I'm bringing up freedreno Vulkan on an Android phone, and my pains are
exactly what Chad said when working on Intel's vulkan for Android in
aa716db0f6 ("intel: Add simple logging façade for Android (v2)"):
On Android, stdio goes to /dev/null. On Android, remote gdb is even
more painful than the usual remote gdb. On Android, nothing works like
you expect and debugging is hell. I need logging.
This patch introduces a small, simple logging API that can easily wrap
Android's API. On non-Android platforms, this logger does nothing
fancy. It follows the time-honored Unix tradition of spewing
everything to stderr with minimal fuss.
My goal here is not perfection. My goal is to make a minimal, clean API,
that people hate merely a little instead of a lot, and that's good
enough to let me bring up Android Vulkan. And it needs to be fast,
which means it must be small. No one wants to their game to miss frames
while aiming a flaming bow into the jaws of an angry robot t-rex, and
thus become t-rex breakfast, because some fool had too much fun desiging
a bloated, ideal logging API.
Compared to trusty fprintf, _mesa_log[ewi]() is actually usable on
Android. Compared to os_log_message(), this has different error levels
and supports format arguments.
The only code change in the move is wrapping flockfile/funlockfile in
!DETECT_OS_WINDOWS, since mingw32 doesn't have it. Windows likely wants
different logging code, anyway.
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6806>
This introduces an analysis pass intended to estimate several
performance statistics of the shader, including cycle count latency
and throughput values, based on static modeling. It has instruction
performance information more comprehensive than the current scheduling
pass for all platforms between Gen4-11, and works on both the FS and
VEC4 back-end.
The most immediate purpose of this pass is to implement a heuristic
meant to determine whether using SIMD32 dispatch for a fragment shader
can be expected to help more than it hurts. In addition this will
allow the effect of passes run after scheduling (e.g. the TGL software
scoreboard pass and the VEC4 dependency control pass) to be visible in
shader-db statistics.
But that isn't the end of the story, other potential applications of
this pass (not part of this MR) I've been playing around with are:
- Implement a similar SIMD16 heuristic allowing the identification of
inefficient SIMD16 fragment shaders.
- Implement similar SIMD16 and SIMD32 heuristics for the compute
shader stage -- Currently compute shader builds always use the
SIMD16 shader if available and never use the SIMD32 shader unless
strictly necessary, which is suboptimal under certain conditions.
- Hook up to the instruction scheduler in order to improve the
accuracy of its timing information.
- Use as heuristic in order to drive the selection of scheduling
modes (Matt was experimenting with that).
- Plug to the TGL software scoreboard pass in order to implement a
more effective SBID token allocation algorithm, since in general
the optimal token allocation depends on the timings of all
instructions in the program.
- Use its bottleneck detection functionality in order to implement a
heuristic computing a more optimal bound for the number of fragment
shader threads executed in parallel (by adjusting the
MaximumNumberofThreadsPerPSD control of 3DSTATE_PS).
As a follow-up I'm planning to submit updated timing information for
Gen12 platforms -- Everything else required to support Gen12 like SWSB
handling is already included in this patch, but there were some IP
concerns regarding the TGL timing parameters since they cannot
currently be obtained with the documentation and hardware which is
publicly available. The timing parameters for any previous Gen7-11
platforms can be obtained by anyone by sampling the timestamp register
using e.g. shader_time, though I have some more convenient
instrumentation coming up.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Where they belong.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Acked-by: Jason Ekstrand <jason@jlekstrand.net>
Acked-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Rafael Antognolli <rafael.antognolli@intel.com>
Reviewed-by: Mark Janes <mark.a.janes@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4344>
This stuff is somewhat specific to the GL extension & drivers. On
Vulkan we won't use this, it also made a rather large file.
v2: Fix Android build (Lionel)
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Acked-by: Jason Ekstrand <jason@jlekstrand.net>
Acked-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Rafael Antognolli <rafael.antognolli@intel.com>
Reviewed-by: Mark Janes <mark.a.janes@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4344>
Gen12 does not support RENDER_SURFACE_STATE::SurfaceArray = true &&
RENDER_SURFACE_STATE::Depth = 0. SurfaceArray can only be set to true
if Depth >= 1.
