Re: GEM-related desktop sluggishness due to linear-time arch_get_unmapped_area_topdown()
On Wed, 30 Mar 2011 07:57:49 +1000, Dave Airlie wrote: > On Wed, Mar 30, 2011 at 7:04 AM, Jerome Glisse wrote: > > What i had in mind was something little bit more advance that pwrite, > > somethings that would take width,height,pitch of userpage and would be > > able to perform proper blit. But yes pwrite in intel is kind of > > limited. > > TTM has support for userpage binding we just don't use it. Yes, and I've been experimenting with the same in GEM to great effect in the DDX. The complication remains in managing the CPU synchronisation, which suggests that it would only be useful for STREAM_DRAW objects (and perhaps the sub-region updates to STATIC_DRAW). (And for readback, if retrieving the data were the actual bottleneck.) And I did play with a new pread/pwrite interface that did as you suggest, binding the user pages and performing a blit. But by the time you make the interface asynchronous, it becomes much easier to let the client code create the mapping and be fully aware of the barriers. And yes I do concur that vma bookkeeping does impose significant overheads and I have been removing as many mappings from our drivers as I can; within the limitations of the pwrite interface. -Chris -- Chris Wilson, Intel Open Source Technology Centre ___ dri-devel mailing list dri-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/dri-devel
[Bug 21682] White screen with compiz/AIGLX on X.org server 1.5.2 when running on 16bpp (intel 945GM)
https://bugs.freedesktop.org/show_bug.cgi?id=21682 --- Comment #19 from Tamás Németh 2011-03-30 02:19:50 PDT --- (In reply to comment #18) > Tamas, any news on that one? I just tried it on openSUSE 11.4. Compiz 0.9.x is practically unusable (it crashes in half a minute after starting), however I was able to recognize that the white screen problem is gone in 16bpp but now the color palette of windows totally went wrong and the windows are displayed scaled down four times and every window is composed four of its scaled down image ad then tiled together. Totally unusable. Rendering is correct at higher color depth but as I mentioned earlier, compiz always crashes in half a minute after starting. I still have an openSUSE 11.3 machine. Should I try that too? -- Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email --- You are receiving this mail because: --- You are the assignee for the bug. ___ dri-devel mailing list dri-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/dri-devel
[Bug 21682] White screen with compiz/AIGLX on X.org server 1.5.2 when running on 16bpp (intel 945GM)
https://bugs.freedesktop.org/show_bug.cgi?id=21682 --- Comment #20 from Stefan Dirsch 2011-03-30 02:49:02 PDT --- (In reply to comment #19) > I still have an openSUSE 11.3 machine. Should I try that too? That would be great, yes. -- Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email --- You are receiving this mail because: --- You are the assignee for the bug. ___ dri-devel mailing list dri-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/dri-devel
Re: [RFC] drm platform multi-device support
Are you by any chance working off a stale branch in git ?
Apparently the lines you are modifying now live in drm_platform_set_busid
function in drm_platform.c
-- Ilija
On Tue, 29 Mar 2011, Rob Clark wrote:
From: Rob Clark
Include the minor device id in the bus-id to give userspace a way to open
the correct "cardN" when there are multiple device instances.
---
It may seem a bit strange to have multiple platform device instances, but
we have a use-case for this. And this relatively simple patch does the
trick.
drivers/gpu/drm/drm_ioctl.c |5 +++--
1 files changed, 3 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/drm/drm_ioctl.c b/drivers/gpu/drm/drm_ioctl.c
index 47db4df..075024b 100644
--- a/drivers/gpu/drm/drm_ioctl.c
+++ b/drivers/gpu/drm/drm_ioctl.c
@@ -165,14 +165,15 @@ static int drm_set_busid(struct drm_device *dev, struct
drm_file *file_priv)
drm_unset_busid(dev, master);
if (drm_core_check_feature(dev, DRIVER_USE_PLATFORM_DEVICE)) {
- master->unique_len = 10 + strlen(dev->platformdev->name);
+ master->unique_len = 13 + strlen(dev->platformdev->name);
+ master->unique_size = master->unique_len;
master->unique = kmalloc(master->unique_len + 1, GFP_KERNEL);
if (master->unique == NULL)
return -ENOMEM;
len = snprintf(master->unique, master->unique_len,
- "platform:%s", dev->platformdev->name);
+ "platform:%s:%02d", dev->platformdev->name,
dev->primary->index);
if (len > master->unique_len) {
DRM_ERROR("Unique buffer overflowed\n");
--
1.7.1
___
dri-devel mailing list
dri-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/dri-devel
___
dri-devel mailing list
dri-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/dri-devel
Re: GEM-related desktop sluggishness due to linear-time arch_get_unmapped_area_topdown()
On Wed, Mar 30, 2011 at 3:32 AM, Chris Wilson wrote: > On Wed, 30 Mar 2011 07:57:49 +1000, Dave Airlie wrote: >> On Wed, Mar 30, 2011 at 7:04 AM, Jerome Glisse wrote: >> > What i had in mind was something little bit more advance that pwrite, >> > somethings that would take width,height,pitch of userpage and would be >> > able to perform proper blit. But yes pwrite in intel is kind of >> > limited. >> >> TTM has support for userpage binding we just don't use it. > > Yes, and I've been experimenting with the same in GEM to great effect in > the DDX. The complication remains in managing the CPU synchronisation, > which suggests that it would only be useful for STREAM_DRAW objects (and > perhaps the sub-region updates to STATIC_DRAW). (And for readback, if > retrieving the data were the actual bottleneck.) What do you mean by CPU synchronisation ? In what i had in mind the upload/download would block userspace until operation is, this would make upload/dowload barrier of course it doesn't play well with usecase where you keep uploading/downloading (idea to aleviate that is to allow several download/upload in one ioctl call). > And I did play with a new pread/pwrite interface that did as you suggest, > binding the user pages and performing a blit. But by the time you make the > interface asynchronous, it becomes much easier to let the client code > create the mapping and be fully aware of the barriers. > > And yes I do concur that vma bookkeeping does impose significant overheads > and I have been removing as many mappings from our drivers as I can; within > the limitations of the pwrite interface. > -Chris > > -- > Chris Wilson, Intel Open Source Technology Centre > Cheers, Jerome ___ dri-devel mailing list dri-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/dri-devel
