[Intel-gfx] INSTDONE instrumentation (patch in progress)

Hi guys, I thought I'd attach this, as it is now gone 2AM and I doubt I'm going to finish it "tonight". I was hoping to elicit some initial review to suggest whether the design was sane or not. I'd originally imagined tying the profiling lifetime to the execution / completion of individual batch-

[Intel-gfx] intel_prepare_render(intel); unhelpful?

Hi guys, I was just poking around looking for somewhere quick and dirty to shove my new experimental DRM IOCTL for retrieving IDLE data from the GPU. I was looking at the various breakpoints in the debugger, and found intel_prepare_render() being called more often than I'd like. For instance, in

Re: [Intel-gfx] Fwd: 2.6.36-rc5 i915 regression

On 10/26/10 04:30, Chris Wilson wrote: > On Tue, 26 Oct 2010 01:19:50 +0200, Seblu wrote: >> - after one day, applications become graphically slow. > > You're not alone. Eric has reported that writes to a GTT mapping become > glacially slow after some time, without any distortion in the CPU profi

Re: [Intel-gfx] gpu freeze

On Sat, 2010-10-30 at 20:22 +0200, Alexey Fisher wrote: > What info should i collect if i get GPU hung? > > I currently get one, with kernel 2.6.36-07532-gbbe2e11 > it happened after i started video. Interesting is my PC worked 15 hours, > many things was done it was not first video. So fear i can

[Intel-gfx] gpu freeze

What info should i collect if i get GPU hung? I currently get one, with kernel 2.6.36-07532-gbbe2e11 it happened after i started video. Interesting is my PC worked 15 hours, many things was done it was not first video. So fear i can't reproduce or bisect it. This is first hung for some weeks or mo

[Intel-gfx] Design of a GPU profiling debug interface

I think I'll need some help with this. I'm by no means a kernel programmer, so I'm feeling my way in the dark with this. I want to design an interface so I can synchronise my GPU idle flags polling with batchbuffer execution. I'm imagining at a high level, doing something like this in my applicati

Re: [Intel-gfx] Speed boost disabling RCC clock gating ECO

On Sat, 30 Oct 2010 03:34:37 +0100, Peter Clifton wrote: > Hi guys, > > Just a note on a data-point I found here: > > sudo intel_reg_read 0x21D0 > [sudo] password for pcjc2: > 0x21D0 : 0x307 > > sudo intel_reg_write 0x21D0 0x1000207 > Value before: 0x307 > Value after: 0x207 > > > This boost