aboutsummaryrefslogtreecommitdiff
path: root/drivers/gpu
AgeCommit message (Collapse)Author
2009-09-09PCI/vgaarb: cleanup some warnings + cleanup some comments.Dave Airlie
Fix some warnings reported in linux-next + also cleanup some comment errors noticed by Pekka Paalanen. Signed-off-by: Dave Airlie <airlied@redhat.com> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
2009-09-09PCI/GPU: implement VGA arbitration on LinuxBenjamin Herrenschmidt
Background: Graphic devices are accessed through ranges in I/O or memory space. While most modern devices allow relocation of such ranges, some "Legacy" VGA devices implemented on PCI will typically have the same "hard-decoded" addresses as they did on ISA. For more details see "PCI Bus Binding to IEEE Std 1275-1994 Standard for Boot (Initialization Configuration) Firmware Revision 2.1" Section 7, Legacy Devices. The Resource Access Control (RAC) module inside the X server currently does the task of arbitration when more than one legacy device co-exists on the same machine. But the problem happens when these devices are trying to be accessed by different userspace clients (e.g. two server in parallel). Their address assignments conflict. Therefore an arbitration scheme _outside_ of the X server is needed to control the sharing of these resources. This document introduces the operation of the VGA arbiter implemented for Linux kernel. Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com> Signed-off-by: Dave Airlie <airlied@redhat.com> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
2009-09-08i915: disable interrupts before tearing down GEM stateLinus Torvalds
Reinette Chatre reports a frozen system (with blinking keyboard LEDs) when switching from graphics mode to the text console, or when suspending (which does the same thing). With netconsole, the oops turned out to be BUG: unable to handle kernel NULL pointer dereference at 0000000000000084 IP: [<ffffffffa03ecaab>] i915_driver_irq_handler+0x26b/0xd20 [i915] and it's due to the i915_gem.c code doing drm_irq_uninstall() after having done i915_gem_idle(). And the i915_gem_idle() path will do i915_gem_idle() -> i915_gem_cleanup_ringbuffer() -> i915_gem_cleanup_hws() -> dev_priv->hw_status_page = NULL; but if an i915 interrupt comes in after this stage, it may want to access that hw_status_page, and gets the above NULL pointer dereference. And since the NULL pointer dereference happens from within an interrupt, and with the screen still in graphics mode, the common end result is simply a silently hung machine. Fix it by simply uninstalling the irq handler before idling rather than after. Fixes http://bugzilla.kernel.org/show_bug.cgi?id=13819 Reported-and-tested-by: Reinette Chatre <reinette.chatre@intel.com> Acked-by: Jesse Barnes <jbarnes@virtuousgeek.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-09-08drm/i915: fix mask bits settingZhenyu Wang
eDP is exclusive connector too, and add missing crtc_mask setting for TV. This fixes http://bugzilla.kernel.org/show_bug.cgi?id=14139 Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com> Reported-and-tested-by: Carlos R. Mafra <crmafra2@gmail.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-09-07Merge branch 'drm-fixes' of ↵Linus Torvalds
git://git.kernel.org/pub/scm/linux/kernel/git/airlied/drm-2.6 * 'drm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/airlied/drm-2.6: drm/radeon/kms: add LTE/GTE discard + rv515 two sided stencil register.