[Bug 21682] White screen with compiz/AIGLX on X.org server 1.5.2 when running on 16bpp (intel 945GM)
https://bugs.freedesktop.org/show_bug.cgi?id=21682 --- Comment #21 from Tamás Németh 2011-03-30 06:48:31 PDT --- Created an attachment (id=45041) --> (https://bugs.freedesktop.org/attachment.cgi?id=45041) My normal openSUSE 11.3 desktop (24bpp) -- Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email --- You are receiving this mail because: --- You are the assignee for the bug. ___ dri-devel mailing list dri-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/dri-devel
[Bug 21682] White screen with compiz/AIGLX on X.org server 1.5.2 when running on 16bpp (intel 945GM)
https://bugs.freedesktop.org/show_bug.cgi?id=21682 --- Comment #22 from Tamás Németh 2011-03-30 06:50:47 PDT --- Created an attachment (id=45042) --> (https://bugs.freedesktop.org/attachment.cgi?id=45042) My openSUSE 11.3 desktop distorted at 16bpp -- Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email --- You are receiving this mail because: --- You are the assignee for the bug. ___ dri-devel mailing list dri-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/dri-devel
[Bug 34618] Slow text scrolling on tty after suspend-cycle
https://bugs.freedesktop.org/show_bug.cgi?id=34618 --- Comment #22 from pete...@hottemptation.org 2011-03-30 06:56:43 PDT --- Scrolling is still painfully slow after a suspend-cycle with kernel 2.6.39-rc1. Working with tools like less is impossible :-( Interesting: After a suspend-cycle normally not every virtual-terminal is affected, most times one or two vt still working fast. Here an example: 1. Boot 2. pm-suspend on vt2 3. resume from suspend 4. everyting still works fine! 5. pm-suspend on vt2 6. resume from suspend 7. scrolling is slow on vt1, vt3, vt4 but not on vt2 Sometimes pressing a button helps to speed up the scrolling. Not always, after a logout/login on a vt pressing a button while scrolling maybe doesn't work anymore. -- Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email --- You are receiving this mail because: --- You are the assignee for the bug. ___ dri-devel mailing list dri-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/dri-devel
[Bug 21682] White screen with compiz/AIGLX on X.org server 1.5.2 when running on 16bpp (intel 945GM)
https://bugs.freedesktop.org/show_bug.cgi?id=21682 --- Comment #23 from Tamás Németh 2011-03-30 07:00:01 PDT --- (In reply to comment #20) > (In reply to comment #19) > > I still have an openSUSE 11.3 machine. Should I try that too? > > That would be great, yes. Sadly, openSUSE 11.3 virtually behaves the same way as 11.4. (See my two attachments.) -- Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email --- You are receiving this mail because: --- You are the assignee for the bug. ___ dri-devel mailing list dri-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/dri-devel
Re: GEM-related desktop sluggishness due to linear-time arch_get_unmapped_area_topdown()
On Wed, 30 Mar 2011 09:28:07 -0400, Jerome Glisse wrote: > On Wed, Mar 30, 2011 at 3:32 AM, Chris Wilson > wrote: > > On Wed, 30 Mar 2011 07:57:49 +1000, Dave Airlie wrote: > >> On Wed, Mar 30, 2011 at 7:04 AM, Jerome Glisse wrote: > >> > What i had in mind was something little bit more advance that pwrite, > >> > somethings that would take width,height,pitch of userpage and would be > >> > able to perform proper blit. But yes pwrite in intel is kind of > >> > limited. > >> > >> TTM has support for userpage binding we just don't use it. > > > > Yes, and I've been experimenting with the same in GEM to great effect in > > the DDX. The complication remains in managing the CPU synchronisation, > > which suggests that it would only be useful for STREAM_DRAW objects (and > > perhaps the sub-region updates to STATIC_DRAW). (And for readback, if > > retrieving the data were the actual bottleneck.) > > What do you mean by CPU synchronisation ? In what i had in mind the > upload/download would block userspace until operation is, this would > make upload/dowload barrier of course it doesn't play well with > usecase where you keep uploading/downloading (idea to aleviate that is > to allow several download/upload in one ioctl call). Yes, that is the issue: having to control access to the user pages whilst they are in use by the GPU. A completely synchronous API for performing a single pwrite with the blitter is too slow, much slower than doing an uncached write with the CPU and queueing up multiple blits (as we currently do). The API I ended up with for the pwrite using the BLT was to specify a 2D region (addr, width, height, stride, flags etc) and list of clip rects. At which point I grew disenchanted, and realised that simply creating a bo for mapping user pages was the far better solution. -Chris -- Chris Wilson, Intel Open Source Technology Centre ___ dri-devel mailing list dri-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/dri-devel
Re: [RFC] drm platform multi-device support
Ahh, yes I am.. I apologize, still too much display and gfx accel and
other stuff I depend on isn't in kernel.org tree yet, so I'm just
using ubuntu 2.6.38 tree, ti-omap4 branch, (since this is the most
current kernel that also has everything else I need). But I guess it
is a bit out of date already. Maybe there is a way I could pull all
the patches just from drivers/gpu/drm directory to stay a bit more
current?
