)]}' { "log": [ { "commit": "1794d257fa7bab3ea5162f8abdca749996b65343", "tree": "ea8d035a218a0cb0ee83443b3c25bd8df25567e2", "parents": [ "bbb0aef5cfe95fe9b51a7eeba4a440b69037b01f" ], "author": { "name": "Chris Wilson", "email": "chris@chris-wilson.co.uk", "time": "Sun Apr 17 07:43:32 2011 +0100" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Apr 28 14:56:23 2011 +1000" }, "message": "drm: Export the command-line mode parser\n\nIn the absence of configuration data for providing the fixed mode for\na panel, I would like to be able to pass such modes along a separate\nmodule paramenter. To do so, I then need to parse a modeline from a\nstring, which drm is already capable of. Export that capability to the\ndrivers.\n\nSigned-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "bbb0aef5cfe95fe9b51a7eeba4a440b69037b01f", "tree": "858fdc7ee2b358827888f278fdfd0dddfbacc64b", "parents": [ "5ad3d8831f0c97257460c11ddcc1cc0466c762d4" ], "author": { "name": "Joe Perches", "email": "joe@perches.com", "time": "Sun Apr 17 20:35:52 2011 -0700" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Apr 28 14:53:21 2011 +1000" }, "message": "drm: Verify debug message arguments\n\nAdd __attribute__((format (printf, 4, 5))) to drm_ut_debug_printk\nand fix fallout.\n\nSigned-off-by: Joe Perches \u003cjoe@perches.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "5ad3d8831f0c97257460c11ddcc1cc0466c762d4", "tree": "8ab86a7b1c0ebfcfdbf24270ac1f54ed1bed8a6e", "parents": [ "cb49ba0bb18632faa7c5ce1dcb095eaa70821c1d" ], "author": { "name": "Joe Perches", "email": "joe@perches.com", "time": "Sun Apr 17 20:35:51 2011 -0700" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Apr 28 14:53:02 2011 +1000" }, "message": "drm: Create and use drm_err\n\nReduce drm text size ~1% by using drm_err and\nprintf extension %pV to emit error messages.\n\nRemove unused macro DRM_MEM_ERROR.\n\n$ size drivers/gpu/drm/built-in.o*\n text\t data\t bss\t dec\t hex\tfilename\n 361159\t 9663\t 256\t 371078\t 5a986\tdrivers/gpu/drm/built-in.o.new\n 365416\t 9663\t 256\t 375335\t 5ba27\tdrivers/gpu/drm/built-in.o.old\n\nSigned-off-by: Joe Perches \u003cjoe@perches.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "25985edcedea6396277003854657b5f3cb31a628", "tree": "f026e810210a2ee7290caeb737c23cb6472b7c38", "parents": [ "6aba74f2791287ec407e0f92487a725a25908067" ], "author": { "name": "Lucas De Marchi", "email": "lucas.demarchi@profusion.mobi", "time": "Wed Mar 30 22:57:33 2011 -0300" }, "committer": { "name": "Lucas De Marchi", "email": "lucas.demarchi@profusion.mobi", "time": "Thu Mar 31 11:26:23 2011 -0300" }, "message": "Fix common misspellings\n\nFixes generated by \u0027codespell\u0027 and manually reviewed.\n\nSigned-off-by: Lucas De Marchi \u003clucas.demarchi@profusion.mobi\u003e\n" }, { "commit": "34db18abd376b2075c760c38f0b861aed379415d", "tree": "c4174e39a2f445f17c25ab206d45c66217bbbf85", "parents": [ "e73f88af66fcc50083fae4b7e1c39b469179a97a", "47ae63e0c2e5fdb582d471dc906eb29be94c732f" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Mar 14 14:15:13 2011 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Mar 14 14:15:13 2011 +1000" }, "message": "Merge remote branch \u0027intel/drm-intel-next\u0027 of ../drm-next into drm-core-next\n\n* \u0027intel/drm-intel-next\u0027 of ../drm-next: (755 commits)\n drm/i915: Only wait on a pending flip if we intend to write to the buffer\n drm/i915/dp: Sanity check eDP existence\n drm/i915: Rebind the buffer if its alignment constraints changes with tiling\n drm/i915: Disable GPU semaphores by default\n drm/i915: Do not overflow the MMADDR write FIFO\n Revert \"drm/i915: fix corruptions on i8xx due to relaxed fencing\"\n drm/i915: Don\u0027t save/restore hardware status page address register\n drm/i915: don\u0027t store the reg value for HWS_PGA\n drm/i915: fix memory corruption with GM965 and \u003e4GB RAM\n Linux 2.6.38-rc7\n Revert \"TPM: Long default timeout fix\"\n drm/i915: Re-enable GPU semaphores for SandyBridge mobile\n drm/i915: Replace vblank PM QoS with \"Interrupt-Based AGPBUSY#\"\n Revert \"drm/i915: Use PM QoS to prevent C-State starvation of gen3 GPU\"\n drm/i915: Allow relocation deltas outside of target bo\n drm/i915: Silence an innocuous compiler warning for an unused variable\n fs/block_dev.c: fix new kernel-doc warning\n ACPI: Fix build for CONFIG_NET unset\n mm: \u003casm-generic/pgtable.h\u003e must include \u003clinux/mm_types.h\u003e\n x86: Use u32 instead of long to set reset vector back to 0\n ...\n\nConflicts:\n\tdrivers/gpu/drm/i915/i915_gem.c\n" }, { "commit": "9f35421e09c494c36079d7cf5724ae9f832431d7", "tree": "1c9ab806a43a30df918e9e12505823a387779e34", "parents": [ "e1e84017f95504348308085ae7cd0bfe268caf10" ], "author": { "name": "Ben Skeggs", "email": "bskeggs@redhat.com", "time": "Mon Feb 21 11:17:35 2011 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Mar 04 14:47:30 2011 +1000" }, "message": "drm/core: add ioctl to query device/driver capabilities\n\nWe\u0027re coming to see a need to have a set of generic capability checks in\nthe core DRM, in addition to the driver-specific ioctls that already\nexist.\n\nThis patch defines an ioctl to do as such, but does not yet define any\ncapabilities.\n\n[airlied: drop the driver callback for now.]\n\nSigned-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "1922756124ddd53846877416d92ba4a802bc658f", "tree": "22b5b5e5457d5eab364cd2c860814b29874e56f7", "parents": [ "3c0556e9673f79e386b27768e27d18b3ce29e40e" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Feb 24 08:35:06 2011 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Feb 28 15:24:35 2011 +1000" }, "message": "drm: fix unsigned vs signed comparison issue in modeset ctl ioctl.\n\nThis fixes CVE-2011-1013.\n\nReported-by: Matthiew Herrb (OpenBSD X.org team)\nCc: stable@kernel.org\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "a250b9fdc53a286d32e22f21170382a46b3a3ef5", "tree": "46bca176ce307f951d3e154b7a41ddd43fdee0ac", "parents": [ "8410ea3b95d105a5be5db501656f44bbb91197c1" ], "author": { "name": "Dave Airlie", "email": "airlied@gmail.com", "time": "Wed Dec 15 07:13:55 2010 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Feb 07 13:09:42 2011 +1000" }, "message": "drm: add usb framework\n\nThis adds an initial framework to plug USB graphics devices\ninto the drm/kms subsystem.\n\nI\u0027ve started writing a displaylink driver using this interface.\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "8410ea3b95d105a5be5db501656f44bbb91197c1", "tree": "6cd27f207e50c13ba2f4a78d6323bc23f751e380", "parents": [ "ff72145badb834e8051719ea66e024784d000cb4" ], "author": { "name": "Dave Airlie", "email": "airlied@gmail.com", "time": "Wed Dec 15 03:16:38 2010 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Feb 07 13:09:36 2011 +1000" }, "message": "drm: rework PCI/platform driver interface.\n\nThis abstracts the pci/platform interface out a step further,\nwe can go further but this is far enough for now to allow USB\nto be plugged in.\n\nThe drivers now just call the init code directly for their\ndevice type.\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "ff72145badb834e8051719ea66e024784d000cb4", "tree": "39dc5fc512e3e0836713de9defb91ea8b4033aa2", "parents": [ "1f692a14cbfbeb11f9a9c16f25c8ecb8ab50d3d5" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Feb 07 12:16:14 2011 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Feb 07 12:16:14 2011 +1000" }, "message": "drm: dumb scanout create/mmap for intel/radeon (v3)\n\nThis is just an idea that might or might not be a good idea,\nit basically adds two ioctls to create a dumb and map a dumb buffer\nsuitable for scanout. The handle can be passed to the KMS ioctls to create\na framebuffer.\n\nIt looks to me like it would be useful in the following cases:\na) in development drivers - we can always provide a shadowfb fallback.\nb) libkms users - we can clean up libkms a lot and avoid linking\nto libdrm_*.\nc) plymouth via libkms is a lot easier.\n\nUserspace bits would be just calls + mmaps. We could probably\nmark these handles somehow as not being suitable for acceleartion\nso as top stop people who are dumber than dumb.\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "78c6e170badd22c86a5b50a7eb038a02024b8f03", "tree": "481dba4a9d30cf25ee25cda443a17cb6e209668c", "parents": [ "021a8455bedb01750fa8047c8576e19d5af9a99f" ], "author": { "name": "Chris Wilson", "email": "chris@chris-wilson.co.uk", "time": "Mon Jan 31 10:48:04 2011 +0000" }, "committer": { "name": "Chris Wilson", "email": "chris@chris-wilson.co.uk", "time": "Mon Jan 31 12:38:47 2011 +0000" }, "message": "drm/i915: Suppress spurious vblank interrupts\n\nHugh Dickins found that characters in xterm were going missing and oft\ndelayed. Being the curious type, he managed to associate this with the\nnew high-precision vblank patches; disabling these he found, restored\nthe orderliness of his characters.\n\nThe oddness begins when one realised that Hugh was not using vblanks at\nall on his system (fvwm and some xterms). Instead, all he had to go on\nwere warning of a pipe underrun, curiously enough at around 60Hz. He\npoked and found that in addition to the underrun warning, the hardware\nwas flagging the start of a new frame, a vblank, which in turn was\nkicking off the pending vblank processing code.\n\nThere is little we can do for the underruns on Hugh\u0027s machine, a\nCrestline [965GM], which must have its FIFO watermarks set to 8.\nHowever, we do not need to process the vblank if we know that they are\ndisabled...\n\nReported-by: Hugh Dickins \u003chughd@google.com\u003e\nSigned-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n" }, { "commit": "5bcf719b7db0f9366cedaf102b081f99b1c325ae", "tree": "6929d42e1d61a3b1bd7b130a37a80b99718e14aa", "parents": [ "8d608aa6295242fe4c4b6105b8c59c6a5b232d89" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Tue Dec 07 09:20:40 2010 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Wed Jan 05 13:45:30 2011 +1000" }, "message": "drm/switcheroo: track state of switch in drivers.\n\nWe need to track the state of the switch in drivers, so that after s/r\nwe don\u0027t resume the card we\u0027ve explicitly switched off before. Also\ndon\u0027t allow a userspace open to occur if we\u0027ve switched the gpu off.\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "d8c58fabd75021cdd99abcd96513cb088d41092b", "tree": "f6554ecfb27c0d50f5ae6acae3a7077282813cab", "parents": [ "9c04f015ebc2cc2cca5a4a576deb82a311578edc", "b08ebe7e776e5be0271ed1e1bbb384e1f29dd117" ], "author": { "name": "Chris Wilson", "email": "chris@chris-wilson.co.uk", "time": "Thu Dec 16 21:02:15 2010 +0000" }, "committer": { "name": "Chris Wilson", "email": "chris@chris-wilson.co.uk", "time": "Thu Dec 16 21:02:15 2010 +0000" }, "message": "Merge remote branch \u0027airlied/drm-core-next\u0027 into drm-intel-next\n" }, { "commit": "4af72e2865a23ac090884a421bd1a8b19e247a22", "tree": "a4188f9feed58258a13d8b00c907c19c5045c85e", "parents": [ "40ce6575102b23e432932b5ce41c44bf7cc5023b" ], "author": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Fri Nov 05 18:13:43 2010 +0100" }, "committer": { "name": "Chris Wilson", "email": "chris@chris-wilson.co.uk", "time": "Tue Nov 23 20:14:44 2010 +0000" }, "message": "drm: kill drm_agp_chipset_flush\n\nNo longer used.