)]}' { "log": [ { "commit": "ff82f052d2a187dd0fa0e431ba70eb457c71a40e", "tree": "65bf2788445c3087cdb6c05fe2d40e81629ffd3b", "parents": [ "5ffdb658f605cbc420944e7c7eeec9fbb8a73772" ], "author": { "name": "Jerome Glisse", "email": "jglisse@redhat.com", "time": "Fri Jan 22 15:19:00 2010 +0100" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Feb 01 11:33:11 2010 +1000" }, "message": "drm/radeon/kms: Bailout of blit if error happen \u0026 protect with mutex V3\n\nIf an error happen in r600_blit_prepare_copy report it rather\nthan WARNING and keeping execution. For instance if ib allocation\nfailed we did just warn about but then latter tried to access\nNULL ib ptr causing oops. This patch also protect r600_copy_blit\nwith a mutex as otherwise one process might overwrite blit temporary\ndata with new one possibly leading to GPU lockup.\n\nShould partialy or totaly fix:\nhttps://bugzilla.redhat.com/show_bug.cgi?id\u003d553279\n\nV2 failing blit initialization is not fatal, fallback to memcpy when\nthis happen\nV3 init blit before startup as we pin in startup, remove duplicate\ncode (this one was actualy tested unlike V2)\n\nSigned-off-by: Jerome Glisse \u003cjglisse@redhat.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "5ffdb658f605cbc420944e7c7eeec9fbb8a73772", "tree": "9f091b311f1c434f506d4c34ce373346a77d895b", "parents": [ "c188660f6dbb0df9febe1b841a16c66c28353c15" ], "author": { "name": "Jakob Bornecrantz", "email": "jakob@vmware.com", "time": "Sat Jan 30 03:38:08 2010 +0000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Feb 01 11:29:59 2010 +1000" }, "message": "drm/vmwgfx: Don\u0027t send bad flags to the host\n\nSigned-off-by: Jakob Bornecrantz \u003cjakob@vmware.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "c188660f6dbb0df9febe1b841a16c66c28353c15", "tree": "268604303fd2f2336a686c093fc0165f6cf4fc1d", "parents": [ "8e19a951774a16cf2626292ae06fd2b62630e67e" ], "author": { "name": "Peter Hanzel", "email": "hanzelpeter@gmail.com", "time": "Sat Jan 30 03:38:07 2010 +0000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Feb 01 11:29:31 2010 +1000" }, "message": "drm/vmwgfx: Request SVGA version 2 and bail if not found\n\nThis fixes the driver not loading on older versions of VMware.\n\nSigned-off-by: Peter Hanzel \u003chanzelpeter@gmail.com\u003e\nSigned-off-by: Jakob Bornecrantz \u003cjakob@vmware.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "8e19a951774a16cf2626292ae06fd2b62630e67e", "tree": "b1abc1cf6c493200d45663f71b1ff59ab3cd9477", "parents": [ "110b20c3ddcfa98cc932aef3af2d59b4e0841f08" ], "author": { "name": "Jakob Bornecrantz", "email": "jakob@vmware.com", "time": "Sat Jan 30 03:38:06 2010 +0000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Feb 01 11:29:03 2010 +1000" }, "message": "drm/vmwgfx: Correctly detect 3D\n\nSigned-off-by: Jakob Bornecrantz \u003cjakob@vmware.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "110b20c3ddcfa98cc932aef3af2d59b4e0841f08", "tree": "26e4a4a91b823a2caed67c12b2e5e812a068abe0", "parents": [ "fa5829b36539067f3c675f5d437531dedcfc4ad8" ], "author": { "name": "Austin Yuan", "email": "shengquan.yuan@gmail.com", "time": "Thu Jan 21 13:45:40 2010 +0800" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Feb 01 11:27:25 2010 +1000" }, "message": "drm/ttm: remove unnecessary save_flags and ttm_flag_masked in ttm_bo_util.c\n\nSigned-off-by: Austin Yuan \u003cshengquan.yuan@gmail.com\u003e\nAcked-by: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "dd5fde6041d7ea30fcfedcc159a4cec8b01f73b2", "tree": "d0a0331461928ef4d755775b1cedd86d5bcc03c6", "parents": [ "4b866288be6ffaefaad9cec212cb09e3258a68ee" ], "author": { "name": "Richard Kennedy", "email": "richard@rsk.demon.co.uk", "time": "Tue Jan 26 17:10:48 2010 +0000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Feb 01 11:24:50 2010 +1000" }, "message": "drm/ttm: remove padding from ttm_ref_object on 64bit builds\n\nRe-order structure ttm_ref_object to remove 8 bytes of alignment padding\non 64 bit builds, so shrinking its size from 72 to 64 bytes allowing it\nto fit into a smaller slab.\n\nSigned-off-by: Richard Kennedy \u003crichard@rsk.demon.co.uk\u003e\nAcked-by: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "4b866288be6ffaefaad9cec212cb09e3258a68ee", "tree": "81317f5c119f6429324118bb5db9a658d429e224", "parents": [ "2dea2e29b9fad48c759aa406b5ea426bff4339af" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Feb 01 11:22:10 2010 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Feb 01 11:22:10 2010 +1000" }, "message": "drm/radeon/kms: release agp on error.\n\nif we get an error, release the AGP if we\u0027ve acquired it already.\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "2dea2e29b9fad48c759aa406b5ea426bff4339af", "tree": "df7ef0a169d7bb96b326595affef6836e9f588d7", "parents": [ "cdb6e375c3c84558ed2c13c5461b22e8f7b7980e" ], "author": { "name": "John Kacur", "email": "jkacur@redhat.com", "time": "Sun Jan 31 20:38:03 2010 +0100" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Feb 01 11:19:31 2010 +1000" }, "message": "drm/kms/radeon/agp: Move the check of the aper_size after drm_acp_acquire and drm_agp_info\n\nFirst call drm_agp_acquire to check if agp has been acquired.\nSecond call drm_agp_info to fill in the info data struct, including aper_size.\nFinally do the check to see if the aper_size makes sense.\n\nSigned-off-by: John Kacur \u003cjkacur@redhat.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "cdb6e375c3c84558ed2c13c5461b22e8f7b7980e", "tree": "28474a8fd0bff4ea869aef387ddf2c680df54aec", "parents": [ "db78e27de7e29a6db6be7caf607cf803d84094aa" ], "author": { "name": "John Kacur", "email": "jkacur@redhat.com", "time": "Sun Jan 31 20:38:02 2010 +0100" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Feb 01 11:19:08 2010 +1000" }, "message": "drm/kms/radeon/agp: Fix warning, format ‘%d’ expects type ‘int’, but argument 4 has type ‘size_t’\n\n- Fix warning by using %zu instead of %d for size_t\n- Fix spelling mistake, \"to\" should be \"too\".\n\nSigned-off-by: John Kacur \u003cjkacur@redhat.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "db78e27de7e29a6db6be7caf607cf803d84094aa", "tree": "c387de94b95e444205361bc31ce59acd29b10aee", "parents": [ "f28cf33945cc112f8ee835512b7440905dc29ad2" ], "author": { "name": "Francisco Jerez", "email": "currojerez@riseup.net", "time": "Tue Jan 12 18:49:43 2010 +0100" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Feb 01 11:15:37 2010 +1000" }, "message": "drm/ttm: Avoid conflicting reserve_memtype during ttm_tt_set_page_caching.\n\nFixes errors like:\n\u003e reserve_ram_pages_type failed 0x15b7a000-0x15b7b000, track 0x8, req 0x10\nwhen a BO is moved between WC and UC areas.\n\nReported-by: Xavier Chantry \u003cshiningxc@gmail.com\u003e\nSigned-off-by: Francisco Jerez \u003ccurrojerez@riseup.net\u003e\nAcked-by: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "f28cf33945cc112f8ee835512b7440905dc29ad2", "tree": "dc824913b484b5f726f23692ff0e7e100f7a3e7a", "parents": [ "43c33ed87d0f1b900a6a3014db556ecc7f4a989b" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Jan 28 17:15:25 2010 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Feb 01 10:13:15 2010 +1000" }, "message": "drm/kms/radeon: pick digitial encoders smarter. (v3)\n\nbooting a Lenovo W500 with LVDS + DP outputs showed up a TODO we had\non our list, to pick a correct digital encoder block. The LVTMA\nencoder requires the second digital encoder, all others can use any\nencoder at all.\n\nThis fixes the digital encoder selection logic to enable LVDS/DP combos\nto work okay.\n\nV2: fix silly addition of connector dig_block and cleanup the other\nplaces in the code that pick the encoder.\n\nV3: rename to dig_encoder and clean up further - also fix\nthe picking algorithm.\n\ntested on Lenovo W500 + desktop 3650 cards.\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "43c33ed87d0f1b900a6a3014db556ecc7f4a989b", "tree": "b8280b52b5965111e6673ab4bc699cefa3e588d3", "parents": [ "97b94ccb9aa1b82ed7a9a045d0ae5b32c99b84a0" ], "author": { "name": "Dave Airlie", "email": "airlied@linux.ie", "time": "Fri Jan 29 15:55:30 2010 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Feb 01 10:12:26 2010 +1000" }, "message": "drm/radeon/kms: use active device to pick connector for encoder\n\nOn the W500 we have UNIPHY routed to both DVI and DP, this seems\nto always pick the DVI connector which means link training fails.\n\nSwitch to using active device to pick the connector, this seems\nlike it should be safe from a code review, and it fixes things\na bit more here.\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "97b94ccb9aa1b82ed7a9a045d0ae5b32c99b84a0", "tree": "0b2a40ecd440ec9ed36b7ca282fab514cb2358cb", "parents": [ "7087e16286913b41ba9a5186360645b57b8508dd" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Jan 29 15:31:47 2010 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Feb 01 10:12:17 2010 +1000" }, "message": "drm/radeon/kms: fix incorrect logic in DP vs eDP connector checking.\n\nThis makes displayport work again here.\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "7087e16286913b41ba9a5186360645b57b8508dd", "tree": "5a199665f9c04a78c0a3cbb5d1b98785d3edfaf7", "parents": [ "f2ab3a13d2cbe19426c27c35a014c98212e914a5" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Jan 25 16:13:55 2010 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Jan 25 16:13:55 2010 +1000" }, "message": "drm/radeon/kms: preface warning printk with driver name\n\nThis just adds a little more info to the warning for old -ati/mesa\nuserspaces.\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "f2ab3a13d2cbe19426c27c35a014c98212e914a5", "tree": "976a541f1394200ac57ad25c939723dff66c801c", "parents": [ "5fd4df4d475a7fee96fff54f6341192f547984e0" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Jan 25 16:13:12 2010 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Jan 25 16:13:12 2010 +1000" }, "message": "drm/radeon/kms: drop unnecessary printks.