2009-09-07drm/radeon/kms: add LTE/GTE discard + rv515 two sided stencil register.Dave Airlie
This adds some rv350+ register for LTE/GTE discard, and enables the rv515 two sided stencil register. It also disables the DEPTHXY_OFFSET register which can be used to workaround the CS checker. Moves rs690 to proper place in rs600 and uses correct table on rs600. Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-09-02drm/i915: Unref old_obj on get_fence_reg() error pathChris Wilson
Remember to release the local reference if we fail to wait on the rendering. (Also whilst in the vicinity add some whitespace so that the phasing of the operations is clearer.) Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Signed-off-by: Eric Anholt <eric@anholt.net>
2009-09-02drm/i915: increase default latency constant (v2 w/comment)Jesse Barnes
Some i915/i945 platforms have a fairly high memory latency in certain situations, so increase our constant a bit to avoid FIFO underruns. The effect should be positive on other platforms as well; we'll have a bit more insurance against a busy memory subsystem due to the extra FIFO entries. Fixes fdo bug #23368. Needed for 2.6.31. Tested-by: Sven Arvidsson <sa@whiz.se> Tested-by: Chris Wilson <chris@chris-wilson.co.uk> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org> Signed-off-by: Eric Anholt <eric@anholt.net>
2009-08-31Merge branch 'for-linus' of ↵Linus Torvalds
git://git.kernel.org/pub/scm/linux/kernel/git/anholt/drm-intel * 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/anholt/drm-intel: drm/i915: Improve CRTDDC mapping by using VBT info drm/i915: Fix CPU-spinning hangs related to fence usage by using an LRU. drm/i915: Set crtc/clone mask in different output devices drm/i915: Always use SDVO_B detect bit for SDVO output detection. drm/i915: Fix typo that broke SVID1 in intel_sdvo_multifunc_encoder() drm/i915: Check if BIOS enabled dual-channel LVDS on 8xx, not only on 9xx drm/i915: Set the multiplier for SDVO on G33 platform
2009-08-29drm/i915: Improve CRTDDC mapping by using VBT infoDavid Müller (ELSOFT AG)
Use VBT information to determine which DDC bus to use for CRTDCC. Fall back to GPIOA if VBT info is not available. Signed-off-by: David Müller <d.mueller@elsoft.ch> Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org> Signed-off-by: Eric Anholt <eric@anholt.net> Tested on: 855 (David), and 945GM, 965GM, GM45, and G45 (anholt)
2009-08-29drm/i915: Fix CPU-spinning hangs related to fence usage by using an LRU.Eric Anholt
The lack of a proper LRU was partially worked around by taking the fence from the object containing the oldest seqno. But if there are multiple objects inactive, then they don't have seqnos and the first fence reg among them would be chosen. If you were trying to copy data between two mappings, this could result in each page fault stealing the fence from the other argument, and your application hanging. https://bugs.freedesktop.org/show_bug.cgi?id=23566 https://bugs.freedesktop.org/show_bug.cgi?id=23220 https://bugs.freedesktop.org/show_bug.cgi?id=23253 https://bugs.freedesktop.org/show_bug.cgi?id=23366 Cc: Stable Team <stable@kernel.org> Signed-off-by: Eric Anholt <eric@anholt.net> Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org> Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
2009-08-24drm/i915: Set crtc/clone mask in different output devicesMa Ling
Based on Bspec each encoder has different sharing pipe property, i.e. Integrated or SDVO TV both will occupy one pipe exclusively, and sdvo-non-tv and crt are allowed to share one. The patch moves sharing judgment into differnet output functions, and sets the right clone bit. This fixes both HDMI outputs choosing the same pipe. https://bugs.freedesktop.org/show_bug.cgi?id=22247 Signed-off-by: Ma Ling <ling.ma@intel.com> Reviewed-by : Jesse Barnes <jbarnes@virtuousgeek.org> Signed-off-by: Zhao Yakui <yakui.zhao@intel.com> Signed-off-by: Eric Anholt <eric@anholt.net>
2009-08-24drm/i915: Always use SDVO_B detect bit for SDVO output detection.Ma Ling
