Mark Menzynski [Tue, 10 Dec 2019 10:28:11 +0000 (11:28 +0100)]
util/blob: Add overwrite function for uint8
Overwrite function for this type was missing and I needed it for my project.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Signed-off-by: Mark Menzynski <mmenzyns@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3903>
Vasily Khoruzhick [Thu, 19 Mar 2020 02:27:12 +0000 (19:27 -0700)]
lima: add support for R and RG formats
Unfortunately these are not supported natively for sampling
so we have to lower them.
Reviewed-by: Erico Nunes <nunes.erico@gmail.com>
Signed-off-by: Vasily Khoruzhick <anarsoul@gmail.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4241>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4241>
Vasily Khoruzhick [Thu, 19 Mar 2020 02:21:57 +0000 (19:21 -0700)]
lima: split pixel and texel format tables
This is preparation for the next commit where we may need different
swap_r_b flags for pixel and texel formats.
Reviewed-by: Erico Nunes <nunes.erico@gmail.com>
Signed-off-by: Vasily Khoruzhick <anarsoul@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4241>
Erik Faye-Lund [Tue, 3 Mar 2020 18:37:37 +0000 (19:37 +0100)]
zink/spirv: do not use bitwise operations on booleans
According to the SPIR-V specification, these operations require
integer-types. When bit_size is 1, we use booleans, which makes us emit
illegal code.
So let's fix the emitting to check if the first source is one bit wide.
For inot we can take a short-cut, and check the destination instead.
This doesn't work for ieq and ine, so let's not bother to do this
BINOP_LOG.
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4036>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4036>
Michel Dänzer [Wed, 18 Mar 2020 17:41:43 +0000 (18:41 +0100)]
gitlab-ci: Restrict s390x/ppc64el jobs to packet runners
They are hitting timeouts on the gstreamer runners now... *sigh*
Reviewed-by: Adam Jackson <ajax@redhat.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4233>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4233>
Rhys Perry [Thu, 19 Mar 2020 20:13:52 +0000 (20:13 +0000)]
radv/winsys: set has_syncobj_wait_for_submit in the null winsys
Needed for Vulkan 1.1+
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4249>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4249>
Lionel Landwerlin [Thu, 19 Mar 2020 19:30:01 +0000 (21:30 +0200)]
intel: add new TGL pci ids
Update following kernel : https://patchwork.freedesktop.org/patch/357921/
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Bspec: 44455
Acked-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Scott D Phillips <scott.d.phillips@intel.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4248>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4248>
Samuel Pitoiset [Wed, 18 Mar 2020 17:49:23 +0000 (18:49 +0100)]
radv: fix optional pSizes parameter when binding streamout buffers
The Vulkan spec 1.2.135 says:
"pSizes is an optional array of buffer sizes, specifying the maximum
number of bytes to capture to the corresponding transform feedback
buffer. If pSizes is NULL, or the value of the pSizes array element
is VK_WHOLE_SIZE, then the maximum bytes captured will be the size
of the corresponding buffer minus the buffer offset."
Closes: https://gitlab.freedesktop.org/mesa/mesa/issues/2650
Fixes:
b4eb029062a ("radv: implement VK_EXT_transform_feedback")
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4232>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4232>
Caio Marcelo de Oliveira Filho [Thu, 19 Mar 2020 00:14:19 +0000 (17:14 -0700)]
mesa/main: Fix overflow in validation of DispatchComputeGroupSizeARB
An uint64_t can store the result of multiplying two GLuint (uint32_t),
so use that property to check for overflow when calculating the total.
Change the error message so we don't need to care about the actual
total -- which means we don't need a larger than 64-bit value to hold
it.
Fixes:
45ab63c0cb2 ("mesa/main: add support for ARB_compute_variable_groups_size")
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4240>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4240>
Marek Olšák [Fri, 20 Mar 2020 01:01:24 +0000 (21:01 -0400)]
driconf: enable glthread for "From The Depths"
25% perf improvement
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4254>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4254>
Marek Olšák [Fri, 13 Mar 2020 20:10:56 +0000 (16:10 -0400)]
winsys/radeon: change to 3-space indentation
Acked-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4192>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4192>
Marek Olšák [Thu, 5 Mar 2020 21:03:19 +0000 (16:03 -0500)]
glthread: don't declare unmarshal functions as inline
They are never inlined.
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4251>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4251>
Marek Olšák [Thu, 5 Mar 2020 20:59:49 +0000 (15:59 -0500)]
glthread: clean up debug_print_sync code
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4251>
Marek Olšák [Thu, 5 Mar 2020 20:57:46 +0000 (15:57 -0500)]
glthread: remove debug_print_marshal function
We don't need to print every function we execute.
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4251>
Marek Olšák [Wed, 4 Mar 2020 20:02:15 +0000 (15:02 -0500)]
glthread: don't execute any custom VAO and BindBuffer code in the Core profile
It's not needed, because user pointers can never occur there.
