gmane.comp.video.dri.devel

[Bug 50360] flightgear has absymal performance with the llvm shadercompiler

2012-05-25T22:21:57

https://bugs.freedesktop.org/show_bug.cgi?id=50360

--- Comment #1 from Vadim Girlin <ptpzz< at >yandex.ru> 2012-05-25 15:21:57 PDT ---
Created attachment 62113
  --> https://bugs.freedesktop.org/attachment.cgi?id=62113
patch

Does this patch help?

[Bug 17902] [830M missing dvo driver] need support for DVO-LVDS viana2501

2012-05-25T21:51:23

https://bugs.freedesktop.org/show_bug.cgi?id=17902

--- Comment #72 from thor< at >math.tu-berlin.de 2012-05-25 14:51:23 PDT ---
Ok, I managed to cross-compile the patched kernel sources on a much faster x64
machine, so I now have an initrd and a vmlinuz with the patch in it. I won't
have access to the machine until tomorrow.

Is this already supposed to do anything interesting, or would anyone just be
interested in the debug output?

I also found a datasheet from national semiconductor for the DS90C2501, is this
the right one?

[Bug 43207] radeon driver on HD6570 shows pixel noise

2012-05-25T20:10:06

https://bugzilla.kernel.org/show_bug.cgi?id=43207





--- Comment #9 from Vladislav Tcendrovskii <ccvs< at >mail.ru>  2012-05-25 20:10:06 ---
I've tried this patch on kernel 3.4, but the only visible result - switching
from noisy X screen to framebuffer console is much faster.

After such switching noise disappears.

[Bug 50360] New: flightgear has absymal performance with the llvmshader compiler

2012-05-25T18:08:15

https://bugs.freedesktop.org/show_bug.cgi?id=50360

             Bug #: 50360
           Summary: flightgear has absymal performance with the llvm
                    shader compiler
    Classification: Unclassified
           Product: Mesa
           Version: git
          Platform: Other
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: medium
         Component: Drivers/Gallium/r600
        AssignedTo: dri-devel< at >lists.freedesktop.org
        ReportedBy: aaalmosss< at >gmail.com


I get ~3 fps with the LLVM compiler, while I get ~20 fps with `R600_LLVM=0
fgfs`. The only setting on the 'rendering options' dialog that affects
framerate is 'material shaders', when it's off, I get ~50 fps with both
compilers.

Other games perform around the same with the two compilers.

[Bug 50232] screen redraw is wrong in sauerbraten with the llvmcompiler

2012-05-25T17:55:10

https://bugs.freedesktop.org/show_bug.cgi?id=50232

almos <aaalmosss< at >gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #3 from almos <aaalmosss< at >gmail.com> 2012-05-25 10:55:10 PDT ---
I see the fix was pushed to mesa master. Closing.

[Bug 50230] offset mapping in nexuiz results in bad texturing withthe llvm compiler

2012-05-25T17:54:31

https://bugs.freedesktop.org/show_bug.cgi?id=50230

almos <aaalmosss< at >gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #11 from almos <aaalmosss< at >gmail.com> 2012-05-25 10:54:31 PDT ---
I see the fixes were pushed to mesa master. Closing.

Re: [Intel-gfx] [PATCH 02/14] drm: handle HDP and polled connectorsseparately

Adam Jackson — 2012-05-25T17:26:46

"HPD".


Here too.  Also in the subject.

- ajax

RE: [RFC] Synchronizing access to buffers shared with dma-bufbetweendrivers/devices

Tom Cooksey — 2012-05-25T16:48:58

I'm not sure I know what you mean by linear implicit dependencies?



Again, not sure what you mean here? Do you mean that userspace can
Submit a piece of work to a driver which depends on something
else happening in userpsace?



Yes, I guess this is the critical point - this approach assumes that
when a client starts using a resource, it will only do so for a finite
amount of time. If userspace wanted to participate in the scheme, we
would probably need some kind of timeout, otherwise userspace could
prevent other devices from accessing a resource.



This is the point I was trying to make. With explicit fence objects
you do have to add a new parameter, whereas with this kds implicit
approach you do not - the buffer itself becomes the sync object.




The intention was to use this mechanism for synchronizing between
drivers rather than between userspace processes, I think the userspace
access is somewhat an afterthought which will probably need some more
thought. In the example you give, app A making a sy

Re: Oops with Radeon/Uninorth on Maple

Alex Deucher — 2012-05-25T15:06:23

On Thu, May 24, 2012 at 2:18 AM, Dmitry Eremin-Solenikov
<dmitry_eremin< at >mentor.com> wrote:

For future reference, you can disable AGP by loading radeon with the
agpmode parameter set to -1, e.g., radeon.agpmode=-1

No need to edit the source.

Alex

Re: [PATCH] gpu: schedule gma500 stub driver for feature removal

Alan Cox — 2012-05-25T14:13:54

On Fri, 25 May 2012 04:53:50 +0900
Greg KH <gregkh< at >linuxfoundation.org> wrote:


If there are distros not currently ahipping the GMA500 driver or people
updating old kernels using vesa and not having the right server bits it
might cause them problems - giving it 6 months isn't a bad idea IMHO

Re: [RFC] Synchronizing access to buffers shared with dma-buf betweendrivers/devices

Pauli — 2012-05-25T13:57:28

It is easy to cause cyclic dependencies with implicit fences unless you 
are very sure that client can only cause linear implicit dependencies. 
But clients already have synchronization dependencies with userspace. 
That makes implicit synchronization possibly cause unexpected deadlocks.

