[PATCH] drm/ttm: callback move_notify any time bo placement change v2

On 11/18/2011 06:32 PM, j.glisse at gmail.com wrote: > From: Jerome Glisse > > Previously we were calling back move_notify in error path when the > bo is returned to it's original position or when destroy the bo. > When destroying the bo set the new mem placement as NULL when calling > back in the

[PATCH] drm/ttm: pass buffer object for bind/unbind callback

On 11/18/2011 06:26 PM, Ben Skeggs wrote: > On Fri, 2011-11-18 at 15:30 +0100, Thomas Hellstrom wrote: > >> On 11/18/2011 02:15 PM, Ben Skeggs wrote: >> >>> On Fri, 2011-11-18 at 08:57 +0100, Thomas Hellstrom wrote: >>> >>> Jerome, I don't like this change for the

[Intel-gfx] [PATCH] drm/i915: By default, enable RC6 on IVB and SNB when reasonable

On 11/18/2011 10:41 PM, Keith Packard wrote: > RC6 should always work on IVB, and should work on SNB whenever IO > remapping is disabled. Make the default value for the parameter turn > it on in these cases. Setting the value to either 0 or 1 will force > the specified behavior. > >

[PATCH] drm/ttm: pass buffer object for bind/unbind callback

On Fri, 2011-11-18 at 08:57 +0100, Thomas Hellstrom wrote: > Jerome, > > I don't like this change for the following reasons -snip- > > > > One can take advantage of move notify callback but, there are > > corner case where bind/unbind might be call without move notify > > being call (in error

DRM KMS Modesetting

On Fri, Nov 18, 2011 at 10:41 PM, Jesse Barnes wrote: > On Fri, 18 Nov 2011 22:35:13 +0100 > David Herrmann wrote: > >> 2011/11/15 Kristian H?gsberg : >> > 2011/11/15 David Herrmann : >> >> 2011/11/15 Kristian H?gsberg : >> >>> On Mon, Nov 14, 2011 at 5:54 PM, Jesse Barnes > >>>

[PULL] drm-intel-fixes

: not available URL: <http://lists.freedesktop.org/archives/dri-devel/attachments/20111118/4e3efcd7/attachment-0001.pgp>

[PATCH] drm/i915: By default, enable RC6 on IVB and SNB when reasonable

RC6 should always work on IVB, and should work on SNB whenever IO remapping is disabled. Make the default value for the parameter turn it on in these cases. Setting the value to either 0 or 1 will force the specified behavior. Signed-off-by: Keith Packard --- drivers/gpu/drm/i915/i915_drv.c

DRM KMS Modesetting

2011/11/15 Kristian H?gsberg : > 2011/11/15 David Herrmann : >> 2011/11/15 Kristian H?gsberg : >>> On Mon, Nov 14, 2011 at 5:54 PM, Jesse Barnes >>> wrote: On Mon, 14 Nov 2011 21:47:09 +0100 David Herrmann wrote: > > I had to modify the resolution the test was searching for >

[Bug 43073] Trine not working on Radeon HD6520G

https://bugs.freedesktop.org/show_bug.cgi?id=43073 Sven Arvidsson changed: What|Removed |Added CC||sa at whiz.se --- Comment #3 from Sven

ttm: merge ttm_backend & ttm_tt, introduce ttm dma allocator V7

On Fri, Nov 18, 2011 at 08:04:47PM -0500, j.glisse at gmail.com wrote: > Important fix to patch 14, fix accounting of ghost bo. When creating a > ghost bo we don't account it, so set its acc_size to 0 so that when > ghost is release we don't overfree. > > I wonder how i didn't run into this

[PATCH 14/14] drm/ttm: simplify memory accounting for ttm user v2

From: Jerome Glisse Provide helper function to compute the kernel memory size needed for each buffer object. Move all the accounting inside ttm, simplifying driver and avoiding code duplication accross them. v2 fix accounting of ghost object, one would have thought that i

[PATCH 13/14] drm/ttm: isolate dma data from ttm_tt V4

From: Jerome Glisse Move dma data to a superset ttm_dma_tt structure which herit from ttm_tt. This allow driver that don't use dma functionalities to not have to waste memory for it. V2 Rebase on top of no memory account changes (where/when is my delorean when i need it

[PATCH 12/14] drm/nouveau: enable the ttm dma pool when swiotlb is active V3

From: Konrad Rzeszutek Wilk If the card is capable of more than 32-bit, then use the default TTM page pool code which allocates from anywhere in the memory. Note: If the 'ttm.no_dma' parameter is set, the override is ignored and the default TTM pool is used. V2 use

[PATCH 11/14] drm/radeon/kms: enable the ttm dma pool if swiotlb is on V4

From: Konrad Rzeszutek Wilk With the exception that we do not handle the AGP case. We only deal with PCIe cards such as ATI ES1000 or HD3200 that have been detected to only do DMA up to 32-bits. V2 force dma32 if we fail to set bigger dma mask V3 Rebase on top of no