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4251>
Marek Olšák [Wed, 4 Mar 2020 19:46:56 +0000 (14:46 -0500)]
glthread: track VAOs created by CreateVertexArrays
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4251>
Marek Olšák [Thu, 20 Feb 2020 03:25:07 +0000 (22:25 -0500)]
glthread: enable display lists
They seem to work fine.
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4251>
Marek Olšák [Fri, 6 Mar 2020 03:21:11 +0000 (22:21 -0500)]
glthread: align the batch buffer to 8 bytes for pointers and doubles again
This was changed when I switched to types from size_t to int.
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4251>
Marek Olšák [Mon, 9 Mar 2020 23:37:39 +0000 (19:37 -0400)]
mesa: remove redundant api_loopback functions
vbo_attrib_tmp.h implements them, so this loopback code isn't needed
and shouldn't be used.
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4123>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4123>
Marek Olšák [Sun, 8 Mar 2020 22:47:56 +0000 (18:47 -0400)]
mesa: use vbo_attrib_tmp.h to generate display list vertex attrib functions
This removes about 1150 lines of code.
The diff is messy, but the new code really starts with save_Attr32bit and
below. Ignore false Eval/Material/Begin changes etc. Git can't figure out
what was really changed. I didn't change them.
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4123>
Jason Ekstrand [Wed, 18 Mar 2020 18:29:06 +0000 (13:29 -0500)]
anv: Only add END_OF_PIPE_SYNC if we actually have AUX_INVAL
Fixes:
43dc842cb91c "anv: Wait for the GPU to be idle before..."
Reviewed-by: Rafael Antognolli <rafael.antognolli@intel.com>
Reviewed-by: D Scott Phillips <d.scott.phillips@intel.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4234>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4234>
Eric Anholt [Tue, 17 Mar 2020 23:03:11 +0000 (16:03 -0700)]
freedreno: Switch to exposing only half-integer pixel centers.
This is what the HW provides us. If we need integer pixel centers, we
want the state tracker to do the lowering pass so that it gets to optimize
on the subtract. This is also the shader instructions that the blob is
doing on GLES, and is what Vulkan wants too, as was noted in MR !4172.
shader-db on a630:
total instructions in shared programs: 186689 -> 186168 (-0.28%)
total nops in shared programs: 66253 -> 66139 (-0.17%)
total non-nops in shared programs: 120436 -> 120029 (-0.34%)
total dwords in shared programs: 292192 -> 291168 (-0.35%)
total last-baryf in shared programs: 4810 -> 4734 (-1.58%)
total full in shared programs: 10176 -> 10195 (0.19%)
total constlen in shared programs: 54589 -> 54575 (-0.03%)
total sstall in shared programs: 24582 -> 24802 (0.89%)
total (ss) in shared programs: 3921 -> 3925 (0.10%)
total (sy) in shared programs: 1934 -> 1923 (-0.57%)
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4223>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4223>
John Stultz [Thu, 12 Mar 2020 22:42:29 +0000 (22:42 +0000)]
r600: Fix build error in sfn_nir_lower_fs_out_to_vector.cpp
In trying a full build under AOSP, I ran into the following error:
In file included from external/mesa3d/src/gallium/drivers/r600/sfn/sfn_nir_lower_fs_out_to_vector.cpp:33:
external/libcxx/include/set:942:26: error: the specified comparator type does not provide a const call operator [-Werror,-Wuser-defined-warnings]
static_assert(sizeof(__diagnose_non_const_comparator<_Key, _Compare>()), "");
^
external/mesa3d/src/gallium/drivers/r600/sfn/sfn_nir_lower_fs_out_to_vector.cpp:78:34: note: in instantiation of template class 'std::__1::multiset<nir_intrinsic_ins
tr *, r600::nir_intrinsic_instr_less, std::__1::allocator<nir_intrinsic_instr *> >' requested here
using InstrSubSet = std::pair<InstrSet::iterator, InstrSet::iterator>;
^
external/libcxx/include/__tree:967:5: note: from 'diagnose_if' attribute on '__diagnose_non_const_comparator<nir_intrinsic_instr *, r600::nir_intrinsic_instr_less>':
_LIBCPP_DIAGNOSE_WARNING(!std::__invokable<_Compare const&, _Tp const&, _Tp const&>::value,
^ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
external/libcxx/include/__config:1244:21: note: expanded from macro '_LIBCPP_DIAGNOSE_WARNING'
__attribute__((diagnose_if(__VA_ARGS__, "warning")))
^ ~~~~~~~~~~~
1 error generated.
Which is pretty opaque to me, but searching the web suggested
adding a cost, which seems to resovle it.