\n\nThese printks aren\u0027t required anymore.\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "5fd4df4d475a7fee96fff54f6341192f547984e0", "tree": "d6a5c1f2c9cf8cca953aabed543f86c732e2adee", "parents": [ "d796d8446fe0ff4442c1a1d0b10c861f17ac8168" ], "author": { "name": "Zhenyu Wang", "email": "zhenyuw@linux.intel.com", "time": "Mon Jan 18 16:47:04 2010 +0800" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Jan 25 16:04:47 2010 +1000" }, "message": "drm: fix regression in fb blank handling\n\ncommit 731b5a15a3b1474a41c2ca29b4c32b0f21bc852e\nAuthor: James Simmons \u003cjsimmons@infradead.org\u003e\nDate: Thu Oct 29 20:39:07 2009 +0000\n\n drm/kms: properly handle fbdev blanking\n\nuses DRM_MODE_DPMS_ON for FB_BLANK_NORMAL, but DRM_MODE_DPMS_ON\nis actually for turning output on instead of blank.\n\nThis makes fb blank broken on my T61, it put LVDS on but leave\npipe disabled which made screen totally white or caused some\n\u0027burning\u0027 effect.\n\n[airlied: James objects to this but at this point in 2.6.33,\nI can\u0027t see a patch that will fix this properly like he wants coming\nin time and otherwise this is a regression - proper fix for 2.6.34\nhopefully.]\n\nCc: James Simmons \u003cjsimmons@infradead.org\u003e\nSigned-off-by: Zhenyu Wang \u003czhenyuw@linux.intel.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "d796d8446fe0ff4442c1a1d0b10c861f17ac8168", "tree": "c3457449a8c9bf019fb81becd042a3f64ee0b972", "parents": [ "8ba5152a3acd5914cade42a1c8c9dc58ad8d1a89" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Jan 25 13:08:08 2010 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Jan 25 16:04:42 2010 +1000" }, "message": "drm/radeon/kms: make hibernate work on IGPs\n\nThis is the least invasive fix without migrating the radeon driver\nto pm_ops from what I can see. We just always migrate VRAM objects\non IGPs for now and we can fix it up later to migrate depending\non STR vs STD.\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "8ba5152a3acd5914cade42a1c8c9dc58ad8d1a89", "tree": "7e5433d4d03d753a152a9e1eec5c6e4c6958709a", "parents": [ "0eaddb28d3460f13102cd82cf58af875bfb1bd53" ], "author": { "name": "Thomas Hellstrom", "email": "thellstrom@vmware.com", "time": "Sat Jan 16 16:05:05 2010 +0100" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Jan 25 16:04:39 2010 +1000" }, "message": "drm/vmwgfx: Optimize memory footprint for DMA buffers.\n\nUse VRAM whenever there is free space for DMA buffers,\nbut use system GMR memory if using VRAM would cause an eviction.\n\nThis significantly reduces the guest system memory usage for\nVMs with a large amount of VRAM allocated.\n\nSigned-off-by: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "0eaddb28d3460f13102cd82cf58af875bfb1bd53", "tree": "5fe3e9f8c1e7d6c5c1fd2c2bf445a3d7cc9d33e2", "parents": [ "9299795c6e1e11b2d1e6f53a03902dc80cfb3320" ], "author": { "name": "Thomas Hellstrom", "email": "thellstrom@vmware.com", "time": "Sat Jan 16 16:05:04 2010 +0100" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Jan 25 16:04:30 2010 +1000" }, "message": "drm/ttm: Allow system memory as a busy placement.\n\nThis is needed to fix a vmwgfx memory usage bug.\n\nSigned-off-by: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "9299795c6e1e11b2d1e6f53a03902dc80cfb3320", "tree": "abfda0d16d1c8239e2be8294a8e5c02190d9384a", "parents": [ "8d586fe65a33b1a3a2a2539119248ce12f4bab50", "38678d3557420a1c40f7ad5a04a46a7de7a305b9" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Jan 25 16:04:21 2010 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Jan 25 16:04:21 2010 +1000" }, "message": "Merge remote branch \u0027korg/drm-radeon-next\u0027 into drm-linus\n\n* korg/drm-radeon-next:\n drm/radeon/kms: fix legacy get_engine/memory clock\n drm/radeon/kms/atom: atom parser fixes\n drm/radeon/kms: clean up atombios pll code\n drm/radeon/kms: clean up pll struct\n drm/radeon/kms/atom: fix crtc lock ordering\n drm/radeon: r6xx/r7xx possible security issue, system ram access\n drm/radeon/kms: r600/r700 don\u0027t test ib if ib initialization fails\n drm/radeon/kms: Forbid creation of framebuffer with no valid GEM object\n drm/radeon/kms: r600 handle irq vector ring overflow\n drm/radeon/kms: r600/r700 don\u0027t process IRQ if not initialized\n drm/radeon/kms: r600/r700 disable irq at suspend\n drm/radeon/kms/r4xx: cleanup atom path\n drm/radeon/kms: fix atombios_crtc_set_base\n drm/radeon/kms/atom: upstream parser updates\n drm/radeon/kms/atom: fix some parser bugs\n drm/radeon/kms: fix hardcoded mmio size in register functions\n drm/radeon/kms/r100: fix bug in CS parser\n drm/radeon/kms/r200: fix bug in CS parser\n drm/radeon/kms/r200: fix bug in CS parser\n" }, { "commit": "8d586fe65a33b1a3a2a2539119248ce12f4bab50", "tree": "05b20c54ec304069002358c7e7682237099ff5d2", "parents": [ "1a961ce09fe39df9a1b796df98794fd32c76c413", "162265367a96d381f07066581d65e52627b08618" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Jan 25 16:04:11 2010 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Jan 25 16:04:11 2010 +1000" }, "message": "Merge remote branch \u0027nouveau/for-airlied\u0027 of ../drm-nouveau-next into drm-linus\n\n* \u0027nouveau/for-airlied\u0027 of ../drm-nouveau-next:\n drm/nv50: prevent switching off SOR when in use for DVI-over-DP\n drm/nv50: fail auxch transaction if reply count not what we expect\n drm/nouveau: fix failure path if userspace specifies no valid memtypes\n drm/nouveau: report LVDS as disconnected if lid closed\n drm/nv50: prevent accidently turning off encoders we\u0027re actually using\n drm/nv50: fix alignment of per-channel fifo cache\n drm/nouveau: Evict buffers in VRAM before freeing sgdma\n drm/nouveau: Acknowledge DMA_VTX_PROTECTION PGRAPH interrupts\n drm/nouveau: fix thinko in nv04_instmem.c\n drm/nouveau: fix a race condition in nouveau_dma_wait()\n" }, { "commit": "1a961ce09fe39df9a1b796df98794fd32c76c413", "tree": "06ab04dd5f4fb780f23a8468cf2b65d152c3db1f", "parents": [ "8471a26b9c36c965d278020cc0699e2e95d120e5" ], "author": { "name": "Luca Barbieri", "email": "luca@luca-barbieri.com", "time": "Wed Jan 20 20:01:30 2010 +0100" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Jan 25 11:43:54 2010 +1000" }, "message": "drm/ttm: Fix race condition in ttm_bo_delayed_delete (v3, final)\n\nResending this with Thomas Hellstrom\u0027s signoff for merging into 2.6.33\n\nttm_bo_delayed_delete has a race condition, because after we do:\nkref_put(\u0026nentry-\u003elist_kref, ttm_bo_release_list);\n\nwe are not holding the list lock and not holding any reference to\nobjects, and thus every bo in the list can be removed and freed at\nthis point.\n\nHowever, we then use the next pointer we stored, which is not guaranteed\nto be valid.\n\nThis was apparently the cause of some Nouveau oopses I experienced.\n\nThis patch rewrites the function so that it keeps the reference to nentry\nuntil nentry itself is freed and we already got a reference to nentry-\u003enext.\n\nv2 updated by me according to Thomas Hellstrom\u0027s feedback.\nv3 proposed by Thomas Hellstrom. Commit comment updated by me.\n\nBoth updates fixed minor efficiency/style issues only and all three versions\nshould be correct.\n\nSigned-off-by: Luca Barbieri \u003cluca@luca-barbieri.com\u003e\nSigned-off-by: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "162265367a96d381f07066581d65e52627b08618", "tree": "d39acdcba8c392355f95a9a9e39a0ba84ae1f3a8", "parents": [ "0107bae01add219e1a146f2b412d64a7f1946028" ], "author": { "name": "Ben Skeggs", "email": "bskeggs@redhat.com", "time": "Fri Jan 22 10:57:01 2010 +1000" }, "committer": { "name": "Ben Skeggs", "email": "bskeggs@redhat.com", "time": "Mon Jan 25 10:35:33 2010 +1000" }, "message": "drm/nv50: prevent switching off SOR when in use for DVI-over-DP\n\nAnother hack because of us exposing each encoder block\u0027s function as\nan encoder rather than exposing a single encoder that deals with them\nall.\n\nA proper fix will come, it\u0027s just rather invasive so this hack will\ndo until then.\n\nSigned-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n" }, { "commit": "0107bae01add219e1a146f2b412d64a7f1946028", "tree": "7432b1d6c37f53ba03a65fe2fe2a102336bd6c67", "parents": [ "0208843dd59330c30e0773c77d5362260aa340e0" ], "author": { "name": "Ben Skeggs", "email": "bskeggs@redhat.com", "time": "Fri Jan 22 09:10:05 2010 +1000" }, "committer": { "name": "Ben Skeggs", "email": "bskeggs@redhat.com", "time": "Mon Jan 25 10:35:25 2010 +1000" }, "message": "drm/nv50: fail auxch transaction if reply count not what we expect\n\nSigned-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n" }, { "commit": "0208843dd59330c30e0773c77d5362260aa340e0", "tree": "c37f2dffe30c252b7f0a46fe5e8f6462227fd98f", "parents": [ "a1470890f29da73783e932078ed911f5871e3ad3" ], "author": { "name": "Ben Skeggs", "email": "bskeggs@redhat.com", "time": "Thu Jan 21 15:03:23 2010 +1000" }, "committer": { "name": "Ben Skeggs", "email": "bskeggs@redhat.com", "time": "Mon Jan 25 10:35:19 2010 +1000" }, "message": "drm/nouveau: fix failure path if userspace specifies no valid memtypes\n\nWe need to add the buffer to the list even if we fail, otherwise the\nvalidate_fini() call won\u0027t unreserve + unreference the GEM object,\nmaking TTM very unhappy.\n\nSigned-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n" }, { "commit": "a1470890f29da73783e932078ed911f5871e3ad3", "tree": "689f07958c378d173cdafab61365324f750f8921", "parents": [ "58d65b84dee524b0a4a8b747b8eb8bbf2fb16d9d" ], "author": { "name": "Ben Skeggs", "email": "bskeggs@redhat.com", "time": "Mon Jan 18 11:42:37 2010 +1000" }, "committer": { "name": "Ben Skeggs", "email": "bskeggs@redhat.com", "time": "Mon Jan 25 10:35:13 2010 +1000" }, "message": "drm/nouveau: report LVDS as disconnected if lid closed\n\nAlso adds a module option to ignore the status reported via ACPI, in case\nwe hit systems with broken ACPI.