Explicit synchronization is easier to debug because developer using 
explicit synchronization can track the dependencies in userspace. But of 
course that makes userspace API harder to use than API using implicitly 
synchronization.

But implicit synchronization can avoid client deadlock issues. Providing 
if client may never block "fence" from triggering in finite time when it 
is granted access. The page flip can be synchronized in that manner if 
client can't block HW from processing queued rendering.

You were talking about adding new parameter to page flip ioctl. I fail 
to see need for it because page flip already has fb object as parameter 
that should map to the implicit synchronization fence through dma_buf.


But

[Bug 43272] pink/purple line displays to the left of the screen

2012-05-25T13:42:09

https://bugzilla.kernel.org/show_bug.cgi?id=43272


Alan <alan< at >lxorguk.ukuu.org.uk> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |CLOSED
                 CC|                            |alan< at >lxorguk.ukuu.org.uk

Re: [Linaro-mm-sig] [PATCH] dma-buf: fix disabled vmap function

Sumit Semwal — 2012-05-25T09:18:55

Hi Dave,

On 25 May 2012 14:34, Dave Airlie <airlied< at >gmail.com> wrote:
I fixed this as part of rebasing while applying your vmap patch to my
for-next - so it is already in my pull request.

Best regards,
~Sumit.

RE: [PATCH 08/10] drm/radeon: replace pflip and sw_int counterswith atomics

Koenig, Christian — 2012-05-24T12:29:42

AFAIK it is, at least for the platforms we care about. But since this depends on the vertical refresh frequency even a 8bit counter should do fine.

On the other hand it was an accident that those patches hit the maillinglist in the first place, cause only the first four where supposed to be send out (my fault, sorry). This one isn't tested beside compiling, and I don't think it will work out of the box.

Cheers,
Christian.

Re: [patch] drm/udl: unlock before returning in udl_gem_mmap()

Dan Carpenter — 2012-05-24T11:00:41

This is still present in linux-next.

regards,
dan carpenter

[Bug 12882] [855GM missing DVO driver] s-video not detected

2012-05-25T12:09:29

https://bugs.freedesktop.org/show_bug.cgi?id=12882

Daniel Vetter <daniel< at >ffwll.ch> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
            Summary|[855GM missng DVO driver]   |[855GM missing DVO driver]
                   |s-video not detected        |s-video not detected

[Bug 17902] [830M missing dvo driver] need support for DVO-LVDS viana2501

2012-05-25T12:07:48

https://bugs.freedesktop.org/show_bug.cgi?id=17902

Daniel Vetter <daniel< at >ffwll.ch> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
            Summary|[830M] need support for     |[830M missing dvo driver]
                   |DVO-LVDS via na2501         |need support for DVO-LVDS
                   |                            |via na2501

[Bug 12882] [855GM missng DVO driver] s-video not detected

2012-05-25T12:06:15

https://bugs.freedesktop.org/show_bug.cgi?id=12882

Daniel Vetter <daniel< at >ffwll.ch> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
            Summary|[855GM DVO] s-video not     |[855GM missng DVO driver]
                   |detected (DVO TV not        |s-video not detected
                   |supported yet)              |

[Bug 30845] [855 missing dvo driver] DVI output unsupported

2012-05-25T12:05:59

https://bugs.freedesktop.org/show_bug.cgi?id=30845

Daniel Vetter <daniel< at >ffwll.ch> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Severity|normal                      |enhancement
           Priority|medium                      |lowest
            Summary|[855] DVI output            |[855 missing dvo driver]
                   |unsupported                 |DVI output unsupported

--- Comment #6 from Daniel Vetter <daniel< at >ffwll.ch> 2012-05-25 05:05:59 PDT ---
Downgrading all missing dvo driver bugs to feature request level.

[Bug 17902] [830M] need support for DVO-LVDS via na2501

2012-05-25T12:01:20

https://bugs.freedesktop.org/show_bug.cgi?id=17902

Chris Wilson <chris< at >chris-wilson.co.uk> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |thor< at >math.tu-berlin.de

--- Comment #71 from Chris Wilson <chris< at >chris-wilson.co.uk> 2012-05-25 05:01:20 PDT ---
*** Bug 50303 has been marked as a duplicate of this bug. ***

[RFC] Synchronizing access to buffers shared with dma-buf betweendrivers/devices

Tom Cooksey — 2012-05-25T11:08:29

Hi All,

I realise it's been a while since this was last discussed, however I'd like
to bring up kernel-side synchronization again. By kernel-side
synchronization, I mean allowing multiple drivers/devices wanting to access
the same buffer to do so without bouncing up to userspace to resolve
dependencies such as "the display controller can't start scanning out a
buffer until the GPU has finished rendering into it". As such, this is
really just an optimization which reduces latency between E.g. The GPU
finishing a rendering job and that buffer being scanned out. I appreciate
this particular example is already solved on desktop graphics cards as the
display controller and 3D core are both controlled by the same driver, so no
"generic" mechanism is needed. However on ARM SoCs, the 3D core (like an ARM
Mali) and display controller tend to be driven by separate drivers, so some
mechanism is needed to allow both drivers to synchronize their access to
buffers.

There are multiple ways synchronization can be achieved