[Intel-gfx] [PATCH 2/6] drm/i915: simplify testing for the global default context
david.s.gordon at intel.com
Tue Dec 22 03:35:08 PST 2015
On 22/12/15 09:05, Chris Wilson wrote:
> On Mon, Dec 21, 2015 at 04:04:41PM +0000, Dave Gordon wrote:
>> There are quite a number of places where the driver tests whether a
>> given context is or is not the global default context, usually by
>> checking whether an engine's default_pointer points to the context. Now
>> that we have a 'is_global_default' flag in the context itself, these can
>> be rewritten to use it. This makes the logic more obvious, and usually
>> saves at least one memory reference.
> All these places do not need to exist. Please just fix execlists.
The patchset "to fix execlists" in one go would be too large to be
accepted here and would take too long to develop, given the nature of
the moving target. Ergo, we can fix execlists only by taking every
opportunity to move towards a clearer design, even though each step
fails to "fix execlists" on its own.
We therefore have to judge each patch on the basis of "does it make
things better or worse", not "does it fix all known problems". IMHO,
this patch (and the rest of the set) are small steps towards a better
design, and you should therefore support their adoption, unless of
course you think it actually makes things worse - in which case, point
out what's worse and I'll change it.
More information about the Intel-gfx