\n\nSigned-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n" }, { "commit": "38678d3557420a1c40f7ad5a04a46a7de7a305b9", "tree": "cfd81d00cfe47373921bb0c23589302cc7f1b576", "parents": [ "947bfc83043181e86e3e3134da158e479c9bcc51" ], "author": { "name": "Alex Deucher", "email": "alexdeucher@gmail.com", "time": "Thu Jan 21 19:28:18 2010 -0500" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Sun Jan 24 17:25:38 2010 +1000" }, "message": "drm/radeon/kms: fix legacy get_engine/memory clock\n\nFix a bad shift in the post div.\n\nShould fix fdo bug 26145\n\nSigned-off-by: Alex Deucher \u003calexdeucher@gmail.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "947bfc83043181e86e3e3134da158e479c9bcc51", "tree": "d3a7111bf9d8e6dd55519d7a28762d0dffd581db", "parents": [ "4eaeca33512b9774c25507b9a9bdcfe3791a5cc5" ], "author": { "name": "Alex Deucher", "email": "alexdeucher@gmail.com", "time": "Thu Jan 21 17:14:49 2010 -0500" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Sun Jan 24 17:25:05 2010 +1000" }, "message": "drm/radeon/kms/atom: atom parser fixes\n\nOnly reset the reg block on the initial execute\ntable call; nested calls require the reg block not be\nreset on each call. Also reset the fb window and\nio mode. This matches the upstream parser behavior.\n\nSigned-off-by: Alex Deucher \u003calexdeucher@gmail.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "4eaeca33512b9774c25507b9a9bdcfe3791a5cc5", "tree": "d171b1cf87604c4af8893a299cb872c950365aac", "parents": [ "fc10332b8ac5ca32d11f898027d84c007543bd80" ], "author": { "name": "Alex Deucher", "email": "alexdeucher@gmail.com", "time": "Tue Jan 19 17:32:27 2010 -0500" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Sun Jan 24 17:24:29 2010 +1000" }, "message": "drm/radeon/kms: clean up atombios pll code\n\n- split pll adjust into a separate function\n- use a union for SetPixelClock params\n\nSigned-off-by: Alex Deucher \u003calexdeucher@gmail.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "fc10332b8ac5ca32d11f898027d84c007543bd80", "tree": "28b24fb83e4c4bfc71286ce93f01271fa5027c95", "parents": [ "a348c84d953f61c776e53cde0a63a4e407a23c18" ], "author": { "name": "Alex Deucher", "email": "alexdeucher@gmail.com", "time": "Tue Jan 19 17:16:10 2010 -0500" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Sun Jan 24 17:24:23 2010 +1000" }, "message": "drm/radeon/kms: clean up pll struct\n\n- add a new flag for fixed post div\n- pull the pll flags into the struct\n\nSigned-off-by: Alex Deucher \u003calexdeucher@gmail.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "a348c84d953f61c776e53cde0a63a4e407a23c18", "tree": "92c2b62e05bc4bac4895eb5eb31f70d0ea25f345", "parents": [ "c8c15ff1e90bfc4a2db1ba77a01b3b2783e723fc" ], "author": { "name": "Alex Deucher", "email": "alexdeucher@gmail.com", "time": "Thu Jan 21 16:50:30 2010 -0500" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Sun Jan 24 17:23:38 2010 +1000" }, "message": "drm/radeon/kms/atom: fix crtc lock ordering\n\nThis makes crtc_prepare and crtc_commit match.\n\nSigned-off-by: Alex Deucher \u003calexdeucher@gmail.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "c8c15ff1e90bfc4a2db1ba77a01b3b2783e723fc", "tree": "21930390abaa2581fef69aef688906b3f72934f5", "parents": [ "db96380ea26fcc31ab37189aedeabd12894b1431" ], "author": { "name": "Jerome Glisse", "email": "jglisse@redhat.com", "time": "Mon Jan 18 13:01:36 2010 +0100" }, "committer": { "name": "Dave Airlie", "email": "airlied@linux.ie", "time": "Thu Jan 21 08:49:32 2010 +1000" }, "message": "drm/radeon: r6xx/r7xx possible security issue, system ram access\n\nThis patch workaround a possible security issue which can allow\nuser to abuse drm on r6xx/r7xx hw to access any system ram memory.\nThis patch doesn\u0027t break userspace, it detect \"valid\" old use of\nCB_COLOR[0-7]_FRAG \u0026 CB_COLOR[0-7]_TILE registers and overwritte\nthe address these registers are pointing to with the one of the\nlast color buffer. This workaround will work for old mesa \u0026\nxf86-video-ati and any old user which did use similar register\nprogramming pattern as those (we expect that there is no others\nuser of those ioctl except possibly a malicious one). This patch\nadd a warning if it detects such usage, warning encourage people\nto update their mesa \u0026 xf86-video-ati. New userspace will submit\nproper relocation.\n\nFix for xf86-video-ati / mesa (this kernel patch is enough to\nprevent abuse, fix for userspace are to set proper cs stream and\navoid kernel warning) :\nhttp://cgit.freedesktop.org/xorg/driver/xf86-video-ati/commit/?id\u003d95d63e408cc88b6934bec84a0b1ef94dfe8bee7b\nhttp://cgit.freedesktop.org/mesa/mesa/commit/?id\u003d46dc6fd3ed5ef96cda53641a97bc68c3bc104a9f\n\nAbusing this register to perform system ram memory is not easy,\nhere is outline on how it could be achieve. First attacker must\nhave access to the drm device and be able to submit command stream\nthrought cs ioctl. Then attacker must build a proper command stream\nfor r6xx/r7xx hw which will abuse the FRAG or TILE buffer to\noverwrite the GPU GART which is in VRAM. To achieve so attacker\nas to setup CB_COLOR[0-7]_FRAG or CB_COLOR[0-7]_TILE to point\nto the GPU GART, then it has to find a way to write predictable\nvalue into those buffer (with little cleverness i believe this\ncan be done but this is an hard task). Once attacker have such\nprogram it can overwritte GPU GART to program GPU gart to point\nanywhere in system memory. It then can reusse same method as he\nused to reprogram GART to overwritte the system ram through the\nGART mapping. In the process the attacker has to be carefull to\nnot overwritte any sensitive area of the GART table, like ring\nor IB gart entry as it will more then likely lead to GPU lockup.\nBottom line is that i think it\u0027s very hard to use this flaw\nto get system ram access but in theory one can achieve so.\n\nSide note: I am not aware of anyone ever using the GPU as an\nattack vector, nevertheless we take great care in the opensource\ndriver to try to detect and forbid malicious use of GPU. I don\u0027t\nthink the closed source driver are as cautious as we are.\n\nSigned-off-by: Jerome Glisse \u003cjglisse@redhat.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@linux.ie\u003e\n" }, { "commit": "db96380ea26fcc31ab37189aedeabd12894b1431", "tree": "ea5b5a5959b81fd2d60a705f96d0d1adeab86881", "parents": [ "7e71c9e2e7704ebf044d4a964e02fbd2098a173f" ], "author": { "name": "Jerome Glisse", "email": "jglisse@redhat.com", "time": "Sun Jan 17 21:21:56 2010 +0100" }, "committer": { "name": "Dave Airlie", "email": "airlied@linux.ie", "time": "Thu Jan 21 08:47:01 2010 +1000" }, "message": "drm/radeon/kms: r600/r700 don\u0027t test ib if ib initialization fails\n\nIf ib initialization failed don\u0027t try to test ib as it will result\nin an oops (accessing NULL ib buffer ptr).\n\nSigned-off-by: Jerome Glisse \u003cjglisse@redhat.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@linux.ie\u003e\n" }, { "commit": "7e71c9e2e7704ebf044d4a964e02fbd2098a173f", "tree": "f0231cd483dc1f4e6521b46802169baee4fda054", "parents": [ "7924e5eb8fe422d6b1ce3b3e2be749a480dfcdd9" ], "author": { "name": "Jerome Glisse", "email": "jglisse@redhat.com", "time": "Sun Jan 17 21:21:41 2010 +0100" }, "committer": { "name": "Dave Airlie", "email": "airlied@linux.ie", "time": "Thu Jan 21 08:46:27 2010 +1000" }, "message": "drm/radeon/kms: Forbid creation of framebuffer with no valid GEM object\n\nThis will avoid oops if at later point the fb is use. Trying to create\na framebuffer with no valid GEM object is bogus and should be forbidden\nas this patch does.\n\nSigned-off-by: Jerome Glisse \u003cjglisse@redhat.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@linux.ie\u003e\n" }, { "commit": "7924e5eb8fe422d6b1ce3b3e2be749a480dfcdd9", "tree": "9fbe9d6b1d23b01300e95bebd09b4ee4a3d1e697", "parents": [ "79c2bbc505751bb5130ac753251fc9a0eb37bb12" ], "author": { "name": "Jerome Glisse", "email": "jglisse@redhat.com", "time": "Fri Jan 15 14:44:39 2010 +0100" }, "committer": { "name": "Dave Airlie", "email": "airlied@linux.ie", "time": "Thu Jan 21 08:45:01 2010 +1000" }, "message": "drm/radeon/kms: r600 handle irq vector ring overflow\n\nIn some rare case i faced an irq overflow quickly followed by\na GPU lockup (hard hang) this patch try to deal with irq vector\nring overflow, so far haven\u0027t been able to reproduce it with\nthe patch.\n\nSigned-off-by: Jerome Glisse \u003cjglisse@redhat.com\u003e\nReviewed-by: Alex Deucher \u003calexdeucher@gmail.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@linux.ie\u003e\n" }, { "commit": "79c2bbc505751bb5130ac753251fc9a0eb37bb12", "tree": "1249938228d4630b9eeabf0d0619034c9666a3af", "parents": [ "0c45249f419d8b86abe0e51c6627ca4b085e8c23" ], "author": { "name": "Jerome Glisse", "email": "jglisse@redhat.com", "time": "Fri Jan 15 14:44:38 2010 +0100" }, "committer": { "name": "Dave Airlie", "email": "airlied@linux.ie", "time": "Thu Jan 21 08:43:56 2010 +1000" }, "message": "drm/radeon/kms: r600/r700 don\u0027t process IRQ if not initialized\n\nIn some rare case the wptr returned from the hw wasn\u0027t 0 and leaded\nto trick r600_process_irq that their were irq to process. Add a\ncheck to bail out if irq hasn\u0027t been initialized this will avoid\noops provoqued by the rare wptr !\u003d 0 on initialization.\n\nSigned-off-by: Jerome Glisse \u003cjglisse@redhat.com\u003e\nReviewed-by: Alex Deucher \u003calexdeucher@gmail.