But anyways, if everyone is ok with the idea to change the bus-id this
way, then I could resubmit a patch with the same change in
drm_platform_set_busid()
BR,
-R
On Wed, Mar 30, 2011 at 7:57 AM, Ilija Hadzic
wrote:
>
> Are you by any chance working off a stale branch in git ?
>
> Apparently the lines you are modifying now live in drm_platform_set_busid
> function in drm_platform.c
>
> -- Ilija
>
> On Tue, 29 Mar 2011, Rob Clark wrote:
>
>> From: Rob Clark
>>
>> Include the minor device id in the bus-id to give userspace a way to open
>> the correct "cardN" when there are multiple device instances.
>> ---
>> It may seem a bit strange to have multiple platform device instances, but
>> we have a use-case for this. And this relatively simple patch does the
>> trick.
>>
>> drivers/gpu/drm/drm_ioctl.c | 5 +++--
>> 1 files changed, 3 insertions(+), 2 deletions(-)
>>
>> diff --git a/drivers/gpu/drm/drm_ioctl.c b/drivers/gpu/drm/drm_ioctl.c
>> index 47db4df..075024b 100644
>> --- a/drivers/gpu/drm/drm_ioctl.c
>> +++ b/drivers/gpu/drm/drm_ioctl.c
>> @@ -165,14 +165,15 @@ static int drm_set_busid(struct drm_device *dev,
>> struct drm_file *file_priv)
>> drm_unset_busid(dev, master);
>>
>> if (drm_core_check_feature(dev, DRIVER_USE_PLATFORM_DEVICE)) {
>> - master->unique_len = 10 + strlen(dev->platformdev->name);
>> + master->unique_len = 13 + strlen(dev->platformdev->name);
>> + master->unique_size = master->unique_len;
>> master->unique = kmalloc(master->unique_len + 1,
>> GFP_KERNEL);
>>
>> if (master->unique == NULL)
>> return -ENOMEM;
>>
>> len = snprintf(master->unique, master->unique_len,
>> - "platform:%s", dev->platformdev->name);
>> + "platform:%s:%02d", dev->platformdev->name,
>> dev->primary->index);
>>
>> if (len > master->unique_len) {
>> DRM_ERROR("Unique buffer overflowed\n");
>> --
>> 1.7.1
>>
>> ___
>> dri-devel mailing list
>> dri-devel@lists.freedesktop.org
>> http://lists.freedesktop.org/mailman/listinfo/dri-devel
>>
> ___
> dri-devel mailing list
> dri-devel@lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/dri-devel
>
___
dri-devel mailing list
dri-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/dri-devel
Re: GEM-related desktop sluggishness due to linear-time arch_get_unmapped_area_topdown()
On Wed, Mar 30, 2011 at 10:07 AM, Chris Wilson wrote: > On Wed, 30 Mar 2011 09:28:07 -0400, Jerome Glisse wrote: >> On Wed, Mar 30, 2011 at 3:32 AM, Chris Wilson >> wrote: >> > On Wed, 30 Mar 2011 07:57:49 +1000, Dave Airlie wrote: >> >> On Wed, Mar 30, 2011 at 7:04 AM, Jerome Glisse wrote: >> >> > What i had in mind was something little bit more advance that pwrite, >> >> > somethings that would take width,height,pitch of userpage and would be >> >> > able to perform proper blit. But yes pwrite in intel is kind of >> >> > limited. >> >> >> >> TTM has support for userpage binding we just don't use it. >> > >> > Yes, and I've been experimenting with the same in GEM to great effect in >> > the DDX. The complication remains in managing the CPU synchronisation, >> > which suggests that it would only be useful for STREAM_DRAW objects (and >> > perhaps the sub-region updates to STATIC_DRAW). (And for readback, if >> > retrieving the data were the actual bottleneck.) >> >> What do you mean by CPU synchronisation ? In what i had in mind the >> upload/download would block userspace until operation is, this would >> make upload/dowload barrier of course it doesn't play well with >> usecase where you keep uploading/downloading (idea to aleviate that is >> to allow several download/upload in one ioctl call). > > Yes, that is the issue: having to control access to the user pages whilst > they are in use by the GPU. A completely synchronous API for performing > a single pwrite with the blitter is too slow, much slower than doing an > uncached write with the CPU and queueing up multiple blits (as we > currently do). > > The API I ended up with for the pwrite using the BLT was to specify a 2D > region (addr, width, height, stride, flags etc) and list of clip rects. At > which point I grew disenchanted, and realised that simply creating a bo > for mapping user pages was the far better solution. > -Chris > > -- > Chris Wilson, Intel Open Source Technology Centre > What kind of usage didn't played well with synchronous upload/download ? X, GL ? Cheers, Jerome ___ dri-devel mailing list dri-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/dri-devel
Re: GEM-related desktop sluggishness due to linear-time arch_get_unmapped_area_topdown()
On Wed, 30 Mar 2011 11:07:08 -0400, Jerome Glisse wrote: > What kind of usage didn't played well with synchronous upload/download ? X, > GL ? Performing fallback rendering for X using a shadow buffer. -Chris -- Chris Wilson, Intel Open Source Technology Centre ___ dri-devel mailing list dri-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/dri-devel
[Bug 35280] [RADEON:KMS:R600G] kwin freezes at drawing combo boxes
https://bugs.freedesktop.org/show_bug.cgi?id=35280 Eugene Shalygin changed: What|Removed |Added Status|NEW |RESOLVED Resolution||WORKSFORME --- Comment #1 from Eugene Shalygin 2011-03-30 08:48:37 PDT --- Seems to be working with 2.6.38 kernel -- Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email --- You are receiving this mail because: --- You are the assignee for the bug. ___ dri-devel mailing list dri-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/dri-devel
[Bug 35434] [RADEON:KMS:R600G] etqw: broken ground textures + sporadic junk being displayed
https://bugs.freedesktop.org/show_bug.cgi?id=35434 --- Comment #3 from Benjamin Bellec 2011-03-30 10:41:14 PDT --- I also have corrupted textures. Buildings textures doesn't have problem, only the ground textures. RV770 / 2.6.39-rc1 / F15 updated -- Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email --- You are receiving this mail because: --- You are the assignee for the bug. ___ dri-devel mailing list dri-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/dri-devel
