[Intel-gfx] [PATCH 25/33] drm/i915: Use VMA for wa_ctx tracking

[Chris Wilson]

On Thu, Aug 11, 2016 at 01:53:40PM +0300, Joonas Lahtinen wrote:
> On su, 2016-08-07 at 15:45 +0100, Chris Wilson wrote:
> > @@ -2019,9 +2023,9 @@ populate_lr_context(struct i915_gem_context *ctx,
> >  			       RING_INDIRECT_CTX(engine->mmio_base), 0);
> >  		ASSIGN_CTX_REG(reg_state, CTX_RCS_INDIRECT_CTX_OFFSET,
> >  			       RING_INDIRECT_CTX_OFFSET(engine->mmio_base), 0);
> > -		if (engine->wa_ctx.obj) {
> > +		if (engine->wa_ctx.vma) {
> >  			struct i915_ctx_workarounds *wa_ctx = &engine->wa_ctx;
> > -			uint32_t ggtt_offset = i915_gem_obj_ggtt_offset(wa_ctx->obj);
> > +			u32 ggtt_offset = wa_ctx->vma->node.start;
> 
> lower_32_bits()?

I considered, I didn't to keep the changes to a minimum plus I've a
slight unease about making it seem like we don't care about the upper 32
bits.

static inline u32 i915_ggtt_offset(vma)
{
	GEM_BUG_ON(upper_32_bits(vma->node.start));
	return lower_32_bits(vma->node.start);
}

is possibly overkill but stops me feeling uneasy about the
seeming truncation. Is this something that UBSAN detects?
-Chris

-- 
Chris Wilson, Intel Open Source Technology Centre