By recording the location of every request in the ringbuffer, we know
that in order to retire the request the GPU must have finished reading
it and so the GPU head is now beyond the tail of the request. We can
therefore provide a conservative estimate of where the GPU is reading
from in order to avoid having to read back the ring buffer registers
when polling for space upon starting a new write into the ringbuffer.
A secondary effect is that this allows us to convert
intel_ring_buffer_wait() to use i915_wait_request() and so consolidate
upon the single function to handle the complicated task of waiting upon
the GPU. A necessary precaution is that we need to make that wait
uninterruptible to match the existing conditions as all the callers of
intel_ring_begin() have not been audited to handle ERESTARTSYS
correctly.
By using a conservative estimate for the head, and always processing all
outstanding requests first, we prevent a race condition between using
the estimate and direct reads of I915_RING_HEAD which could result in
the value of the head going backwards, and the tail overflowing once
again. We are also careful to mark any request that we skip over in
order to free space in ring as consumed which provides a
self-consistency check.
Given sufficient abuse, such as a set of unthrottled GPU bound
cairo-traces, avoiding the use of I915_RING_HEAD gives a 10-20% boost on
Sandy Bridge (i5-2520m):
firefox-paintball 18927ms -> 15646ms: 1.21x speedup
firefox-fishtank 12563ms -> 11278ms: 1.11x speedup
which is a mild consolation for the performance those traces achieved from
exploiting the buggy autoreported head.
v2: Add a few more comments and make request->tail a conservative
estimate as suggested by Daniel Vetter.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
[danvet: resolve conflicts with retirement defering and the lack of
the autoreport head removal (that will go in through -fixes).]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
************************************************************
* For the very latest on DRI development, please see: *
* http://dri.freedesktop.org/ *
************************************************************
The Direct Rendering Manager (drm) is a device-independent kernel-level
device driver that provides support for the XFree86 Direct Rendering
Infrastructure (DRI).
The DRM supports the Direct Rendering Infrastructure (DRI) in four major
ways:
1. The DRM provides synchronized access to the graphics hardware via
the use of an optimized two-tiered lock.
2. The DRM enforces the DRI security policy for access to the graphics
hardware by only allowing authenticated X11 clients access to
restricted regions of memory.
3. The DRM provides a generic DMA engine, complete with multiple
queues and the ability to detect the need for an OpenGL context
switch.
4. The DRM is extensible via the use of small device-specific modules
that rely extensively on the API exported by the DRM module.
Documentation on the DRI is available from:
http://dri.freedesktop.org/wiki/Documentation
http://sourceforge.net/project/showfiles.php?group_id=387
http://dri.sourceforge.net/doc/
For specific information about kernel-level support, see:
The Direct Rendering Manager, Kernel Support for the Direct Rendering
Infrastructure
http://dri.sourceforge.net/doc/drm_low_level.html
Hardware Locking for the Direct Rendering Infrastructure
http://dri.sourceforge.net/doc/hardware_locking_low_level.html
A Security Analysis of the Direct Rendering Infrastructure
http://dri.sourceforge.net/doc/security_low_level.html