[Intel-gfx] [PATCH 4/6] drm/i915: Serialize all register access

2013-07-16 Thread Chris Wilson
In theory, the different register blocks were meant to be only ever touched when holding either the struct_mutex, mode_config.lock or even a specific localised lock. This does not seem to be the case, and the hardware reacts extremely badly if we attempt to concurrently access two registers within

Re: [Intel-gfx] [PATCH 4/6] drm/i915: Serialize all register access

2013-07-16 Thread Chris Wilson
On Sun, Jul 14, 2013 at 01:23:52PM -0700, Ben Widawsky wrote: > On Fri, Jul 12, 2013 at 06:08:25PM +0100, Chris Wilson wrote: > > In theory, the different register blocks were meant to be only ever > > touched when holding either the struct_mutex, mode_config.lock or even a > > specific localised l

Re: [Intel-gfx] [PATCH 4/6] drm/i915: Serialize all register access

2013-07-15 Thread Ben Widawsky
On Fri, Jul 12, 2013 at 09:52:46PM +0100, Chris Wilson wrote: > On Fri, Jul 12, 2013 at 01:37:28PM -0700, Ben Widawsky wrote: > > On Fri, Jul 12, 2013 at 03:02:34PM +0100, Chris Wilson wrote: > > > In theory, the different register blocks were meant to be only ever > > > touched when holding either

Re: [Intel-gfx] [PATCH 4/6] drm/i915: Serialize all register access

2013-07-14 Thread Ben Widawsky
On Fri, Jul 12, 2013 at 06:08:25PM +0100, Chris Wilson wrote: > In theory, the different register blocks were meant to be only ever > touched when holding either the struct_mutex, mode_config.lock or even a > specific localised lock. This does not seem to be the case, and the > hardware reacts extr

Re: [Intel-gfx] [PATCH 4/6] drm/i915: Serialize all register access

2013-07-12 Thread Chris Wilson
On Fri, Jul 12, 2013 at 01:37:28PM -0700, Ben Widawsky wrote: > On Fri, Jul 12, 2013 at 03:02:34PM +0100, Chris Wilson wrote: > > In theory, the different register blocks were meant to be only ever > > touched when holding either the struct_mutex, mode_config.lock or even a > > specific localised l

Re: [Intel-gfx] [PATCH 4/6] drm/i915: Serialize all register access

2013-07-12 Thread Ben Widawsky
On Fri, Jul 12, 2013 at 03:02:34PM +0100, Chris Wilson wrote: > In theory, the different register blocks were meant to be only ever > touched when holding either the struct_mutex, mode_config.lock or even a > specific localised lock. This does not seem to be the case, and the > hardware reacts extr

[Intel-gfx] [PATCH 4/6] drm/i915: Serialize all register access

2013-07-12 Thread Chris Wilson
In theory, the different register blocks were meant to be only ever touched when holding either the struct_mutex, mode_config.lock or even a specific localised lock. This does not seem to be the case, and the hardware reacts extremely badly if we attempt to concurrently access two registers within

[Intel-gfx] [PATCH 4/6] drm/i915: Serialize all register access

2013-07-12 Thread Chris Wilson
In theory, the different register blocks were meant to be only ever touched when holding either the struct_mutex, mode_config.lock or even a specific localised lock. This does not seem to be the case, and the hardware reacts extremely badly if we attempt to concurrently access two registers within

[Intel-gfx] [PATCH 4/6] drm/i915: Serialize all register access

2013-07-12 Thread Chris Wilson
In theory, the different register blocks were meant to be only ever touched when holding either the struct_mutex, mode_config.lock or even a specific localised lock. This does not seem to be the case, and the hardware reacts extremely badly if we attempt to concurrently access two registers within