We workaround this limitation by adding the max(value, 1) snippet in
the shaders on the 3 components for texture array sizes.
Tested on Gen9 with the following Vulkan CTS tests :
dEQP-VK.image.image_size.2d_array.*
v2: Drop debug print (Tapani)
Switch to GEN:BUG instead of Wa_
v3: Fix dEQP-VK.image.image_size.1d_array.* cases (Lionel)
v4: Fix dEQP-VK.glsl.texture_functions.query.texturesize.* cases
(Missing tex_op handling) (Lionel)
v5: Missing break statement (Lionel)
v6: Fixup comment (Tapani)
v7: Fixup comment again (Tapani)
v8: Don't use sample_dim as index (Jason)
Rename pass
Simplify control flow
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com> (v7)
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3362>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3362>
This substantially reworks both the state setup side of push constant
handling and the pipeline compile side. The fundamental change here is
that we're no longer respecting the prog_data::param array and instead
are just instructing the back-end compiler to leave the array alone.
This makes the state setup side substantially simpler because we can now
just memcpy the whole block of push constants and don't have to
upload one DWORD at a time.
This also means that we can compute the full push constant layout
up-front and just trust the back-end compiler to not mess with it.
Maybe one day we'll decide that the back-end compiler can do useful
things there again but for now, this is functionally no different from
what we had before this commit and makes the NIR handling cleaner.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
It turns off that emitting push constants is one of the hottest paths in
the driver and ANY work we do there costs us. By pre-computing things a
bit ahead of time, we shave 5% off the runtime of a CPU-limited example
running with the Dawn WebGPU implementation.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
v2: Introduce the appropriate pipe controls
Properly deal with changes in metric sets (using execbuf parameter)
Record marker at query end
v3: Fill out PerfCntr1&2
v4: Introduce vkUninitializePerformanceApiINTEL
v5: Use new execbuf extension mechanism
v6: Fix comments in genX_query.c (Rafael)
Use PIPE_CONTROL workarounds (Rafael)
Refactor on the last kernel series update (Lionel)
v7: Only I915_PERF_IOCTL_CONFIG when perf stream is already opened (Lionel)
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Rafael Antognolli <rafael.antognolli@intel.com>
Gen12+ hardware lacks the register scoreboard logic that used to
guarantee data coherency between register reads and writes in previous
generations. This lowering pass runs after register allocation in
order to make up for it.
It works by performing global dataflow analysis in order to determine
the set of potential dependencies of every instruction in the shader,
and then inserts any required SWSB annotations and additional SYNC
instructions in order to guarantee data coherency.
v2: Drop unnecessary _safe list iteration (Caio).
v3: Temporarily workaround potential WaR hazard between FPU
instruction and subsequent out-of-order write, pending
clarification from the hardware team. Drop redundant tracking of
implicit access of acc0-1, since the hardware guarantees coherency
of these (but not the other accumulators...).
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
This rewrites the current opcode description tables as a more compact
flat data structure. The purpose is to allow efficient constant-time
look-up by either HW or IR opcode, which will allow us to drop the
hard-coded correspondence between HW and IR opcodes -- See the next
commits for the rationale.
brw_eu.c is now built as C++ source so we can take advantage of
pointers to member in order to make the look-up function work
regardless of the opcode_desc member used as look-up key.
v2: Optimize devinfo struct comparison (Caio)
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Some conversions are not directly supported in hardware and need to be
split in two conversion instructions going through an intermediary type.