[PATCH 10/14] drm/ttm: provide dma aware ttm page pool code V9

From: Konrad Rzeszutek Wilk In TTM world the pages for the graphic drivers are kept in three different pools: write combined, uncached, and cached (write-back). When the pages are used by the graphic driver the graphic adapter via its built in MMU (or AGP) programs these

[PATCH 09/14] drm/ttm: introduce callback for ttm_tt populate & unpopulate V4

From: Jerome Glisse Move the page allocation and freeing to driver callback and provide ttm code helper function for those. Most intrusive change, is the fact that we now only fully populate an object this simplify some of code designed around the page fault design. V2

[PATCH 08/14] drm/ttm: merge ttm_backend and ttm_tt V5

From: Jerome Glisse ttm_backend will only exist with a ttm_tt, and ttm_tt will only be of interest when bound to a backend. Merge them to avoid code and data duplication. V2 Rebase on top of memory accounting overhaul V3 Rebase on top of more memory accounting changes V4

[PATCH 07/14] drm/ttm: page allocation use page array instead of list

From: Jerome Glisse Use the ttm_tt pages array for pages allocations, move the list unwinding into the page allocation functions. Signed-off-by: Jerome Glisse --- drivers/gpu/drm/ttm/ttm_page_alloc.c | 85 +- drivers/gpu/drm/ttm/ttm_tt.c

[PATCH 06/14] drm/ttm: test for dma_address array allocation failure

From: Jerome Glisse Signed-off-by: Jerome Glisse Reviewed-by: Konrad Rzeszutek Wilk Reviewed-by: Thomas Hellstrom --- drivers/gpu/drm/ttm/ttm_tt.c |2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/drivers/gpu/drm/ttm/ttm_tt.c

[PATCH 05/14] drm/ttm: use ttm put pages function to properly restore cache attribute

From: Jerome Glisse On failure we need to make sure the page we free has wb cache attribute. Do this pas call the proper ttm page helper function. Signed-off-by: Jerome Glisse Reviewed-by: Konrad Rzeszutek Wilk Reviewed-by: Thomas Hellstrom ---

[PATCH 04/14] drm/ttm: remove unused backend flags field

From: Jerome Glisse This field is not use by any of the driver just drop it. Signed-off-by: Jerome Glisse Reviewed-by: Konrad Rzeszutek Wilk Reviewed-by: Thomas Hellstrom --- drivers/gpu/drm/radeon/radeon_ttm.c |1 - include/drm/ttm/ttm_bo_driver.h |2 -- 2

[PATCH 03/14] drm/ttm: remove split btw highmen and lowmem page

From: Jerome Glisse Split btw highmem and lowmem page was rendered useless by the pool code. Remove it. Note further cleanup would change the ttm page allocation helper to actualy take an array instead of relying on list this could drasticly reduce the number of function call

[PATCH 02/14] drm/ttm: remove userspace backed ttm object support

From: Jerome Glisse This was never use in none of the driver, properly using userspace page for bo would need more code (vma interaction mostly). Removing this dead code in preparation of ttm_tt & backend merge. Signed-off-by: Jerome Glisse Reviewed-by: Konrad Rzeszutek

[PATCH 01/14] swiotlb: Expose swiotlb_nr_tlb function to modules

From: Konrad Rzeszutek Wilk As a mechanism to detect whether SWIOTLB is enabled or not. We also fix the spelling - it was swioltb instead of swiotlb. CC: FUJITA Tomonori [v1: Ripped out swiotlb_enabled] Signed-off-by: Konrad Rzeszutek Wilk ---

ttm: merge ttm_backend & ttm_tt, introduce ttm dma allocator V7

Important fix to patch 14, fix accounting of ghost bo. When creating a ghost bo we don't account it, so set its acc_size to 0 so that when ghost is release we don't overfree. I wonder how i didn't run into this before. Patch are also at http://people.freedesktop.org/~glisse/ttmdma/ Cheers,

[Bug 43073] Trine not working on Radeon HD6520G

https://bugs.freedesktop.org/show_bug.cgi?id=43073 --- Comment #2 from Sandeep 2011-11-18 11:27:49 PST --- Created attachment 53674 --> https://bugs.freedesktop.org/attachment.cgi?id=53674 Output from glxinfo -- Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email ---

[Bug 43073] Trine not working on Radeon HD6520G

https://bugs.freedesktop.org/show_bug.cgi?id=43073 --- Comment #1 from Sandeep 2011-11-18 11:26:57 PST --- Created attachment 53673 --> https://bugs.freedesktop.org/attachment.cgi?id=53673 trine log -- Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email --- You are

[Bug 43073] New: Trine not working on Radeon HD6520G

https://bugs.freedesktop.org/show_bug.cgi?id=43073 Bug #: 43073 Summary: Trine not working on Radeon HD6520G Classification: Unclassified Product: DRI Version: XOrg CVS Platform: x86-64 (AMD64) OS/Version: Linux (All)

