Now that we're using load/store_reg intrinsics, the previous checks for
registers aren't what we want. Instead, we need to be looking for a mov
or vec where both the destination and a source are load/store_reg with
matching decl_reg.
Fixes: b8209d69ff ("intel/fs: Add support for new-style registers")
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24310>
CID 1528164 (#1 of 1): Unintentional integer overflow (OVERFLOW_BEFORE_WIDEN)
overflow_before_widen: Potentially overflowing expression
pool->n_passes * pool->khr_perf_preamble_stride with type
unsigned int (32 bits, unsigned) is evaluated using 32-bit arithmetic,
and then used in a context that expects an expression of type uint64_t (64 bits, unsigned).
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: José Roberto de Souza <jose.souza@intel.com>
Signed-off-by: Illia Polishchuk <illia.a.polishchuk@globallogic.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20893>
When a default value of a struct's field, which is in the
higher half of the first dword, is specified in a gen xml
file, setting op mask makes decoder treat the field as a
header (intel_field_is_header()). As a result, it won't
output the field in batch dump. This is not a common case
but can happen once a gen xml file includes such fields.
The op mask is only meaningful to instructions, so we fix
the above issue by not setting op mask of structs (also
registers).
Signed-off-by: Jianxun Zhang <jianxun.zhang@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24268>
ISL's state-machine of CCS_D describes full resolves as leaving the aux
buffer in the pass-through state. Hardware doesn't behave this way on
gfx8 however. On that platform, full resolves transition the aux buffer
to the resolved state. This was verified by dumping the CCS before and
after a full resolve on BDW (gfx7 is simply assumed to behave the same).
Ambiguate after resolving to match driver expectations.
Prevents iris from failing piglit's fcc-write-after-clear on BDW with a
future patch which relies on fast-clear encodings being removed after a
resolve. The avoided failure is:
Testing implicit read of partial block UNORM -> SNORM
Probe color at (0,1,0)
Expected: 1.000000 1.000000 1.000000 1.000000
Observed: 0.000000 0.000000 0.000000 0.000000
Cc: mesa-stable
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23676>
The field is no longer consumed by brw_complie_* and is instead handled
directly by the crocus driver. Therefore, it's safe to leave it zero
and not even bother setting it. This removes our reliance on the
SWIZZLE_* macros in prog_instructions.h.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24288>
Both per-primitive and per-vertex space is allocated in MUE in 8 dword
chunks and those 8-dword chunks (granularity of
3DSTATE_SBE_MESH.Per[Primitive|Vertex]URBEntryOutputReadLength)
are passed to fragment shaders as inputs (either non-interpolated
for per-primitive and flat vertex attributes or interpolated
for non-flat vertex attributes).
Some attributes have a special meaning and must be placed in separate
8/16-dword slot called Primitive Header or Vertex Header.
Primitive Header contains 4 such attributes (Cull Primitive,
ViewportIndex, RTAIndex, CPS), leaving 4 dwords (the rest of 8-dword
slot) potentially unused.
Vertex Header is similar - it starts with 3 unused dwords, 1 dword for
Point Size (but if we declare that shader doesn't produce Point Size
then we can reuse it), followed by 4 dwords for Position and optionally
8 dwords for clip distances.
This means we have an interesting optimization problem - we can put
some user attributes into holes in Primitive and Vertex Headers, which
may lead to smaller MUE size and potentially more mesh threads running
in parallel, but we have to be careful to use those holes only when
we need it, otherwise we could force HW to pass too much data to
fragment shader.
Example 1:
Let's assume that Primitive Header is enabled and user defined
12 dwords of per-primitive attributes.
Without packing we would consume 8 + ALIGN(12, 8) = 24 dwords of
MUE space and pass ALIGN(12, 8) = 16 dwords to fragment shader.
With packing, we'll consume 4 + 4 + ALIGN(12 - 4, 8) = 16 dwords of
MUE space and pass ALIGN(4, 8) + ALIGN(12 - 4, 8) = 16 dwords to
fragment shader.