\n\nCc: Dave Airlie \u003cairlied@gmail.com\u003e\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nSigned-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n" }, { "commit": "27641c3f003e7f3b6585c01d8a788883603eb262", "tree": "e26eb639c4b5f429d30033d598544f82479d1ea3", "parents": [ "3561d43fd289f590fdae672e5eb831b8d5cf0bf6" ], "author": { "name": "Mario Kleiner", "email": "mario.kleiner@tuebingen.mpg.de", "time": "Sat Oct 23 04:20:23 2010 +0200" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Nov 22 11:45:05 2010 +1000" }, "message": "drm/vblank: Add support for precise vblank timestamping.\n\nThe DRI2 swap \u0026 sync implementation needs precise\nvblank counts and precise timestamps corresponding\nto those vblank counts. For conformance to the OpenML\nOML_sync_control extension specification the DRM\ntimestamp associated with a vblank count should\ncorrespond to the start of video scanout of the first\nscanline of the video frame following the vblank\ninterval for that vblank count.\n\nTherefore we need to carry around precise timestamps\nfor vblanks. Currently the DRM and KMS drivers generate\ntimestamps ad-hoc via do_gettimeofday() in some\nplaces. The resulting timestamps are sometimes not\nvery precise due to interrupt handling delays, they\ndon\u0027t conform to OML_sync_control and some are wrong,\nas they aren\u0027t taken synchronized to the vblank.\n\nThis patch implements support inside the drm core\nfor precise and robust timestamping. It consists\nof the following interrelated pieces.\n\n1. Vblank timestamp caching:\n\nA per-crtc ringbuffer stores the most recent vblank\ntimestamps corresponding to vblank counts.\n\nThe ringbuffer can be read out lock-free via the\naccessor function:\n\nstruct timeval timestamp;\nvblankcount \u003d drm_vblank_count_and_time(dev, crtcid, \u0026timestamp).\n\nThe function returns the current vblank count and\nthe corresponding timestamp for start of video\nscanout following the vblank interval. It can be\nused anywhere between enclosing drm_vblank_get(dev, crtcid)\nand drm_vblank_put(dev,crtcid) statements. It is used\ninside the drmWaitVblank ioctl and in the vblank event\nqueueing and handling. It should be used by kms drivers for\ntimestamping of bufferswap completion.\n\nThe timestamp ringbuffer is reinitialized each time\nvblank irq\u0027s get reenabled in drm_vblank_get()/\ndrm_update_vblank_count(). It is invalidated when\nvblank irq\u0027s get disabled.\n\nThe ringbuffer is updated inside drm_handle_vblank()\nat each vblank irq.\n\n2. Calculation of precise vblank timestamps:\n\ndrm_get_last_vbltimestamp() is used to compute the\ntimestamp for the end of the most recent vblank (if\ninside active scanout), or the expected end of the\ncurrent vblank interval (if called inside a vblank\ninterval). The function calls into a new optional kms\ndriver entry point dev-\u003edriver-\u003eget_vblank_timestamp()\nwhich is supposed to provide the precise timestamp.\nIf a kms driver doesn\u0027t implement the entry point or\nif the call fails, a simple do_gettimeofday() timestamp\nis returned as crude approximation of the true vblank time.\n\nA new drm module parameter drm.timestamp_precision_usec\nallows to disable high precision timestamps (if set to\nzero) or to specify the maximum acceptable error in\nthe timestamps in microseconds.\n\nKms drivers could implement their get_vblank_timestamp()\nfunction in a gpu specific way, as long as returned\ntimestamps conform to OML_sync_control, e.g., by use\nof gpu specific hardware timestamps.\n\nOptionally, kms drivers can simply wrap and use the new\nutility function drm_calc_vbltimestamp_from_scanoutpos().\nThis function calls a new optional kms driver function\ndev-\u003edriver-\u003eget_scanout_position() which returns the\ncurrent horizontal and vertical video scanout position\nof the crtc. The scanout position together with the\ndrm_display_timing of the current video mode is used\nto calculate elapsed time relative to start of active scanout\nfor the current video frame. This elapsed time is subtracted\nfrom the current do_gettimeofday() time to get the timestamp\ncorresponding to start of video scanout. Currently\nnon-interlaced, non-doublescan video modes, with or\nwithout panel scaling are handled correctly. Interlaced/\ndoublescan modes are tbd in a future patch.\n\n3. Filtering of redundant vblank irq\u0027s and removal of\nsome race-conditions in the vblank irq enable/disable path:\n\nSome gpu\u0027s (e.g., Radeon R500/R600) send spurious vblank\nirq\u0027s outside the vblank if vblank irq\u0027s get reenabled.\nThese get detected by use of the vblank timestamps and\nfiltered out to avoid miscounting of vblanks.\n\nSome race-conditions between the vblank irq enable/disable\nfunctions, the vblank irq handler and the gpu itself (updating\nits hardware vblank counter in the \"wrong\" moment) are\nfixed inside vblank_disable_and_save() and\ndrm_update_vblank_count() by use of the vblank timestamps and\na new spinlock dev-\u003evblank_time_lock.\n\nThe time until vblank irq disable is now configurable via\na new drm module parameter drm.vblankoffdelay to allow\nexperimentation with timeouts that are much shorter than\nthe current 5 seconds and should allow longer vblank off\nperiods for better power savings.\n\nFollowup patches will use these new functions to\nimplement precise timestamping for the intel and radeon\nkms drivers.\n\nSigned-off-by: Mario Kleiner \u003cmario.kleiner@tuebingen.mpg.de\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "0f8c6d7ca9257d6a01671ab69b897860d3ae9bc0", "tree": "e06b567e620e17d5ce983610ef3721d860d0cf27", "parents": [ "13b2928933919c5344716d49620a52493a243f8c" ], "author": { "name": "Chris Wilson", "email": "chris@chris-wilson.co.uk", "time": "Mon Nov 01 12:38:44 2010 +0000" }, "committer": { "name": "Chris Wilson", "email": "chris@chris-wilson.co.uk", "time": "Mon Nov 01 12:38:44 2010 +0000" }, "message": "drm/i915: Move the invalidate|flush information out of the device struct\n\n... and into a local structure scoped for the single function in which\nit is used.\n\nSigned-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n" }, { "commit": "fb7ba2114bcd8bb51640c20bc68f89164b29b9ed", "tree": "80b4a779130a477680a72109257fb8c19d66cf22", "parents": [ "9a170caed6fce89da77852575a7eee7dbadee332", "abb295f3b3db602f91accf58b526b30b48673af1" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Wed Oct 06 11:10:48 2010 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Wed Oct 06 11:10:48 2010 +1000" }, "message": "Merge remote branch \u0027korg/drm-fixes\u0027 into drm-vmware-next\n\nnecessary for some of the vmware fixes to be pushed in.\n\nConflicts:\n\tdrivers/gpu/drm/drm_gem.c\n\tdrivers/gpu/drm/i915/intel_fb.c\n\tinclude/drm/drmP.h\n" }, { "commit": "73aa808f10effc280e6eb70267314542a7c29426", "tree": "e8c7d0e25bb5d0658c47ff0339aec9a65367bbc2", "parents": [ "6a04002bea137d2c6359228316d9c827806e475f" ], "author": { "name": "Chris Wilson", "email": "chris@chris-wilson.co.uk", "time": "Thu Sep 30 11:46:12 2010 +0100" }, "committer": { "name": "Chris Wilson", "email": "chris@chris-wilson.co.uk", "time": "Fri Oct 01 14:45:20 2010 +0100" }, "message": "drm: Move the GTT accounting to i915\n\nOnly drm/i915 does the bookkeeping that makes the information useful,\nand the information maintained is driver specific, so move it out of the\ncore and into its single user.\n\nSigned-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nCc: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "39b4d07aa3583ceefe73622841303a0a3e942ca1", "tree": "d42f6e782f331b1d967f50ca3a02b9e51ea88515", "parents": [ "29d08b3efddca628b0360411ab2b85f7b1723f48" ], "author": { "name": "Chris Wilson", "email": "chris@chris-wilson.co.uk", "time": "Thu Sep 30 09:10:26 2010 +0100" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Oct 01 21:08:45 2010 +1000" }, "message": "drm: Hold the mutex when dropping the last GEM reference (v2)\n\nIn order to be fully threadsafe we need to check that the drm_gem_object\nrefcount is still 0 after acquiring the mutex in order to call the free\nfunction. Otherwise, we may encounter scenarios like:\n\nThread A: Thread B:\ndrm_gem_close\nunreference_unlocked\nkref_put mutex_lock\n... i915_gem_evict\n... kref_get -\u003e BUG\n... i915_gem_unbind\n... kref_put\n... i915_gem_object_free\n... mutex_unlock\nmutex_lock\ni915_gem_object_free -\u003e BUG\ni915_gem_object_unbind\nkfree\nmutex_unlock\n\nNote that no driver is currently using the free_unlocked vfunc and it is\nscheduled for removal, hasten that process.\n\nBugzilla: https://bugs.freedesktop.org/show_bug.cgi?id\u003d30454\nReported-and-Tested-by: Magnus Kessler \u003cMagnus.Kessler@gmx.net\u003e\nSigned-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nCc: stable@kernel.org\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "29d08b3efddca628b0360411ab2b85f7b1723f48", "tree": "f89a8a3069ebe9828b8a08c4b123f52625bf0bc7", "parents": [ "130b9851933e6da636502cd85e1ba8f45f862e8c" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Sep 27 16:17:17 2010 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Oct 01 09:17:44 2010 +1000" }, "message": "drm/gem: handlecount isn\u0027t really a kref so don\u0027t make it one.\n\nThere were lots of places being inconsistent since handle count\nlooked like a kref but it really wasn\u0027t.\n\nFix this my just making handle count an atomic on the object,\nand have it increase the normal object kref.\n\nNow i915/radeon/nouveau drivers can drop the normal reference on\nuserspace object creation, and have the handle hold it.\n\nThis patch fixes a memory leak or corruption on unload, because\nthe driver had no way of knowing if a handle had been actually\nadded for this object, and the fbcon object needed to know this\nto clean itself up properly.\n\nReviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "31dfbc93923c0aaa0440b809f80ff2830c6a531a", "tree": "9527de1c84f266ea23d6009c454d7cf1d1769889", "parents": [ "73758a5d51280ca0613b8380fc07351f4d64f9c8" ], "author": { "name": "Chris Wilson", "email": "chris@chris-wilson.co.uk", "time": "Mon Sep 27 21:28:30 2010 +0100" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Tue Sep 28 09:14:34 2010 +1000" }, "message": "drm: Prune GEM vma entries\n\nHook the GEM vm open/close ops into the generic drm vm open/close so\nthat the private vma entries are created and destroy appropriately.\nFixes the leak of the drm_vma_entries during the lifetime of the filp.\n\nReported-by: Matt Mackall \u003cmpm@selenic.com\u003e\nCc: Jesse Barnes \u003cjbarnes@virtuousgeek.org\u003e\nSigned-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nAcked-by: Jesse Barnes \u003cjbarnes@virtuousgeek.org\u003e\nCc: stable@kernel.org\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "cbc60ca04b342a4e1f2a1086a7277c077f07dbed", "tree": "3979a1304b20cc90a7ec1dce593657435aa48c8d", "parents": [ "df8fcb09667c1b2c9dcf65de23f0bfa851e8138e" ], "author": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Mon Aug 23 22:53:28 2010 +0200" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Aug 30 09:44:56 2010 +1000" }, "message": "drm: kill get_reg_ofs callback\n\nEvery driver used the default implementation. Fold that one into\nthe only callsite and drop the callback.\n\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "df8fcb09667c1b2c9dcf65de23f0bfa851e8138e", "tree": "c49982905d4f9e7bffb24b9ca8d54b47c406302e", "parents": [ "89c372647d1d698a96e2189ef4312a977b939839" ], "author": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Mon Aug 23 22:53:32 2010 +0200" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Aug 30 09:44:54 2010 +1000" }, "message": "drm: kill dev-\u003etimer\n\nTotally unused.