[PATCH] drm/ttm: callback move_notify any time bo placement change v3

From: Jerome Glisse Previously we were calling back move_notify in error path when the bo is returned to it's original position or when destroy the bo. When destroying the bo set the new mem placement as NULL when calling back in the driver. Updating nouveau to deal with

[PATCH] drm/ttm: pass buffer object for bind/unbind callback

On Fri, Nov 18, 2011 at 6:25 PM, Jerome Glisse wrote: > On Fri, Nov 18, 2011 at 06:14:02PM -0500, Jerome Glisse wrote: >> On Fri, Nov 18, 2011 at 11:48:58PM +0100, Thomas Hellstrom wrote: >> > On 11/18/2011 06:26 PM, Ben Skeggs wrote: >> > >On Fri, 2011-11-18 at 15:30 +0100, Thomas Hellstrom

[PATCH] drm/ttm: pass buffer object for bind/unbind callback

On Fri, Nov 18, 2011 at 06:14:02PM -0500, Jerome Glisse wrote: > On Fri, Nov 18, 2011 at 11:48:58PM +0100, Thomas Hellstrom wrote: > > On 11/18/2011 06:26 PM, Ben Skeggs wrote: > > >On Fri, 2011-11-18 at 15:30 +0100, Thomas Hellstrom wrote: > > >>On 11/18/2011 02:15 PM, Ben Skeggs wrote: > > >>>On

[PATCH] drm/ttm: pass buffer object for bind/unbind callback

On Fri, Nov 18, 2011 at 11:48:58PM +0100, Thomas Hellstrom wrote: > On 11/18/2011 06:26 PM, Ben Skeggs wrote: > >On Fri, 2011-11-18 at 15:30 +0100, Thomas Hellstrom wrote: > >>On 11/18/2011 02:15 PM, Ben Skeggs wrote: > >>>On Fri, 2011-11-18 at 08:57 +0100, Thomas Hellstrom wrote: > >>> >

[PATCH 3/3] drm/i915: hot removal notification to HDMI audio driver

Update: Just tested DP and its working! However, it's found that hot plug under X won't work... The previous hot plug tests are done in KMS console which are all fine: on re-inserting the monitor, ->mode_set will be called and HDMI/DP audio will be re-enabled and ELD be transfered. However when

[PATCH] drm/ttm: pass buffer object for bind/unbind callback

On 11/18/2011 03:56 PM, Jerome Glisse wrote: > On Fri, Nov 18, 2011 at 03:30:03PM +0100, Thomas Hellstrom wrote: > >> On 11/18/2011 02:15 PM, Ben Skeggs wrote: >> >>> On Fri, 2011-11-18 at 08:57 +0100, Thomas Hellstrom wrote: >>> Jerome, I don't like this change

[Bug 43000] huge performance regression in ut2004 since 7.11

https://bugs.freedesktop.org/show_bug.cgi?id=43000 --- Comment #7 from almos 2011-11-18 07:51:26 PST --- (In reply to comment #6) > (In reply to comment #5) > > 7.12-dev git-08b288b: nvidia logo 60fps (capped at refresh rate??) > > If so, the environment variable vblank_mode=0 should disable

[Bug 40935] radeon lockup on resume

https://bugs.freedesktop.org/show_bug.cgi?id=40935 --- Comment #5 from Michal Suchanek 2011-11-18 07:36:11 PST --- hmm, I it looks like the driver resets the card quite a few times over some hours until it locks up permanently (or at least long enough for the next reset to fail) or the driver

[PATCH] drm/ttm: pass buffer object for bind/unbind callback

On 11/18/2011 02:15 PM, Ben Skeggs wrote: > On Fri, 2011-11-18 at 08:57 +0100, Thomas Hellstrom wrote: > >> Jerome, >> >> I don't like this change for the following reasons >> > -snip- > > >>> One can take advantage of move notify callback but, there are >>> corner case where

[Bug 43000] huge performance regression in ut2004 since 7.11

https://bugs.freedesktop.org/show_bug.cgi?id=43000 --- Comment #6 from Michel D?nzer 2011-11-18 07:25:27 PST --- (In reply to comment #5) > 7.12-dev git-08b288b: nvidia logo 60fps (capped at refresh rate??) If so, the environment variable vblank_mode=0 should disable it. -- Configure

[Bug 43000] huge performance regression in ut2004 since 7.11

https://bugs.freedesktop.org/show_bug.cgi?id=43000 --- Comment #5 from almos 2011-11-18 07:00:34 PST --- Now I compiled a 32bit r600g from mesa git on a 32 bit machine. The ad hoc benchmark results now with swapbufferwait disabled: 7.11: nvidia logo 200-300fps, menu 70-200fps, ons-torlan looking