Reviewed-by: Gert Wollny <gert.wollny@collabora.com>
Signed-off-by: John Stultz <john.stultz@linaro.org>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4175>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4175>
John Stultz [Thu, 12 Mar 2020 21:21:50 +0000 (21:21 +0000)]
vc4_bufmgr: Remove duplicative VC definition
This is already defined in
src/broadcom/cle/v3d_packet_helpers.h:42:9
And was causing build issues in AOSP when building with mmma
Reviewed-by: Eric Anholt <eric@anholt.net>
Signed-off-by: John Stultz <john.stultz@linaro.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4175>
John Stultz [Thu, 12 Mar 2020 22:21:29 +0000 (22:21 +0000)]
etnaviv: Avoid shift overflow
Building with AOSP I'm seeing:
external/mesa3d/src/gallium/drivers/etnaviv/etnaviv_screen.c:245:31: error: signed shift result (0x100000000) requires 34 bits to represent, but 'int' only has 32 bits [-Werror,-Wshift-overflow]
system_memory = 4096 << 20;
system_memory is a uint_64t, so this patch addresses the issue
by casting 4096 to a unint_64t before the shift is done.
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Signed-off-by: John Stultz <john.stultz@linaro.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4175>
John Stultz [Fri, 26 Apr 2019 16:35:13 +0000 (09:35 -0700)]
Android.mk: Tweak MESA_ENABLE_LLVM checks
Change the MESA_ENABLE_LLVM checks in Android.mk
files in order to get mesa3d to build w/ AOSP
using mmma.
This tries to re-create a change that was introduced
in the following merge in the AOSP branch:
69f2c0128d2b Merge branch 'aosp/upstream-18.0'
Acked-by: Tapani Pälli <tapani.palli@intel.com>
Acked-by: Mauro Rossi <issor.oruam@gmail.com>
Signed-off-by: John Stultz <john.stultz@linaro.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4175>
Jason Ekstrand [Wed, 4 Mar 2020 20:37:03 +0000 (14:37 -0600)]
intel/iris: Always initialize CCS to 0
Previously, we were initializing the CCS to 0xFF for MCS+CCS due to a
misunderstanding of the following lines in the bspec:
The following are the general SW requirements for MCS buffer clear
functionality:
...
- If Software wants to enable Color Compression without Fast
clear, Software needs to initialize MCS with zeros.
- Lossless compression and CCS initialized to all F (using HW
Fast Clear or SW direct Clear) on the same surface is not
supported.
The first line does not refer to the CCS as the comment author supposed
but refers to the MCS as the comment says. It means that if you want to
use MCS compression without a fast-clear, you should initialize the MCS
to 0x00. This is because the value 0x00 in the MCS means "all data is
in plane 0" which is a perfectly valid non-fast-clear initialization.
It's also the value the MCS should be in if you do a RECTLIST slow-clear
where the primitive fully covers each pixel such that the same value is
written to all samples.
The second line in the above quote seems to imply that CCS fast-clear is
incompatible with MCS fast-clear. In particular, MCS+CCS fast-clear
uses a 0xff value in the MCS (like on Gen7-11) and leaves the CCS in
either the compressed or the pass-through state. Therefore, we should
initialize the CCS to 0x00 even for MCS+CCS surfaces.
Reviewed-by: Sagar Ghuge<sagar.ghuge@intel.com>
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4074>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4074>
Lionel Landwerlin [Wed, 18 Mar 2020 18:03:53 +0000 (20:03 +0200)]
isl: drop min row pitch alignment when set by the driver
When the caller of the isl_surf_init() specifies a row pitch, do not
consider the minimum CCS requirement if it's incompatible with the
caller's value.
isl_surf_get_ccs_surf() will check that the main surface alignment
matches CCS expectations.
v2: Simplify checks (Nanley)
v3: Add Comment about isl_surf_get_ccs_surf() (Jason)
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Cc: <mesa-stable@lists.freedesktop.org>
Fixes:
a3f6db2c4e92 ("isl: drop CCS row pitch requirement for linear surfaces")
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4243>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4243>
Lionel Landwerlin [Thu, 20 Feb 2020 14:16:33 +0000 (16:16 +0200)]
isl: only apply main surface ccs pitch constraint with CCS
We could be creating a Y-tiled surface that isn't going to use CCS
(this could be the case when clearly indicated through modifiers).
Don't apply the main surface pitch alignment constraint in that case.
v2: Use logical NOT (Sagar)
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Fixes:
a3f6db2c4e92 ("isl: drop CCS row pitch requirement for linear surfaces")
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4243>
Lionel Landwerlin [Tue, 15 Oct 2019 11:37:45 +0000 (14:37 +0300)]
isl: properly filter supported display modifiers on Gen9+
Y tiling is supported for display on Gen9+ so don't filter it from the
possible flags.
v2: Drop Yf from display supported tilings on Gen12+ (Jason)
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Cc: <mesa-stable@lists.freedesktop.org>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4243>
Lionel Landwerlin [Mon, 14 Oct 2019 18:26:18 +0000 (21:26 +0300)]
isl: implement linear tiling row pitch requirement for display
We're missing a requirement for alignment of row pitch for the display
HW. In linear tiling, the row pitch must be a 64bytes aligned.
v2: Use correct formula to align to 64bytes (Chad)
v3: Matching {} (Jason)
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Cc: <mesa-stable@lists.freedesktop.org>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4243>
Eric Anholt [Wed, 18 Mar 2020 16:51:03 +0000 (09:51 -0700)]
ci: Only run the freedreno baremetal tests when freedreno/core changes.