\n\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "89c372647d1d698a96e2189ef4312a977b939839", "tree": "a9ac9335b7c9079c6d119c4de318af3e83368512", "parents": [ "8f879194f88742d9c452f669482b6d6abdc1e1e7" ], "author": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Mon Aug 23 22:53:36 2010 +0200" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Aug 30 09:44:40 2010 +1000" }, "message": "drm: kill agp indirection mess\n\nThere\u0027s no point in jumping through two indirections. So kill one\nand call the kernels agp functions directly.\n\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "8f879194f88742d9c452f669482b6d6abdc1e1e7", "tree": "d50a7d96a2150c2c7df7b116f4eb478a2c28f6cc", "parents": [ "4ac5ec40ec70022e4dea8cc6254d2dadd1e43d57" ], "author": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Mon Aug 23 22:53:31 2010 +0200" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Aug 30 09:39:11 2010 +1000" }, "message": "drm: replace drawable ioctl by noops\n\nThe information supplied by userspace through these ioctls is only\naccessible by dev-\u003edrw_idr. But there\u0027s no in-tree user of that.\nAlso userspace does not really care about return values of these ioctls,\neither. Only hw/xfree86/dri/dri.c from the xserver actually checks the\nreturn from adddraw and keeps on trying to create a kernel drawable\nevery time somebody creates a dri drawable. But since that\u0027s now a noop,\nwho cares.\n\nTherefore it\u0027s safe to replace these three ioctls with noops and rip\nout the implementation.\n\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nReviewed-by: Kristian Høgsberg \u003ckrh@bitplanet.net\u003e\nReviewed-by: Michel Dänzer \u003cmichel@daenzer.net\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "4ac5ec40ec70022e4dea8cc6254d2dadd1e43d57", "tree": "fbe2a7b487a0b0d13275c45c2c7570945370a26a", "parents": [ "793a97e4cc38f834e0488ccc1ecbfe52ff6f5b84" ], "author": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Mon Aug 23 22:53:34 2010 +0200" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Aug 30 09:39:00 2010 +1000" }, "message": "drm: don\u0027t export dri1 locking functions\n\nOnly used by ioctl, not by any in-tree drivers.\n\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "793a97e4cc38f834e0488ccc1ecbfe52ff6f5b84", "tree": "418e5c4eef9c7566625829ddc823fa36c6a0d65e", "parents": [ "690bb51b54a986e48c7b8b2dba51a3cd262a7266" ], "author": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Mon Aug 23 22:53:27 2010 +0200" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Aug 30 09:38:50 2010 +1000" }, "message": "drm: kill drm_map_ofs callbacks\n\nAll drivers happily copy\u0026pasted the default implementation without\nchecking whether this callback is used at all. It\u0027s not. Sigh.\n\nKill it.\n\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "690bb51b54a986e48c7b8b2dba51a3cd262a7266", "tree": "7943c7eaf853d987dedff50ec507d25cd5da9c11", "parents": [ "a2a273c94357ffd24e635cf9ec9b2e5c6f02b63b" ], "author": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Mon Aug 23 22:53:35 2010 +0200" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Aug 30 09:38:39 2010 +1000" }, "message": "drm: drop return value of drm_free_agp\n\nNo caller (rightly) cares about it, so drop it.\n\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "a2a273c94357ffd24e635cf9ec9b2e5c6f02b63b", "tree": "87c217ff0f3dee41cd994f437a8ac5069174932d", "parents": [ "b3da8f7d2d1fa81fb65cb3f5d9e50dde40a83182" ], "author": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Mon Aug 23 22:53:30 2010 +0200" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Aug 30 09:38:37 2010 +1000" }, "message": "drm: don\u0027t export drm_get_drawable_info\n\nNot used by any in-tree user. So drop it.\n\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "b3da8f7d2d1fa81fb65cb3f5d9e50dde40a83182", "tree": "3fcc266b657a1b16f644eb214e3c774047fdea6f", "parents": [ "fd2e7931cdefa8e9acf63f0a4efd61ae0f89e77b" ], "author": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Mon Aug 23 22:53:29 2010 +0200" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Aug 30 09:38:25 2010 +1000" }, "message": "drm: kill context_ctor callback\n\nIt\u0027s not used by any driver. The destructor callback is unfortunately\nused by the via driver in a rather convoluted piece of code used\nto reimplement something resembling broken futexes. I didn\u0027t dare\nto touch this code. But at least kill the needless NULL assignemt\nin the sis driver.\n\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "fd2e7931cdefa8e9acf63f0a4efd61ae0f89e77b", "tree": "ea2d64ccff37e5d97ed94abab8851ac89dd5d1da", "parents": [ "23ddc0243d7313942b94f1a2e44e6394f7bb996e" ], "author": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Mon Aug 23 22:53:33 2010 +0200" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Aug 30 09:38:18 2010 +1000" }, "message": "drm: kill gem_free_object_unlocked driver callback\n\nNot used by any current driver.\n\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "23ddc0243d7313942b94f1a2e44e6394f7bb996e", "tree": "33519710bd456a48fb8f7a7e572618d5ddc8471b", "parents": [ "be72ae26b11478c00c64858c86b647b438791671" ], "author": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Mon Aug 23 22:53:25 2010 +0200" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Aug 30 09:38:12 2010 +1000" }, "message": "drm: kill dma_ready callbacks\n\nNot used by any driver. So drop it.\n\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "be72ae26b11478c00c64858c86b647b438791671", "tree": "d4490daab3f4c193e7e4768f083c13ca77ac98eb", "parents": [ "1da3f87ebb7edb3e0b829ec4bbe5fb3d9d93986f" ], "author": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Mon Aug 23 22:53:26 2010 +0200" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Aug 30 09:38:08 2010 +1000" }, "message": "drm: kill procfs callbacks\n\nNot used by any driver (rightly so!). Kill them.\n\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "1da3f87ebb7edb3e0b829ec4bbe5fb3d9d93986f", "tree": "af611e50ee6120297e56eed52c041f2ba008ed6d", "parents": [ "59506197c7d0a94088dc15d488d8c964f909a6b1" ], "author": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Mon Aug 23 22:53:24 2010 +0200" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Aug 30 09:37:56 2010 +1000" }, "message": "drm: kill kernel_context_switch callbacks\n\nNot used by any in-kernel driver. So drop it.\n\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "1b2f1489633888d4a06028315dc19d65768a1c05", "tree": "8d5d3bfec854d6c96c19a781bf9c873409b74210", "parents": [ "b9f0aee83335db1f3915f4e42a5e21b351740afd" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Sat Aug 14 20:20:34 2010 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Tue Aug 17 14:52:25 2010 +1000" }, "message": "drm: block userspace under allocating buffer and having drivers overwrite it (v2)\n\nWith the current screwed but its ABI, ioctls for the drm, Linus pointed out that we could allow userspace to specify the allocation size, but we pass it to the driver which then uses it blindly to store a struct. Now if userspace specifies the allocation size as smaller than the driver needs, the driver can possibly overwrite memory.\n\nThis patch restructures the driver ioctls so we store the structure size we are expecting, and make sure we allocate at least that size. The copy from/to userspace are still restricted to the size the user specifies, this allows ioctl structs to grow on both sides of the equation.\n\nUp until now we didn\u0027t really use the DRM_IOCTL defines in the kernel, so this cleans them up and adds them for nouveau.\n\nv2:\nfix nouveau pushbuf arg (thanks to Ben for pointing it out)\n\nReported-by: Linus Torvalds \u003ctorvalds@linuxfoundation.org\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "c17c2f892e23a3c70281a4e79bd7c2c226b1a561", "tree": "73e048ae853103479e3ba747809fc8c38f6b7de8", "parents": [ "c3b6ef8633e75bdcde8e845548e3b95e9a21cdf6" ], "author": { "name": "Benjamin Herrenschmidt", "email": "benh@kernel.crashing.org", "time": "Fri Aug 06 13:55:10 2010 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Tue Aug 10 08:20:20 2010 +1000" }, "message": "drm: Fix support for PCI domains\n\n(For some reason I thought that went in ages ago ...)\n\nThis fixes support for PCI domains in what should hopefully be a backward\ncompatible way along with a change to libdrm.\n\nWhen the interface version is set to 1.4, we assume userspace understands\ndomains and the world is at peace. We thus pass proper domain numbers\ninstead of 0 to userspace.\n\nThe newer libdrm will then try 1.4 first, and fallback to 1.1, along with\nignoring domains in the later case (well, except on alpha of course)\n\nSigned-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "58374713c9dfb4d231f8c56cac089f6fbdedc2ec", "tree": "9ba01c3990b5a8625437d13722595f5063f2d915", "parents": [ "a1e09b62592eb57e25f8d076ffa5b7bef18be812" ], "author": { "name": "Arnd Bergmann", "email": "arnd@arndb.de", "time": "Sat Jul 10 23:51:39 2010 +0200" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Aug 05 11:54:40 2010 +1000" }, "message": "drm: kill BKL from common code\n\nThis restricts the use of the big kernel lock to the i830 and i810\ndevice drivers. The three remaining users in common code (open, ioctl\nand release) get converted to a new mutex, the drm_global_mutex,\nmaking the locking stricter than the big kernel lock.\n\nThis may have a performance impact, but only in those cases that\ncurrently don\u0027t use DRM_UNLOCKED flag in the ioctl list and would\nbenefit from that anyway.\n\nThe reason why i810 and i830 cannot use drm_global_mutex in their\nmmap functions is a lock-order inversion problem between the current\nuse of the BKL and mmap_sem in these drivers. Since the BKL has\nrelease-on-sleep semantics, it\u0027s harmless but it would cause trouble\nif we replace the BKL with a mutex.\n\nInstead, these drivers get their own ioctl wrappers that take the\nBKL around every ioctl call and then set their own handlers as\nDRM_UNLOCKED.\n\nSigned-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\nCc: David Airlie \u003cairlied@linux.ie\u003e\nCc: dri-devel@lists.freedesktop.org\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "ba4420c224c2808f2661cf8428f43ceef7a73a4a", "tree": "67b32498483c797c76e8f504a83746bdef35db7d", "parents": [ "2581afccadd347bf97c3a5620ba72c99aca8c355" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Tue Mar 09 10:56:52 2010 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Wed Aug 04 09:46:06 2010 +1000" }, "message": "drm: move ttm global code to core drm\n\nI wrote this for the prime sharing work, but I also noticed other external\nnon-upstream drivers from a large company carrying a similiar patch, so I\nmay as well ship it in master.\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "102e73463e1bb921b3a829a1392a29dba16bbabd", "tree": "8c569c1a6a7fd3ffc21cf1780bd8fa62be5b23b2", "parents": [ "5c8d7171cc4984351af802a525675d50ae555a7b", "e5510fac98a706c424034950f55bb5e819c46f51" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Wed Jul 07 18:38:44 2010 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Wed Jul 07 18:38:44 2010 +1000" }, "message": "Merge branch \u0027drm-tracepoints\u0027 into drm-testing\n" }, { "commit": "b9c2c9ae882f058084e13e339925dbf8d2d20271", "tree": "2c3baccc55ef03477a4fc3e08473ea35931e56ec", "parents": [ "ac2874b980e05ed7a4ea8fed7b0a92428b51ce58" ], "author": { "name": "Jesse Barnes", "email": "jbarnes@virtuousgeek.org", "time": "Thu Jul 01 16:48:09 2010 -0700" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Jul 02 14:03:24 2010 +1000" }, "message": "drm: add per-event vblank event trace points\n\nAllows us to track each process that requests and completes events.