[PATCH 2/2] drm: add an fb creation ioctl that takes a pixel format v5

s > + * offeset[1]. Note that offset[0] will generally > + * be 0. > + */ > + __u32 handles[4]; > + __u32 pitches[4]; /* pitch for each plane */ > + __u32 offsets[4]; /* offset of each plane */ > +}; Regards, - Seung-Woo Kim -- next part -- An HTML attachment was scrubbed... URL: <http://lists.freedesktop.org/archives/dri-devel/attachments/2018/377b831f/attachment.html>

DRM KMS Modesetting

nnot tell what fixed it as I recompiled with other > configure options. Previously I used --disable-glx-tls and > --enable-gallium-drivers=i915. Now I use --enable-glx-tls and no i915 > in gallium but only --with-dri-drivers=i915. > My laptop takes about 30min to compile mesa so I haven't tracked it down, yet. Ah it's probably gallium vs classic then. We don't support the gallium driver at all, so I'm not sure what state it's in. -- Jesse Barnes, Intel Open Source Technology Center -- next part -- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 836 bytes Desc: not available URL: <http://lists.freedesktop.org/archives/dri-devel/attachments/2018/f588290f/attachment.pgp>

[Bug 43000] huge performance regression in ut2004 since 7.11

https://bugs.freedesktop.org/show_bug.cgi?id=43000 --- Comment #4 from almos 2011-11-18 05:25:58 PST --- OK, now I found out the reason: I still haven't got used to my new 64bit system. ut2004 is 32 bit, and when I set LIBGL_DRIVERS_PATH libGL.so reverts to indirect rendering. Ouch. What's

RFC: Radeon multi ring support branch

On 17.11.2011 17:58, Jerome Glisse wrote: > 2011/11/17 Christian K?nig: >> On 16.11.2011 01:24, Jerome Glisse wrote: >>> Well as we don't specify on which value semaphore should wait on, i am >>> prety sure the first ring to increment the semaphore will unblock all >>> waiter. So if you have ring1

[PATCH] drm/ttm: callback move_notify any time bo placement change v2

From: Jerome Glisse Previously we were calling back move_notify in error path when the bo is returned to it's original position or when destroy the bo. When destroying the bo set the new mem placement as NULL when calling back in the driver. Updating nouveau to deal with

[Intel-gfx] [PATCH] drm/i915: Hook up Ivybridge eDP

On 11/17/11 9:13 PM, Keith Packard wrote: > On Thu, 17 Nov 2011 17:45:40 -0500, Adam Jackson wrote: > >> Your silicon people worry me. > > In any case, the changes are mostly to move bits around so that there > are two bits for pipe select Yeah, the patch looks like it could be correct, just

[PATCH] drm/ttm: callback move_notify any time bo placement change

From: Jerome Glisse Previously we were calling back move_notify in error path when the bo is returned to it's original position or when destroy the bo. When destroying the bo set the new mem placement as NULL when calling back in the driver. Updating nouveau to deal with

RFC: Radeon multi ring support branch

On Fri, Nov 18, 2011 at 07:44:19AM -0500, Alex Deucher wrote: > 2011/11/17 Alex Deucher : > > 2011/11/17 Christian K?nig : > >> On 16.11.2011 01:24, Jerome Glisse wrote: > >>> > >>> Well as we don't specify on which value semaphore should wait on, i am > >>> prety sure the first ring to increment

RFC: Radeon multi ring support branch

On Fri, Nov 18, 2011 at 09:21:50AM -0500, Jerome Glisse wrote: > 2011/11/18 Christian K?nig : > > On 17.11.2011 17:58, Jerome Glisse wrote: > >> > >> 2011/11/17 Christian K?nig: > >>> > >>> On 16.11.2011 01:24, Jerome Glisse wrote: > > Well as we don't specify on which value semaphore

[PATCH] drm/ttm: pass buffer object for bind/unbind callback

On Fri, Nov 18, 2011 at 04:06:05PM +0100, Thomas Hellstrom wrote: > On 11/18/2011 03:56 PM, Jerome Glisse wrote: > >On Fri, Nov 18, 2011 at 03:30:03PM +0100, Thomas Hellstrom wrote: > >>On 11/18/2011 02:15 PM, Ben Skeggs wrote: > >>>On Fri, 2011-11-18 at 08:57 +0100, Thomas Hellstrom wrote: >

[PATCH] drm/ttm: pass buffer object for bind/unbind callback

On Fri, Nov 18, 2011 at 03:30:03PM +0100, Thomas Hellstrom wrote: > On 11/18/2011 02:15 PM, Ben Skeggs wrote: > >On Fri, 2011-11-18 at 08:57 +0100, Thomas Hellstrom wrote: > >>Jerome, > >> > >>I don't like this change for the following reasons > >-snip- > > > >>>One can take advantage of move

[PATCH 3/3] drm/i915: hot removal notification to HDMI audio driver

dn't have audio routed to it by default). -- keith.packard at intel.com -- next part -- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 827 bytes Desc: not available URL: <http://lists.freedesktop.org/archives/dri-devel/at