Same as we do for a630 (docker) tests.
Reviewed-by: Michel Dänzer <mdaenzer@redhat.com>
Acked-by: Rob Clark <robdclark@chromium.org>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4229>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4229>
Dylan Baker [Thu, 5 Mar 2020 23:48:42 +0000 (15:48 -0800)]
docs/release-calendar: Add calendar for 20.1 Release candidates
It's time to start getting the calendar going for 20.1 so that everyone
is clear on when the close date for new features is. Eric Engstrom has
agreed to help out with the 20.1 series, and will be the primary point,
he's also helping out with a few of the 20.0.x point releases.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Acked-by: Eric Engestrom <eric@engestrom.ch>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4077>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4077>
Rhys Perry [Fri, 13 Mar 2020 16:38:41 +0000 (16:38 +0000)]
radv: call nir_shader_gather_info again
pipeline-db (Navi, ACO):
Totals from affected shaders:
SGPRS: 11840 -> 11840 (0.00 %)
VGPRS: 19012 -> 19124 (0.59 %)
Spilled SGPRs: 0 -> 0 (0.00 %)
Spilled VGPRs: 0 -> 0 (0.00 %)
Scratch size: 3696 -> 3696 (0.00 %) dwords per thread
Code Size: 998680 -> 921388 (-7.74 %) bytes
LDS: 19646 -> 19646 (0.00 %) blocks
Max Waves: 3398 -> 3401 (0.09 %)
pipeline-db (Navi, LLVM):
Totals from affected shaders:
SGPRS: 17016 -> 17128 (0.66 %)
VGPRS: 19564 -> 14876 (-23.96 %)
Spilled SGPRs: 0 -> 0 (0.00 %)
Spilled VGPRs: 0 -> 0 (0.00 %)
Scratch size: 3872 -> 3872 (0.00 %) dwords per thread
Code Size: 820416 -> 743576 (-9.37 %) bytes
LDS: 0 -> 0 (0.00 %) blocks
Max Waves: 3367 -> 3534 (4.96 %)
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-By: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4193>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4193>
Rhys Perry [Fri, 13 Mar 2020 19:31:03 +0000 (19:31 +0000)]
nir/gather_info: handle emit_vertex_with_counter
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-By: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
CC: <mesa-stable@lists.freedesktop.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4193>
Tomasz Pyra [Tue, 10 Mar 2020 12:00:03 +0000 (13:00 +0100)]
gallium/swr: spin-lock performance improvement
Currently, the worker threads are very aggresively
polling for new tasks. If the work is not constantly
fed into the pipeline (which is a case for most of
interactive applications), this creates unnecessary
memory pressure and is using CPU cycles that
could otherwise be used by the applications.
The change implements simple back off mechanism
to help with this problem
Change by Tomasz Pyra (tomasz.pyra@intel.com)
Reviewed-by: Alok Hota <alok.hota@intel.com>
Reviewed-by: Jan Zielinski <jan.zielinski@intel.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4226>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4226>
Pierre-Eric Pelloux-Prayer [Fri, 13 Mar 2020 09:04:16 +0000 (10:04 +0100)]
radeonsi: enable glsl_zero_init for Curse of the Dead Gods
Closes: https://gitlab.freedesktop.org/mesa/mesa/issues/2598
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4214>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4214>
Marek Olšák [Fri, 13 Mar 2020 21:54:35 +0000 (17:54 -0400)]
nir: fix clip/cull_distance_array_size in nir_lower_clip_cull_distance_arrays
This fixes a GPU hang on radeonsi.
It only works if optimizations have already been run.