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@linux.ie\u003e\n" }, { "commit": "0c45249f419d8b86abe0e51c6627ca4b085e8c23", "tree": "bb50e5f91fbb7c2db683a3d1e84965dc83612acc", "parents": [ "615e0cb67968c94fd9e53797985843a5b816dec4" ], "author": { "name": "Jerome Glisse", "email": "jglisse@redhat.com", "time": "Fri Jan 15 14:44:37 2010 +0100" }, "committer": { "name": "Dave Airlie", "email": "airlied@linux.ie", "time": "Thu Jan 21 08:42:59 2010 +1000" }, "message": "drm/radeon/kms: r600/r700 disable irq at suspend\n\nTo avoid hw doing anythings after we disabled PCIE GART, fully\ndisable IRQ at suspend. Also cleanup a bit the ih structure\nand process function.\n\nSigned-off-by: Jerome Glisse \u003cjglisse@redhat.com\u003e\nReviewed-by: Alex Deucher \u003calexdeucher@gmail.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@linux.ie\u003e\n" }, { "commit": "615e0cb67968c94fd9e53797985843a5b816dec4", "tree": "7a6505324def47d4ecf295084ab2db49c0f336fe", "parents": [ "54f088a9603dbee88809cb2ddfd8dc1ef8a74be5" ], "author": { "name": "Alex Deucher", "email": "alexdeucher@gmail.com", "time": "Wed Jan 20 16:22:53 2010 -0500" }, "committer": { "name": "Dave Airlie", "email": "airlied@linux.ie", "time": "Thu Jan 21 08:40:08 2010 +1000" }, "message": "drm/radeon/kms/r4xx: cleanup atom path\n\nmost of radeon_legacy_atom_set_surface() is taken care\nof in atombios_set_base(), so remove the duplicate\nsetup and move the remaining bits (DISP_MERGE setup and\nFP2 sync) to atombios_crtc.c where they are used.\n\nSigned-off-by: Alex Deucher \u003calexdeucher@gmail.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@linux.ie\u003e\n" }, { "commit": "54f088a9603dbee88809cb2ddfd8dc1ef8a74be5", "tree": "81e03ac860920dba11ed452c59a7e75b2b37ddfe", "parents": [ "e2f8e87089427d3a203538dc73058b90e82b11f6" ], "author": { "name": "Alex Deucher", "email": "alexdeucher@gmail.com", "time": "Tue Jan 19 16:34:01 2010 -0500" }, "committer": { "name": "Dave Airlie", "email": "airlied@linux.ie", "time": "Thu Jan 21 08:39:02 2010 +1000" }, "message": "drm/radeon/kms: fix atombios_crtc_set_base\n\nMake it call the proper backend depending on the\nGPU family. Right now r4xx cards with atombios modesetting\nenabled were using the avivo crtc base code. This also\nallows us to add support for new asics more easily.\n\nSigned-off-by: Alex Deucher \u003calexdeucher@gmail.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@linux.ie\u003e\n" }, { "commit": "e2f8e87089427d3a203538dc73058b90e82b11f6", "tree": "0e86c59445979eeee01a18979b3505c0bbd39f34", "parents": [ "9f53e7931610cbd1715fd2d2c0f9a853e17f02d8" ], "author": { "name": "Alex Deucher", "email": "alexdeucher@gmail.com", "time": "Tue Jan 19 12:45:29 2010 -0500" }, "committer": { "name": "Dave Airlie", "email": "airlied@linux.ie", "time": "Thu Jan 21 08:20:28 2010 +1000" }, "message": "drm/radeon/kms/atom: upstream parser updates\n\nSigned-off-by: Alex Deucher \u003calexdeucher@gmail.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@linux.ie\u003e\n" }, { "commit": "9f53e7931610cbd1715fd2d2c0f9a853e17f02d8", "tree": "469c4db5900da5bc93aa158c4d9f33dff6e03a84", "parents": [ "07bec2df01cde5590a1700b992d12de3f46b12bc" ], "author": { "name": "Alex Deucher", "email": "alexdeucher@gmail.com", "time": "Tue Jan 19 12:38:48 2010 -0500" }, "committer": { "name": "Dave Airlie", "email": "airlied@linux.ie", "time": "Thu Jan 21 08:19:55 2010 +1000" }, "message": "drm/radeon/kms/atom: fix some parser bugs\n\n- add support for inline src params\n- fix shift_left/shift_right and shl/shr ops\n shift_* ops use inline src params, shl/r use full params\n- fix mask op (uses inline params)\n\nSigned-off-by: Alex Deucher \u003calexdeucher@gmail.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@linux.ie\u003e\n" }, { "commit": "07bec2df01cde5590a1700b992d12de3f46b12bc", "tree": "29208b029d02dbe02f11b2449a2d8529505192ce", "parents": [ "cf57fc7aa2ac61d02a29550b99db6a01ccd7917c" ], "author": { "name": "Alex Deucher", "email": "alexdeucher@gmail.com", "time": "Wed Jan 13 19:09:12 2010 -0500" }, "committer": { "name": "Dave Airlie", "email": "airlied@linux.ie", "time": "Thu Jan 21 08:19:42 2010 +1000" }, "message": "drm/radeon/kms: fix hardcoded mmio size in register functions\n\nnewer asics have large mmio apertures\n\nSigned-off-by: Alex Deucher \u003calexdeucher@gmail.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@linux.ie\u003e\n" }, { "commit": "cf57fc7aa2ac61d02a29550b99db6a01ccd7917c", "tree": "f266175928615dc0543d0ee4911f939e3e8d549c", "parents": [ "828153e29206196552c2cc34461ad7e5b93e3f22" ], "author": { "name": "Alex Deucher", "email": "alexdeucher@gmail.com", "time": "Mon Jan 18 20:20:07 2010 -0500" }, "committer": { "name": "Dave Airlie", "email": "airlied@linux.ie", "time": "Thu Jan 21 08:18:18 2010 +1000" }, "message": "drm/radeon/kms/r100: fix bug in CS parser\n\nThe first dword of PACKET3_3D_DRAW_IMMD maps to\nSE_VTX_FMT so the vertex size is part of the draw\npacket.\n\nThis patch fixes a possible case where you have a\ncommand buffer that does not contain SE_VTX_FMT\nregister write, but does contain PACKET3_3D_DRAW_IMMD.\n\nSigned-off-by: Alex Deucher \u003calexdeucher@gmail.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@linux.ie\u003e\n" }, { "commit": "828153e29206196552c2cc34461ad7e5b93e3f22", "tree": "292169ea4c4a665b5502a1138655ea5d783dccef", "parents": [ "f3d1ccc14f37a07614c67a270f05a8d94ec3667c" ], "author": { "name": "Andrew Randrianasulu", "email": "randrik@mail.ru", "time": "Wed Jan 20 11:56:07 2010 -0500" }, "committer": { "name": "Dave Airlie", "email": "airlied@linux.ie", "time": "Thu Jan 21 08:17:27 2010 +1000" }, "message": "drm/radeon/kms/r200: fix bug in CS parser\n\nAdd missing vertex shader regs for r200.\n\nfixed fdo bug 26061\n\nagd5f: use official reg names\n\nSigned-off-by: Alex Deucher \u003calexdeucher@gmail.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@linux.ie\u003e\n" }, { "commit": "f3d1ccc14f37a07614c67a270f05a8d94ec3667c", "tree": "22a22e8fcac232f351a3884b2948431408b0e50a", "parents": [ "700a0cc088a42a2ed92c6f961534fdb38588af87" ], "author": { "name": "Andrew Randrianasulu", "email": "randrik@mail.ru", "time": "Wed Jan 20 11:36:30 2010 -0500" }, "committer": { "name": "Dave Airlie", "email": "airlied@linux.ie", "time": "Thu Jan 21 08:17:13 2010 +1000" }, "message": "drm/radeon/kms/r200: fix bug in CS parser\n\nThe checks for CUBE and 3D textures were inverted.\n\nfixes fdo bug 24159\n\nagd5f: added comments for clarity.\n\nSigned-off-by: Alex Deucher \u003calexdeucher@gmail.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@linux.ie\u003e\n" }, { "commit": "58d65b84dee524b0a4a8b747b8eb8bbf2fb16d9d", "tree": "de9c4ef82d3c1a51d3fc0267b499e3dfcb02528a", "parents": [ "134f248bea4bf5c3169b4950eb49c6651b09eb0e" ], "author": { "name": "Ben Skeggs", "email": "bskeggs@redhat.com", "time": "Mon Jan 18 08:52:35 2010 +1000" }, "committer": { "name": "Ben Skeggs", "email": "bskeggs@redhat.com", "time": "Mon Jan 18 09:56:27 2010 +1000" }, "message": "drm/nv50: prevent accidently turning off encoders we\u0027re actually using\n\nOn most cards the DisplayPort connector is created with 2 encoders sharing\na single SOR (for native DP, and for DVI-over-DP). The previous logic\nfor turning off unused encoders didn\u0027t take into account that we could\nhave multiple drm_encoders on a single hw encoder and ended up turning off\nencoders that were actually being used still.\n\nThis patch fixes that issue. We probably want to look at something a bit\nbetter later on, and only expose one drm_encoder per hw encoder block.\n\nSigned-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n" }, { "commit": "134f248bea4bf5c3169b4950eb49c6651b09eb0e", "tree": "db66d0ce739c94538603f491017063c0ac8e0413", "parents": [ "71666475018a3024fb49499096d26a9350349e8b" ], "author": { "name": "Ben Skeggs", "email": "bskeggs@redhat.com", "time": "Mon Jan 18 08:33:04 2010 +1000" }, "committer": { "name": "Ben Skeggs", "email": "bskeggs@redhat.com", "time": "Mon Jan 18 09:56:19 2010 +1000" }, "message": "drm/nv50: fix alignment of per-channel fifo cache\n\nGPU pointer to the structure is shifted right by 10 bits, so we need to\nalign to 1024 bytes, not 256.\n\nReported-by: Maarten Maathuis \u003cmadman2003@gmail.com\u003e\nSigned-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n" }, { "commit": "71666475018a3024fb49499096d26a9350349e8b", "tree": "c10a83f9dcb7fbeb52ac4fd7fc2b8374f128ebc2", "parents": [ "d051bbb22e9d8e87e2a5b8efb176d1bfd0f7feee" ], "author": { "name": "Luca Barbieri", "email": "luca@luca-barbieri.com", "time": "Sat Jan 16 15:30:15 2010 +0100" }, "committer": { "name": "Ben Skeggs", "email": "bskeggs@redhat.com", "time": "Mon Jan 18 09:56:13 2010 +1000" }, "message": "drm/nouveau: Evict buffers in VRAM before freeing sgdma\n\nCurrently, we take down the sgdma engine without evicting all buffers\nfrom VRAM.\n\nThe TTM device release will try to evict anything in VRAM to GART\nmemory, but this will fail since sgdma has already been taken down.\n\nThis causes an infinite loop in kernel mode on module unload.\nIt usually doesn\u0027t happen because there aren\u0027t any buffer on close.\nHowever, if the GPU is locked up, this condition is easily triggered.\n\nThis patch fixes it in the simplest way possible by cleaning VRAM\nright before cleaning SGDMA memory.\n\nSigned-off-by: Luca Barbieri \u003cluca@luca-barbieri.com\u003e\nSigned-off-by: Francisco Jerez \u003ccurrojerez@riseup.net\u003e\nSigned-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n" }, { "commit": "d051bbb22e9d8e87e2a5b8efb176d1bfd0f7feee", "tree": "1a03f3b39bddfad3c40184a6eb9947fece59664a", "parents": [ "0d92971dd6973d865a873c24df2d7bfe06ec4486" ], "author": { "name": "Luca Barbieri", "email": "luca@luca-barbieri.com", "time": "Sat Jan 16 15:27:51 2010 +0100" }, "committer": { "name": "Ben Skeggs", "email": "bskeggs@redhat.com", "time": "Mon Jan 18 09:56:03 2010 +1000" }, "message": "drm/nouveau: Acknowledge DMA_VTX_PROTECTION PGRAPH interrupts\n\nCurrently Nouveau is unable to dismiss DMA_VTX_PROTECTION errors,\nwhich results in an infinite loop in the interrupt handler.