RFC: Radeon multi ring support branch

2011/11/18 Christian K?nig : > On 17.11.2011 17:58, Jerome Glisse wrote: >> >> 2011/11/17 Christian K?nig: >>> >>> On 16.11.2011 01:24, Jerome Glisse wrote: Well as we don't specify on which value semaphore should wait on, i am prety sure the first ring to increment the semaphore

[PATCH] drm/ttm: pass buffer object for bind/unbind callback

Jerome, I don't like this change for the following reasons 1) This is really a layer violation. It's like passing a state tracker object down to the pipe driver i Gallium, so that eventually the winsys can access it. 2) TTM, as you say, doesn't really care about GPU virtual maps. It cares

RFC: Radeon multi ring support branch

2011/11/17 Alex Deucher : > 2011/11/17 Christian K?nig : >> On 16.11.2011 01:24, Jerome Glisse wrote: >>> >>> Well as we don't specify on which value semaphore should wait on, i am >>> prety sure the first ring to increment the semaphore will unblock all >>> waiter. So if you have ring1 that want

[Bug 43016] DVI is wrongly detected as Display Port

https://bugs.freedesktop.org/show_bug.cgi?id=43016 --- Comment #6 from samit vats 2011-11-17 23:26:03 PST --- (In reply to comment #3) > Please attach your dmesg output and a copy of your video bios. To get a copy > of the vbios: > > (as root) > (use lspci to get the bus id) > cd

[Bug 43016] DVI is wrongly detected as Display Port

https://bugs.freedesktop.org/show_bug.cgi?id=43016 --- Comment #4 from samit vats 2011-11-17 23:24:20 PST --- Created attachment 53643 --> https://bugs.freedesktop.org/attachment.cgi?id=53643 dmesg -- Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email --- You are

[Bug 43018] (NI Turks Pro) : UT2004-demo crashes after few minutes of launch

https://bugs.freedesktop.org/show_bug.cgi?id=43018 --- Comment #4 from samit vats 2011-11-17 22:56:38 PST --- Created attachment 53642 --> https://bugs.freedesktop.org/attachment.cgi?id=53642 backtrace -- Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email --- You

[Bug 43018] (NI Turks Pro) : UT2004-demo crashes after few minutes of launch

https://bugs.freedesktop.org/show_bug.cgi?id=43018 --- Comment #3 from samit vats 2011-11-17 22:56:04 PST --- (In reply to comment #2) > Does it still happen with current Git? There have been some regression fixes > since commit f1a677c. > > If yes, can you bisect, or at least get a backtrace

Re: [PATCH] drm/ttm: pass buffer object for bind/unbind callback

Jerome, I don't like this change for the following reasons 1) This is really a layer violation. It's like passing a state tracker object down to the pipe driver i Gallium, so that eventually the winsys can access it. 2) TTM, as you say, doesn't really care about GPU virtual maps. It cares

Re: RFC: Radeon multi ring support branch

2011/11/17 Alex Deucher alexdeuc...@gmail.com: 2011/11/17 Christian König deathsim...@vodafone.de: On 16.11.2011 01:24, Jerome Glisse wrote: Well as we don't specify on which value semaphore should wait on, i am prety sure the first ring to increment the semaphore will unblock all waiter. So

[Bug 43000] huge performance regression in ut2004 since 7.11

https://bugs.freedesktop.org/show_bug.cgi?id=43000 --- Comment #4 from almos aaalmo...@gmail.com 2011-11-18 05:25:58 PST --- OK, now I found out the reason: I still haven't got used to my new 64bit system. ut2004 is 32 bit, and when I set LIBGL_DRIVERS_PATH libGL.so reverts to indirect rendering.

Re: RFC: Radeon multi ring support branch

2011/11/18 Christian König deathsim...@vodafone.de: On 17.11.2011 17:58, Jerome Glisse wrote: 2011/11/17 Christian Königdeathsim...@vodafone.de: On 16.11.2011 01:24, Jerome Glisse wrote: Well as we don't specify on which value semaphore should wait on, i am prety sure the first ring to

Re: [PATCH] drm/ttm: pass buffer object for bind/unbind callback

On 11/18/2011 02:15 PM, Ben Skeggs wrote: On Fri, 2011-11-18 at 08:57 +0100, Thomas Hellstrom wrote: Jerome, I don't like this change for the following reasons -snip- One can take advantage of move notify callback but, there are corner case where bind/unbind might be call

Re: [PATCH] drm/ttm: pass buffer object for bind/unbind callback

On Fri, Nov 18, 2011 at 03:30:03PM +0100, Thomas Hellstrom wrote: On 11/18/2011 02:15 PM, Ben Skeggs wrote: On Fri, 2011-11-18 at 08:57 +0100, Thomas Hellstrom wrote: Jerome, I don't like this change for the following reasons -snip- One can take advantage of move notify callback but,