Cc: 19.3 20.0 <mesa-stable@lists.freedesktop.org>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4194>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4194>
Alyssa Rosenzweig [Thu, 19 Mar 2020 03:12:38 +0000 (23:12 -0400)]
pan/bi: Pack BI_BLEND
MRT not yet supported to keep things easy.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4242>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4242>
Alyssa Rosenzweig [Thu, 19 Mar 2020 03:12:23 +0000 (23:12 -0400)]
pan/bi: Flesh out BI_BLEND
It ingests the output of ATEST, whatever that actually is.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4242>
Alyssa Rosenzweig [Thu, 19 Mar 2020 03:03:07 +0000 (23:03 -0400)]
pan/bi: Add ATEST packing
Only fp32 for now.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4242>
Alyssa Rosenzweig [Thu, 19 Mar 2020 03:02:12 +0000 (23:02 -0400)]
pan/bi: Flesh out ATEST in IR
ATEST actually takes two sources and has a destination. Although the
details are a little funny, we should still model this correctly.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4242>
Alyssa Rosenzweig [Wed, 18 Mar 2020 17:43:10 +0000 (13:43 -0400)]
pan/bi: Track clause types during scheduling
There's an easy mapping for this, so let's do it. Note we do this at
schedule-time instead of emit since we'll need to lookahead clause
types. The alternative is a prepass running after schedule but before
codegen, but there's no reason not to just stick it here when we're
preparing bi_clause in the first place.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4242>
Alyssa Rosenzweig [Wed, 18 Mar 2020 17:42:12 +0000 (13:42 -0400)]
pan/bi: Don't hide SCHED_ADD inside HI_LATENCY
It makes bitwise property checking annoying.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4242>
Alyssa Rosenzweig [Wed, 18 Mar 2020 17:23:00 +0000 (13:23 -0400)]
pan/bi: Pretty-print clause types in disassembler
Also note that type=1 is for load_vary.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4242>
Alyssa Rosenzweig [Wed, 18 Mar 2020 16:18:30 +0000 (12:18 -0400)]
pan/bi: Route through clause header
We already track almost all the information we need, let's dump it onto
the wire now.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4242>
Alyssa Rosenzweig [Wed, 18 Mar 2020 16:08:28 +0000 (12:08 -0400)]
pan/bi: Skip over data registers in port assignment
They bypass the usual mechanism entirely, let's add some props to
describe this and respect them.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4242>
Alyssa Rosenzweig [Wed, 18 Mar 2020 15:55:45 +0000 (11:55 -0400)]
pan/bi: Emit load_vary ops
Annoyingly long code to do so, but this should theoretically work for
both direct and indirect load_vary. Still need to handle destination.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4242>
Alyssa Rosenzweig [Wed, 18 Mar 2020 15:55:10 +0000 (11:55 -0400)]
pan/bi: Pass second src for load_vary ops
For direct, this is just 0, but for indirct, this is a sample mask
preloaded in R61. Handle this at code emit time instead of trying to do
crazy monkeypatching later.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4242>
Alyssa Rosenzweig [Wed, 18 Mar 2020 15:54:25 +0000 (11:54 -0400)]
pan/bi: Generalize bi_get_src a bit
Allow it to work with ADD ops and stub out some immediate fetching
infrastructure (currently only works with 0).
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4242>
Alyssa Rosenzweig [Wed, 18 Mar 2020 15:08:07 +0000 (11:08 -0400)]
pan/bi: List ADD classes in bi_pack_add
Handling will be... somewhat tricky.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4242>
Alyssa Rosenzweig [Wed, 18 Mar 2020 15:02:57 +0000 (11:02 -0400)]
pan/bi: Pack fadd32
Choo choo.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4242>
Alyssa Rosenzweig [Wed, 18 Mar 2020 13:58:59 +0000 (09:58 -0400)]
pan/bi: Pack BI_FMA ops
This is our first instruction we've emitted, requiring us to pipe
through registes/ports and various details from the IR. It's quite a bit
of code, but overall I'm happy with this structure. With some tedium we
should be able to emit the rest of the ALU ops this way, too.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4242>
Alyssa Rosenzweig [Wed, 18 Mar 2020 13:58:42 +0000 (09:58 -0400)]
pan/bi: Add struct bifrost_fma_fma
So we can pack regular FMA ops.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4242>
Alyssa Rosenzweig [Wed, 18 Mar 2020 13:57:32 +0000 (09:57 -0400)]
pan/bi: Model 3-bit Bifrost srcs in IR
We'll want to set these manually for schedule-time passthrough, as well
as use the enum for packing.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4242>
Alyssa Rosenzweig [Wed, 18 Mar 2020 13:22:58 +0000 (09:22 -0400)]
pan/bi: Route through first_instruction field
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4242>
Alyssa Rosenzweig [Wed, 18 Mar 2020 03:01:48 +0000 (23:01 -0400)]
pan/bi: Assign registers to ports
Now that we can pack registers given the assigned ports, and we can
assign registers from the indices, the missing link is assigning ports
from the registers, and now finally we get some real data showing up in
a disassembly exercising lots of different code paths.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4242>
Alyssa Rosenzweig [Wed, 18 Mar 2020 02:42:54 +0000 (22:42 -0400)]
pan/bi: Add missing __attribute__((packed))
That this code worked before makes me rather nervous...
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4242>
Alyssa Rosenzweig [Wed, 18 Mar 2020 02:17:39 +0000 (22:17 -0400)]
pan/bi: Pack register fields
Now that we have ctrl, the rest is natural... sorta.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4242>
Alyssa Rosenzweig [Wed, 18 Mar 2020 01:35:44 +0000 (21:35 -0400)]
pan/bi: Add packing for register control field
Filling in some gaps based on intuition from the bit patterns but this
should be vaguely right. More investigation needed down the line.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4242>
Alyssa Rosenzweig [Thu, 12 Mar 2020 18:33:32 +0000 (14:33 -0400)]
pan/bi: Sketch out instruction word packing
Instructions are 78-bits with some seriously suspicious packing
requirements but hey, gotta save 'em bits.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4242>
Alyssa Rosenzweig [Thu, 12 Mar 2020 18:16:22 +0000 (14:16 -0400)]
pan/bi: Setup initial clause packing
At the moment, we just iterate the clauses in the post-RA, post-sched IR
and generate a dummy clause corresponding, passing the results to the
disassembler to verify.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4242>
Dylan Baker [Wed, 18 Mar 2020 21:48:40 +0000 (14:48 -0700)]
docs: update calendar, add news item, and link releases notes for 20.0.2
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4236>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4236>
Dylan Baker [Wed, 18 Mar 2020 21:40:36 +0000 (14:40 -0700)]
docs/relnotes: Add sha256 sums for 20.0.2
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4236>
Dylan Baker [Wed, 18 Mar 2020 21:22:17 +0000 (14:22 -0700)]
Docs: Add release notes for 20.0.2
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4236>
Eric Anholt [Wed, 18 Mar 2020 17:37:13 +0000 (10:37 -0700)]
ci: Disable tests that showed intermittent fails on a530 in day 1.
