parents": [ "9c61a32d31a55c8c6e590d83ae5645e14fde09f2" ], "author": { "name": "Ben Widawsky", "email": "ben@bwidawsk.net", "time": "Fri Jan 18 12:30:33 2013 -0800" }, "committer": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Sun Jan 20 13:11:12 2013 +0100" }, "message": "drm/i915: Needs_dmar, not\n\nThe reasoning behind our code taking two paths depending upon whether or\nnot we may have been configured for IOMMU isn\u0027t clear to me. It should\nalways be safe to use the pci mapping functions as they are designed to\nabstract the decision we were handling in i915.\n\nAside from simpler code, removing another member for the intel_gtt\nstruct is a nice motivation.\n\nI ran this by Chris, and he wasn\u0027t concerned about the extra kzalloc,\nand memory references vs. page_to_phys calculation in the case without\nIOMMU.\n\nv2: Update commit message\n\nv3: Remove needs_dmar addition from Zhenyu upstream\n\nThis reverts (and then other stuff)\ncommit 20652097dadd9a7fb4d652f25466299974bc78f9\nAuthor: Zhenyu Wang \u003czhenyuw@linux.intel.com\u003e\nDate: Thu Dec 13 23:47:47 2012 +0800\n\n drm/i915: Fix missed needs_dmar setting\n\nReviewed-by: Rodrigo Vivi \u003crodrigo.vivi@gmail.com\u003e (v2)\nCc: Zhenyu Wang \u003czhenyuw@linux.intel.com\u003e\nSigned-off-by: Ben Widawsky \u003cben@bwidawsk.net\u003e\n[danvet: Squash in follow-up fix to remove the bogus hunk which\ndeleted the dma_mask configuration for gen6+.]\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n" }, { "commit": "9c61a32d31a55c8c6e590d83ae5645e14fde09f2", "tree": "3e0b76fa2c8b3a598b80e546821f659fae2530e2", "parents": [ "a81cc00c11ab6816fbcb7dd99a60b50e71765d25" ], "author": { "name": "Ben Widawsky", "email": "ben@bwidawsk.net", "time": "Fri Jan 18 12:30:32 2013 -0800" }, "committer": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Sun Jan 20 13:11:11 2013 +0100" }, "message": "drm/i915: Remove scratch page from shared\n\nWe already had a mapping in both (minus the phys_addr in AGP).\n\nReviewed-by: Rodrigo Vivi \u003crodrigo.vivi@gmail.com\u003e\nSigned-off-by: Ben Widawsky \u003cben@bwidawsk.net\u003e\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n" }, { "commit": "a81cc00c11ab6816fbcb7dd99a60b50e71765d25", "tree": "506c92848c259bfa397e69d805508511e2024458", "parents": [ "abedc077b45eff0b5a8630af8431ad5d59213582" ], "author": { "name": "Ben Widawsky", "email": "ben@bwidawsk.net", "time": "Fri Jan 18 12:30:31 2013 -0800" }, "committer": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Sun Jan 20 13:11:11 2013 +0100" }, "message": "drm/i915: Cut out the infamous ILK w/a from AGP layer\n\nAnd, move it to where the rest of the logic is.\n\nThere is some slight functionality changes. There was extra paranoid\nchecks in AGP code making sure we never do idle maps on gen2 parts. That\nwas not duplicated as the simple PCI id check should do the right thing.\n\nv2: use IS_GEN5 \u0026\u0026 IS_MOBILE check instead. For now, this is the same as\nIS_IRONLAKE_M but is more future proof. The workaround docs hint that\nmore than one platform may be effected, but we\u0027ve never seen such a\nplatform in the wild. (Rodrigo, Daniel)\n\nReviewed-by: Rodrigo Vivi \u003crodrigo.vivi@gmail.com\u003e (v1)\nCc: Dave Airlie \u003cairlied@redhat.com\u003e\nSigned-off-by: Ben Widawsky \u003cben@bwidawsk.net\u003e\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n" }, { "commit": "b1edd6a6ecd436af33f210a5c75e0249466fd200", "tree": "12ea6bb801282488ecd38f6a4001867b26289515", "parents": [ "55bc60db5988c8366751d3d04dd690698a53412c" ], "author": { "name": "Ville Syrjälä", "email": "ville.syrjala@linux.intel.com", "time": "Thu Jan 17 16:31:30 2013 +0200" }, "committer": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Sun Jan 20 13:09:44 2013 +0100" }, "message": "drm/edid: Add drm_rgb_quant_range_selectable()\n\ndrm_rgb_quant_range_selectable() will report whether the monitor\nclaims to support for RGB quantization range selection.\n\nThe information can be found in the CEA Video capability block.\n\nv2: s/quantzation/quantization/ in the comment\n\nReviewed-by: Paulo Zanoni \u003cpaulo.r.zanoni@intel.com\u003e\nSigned-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nAcked-by: David Airlie \u003cairlied@linux.ie\u003e\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n" } ], "next": "b5cc6c0387b2f8d269c1df1e68c97c958dd22fed" }