With the new callback, Jason's newer batch decoder infrastructure
should be able to do just as well as the old open coded INTEL_DEBUG=bat
handling, with much less code. If there are any limitations, we'd like
to improve the common code rather than doing one-off hacks here.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Given an arbitrary batch, we don't always know what the size of certain
things are, such as how many entries are in a binding table. But it's
easy for the driver to track that information, so with a simple callback
we can calculate this correctly for INTEL_DEBUG=bat.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Making these part of libintel_common allows us to use them in the DRI
driver. The standalone tool binaries already link against the common
library, too, so it's no harder for them.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
This unfortunately makes it malloc/realloc on every new batch, rather
than once at startup. But it ensures that the shadow buffer's size will
absolutely match the BO size. Otherwise, as we tune BATCH_SZ/STATE_SZ
or bufmgr cache bucket sizes, we may get a BO size that's rounded up,
and fail to allocate the shadow buffer large enough.
This doesn't fix any bugs today, as BATCH_SZ/STATE_SZ are the size of
a cache bucket, but it's better to be safe than sorry.
Reported-by: James Xiong <james.xiong@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
The gen_field_iterator only iterates the fields of a given gen_group.
If we want to iterate the fields of another gen_group contained as
field, we need to do it manually.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Struct fields might span several dwords, but iter_dword is incremented
up to the last dword of the current field before we print out the
struct's fields. We can't use iter_dword for computing the offset into
the pointer of data to decode.
v2: Fix displayed offset number (Ken)
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
<register> & <struct> elements always have fixed length. The
get_length() method implies that we're dealing with an instruction in
which the length is encoded into the variable data but the field
iterator uses it without checking what kind of gen_group it is dealing
with.
Let's make get_length() report the correct length regardless of the
gen_group (register, struct or instruction).
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
The Vertex Elements are now:
* VE 1: <BaseVertex/firstvertex, BaseInstance, VertexID, InstanceID>
* VE 2: <DrawID, is-indexed-draw, 0, 0>
VE1 is it kept as it was before, VE2 additionally contains the new
system value.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
This VS system value contains if the draw command used to start the
rendering was an indexed draw command or a non-indexed one
(~0/0 respectively). Useful to calculate the gl_BaseVertex as:
(SYSTEM_VALUE_IS_INDEXED_DRAW & SYSTEM_VALUE_FIRST_VERTEX).
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
As the implementation is conservative, we can now enable it
by default. It can be disabled with RADV_DEBUG=nooutoforder.
Don't expect much more than 1% of improvements, but the gain
seems consistent.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Both the internal documentation and the results of testing this in the
CI suggest that this is unnecessary. Add the fixes tag because this
reduces an internal benchmark's startup time by about 17 seconds
(reported by Eero).
Fixes: 710b1d2e66 "i965/tex_image: Flush certain subnormal ASTC channel values"
Tested-by: Eero Tamminen <eero.t.tamminen@intel.com>
Acked-by: Kenneth Graunke <kenneth@whitecape.org>
This is needed for gfx9 and later for all fmask surface index.
(Mentioned by Marek on irc)
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Subpixel precision bias, dilation and the post-snap mode are supported on
GM200 and newer. The pre-snap mode is supported for triangle primitives on
GP100.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Although the specs are written against compatibility GL 4.3 and allows core
profile and GLES2+, it is exposed for GL 1.0+ and GLES1 and GLES2+.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Brian Paul <brianp@vmware.com>
To silence warnings about unhandled switch values.
Untested otherwise.
v2: move the INT/UINT8 cases after the INT/UINT16 cases, per Eric.
Reviewed-by: Eric Anholt <eric@anholt.net>
With this we should have no passes in src/compiler/nir with any
dependencies on headers from core GL Mesa.
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
A bo's ref_count was not being initialized when imported from an fd.
Therefore, we would fail to free the resource during VkFreeMemory().
This patch fixes applications like hifi VR in threaded mode, which
perform frequent imports/releases of IPC shared memory.
Signed-off-by: Andres Rodriguez <andresx7@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
CC: 18.0 18.1 <mesa-stable@lists.freedesktop.org>
Similar to the transformation applied to linear_to_ytiled, also align
each readback from the ytiled source to a cacheline (i.e. transfer a
whole cacheline from the source before moving on to the next column).
This will allow us to utilize movntqda (_mm_stream_si128) in a
subsequent patch to obtain near WB readback performance when accessing
the uncached ytiled memory, an order of magnitude improvement.
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
When mapping a region of the mipmap_tree, record which complementary
method to use to unmap it afterwards. By doing so we can avoid
duplicating the decision tree used when mapping and thereby eliminate
trivial errors that can be introduced if the two if-chains become out of
sync.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Scott D Phillips <scott.d.phillips@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
If loading 64-bit vec3 values, a 4 component load would be followed
by a 2 component load and the resulting shuffle would fail as it
requires 2 4 components. This just expands the second results
vector out to 4 components.
This fixes 100 CTS tests:
dEQP-VK.spirv_assembly.type.vec3.*64*
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
We want to flag EXEC_OBJECT_CAPTURE, but we ought to preserve any
existing kflags. Today, there are none (as the program cache doesn't
support 48-bit addressing), but once we start using softpin, we'll
need to preserve EXEC_OBJECT_PINNED.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
We were trying to mark batch buffers with EXEC_OBJECT_CAPTURE, and
accidentally stomped EXEC_OBJECT_SUPPORTS_48B_ADDRESS in the process.
There's no reason to restrict batch buffers to the lower 4GB.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
The previous logic of the supports_48b_addresses wasn't actually
checking if i915.ko was running with full_48bit_ppgtt. The ENOENT
it was checking for was actually coming from the invalid context
id provided in the test execbuffer. There is no path in the
kernel driver where the presence of
EXEC_OBJECT_SUPPORTS_48B_ADDRESS leads to an error.
Instead, check the default context's GTT_SIZE param for a value
greater than 4 GiB
v2 (Ken): Fix in i965 as well.
v3 Check GTT_SIZE instead of HAS_ALIASING_PPGTT (Chris Wilson)
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
The blit here involves scaling since it's copying from I8 format to R8G8 format.
Half of source will be filtered out with PIPE_TEX_FILTER_NEAREST instruction, it
looks that GPU always uses the second half as source. Currently we use "1" as
the start point of x for R, then causing 1 source pixel of U component shift to
right. So "-1" should be the start point for U component.
Cc: 18.0 18.1 <mesa-stable@lists.freedesktop.org>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Lionel Landwerlin