Link: https://gitlab.freedesktop.org/mesa/mesa/issues/2649
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4231>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4231>
Eric Anholt [Wed, 18 Mar 2020 17:35:17 +0000 (10:35 -0700)]
ci: Ban the recent popular freedreno a630 flakes.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4231>
Samuel Pitoiset [Wed, 18 Mar 2020 14:03:45 +0000 (15:03 +0100)]
radv: remove wrong assert that checks compute subgroup size
Ooops. For some reasons, I have been confused with Wave32 on GFX10,
but it's still possible to require a specific subgroup size if
only Wave64 is supported.
Fixes:
672d1061998 ("radv/gfx10: fix required subgroup size with VK_EXT_subgroup_size_control")
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4227>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4227>
Jason Ekstrand [Tue, 17 Mar 2020 18:12:35 +0000 (13:12 -0500)]
anv: Swizzle fast-clear values
Starting with Gen12, we can fast-clear a lot more surface formats and we
are suddenly in the position of having to fast-clear surfaces with
formats with an implicit swizzle such as VK_FORMAT_R4G4B4A4_UNORM_PACK16
which is represented as ISL_FORMAT_A4B4G4R4 with a BGRA swizzle. In
order for blorp to do the fast-clear color conversion for us, it needs
a properly swizzled color.
This fixes the following Vulkan CTS groups on TGL:
- dEQP-VK.pipeline.blend.format.b4g4r4a4_unorm_pack16.*
- dEQP-VK.api.image_clearing.core.clear_color_image.*.b4g4r4a4*
Cc: mesa-stable@lists.freedesktop.org
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4218>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4218>
Jason Ekstrand [Tue, 17 Mar 2020 18:36:40 +0000 (13:36 -0500)]
intel/blorp: Add support for swizzling fast-clear colors
Cc: mesa-stable@lists.freedesktop.org
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4218>
Ian Romanick [Thu, 5 Mar 2020 07:00:24 +0000 (23:00 -0800)]
soft-fp64: Split a block that was missing a cast on a comparison
This function has code like:
if (0x7FD <= zExp) {
if ((0x7FD < zExp) ||
((zExp == 0x7FD) &&
(0x001FFFFFu == zFrac0 && 0xFFFFFFFFu == zFrac1) &&
increment)) {
...
return ...;
}
if (zExp < 0) {
I saw that, and I thought, "Uh... what? Dead code?" I thought it was a
bit fishy, so I grabbed the Berkeley SoftFloat Library 3e code, and
there is similar code in softfloat_roundPackToF64
(source/s_roundPackToF64.c), but it has an extra (uint16_t) cast in the
first comparison. This is basicially a shortcut for
if (zExp < 0 || zExp >= 0x7FD) {
So, having the nesting kind of makes sense. On a CPU, nesting the flow
control can be an optimization. On a GPU, it's just fail. Split the
block so that we don't need the uint16_t cast magic.
Results on the 308 shaders extracted from the fp64 portion of the OpenGL
CTS:
Tiger Lake and Ice Lake had similar results. (Tiger Lake shown)
total instructions in shared programs: 683638 -> 658127 (-3.73%)
instructions in affected programs: 666839 -> 641328 (-3.83%)
helped: 92
HURT: 0
helped stats (abs) min: 26 max: 2456 x̄: 277.29 x̃: 144
helped stats (rel) min: 3.21% max: 4.22% x̄: 3.79% x̃: 3.90%
95% mean confidence interval for instructions value: -345.84 -208.75
95% mean confidence interval for instructions %-change: -3.86% -3.73%
Instructions are helped.
total cycles in shared programs: 5458858 -> 5344600 (-2.09%)
cycles in affected programs: 5360114 -> 5245856 (-2.13%)
helped: 92
HURT: 0
helped stats (abs) min: 126 max: 10300 x̄: 1241.93 x̃: 655
helped stats (rel) min: 1.71% max: 2.37% x̄: 2.12% x̃: 2.17%
95% mean confidence interval for cycles value: -1539.93 -943.94
95% mean confidence interval for cycles %-change: -2.16% -2.08%
Cycles are helped.
