On 11/26/2014 11:44 PM, Chris Wilson wrote:
On Wed, Nov 26, 2014 at 01:53:34PM -0800, michael.h.nguyen@xxxxxxxxx wrote:
From: "Michael H. Nguyen" <michael.h.nguyen@xxxxxxxxx>
This is v5 of the series sent here:
http://lists.freedesktop.org/archives/intel-gfx/2014-November/055141.html
This version incorporates the following feedback from v4.
- 0/7 Move 'pending_read_domains |= I915_GEM_DOMAIN_COMMAND' after the
parser (danvet)
- 1/7 Move purged check inside the loop (danvet)
- 6/7 Move 'shadow_batch_obj->madv = I915_MADV_WILLNEED' inside _get
fnc (danvet)
- 7/7 Move pin/unpin calls inside i915_parse_cmds() (Chris W)
Issue: VIZ-4719
Brad Volkin (7):
drm/i915: Implement a framework for batch buffer pools
drm/i915: Use batch pools with the command parser
drm/i915: Add a batch pool debugfs file
drm/i915: Add batch pool details to i915_gem_objects debugfs
drm/i915: Use batch length instead of object size in command parser
drm/i915: Mark shadow batch buffers as purgeable
drm/i915: Tidy up execbuffer command parsing code
This still does not incorporate the feedback from the last N cycles.
Chiefly: A single cache list, madvise on creation, and squash the
framework and debugging patches into one.
Re: single cache list
OK. Found the feedback in an old rev.
Re: madvise on creation
Were you referring to this?
from
http://lists.freedesktop.org/archives/intel-gfx/2014-November/055060.htm
obj = i915_gem_obj_alloc();
i915_gem_object_get_pages(obj);
obj->madv = I915_MADV_DONTNEED;
If so, I don't understand . _get is returning obj and it'll be needed so
would expect to set 'obj->madv = I915_MADV_WILLNEED' which is the case now.
For this feedback, would appreciate it if you could provide comments or
proposed solutions in line w/ source. That would be helpful for me to
get whatever context I'm missing.
Re: Squash
OK. I didn't see this feedback. Will do.
Thanks,
-Mike
-Chris
_______________________________________________
Intel-gfx mailing list
Intel-gfx@xxxxxxxxxxxxxxxxxxxxx
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
