As the discussion in the intel-gfx mailing list, http://lists.freedesktop.org/archives/intel-gfx/2009-December/004991.html, memory self refresh need to be enabled, it can save idle power. Jesse, I test your patch (http://lists.freedesktop.org/archives/intel-gfx/2009-December/005059.html) in moblin2.1 + 2.6.32 kernel, and see the screen frozen. Would you please give it a try ? Thanks.
I tried on Asus eeepc 1000H and Samsung N130, both of them have the same frozen issue. I will try on Aspire One.
Test on acer aspire one still fails with this patch, diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index 099f420..03c897c 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -2577,7 +2577,9 @@ static void i9xx_update_wm(struct drm_device *dev, int planea_clock, srwm = total_size - sr_entries; if (srwm < 0) srwm = 1; - I915_WRITE(FW_BLC_SELF, FW_BLC_SELF_EN | (srwm & 0x3f)); + //I915_WRITE(FW_BLC_SELF, FW_BLC_SELF_EN | (srwm & 0x3f)); + I915_WRITE(FW_BLC_SELF, ((1 << 31) | (1 << 16) | (1 << 15) | (srwm & 0xff))); + POSTING_READ(FW_BLC_SELF); } DRM_DEBUG("Setting FIFO watermarks - A: %d, B: %d, C: %d, SR %d\n", I got below dmesg. [ 15.698063] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung [ 15.698087] render error detected, EIR: 0x00000000 [ 15.698100] i915: Waking up sleeping processes [ 15.698150] [drm:i915_wait_request] *ERROR* i915_wait_request returns -5 (awaiting 1443 at 1435) [ 15.698468] [drm:i915_gem_execbuffer] *ERROR* Execbuf while wedged [ 15.698933] reboot required [ 15.699360] [drm:i915_gem_execbuffer] *ERROR* Execbuf while wedged [ 15.699545] [drm:i915_gem_execbuffer] *ERROR* Execbuf while wedged [ 15.700120] [drm:i915_gem_execbuffer] *ERROR* Execbuf while wedged [ 15.700642] [drm:i915_gem_execbuffer] *ERROR* Execbuf while wedged [ 15.703888] [drm:i915_gem_execbuffer] *ERROR* Execbuf while wedged [ 15.704880] [drm:i915_gem_execbuffer] *ERROR* Execbuf while wedged [ 15.705598] render error detected, EIR: 0x00000010 [ 15.705608] page table error [ 15.705614] PGTBL_ER: 0x00000002 [ 15.705623] [drm:i915_handle_error] *ERROR* EIR stuck: 0x00000010, masking [ 15.705647] render error detected, EIR: 0x00000010 [ 15.705653] page table error [ 15.705658] PGTBL_ER: 0x00000002 [ 15.705730] reboot required [ 15.766980] [drm:i915_gem_execbuffer] *ERROR* Execbuf while wedged [ 15.767214] [drm:i915_gem_execbuffer] *ERROR* Execbuf while wedged [ 15.767272] [drm:i915_gem_execbuffer] *ERROR* Execbuf while wedged [ 15.767304] [drm:i915_gem_execbuffer] *ERROR* Execbuf while wedged [ 15.767347] [drm:i915_gem_execbuffer] *ERROR* Execbuf while wedged [ 15.767388] [drm:i915_gem_execbuffer] *ERROR* Execbuf while wedged [ 15.772580] [drm:i915_gem_execbuffer] *ERROR* Execbuf while wedged [ 15.772735] [drm:i915_gem_execbuffer] *ERROR* Execbuf while wedged [ 15.773062] [drm:i915_gem_execbuffer] *ERROR* Execbuf while wedged [ 15.773156] [drm:i915_gem_execbuffer] *ERROR* Execbuf while wedged [ 15.778592] [drm:i915_gem_execbuffer] *ERROR* Execbuf while wedged
Jesse, I retest your patch on latest 2.6.33-rc3, still met same problem.
We should probably go ahead with your patch; I replied to Eric that it's ok to apply it, you should ping him again.
You fixed this one, thanks: commit 72ddbd39f39b3cc5f1f661f97e67f2a497c4d57a Author: Li Peng <peng.li@linux.intel.com> Date: Tue Jan 19 21:32:57 2010 +0800
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.