Doing this at the NIR level simplifies a bit the complexity in the backend.
v2:
- Consider fp16 rounding conversion opcodes
- Properly handle swizzles on conversion sources.
v3
- Run the pass earlier, right after nir_opt_algebraic_late (Jason)
- NIR alu output types already have the bit-size (Jason)
- Use 'is_conversion' to identify conversion operations (Jason)
v4:
- Be careful about the intermediate types we use so we don't lose
range and avoid incorrect rounding semantics (Jason)
Reviewed-by: Topi Pohjolainen <topi.pohjolainen@intel.com> (v1)
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
We'll want to reuse those structures later on.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Mark Janes <mark.a.janes@intel.com>
We would like to reuse performance query metrics in other APIs. Let's
make the query code dealing with the processing of raw counters into
human readable values API agnostic.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Mark Janes <mark.a.janes@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
The i965 driver has a bunch of code to compare two sets of program keys
and print out the differences. This can be useful for debugging why a
shader needed to be recompiled on the fly due to non-orthogonal state
dependencies. anv doesn't do recompiles, so we didn't need to share
this in the past - but I'd like to use it in iris.
This moves the bulk of the code to the compiler where it can be reused.
To make that possible, we need to decouple it from i965 - we can't get
at the brw program cache directly, nor use brw_context to print things.
Instead, we use compiler->shader_perf_log(), and simply pass in keys.
We put all of this debugging code in brw_debug_recompile.c, and only
export a single function, for simplicity. I also tidied the code a
bit while moving it, now that it all lives in one file.
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
libintel_common depends on libintel_compiler, but it contains debug
functionality that is needed by libintel_compiler. Break the circular
dependency by moving gen_debug files to libintel_dev.
Suggested-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
All of the actual abstraction (except possibly setting size_written)
happens as part of the logical opcodes. The only thing that the surface
builder is providing at this point is extra levels of functions to call
through. I'm going to be adding bindless image support soon and all the
extra abstraction here is just getting in the way.
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Patch moves intel_tiled_memcpy[_sse41] libraries to isl, renames some
functions and types and makes the required build system changes for
meson, automake and Android. No functional changes are introduced.
v2: code cleanups, move isl_get_memcpy_type to i965 (Jason)
v3: move isl_mem_copy_fn to priv header, cleanups (Jason, Dylan)
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Dylan Baker <dylan@pnwbakers.com>
Acked-by: Kenneth Graunke <kenneth@whitecape.org>
This legalization pass is meant to handle situations where the source
or destination regioning controls of an instruction are unsupported by
the hardware and need to be lowered away into separate instructions.
This should be more reliable and future-proof than the current
approach of handling CHV/BXT restrictions manually all over the
visitor. The same mechanism is leveraged to lower unsupported type
conversions easily, which obsoletes the lower_conversions pass.
v2: Give conditional modifiers the same treatment as predicates for
SEL instructions in lower_dst_modifiers() (Iago). Special-case a
couple of other instructions with inconsistent conditional mod
semantics in lower_dst_modifiers() (Curro).
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
We have a bunch of code to do this in the back-end compiler but it's
fairly specific to typed surface messages and the way we emit them.
This breaks it out into NIR were it's easier to do things a bit more
generally. It also means we can easily share the code between the vec4
and FS back-ends if we wish.
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
This reduces the amount of #ifdef ANDROID we'll have to have inside
the driver. Potentially offering better coverage of the android
extensions.
v2: Move anv_android.h include before anv_entrypoints.h (Tapani)
Fix autotools android build (Lionel)
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
This commit moves our storage image format conversion codegen into NIR
instead of doing it in the back-end. This has the advantage of letting
us run it through NIR's optimizer which is pretty effective at shrinking
things down. In the common case of rgba8, the number of instructions
emitted after NIR is done with it is half of what it was with the
lowering happening in the back-end. On the downside, the back-end's
lowering is able to directly use predicates and the NIR lowering has to
use IFs.
Shader-db results on Kaby Lake:
total instructions in shared programs: 15166910 -> 15166872 (<.01%)
instructions in affected programs: 5895 -> 5857 (-0.64%)
helped: 15
HURT: 0
Clearly, we don't have that much image_load_store happening in the
shaders in shader-db....