[Bug 35434] [RADEON:KMS:R600G] etqw: broken ground textures + sporadic junk being displayed
https://bugs.freedesktop.org/show_bug.cgi?id=35434 Benjamin Bellec changed: What|Removed |Added Platform|x86-64 (AMD64) |All -- Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email --- You are receiving this mail because: --- You are the assignee for the bug. ___ dri-devel mailing list dri-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/dri-devel
[Bug 31782] nouveau: lockdep spew
https://bugzilla.kernel.org/show_bug.cgi?id=31782 Andrew Morton changed: What|Removed |Added CC||a...@linux-foundation.org --- Comment #1 from Andrew Morton 2011-03-30 21:26:16 --- Kernel version is unclear. Is this a 2.6.37->2.6.38 regression, or a post-2.6.38 regression? Thanks. -- Configure bugmail: https://bugzilla.kernel.org/userprefs.cgi?tab=email --- You are receiving this mail because: --- You are watching the assignee of the bug. -- Create and publish websites with WebMatrix Use the most popular FREE web apps or write code yourself; WebMatrix provides all the features you need to develop and publish your website. http://p.sf.net/sfu/ms-webmatrix-sf -- ___ Dri-devel mailing list dri-de...@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dri-devel ___ dri-devel mailing list dri-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/dri-devel
[Bug 31782] nouveau: lockdep spew
https://bugzilla.kernel.org/show_bug.cgi?id=31782 Lucas Stach changed: What|Removed |Added CC||d...@lynxeye.de --- Comment #2 from Lucas Stach 2011-03-30 22:18:22 --- I think this problem was introduced by 6f70a4c3d19e8e8e1047a4dbf0ca910fed39f619. This commit fixes a real problem, so it shouldn't be reverted, but still this issue needs to be fixed. It seems Ben isn't registered here at kernel bugzilla. I will drop him a mail about this. -- Configure bugmail: https://bugzilla.kernel.org/userprefs.cgi?tab=email --- You are receiving this mail because: --- You are watching the assignee of the bug. -- Create and publish websites with WebMatrix Use the most popular FREE web apps or write code yourself; WebMatrix provides all the features you need to develop and publish your website. http://p.sf.net/sfu/ms-webmatrix-sf -- ___ Dri-devel mailing list dri-de...@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dri-devel ___ dri-devel mailing list dri-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/dri-devel
[Bug 31782] nouveau: lockdep spew
https://bugzilla.kernel.org/show_bug.cgi?id=31782 Rafael J. Wysocki changed: What|Removed |Added CC||flor...@mickler.org, ||maciej.rute...@gmail.com, ||r...@sisk.pl Blocks||27352 -- Configure bugmail: https://bugzilla.kernel.org/userprefs.cgi?tab=email --- You are receiving this mail because: --- You are watching the assignee of the bug. -- Create and publish websites with WebMatrix Use the most popular FREE web apps or write code yourself; WebMatrix provides all the features you need to develop and publish your website. http://p.sf.net/sfu/ms-webmatrix-sf -- ___ Dri-devel mailing list dri-de...@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dri-devel ___ dri-devel mailing list dri-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/dri-devel
[Bug 31782] nouveau: lockdep spew
https://bugzilla.kernel.org/show_bug.cgi?id=31782 --- Comment #3 from Lucas Stach 2011-03-31 06:08:35 --- Created an attachment (id=52722) --> (https://bugzilla.kernel.org/attachment.cgi?id=52722) slight lock reordering from Ben Could you please test the attached patch? It should fix the problem, but as I'm physically away from my nvidia machine I could not test it myself. -- Configure bugmail: https://bugzilla.kernel.org/userprefs.cgi?tab=email --- You are receiving this mail because: --- You are watching the assignee of the bug. -- Create and publish websites with WebMatrix Use the most popular FREE web apps or write code yourself; WebMatrix provides all the features you need to develop and publish your website. http://p.sf.net/sfu/ms-webmatrix-sf -- ___ Dri-devel mailing list dri-de...@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dri-devel ___ dri-devel mailing list dri-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/dri-devel
GEM-related desktop sluggishness due to linear-time arch_get_unmapped_area_topdown()
On Wed, Mar 30, 2011 at 7:04 AM, Jerome Glisse wrote: > On Tue, Mar 29, 2011 at 4:26 PM, Daniel Vetter wrote: >> On Tue, Mar 29, 2011 at 03:45:34PM -0400, Jerome Glisse wrote: >>> Short lived & small bo would definitly doesn't work well for this kind >>> of API, it would all be a function of the ioctl cost. But i am not >>> sure the drawback would be that big, intel tested with pread/pwrite >>> and gived up don't remember why. For the vbo case you describe the >>> scheme i was thinking would be : allocate bo and on buffer data call >>> upload to the allocated bo using the bind user page feature that would >>> mean zero extra copy operation. For the fire forget case of vbo, >>> likely somekind of transient buffer would be more appropriate. >> >> Just to clarify: Uploads to linear buffers are all done with pwrite (due >> to an api foobar, it's not so great for 2d/tiled stuff). It's amazing how >> much faster that is: Switching vbo's from mmpa to pwrite has given a few >> percent more fps in openarena in i915g! As long as the chunk you're gonna >> write fits into L1 cache, it's probably a net win. >> -Daniel > > What i had in mind was something little bit more advance that pwrite, > somethings that would take width,height,pitch of userpage and would be > able to perform proper blit. But yes pwrite in intel is kind of > limited. TTM has support for userpage binding we just don't use it. Dave.