[Bug 43000] huge performance regression in ut2004 since 7.11

https://bugs.freedesktop.org/show_bug.cgi?id=43000 --- Comment #5 from almos aaalmo...@gmail.com 2011-11-18 07:00:34 PST --- Now I compiled a 32bit r600g from mesa git on a 32 bit machine. The ad hoc benchmark results now with swapbufferwait disabled: 7.11: nvidia logo 200-300fps, menu 70-200fps,

Re: [PATCH] drm/ttm: pass buffer object for bind/unbind callback

On Fri, Nov 18, 2011 at 04:06:05PM +0100, Thomas Hellstrom wrote: On 11/18/2011 03:56 PM, Jerome Glisse wrote: On Fri, Nov 18, 2011 at 03:30:03PM +0100, Thomas Hellstrom wrote: On 11/18/2011 02:15 PM, Ben Skeggs wrote: On Fri, 2011-11-18 at 08:57 +0100, Thomas Hellstrom wrote: Jerome, I

[Bug 43000] huge performance regression in ut2004 since 7.11

https://bugs.freedesktop.org/show_bug.cgi?id=43000 --- Comment #6 from Michel Dänzer mic...@daenzer.net 2011-11-18 07:25:27 PST --- (In reply to comment #5) 7.12-dev git-08b288b: nvidia logo 60fps (capped at refresh rate??) If so, the environment variable vblank_mode=0 should disable it. --

[Bug 40935] radeon lockup on resume

https://bugs.freedesktop.org/show_bug.cgi?id=40935 --- Comment #5 from Michal Suchanek hramr...@gmail.com 2011-11-18 07:36:11 PST --- hmm, I it looks like the driver resets the card quite a few times over some hours until it locks up permanently (or at least long enough for the next reset to

Re: RFC: Radeon multi ring support branch

On Fri, Nov 18, 2011 at 09:21:50AM -0500, Jerome Glisse wrote: 2011/11/18 Christian König deathsim...@vodafone.de: On 17.11.2011 17:58, Jerome Glisse wrote: 2011/11/17 Christian Königdeathsim...@vodafone.de: On 16.11.2011 01:24, Jerome Glisse wrote: Well as we don't specify on which

[Bug 43000] huge performance regression in ut2004 since 7.11

https://bugs.freedesktop.org/show_bug.cgi?id=43000 --- Comment #7 from almos aaalmo...@gmail.com 2011-11-18 07:51:26 PST --- (In reply to comment #6) (In reply to comment #5) 7.12-dev git-08b288b: nvidia logo 60fps (capped at refresh rate??) If so, the environment variable vblank_mode=0

Re: [Intel-gfx] [PATCH] drm/i915: Hook up Ivybridge eDP

On 11/17/11 9:13 PM, Keith Packard wrote: On Thu, 17 Nov 2011 17:45:40 -0500, Adam Jacksona...@redhat.com wrote: Your silicon people worry me. In any case, the changes are mostly to move bits around so that there are two bits for pipe select Yeah, the patch looks like it could be correct,

Re: [PATCH] drm/ttm: pass buffer object for bind/unbind callback

On Fri, 2011-11-18 at 15:30 +0100, Thomas Hellstrom wrote: On 11/18/2011 02:15 PM, Ben Skeggs wrote: On Fri, 2011-11-18 at 08:57 +0100, Thomas Hellstrom wrote: Jerome, I don't like this change for the following reasons -snip- One can take advantage of move notify

[PATCH] drm/ttm: callback move_notify any time bo placement change v2

From: Jerome Glisse jgli...@redhat.com Previously we were calling back move_notify in error path when the bo is returned to it's original position or when destroy the bo. When destroying the bo set the new mem placement as NULL when calling back in the driver. Updating nouveau to deal with NULL

Re: [PATCH 3/3] drm/i915: hot removal notification to HDMI audio driver

On Fri, 18 Nov 2011 17:37:40 +0800, Wu Fengguang fengguang...@intel.com wrote: However when in X, -mode_set won't be called at all. Only -get_modes and -detect are called... The desktop software will call mode_set when it configures the monitor. Otherwise, it's not being used (and so

[Bug 43073] New: Trine not working on Radeon HD6520G

https://bugs.freedesktop.org/show_bug.cgi?id=43073 Bug #: 43073 Summary: Trine not working on Radeon HD6520G Classification: Unclassified Product: DRI Version: XOrg CVS Platform: x86-64 (AMD64) OS/Version: Linux (All)

[Bug 43073] Trine not working on Radeon HD6520G

https://bugs.freedesktop.org/show_bug.cgi?id=43073 --- Comment #1 from Sandeep sandy.8...@gmail.com 2011-11-18 11:26:57 PST --- Created attachment 53673 -- https://bugs.freedesktop.org/attachment.cgi?id=53673 trine log -- Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email

