[PATCH 2/2] drm/radeon: implement dynamic PTs allocation via SA

Hi Michel. 2012/9/27 Michel D?nzer : > On Die, 2012-09-25 at 18:44 +0400, Dmitry Cherkasov wrote: >> make dynamic allocation of page tables on demand in radeon_vm_update_pte >> >> Signed-off-by: Dmitry Cherkasov > > This change caused many piglit tests to fail for me on SI with > > radeon

Re: [PATCH 2/2] drm/radeon: implement dynamic PTs allocation via SA

Hi Michel. 2012/9/27 Michel Dänzer : > On Die, 2012-09-25 at 18:44 +0400, Dmitry Cherkasov wrote: >> make dynamic allocation of page tables on demand in radeon_vm_update_pte >> >> Signed-off-by: Dmitry Cherkasov > > This change caused many piglit tests to fail for me on SI with > > radeon

Re: [PATCH 2/2] drm/radeon: implement dynamic PTs allocation via SA

I've found introduced error in radeon_vm_free_pt, so will fix in next version of this patch. This one: --- a/drivers/gpu/drm/radeon/radeon_gart.c +++ b/drivers/gpu/drm/radeon/radeon_gart.c @@ -505,15 +505,6 @@ static void radeon_vm_free_pt(struct radeon_device *rdev, int driver_table_ent

[PATCH 2/2] drm/radeon: implement dynamic PTs allocation via SA

I've found introduced error in radeon_vm_free_pt, so will fix in next version of this patch. This one: --- a/drivers/gpu/drm/radeon/radeon_gart.c +++ b/drivers/gpu/drm/radeon/radeon_gart.c @@ -505,15 +505,6 @@ static void radeon_vm_free_pt(struct radeon_device *rdev, int driver_table_ent

Re: [PATCH] Add 2-level GPUVM pagetables support to radeon driver.

Michel, > E.g. reflect from Mesa Demos wasn't able to sustain 60 fps in > fullscreen. > I've tried reflect with my patches on cayman and it sustained 60 fps in maximized window @ 1680x1080 I'll test when i have tahiti machine ready. -- With best regards, Dmitry

[PATCH] Add 2-level GPUVM pagetables support to radeon driver.

Michel, > E.g. reflect from Mesa Demos wasn't able to sustain 60 fps in > fullscreen. > I've tried reflect with my patches on cayman and it sustained 60 fps in maximized window @ 1680x1080 I'll test when i have tahiti machine ready. -- With best regards, Dmitry

[PATCH] Add 2-level GPUVM pagetables support to radeon driver.

> With that fixed, it seems to work on SI, but seems to slow things down > significantly. Have you noticed that as well? Any idea what might be the > reason? > Thanks I'll put it up to the patch. I had everything running slow on cayman when having lots of debugging output, removing it fixed the sl

Re: [PATCH] Add 2-level GPUVM pagetables support to radeon driver.

> With that fixed, it seems to work on SI, but seems to slow things down > significantly. Have you noticed that as well? Any idea what might be the > reason? > Thanks I'll put it up to the patch. I had everything running slow on cayman when having lots of debugging output, removing it fixed the sl

Re: [PATCH] Add 2-level GPUVM pagetables support to radeon driver.

Christian, > as I already noted in our internal discussion, the first step to > hierarchical page table support should be to cleanup the set_page interface. > Please see the attached patch, it does exactly this. I suggest that you > rebase on it and try to don't touch the chipset specific code in

[PATCH] Add 2-level GPUVM pagetables support to radeon driver.

Christian, > as I already noted in our internal discussion, the first step to > hierarchical page table support should be to cleanup the set_page interface. > Please see the attached patch, it does exactly this. I suggest that you > rebase on it and try to don't touch the chipset specific code in