[PATCH] drm: export drm_find_cea_extension to drivers
From: Ben Skeggs
Nouveau needs access to this structure to build an ELD block for use
by the HDA audio codec.
Signed-off-by: Ben Skeggs
---
drivers/gpu/drm/drm_edid.c |3 ++-
include/drm/drm_crtc.h |1 +
2 files changed, 3 insertions(+), 1 deletions(-)
diff --git a/drivers/gpu/drm/drm_edid.c b/drivers/gpu/drm/drm_edid.c
index a245d17..0c39b09 100644
--- a/drivers/gpu/drm/drm_edid.c
+++ b/drivers/gpu/drm/drm_edid.c
@@ -1288,7 +1288,7 @@ add_detailed_modes(struct drm_connector *connector,
struct edid *edid,
/**
* Search EDID for CEA extension block.
*/
-static u8 *drm_find_cea_extension(struct edid *edid)
+u8 *drm_find_cea_extension(struct edid *edid)
{
u8 *edid_ext = NULL;
int i;
@@ -1309,6 +1309,7 @@ static u8 *drm_find_cea_extension(struct edid *edid)
return edid_ext;
}
+EXPORT_SYMBOL(drm_find_cea_extension);
/**
* drm_detect_hdmi_monitor - detect whether monitor is hdmi.
diff --git a/include/drm/drm_crtc.h b/include/drm/drm_crtc.h
index 801be59..5422004 100644
--- a/include/drm/drm_crtc.h
+++ b/include/drm/drm_crtc.h
@@ -778,6 +778,7 @@ extern int drm_mode_gamma_get_ioctl(struct drm_device *dev,
void *data, struct drm_file *file_priv);
extern int drm_mode_gamma_set_ioctl(struct drm_device *dev,
void *data, struct drm_file *file_priv);
+extern u8 *drm_find_cea_extension(struct edid *edid);
extern bool drm_detect_hdmi_monitor(struct edid *edid);
extern bool drm_detect_monitor_audio(struct edid *edid);
extern int drm_mode_page_flip_ioctl(struct drm_device *dev,
--
1.7.4.2
GEM-related desktop sluggishness due to linear-time arch_get_unmapped_area_topdown()
On Wed, 30 Mar 2011 07:57:49 +1000, Dave Airlie wrote: > On Wed, Mar 30, 2011 at 7:04 AM, Jerome Glisse wrote: > > What i had in mind was something little bit more advance that pwrite, > > somethings that would take width,height,pitch of userpage and would be > > able to perform proper blit. But yes pwrite in intel is kind of > > limited. > > TTM has support for userpage binding we just don't use it. Yes, and I've been experimenting with the same in GEM to great effect in the DDX. The complication remains in managing the CPU synchronisation, which suggests that it would only be useful for STREAM_DRAW objects (and perhaps the sub-region updates to STATIC_DRAW). (And for readback, if retrieving the data were the actual bottleneck.) And I did play with a new pread/pwrite interface that did as you suggest, binding the user pages and performing a blit. But by the time you make the interface asynchronous, it becomes much easier to let the client code create the mapping and be fully aware of the barriers. And yes I do concur that vma bookkeeping does impose significant overheads and I have been removing as many mappings from our drivers as I can; within the limitations of the pwrite interface. -Chris -- Chris Wilson, Intel Open Source Technology Centre
[Bug 21682] White screen with compiz/AIGLX on X.org server 1.5.2 when running on 16bpp (intel 945GM)
https://bugs.freedesktop.org/show_bug.cgi?id=21682 --- Comment #19 from Tam?s N?meth 2011-03-30 02:19:50 PDT --- (In reply to comment #18) > Tamas, any news on that one? I just tried it on openSUSE 11.4. Compiz 0.9.x is practically unusable (it crashes in half a minute after starting), however I was able to recognize that the white screen problem is gone in 16bpp but now the color palette of windows totally went wrong and the windows are displayed scaled down four times and every window is composed four of its scaled down image ad then tiled together. Totally unusable. Rendering is correct at higher color depth but as I mentioned earlier, compiz always crashes in half a minute after starting. I still have an openSUSE 11.3 machine. Should I try that too? -- Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email --- You are receiving this mail because: --- You are the assignee for the bug.
[Bug 21682] White screen with compiz/AIGLX on X.org server 1.5.2 when running on 16bpp (intel 945GM)
https://bugs.freedesktop.org/show_bug.cgi?id=21682 --- Comment #20 from Stefan Dirsch 2011-03-30 02:49:02 PDT --- (In reply to comment #19) > I still have an openSUSE 11.3 machine. Should I try that too? That would be great, yes. -- Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email --- You are receiving this mail because: --- You are the assignee for the bug.
[RFC] drm platform multi-device support
Are you by any chance working off a stale branch in git ?
Apparently the lines you are modifying now live in drm_platform_set_busid
function in drm_platform.c
-- Ilija
On Tue, 29 Mar 2011, Rob Clark wrote:
> From: Rob Clark
>
> Include the minor device id in the bus-id to give userspace a way to open
> the correct "cardN" when there are multiple device instances.
> ---
> It may seem a bit strange to have multiple platform device instances, but
> we have a use-case for this. And this relatively simple patch does the
> trick.