16/16 is better than 24/16, so packing makes sense.
Example 2:
Now let's assume that Primitive Header is enabled and user defined
16 dwords of per-primitive attributes.
Without packing we would consume 8 + ALIGN(16, 8) = 24 dwords of
MUE space and pass ALIGN(16, 16) = 16 dwords to fragment shader.
With packing, we'll consume 4 + 4 + ALIGN(16 - 4, 8) = 24 dwords of
MUE space and pass ALIGN(4, 8) + ALIGN(16 - 4, 8) = 24 dwords to
fragment shader.
24/24 is worse than 24/16, so packing doesn't make sense.
This change doesn't affect vk_meshlet_cadscene in default configuration,
but it speeds it up by up to 25% with "-extraattributes N", where
N is some small value divisible by 2 (by default N == 1) and we
are bound by URB size.
Reviewed-by: Ivan Briano <ivan.briano@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20407>
Instead of using 4 dwords for each output slot, use only the amount
of memory actually needed by each variable.
There are some complications from this "obvious" idea:
- flat and non-flat variables can't be merged into the same vec4 slot,
because flat inputs mask has vec4 stride
- multi-slot variables can have different layout:
float[N] requires N 1-dword slots, but
i64vec3 requires 1 fully occupied 4-dword slot followed by 2-dword slot
- some output variables occur both in single-channel/component split
and combined variants
- crossing vec4 boundary requires generating more writes, so avoiding them
if possible is beneficial
This patch fixes some issues with arrays in per-vertex and per-primitive data
(func.mesh.ext.outputs.*.indirect_array.q0 in crucible)
and by reduction in single MUE size it allows spawning more threads at
the same time.
Note: this patch doesn't improve vk_meshlet_cadscene performance because
default layout is already optimal enough.
Reviewed-by: Ivan Briano <ivan.briano@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20407>
anv no longer needs to track if the CFE state is valid since we ensure
that the state is valid at pipeline creation time.
Signed-off-by: Rohan Garg <rohan.garg@intel.com>
Reviewed-by: José Roberto de Souza <jose.souza@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23934>
Fixes: 90a39cac87 ("intel/blorp: Emit compute program based on BLORP_BATCH_USE_COMPUTE")
Signed-off-by: Rohan Garg <rohan.garg@intel.com>
Reviewed-by: José Roberto de Souza <jose.souza@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23934>
Since !24096 landed, we can just use ycbcr_info to get information
of an image of the P010 format.
Signed-off-by: Hyunjun Ko <zzoon@igalia.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24265>
This enables the DG2 render compression modifier in anv. When I tested
this against vkcube, I observed that the full resolve which happened at
the end of every frame was converted to a partial resolve, allowing the
framebuffer to retain compression.
According to Caleb Callaway's testing, enabling this modifier positively
impacts the FPS of the following game benchmarks:
- Strange Brigade.vk-g6 +12.78%
- Strange Brigade.dx12vk-g6 + 9.33%
- Shadow of the Tomb Raider.vk-g6-lx + 2.37%
- Dota 2 (replay Jul 2020).vk-g6 + 2.28%
Thanks to Felix Degrood for pointing out that Strange Brigade would
benefit from this optimization.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24120>
Replace the aux_usage field with two booleans: one for render
compression and one for media compression.
This more accurately describes how CCS_E is used on gfx12. On those
platforms, the FCV feature may be enabled or disabled, but ISL's
modifier table has been using the FCV aux-usage for every gfx12 render
compression modifier. Instead, set the newly-added render compression
boolean to true.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24120>
We're going to enable the DG2 modifier. Account for the reduced plane
count that exists with it.
Also add an assert to make it clearer that the aux in use is CCS.
Otherwise, it may not be obvious because of the generic compression
names being used here.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24120>
We're going to enable the DG2_RC_CCS modifier in anv. Add and use this
function to prepare for the new plane count that comes with that
modifier.
iris is left alone for now because it supports more modifiers than
isl_drm_modifier_get_score is aware of.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24120>
Before this change, anv_get_image_format_features2 reported support for
ASTC formats with any modifier (even those not supported by anv). But,
we didn't intend to support that compressed image format with modifiers.