Fixes:
f111d72596c ("glsl: Add "built-in" functions to do add(fp64, fp64)")
Reviewed-by: Matt Turner <mattst88@gmail.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4142>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4142>
Ian Romanick [Wed, 4 Mar 2020 21:00:44 +0000 (13:00 -0800)]
soft-fp64/fadd: Common code optimization for differing sign case
This is basically the same ideas from the previous 4 commits applied
to the aSign != bSign part... and all smashed into one commit.
The shader hurt for spill and / or fills is from
KHR-GL46.gpu_shader_fp64.builtin.inverse_dmat4.
Results on the 308 shaders extracted from the fp64 portion of the OpenGL
CTS:
Tiger Lake
total instructions in shared programs: 787258 -> 683638 (-13.16%)
instructions in affected programs: 725435 -> 621815 (-14.28%)
helped: 74
HURT: 0
helped stats (abs) min: 152 max: 10261 x̄: 1400.27 x̃: 975
helped stats (rel) min: 11.61% max: 20.92% x̄: 15.40% x̃: 14.86%
95% mean confidence interval for instructions value: -1740.11 -1060.43
95% mean confidence interval for instructions %-change: -16.01% -14.79%
Instructions are helped.
total cycles in shared programs: 6483227 -> 5458858 (-15.80%)
cycles in affected programs: 6051245 -> 5026876 (-16.93%)
helped: 74
HURT: 0
helped stats (abs) min: 1566 max: 95474 x̄: 13842.82 x̃: 9757
helped stats (rel) min: 13.94% max: 23.26% x̄: 17.98% x̃: 17.57%
95% mean confidence interval for cycles value: -17104.25 -10581.40
95% mean confidence interval for cycles %-change: -18.61% -17.35%
Cycles are helped.
total spills in shared programs: 553 -> 445 (-19.53%)
spills in affected programs: 553 -> 445 (-19.53%)
helped: 1
HURT: 0
total fills in shared programs: 1307 -> 1323 (1.22%)
fills in affected programs: 1307 -> 1323 (1.22%)
helped: 0
HURT: 1
Ice Lake
total instructions in shared programs: 781216 -> 678470 (-13.15%)
instructions in affected programs: 720088 -> 617342 (-14.27%)
helped: 74
HURT: 0
helped stats (abs) min: 153 max: 8863 x̄: 1388.46 x̃: 975
helped stats (rel) min: 11.24% max: 21.03% x̄: 15.47% x̃: 15.01%
95% mean confidence interval for instructions value: -1703.57 -1073.35
95% mean confidence interval for instructions %-change: -16.09% -14.85%
Instructions are helped.
total cycles in shared programs: 6464085 -> 5453997 (-15.63%)
cycles in affected programs: 6031771 -> 5021683 (-16.75%)
helped: 74
HURT: 0
helped stats (abs) min: 1552 max: 90317 x̄: 13649.84 x̃: 9650
helped stats (rel) min: 13.84% max: 23.11% x̄: 17.83% x̃: 17.41%
95% mean confidence interval for cycles value: -16802.89 -10496.79
95% mean confidence interval for cycles %-change: -18.46% -17.21%
Cycles are helped.
total spills in shared programs: 279 -> 368 (31.90%)
spills in affected programs: 279 -> 368 (31.90%)
helped: 0
HURT: 1
total fills in shared programs: 973 -> 1155 (18.71%)
fills in affected programs: 973 -> 1155 (18.71%)
helped: 0
HURT: 1
Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4142>
Ian Romanick [Wed, 4 Mar 2020 20:39:36 +0000 (12:39 -0800)]
soft-fp64/fadd: Move common code out of both branches of an if-statement
The previous two commits were just setting the scene for this change.
The mix(..., __propagateFloat64NaN(a, b), propagate) statements are not
identical in the two halves, but they are equivalent. The first clause
of the mix in the else-branch is trivally ±Inf. The first clause in the
then-branch __packFloat64(aSign, aExp, aFracHi, aFracLo). The
preceeding conditions prove that aExp=0x7ff, aFracHi=0, and aFracLo=0.
Results on the 308 shaders extracted from the fp64 portion of the OpenGL
CTS:
Tiger Lake and Ice Lake had similar results. (Tiger Lake shown)
total instructions in shared programs: 819560 -> 787258 (-3.94%)
instructions in affected programs: 757737 -> 725435 (-4.26%)
helped: 74
HURT: 0
helped stats (abs) min: 43 max: 3545 x̄: 436.51 x̃: 296
helped stats (rel) min: 3.54% max: 6.16% x̄: 4.52% x̃: 4.36%
95% mean confidence interval for instructions value: -548.42 -324.61
95% mean confidence interval for instructions %-change: -4.68% -4.37%
Instructions are helped.
total cycles in shared programs: 6817254 -> 6483227 (-4.90%)
cycles in affected programs: 6385272 -> 6051245 (-5.23%)
helped: 74
HURT: 0
helped stats (abs) min: 430 max: 33271 x̄: 4513.88 x̃: 3047
helped stats (rel) min: 4.28% max: 7.45% x̄: 5.48% x̃: 5.31%
95% mean confidence interval for cycles value: -5610.46 -3417.30
95% mean confidence interval for cycles %-change: -5.65% -5.32%
Cycles are helped.
total spills in shared programs: 591 -> 553 (-6.43%)
spills in affected programs: 591 -> 553 (-6.43%)
helped: 1
HURT: 0
total fills in shared programs: 1353 -> 1307 (-3.40%)
fills in affected programs: 1353 -> 1307 (-3.40%)
helped: 1
HURT: 0
Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4142>
Ian Romanick [Wed, 4 Mar 2020 20:31:02 +0000 (12:31 -0800)]
soft-fp64/fadd: Use absolute value of expDiff
In one branch we know that expDiff is already positive.