>
> drivers/gpu/drm/drm_ioctl.c |5 +++--
> 1 files changed, 3 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/gpu/drm/drm_ioctl.c b/drivers/gpu/drm/drm_ioctl.c
> index 47db4df..075024b 100644
> --- a/drivers/gpu/drm/drm_ioctl.c
> +++ b/drivers/gpu/drm/drm_ioctl.c
> @@ -165,14 +165,15 @@ static int drm_set_busid(struct drm_device *dev, struct
> drm_file *file_priv)
> drm_unset_busid(dev, master);
>
> if (drm_core_check_feature(dev, DRIVER_USE_PLATFORM_DEVICE)) {
> - master->unique_len = 10 + strlen(dev->platformdev->name);
> + master->unique_len = 13 + strlen(dev->platformdev->name);
> + master->unique_size = master->unique_len;
> master->unique = kmalloc(master->unique_len + 1, GFP_KERNEL);
>
> if (master->unique == NULL)
> return -ENOMEM;
>
> len = snprintf(master->unique, master->unique_len,
> - "platform:%s", dev->platformdev->name);
> + "platform:%s:%02d", dev->platformdev->name,
> dev->primary->index);
>
> if (len > master->unique_len) {
> DRM_ERROR("Unique buffer overflowed\n");
> --
> 1.7.1
>
> ___
> dri-devel mailing list
> dri-devel at lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/dri-devel
>
GEM-related desktop sluggishness due to linear-time arch_get_unmapped_area_topdown()
On Wed, Mar 30, 2011 at 3:32 AM, Chris Wilson wrote: > On Wed, 30 Mar 2011 07:57:49 +1000, Dave Airlie wrote: >> On Wed, Mar 30, 2011 at 7:04 AM, Jerome Glisse wrote: >> > What i had in mind was something little bit more advance that pwrite, >> > somethings that would take width,height,pitch of userpage and would be >> > able to perform proper blit. But yes pwrite in intel is kind of >> > limited. >> >> TTM has support for userpage binding we just don't use it. > > Yes, and I've been experimenting with the same in GEM to great effect in > the DDX. The complication remains in managing the CPU synchronisation, > which suggests that it would only be useful for STREAM_DRAW objects (and > perhaps the sub-region updates to STATIC_DRAW). (And for readback, if > retrieving the data were the actual bottleneck.) What do you mean by CPU synchronisation ? In what i had in mind the upload/download would block userspace until operation is, this would make upload/dowload barrier of course it doesn't play well with usecase where you keep uploading/downloading (idea to aleviate that is to allow several download/upload in one ioctl call). > And I did play with a new pread/pwrite interface that did as you suggest, > binding the user pages and performing a blit. But by the time you make the > interface asynchronous, it becomes much easier to let the client code > create the mapping and be fully aware of the barriers. > > And yes I do concur that vma bookkeeping does impose significant overheads > and I have been removing as many mappings from our drivers as I can; within > the limitations of the pwrite interface. > -Chris > > -- > Chris Wilson, Intel Open Source Technology Centre > Cheers, Jerome
[Bug 21682] White screen with compiz/AIGLX on X.org server 1.5.2 when running on 16bpp (intel 945GM)
https://bugs.freedesktop.org/show_bug.cgi?id=21682 --- Comment #21 from Tam?s N?meth 2011-03-30 06:48:31 PDT --- Created an attachment (id=45041) --> (https://bugs.freedesktop.org/attachment.cgi?id=45041) My normal openSUSE 11.3 desktop (24bpp) -- Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email --- You are receiving this mail because: --- You are the assignee for the bug.
[Bug 21682] White screen with compiz/AIGLX on X.org server 1.5.2 when running on 16bpp (intel 945GM)
https://bugs.freedesktop.org/show_bug.cgi?id=21682 --- Comment #22 from Tam?s N?meth 2011-03-30 06:50:47 PDT --- Created an attachment (id=45042) --> (https://bugs.freedesktop.org/attachment.cgi?id=45042) My openSUSE 11.3 desktop distorted at 16bpp -- Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email --- You are receiving this mail because: --- You are the assignee for the bug.
[Bug 34618] Slow text scrolling on tty after suspend-cycle
https://bugs.freedesktop.org/show_bug.cgi?id=34618 --- Comment #22 from peterle at hottemptation.org 2011-03-30 06:56:43 PDT --- Scrolling is still painfully slow after a suspend-cycle with kernel 2.6.39-rc1. Working with tools like less is impossible :-( Interesting: After a suspend-cycle normally not every virtual-terminal is affected, most times one or two vt still working fast. Here an example: 1. Boot 2. pm-suspend on vt2 3. resume from suspend 4. everyting still works fine! 5. pm-suspend on vt2 6. resume from suspend 7. scrolling is slow on vt1, vt3, vt4 but not on vt2 Sometimes pressing a button helps to speed up the scrolling. Not always, after a logout/login on a vt pressing a button while scrolling maybe doesn't work anymore. -- Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email --- You are receiving this mail because: --- You are the assignee for the bug.
[Bug 21682] White screen with compiz/AIGLX on X.org server 1.5.2 when running on 16bpp (intel 945GM)
https://bugs.freedesktop.org/show_bug.cgi?id=21682 --- Comment #23 from Tam?s N?meth 2011-03-30 07:00:01 PDT --- (In reply to comment #20) > (In reply to comment #19) > > I still have an openSUSE 11.3 machine. Should I try that too? > > That would be great, yes. Sadly, openSUSE 11.3 virtually behaves the same way as 11.4. (See my two attachments.) -- Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email --- You are receiving this mail because: --- You are the assignee for the bug.