\n\nSigned-off-by: Jesse Barnes \u003cjbarnes@virtuousgeek.org\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "dcdb167402cbdca1d021bdfa5f63995ee0a79317", "tree": "3cd0ab1189d972b59daaaf863f518d7b94d2de7a", "parents": [ "01d73a6967f12fe6c4bbde1834a9fe662264a2eb" ], "author": { "name": "Jordan Crouse", "email": "jcrouse@codeaurora.org", "time": "Thu May 27 13:40:25 2010 -0600" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Tue Jun 01 10:07:39 2010 +1000" }, "message": "drm: Add support for platform devices to register as DRM devices\n\nAllow platform devices without PCI resources to be DRM devices.\n\n[airlied: fixup warnings with dev pointers]\n\nSigned-off-by: Jordan Crouse \u003cjcrouse@codeaurora.org\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "01d73a6967f12fe6c4bbde1834a9fe662264a2eb", "tree": "32e4a3171cd371dfc037af52e181e6c8454a961b", "parents": [ "cf22f20ade30f8c03955324aaf27b1049e182600" ], "author": { "name": "Jordan Crouse", "email": "jcrouse@codeaurora.org", "time": "Thu May 27 13:40:24 2010 -0600" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Tue Jun 01 10:07:24 2010 +1000" }, "message": "drm: Remove drm_resource wrappers\n\nRemove the drm_resource wrappers and directly use the\nactual PCI and/or platform functions in their place.\n\n[airlied: fixup nouveau properly to build]\n\nSigned-off-by: Jordan Crouse \u003cjcrouse@codeaurora.org\u003e\nReviewed-by: Matt Turner \u003cmattst88@gmail.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "fd632aa34c8592fb1d37fc83cbffa827bc7dd42c", "tree": "5c6a0fc186e297590aee2c050b837a721db337b2", "parents": [ "1d397043bcc2c8cdccb584a8ef73131f28f18e4c" ], "author": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Fri Apr 09 19:05:05 2010 +0000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Tue Apr 20 13:19:33 2010 +1000" }, "message": "drm: free core gem object from driver callbacks\n\nWhen drivers embed the core gem object into their own structures,\nthey\u0027ll have to do this. Temporarily this results in an ugly\n\nkfree(gem_obj);\n\nin every gem driver.\n\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nAcked-by: Eric Anholt \u003ceric@anholt.net\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "1d397043bcc2c8cdccb584a8ef73131f28f18e4c", "tree": "564c09e78bffef21ff65988ea76380db132cc02a", "parents": [ "153549b8b63d71a9c5d8cbde887097b995c32bd6" ], "author": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Fri Apr 09 19:05:04 2010 +0000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Tue Apr 20 13:19:25 2010 +1000" }, "message": "drm: extract drm_gem_object_init\n\nThis function can be used by drivers who allocate the drm gem object\non their own. No functional change in here, just preparation.\n\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nAcked-by: Eric Anholt \u003ceric@anholt.net\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "336f5899d287f06d8329e208fc14ce50f7ec9698", "tree": "9b762d450d5eb248a6ff8317badb7e223d93ed58", "parents": [ "a4ab2773205e8b94c18625455f85e3b6bb9d7ad6", "db217dece3003df0841bacf9556b5c06aa097dae" ], "author": { "name": "Tejun Heo", "email": "tj@kernel.org", "time": "Mon Apr 05 11:37:28 2010 +0900" }, "committer": { "name": "Tejun Heo", "email": "tj@kernel.org", "time": "Mon Apr 05 11:37:28 2010 +0900" }, "message": "Merge branch \u0027master\u0027 into export-slabh\n" }, { "commit": "5a0e3ad6af8660be21ca98a971cd00f331318c05", "tree": "5bfb7be11a03176a87296a43ac6647975c00a1d1", "parents": [ "ed391f4ebf8f701d3566423ce8f17e614cde9806" ], "author": { "name": "Tejun Heo", "email": "tj@kernel.org", "time": "Wed Mar 24 17:04:11 2010 +0900" }, "committer": { "name": "Tejun Heo", "email": "tj@kernel.org", "time": "Tue Mar 30 22:02:32 2010 +0900" }, "message": "include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h\n\npercpu.h is included by sched.h and module.h and thus ends up being\nincluded when building most .c files. percpu.h includes slab.h which\nin turn includes gfp.h making everything defined by the two files\nuniversally available and complicating inclusion dependencies.\n\npercpu.h -\u003e slab.h dependency is about to be removed. Prepare for\nthis change by updating users of gfp and slab facilities include those\nheaders directly instead of assuming availability. As this conversion\nneeds to touch large number of source files, the following script is\nused as the basis of conversion.\n\n http://userweb.kernel.org/~tj/misc/slabh-sweep.py\n\nThe script does the followings.\n\n* Scan files for gfp and slab usages and update includes such that\n only the necessary includes are there. ie. if only gfp is used,\n gfp.h, if slab is used, slab.h.\n\n* When the script inserts a new include, it looks at the include\n blocks and try to put the new include such that its order conforms\n to its surrounding. It\u0027s put in the include block which contains\n core kernel includes, in the same order that the rest are ordered -\n alphabetical, Christmas tree, rev-Xmas-tree or at the end if there\n doesn\u0027t seem to be any matching order.\n\n* If the script can\u0027t find a place to put a new include (mostly\n because the file doesn\u0027t have fitting include block), it prints out\n an error message indicating which .h file needs to be added to the\n file.\n\nThe conversion was done in the following steps.\n\n1. The initial automatic conversion of all .c files updated slightly\n over 4000 files, deleting around 700 includes and adding ~480 gfp.h\n and ~3000 slab.h inclusions. The script emitted errors for ~400\n files.\n\n2. Each error was manually checked. Some didn\u0027t need the inclusion,\n some needed manual addition while adding it to implementation .h or\n embedding .c file was more appropriate for others. This step added\n inclusions to around 150 files.\n\n3. The script was run again and the output was compared to the edits\n from #2 to make sure no file was left behind.\n\n4. Several build tests were done and a couple of problems were fixed.\n e.g. lib/decompress_*.c used malloc/free() wrappers around slab\n APIs requiring slab.h to be added manually.\n\n5. The script was run on all .h files but without automatically\n editing them as sprinkling gfp.h and slab.h inclusions around .h\n files could easily lead to inclusion dependency hell. Most gfp.h\n inclusion directives were ignored as stuff from gfp.h was usually\n wildly available and often used in preprocessor macros. Each\n slab.h inclusion directive was examined and added manually as\n necessary.\n\n6. percpu.h was updated not to include slab.h.\n\n7. Build test were done on the following configurations and failures\n were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my\n distributed build env didn\u0027t work with gcov compiles) and a few\n more options had to be turned off depending on archs to make things\n build (like ipr on powerpc/64 which failed due to missing writeq).\n\n * x86 and x86_64 UP and SMP allmodconfig and a custom test config.\n * powerpc and powerpc64 SMP allmodconfig\n * sparc and sparc64 SMP allmodconfig\n * ia64 SMP allmodconfig\n * s390 SMP allmodconfig\n * alpha SMP allmodconfig\n * um on x86_64 SMP allmodconfig\n\n8. percpu.h modifications were reverted so that it could be applied as\n a separate patch and serve as bisection point.\n\nGiven the fact that I had only a couple of failures from tests on step\n6, I\u0027m fairly confident about the coverage of this conversion patch.\nIf there is a breakage, it\u0027s likely to be something in one of the arch\nheaders which should be easily discoverable easily on most builds of\nthe specific arch.\n\nSigned-off-by: Tejun Heo \u003ctj@kernel.org\u003e\nGuess-its-ok-by: Christoph Lameter \u003ccl@linux-foundation.org\u003e\nCc: Ingo Molnar \u003cmingo@redhat.com\u003e\nCc: Lee Schermerhorn \u003cLee.Schermerhorn@hp.com\u003e\n" }, { "commit": "72e942dd846f98e2d35aad5436d77a878ef05c5e", "tree": "bed31b516da9fd80058800caa6161b6e0f880595", "parents": [ "d424b925f7092b9d95e0a8556872349abe79d9b6" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Tue Mar 09 06:33:26 2010 +0000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Mar 15 10:31:43 2010 +1000" }, "message": "drm/ttm: use drm calloc large and free large\n\nNow that the drm core can do this, lets just use it, split the code out\nso TTM doesn\u0027t have to drag all of drmP.h in.\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "c3ae90c099bb62387507e86da7cf799850444b08", "tree": "4f546a65e35fddea09a97fbfd28ce724ea86f7c2", "parents": [ "77c1ff3982c6b36961725dd19e872a1c07df7f3b" ], "author": { "name": "Luca Barbieri", "email": "luca@luca-barbieri.com", "time": "Tue Feb 09 05:49:11 2010 +0000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Feb 11 14:21:24 2010 +1000" }, "message": "drm: introduce drm_gem_object_[handle_]unreference_unlocked\n\nThis patch introduces the drm_gem_object_unreference_unlocked\nand drm_gem_object_handle_unreference_unlocked functions that\ndo not require holding struct_mutex.\n\ndrm_gem_object_unreference_unlocked calls the new\n-\u003egem_free_object_unlocked entry point if available, and\notherwise just takes struct_mutex and just calls -\u003egem_free_object\n\nSigned-off-by: Luca Barbieri \u003cluca@luca-barbieri.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "e6be8d9d17bd44061116f601fe2609b3ace7aa69", "tree": "85b8acc2fd4734724b9d5202016743461e02ee07", "parents": [ "29ebdf925c2c45f6531a953c6c5c8e4d3b4ac2dc" ], "author": { "name": "Zhenyu Wang", "email": "zhenyu.z.wang@intel.com", "time": "Tue Jan 05 11:25:05 2010 +0800" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Jan 07 13:15:50 2010 +1000" }, "message": "drm: remove address mask param for drm_pci_alloc()\n\ndrm_pci_alloc() has input of address mask for setting pci dma\nmask on the device, which should be properly setup by drm driver.\nAnd leave it as a param for drm_pci_alloc() would cause confusion\nor mistake would corrupt the correct dma mask setting, as seen on\nintel hw which set wrong dma mask for hw status page. So remove\nit from drm_pci_alloc() function.\n\nSigned-off-by: Zhenyu Wang \u003czhenyuw@linux.intel.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "ed8b67040965e4fe695db333d5914e18ea5f146f", "tree": "6fd9f50130f7ccaccf8070f7821ea8eb6743854c", "parents": [ "dcd6dfcfe959bade75fbf49499775985d2cac5d5" ], "author": { "name": "Arnd Bergmann", "email": "arnd@arndb.de", "time": "Wed Dec 16 22:17:09 2009 +0000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Dec 18 11:22:31 2009 +1000" }, "message": "drm: convert drm_ioctl to unlocked_ioctl\n\ndrm_ioctl is called with the Big Kernel Lock held,\nwhich shows up very high in statistics on vfs_ioctl.\n\nMoving the lock into the drm_ioctl function itself\nmakes sure we blame the right subsystem and it gets\nus one step closer to eliminating the locked version\nof fops-\u003eioctl.\n\nSince drm_ioctl does not require the lock itself,\nwe only need to hold it while calling the specific\nhandler. The 32 bit conversion handlers do not\ninteract with any other code, so they don\u0027t need\nthe BKL here either and can just call drm_ioctl.