[Bug 43073] Trine not working on Radeon HD6520G

https://bugs.freedesktop.org/show_bug.cgi?id=43073 --- Comment #2 from Sandeep sandy.8...@gmail.com 2011-11-18 11:27:49 PST --- Created attachment 53674 -- https://bugs.freedesktop.org/attachment.cgi?id=53674 Output from glxinfo -- Configure bugmail:

[Bug 43073] Trine not working on Radeon HD6520G

https://bugs.freedesktop.org/show_bug.cgi?id=43073 Sven Arvidsson s...@whiz.se changed: What|Removed |Added CC||s...@whiz.se --- Comment

Re: DRM KMS Modesetting

2011/11/15 Kristian Høgsberg k...@bitplanet.net: 2011/11/15 David Herrmann dh.herrm...@googlemail.com: 2011/11/15 Kristian Høgsberg k...@bitplanet.net: On Mon, Nov 14, 2011 at 5:54 PM, Jesse Barnes jbar...@virtuousgeek.org wrote: On Mon, 14 Nov 2011 21:47:09 +0100 David Herrmann

Re: DRM KMS Modesetting

On Fri, 18 Nov 2011 22:35:13 +0100 David Herrmann dh.herrm...@googlemail.com wrote: 2011/11/15 Kristian Høgsberg k...@bitplanet.net: 2011/11/15 David Herrmann dh.herrm...@googlemail.com: 2011/11/15 Kristian Høgsberg k...@bitplanet.net: On Mon, Nov 14, 2011 at 5:54 PM, Jesse Barnes

Re: DRM KMS Modesetting

On Fri, Nov 18, 2011 at 10:41 PM, Jesse Barnes jbar...@virtuousgeek.org wrote: On Fri, 18 Nov 2011 22:35:13 +0100 David Herrmann dh.herrm...@googlemail.com wrote: 2011/11/15 Kristian Høgsberg k...@bitplanet.net: 2011/11/15 David Herrmann dh.herrm...@googlemail.com: 2011/11/15 Kristian

Re: [PATCH] drm/ttm: pass buffer object for bind/unbind callback

On 11/18/2011 06:26 PM, Ben Skeggs wrote: On Fri, 2011-11-18 at 15:30 +0100, Thomas Hellstrom wrote: On 11/18/2011 02:15 PM, Ben Skeggs wrote: On Fri, 2011-11-18 at 08:57 +0100, Thomas Hellstrom wrote: Jerome, I don't like this change for the following reasons

Re: [PATCH] drm/ttm: callback move_notify any time bo placement change v2

On 11/18/2011 06:32 PM, j.gli...@gmail.com wrote: From: Jerome Glissejgli...@redhat.com Previously we were calling back move_notify in error path when the bo is returned to it's original position or when destroy the bo. When destroying the bo set the new mem placement as NULL when calling back

Re: [PATCH] drm/ttm: pass buffer object for bind/unbind callback

On Fri, Nov 18, 2011 at 11:48:58PM +0100, Thomas Hellstrom wrote: On 11/18/2011 06:26 PM, Ben Skeggs wrote: On Fri, 2011-11-18 at 15:30 +0100, Thomas Hellstrom wrote: On 11/18/2011 02:15 PM, Ben Skeggs wrote: On Fri, 2011-11-18 at 08:57 +0100, Thomas Hellstrom wrote: Jerome, I don't like

Re: [PATCH] drm/ttm: pass buffer object for bind/unbind callback

On Fri, Nov 18, 2011 at 06:14:02PM -0500, Jerome Glisse wrote: On Fri, Nov 18, 2011 at 11:48:58PM +0100, Thomas Hellstrom wrote: On 11/18/2011 06:26 PM, Ben Skeggs wrote: On Fri, 2011-11-18 at 15:30 +0100, Thomas Hellstrom wrote: On 11/18/2011 02:15 PM, Ben Skeggs wrote: On Fri,

Re: [PATCH] drm/ttm: pass buffer object for bind/unbind callback

On Fri, Nov 18, 2011 at 6:25 PM, Jerome Glisse j.gli...@gmail.com wrote: On Fri, Nov 18, 2011 at 06:14:02PM -0500, Jerome Glisse wrote: On Fri, Nov 18, 2011 at 11:48:58PM +0100, Thomas Hellstrom wrote: On 11/18/2011 06:26 PM, Ben Skeggs wrote: On Fri, 2011-11-18 at 15:30 +0100, Thomas

Re: [PATCH] drm/ttm: pass buffer object for bind/unbind callback

On Fri, 2011-11-18 at 23:48 +0100, Thomas Hellstrom wrote: On 11/18/2011 06:26 PM, Ben Skeggs wrote: On Fri, 2011-11-18 at 15:30 +0100, Thomas Hellstrom wrote: On 11/18/2011 02:15 PM, Ben Skeggs wrote: On Fri, 2011-11-18 at 08:57 +0100, Thomas Hellstrom wrote:

ttm: merge ttm_backend ttm_tt, introduce ttm dma allocator V7

Important fix to patch 14, fix accounting of ghost bo. When creating a ghost bo we don't account it, so set its acc_size to 0 so that when ghost is release we don't overfree. I wonder how i didn't run into this before. Patch are also at http://people.freedesktop.org/~glisse/ttmdma/ Cheers,