\n\nThese errors are caused both by bugs in the Gallium driver and by\nuser-specified index buffers with out of bounds indices.\n\nBy mmio-tracing the nVidia drivers, I found out how this is done.\nOn DMA_VTX_PROTECTION, The nVidia driver reads the register 0x402000,\nalways getting the value 4, and then writes 4 back to 0x402000.\n\nThis patch adds that logic by reading 0x402000 and writing the same\nvalue back.\nIt\u0027s unclear what should happen if the value read is not 4, and\nthe current approach might not be the correct one.\n\nTo test this, modify mesa/progs/trivial/vbo-drawrange.c, defining\nELTOBJ to 1 and replacing indices with huge out of bounds integers.\n\nWithout this patch, the GPU and/or kernel should lock up.\nWith this patch, it should misrender as expected but not lock up.\n\nThe errors are still logged since they are useful for development.\n\nThis has been tested on NV49 and may not work on other cards.\n\nTo find out how things work on other cards, run the aforementioned\ntest using the blob with mmiotrace and grep for a read of the PGRAPH\nsource register.\n\nSigned-off-by: Luca Barbieri \u003cluca@luca-barbieri.com\u003e\nSigned-off-by: Francisco Jerez \u003ccurrojerez@riseup.net\u003e\nSigned-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n" }, { "commit": "0d92971dd6973d865a873c24df2d7bfe06ec4486", "tree": "5f9ae19e7ac4a7d4019586e22cc8a2fed2a8d892", "parents": [ "ba59953d281747b1f7518a60f0ba8ff671cd0d65" ], "author": { "name": "Ben Skeggs", "email": "bskeggs@redhat.com", "time": "Fri Jan 15 12:21:37 2010 +1000" }, "committer": { "name": "Ben Skeggs", "email": "bskeggs@redhat.com", "time": "Mon Jan 18 09:55:54 2010 +1000" }, "message": "drm/nouveau: fix thinko in nv04_instmem.c\n\nSigned-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n" }, { "commit": "ba59953d281747b1f7518a60f0ba8ff671cd0d65", "tree": "6019422fd736debff175e88cd57ee300a8532463", "parents": [ "12f735b79f0ad63964dedabed3eee8a581bb66a5" ], "author": { "name": "Ben Skeggs", "email": "bskeggs@redhat.com", "time": "Fri Jan 15 12:08:57 2010 +1000" }, "committer": { "name": "Ben Skeggs", "email": "bskeggs@redhat.com", "time": "Mon Jan 18 09:55:48 2010 +1000" }, "message": "drm/nouveau: fix a race condition in nouveau_dma_wait()\n\nCan be triggered easily on certain cards (NV46 and NV50 of mine) by\nrunning \"dmesg\", the DRM\u0027s channel will lockup.\n\nSigned-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n" }, { "commit": "8471a26b9c36c965d278020cc0699e2e95d120e5", "tree": "d0d2453ad6a756ddc772aa74b8ac35980d75ad87", "parents": [ "79b7dcb2af7da728819cf060b94ce8c9ead36f7e", "700a0cc088a42a2ed92c6f961534fdb38588af87" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Jan 15 15:40:39 2010 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Jan 15 15:40:39 2010 +1000" }, "message": "Merge remote branch \u0027korg/drm-radeon-testing\u0027 into drm-linus\n\n* korg/drm-radeon-testing:\n drm/radeon/kms: Use radeon_agp_disable when disabling AGP\n drm/radeon/kms: Disable AGP is aperture size \u003c 32M\n drm/radeon/kms: Fix r600 blit cleanup path\n drm/radeon/kms: Do not unpin buffer in fb destruction\n drm/radeon/kms: fix displayport-\u003edvi connector DDC.\n" }, { "commit": "79b7dcb2af7da728819cf060b94ce8c9ead36f7e", "tree": "278d7cd2ef10b708dbb80a94c111b5c158ac68a6", "parents": [ "1c974dc21898f9abbcb5e47ae9ddb7e6b473de2e" ], "author": { "name": "Jerome Glisse", "email": "jglisse@redhat.com", "time": "Thu Jan 14 19:02:20 2010 +0100" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Jan 15 13:29:52 2010 +1000" }, "message": "drm: EDID accept separate sync video mode\n\nX is accepting such video mode, do the same. Pointed out by Joshua Roys\non IRC. Fix https://bugzilla.redhat.com/show_bug.cgi?id\u003d540024\n\n[fix printf to use composite not integrated :- airlied]\n\nSigned-off-by: Jerome Glisse \u003cjglisse@redhat.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "1c974dc21898f9abbcb5e47ae9ddb7e6b473de2e", "tree": "60f76a234060d19c4d63335d292fc1b93330746a", "parents": [ "354fb52cb6138de0e6cf84a0f6a7f3467586e390", "12f735b79f0ad63964dedabed3eee8a581bb66a5" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Jan 15 13:22:59 2010 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Jan 15 13:25:38 2010 +1000" }, "message": "Merge remote branch \u0027nouveau/for-airlied\u0027 of ../drm-nouveau-next into drm-linus\n\n* \u0027nouveau/for-airlied\u0027 of ../drm-nouveau-next: (44 commits)\n drm/nouveau: check pushbuffer bounds in ioctl\n drm/nouveau: reserve VGA area for the moment\n drm/nouveau: Unset the EDID connector property when the EDID block goes away.\n drm/nouveau: Fallback to analog load detection when the EDID block is invalid.\n drm/nouveau: fix edid memleak in nouveau_connector\n drm/nouveau: Break some long lines.\n drm/nouveau: add NV18 device id to call_lvds_manufacturer_script\n drm/nv50: Fix typo in PGRAPH initialisation.\n drm/nouveau: less magic DCB 1.5 parsing\n drm/nouveau: assume no nv04 board has a DCB table\n drm/nouveau: remove PRIV0 check in nouveau_mem_close()\n drm/nouveau: wait on fence after bo move if validating for another channel\n drm/nouveau: trust init table registers are safe\n drm/nv50: wait for pgraph to idle before unloading the context\n" }, { "commit": "12f735b79f0ad63964dedabed3eee8a581bb66a5", "tree": "22a774a050cbd0967d0b89788375271f226ba252", "parents": [ "ac8fb975e8c88d312a376b035494be17548d01c6" ], "author": { "name": "Luca Barbieri", "email": "luca@luca-barbieri.com", "time": "Sun Jan 10 20:10:53 2010 +0100" }, "committer": { "name": "Ben Skeggs", "email": "bskeggs@redhat.com", "time": "Fri Jan 15 09:56:50 2010 +1000" }, "message": "drm/nouveau: check pushbuffer bounds in ioctl\n\nCurrently there is no check that the pushbuffer request bounds are inside\nthe TTM BO.\n\nThis allows to instruct the kernel to do relocations on user-selected\naddresses, since the relocation bounds checking relies on the request\nbounds.\n\nThis can oops the kernel accidentally and is easily exploitable.\n\nThis patch adds bound checking and alignment checking for -\u003eoffset and\n-\u003enr_dwords.\n\nIt also makes some variables unsigned, which should have no effect,\nbut prevents possible bounds checking problems.\n\nSigned-off-by: Luca Barbieri \u003cluca@luca-barbieri.com\u003e\nSigned-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n" }, { "commit": "ac8fb975e8c88d312a376b035494be17548d01c6", "tree": "961ea3f6dcbc5483a676daa8c91d6fc27adf068f", "parents": [ "b8780e2ad20feb8d086bc961a678e8fdb4fa296b" ], "author": { "name": "Ben Skeggs", "email": "bskeggs@redhat.com", "time": "Fri Jan 15 09:24:20 2010 +1000" }, "committer": { "name": "Ben Skeggs", "email": "bskeggs@redhat.com", "time": "Fri Jan 15 09:29:38 2010 +1000" }, "message": "drm/nouveau: reserve VGA area for the moment\n\nThis is to prevent things such as GART tables and other important GPU\nstructures being allocated there before we take over fbcon ourselves.\n\nThis is more of a workaround for the moment, a better solution will\nrequire some more invasive changes, but it\u0027ll be done at some point.\n\nSigned-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n" }, { "commit": "b8780e2ad20feb8d086bc961a678e8fdb4fa296b", "tree": "3c3e4196e0019d604f076544ef1b0f6866bb3252", "parents": [ "0ed3165e55fdedee6fc04b757907a2d5ccacdf18" ], "author": { "name": "Francisco Jerez", "email": "currojerez@riseup.net", "time": "Thu Jan 14 15:54:23 2010 +0100" }, "committer": { "name": "Ben Skeggs", "email": "bskeggs@redhat.com", "time": "Fri Jan 15 09:29:35 2010 +1000" }, "message": "drm/nouveau: Unset the EDID connector property when the EDID block goes away.\n\nSigned-off-by: Francisco Jerez \u003ccurrojerez@riseup.net\u003e\nSigned-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n" }, { "commit": "0ed3165e55fdedee6fc04b757907a2d5ccacdf18", "tree": "a4a3b95c12cc49efb32e324fb5e2f8071456876a", "parents": [ "c8ebe275790f2c5efc3ffec515c93b7a139829d3" ], "author": { "name": "Francisco Jerez", "email": "currojerez@riseup.net", "time": "Thu Jan 14 15:47:03 2010 +0100" }, "committer": { "name": "Ben Skeggs", "email": "bskeggs@redhat.com", "time": "Fri Jan 15 09:28:59 2010 +1000" }, "message": "drm/nouveau: Fallback to analog load detection when the EDID block is invalid.\n\nSigned-off-by: Francisco Jerez \u003ccurrojerez@riseup.net\u003e\nSigned-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n" }, { "commit": "c8ebe275790f2c5efc3ffec515c93b7a139829d3", "tree": "82093fe7938d881555112d939a04627292d71233", "parents": [ "3d9aefb8bd1be2b5e07acf0f938c21769f03cc64" ], "author": { "name": "Xavier Chantry", "email": "shiningxc@gmail.com", "time": "Mon Jan 11 22:42:21 2010 +0100" }, "committer": { "name": "Ben Skeggs", "email": "bskeggs@redhat.com", "time": "Fri Jan 15 09:28:48 2010 +1000" }, "message": "drm/nouveau: fix edid memleak in nouveau_connector\n\nThis was spotted by kmemleak.\n\nSigned-off-by: Xavier Chantry \u003cshiningxc@gmail.com\u003e\nSigned-off-by: Francisco Jerez \u003ccurrojerez@riseup.net\u003e\nSigned-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n" }, { "commit": "3d9aefb8bd1be2b5e07acf0f938c21769f03cc64", "tree": "e139993ae09cf8ec638a5c97f76991502ed71de7", "parents": [ "b79d8a65d52d3d80204e1bad30ee57bb5a58ced1" ], "author": { "name": "Francisco Jerez", "email": "currojerez@riseup.net", "time": "Thu Jan 14 14:56:54 2010 +0100" }, "committer": { "name": "Ben Skeggs", "email": "bskeggs@redhat.com", "time": "Fri Jan 15 09:28:10 2010 +1000" }, "message": "drm/nouveau: Break some long lines.