\n\nAs a bonus, this cleans up all the other users\nof drm_ioctl which now no longer have to find\nthe inode or call lock_kernel.\n\n[airlied: squashed the non-driver bits\nof the second patch in here, this provides\nthe flag for drivers to use to select unlocked\nioctls - but doesn\u0027t modify any drivers].\n\nSigned-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\nCc: David Airlie \u003cairlied@linux.ie\u003e\nCc: dri-devel@lists.sourceforge.net\nCc: Frederic Weisbecker \u003cfweisbec@gmail.com\u003e\nCc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "3ff99164f67aae78a2bd2313f65ad55bddb1ffea", "tree": "d6bba03616d1be6ab9e6d9e92641a6f4047e1e15", "parents": [ "1bd049fa895f9c6743f38b52ce14775f5a31ea63", "f2b115e69d46344ae7afcaad5823496d2a0d8650" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Tue Dec 08 14:03:47 2009 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Tue Dec 08 14:03:47 2009 +1000" }, "message": "Merge remote branch \u0027anholt/drm-intel-next\u0027 into drm-linus\n\nThis merges the upstream Intel tree and fixes up numerous conflicts\ndue to patches merged into Linus tree later in -rc cycle.\n\nConflicts:\n\tdrivers/char/agp/intel-agp.c\n\tdrivers/gpu/drm/drm_dp_i2c_helper.c\n\tdrivers/gpu/drm/i915/i915_irq.c\n\tdrivers/gpu/drm/i915/i915_suspend.c\n" }, { "commit": "1a95916f5465ad6c91398f17924949db7e0b5c36", "tree": "e9293ebb277e8cb39dcd57d9c1c07c68a0c8d803", "parents": [ "862302ffe422378a5213f558fc5cdf62c37050a9" ], "author": { "name": "Kristian Høgsberg", "email": "krh@bitplanet.net", "time": "Wed Dec 02 12:13:48 2009 -0500" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Dec 04 08:59:28 2009 +1000" }, "message": "drm: Add compatibility #ifdefs for *BSD\n\nThis let\u0027s use use the linux drm headers as the canonical source for\nlibdrm on all platforms.\n\nSigned-off-by: Kristian Høgsberg \u003ckrh@bitplanet.net\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "862302ffe422378a5213f558fc5cdf62c37050a9", "tree": "13557874eb479023e5a64f12990416045ea60818", "parents": [ "9340d8cfeacd16cef1cbe94527f7baaed7640669" ], "author": { "name": "Thomas Hellstrom", "email": "thellstrom@vmware.com", "time": "Wed Dec 02 18:15:25 2009 +0000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Dec 04 08:55:46 2009 +1000" }, "message": "drm: Add support for drm master_[set|drop] callbacks.\n\nThe vmwgfx driver has a per master rw lock around TTM, to guarantee \nmutual exclusion when needed.\n\nThis is typically when all evictable buffers are evicted due to\n\n1) vt switch\n2) master switch\n3) suspend / resume.\n\nIn the multi-master case, on master switch the new master takes the \npreviously active master lock in write mode, and then evicts all \nbuffers. Any clients to previous masters will then block on that lock \nwhen trying to validate a buffer. fbdev also acts as a virtual master\nwrt this.\n\nSigned-off-by: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\nSigned-off-by: Jakob Bornecrantz \u003cjakob@vmware.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "778c902640530371a169ad1c03566e7c51b09874", "tree": "4230fb7e4f79e00931c1efcf2b88cfec285ae7ff", "parents": [ "27dfaf4f5825a119305db1bc63bef30ed400e376" ], "author": { "name": "Li Peng", "email": "peng.li@linux.intel.com", "time": "Mon Nov 09 12:51:22 2009 +0800" }, "committer": { "name": "Eric Anholt", "email": "eric@anholt.net", "time": "Tue Dec 01 10:27:40 2009 -0800" }, "message": "drm/i915: Fix sync to vblank when VGA output is turned off\n\nIn current vblank-wait implementation, if we turn off VGA output,\ndrm_wait_vblank will still wait on the disabled pipe until timeout,\nbecause vblank on the pipe is assumed be enabled. This would cause\nslow system response on some system such as moblin.\n\nThis patch resolve the issue by adding a drm helper function\ndrm_vblank_off which explicitly clear vblank_enabled[crtc], wake up\nany waiting queue and save last vblank counter before turning off\ncrtc. It also slightly change drm_vblank_get to ensure that we will\nwill return immediately if trying to wait on a disabled pipe.\n\nSigned-off-by: Li Peng \u003cpeng.li@intel.com\u003e\nReviewed-by: Jesse Barnes \u003cjbarnes@virtuousgeek.org\u003e\n[anholt: hand-applied for conflicts with overlay changes]\nSigned-off-by: Eric Anholt \u003ceric@anholt.net\u003e\n" }, { "commit": "f40d6817a5c2bf84f5fe7b5d1a83f1e8f8669951", "tree": "1c515a34a60f65cbfd3cf1a387427d0a9fdf878f", "parents": [ "103a196f4224dc6872081305cf7f82ebf67aa7bd", "46557bef3f3834ac33031c7be27d39d90d507442" ], "author": { "name": "Eric Anholt", "email": "eric@anholt.net", "time": "Tue Dec 01 09:01:54 2009 -0800" }, "committer": { "name": "Eric Anholt", "email": "eric@anholt.net", "time": "Tue Dec 01 09:01:54 2009 -0800" }, "message": "Merge remote branch \u0027airlied/drm-next\u0027 into drm-intel-next\n" }, { "commit": "c8e0f93a381d1d76135e567f13a4418fce66fd95", "tree": "efa286f2e381d60b514c897cd148e5a824cce667", "parents": [ "5b8f0be0dce012d190a53d55240fe3fde6306476" ], "author": { "name": "Eric Anholt", "email": "eric@anholt.net", "time": "Sun Nov 22 03:49:37 2009 +0100" }, "committer": { "name": "Eric Anholt", "email": "eric@anholt.net", "time": "Wed Nov 25 06:36:21 2009 -0800" }, "message": "drm/i915: Replace a calloc followed by copying data over it with malloc.\n\nExecbufs involve quite a bit of payload, to the extent that cache misses\nshow up in the profiles here, and a suspicion that some of those cachelines\nmay get evicted and then reloaded in the subsequent copy.\n\nThis is still abstracted like drm_calloc_large since we want to check for\nsize overflow, and because we want to choose between kmalloc and vmalloc\non the fly. cairo\u0027s interface for malloc-with-calloc\u0027s-args was used as\nthe model.\n\nSigned-off-by: Eric Anholt \u003ceric@anholt.net\u003e\n" }, { "commit": "156822f7175d9ceb9d7e808502d3c5de8841e047", "tree": "1eb79f8e11c1ce4710dcdd83210d1acb5cd95b39", "parents": [ "420a457088669e055e767dfb8468909cd1799cf9" ], "author": { "name": "Andres Salomon", "email": "dilinger@collabora.co.uk", "time": "Tue Nov 17 14:41:23 2009 -0800" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Wed Nov 18 10:02:49 2009 +1000" }, "message": "drm: kill more unused DRM macros\n\nThere are a few more macros in drmP.h that are unused; DRM_GET_PRIV_SAREA,\nDRM_ARRAY_SIZE, and DRM_WAITCOUNT can go away completely.\n\nUnfortunately, DRM_COPY is still used in one place, but we can at least\nmove it to where it\u0027s used. It\u0027s an awful looking macro..\n\n[akpm: fix overeagerness]\nSigned-off-by: Andres Salomon \u003cdilinger@collabora.co.uk\u003e\nCc: Dave Airlie \u003cairlied@linux.ie\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "420a457088669e055e767dfb8468909cd1799cf9", "tree": "6cc3c7e467ff86ea86015ef1acf55019bff91cb7", "parents": [ "4a9216453c8537a7f43a3b1708509b9dd271dc9f" ], "author": { "name": "Andres Salomon", "email": "dilinger@collabora.co.uk", "time": "Tue Nov 17 14:41:23 2009 -0800" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Wed Nov 18 10:02:49 2009 +1000" }, "message": "drm: kill some unused DRM_PROC macros from drmP.h\n\ni915_gem_proc.c appears to have been the last user of the DRM_PROC_*\nmacros, and it has gone away. The macros should die as well.\n\nSigned-off-by: Andres Salomon \u003cdilinger@collabora.co.uk\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "c9a9c5e02aedc1a2815877b0268f886d2640b771", "tree": "51f69f2ec6dcebb019cd6aa22b9e29ee48c397c1", "parents": [ "799dd75b1a8380a967c929a4551895788c374b31" ], "author": { "name": "Kristian Høgsberg", "email": "krh@bitplanet.net", "time": "Sat Sep 12 04:33:34 2009 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Wed Nov 18 10:02:47 2009 +1000" }, "message": "drm: Add async event synchronization for drmWaitVblank\n\nThis patch adds a new flag to the drmWaitVblank ioctl, which asks the drm\nto return immediately and notify userspace when the specified vblank sequence\nhappens by sending an event back on the drm fd.\n\nThe event mechanism works with the other flags supported by the ioctls,\nspecifically, the vblank sequence can be specified relatively or absolutely,\nand works for primary and seconday crtc.\n\nThe signal field of the vblank request is used to provide user data,\nwhich will be sent back to user space in the vblank event.\n\nSigned-off-by: Kristian Høgsberg \u003ckrh@redhat.com\u003e\nReviewed-by: Jesse Barnes \u003cjbarnes@virtuousgeek.org\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "5b8f0be0dce012d190a53d55240fe3fde6306476", "tree": "19fccf22b07730205e9b1d64e8e8f676a12b763a", "parents": [ "43bcd61fae05fc6062b4f117c5adb1a72c9f8c57", "4fe9676d1ae6639b5749140eba052261d363366b" ], "author": { "name": "Eric Anholt", "email": "eric@anholt.net", "time": "Thu Nov 05 15:04:06 2009 -0800" }, "committer": { "name": "Eric Anholt", "email": "eric@anholt.net", "time": "Thu Nov 05 15:04:06 2009 -0800" }, "message": "Merge remote branch \u0027airlied/drm-next\u0027 into drm-intel-next\n" }, { "commit": "c182be37ed7cb04c344501b88b8fdb747016e6cf", "tree": "013ae7e4c5bb512ac8e4132a642f40c27f03390e", "parents": [ "0a5c1e61dbaceb6ce56281a3128a6912b0dcd043" ], "author": { "name": "Kristian Høgsberg", "email": "krh@bitplanet.net", "time": "Sat Sep 12 04:33:34 2009 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Oct 26 13:29:27 2009 +1000" }, "message": "drm: Add async event synchronization for drmWaitVblank\n\nThis patch adds a new flag to the drmWaitVblank ioctl, which asks the drm\nto return immediately and notify userspace when the specified vblank sequence\nhappens by sending an event back on the drm fd.\n\nThe event mechanism works with the other flags supported by the ioctls,\nspecifically, the vblank sequence can be specified relatively or absolutely,\nand works for primary and seconday crtc.\n\nThe signal field of the vblank request is used to provide user data,\nwhich will be sent back to user space in the vblank event.\n\nSigned-off-by: Kristian Høgsberg \u003ckrh@redhat.com\u003e\nReviewed-by: Jesse Barnes \u003cjbarnes@virtuousgeek.org\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "28d520433b6375740990ab99d69b0d0067fd656b", "tree": "d3affea0fb69f84b9c924c03d78c5df23946306f", "parents": [ "aadd4e17452d3d5c2269cd2b000b7de7cfb6c79e" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Sep 21 14:33:58 2009 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Sep 21 15:00:27 2009 +1000" }, "message": "drm/vgaarb: add VGA arbitration support to the drm and kms.\n\nVGA arb requires DRM support for non-kms drivers, to turn on/off\nirqs when disabling the mem/io regions.\n\nVGA arb requires KMS support for GPUs where we can turn off VGA\ndecoding. Currently we know how to do this for intel and radeon\nkms drivers, which allows them to be removed from the arbiter.\n\nThis patch comes from Fedora rawhide kernel.\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "a1a2d1d32250f6fcc317419e9dfb4a5a6946d2e6", "tree": "3ef2ff19e439764d73c78ed77f15cce554920a5f", "parents": [ "3f7dc91adef90a20f806282724c40d68d5b020aa" ], "author": { "name": "Pekka Paalanen", "email": "pq@iki.fi", "time": "Sun Aug 23 12:40:55 2009 +0300" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Aug 27 11:21:08 2009 +1000" }, "message": "drm: GEM handles are u32, not int\n\nSeveral functions in the GEM kernel API used int as handle type, but\nuser API has it __u32 which is also the intended type.\n\nReplace int with u32.