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
This adds support for the KHR_display extension to the anv Vulkan
driver. The driver now attempts to open the master DRM node when the
KHR_display extension is requested so that the common winsys code can
perform the necessary operations.
v2: Make sure primary fd is usable
When KHR_display is selected, we try to open the primary node
instead of the render node in case the user wants to use
KHR_display for presentation. However, if we're actually going
to end up using RandR leases, then we don't care if the
resulting fd can't be used for display, but the kernel also
prevents us from using it for drawing when someone else has
master.
v3:
Simplify addition of VK_USE_PLATFORM_DISPLAY_KHR to vulkan_wsi_args
Suggested-by: Eric Engestrom <eric.engestrom@imgtec.com>
v4:
Adapt primary node usage to new wsi_device_init API
v5:
Adopt Jason Ekstrand's coding conventions
Declare variables at first use, eliminate extra whitespace between
types and names. Wrap lines to 80 columns.
Remove spurious MM_PER_PIXEL define
Suggested-by: Jason Ekstrand <jason.ekstrand@intel.com>
v6:
Open DRM master before initializing WSI layer.
The DRM master FD is passed to the WSI layer during
initialization, so we need to open the device slightly earlier
in the function.
Close DRM master in device_finish.
Use anv_gem_get_param to detect working master_fd instead of
directly using the ioctl.
Suggested-by: Jason Ekstrand <jason.ekstrand@intel.com>
v7:
Add vkCreateDisplayModeKHR. This doesn't actually create
new modes, it only looks to see if the requested parameters
matches an existing mode and returns that.
Suggested-by: Jason Ekstrand <jason.ekstrand@intel.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
A later patch will make use of this in other places. Also, remove
dependency on undefined behavior of left-shifting a signed value.
v2: - move function into a separate header (Chris)
v3: (by Ken) Add new header to the various build systems.
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
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 is part of the device groups extension/feature but it's a decent
chunk of work in its own right so it's worth breaking into its own
patch. The mechanism we use is fairly straightforward: we just push the
base work group id into the shader and add it to the work group id we
get from dispatch.
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Split out the device info so isl doesn't depend on intel/common. Now
it will depend on the new intel/dev device info lib.
This will allow the decoder in intel/common to use isl, allowing us to
apply Ken's patch that removes the genxml duplication of surface
formats.
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
For also using it in radv. I moved the remaining stubs back to
anv_device.c as they were just trivial.
This does not move the vk_errorf/anv_perf_warn or the object
type macros, as those depend on anv types and logging.
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Unnecessary GRF bank conflicts increase the issue time of ternary
instructions (the overwhelmingly most common of which is MAD) by
roughly 50%, leading to reduced ALU throughput. This pass attempts to
minimize the number of bank conflicts by rearranging the layout of the
GRF space post-register allocation. It's in general not possible to
eliminate all of them without introducing extra copies, which are
typically more expensive than the bank conflict itself.
In a shader-db run on SKL this helps roughly 46k shaders:
total conflicts in shared programs: 1008981 -> 600461 (-40.49%)
conflicts in affected programs: 816222 -> 407702 (-50.05%)
helped: 46234
HURT: 72
The running time of shader-db itself on SKL seems to be increased by
roughly 2.52%±1.13% with n=20 due to the additional work done by the
compiler back-end.
On earlier generations the pass is somewhat less effective in relative
terms because the hardware incurs a bank conflict anytime the last two
sources of the instruction are duplicate (e.g. while trying to square
a value using MAD), which is impossible to avoid without introducing
copies. E.g. for a shader-db run on SNB:
total conflicts in shared programs: 944636 -> 623185 (-34.03%)
conflicts in affected programs: 853258 -> 531807 (-37.67%)
helped: 31052
HURT: 19
And on BDW:
total conflicts in shared programs: 1418393 -> 987539 (-30.38%)
conflicts in affected programs: 1179787 -> 748933 (-36.52%)
helped: 47592
HURT: 70
On SKL GT4e this improves performance of GpuTest Volplosion by 3.64%
±0.33% with n=16.
NOTE: This patch intentionally disregards some i965 coding conventions
for the sake of reviewability. This is addressed by the next
squash patch which introduces an amount of (for the most part
boring) boilerplate that might distract reviewers from the
non-trivial algorithmic details of the pass.
The following patch is squashed in:
SQUASH: intel/fs/bank_conflicts: Roll back to the nineties.
Acked-by: Matt Turner <mattst88@gmail.com>
It was the only file named intel_* in the compiler.
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