\n\nSigned-off-by: Francisco Jerez \u003ccurrojerez@riseup.net\u003e\nSigned-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n" }, { "commit": "b79d8a65d52d3d80204e1bad30ee57bb5a58ced1", "tree": "071cc6aaa24e1307ce066d1d7027319efbbe0ad7", "parents": [ "716abaa8e52548e8863831cb32984655f46b5566" ], "author": { "name": "Andrea Tacconi", "email": "tacconet@libero.it", "time": "Thu Jan 14 14:52:37 2010 +0100" }, "committer": { "name": "Ben Skeggs", "email": "bskeggs@redhat.com", "time": "Fri Jan 15 09:28:02 2010 +1000" }, "message": "drm/nouveau: add NV18 device id to call_lvds_manufacturer_script\n\nThis fixes imac black screen (NV18 card)\n\nSigned-off-by: Andrea Tacconi \u003ctacconet@libero.it\u003e\nSigned-off-by: Francisco Jerez \u003ccurrojerez@riseup.net\u003e\nSigned-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n" }, { "commit": "716abaa8e52548e8863831cb32984655f46b5566", "tree": "4b54978a0285490f2b2a26b7f45e5e4414dc1deb", "parents": [ "b0d2de860bd621959bc826ffd42618fe1de37a61" ], "author": { "name": "Marcin Kościelnicki", "email": "koriakin@0x04.net", "time": "Tue Jan 12 18:21:56 2010 +0000" }, "committer": { "name": "Ben Skeggs", "email": "bskeggs@redhat.com", "time": "Thu Jan 14 18:49:05 2010 +1000" }, "message": "drm/nv50: Fix typo in PGRAPH initialisation.\n\nThis enables streamout functionality.\n\nSigned-off-by: Marcin Kościelnicki \u003ckoriakin@0x04.net\u003e\nSigned-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n" }, { "commit": "b0d2de860bd621959bc826ffd42618fe1de37a61", "tree": "0d9d506c8cbc698489e7c0569a63f0089ce9d1e9", "parents": [ "ed42f8240cfea13580fe91195e52c5247275e7df" ], "author": { "name": "Ben Skeggs", "email": "bskeggs@redhat.com", "time": "Thu Jan 14 17:53:04 2010 +1000" }, "committer": { "name": "Ben Skeggs", "email": "bskeggs@redhat.com", "time": "Thu Jan 14 18:48:59 2010 +1000" }, "message": "drm/nouveau: less magic DCB 1.5 parsing\n\nThis in the very least matches the parsing of all the previously known\nentries, and hopefully (at least closer to) correct for any we haven\u0027t\nseen yet.\n\nSigned-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n" }, { "commit": "ed42f8240cfea13580fe91195e52c5247275e7df", "tree": "69b0d833d0da2c7c41cabbd10c9904f1bcc78bf5", "parents": [ "77cb5c75e89ccfeb993d9dec37f8edb3a855e9f4" ], "author": { "name": "Ben Skeggs", "email": "bskeggs@redhat.com", "time": "Thu Jan 14 15:58:10 2010 +1000" }, "committer": { "name": "Ben Skeggs", "email": "bskeggs@redhat.com", "time": "Thu Jan 14 18:48:52 2010 +1000" }, "message": "drm/nouveau: assume no nv04 board has a DCB table\n\nThere\u0027s a report of a TNT2 where the DCB table pointer is *not* NULL\n(it contains a part of a VBIOS data string), and we assume this means\na DCB table is present, causing all kinds of hilarity.\n\nSigned-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n" }, { "commit": "77cb5c75e89ccfeb993d9dec37f8edb3a855e9f4", "tree": "77bd8a25ba752b26fe54004433ab344dc2c7b5cd", "parents": [ "e147eae80afe98b86a7921a40a709584e6603e34" ], "author": { "name": "Ben Skeggs", "email": "bskeggs@redhat.com", "time": "Wed Jan 13 11:22:33 2010 +1000" }, "committer": { "name": "Ben Skeggs", "email": "bskeggs@redhat.com", "time": "Thu Jan 14 18:48:46 2010 +1000" }, "message": "drm/nouveau: remove PRIV0 check in nouveau_mem_close()\n\nWe don\u0027t setup PRIV0 anymore, so this is unnecessary.\n\nSigned-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n" }, { "commit": "e147eae80afe98b86a7921a40a709584e6603e34", "tree": "b40bdaf039d5d2b2a90794bff102cf9a42abd745", "parents": [ "9855e584d296a32bbcc1bb3d29b7e0ed078b03cc" ], "author": { "name": "Ben Skeggs", "email": "bskeggs@redhat.com", "time": "Tue Jan 12 15:28:19 2010 +1000" }, "committer": { "name": "Ben Skeggs", "email": "bskeggs@redhat.com", "time": "Thu Jan 14 18:48:38 2010 +1000" }, "message": "drm/nouveau: wait on fence after bo move if validating for another channel\n\nNot an ideal solution, but it\u0027ll do for the moment for correctness. We\nneed to come up with a nicer way to manage inter-channel sync, the hw\nis unfortunately a little lacking in this area.\n\nShould fix some resume corruption, as well as corruption that may be seen\nwhile under memory pressure.\n\nSigned-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n" }, { "commit": "9855e584d296a32bbcc1bb3d29b7e0ed078b03cc", "tree": "7fbcc6ccdcc3dce65ac2b22ccd1c2f2572770931", "parents": [ "0a90dc51aa3162abb7cd64323ba2e6674e1c5ff4" ], "author": { "name": "Ben Skeggs", "email": "bskeggs@redhat.com", "time": "Tue Jan 12 13:02:19 2010 +1000" }, "committer": { "name": "Ben Skeggs", "email": "bskeggs@redhat.com", "time": "Thu Jan 14 18:48:31 2010 +1000" }, "message": "drm/nouveau: trust init table registers are safe\n\nApparently the original reason for checking this was there were known\nregister accesses that caused hangs on some chipsets. This was more\nthan likely because of incorrect parsing of previous opcodes, and I\nhardly think aborting a script half way through is going to be any\nbetter (in fact, we have had bug reports where this has been the cause\nof s/r failures among other things).\n\nThis patch (which has been in Fedora 12 for a long time now) removes\nall checking for known register ranges, and just leaves the check to\nensure the access is within the mapped aperture to avoid an oops.\n\nSigned-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n" }, { "commit": "0a90dc51aa3162abb7cd64323ba2e6674e1c5ff4", "tree": "31fe16e9d83d601e3a60f0f924966a6023aee2b2", "parents": [ "fe5a64acbfab4f79ccd18972c3d4c1cae31f3b8d" ], "author": { "name": "Maarten Maathuis", "email": "madman2003@gmail.com", "time": "Mon Jan 11 21:18:53 2010 +0100" }, "committer": { "name": "Ben Skeggs", "email": "bskeggs@redhat.com", "time": "Thu Jan 14 18:48:21 2010 +1000" }, "message": "drm/nv50: wait for pgraph to idle before unloading the context\n\nThis should fix the problem with gpu hangs people have had when closing\nchannels.\n\nSigned-off-by: Maarten Maathuis \u003cmadman2003@gmail.com\u003e\nSigned-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n" }, { "commit": "354fb52cb6138de0e6cf84a0f6a7f3467586e390", "tree": "fb82e886349c36bcf21afb51bf5fad9ead47d0a9", "parents": [ "d9f36a0051b7c0382107cb0342af1126a6eb627d" ], "author": { "name": "Thomas Hellstrom", "email": "thellstrom@vmware.com", "time": "Wed Jan 13 22:28:45 2010 +0100" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Jan 14 12:20:04 2010 +1000" }, "message": "drm/ttm: Make sure system buffer objects has offset \u003d\u003d 0.\n\nThis is a convention that the vmwgfx driver has come to rely on.\n\nSigned-off-by: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "d9f36a0051b7c0382107cb0342af1126a6eb627d", "tree": "e39118bbf806914b7124ee9ab4b0b51a585d493c", "parents": [ "e99e1e7893ac80fe769477cb1ddd4b482cef8902" ], "author": { "name": "Thomas Hellstrom", "email": "thellstrom@vmware.com", "time": "Wed Jan 13 22:28:43 2010 +0100" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Jan 14 12:19:57 2010 +1000" }, "message": "drm/vmwgfx: Implement basic pm operations.\n\nCurrently we really only support S3, since the device doesn\u0027t support\nsaving of the 3D state.\n\nOn S3/S4, move all buffer objects to swappable memory and take down\nGMR bindings. We need to do that from a PM notifier since we can\u0027t\ndo persistant memory allocations from the standard PM callbacks.\n\nSigned-off-by: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "e99e1e7893ac80fe769477cb1ddd4b482cef8902", "tree": "16818131508809a7e0366447b627b604f7e7076d", "parents": [ "476d51dbdbaa2e61fa4899459c658f476eee3fd9" ], "author": { "name": "Thomas Hellstrom", "email": "thellstrom@vmware.com", "time": "Wed Jan 13 22:28:42 2010 +0100" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Jan 14 12:19:50 2010 +1000" }, "message": "drm/ttm: Export symbols needed for vmwgfx suspend / resume operations.\n\nSigned-off-by: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "476d51dbdbaa2e61fa4899459c658f476eee3fd9", "tree": "4917e97ae7d01226296bc665b0f0d91e2f7491c1", "parents": [ "3f09ea4ecdcbcea05541f83e557d6ce2e56626d8" ], "author": { "name": "Thomas Hellstrom", "email": "thellstrom@vmware.com", "time": "Wed Jan 13 22:28:41 2010 +0100" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Jan 14 12:19:01 2010 +1000" }, "message": "drm/vmwgfx: Implement a swap_notify callback.\n\nUnbind GMR bindings on the buffer about to be swapped out.\n\nSigned-off-by: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "3f09ea4ecdcbcea05541f83e557d6ce2e56626d8", "tree": "445e4519b2e8a39136f611f07db670d1e10ace6e", "parents": [ "effe1105bef07f42366c20eac41b80ff9fcf675e" ], "author": { "name": "Thomas Hellstrom", "email": "thellstrom@vmware.com", "time": "Wed Jan 13 22:28:40 2010 +0100" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Jan 14 12:18:54 2010 +1000" }, "message": "drm/ttm: Add a swap_notify callback.\n\nThis is needed for a bugfix in the vmwgfx driver.\nDrivers may have GPU bindings on buffers that core TTM is not aware of,\nand TTM may view those buffers as ordinary system memory buffers.\nAdd a notifier to such drivers when TTM is about to move the buffer\ncontents out to swappable memory. The driver must then release any\nprivate GPU bindings on those buffers.\n\nSigned-off-by: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "effe1105bef07f42366c20eac41b80ff9fcf675e", "tree": "d4341a955614b8be1732f4ffc37f39016dffa61d", "parents": [ "7704befbd599e9c9524f640e14658ca8ed9d8717" ], "author": { "name": "Thomas Hellstrom", "email": "thellstrom@vmware.com", "time": "Wed Jan 13 22:28:39 2010 +0100" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Jan 14 12:18:43 2010 +1000" }, "message": "drm/vmwgfx: Use bo_driver::move_notify to unbind GMRs.