\n\nSigned-off-by: Pekka Paalanen \u003cpq@iki.fi\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "87fdff81cd2d770f0adc742e21eb5e062ad20def", "tree": "5fe04660ad959d7d15eeb7969dfb21c4f4965ff1", "parents": [ "f940f37f022f7392ab81a35516222cbd46110b42" ], "author": { "name": "Zhao Yakui", "email": "yakui.zhao@intel.com", "time": "Mon Jul 20 13:48:06 2009 +0800" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Tue Aug 04 14:02:41 2009 +1000" }, "message": "DRM: Add the explanation about DRM debug level\n\nAdd the explanation about DRM debug level in the drmP header file. This is to\nexplain how/where to use the different DRM debug level.\n\nSigned-off-by: Zhao Yakui \u003cyakui.zhao@intel.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "f940f37f022f7392ab81a35516222cbd46110b42", "tree": "1dac5677cee83eb6f9cf38bf6b8f0419c5df01e4", "parents": [ "8a4c47f346cc7a12d0897c05eb3cc1add26b487f" ], "author": { "name": "Zhao Yakui", "email": "yakui.zhao@intel.com", "time": "Mon Jul 20 13:48:05 2009 +0800" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Tue Aug 04 14:02:39 2009 +1000" }, "message": "drm: Remove the macro defintion of DRM_DEBUG_MODE\n\nTwo macro definitions of DRM_DEBUG_KMS/MODE can be used to add the debug\ninfo related with KMS. It is confusing.\nSo remove the macro definition of DRM_DEBUG_MODE. Instead it can be replaced\nby the DRM_DEBUG_KMS.\n\nSigned-off-by: Zhao Yakui \u003cyakui.zhao@intel.com\u003e\nAcked-by: Ian Romanick \u003cian.d.romanick@intel.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "8a4c47f346cc7a12d0897c05eb3cc1add26b487f", "tree": "3beb08565466a2aa5649debd9dad411388afa53a", "parents": [ "5c61259e6c7290082108e93815f7f72f27da14f4" ], "author": { "name": "Zhao Yakui", "email": "yakui.zhao@intel.com", "time": "Mon Jul 20 13:48:04 2009 +0800" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Tue Aug 04 14:02:31 2009 +1000" }, "message": "drm: Remove the unused prefix in DRM_DEBUG_KMS/DRIVER/MODE\n\nWe will have to add a prefix when using the macro defintion of DRM_DEBUG_KMS\n/DRM_DEBUG_DRIVER/MODE. It is not convenient. We should use the DRM_NAME\nas default prefix.\nSo remove the prefix in the macro definition of DRM_DEBUG_KMS/DRIVER/MODE.\n\nSigned-off-by: Zhao Yakui \u003cyakui.zhao@intel.com\u003e\nAcked-by: Ian Romanick \u003cian.d.romanick@intel.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "9a298b2acd771d8a5c0004d8f8e4156c65b11f6b", "tree": "f8bf6cfe675492cb881abb50d9aa9aa8f38733d6", "parents": [ "52dc7d32b88156248167864f77a9026abe27b432" ], "author": { "name": "Eric Anholt", "email": "eric@anholt.net", "time": "Tue Mar 24 12:23:04 2009 -0700" }, "committer": { "name": "Eric Anholt", "email": "eric@anholt.net", "time": "Thu Jun 18 13:00:33 2009 -0700" }, "message": "drm: Remove memory debugging infrastructure.\n\nIt hasn\u0027t been used in ages, and having the user tell your how much\nmemory is being freed at free time is a recipe for disaster even if it\nwas ever used.\n\nSigned-off-by: Eric Anholt \u003ceric@anholt.net\u003e\n" }, { "commit": "249d6048ca98b5452105b0824abac1275661b8e3", "tree": "5e8e89288ec6f6a278fcb819ea49d0d9984576fd", "parents": [ "715cbb05c935e8a4306a730d14a72d5af881523e" ], "author": { "name": "Jerome Glisse", "email": "glisse@freedesktop.org", "time": "Wed Apr 08 17:11:16 2009 +0200" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Jun 12 15:56:31 2009 +1000" }, "message": "drm: Split out the mm declarations in a separate header. Add atomic operations.\n\nthis is a TTM preparation patch, it rearranges the mm and\nadd operations needed to do mm operations in atomic context.\n\nSigned-off-by: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "fbe0efb869efde8d847ede3a925230ef88910086", "tree": "fb7df90bd25a77555db351a6db632e52aab311e2", "parents": [ "61f11699e7a92d932b31ded3715ad4f70eb26ef2" ], "author": { "name": "Kristian Høgsberg", "email": "krh@redhat.com", "time": "Tue Jun 09 01:50:41 2009 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Jun 12 15:37:28 2009 +1000" }, "message": "drm_calloc_large: check right size, check integer overflow, use GFP_ZERO\n\nPreviously we would check size instead of size * nmemb, and so would\nnever hit the vmalloc path. Also add integer overflow check as in kcalloc,\nand allocate GFP_ZERO pages instead of memset()ing them.\n\nSigned-off-by: Kristian Høgsberg \u003ckrh@redhat.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "4fefcb27050b98c97b1c32bc710fc2f874449dee", "tree": "1983f5ca546dab13d56ad8207762fee6088c9079", "parents": [ "df4f7fe7bd516b3833e25c692c3970e22038a6ca" ], "author": { "name": "yakui_zhao", "email": "yakui.zhao@intel.com", "time": "Tue Jun 02 14:09:47 2009 +0800" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Jun 11 18:36:36 2009 +1000" }, "message": "drm: add separate drm debugging levels\n\nNow all the DRM debug info will be reported if the boot option of\n\"drm.debug\u003d1\" is added. Sometimes it is inconvenient to get the debug\ninfo in KMS mode. We will get too much unrelated info.\n\nThis will separate several DRM debug levels and the debug level can be used\nto print the different debug info. And the debug level is controlled by the\nmodule parameter of drm.debug\n\nIn this patch it is divided into four debug levels;\n \tdrm_core, drm_driver, drm_kms, drm_mode.\n\nAt the same time we can get the different debug info by changing the debug\nlevel. This can be done by adding the module parameter. Of course it can\nbe changed through the /sys/module/drm/parameters/debug after the system is\nbooted.\n\nFour debug macro definitions are provided.\n\tDRM_DEBUG(fmt, args...)\n\tDRM_DEBUG_DRIVER(prefix, fmt, args...)\n\tDRM_DEBUG_KMS(prefix, fmt, args...)\n\tDRM_DEBUG_MODE(prefix, fmt, args...)\n\nWhen the boot option of \"drm.debug\u003d4\" is added, it will print the debug info\nusing DRM_DEBUG_KMS macro definition.\nWhen the boot option of \"drm.debug\u003d6\" is added, it will print the debug info\nusing DRM_DEBUG_KMS/DRM_DEBUG_DRIVER.\n\nSometimes we expect to print the value of an array.\nFor example: SDVO command,\nIn such case the following four DRM debug macro definitions are added:\n\tDRM_LOG(fmt, args...)\n\tDRM_LOG_DRIVER(fmt, args...)\n\tDRM_LOG_KMS(fmt, args...)\n\tDRM_LOG_MODE(fmt, args...)\n\nSigned-off-by: Zhao Yakui \u003cyakui.zhao@intel.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "dcae3626d031fe6296b1e96a16f986193a41f840", "tree": "0c9040b8569f47afaaf409d83d720889ba2d25f1", "parents": [ "03d6069912babc07a3da20e715dd6a5dc8f0f867" ], "author": { "name": "Roel Kluin", "email": "roel.kluin@gmail.com", "time": "Wed Jun 10 12:43:48 2009 -0700" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Jun 11 16:10:27 2009 +1000" }, "message": "drm: fix LOCK_TEST_WITH_RETURN macro\n\nWhen this macro isn\u0027t called with \u0027file_priv\u0027 this will result in a build\nfailure.\n\nSigned-off-by: Roel Kluin \u003croel.kluin@gmail.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "8e7d2b2c6ecd3c21a54b877eae3d5be48292e6b5", "tree": "704b3affc3b617be2bcd63ddd9865e9adceeb031", "parents": [ "1406de8e11eb043681297adf86d6892ff8efc27a" ], "author": { "name": "Jesse Barnes", "email": "jbarnes@virtuousgeek.org", "time": "Fri May 08 16:13:25 2009 -0700" }, "committer": { "name": "Eric Anholt", "email": "eric@anholt.net", "time": "Tue May 19 10:07:14 2009 -0700" }, "message": "drm/i915: allocate large pointer arrays with vmalloc\n\nFor awhile now, many of the GEM code paths have allocated page or\nobject arrays with the slab allocator. This is nice and fast, but\nwon\u0027t work well if memory is fragmented, since the slab allocator works\nwith physically contiguous memory (i.e. order \u003e 2 allocations are\nlikely to fail fairly early after booting and doing some work).\n\nThis patch works around the issue by falling back to vmalloc for\n\u003ePAGE_SIZE allocations. This is ugly, but much less work than chaining\na bunch of pages together by hand (suprisingly there\u0027s not a bunch of\ngeneric kernel helpers for this yet afaik). vmalloc space is somewhat\nprecious on 32 bit kernels, but our allocations shouldn\u0027t be big enough\nto cause problems, though they\u0027re routinely more than a page.\n\nNote that this patch doesn\u0027t address the unchecked\nalloc-based-on-ioctl-args in GEM; that needs to be fixed in a separate\npatch.\n\nAlso, I\u0027ve deliberately ignored the DRM\u0027s \"area\" junk. I don\u0027t think\nanyone actually uses it anymore and I\u0027m hoping it gets ripped out soon.\n\n[Updated: removed size arg to new free function. We could unify the\nfree functions as well once the DRM mem tracking is ripped out.]\n\nfd.o bug #20152 (part 1/3)\n\nSigned-off-by: Jesse Barnes \u003cjbarnes@virtuousgeek.org\u003e\nSigned-off-by: Eric Anholt \u003ceric@anholt.net\u003e\n" }, { "commit": "c972d750e4fa3bfee6e7d3635729bf8c9cbb8f0a", "tree": "21636305d6828368d5b4d60a34ab0f1e3dca4a51", "parents": [ "40fc6eab599d0087a75fc77eaaf04d769b667f6d" ], "author": { "name": "Richard Kennedy", "email": "richard@rsk.demon.co.uk", "time": "Wed Mar 18 17:26:44 2009 +0000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Sun Mar 29 18:31:45 2009 +1000" }, "message": "drm: reorder struct drm_ioctl_desc to save space on 64 bit builds\n\nshrinks drm_ioctl_desc from 24 bytes to 16 bytes by reordering members\nto remove padding.\n\nupdates DRM_IOCTL_DEF macro to initialise structure members by name to\nhandle the structure reorder.\n\nThe applied patch reduces data used in drm.ko from 10440 to 9032\n\nSigned-off-by: Richard Kennedy \u003crichard@rsk.demon.co.uk\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "955b12def42e83287c1bdb1411d99451753c1391", "tree": "5dc5025f500b7a0a86c70ea79f2294e94902e3b2", "parents": [ "dd8d7cb49e6e61da96ca44174b063081892c4dc6" ], "author": { "name": "Ben Gamari", "email": "bgamari@gmail.com", "time": "Tue Feb 17 20:08:49 2009 -0500" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Mar 13 14:24:07 2009 +1000" }, "message": "drm: Convert proc files to seq_file and introduce debugfs\n\nThe old mechanism to formatting proc files is extremely ugly. The\nseq_file API was designed specifically for cases like this and greatly\nsimplifies the process.\n\nAlso, most of the files in /proc really don\u0027t belong there. This patch\nintroduces the infrastructure for putting these into debugfs and exposes\nall of the proc files in debugfs as well.\n\nThis contains the i915 hooks rewrite as well, to make bisectability better.\n\nSigned-off-by: Ben Gamari \u003cbgamari@gmail.com\u003e\nSigned-off-by: Eric Anholt \u003ceric@anholt.net\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "8e1004580e0c862cb6bbe2ff8e496f846c54052f", "tree": "f75ca2cebf77d2746613af8e5b61998057df5387", "parents": [ "112b715e8e2f9ef7b96930888bb099ce10b4c3cc" ], "author": { "name": "Kristian Høgsberg", "email": "krh@redhat.com", "time": "Mon Jan 05 16:10:05 2009 -0500" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Mar 13 14:23:58 2009 +1000" }, "message": "drm: Drop unused and broken dri_library_name sysfs attribute.\n\nThe kernel shouldn\u0027t be in the business of telling user space which\ndriver to load. The kernel defers mapping PCI IDs to module names\nto user space and we should do the same for DRI drivers.\n\nAnd in fact, that\u0027s how it does work today. Nothing uses the\ndri_library_name attribute, and the attribute is in fact broken.\nFor intel devices, it falls back to the default behaviour of returning\nthe kernel module name as the DRI driver name, which doesn\u0027t work for\ni965 devices. Nobody has ever hit this problem or filed a bug about this.