After the following commit is shipped, the SDVO C detection will depend on the SDVO_C/DP detion bit. commit 13520b051e8888dd3af9bda639d83e7df76613d1 Author: Kristian Høgsberg <krh@redhat.com> Date: Fri Mar 13 15:42:14 2009 -0400 drm/i915: Read the right SDVO register when detecting SVDO/HDMI. According to the spec we should continue to detect the SDVO_B/C based on the SDVO_B detection bit. The new detection bit on G4X platform is for the HDMI_C detection rather than SDVO_C detection. https://bugs.freedesktop.org/show_bug.cgi?id=20639 Signed-off-by: Ma Ling <ling.ma@intel.com> Acked-by: Zhao Yakui <yakui.zhao@intel.com> Signed-off-by: Eric Anholt <eric@anholt.net>
2009-08-24drm/i915: Fix typo that broke SVID1 in intel_sdvo_multifunc_encoder()Roel Kluin
Bit SDVO_OUTPUT_SVID0 was tested twice Signed-off-by: Roel Kluin <roel.kluin@gmail.com> Signed-off-by: Eric Anholt <eric@anholt.net>
2009-08-24drm/i915: Check if BIOS enabled dual-channel LVDS on 8xx, not only on 9xxBruno Prémont
Commit 0c2e39525b3b53a97a0202c5f35058147e53977e is not sufficient to get fd.o bug #20115 fixed. In addition intel_find_best_PLL() must not only rely on BIOS settings for i9xx chips but also for i8xx, so drop the IS_I9XX() check. Signed-off-by: Bruno Prémont <bonbons@linux-vserver.org> Signed-off-by: Eric Anholt <eric@anholt.net>
2009-08-24drm/i915: Set the multiplier for SDVO on G33 platformSean Young
http://bugs.freedesktop.org/show_bug.cgi?id=21417 Signed-off-by: Sean Young <sean@mess.org> Reviewed-by: Zhao Yakui <yakui.zhao@intel.com> Acked-by: Zhenyu Wang <zhenyuw@linux.intel.com> Signed-off-by: Eric Anholt <eric@anholt.net>
2009-08-23drm/radeon/kms: Fix radeon_gem_busy_ioctl harder.Michel Dänzer
It was mixing up TTM placement values and flags. Signed-off-by: Michel Dänzer <daenzer@vmware.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-08-21drm/radeon: add GET_PARAM/INFO support for Z pipesAlex Deucher
Needed for occlusion queries on rv530 chips. Signed-off-by: Alex Deucher <alexdeucher@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-08-21drm/radeon/kms: add r100/r200 OQ support.Dave Airlie
This adds the relocation necessary for OQ support on the r100/r200 chipsets. Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-08-21drm: Fix sysfs device confusion.Thomas Hellstrom
The drm sysfs class suspend / resume methods could not distinguish between different device types wich could lead to illegal type casts. Use struct device_type and make sure the class suspend / resume callbacks are aware of those. There is no per device-type suspend / resume. Only new-style PM. Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org> Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-08-21drm/radeon/kms: implement the bo busy ioctl properly.Dave Airlie
The previous patch assumes the ioctl already existed, when it actually didn't. It also didn't return the correct error code. Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-08-19drm/kms: teardown crtc correctly when fb is destroyed.Dave Airlie
If userspace destroys a framebuffer that is in use on a crtc, don't just null it out, tear down the crtc properly so the hw gets turned off. Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-08-19drm/kms/radeon: cleanup combios TV table like DDX.Dave Airlie
The fallback case wasn't getting executed properly if there was no TV table, which my T42 M7 hasn't got. Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-08-19drm/radeon/kms: memset the allocated framebuffer before using it.Dave Airlie
This gets rid of some ugliness, we shuold probably find a way for the GPU to zero this. Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-08-19drm/radeon/kms: although LVDS might be possible on crtc 1 don't do it.Dave Airlie
LVDS always requests RMX_FULL, we need to fix it so that doesn't happen before we can enable LVDS on crtc 1. Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-08-17drm/radeon/kms: implement bo busy check + current domainDave Airlie
This implements the busy ioctl along with a current domain check. returns 0 or -EBUSY puts the current domain no matter what the answer. Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-08-16drm/radeon/kms: cut down indirects in register accesses.Dave Airlie