\n\nThis was previously done explicitly for overlay- and fb buffers.\nNow it\u0027s done for any buffer leaving the SYSTEM memory region.\n\nSigned-off-by: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "7704befbd599e9c9524f640e14658ca8ed9d8717", "tree": "9ee43b851160e50234d8bd67967e7eb07104d8b9", "parents": [ "50ec3b7c35fda131e92abc6f9d93a230b7e009eb" ], "author": { "name": "Thomas Hellstrom", "email": "thellstrom@vmware.com", "time": "Wed Jan 13 22:28:38 2010 +0100" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Jan 14 12:18:02 2010 +1000" }, "message": "drm/vmwgfx: Make fence sequences continous across a VT switch.\n\nA vt switch in stealth mode would take down the FIFO, and re-\ninitialize fence sequence numbers. This patch\nsaves the current state of the fence sequence when the FIFO is\ndisabled.\n\nSigned-off-by: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "50ec3b7c35fda131e92abc6f9d93a230b7e009eb", "tree": "3fb8bbe2794bfb76f56549d070ed53f806f3bf94", "parents": [ "df1c93bae2b671e2aa50899a120af20afdbd504b" ], "author": { "name": "Thomas Hellstrom", "email": "thellstrom@vmware.com", "time": "Wed Jan 13 22:28:37 2010 +0100" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Jan 14 12:17:56 2010 +1000" }, "message": "drm/vmwgfx: Fix an error path causing an oops.\n\nAn error happening before the snooper.image member had been set up\nwould cause a kfree of an arbitrary pointer. Set up the snooper.image\nmember early.\n\nSigned-off-by: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "df1c93bae2b671e2aa50899a120af20afdbd504b", "tree": "20eef5d14f9605ffe7c761d70035556c97d06fc0", "parents": [ "194fda0dd83623f7927d505e39008c73fbc1c141" ], "author": { "name": "Thomas Hellstrom", "email": "thellstrom@vmware.com", "time": "Wed Jan 13 22:28:36 2010 +0100" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Jan 14 12:17:44 2010 +1000" }, "message": "drm/vmwgfx: Don\u0027t promote updates from GMR-backed scanouts to fullscreen.\n\nThat\u0027s unnecessary since partial screen updates from GMRs are fast.\nAlso fix cliprect pointer dereferencing\n\nSigned-off-by: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "700a0cc088a42a2ed92c6f961534fdb38588af87", "tree": "81072f73651c411080d82d7c6466dfe46f4ce0c4", "parents": [ "0a3f316370da07fcdb11fde4f5a683d60713b7fd" ], "author": { "name": "Jerome Glisse", "email": "jglisse@redhat.com", "time": "Wed Jan 13 15:16:38 2010 +0100" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Jan 14 11:53:55 2010 +1000" }, "message": "drm/radeon/kms: Use radeon_agp_disable when disabling AGP\n\nUse same common function to disable agp so we replace the GART\ncallback by the proper one when we do so. This fix oops if\nradeon_agp_init report failure.\n\nThis patch also move radeon_agp_init out of *_mc_init for r600\n\u0026 rv770 so that we can have a similar behavior than for previous\nhw, ie if agp_init fails it will fallback to GPU GART and disable\nAGP.\n\nSigned-off-by: Jerome Glisse \u003cjglisse@redhat.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "0a3f316370da07fcdb11fde4f5a683d60713b7fd", "tree": "8b1fc890b7ee33f618ffa969156f26020e4c88a2", "parents": [ "30d2d9a54d48e4fefede0389ded1b6fc2d44a522" ], "author": { "name": "Jerome Glisse", "email": "jglisse@redhat.com", "time": "Wed Jan 13 15:16:39 2010 +0100" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Jan 14 11:53:45 2010 +1000" }, "message": "drm/radeon/kms: Disable AGP is aperture size \u003c 32M\n\nradeon KMS need a GART of at least 32M to properly work. This patch\ncheck the AGP aperture size and disable if it\u0027s less than 32M. Note\nthan unlike non KMS path we don\u0027t staticaly allocate AGP memory so\nwe are not wasting memory not used by graphic processing.\n\nSigned-off-by: Jerome Glisse \u003cjglisse@redhat.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "30d2d9a54d48e4fefede0389ded1b6fc2d44a522", "tree": "f3cc8c5bd05bf6ae4605522f7f57d967fd25f30f", "parents": [ "6398d42454ce1671b3422c34fd0f600c9c76c8b1" ], "author": { "name": "Jerome Glisse", "email": "jglisse@redhat.com", "time": "Wed Jan 13 10:29:27 2010 +0100" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Jan 14 11:53:36 2010 +1000" }, "message": "drm/radeon/kms: Fix r600 blit cleanup path\n\nr600 blit cleanup path need to check if a bo was allocated before\ntrying to free or unpin it. This patch add this check and avoid\noops when the initialization on r6xx or r7xx hw fails.\n\nSigned-off-by: Jerome Glisse \u003cjglisse@redhat.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "6398d42454ce1671b3422c34fd0f600c9c76c8b1", "tree": "d52d909cc653b1fb9a6553ca973900d1bea2aabf", "parents": [ "7a15cbd40108a670baae71cbd3cec62b65891fa6" ], "author": { "name": "Jerome Glisse", "email": "jglisse@redhat.com", "time": "Wed Jan 13 12:32:36 2010 +0100" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Jan 14 11:53:21 2010 +1000" }, "message": "drm/radeon/kms: Do not unpin buffer in fb destruction\n\nIt\u0027s not necessary to unpin buffer in fb destruction. pin/unpin\nneed to be balanced and we don\u0027t pin in fb creation. We pin when\nan fb is associated to a crtc and unpin when the fb is disassociated\nfrom the crtc.\n\nNote:\nMaybe we should take reference on fb in set_base callback so fb\ndoesn\u0027t disappear until it\u0027s unbind from ctrc.\n\nSigned-off-by: Jerome Glisse \u003cjglisse@redhat.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "7a15cbd40108a670baae71cbd3cec62b65891fa6", "tree": "cafb93efc353d4be301ee039b8eca1ca25494cb2", "parents": [ "9270eb1b496cb002d75f49ef82c9ef4cbd22a5a0" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Jan 14 11:42:17 2010 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Jan 14 11:51:56 2010 +1000" }, "message": "drm/radeon/kms: fix displayport-\u003edvi connector DDC.\n\nIt appears that attempting AUXCH DDC breaks the subsequent attempt\nto do DDC over the i2c lines, so use the sink type to determine\nif we should be doing AUXCH or i2c DDC.\n\nThis fixes my DVI monitor plugged into DP-\u003eDVI convertor.\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "194fda0dd83623f7927d505e39008c73fbc1c141", "tree": "80ad7b8eb803ec8712465e74e584ad846c8b35a8", "parents": [ "ef14587706521287f1c7ea3326e732f7d86dd096", "9270eb1b496cb002d75f49ef82c9ef4cbd22a5a0" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Wed Jan 13 16:17:38 2010 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Wed Jan 13 16:17:41 2010 +1000" }, "message": "Merge remote branch \u0027korg/drm-radeon-next\u0027 into drm-linus\n\n* korg/drm-radeon-next\n drm/radeon/kms: only evict to GTT if CP is ready\n drm/radeon/kms: Fix crash getting TV info with no BIOS.\n drm/radeon/kms/rv100: reject modes \u003e 135 Mhz on DVI (v2)\n drm/radeon/kms/r6xx+: make irq handler less verbose\n drm/radeon/kms: fix up LVDS handling on macs (v2)\n" }, { "commit": "ef14587706521287f1c7ea3326e732f7d86dd096", "tree": "73b257c6d393f073d2ea7358bf99d8864ca67de5", "parents": [ "70a94d6a35072b62f808155f117f00485a395f03" ], "author": { "name": "Dave Young", "email": "hidave.darkstar@gmail.com", "time": "Wed Jan 13 13:38:59 2010 +0800" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Wed Jan 13 16:16:05 2010 +1000" }, "message": "drm: change drm set mode messages as DRM_DEBUG\n\nFollowing drm info repeat 207 times during one hour, it\u0027s quite annoying\n[ 1266.286747] [drm] TV-19: set mode NTSC 480i 0\n\nChange from DRM_INFO to DRM_DEBUG\n\nSigned-off-by: Dave Young \u003chidave.darkstar@gmail.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "70a94d6a35072b62f808155f117f00485a395f03", "tree": "a425defdab10f053ee863f43b0aad016fbd0c945", "parents": [ "f22d6ddaeb8126623d62c828a4d4a96dfc4cbc5c" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Wed Jan 13 16:15:11 2010 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Wed Jan 13 16:15:11 2010 +1000" }, "message": "drm: fix crtc no modes printf + typo\n\nToralf Förster pointed out the typo, the fact I forget the if\nstatement is purely personal fail.\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "9270eb1b496cb002d75f49ef82c9ef4cbd22a5a0", "tree": "10ad9294e12834b7ff4ea9d91ea5e0bc73744c55", "parents": [ "11f3b59e3654c66c4e8ef2c48f8138b78bf440da" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Wed Jan 13 09:21:49 2010 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Wed Jan 13 13:25:14 2010 +1000" }, "message": "drm/radeon/kms: only evict to GTT if CP is ready\n\nTesting GTT ready might be more correct but cp.ready\nworks fine and has been tested on irc by 2-3 ppl.\n\nfixes bug k.org 15035 and fd.o 25733\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "11f3b59e3654c66c4e8ef2c48f8138b78bf440da", "tree": "0d279bd7a56537e199cbdc9e9c6d9cd783b8c869", "parents": [ "1b24203e51072b6e76aff8c74bdd67eb3b34a724" ], "author": { "name": "Michel Dänzer", "email": "daenzer@vmware.com", "time": "Mon Jan 11 08:58:38 2010 +0100" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Wed Jan 13 13:25:10 2010 +1000" }, "message": "drm/radeon/kms: Fix crash getting TV info with no BIOS.\n\nSigned-off-by: Michel Dänzer \u003cdaenzer@vmware.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "1b24203e51072b6e76aff8c74bdd67eb3b34a724", "tree": "f4700c8190369d3e331edd0f7d89f8e523a7d416", "parents": [ "b042589ca038e647fa1e2bb4e7ac3963688479b8" ], "author": { "name": "Alex Deucher", "email": "alexdeucher@gmail.com", "time": "Mon Jan 11 15:02:31 2010 -0500" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Wed Jan 13 13:25:10 2010 +1000" }, "message": "drm/radeon/kms/rv100: reject modes \u003e 135 Mhz on DVI (v2)\n\nDue to heat issues. Fixes fdo bug 25992\n\nv2: fix typo noticed by Maarten Maathuis\n\nSigned-off-by: Alex Deucher \u003calexdeucher@gmail.