\n\nSigned-off-by: Kristian Høgsberg \u003ckrh@redhat.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@linux.ie\u003e\n" }, { "commit": "112b715e8e2f9ef7b96930888bb099ce10b4c3cc", "tree": "1058edb8beb6dd60a794d2333e43d37cc7116f06", "parents": [ "41c2e75e60200a860a74b7c84a6375c105e7437f" ], "author": { "name": "Kristian Høgsberg", "email": "krh@bitplanet.net", "time": "Sun Jan 04 16:55:33 2009 -0500" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Mar 13 14:23:58 2009 +1000" }, "message": "drm: claim PCI device when running in modesetting mode.\n\nUnder kernel modesetting, we manage the device at all times, regardless\nof VT switching and X servers, so the only decent thing to do is to\nclaim the PCI device. In that case, we call the suspend/resume hooks\ndirectly from the pci driver hooks instead of the current class device detour.\n\nSigned-off-by: Kristian Høgsberg \u003ckrh@redhat.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@linux.ie\u003e\n" }, { "commit": "41c2e75e60200a860a74b7c84a6375c105e7437f", "tree": "18e97662d6859eead4de816e121d001b34a7352a", "parents": [ "f77d390c9779c496aa5b99ec832996fb76bb1d13" ], "author": { "name": "Benjamin Herrenschmidt", "email": "benh@kernel.crashing.org", "time": "Mon Feb 02 16:55:47 2009 +1100" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Mar 13 14:23:57 2009 +1000" }, "message": "drm: Make drm_local_map use a resource_size_t offset\n\nThis changes drm_local_map to use a resource_size for its \"offset\"\nmember instead of an unsigned long, thus allowing 32-bit machines\nwith a \u003e32-bit physical address space to be able to store there\ntheir register or framebuffer addresses when those are above 4G,\nsuch as when using a PCI video card on a recent AMCC 440 SoC.\n\nThis patch isn\u0027t as \"trivial\" as it sounds: A few functions needed\nto have some unsigned long/int changed to resource_size_t and a few\nprintk\u0027s had to be adjusted.\n\nBut also, because userspace isn\u0027t capable of passing such offsets,\nI had to modify drm_find_matching_map() to ignore the offset passed\nin for maps of type _DRM_FRAMEBUFFER or _DRM_REGISTERS.\n\nIf we ever support multiple _DRM_FRAMEBUFFER or _DRM_REGISTERS maps\nfor a given device, we might have to change that trick, but I don\u0027t\nthink that happens on any current driver.\n\nSigned-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nSigned-off-by: Dave Airlie \u003cairlied@linux.ie\u003e\n" }, { "commit": "f77d390c9779c496aa5b99ec832996fb76bb1d13", "tree": "3e3cccc4ac9416457a944d13b31ef942432717f2", "parents": [ "d883f7f1b75c8dcafa891f7b9e69c5a2f0ff6d66" ], "author": { "name": "Benjamin Herrenschmidt", "email": "benh@kernel.crashing.org", "time": "Mon Feb 02 16:55:46 2009 +1100" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Mar 13 14:23:56 2009 +1000" }, "message": "drm: Split drm_map and drm_local_map\n\nOnce upon a time, the DRM made the distinction between the drm_map\ndata structure exchanged with user space and the drm_local_map used\nin the kernel.\n\nFor some reasons, while the BSD port still has that \"feature\", the\nlinux part abused drm_map for kernel internal usage as the local\nmap only existed as a typedef of the struct drm_map.\n\nThis patch fixes it by declaring struct drm_local_map separately\n(though its content is currently identical to the userspace variant),\nand changing the kernel code to only use that, except when it\u0027s a\nuser\u003c-\u003ekernel interface (ie. ioctl).\n\nThis allows subsequent changes to the in-kernel format\n\nI\u0027ve also replaced the use of drm_local_map_t with struct drm_local_map\nin a couple of places. Mostly by accident but they are the same (the\nformer is a typedef of the later) and I have some remote plans and\nhalf finished patch to completely kill the drm_local_map_t typedef\nso I left those bits in.\n\nSigned-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nAcked-by: Eric Anholt \u003ceric@anholt.net\u003e\nSigned-off-by: Dave Airlie \u003cairlied@linux.ie\u003e\n" }, { "commit": "d883f7f1b75c8dcafa891f7b9e69c5a2f0ff6d66", "tree": "765a9bf6d3322ecfa42819622981baea2480ba41", "parents": [ "041b62374c7fedc11a8a1eeda2868612d3d1436c" ], "author": { "name": "Benjamin Herrenschmidt", "email": "benh@kernel.crashing.org", "time": "Mon Feb 02 16:55:45 2009 +1100" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Mar 13 14:23:56 2009 +1000" }, "message": "drm: Use resource_size_t for drm_get_resource_{start, len}\n\nThe DRM uses its own wrappers to obtain resources from PCI devices,\nwhich currently convert the resource_size_t into an unsigned long.\n\nThis is broken on 32-bit platforms with \u003e32-bit physical address\nspace.\n\nThis fixes them, along with a few occurences of unsigned long used\nto store such a resource in drivers.\n\nSigned-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nSigned-off-by: Dave Airlie \u003cairlied@linux.ie\u003e\n" }, { "commit": "ab00b3e5210954cbaff9207db874a9f03197e3ba", "tree": "68359cf341eb58cefa9d8c91200e6d137ff2007c", "parents": [ "496818f08a78476abdb307e241911536221239fc" ], "author": { "name": "Jesse Barnes", "email": "jbarnes@virtuousgeek.org", "time": "Wed Feb 11 14:01:46 2009 -0800" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Feb 20 12:21:13 2009 +1000" }, "message": "drm/i915: Keep refs on the object over the lifetime of vmas for GTT mmap.\n\nThis fixes potential fault at fault time if the object was unreferenced\nwhile the mapping still existed. Now, while the mmap_offset only lives\nfor the lifetime of the object, the object also stays alive while a vma\nexists that needs it.\n\nSigned-off-by: Jesse Barnes \u003cjbarnes@virtuousgeek.org\u003e\nSigned-off-by: Eric Anholt \u003ceric@anholt.net\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "30b23634084d95781f7611c0713cb551a0c0a152", "tree": "fd4185cb0f0b46ca0069251144619b6e9e2031b0", "parents": [ "152a649b647a8ef47bb74ff9e11850fa6001bedc" ], "author": { "name": "Eric Anholt", "email": "eric@anholt.net", "time": "Tue Jan 27 21:19:41 2009 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Jan 28 07:50:14 2009 -0800" }, "message": "drm: Rip out the racy, unused vblank signal code.\n\nSchedule a vblank signal, kill the process, and we\u0027ll go walking over freed\nmemory. Given that no open-source userland exists using this, nor have I\never heard of a consumer, just let this code die.\n\nSigned-off-by: Eric Anholt \u003ceric@anholt.net\u003e\nRequested-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nAcked-by: Dave Airlie \u003cairlied@linux.ie\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "fede5c91c4a8a7701d205b2b84b9835ddc7d6f02", "tree": "a80b546790a405b5a0b60e843c06b5697c0b2226", "parents": [ "f51c5b6e623cc737d47f513cbb893cec914f0bd2" ], "author": { "name": "Eric Anholt", "email": "eric@anholt.net", "time": "Fri Dec 19 17:23:38 2008 -0800" }, "committer": { "name": "Dave Airlie", "email": "airlied@linux.ie", "time": "Mon Dec 29 17:47:27 2008 +1000" }, "message": "drm: Add a debug node for vblank state.\n\nSigned-off-by: Eric Anholt \u003ceric@anholt.net\u003e\nSigned-off-by: Dave Airlie \u003cairlied@linux.ie\u003e\n" }, { "commit": "f453ba0460742ad027ae0c4c7d61e62817b3e7ef", "tree": "29e6ecacd6e8971aa62e1825d77f2c1876ac3eb2", "parents": [ "de151cf67ce52ed2d88083daa5e60c7858947329" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Nov 07 14:05:41 2008 -0800" }, "committer": { "name": "Dave Airlie", "email": "airlied@linux.ie", "time": "Mon Dec 29 17:47:23 2008 +1000" }, "message": "DRM: add mode setting support\n\nAdd mode setting support to the DRM layer.\n\nThis is a fairly big chunk of work that allows DRM drivers to provide\nfull output control and configuration capabilities to userspace. It was\nmotivated by several factors:\n - the fb layer\u0027s APIs aren\u0027t suited for anything but simple\n configurations\n - coordination between the fb layer, DRM layer, and various userspace\n drivers is poor to non-existent (radeonfb excepted)\n - user level mode setting drivers makes displaying panic \u0026 oops\n messages more difficult\n - suspend/resume of graphics state is possible in many more\n configurations with kernel level support\n\nThis commit just adds the core DRM part of the mode setting APIs.\nDriver specific commits using these new structure and APIs will follow.\n\nCo-authors: Jesse Barnes \u003cjbarnes@virtuousgeek.org\u003e, Jakob Bornecrantz \u003cjakob@tungstengraphics.com\u003e\nContributors: Alan Hourihane \u003calanh@tungstengraphics.com\u003e, Maarten Maathuis \u003cmadman2003@gmail.com\u003e\n\nSigned-off-by: Jesse Barnes \u003cjbarnes@virtuousgeek.org\u003e\nSigned-off-by: Eric Anholt \u003ceric@anholt.net\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "a2c0a97b784f837300f7b0869c82ab712c600952", "tree": "aca1cdf3d32e1cfa7387350483f6a70c74a24ffd", "parents": [ "a9587470f753d670d910293ecbf1c7b66c99de50" ], "author": { "name": "Jesse Barnes", "email": "jbarnes@virtuousgeek.org", "time": "Wed Nov 05 10:31:53 2008 -0800" }, "committer": { "name": "Dave Airlie", "email": "airlied@linux.ie", "time": "Mon Dec 29 17:47:22 2008 +1000" }, "message": "drm: GEM mmap support\n\nAdd core support for mapping of GEM objects. Drivers should provide a\nvm_operations_struct if they want to support page faulting of objects.\nThe code for handling GEM object offsets was taken from TTM, which was\nwritten by Thomas Hellström.\n\nSigned-off-by: Jesse Barnes \u003cjbarnes@virtuousgeek.org\u003e\nSigned-off-by: Eric Anholt \u003ceric@anholt.net\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "1147c9cdd0f60f09a98702a9f865176af18a989f", "tree": "4f3c33102566475cd145cf0235c1738d07b8b715", "parents": [ "7c1c2871a6a3a114853ec6836e9035ac1c0c7f7a" ], "author": { "name": "Vegard Nossum", "email": "vegard.nossum@gmail.com", "time": "Tue Dec 02 13:38:47 2008 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@linux.ie", "time": "Mon Dec 29 17:47:22 2008 +1000" }, "message": "drm: fix leak of uninitialized data to userspace\n\n...so drm_getunique() is trying to copy some uninitialized data to\nuserspace. The ECX register contains the number of words that are\nleft to copy -- so there are 5 * 4 \u003d 20 bytes left. The offset of the\nfirst uninitialized byte (counting from the start of the string) is\nalso 20 (i.e. 0xf65d2294\u0026((1 \u003c\u003c 5)-1) \u003d\u003d 20). So somebody tried to\ncopy 40 bytes when the string was only 19 long.\n\nIn drm_set_busid() we have this code:\n\n dev-\u003eunique_len \u003d 40;\n dev-\u003eunique \u003d drm_alloc(dev-\u003eunique_len + 1, DRM_MEM_DRIVER);\n ...\n len \u003d snprintf(dev-\u003eunique, dev-\u003eunique_len, pci:%04x:%02x:%02x.%d\",\n\n...so it seems that dev-\u003eunique is never updated to reflect the\nactual length of the string. The remaining bytes (20 in this case)\nare random uninitialized bytes that are copied into userspace.\n\nThis patch fixes the problem by setting dev-\u003eunique_len after the\nsnprintf().\n\nairlied- I\u0027ve had to fix this up to store the alloced size so\nwe have it for drm_free later.\n\nReported-by: Sitsofe Wheeler \u003csitsofe@yahoo.com\u003e\nSigned-off-by: Vegard Nossum \u003cvegardno@thuin.ifi.uio.no\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "7c1c2871a6a3a114853ec6836e9035ac1c0c7f7a", "tree": "1b5debcc86ff20bd5e11b42ea5c52da42214e376", "parents": [ "e7f7ab45ebcb54fd5f814ea15ea079e079662f67" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Nov 28 14:22:24 2008 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@linux.ie", "time": "Mon Dec 29 17:47:22 2008 +1000" }, "message": "drm: move to kref per-master structures.