GEM-related desktop sluggishness due to linear-time arch_get_unmapped_area_topdown()
On Wed, 30 Mar 2011 09:28:07 -0400, Jerome Glisse wrote: > On Wed, Mar 30, 2011 at 3:32 AM, Chris Wilson > wrote: > > On Wed, 30 Mar 2011 07:57:49 +1000, Dave Airlie > > wrote: > >> On Wed, Mar 30, 2011 at 7:04 AM, Jerome Glisse > >> wrote: > >> > What i had in mind was something little bit more advance that pwrite, > >> > somethings that would take width,height,pitch of userpage and would be > >> > able to perform proper blit. But yes pwrite in intel is kind of > >> > limited. > >> > >> TTM has support for userpage binding we just don't use it. > > > > Yes, and I've been experimenting with the same in GEM to great effect in > > the DDX. The complication remains in managing the CPU synchronisation, > > which suggests that it would only be useful for STREAM_DRAW objects (and > > perhaps the sub-region updates to STATIC_DRAW). (And for readback, if > > retrieving the data were the actual bottleneck.) > > What do you mean by CPU synchronisation ? In what i had in mind the > upload/download would block userspace until operation is, this would > make upload/dowload barrier of course it doesn't play well with > usecase where you keep uploading/downloading (idea to aleviate that is > to allow several download/upload in one ioctl call). Yes, that is the issue: having to control access to the user pages whilst they are in use by the GPU. A completely synchronous API for performing a single pwrite with the blitter is too slow, much slower than doing an uncached write with the CPU and queueing up multiple blits (as we currently do). The API I ended up with for the pwrite using the BLT was to specify a 2D region (addr, width, height, stride, flags etc) and list of clip rects. At which point I grew disenchanted, and realised that simply creating a bo for mapping user pages was the far better solution. -Chris -- Chris Wilson, Intel Open Source Technology Centre
[RFC] drm platform multi-device support
Ahh, yes I am.. I apologize, still too much display and gfx accel and
other stuff I depend on isn't in kernel.org tree yet, so I'm just
using ubuntu 2.6.38 tree, ti-omap4 branch, (since this is the most
current kernel that also has everything else I need). But I guess it
is a bit out of date already. Maybe there is a way I could pull all
the patches just from drivers/gpu/drm directory to stay a bit more
current?
But anyways, if everyone is ok with the idea to change the bus-id this
way, then I could resubmit a patch with the same change in
drm_platform_set_busid()
BR,
-R
On Wed, Mar 30, 2011 at 7:57 AM, Ilija Hadzic
wrote:
>
> Are you by any chance working off a stale branch in git ?
>
> Apparently the lines you are modifying now live in drm_platform_set_busid
> function in drm_platform.c
>
> -- Ilija
>
> On Tue, 29 Mar 2011, Rob Clark wrote:
>
>> From: Rob Clark
>>
>> Include the minor device id in the bus-id to give userspace a way to open
>> the correct "cardN" when there are multiple device instances.
>> ---
>> It may seem a bit strange to have multiple platform device instances, but
>> we have a use-case for this. ?And this relatively simple patch does the
>> trick.
>>
>> drivers/gpu/drm/drm_ioctl.c | ? ?5 +++--
>> 1 files changed, 3 insertions(+), 2 deletions(-)
>>
>> diff --git a/drivers/gpu/drm/drm_ioctl.c b/drivers/gpu/drm/drm_ioctl.c
>> index 47db4df..075024b 100644
>> --- a/drivers/gpu/drm/drm_ioctl.c
>> +++ b/drivers/gpu/drm/drm_ioctl.c
>> @@ -165,14 +165,15 @@ static int drm_set_busid(struct drm_device *dev,
>> struct drm_file *file_priv)
>> ? ? ? ? ? ? ? ?drm_unset_busid(dev, master);
>>
>> ? ? ? ?if (drm_core_check_feature(dev, DRIVER_USE_PLATFORM_DEVICE)) {
>> - ? ? ? ? ? ? ? master->unique_len = 10 + strlen(dev->platformdev->name);
>> + ? ? ? ? ? ? ? master->unique_len = 13 + strlen(dev->platformdev->name);
>> + ? ? ? ? ? ? ? master->unique_size = master->unique_len;
>> ? ? ? ? ? ? ? ?master->unique = kmalloc(master->unique_len + 1,
>> GFP_KERNEL);
>>
>> ? ? ? ? ? ? ? ?if (master->unique == NULL)
>> ? ? ? ? ? ? ? ? ? ? ? ?return -ENOMEM;
>>
>> ? ? ? ? ? ? ? ?len = snprintf(master->unique, master->unique_len,
>> - ? ? ? ? ? ? ? ? ? ? ? "platform:%s", dev->platformdev->name);
>> + ? ? ? ? ? ? ? ? ? ? ? "platform:%s:%02d", dev->platformdev->name,
>> dev->primary->index);
>>
>> ? ? ? ? ? ? ? ?if (len > master->unique_len) {
>> ? ? ? ? ? ? ? ? ? ? ? ?DRM_ERROR("Unique buffer overflowed\n");
>> --
>> 1.7.1
>>
>> ___
>> dri-devel mailing list
>> dri-devel at lists.freedesktop.org
>> http://lists.freedesktop.org/mailman/listinfo/dri-devel
>>
> ___
> dri-devel mailing list
> dri-devel at lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/dri-devel
>
GEM-related desktop sluggishness due to linear-time arch_get_unmapped_area_topdown()