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "b042589ca038e647fa1e2bb4e7ac3963688479b8", "tree": "0389d3223132139c2894edc603ddfb418d16e081", "parents": [ "3890ddf56dbc0f804953198e65a7e406ed654576" ], "author": { "name": "Alex Deucher", "email": "alexdeucher@gmail.com", "time": "Mon Jan 11 19:47:38 2010 -0500" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Wed Jan 13 13:25:09 2010 +1000" }, "message": "drm/radeon/kms/r6xx+: make irq handler less verbose\n\nUnhandled vectors can be safely ignored, no need\nto spam the kernel log by default.\n\nSigned-off-by: Alex Deucher \u003calexdeucher@gmail.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "3890ddf56dbc0f804953198e65a7e406ed654576", "tree": "18b05e3fdc4f0fc59cd65891c3197acd7ad5cefc", "parents": [ "804c7559e9376c3ba78ae15a30337b1e24f8ae80" ], "author": { "name": "Alex Deucher", "email": "alexdeucher@gmail.com", "time": "Tue Jan 12 11:16:57 2010 -0500" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Wed Jan 13 13:25:08 2010 +1000" }, "message": "drm/radeon/kms: fix up LVDS handling on macs (v2)\n\nBased on radeonfb code and recent ddx fix.\n\nv2: minor formatting fix from Michel Dänzer\n\nSigned-off-by: Alex Deucher \u003calexdeucher@gmail.com\u003e\nReviewed-by: Michel Dänzer \u003cmichel@daenzer.net\u003e\nTested-by: Michel Dänzer \u003cmichel@daenzer.net\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "f22d6ddaeb8126623d62c828a4d4a96dfc4cbc5c", "tree": "f866bf0e2445144208a9884de89b50ca94c43be2", "parents": [ "0c9d2c418aa4a45534943c4c9a1c8dda82d3b481", "40c2298bdcc8b766a39964c44e9a74d16aa95d53" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Jan 11 14:43:16 2010 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Jan 11 14:43:16 2010 +1000" }, "message": "Merge branch \u0027for-airlied\u0027 of /ssd/git/drm-nouveau-next into drm-linus\n\n* \u0027for-airlied\u0027 of /ssd/git/drm-nouveau-next: (28 commits)\n drm/nv04: Fix set_operation software method.\n drm/nouveau: initialise DMA tracking parameters earlier\n drm/nouveau: use dma.max rather than pushbuf size for checking GET validity\n drm/nv04: differentiate between nv04/nv05\n drm/nouveau: Fix null deref in nouveau_fence_emit due to deleted fence\n drm/nv50: prevent a possible ctxprog hang\n drm/nouveau: have ttm\u0027s fault handler called directly\n drm/nv50: restore correct cache1 get/put address on fifoctx load\n drm/nouveau: create function for \"dealing\" with gpu lockup\n drm/nouveau: remove unused nouveau_channel_idle() function\n drm/nouveau: fix handling of fbcon colours in 8bpp\n drm/nv04: Context switching fixes.\n drm/nouveau: Use the software object for fencing.\n drm/nouveau: Allocate a per-channel instance of NV_SW.\n drm/nv50: make the blocksize depend on vram size\n drm/nouveau: better alignment of bo sizes and use roundup instead of ALIGN\n drm/nouveau: Don\u0027t skip card take down on nv0x.\n drm/nouveau: Implement nv42-nv43 TV load detection.\n drm/nouveau: Clean up the nv17-nv4x load detection code a bit.\n drm/nv50: fix fillrect color\n ...\n" }, { "commit": "0c9d2c418aa4a45534943c4c9a1c8dda82d3b481", "tree": "7844c421b7bd20b9b4e9c17b425ee050653eb407", "parents": [ "94fd163d86b049842856864cdeac318131ec576d", "804c7559e9376c3ba78ae15a30337b1e24f8ae80" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Jan 11 14:42:58 2010 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Jan 11 14:42:58 2010 +1000" }, "message": "Merge remote branch \u0027korg/drm-radeon-next\u0027 into drm-linus\n\n* korg/drm-radeon-next:\n drm/radeon/kms: add additional safe regs for r4xx/rs6xx and r5xx\n drm/radeon/kms: Don\u0027t try to enable IRQ if we have no handler installed\n drm: Avoid calling vblank function is vblank wasn\u0027t initialized\n drm/radeon: mkregtable.c: close a file before exit\n drm/radeon/kms: Make sure we release AGP device if we acquired it\n drm/radeon/kms: Schedule host path read cache flush through the ring V2\n drm/radeon/kms: Workaround RV410/R420 CP errata (V3)\n drm/radeon/kms: detect sideport memory on IGP chips\n drm/radeon: fix a couple of array index errors\n drm/radeon/kms: add support for eDP (embedded DisplayPort)\n drm: Add eDP connector type\n drm/radeon/kms: pull in the latest upstream ObjectID.h changes\n drm/radeon/kms: whitespace changes to ObjectID.h\n drm/radeon/kms: fix typo in atom connector type handling\n" }, { "commit": "40c2298bdcc8b766a39964c44e9a74d16aa95d53", "tree": "e3da8a0aa4cf15e256add5b742c7df6d6a2064be", "parents": [ "c63834e1db41b59d6c7bfb1d2a549c027a42a877" ], "author": { "name": "Marcin Kościelnicki", "email": "koriakin@0x04.net", "time": "Sun Jan 10 17:09:14 2010 +0000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Jan 11 14:41:19 2010 +1000" }, "message": "drm/nv04: Fix set_operation software method.\n\nSigned-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n" }, { "commit": "c63834e1db41b59d6c7bfb1d2a549c027a42a877", "tree": "daa0c39043527695dc4c87e1eb4c66acfe50d5db", "parents": [ "400f14ac4ef02b2f77c9d0e3ad7d66e2f6c8e663" ], "author": { "name": "Ben Skeggs", "email": "bskeggs@redhat.com", "time": "Fri Jan 08 10:57:39 2010 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Jan 11 14:41:18 2010 +1000" }, "message": "drm/nouveau: initialise DMA tracking parameters earlier\n\nSigned-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n" }, { "commit": "400f14ac4ef02b2f77c9d0e3ad7d66e2f6c8e663", "tree": "0b8174d1a3ed384462b2d751b31ca6a0606efd4c", "parents": [ "cc6e496587502057af02139931736b0b7a49f637" ], "author": { "name": "Ben Skeggs", "email": "bskeggs@redhat.com", "time": "Fri Jan 08 10:53:40 2010 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Jan 11 14:41:18 2010 +1000" }, "message": "drm/nouveau: use dma.max rather than pushbuf size for checking GET validity\n\nSome upcoming G80 DMA changes will depend on this, but it\u0027s split out for\nbisectibility just in case it causes some unexpected issues.\n\nSigned-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n" }, { "commit": "cc6e496587502057af02139931736b0b7a49f637", "tree": "7aea86122a8efb91b2a3fa50f1fb97ede0cf6a5d", "parents": [ "d6126c5c8b2019658aadc9754dca80a7573dbff5" ], "author": { "name": "Ben Skeggs", "email": "bskeggs@redhat.com", "time": "Thu Jan 07 13:47:57 2010 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Jan 11 14:41:17 2010 +1000" }, "message": "drm/nv04: differentiate between nv04/nv05\n\nSigned-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n" }, { "commit": "d6126c5c8b2019658aadc9754dca80a7573dbff5", "tree": "8c16737a07bf7b80b8048c3f13802c74ef5662b8", "parents": [ "dc8d76cac942e7344a72ad18afb90fa46cf20bb4" ], "author": { "name": "Luca Barbieri", "email": "luca@luca-barbieri.com", "time": "Wed Jan 06 04:02:45 2010 +0100" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Jan 11 14:41:17 2010 +1000" }, "message": "drm/nouveau: Fix null deref in nouveau_fence_emit due to deleted fence\n\nCurrently Nouveau will unvalidate all buffers if it is forced to wait on\none, and then start revalidating from the beginning. While doing so, it\ndestroys the operation fence, causing nouveau_fence_emit to crash.\n\nThis patch fixes this bug by taking the fence object out of validate_op\nand creating it just before emit. The fence pointer is initialized to 0\nand unref\u0027ed unconditionally.\n\nIn addition to fixing the bug, this prevents its reintroduction and\nsimplifies the code.\n\nSigned-off-by: Luca Barbieri \u003cluca@luca-barbieri.com\u003e\nSigned-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n" }, { "commit": "dc8d76cac942e7344a72ad18afb90fa46cf20bb4", "tree": "0516fdca404bc9a87271710a93159bfda5a04c77", "parents": [ "1959ca80e1f88b82c1cb7227f437910768ab0c94" ], "author": { "name": "Ben Skeggs", "email": "bskeggs@redhat.com", "time": "Wed Jan 06 12:00:02 2010 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Jan 11 14:41:16 2010 +1000" }, "message": "drm/nv50: prevent a possible ctxprog hang\n\nThe below is mainly an educated guess at what\u0027s going on, docs would\nsure be handy... NVIDIA? :P\n\nIt appears it\u0027s possible for a ctxprog to run even while a GPU exception\nis pending. The GF8 and up ctxprogs appear to have a small snippet of\ncode which detects this, and stalls the ctxprog until it\u0027s been handled,\nwhich essentially looks like:\n\n\tif (r2 \u0026 0x00008000) {\n\t\tr0 |\u003d 0x80000000;\n\t\twhile (r0 \u0026 0x80000000) {}\n\t}\n\nI don\u0027t know of any way that flag would get cleared unless the driver\nintervenes (and indeed, in the cases I\u0027ve seen the hang, nothing steps\nin to automagically clear it for us). This patch causes the driver to\nclear the flag during the PGRAPH IRQ handler.\n\nSigned-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n" }, { "commit": "1959ca80e1f88b82c1cb7227f437910768ab0c94", "tree": "1bc31d7c764f156dc3c92cb77c7aedcce16d9f62", "parents": [ "a908b96c22883f967e4ddf5aa5b35e3b4a0629a5" ], "author": { "name": "Ben Skeggs", "email": "bskeggs@redhat.com", "time": "Mon Jan 04 15:52:20 2010 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Jan 11 14:41:15 2010 +1000" }, "message": "drm/nouveau: have ttm\u0027s fault handler called directly\n\nThere\u0027s no good reason for us to have our own anymore, this is left over\nfrom an early port to these TTM interfaces.\n\nSigned-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n" }, { "commit": "a908b96c22883f967e4ddf5aa5b35e3b4a0629a5", "tree": "c98c182736efc6eb49bffd55052a0fce46edb585", "parents": [ "c03ec7f91fcf20af177dbc728d518fb462bad42d" ], "author": { "name": "Ben Skeggs", "email": "bskeggs@redhat.com", "time": "Tue Jan 05 09:41:05 2010 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Mon Jan 11 14:41:15 2010 +1000" }, "message": "drm/nv50: restore correct cache1 get/put address on fifoctx load\n\nSigned-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n" } ], "next": "c03ec7f91fcf20af177dbc728d518fb462bad42d" }