This is around 3x or so speedup, since we would read wide rows at a time, and
clflush each tile 8 times as a result. We'll want code related to this anyway
when we do fault-based per-page clflushing for sw fallbacks.
This increases overhead for the large-readpixels case due to the repeated
page cache accessing, but greatly reduces overhead for the small-readpixels
case.
One of our systems has been returning 0xffffffff from all MCHBAR reads, which
means we'll need to figure out why, or add an alternate detection method.
Various chips have exciting interactions between the CPU and the GPU's
different ways of accessing interleaved memory, so we need some kernel
assistance in determining how it works.
Only fully tested on GM965 so far.
Clean up queues, free objects. On the next entervt, unmark the hardware to
let the user try again (presumably after resetting the chip). Someday we'll
automatically recover...
While waiting for the hardware to idle on leavevt or lastclose, poll
for the sync sequence number instead of waiting for an interrupt. This
allows the code to bail if the hardware hangs for some reason. Also, this
avoids issues with signals as the exisiting wait function is interruptible.
find_or_create_page doesn't quite set up pages correctly; any newly created
pages aren't hooked into the shmem object quite right; user space mmaps of
those pages end up mapping pages full of zeros which then get written to the
real pages inappropriately. This patch requires that the kernel export
shmem_getpage.
When a software fallback has completed, usermode must notify the kernel so
that any scanout buffers can be synchronized. This ioctl should be called
whenever a fallback completes to flush CPU and chipset caches.
This fixes registration when MSI is set up after the stub function fills in
dev->irq. Otherwise /proc/interrupts would report attachment to the fasteoi
interrupt. dev->irq is still exposed (and updated at IRQ setup)
for the drivers that use it for whatever reason.
In leavevt_ioctl, queue an MI_FLUSH and then block waiting for it to
complete. This will empty the active and flushing lists. That leaves only
the inactive list to evict.
Pin/unpin need to know whether to remove/add objects from the inactive list,
inactive objects cannot be in any GPU write domain as those would be on the
flushing list instead. However, inactive objects may be in the CPU write
domain.
Now that gem_object_unbind waits for rendering to complete, objects should
not be active when they are being pulled from the GTT. BUG_ON if this is
broken.
Record the last execbuffer sequence for each client.
Record that sequence in the throttle ioctl as the 'throttle sequence'.
Wait for the last throttle sequence in the throttle ioctl.
When i915_wait_request clears object from the active list, it may end up
freeing them and not moving them to the inactive list. This ends up
unbinding objects from the GTT without there ever being new objects visible
to i915_gem_evict_something on the inactive list. As the only success
condition required the presence of objects on the inactive list, this would
falsely assume that no GTT space had been made available, and end up
returning -ENOMEM to the application.
We want request retirement to occur about once a second when the request
queue is non-empty. This was done with a timer that queued a work_struct,
using a delayed_work instead makes a lot more sense.
This is the create (may want location flags), pread/pwrite/mmap
(performance tuning hints), and set_domain (will 32 bits be enough for
everyone?) ioctls. Left in the generic set are just flink/open/close.
The 2D driver must be updated for this change, and API but not ABI is broken
for 3D. The driver version is bumped to mark this.
This requires that the X Server use the execbuf interface for buffer
submission, as it no longer has direct access to the ring. This is
therefore a flag day for the gem interface.
This also adds enter/leavevt ioctls for use by the X Server. These would
get stubbed out in a modesetting implementation, but are required while
in an environment where the device's state is only managed by the DRM while
X has the VT.
Eric Anholt
Ian Romanick
Keith Packard