[PATCH 01/14] swiotlb: Expose swiotlb_nr_tlb function to modules

From: Konrad Rzeszutek Wilk konrad.w...@oracle.com As a mechanism to detect whether SWIOTLB is enabled or not. We also fix the spelling - it was swioltb instead of swiotlb. CC: FUJITA Tomonori fujita.tomon...@lab.ntt.co.jp [v1: Ripped out swiotlb_enabled] Signed-off-by: Konrad Rzeszutek Wilk

[PATCH 02/14] drm/ttm: remove userspace backed ttm object support

From: Jerome Glisse jgli...@redhat.com This was never use in none of the driver, properly using userspace page for bo would need more code (vma interaction mostly). Removing this dead code in preparation of ttm_tt backend merge. Signed-off-by: Jerome Glisse jgli...@redhat.com Reviewed-by:

[PATCH 03/14] drm/ttm: remove split btw highmen and lowmem page

From: Jerome Glisse jgli...@redhat.com Split btw highmem and lowmem page was rendered useless by the pool code. Remove it. Note further cleanup would change the ttm page allocation helper to actualy take an array instead of relying on list this could drasticly reduce the number of function call

[PATCH 04/14] drm/ttm: remove unused backend flags field

From: Jerome Glisse jgli...@redhat.com This field is not use by any of the driver just drop it. Signed-off-by: Jerome Glisse jgli...@redhat.com Reviewed-by: Konrad Rzeszutek Wilk konrad.w...@oracle.com Reviewed-by: Thomas Hellstrom thellst...@vmware.com --- drivers/gpu/drm/radeon/radeon_ttm.c |

[PATCH 05/14] drm/ttm: use ttm put pages function to properly restore cache attribute

From: Jerome Glisse jgli...@redhat.com On failure we need to make sure the page we free has wb cache attribute. Do this pas call the proper ttm page helper function. Signed-off-by: Jerome Glisse jgli...@redhat.com Reviewed-by: Konrad Rzeszutek Wilk konrad.w...@oracle.com Reviewed-by: Thomas

[PATCH 06/14] drm/ttm: test for dma_address array allocation failure

From: Jerome Glisse jgli...@redhat.com Signed-off-by: Jerome Glisse jgli...@redhat.com Reviewed-by: Konrad Rzeszutek Wilk konrad.w...@oracle.com Reviewed-by: Thomas Hellstrom thellst...@vmware.com --- drivers/gpu/drm/ttm/ttm_tt.c |2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff

[PATCH 07/14] drm/ttm: page allocation use page array instead of list

From: Jerome Glisse jgli...@redhat.com Use the ttm_tt pages array for pages allocations, move the list unwinding into the page allocation functions. Signed-off-by: Jerome Glisse jgli...@redhat.com --- drivers/gpu/drm/ttm/ttm_page_alloc.c | 85 +-

[PATCH 08/14] drm/ttm: merge ttm_backend and ttm_tt V5

From: Jerome Glisse jgli...@redhat.com ttm_backend will only exist with a ttm_tt, and ttm_tt will only be of interest when bound to a backend. Merge them to avoid code and data duplication. V2 Rebase on top of memory accounting overhaul V3 Rebase on top of more memory accounting changes V4

[PATCH 09/14] drm/ttm: introduce callback for ttm_tt populate unpopulate V4

From: Jerome Glisse jgli...@redhat.com Move the page allocation and freeing to driver callback and provide ttm code helper function for those. Most intrusive change, is the fact that we now only fully populate an object this simplify some of code designed around the page fault design. V2 Rebase

[PATCH 10/14] drm/ttm: provide dma aware ttm page pool code V9

From: Konrad Rzeszutek Wilk konrad.w...@oracle.com In TTM world the pages for the graphic drivers are kept in three different pools: write combined, uncached, and cached (write-back). When the pages are used by the graphic driver the graphic adapter via its built in MMU (or AGP) programs these

[PATCH 11/14] drm/radeon/kms: enable the ttm dma pool if swiotlb is on V4

From: Konrad Rzeszutek Wilk konrad.w...@oracle.com With the exception that we do not handle the AGP case. We only deal with PCIe cards such as ATI ES1000 or HD3200 that have been detected to only do DMA up to 32-bits. V2 force dma32 if we fail to set bigger dma mask V3 Rebase on top of no memory

[PATCH 12/14] drm/nouveau: enable the ttm dma pool when swiotlb is active V3

From: Konrad Rzeszutek Wilk konrad.w...@oracle.com If the card is capable of more than 32-bit, then use the default TTM page pool code which allocates from anywhere in the memory. Note: If the 'ttm.no_dma' parameter is set, the override is ignored and the default TTM pool is used. V2 use

  1   2   >