With this change, the format feature function reports no support for
modifiers on ASTC-formatted images.
This prevents the next patch from causing assertion failures due to
unsupported modifiers.
Fixes: 355f318843 ("anv: Allow transfer-only linear ASTC images")
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24120>
Save lowest dword of shader source sha1 in pipeline object for use
later as hash for uniquely identifying shader in debug outputs.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23942>
Use a struct for various common parameters rather than per stage
structure or arguments to stage specific entrypoints.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Felix DeGrood <felix.j.degrood@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23942>
Xe KMD only requires userptr to be bound to VM, so here reusing
workaround_bo->gem_handle id to all userptr bos in Xe version of
gem_create_userptr(). The Xe version of gem_close() will make sure
that workaround_bo->gem_handle is not closed when userptr bos
are closed.
With the same gem_handle for all userptr bos, it was also necessary
skip the anv_device_lookup_bo() and manually allocate memory to store
anv_bo in host heap memory, what lead to some small changes in
anv_device_release_bo() as well.
The remaining changes are the support to VM bind userptr bos and the
gem_vm_bind() call in anv_device_import_bo_from_host_ptr().
Fixes: dEQP-VK.memory.external_memory_host*
Signed-off-by: José Roberto de Souza <jose.souza@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23787>
struct anv_bo will be needed in the next patch to properly handle
closure of userptr bos.
Signed-off-by: José Roberto de Souza <jose.souza@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23787>
Xe support of userptr will be implemented in the next patch,
this is just moving the i915 and stub functions to KMD backend.
Signed-off-by: José Roberto de Souza <jose.souza@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23787>
Each entry is a uint64_t, L2 and L1 maps 12 bits so:
(1 << 12) = 4096
sizeof(uint64_t) = 8
4096 * 8 = 32768 = 32K
Same value but easier to understand.
Signed-off-by: José Roberto de Souza <jose.souza@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24077>
The expression 'l1_gpu_addr + l1_index * sizeof(*l1_map)' could cause
bit 47 to be set so it needs to be converted to canonical.
Signed-off-by: José Roberto de Souza <jose.souza@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24077>
The bits above index 47 in l1 entry are used to define format,
depth and luminance.
aux_address is formated as canonical, so bits above 47 could all be
set to 1 causing wrong values being set to format, depth and luminance.
intel_aux_get_meta_address_mask() was previously using 2 shifts to
mask out bits above index 47, what is not so obvious and are 2
operations, so here doing a AND with VALID_ADDRESS_MASK to make it
easier to understand.
Signed-off-by: José Roberto de Souza <jose.souza@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24077>
No changes in behavior here, mostly doing this types of renames:
- address to main_address, to know that addresses refers to main
surface address or aux surface address
- gpu to addr
- main_map_addr to main_inc_addr
- aux_dest_addr to aux_inc_addr
Signed-off-by: José Roberto de Souza <jose.souza@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24077>
remove_mapping() duplicated almost half of get_aux_entry(), it is
only dropping the cases were entries are not alocated but during
removal it is expected that entries were already alocated so we can
reuse get_aux_entry() and drop duplicated code.
Signed-off-by: José Roberto de Souza <jose.souza@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24077>
The only user of format_enum is intel_aux_map_get_alignment() that
can easily use information in format->main_page_size.
This allow us to nuke format_enum and remove duplicated information
in intel_aux_map_get_alignment().
Signed-off-by: José Roberto de Souza <jose.souza@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24077>
Wa_14016118574 is not the lineage number for this workaround so
it was updated to Wa_22014412737.
Wa_22014412737 is not applicable for MTL B0 steppings and newer
so using the workaround framework eliminates this pipe_control
instruction for not affected revisions.
Signed-off-by: José Roberto de Souza <jose.souza@intel.com>
Reviewed-by: Sagar Ghuge <sagar.ghuge@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24221>
