The media spin utility is derived from media fill. The purpose
is to create a simple means to keep the render engine (media
pipeline) busy for a controlled amount of time. It does so by
emitting a batch with a single execution thread that spins in
a tight loop the requested number of times. Each spin increments
a counter whose final 32-bit value is written to the destination
buffer on completion for checking. The implementation supports
Gen8, Gen8lp, and Gen9.
v2: Apply the recommendations of igt.cocci.
Signed-off-by: Jeff McGee <jeff.mcgee@intel.com>
Tested-by: Lei Liu <lei.a.liu@intel.com>
Signed-off-by: Thomas Wood <thomas.wood@intel.com>
So we can use it with bare kernel types, without going through libdrm
bos.
v2: Don't forget the object handle. (Tvrtko)
Correct surface pitch calculation. (Tvrtko)
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
v2: Adjust for BB handling changes. (Tvrtko Ursulin)
Correct XY_FAST_COPY_DST_TILING_Yf. (Tvrtko Ursulin)
v3: New tiling modes are not defined in the kernel any more. (Tvrtko Ursulin)
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
This is to add fill operation using GPGPU pipeline which is similar to
current media fill. This can be used to simply verify GPGPU pipeline
and help to enable it on newer HW, currently it works on Gen7 only and
will add support on later platform.
Now this sets very simply thread group dispatch for one thread per
thread group on SIMD16 dispatch. So the fill shader just uses thread
group ID for buffer offset.
v2: No new fill func typedef but adapt to igt_fillfunc_t.
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
Since relocations are variable size, depending upon generation, it is
easier to handle the resizing of the batch request inside the
BEGIN_BATCH macro. This still leaves us with having to resize commands
in a few places - which still need adaption for gen8+.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
This reveal that quite a few locations were writing relocation offsets
but only allowing for 32 bit addresses. To reveal such places in active
tests, we also now double check that we do not use more batch space than
declared.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Only include what the header itself needs. The big fish here is
intel-gpu-tools.h. More will follow.
One ugly thing removed here is the duplicated GEN6_TD_CTL #define, one
of which was broken.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Only the igt core and non-test tools should have asserts to catch
internal errors, tests and helper libraries should all user igt_asert
instead.
Fix things up where assert instead of igt_assert was used.
One tiny step towards header sanity.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Also fix a fumble in the documentation for intel_blt_copy. One thing
we might want to do is unify the parameter ordering here a bit ...
Again gtkdoc fails to pick up the documentation for struct igt_buf :(
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
It's unused. There's still num_tiles getting in the way of things,
but that is used by gem_stress a bit.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Step one to properly namespace the rendercpy/mediafill functions. Als
give the buf_height/width helpers a proper igt_ prefix.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
And move the public interfaces into intel_batchbuffer.[hc].
A bit messy since we are fairly inconsistent with our header #include
handling.
Also exclude rendercopy.h from the documentation.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Instead of a width/height combination. Since I've been lazy with the
math this now only accepts page-aligned copy operations, but that's
all we need really.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
- I didn't bother to document the BLIT batch header macros - I'm not
too happy with them and they're fairly obscure.
- intel_copy_bo could use some interface love, added a FIXME comment
for now.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
I've opted to not use the PIPE_CONTROL w/a for now. I am unclear if it
is actually required (the test does pass).
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
As these files are compiled for every test, the warnings cluttered
the Android build completely.
v2: As suggested by Daniel Vetter, drop some of the fixes and fix
the compilation flags instead.
Signed-off-by: Oscar Mateo <oscar.mateo@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
This provides a macro that allows us to update all the arbitrary blit
commands we have stuck throughout the code. It assumes we don't actually
use 64b relocs (which is currently true). This also allows us to easily find
all the areas we need to update later when we really use the upper dword.
This block was done mostly with a sed job, and represents the easier
in test blit implementations.
v2 by Oscar: s/OUT_BATCH/BEGIN_BATCH in BLIT_COPY_BATCH_START
CC: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Oscar Mateo <oscar.mateo@intel.com>
Needed by an upcoming patch fixing kms_render's blits for fbs that have
other than 32bpp formats.
Based on the corresponding SNA function.
v2:
- fix random ordering of src, dst parameters (Chris)
- pass pitch in bytes rather than pixels (Chris)
Signed-off-by: Imre Deak <imre.deak@intel.com>
This wraps libdrm functionality to exec with contexts. This patch
shouldn't be applied until libdrm for contexts is updated.
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Using BEGIN_BATCH can lead to a nice inf recursion through require_space
-> flush_batch -> BEGIN_BATCH.
Also fix things up to always require BATCH_RESERVED. We need 2 dwords
for the gen5 workaround and 2 dwords for MI_BB_END.
In this case, the code already has a comment about the problem. We don't
need to flood the build with this less than useful message.
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
After full-gtt, gem_tiled_blits doesn't allocate enough to force
eviction. So query the total aperture and accommodate.
Also introduce a similar test that utilizes fences rather than
use the BLT to perform the tiling and detiling.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
A few of the tools can be performed post-mortem from a different system,
so it is useful to be able to compile those tools on those foreign
systems. Obviously, any program to interact with the PCI device or talk
to GEM will fail on a non-Intel system.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