We really don't want to be doing all these indirects, updating the GPU gart table is something we do often so the less overhead the better. Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-08-16drm/radeon/kms: Fix up vertical blank interrupt support.Michel Dänzer
Fixes 3D apps timing out in the WAIT_VBLANK ioctl. AVIVO bits compile-tested only. Signed-off-by: Michel Dänzer <daenzer@vmware.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-08-16drm/radeon/kms: add rv530 R300_SU_REG_DEST + reloc for ZPASS_ADDRDave Airlie
These are needed for Occulsion Query support. Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-08-15drm/edid: fixup detailed timings like the X server.Dave Airlie
this syncs the versioning check with the code the X server uses. Reported-by: Anssi Hannula <anssi.hannula@iki.fi> Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-08-15drm/radeon/kms: Add specific rs690 authorized register tableJerome Glisse
rs690 is r3xx 3D engine with AVIVO modesetting so we need to allow AVIVO register for vline synchronization. This add a specific table to rs690 to handle that. Thanks to Marc (marvin24) for debugging this and kudos to Andre (taiu1) for spotting the origin of the bugs. Signed-off-by: Jerome Glisse <jglisse@redhat.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-08-09drm/i915: silence vblank warningsFrans Pop
these errors are pretty pointless Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org> Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-08-09drm: silence pointless vblank warning.Paul Rolland
Some applications/hardware combinations are triggering the message "failed to acquire vblank counter" to be issued up to 20 times a second, which makes it both useless and dangerous, as this may hide other important messages. This changes makes it only appear when people are debugging. Signed-off-by: Paul Rolland <rol@as2917.net> Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org> Lost-twice-by: Dave Airlie <airlied@redhat.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-08-09drm: When adding probed modes, preserve duplicate mode typesKeith Packard
The code which takes probed modes and adds them to a connector eliminates duplicate modes by comparing them using drm_mode_equal. That function doesn't consider the type bits, which means that any modes which differ only in the type field will be lost. One of the bits in the mode->type field is the DRM_MODE_TYPE_PREFERRED bit. If the mode with that bit is lost, then higher level code will not know which mode to select, causing a random mode to be used instead. This patch simply merges the two mode type bits together; that seems reasonable to me, but perhaps only a subset of the bits should be used? None of these can be user defined as they all come from looking at just the hardware. Signed-off-by: Keith Packard <keithp@keithp.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-08-07Merge branch 'for-linus' of ↵Linus Torvalds
git://git.kernel.org/pub/scm/linux/kernel/git/anholt/drm-intel * 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/anholt/drm-intel: (22 commits) drm/i915: Fix read outside array bounds in restoring the SWF10 range. drm/i915: Use our own workqueue to avoid wedging the system along with the GPU. drm/i915: Add support for dual-channel LVDS on 8xx. drm/i915: Return disconnected for SDVO DVI when there's no digital EDID. drm/i915: Choose real sdvo output according to result from detection drm/i915: Set preferred mode for integrated TV according to TV format drm/i915: fix 845G FIFO size & burst length drm/i915: fix VGA detect on IGDNG drm/i915: Add eDP support on IGDNG mobile chip drm/i915: enable DisplayPort support on IGDNG drm/i915: Fix channel ending action for DP aux transaction drm/i915: fix issue in display pipe setup on IGDNG drm/i915: disable VGA plane reliably drm/I915: Fix offset to DVO timings in LVDS data drm/i915: hdmi detection according by reading edid drm/i915: correct self-refresh calculation in "everything off" case drm/i915: handle FIFO oversubsription correctly drm/i915: FIFO watermark calculation fixes drm/i915: ignore lvds on AOpen Mini PC MP-915 drm/i915: Allow frame buffers up to 4096x4096 on 915/945 class hardware ...