\n\nThis is step one towards having multiple masters sharing a drm\ndevice in order to get fast-user-switching to work.\n\nIt splits out the information associated with the drm master\ninto a separate kref counted structure, and allocates this when\na master opens the device node. It also allows the current master\nto abdicate (say while VT switched), and a new master to take over\nthe hardware.\n\nIt moves the Intel and radeon drivers to using the sarea from\nwithin the new master structures.\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "e7f7ab45ebcb54fd5f814ea15ea079e079662f67", "tree": "53e8206501c0b7a00831ea084b52664f7899d1c0", "parents": [ "4a6908a3a050aacc9c3a2f36b276b46c0629ad91" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Nov 28 13:43:47 2008 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@linux.ie", "time": "Mon Dec 29 17:47:21 2008 +1000" }, "message": "drm: cleanup exit path for module unload\n\nThe current sub-module unload exit path is a mess, it tries\nto abuse the idr. Just keep a list of devices per driver struct\nand free them in-order on rmmod.\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "52440211dcdc52c0b757f8b34d122e11b12cdd50", "tree": "5ad9c526134d046043b04ee56fb34d428a03c890", "parents": [ "6133047aa64d2fd5b3b79dff74f696ded45615b2" ], "author": { "name": "Keith Packard", "email": "keithp@keithp.com", "time": "Tue Nov 18 09:30:25 2008 -0800" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Tue Nov 25 09:49:03 2008 +1000" }, "message": "drm: move drm vblank initialization/cleanup to driver load/unload\n\ndrm vblank initialization keeps track of the changes in driver-supplied\nframe counts across vt switch and mode setting, but only if you let it by\nnot tearing down the drm vblank structure.\n\nSigned-off-by: Keith Packard \u003ckeithp@keithp.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "5d8e6bb7a20b6206e1fe44565efc383a941b81fa", "tree": "088f954cf9d902455b1db968954fb3aed6edb246", "parents": [ "bd95e0a4a6bb9485fe35dda62719663f6ceabae1" ], "author": { "name": "Eric Anholt", "email": "eric@anholt.net", "time": "Tue Nov 04 18:36:29 2008 -0800" }, "committer": { "name": "Dave Airlie", "email": "airlied@linux.ie", "time": "Tue Nov 11 17:44:26 2008 +1000" }, "message": "drm: Remove infrastructure for supporting i915\u0027s vblank swapping.\n\nIt\u0027s not used in any other drivers, and doesn\u0027t look like it will be from\ndrm.git master.\n\nSigned-off-by: Eric Anholt \u003ceric@anholt.net\u003e\nSigned-off-by: Dave Airlie \u003cairlied@linux.ie\u003e\n" }, { "commit": "ba1eb1d825fdef40f69871caf8e5842d00efbbc5", "tree": "73b45769022788532ee169c270e2792a229b1a40", "parents": [ "50aa253d820ad4577e2231202f2c8fd89f9dc4e6" ], "author": { "name": "Keith Packard", "email": "keithp@keithp.com", "time": "Tue Oct 14 19:55:10 2008 -0700" }, "committer": { "name": "Dave Airlie", "email": "airlied@linux.ie", "time": "Sat Oct 18 07:10:53 2008 +1000" }, "message": "i915: Map status page cached for chips with GTT-based HWS location.\n\nThis should improve performance by avoiding uncached reads by the CPU (the\npoint of having a status page), and may improve stability. This patch only\naffects G33, GM45 and G45 chips as those are the only ones using GTT-based\nHWS mappings.\n\nSigned-off-by: Keith Packard \u003ckeithp@keithp.com\u003e\nSigned-off-by: Eric Anholt \u003ceric@anholt.net\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "9bfbd5cb72c9edb8504a4a7a0aa89cdb2fcb4845", "tree": "b59f08475d3366e60e3c93538ad398905c4a74ec", "parents": [ "e0f0754ff6128570dcf38032f5bfb1f195e5bbee" ], "author": { "name": "Jesse Barnes", "email": "jbarnes@virtuousgeek.org", "time": "Mon Sep 15 15:00:33 2008 -0700" }, "committer": { "name": "Dave Airlie", "email": "airlied@linux.ie", "time": "Sat Oct 18 07:10:53 2008 +1000" }, "message": "drm: kill drm_device-\u003eirq\n\nLike the last patch but adds a macro to get at the irq value instead of\ndereferencing pdev directly. Should make things easier for the BSD guys and\nif we ever support non-PCI devices.\n\nSigned-off-by: Jesse Barnes \u003cjbarnes@virtuousgeek.org\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "dbb19d302baa8ba518599701914f600935fc53fa", "tree": "d3391ed8f7c39f8201d3ea1432df6f634770a2e5", "parents": [ "c99b058f132388a666544d293392d52d1def6b12" ], "author": { "name": "Kristian Høgsberg", "email": "krh@redhat.com", "time": "Wed Aug 20 11:04:27 2008 -0400" }, "committer": { "name": "Dave Airlie", "email": "airlied@linux.ie", "time": "Sat Oct 18 07:10:52 2008 +1000" }, "message": "i915 gem: install and uninstall irq handler in entervt and leavevt ioctls.\n\nSigned-off-by: Kristian Høgsberg \u003ckrh@redhat.com\u003e\nSigned-off-by: Eric Anholt \u003ceric@anholt.net\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "673a394b1e3b69be886ff24abfd6df97c52e8d08", "tree": "61ca8299333ab50ffc46cf328b20eb25133392ff", "parents": [ "d1d8c925b71dd6753bf438f9e14a9e5c5183bcc6" ], "author": { "name": "Eric Anholt", "email": "eric@anholt.net", "time": "Wed Jul 30 12:06:12 2008 -0700" }, "committer": { "name": "Dave Airlie", "email": "airlied@linux.ie", "time": "Sat Oct 18 07:10:12 2008 +1000" }, "message": "drm: Add GEM (\"graphics execution manager\") to i915 driver.\n\nGEM allows the creation of persistent buffer objects accessible by the\ngraphics device through new ioctls for managing execution of commands on the\ndevice. The userland API is almost entirely driver-specific to ensure that\nany driver building on this model can easily map the interface to individual\ndriver requirements.\n\nGEM is used by the 2d driver for managing its internal state allocations and\nwill be used for pixmap storage to reduce memory consumption and enable\nzero-copy GLX_EXT_texture_from_pixmap, and in the 3d driver is used to enable\nGL_EXT_framebuffer_object and GL_ARB_pixel_buffer_object.\n\nSigned-off-by: Eric Anholt \u003ceric@anholt.net\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "0a3e67a4caac273a3bfc4ced3da364830b1ab241", "tree": "02a2e5e76d9dffcb556d09b0eee4d34ebe5d81cb", "parents": [ "2df68b439fcb97a4c55f81516206ef4ee325e28d" ], "author": { "name": "Jesse Barnes", "email": "jbarnes@virtuousgeek.org", "time": "Tue Sep 30 12:14:26 2008 -0700" }, "committer": { "name": "Dave Airlie", "email": "airlied@linux.ie", "time": "Sat Oct 18 07:10:11 2008 +1000" }, "message": "drm: Rework vblank-wait handling to allow interrupt reduction.\n\nPreviously, drivers supporting vblank interrupt waits would run the interrupt\nall the time, or all the time that any 3d client was running, preventing the\nCPU from sleeping for long when the system was otherwise idle. Now, interrupts\nare disabled any time that no client is waiting on a vblank event. The new\nmethod uses vblank counters on the chipsets when the interrupts are turned\noff, rather than counting interrupts, so that we can continue to present\naccurate vblank numbers.\n\nCo-author: Michel Dänzer \u003cmichel@tungstengraphics.com\u003e\nSigned-off-by: Jesse Barnes \u003cjbarnes@virtuousgeek.org\u003e\nSigned-off-by: Eric Anholt \u003ceric@anholt.net\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "242e3df80b8d25ed681c278512df0993725f25dd", "tree": "28dcd053d5c5203d8a952f0c1b5f354c3c7d06f7", "parents": [ "50515af207d410c9f228380e529c56f43c3de0bd" ], "author": { "name": "Dave Airlie", "email": "airlied@linux.ie", "time": "Tue Jul 15 15:48:05 2008 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@linux.ie", "time": "Tue Jul 15 15:48:05 2008 +1000" }, "message": "drm/radeon: fixup issue with radeon and PAT support.\n\nWith new userspace libpciaccess we can get a conflicting mapping\non the PCIE GART table in the video RAM. Always try and map it _wc.\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "c0e09200dc0813972442e550a5905a132768e56c", "tree": "d38e635a30ff8b0a2b98b9d7f97cab1501f8209e", "parents": [ "bce7f793daec3e65ec5c5705d2457b81fe7b5725" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu May 29 10:09:59 2008 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Jul 14 10:45:01 2008 +1000" }, "message": "drm: reorganise drm tree to be more future proof.\n\nWith the coming of kernel based modesetting and the memory manager stuff,\nthe everything in one directory approach was getting very ugly and\nstarting to be unmanageable.\n\nThis restructures the drm along the lines of other kernel components.\n\nIt creates a drivers/gpu/drm directory and moves the hw drivers into\nsubdirectores. It moves the includes into an include/drm, and\nsets up the unifdef for the userspace headers we should be exporting.\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "af6061af0d9f84a4665f88186dc1ff9e4fb78330", "tree": "90281b9188338cc702f92329ed3725313d248eea", "parents": [ "c0a18111e571138747a98af18b3a2124df56a0d1" ], "author": { "name": "Dave Airlie", "email": "airlied@linux.ie", "time": "Wed May 07 12:15:39 2008 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@linux.ie", "time": "Wed May 07 12:15:39 2008 +1000" }, "message": "Revert \"drm/vbl rework: rework how the drm deals with vblank.\"\n\nThis reverts commit ac741ab71bb39e6977694ac0cc26678d8673cda4.\n\nOkay this looks like wasn\u0027t as fully baked as I\u0027d led myself to believe.\n\nRevert for now for further baking.\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "bf9d89295233ae2ba7b312c78ee5657307b09f4c", "tree": "3da2fbcd93937bfa1907a893dc0d418c78f3d37b", "parents": [ "8e24eea728068bbeb6a3c500b848f883a20bf225" ], "author": { "name": "Harvey Harrison", "email": "harvey.harrison@gmail.com", "time": "Wed Apr 30 00:55:10 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 30 08:29:54 2008 -0700" }, "message": "drivers/char: replace remaining __FUNCTION__ occurrences\n\n__FUNCTION__ is gcc-specific, use __func__\n\nSigned-off-by: Harvey Harrison \u003charvey.harrison@gmail.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "ac741ab71bb39e6977694ac0cc26678d8673cda4", "tree": "f82e08399a0da5accba930444744b269479185dd", "parents": [ "2c14f28be2a3f2a2e9861b156d64fbe2bc7000c3" ], "author": { "name": "Jesse Barnes", "email": "jbarnes@virtuousgeek.org", "time": "Tue Apr 22 16:03:07 2008 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@linux.ie", "time": "Sat Apr 26 18:01:07 2008 +1000" }, "message": "drm/vbl rework: rework how the drm deals with vblank.\n\nOther Authors: Michel Dänzer \u003cmichel@tungstengraphics.com\u003e\nmga: Ian Romanick \u003cidr@us.ibm.com\u003e\nvia: Thomas Hellstrom \u003cthomas-at-tungstengraphics-dot-com\u003e\n\nThis re-works the DRM internals to provide a better interface for drivers\nto expose vblank on multiple crtcs.\n\nIt also includes work done by Michel on making i915 triple buffering and pageflipping work properly.\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "2c14f28be2a3f2a2e9861b156d64fbe2bc7000c3", "tree": "dbf55c07d1b245a45330e0d879833df29954fe12", "parents": [ "7b832b56bd971348329c3f4c753ca0abfdf3a3d1" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Apr 21 16:47:32 2008 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@linux.ie", "time": "Sat Apr 26 17:55:07 2008 +1000" }, "message": "drm: reorganise minor number handling using backported modesetting code.\n\nrips out the head crap and replaces it with an idr and drm_minor structure\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" } ], "next": "56d110e852b0b1c85ad6c9bfe1cb4473ceb16402" }