On Wed, Mar 30, 2011 at 10:07 AM, Chris Wilson wrote: > On Wed, 30 Mar 2011 09:28:07 -0400, Jerome Glisse > wrote: >> On Wed, Mar 30, 2011 at 3:32 AM, Chris Wilson >> wrote: >> > On Wed, 30 Mar 2011 07:57:49 +1000, Dave Airlie >> > wrote: >> >> On Wed, Mar 30, 2011 at 7:04 AM, Jerome Glisse >> >> wrote: >> >> > What i had in mind was something little bit more advance that pwrite, >> >> > somethings that would take width,height,pitch of userpage and would be >> >> > able to perform proper blit. But yes pwrite in intel is kind of >> >> > limited. >> >> >> >> TTM has support for userpage binding we just don't use it. >> > >> > Yes, and I've been experimenting with the same in GEM to great effect in >> > the DDX. The complication remains in managing the CPU synchronisation, >> > which suggests that it would only be useful for STREAM_DRAW objects (and >> > perhaps the sub-region updates to STATIC_DRAW). (And for readback, if >> > retrieving the data were the actual bottleneck.) >> >> What do you mean by CPU synchronisation ? In what i had in mind the >> upload/download would block userspace until operation is, this would >> make upload/dowload barrier of course it doesn't play well with >> usecase where you keep uploading/downloading (idea to aleviate that is >> to allow several download/upload in one ioctl call). > > Yes, that is the issue: having to control access to the user pages whilst > they are in use by the GPU. A completely synchronous API for performing > a single pwrite with the blitter is too slow, much slower than doing an > uncached write with the CPU and queueing up multiple blits (as we > currently do). > > The API I ended up with for the pwrite using the BLT was to specify a 2D > region (addr, width, height, stride, flags etc) and list of clip rects. At > which point I grew disenchanted, and realised that simply creating a bo > for mapping user pages was the far better solution. > -Chris > > -- > Chris Wilson, Intel Open Source Technology Centre > What kind of usage didn't played well with synchronous upload/download ? X, GL ? Cheers, Jerome
GEM-related desktop sluggishness due to linear-time arch_get_unmapped_area_topdown()
On Wed, 30 Mar 2011 11:07:08 -0400, Jerome Glisse wrote: > What kind of usage didn't played well with synchronous upload/download ? X, > GL ? Performing fallback rendering for X using a shadow buffer. -Chris -- Chris Wilson, Intel Open Source Technology Centre
[Bug 35280] [RADEON:KMS:R600G] kwin freezes at drawing combo boxes
https://bugs.freedesktop.org/show_bug.cgi?id=35280 Eugene Shalygin changed: What|Removed |Added Status|NEW |RESOLVED Resolution||WORKSFORME --- Comment #1 from Eugene Shalygin 2011-03-30 08:48:37 PDT --- Seems to be working with 2.6.38 kernel -- Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email --- You are receiving this mail because: --- You are the assignee for the bug.
[Bug 35434] [RADEON:KMS:R600G] etqw: broken ground textures + sporadic junk being displayed
https://bugs.freedesktop.org/show_bug.cgi?id=35434 --- Comment #3 from Benjamin Bellec 2011-03-30 10:41:14 PDT --- I also have corrupted textures. Buildings textures doesn't have problem, only the ground textures. RV770 / 2.6.39-rc1 / F15 updated -- Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email --- You are receiving this mail because: --- You are the assignee for the bug.
[Bug 35434] [RADEON:KMS:R600G] etqw: broken ground textures + sporadic junk being displayed
https://bugs.freedesktop.org/show_bug.cgi?id=35434 Benjamin Bellec changed: What|Removed |Added Platform|x86-64 (AMD64) |All -- Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email --- You are receiving this mail because: --- You are the assignee for the bug.
[Bug 31782] nouveau: lockdep spew
https://bugzilla.kernel.org/show_bug.cgi?id=31782 Andrew Morton changed: What|Removed |Added CC||akpm at linux-foundation.org --- Comment #1 from Andrew Morton 2011-03-30 21:26:16 --- Kernel version is unclear. Is this a 2.6.37->2.6.38 regression, or a post-2.6.38 regression? Thanks. -- Configure bugmail: https://bugzilla.kernel.org/userprefs.cgi?tab=email --- You are receiving this mail because: --- You are watching the assignee of the bug. -- Create and publish websites with WebMatrix Use the most popular FREE web apps or write code yourself; WebMatrix provides all the features you need to develop and publish your website. http://p.sf.net/sfu/ms-webmatrix-sf -- ___ Dri-devel mailing list Dri-devel at lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dri-devel
[Bug 31782] nouveau: lockdep spew
https://bugzilla.kernel.org/show_bug.cgi?id=31782 Lucas Stach changed: What|Removed |Added CC||dev at lynxeye.de --- Comment #2 from Lucas Stach 2011-03-30 22:18:22 --- I think this problem was introduced by 6f70a4c3d19e8e8e1047a4dbf0ca910fed39f619. This commit fixes a real problem, so it shouldn't be reverted, but still this issue needs to be fixed. It seems Ben isn't registered here at kernel bugzilla. I will drop him a mail about this. -- Configure bugmail: https://bugzilla.kernel.org/userprefs.cgi?tab=email --- You are receiving this mail because: --- You are watching the assignee of the bug. -- Create and publish websites with WebMatrix Use the most popular FREE web apps or write code yourself; WebMatrix provides all the features you need to develop and publish your website. http://p.sf.net/sfu/ms-webmatrix-sf -- ___ Dri-devel mailing list Dri-devel at lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dri-devel
[Bug 31782] nouveau: lockdep spew
https://bugzilla.kernel.org/show_bug.cgi?id=31782 Rafael J. Wysocki changed: What|Removed |Added CC||florian at mickler.org, ||maciej.rutecki at gmail.com, ||rjw at sisk.pl Blocks||27352 -- Configure bugmail: https://bugzilla.kernel.org/userprefs.cgi?tab=email --- You are receiving this mail because: --- You are watching the assignee of the bug. -- Create and publish websites with WebMatrix Use the most popular FREE web apps or write code yourself; WebMatrix provides all the features you need to develop and publish your website. http://p.sf.net/sfu/ms-webmatrix-sf -- ___ Dri-devel mailing list Dri-devel at lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dri-devel