2009-08-07drm/radeon/kms: setup MC/VRAM the same way for suspend/resumeDave Airlie
we should align the GTT after VRAM no matter what, as we can come back from resume and put in a different place and bad things happen. Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-08-06drm/radeon/kms: Fix caching mode selection for GTT objectJerome Glisse
GTT object can either be cached,uncached or wc just let core ttm pick the best mode according to how the bo driver and GTT memory type was initialized. Signed-off-by: Jerome Glisse <jglisse@redhat.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-08-05drm/i915: Fix read outside array bounds in restoring the SWF10 range.Roel Kluin
dev_priv->saveSWF1 is a 16 element array, but this reads up to index 22, and restored values from the wrong registers. Signed-off-by: Roel Kluin <roel.kluin@gmail.com> Signed-off-by: Eric Anholt <eric@anholt.net>
2009-08-05drm/i915: Use our own workqueue to avoid wedging the system along with the GPU.Eric Anholt
Signed-off-by: Eric Anholt <eric@anholt.net>
2009-08-05drm/radeon: Add support for RS880 chipsAlex Deucher
These are new AMD IGP chips Signed-off-by: Alex Deucher <alexdeucher@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-08-04drm/ttm: Read buffer overflowRoel Kluin
Check whether index is within bounds before grabbing the element. Signed-off-by: Roel Kluin <roel.kluin@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-08-04drm/radeon: Read buffer overflowRoel Kluin
Check whether index is within bounds before grabbing the element. Signed-off-by: Roel Kluin <roel.kluin@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-08-04drm/ttm: Fix a sync object leak.Thomas Hellstrom
If there are multiple simultaneous waiters for the same buffer object, a temporary reference to its sync object may be leaked. Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-08-04drm/radeon/kms: fix memory leak in radeon_driver_load_kmsXiaotian Feng
This patch fixes following kmemleak report: unreferenced object 0xffff88022cb53000 (size 4096): comm "work_for_cpu", pid 97, jiffies 4294672345 backtrace: [<ffffffff810eb222>] create_object+0x19f/0x2a0 [<ffffffff810eb422>] kmemleak_alloc+0x26/0x4c [<ffffffff810e363f>] __kmalloc+0x187/0x1b0 [<ffffffffa005f3db>] kzalloc.clone.0+0x13/0x15 [radeon] [<ffffffffa005f403>] radeon_driver_load_kms+0x26/0xe1 [radeon] [<ffffffffa0017432>] drm_get_dev+0x37f/0x480 [drm] [<ffffffffa007f424>] radeon_pci_probe+0x15/0x269 [radeon] [<ffffffff811f8779>] local_pci_probe+0x17/0x1b [<ffffffff8105ffbb>] do_work_for_cpu+0x18/0x2a [<ffffffff81063c38>] kthread+0x8a/0x92 [<ffffffff81012cba>] child_rip+0xa/0x20 [<ffffffffffffffff>] 0xffffffffffffffff Signed-off-by: Xiaotian Feng <dfeng@redhat.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-08-04drm/radeon/kms: fix nomodeset.Dave Airlie
The ordering was wrong to get the nomodeset parameter to work. Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-08-04drm/ttm: Fix a potential comparison of structs.Thomas Hellstrom
On some architectures the comparison may cause a compilation failure. Original partial fix Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> Signed-off-by: Pekka Paalanen <pq@iki.fi> Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-08-04drm/radeon/kms: fix rv515 VRAM initialisation.Dave Airlie
This got missed in the VRAM init re-workings. Signed-of-by: Dave Airlie <airlied@redhat.com>
2009-08-04drm: Catch stop possible NULL pointer referenceJakob Bornecrantz
This was caught by Weiss. Also added some comments to the fb_changed and mode_changed variables to explain what they do. Signed-off-by: Jakob Bornecrantz <jakob@vmware.com> Tested-by: Thomas White <taw@bitwiz.org.uk> Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-08-04drm: Small logic fix in drm_mode_setcrtcJakob Bornecrantz
Match the logic to the comments in the debug message Signed-off-by: Jakob Bornecrantz <jakob@vmware.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-07-30drm/i915: Add support for dual-channel LVDS on 8xx.ling.ma@intel.com
This corresponds to a fix to UMS back in 2007. Fixes fd.o bug #20115. Signed-off-by: Ma Ling <ling.ma@intel.com> Signed-off-by: Eric Anholt <eric@anholt.net>