In the other branch we know the expDiff is negative. Previously in that
branch the code was -(expDiff + 1). This is equvialent to (-expDiff) -
1, and since expDiff is negative, abs(expDiff) - 1.
The main purpose of this commit is to prepare for "soft-fp64/fadd: Move
common code out of both branches of an if-statement".
Results on the 308 shaders extracted from the fp64 portion of the OpenGL
CTS:
Tiger Lake and Ice Lake had similar results. (Tiger Lake shown)
total instructions in shared programs: 818246 -> 819560 (0.16%)
instructions in affected programs: 756423 -> 757737 (0.17%)
helped: 1
HURT: 73
helped stats (abs) min: 1205 max: 1205 x̄: 1205.00 x̃: 1205
helped stats (rel) min: 1.36% max: 1.36% x̄: 1.36% x̃: 1.36%
HURT stats (abs) min: 2 max: 149 x̄: 34.51 x̃: 27
HURT stats (rel) min: 0.14% max: 1.09% x̄: 0.41% x̃: 0.30%
95% mean confidence interval for instructions value: -16.56 52.07
95% mean confidence interval for instructions %-change: 0.30% 0.47%
Inconclusive result (value mean confidence interval includes 0).
total cycles in shared programs: 6816686 -> 6817254 (<.01%)
cycles in affected programs: 6384704 -> 6385272 (<.01%)
helped: 37
HURT: 37
helped stats (abs) min: 30 max: 5790 x̄: 289.05 x̃: 102
helped stats (rel) min: 0.04% max: 0.86% x̄: 0.29% x̃: 0.31%
HURT stats (abs) min: 2 max: 1020 x̄: 304.41 x̃: 232
HURT stats (rel) min: <.01% max: 1.58% x̄: 0.55% x̃: 0.43%
95% mean confidence interval for cycles value: -165.37 180.72
95% mean confidence interval for cycles %-change: <.01% 0.27%
Inconclusive result (value mean confidence interval includes 0).
total spills in shared programs: 705 -> 591 (-16.17%)
spills in affected programs: 705 -> 591 (-16.17%)
helped: 1
HURT: 0
total fills in shared programs: 1501 -> 1353 (-9.86%)
fills in affected programs: 1501 -> 1353 (-9.86%)
helped: 1
HURT: 0
Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4142>
Ian Romanick [Wed, 4 Mar 2020 20:23:13 +0000 (12:23 -0800)]
soft-fp64/fadd: Rename aFrac and bFrac variables
Exchanging aFracHi / bFracHi and aFracLo / bFracLo should not affect the
result of the later call to __add64.
The main purpose of this commit is to prepare for "soft-fp64/fadd: Move
common code out of both branches of an if-statement".
v2: Fix a typo in a comment. Noticed by Matt.
Results on the 308 shaders extracted from the fp64 portion of the OpenGL
CTS:
Tiger Lake and Ice Lake had similar results. (Tiger Lake shown)
total instructions in shared programs: 812094 -> 818246 (0.76%)
instructions in affected programs: 750271 -> 756423 (0.82%)
helped: 0
HURT: 74
HURT stats (abs) min: 7 max: 520 x̄: 83.14 x̃: 59
HURT stats (rel) min: 0.52% max: 1.48% x̄: 0.89% x̃: 0.84%
95% mean confidence interval for instructions value: 63.96 102.31
95% mean confidence interval for instructions %-change: 0.83% 0.95%
Instructions are HURT.
total cycles in shared programs: 6797157 -> 6816686 (0.29%)
cycles in affected programs: 6365175 -> 6384704 (0.31%)
helped: 0
HURT: 74
HURT stats (abs) min: 16 max: 1690 x̄: 263.91 x̃: 181
HURT stats (rel) min: 0.14% max: 0.68% x̄: 0.32% x̃: 0.27%
95% mean confidence interval for cycles value: 199.74 328.07
95% mean confidence interval for cycles %-change: 0.29% 0.36%
Cycles are HURT.
total spills in shared programs: 703 -> 705 (0.28%)
spills in affected programs: 703 -> 705 (0.28%)
helped: 0
HURT: 1
total fills in shared programs: 1499 -> 1501 (0.13%)
fills in affected programs: 1499 -> 1501 (0.13%)
helped: 0
HURT: 1
Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4142>
Ian Romanick [Wed, 4 Mar 2020 20:13:32 +0000 (12:13 -0800)]
soft-fp64/fadd: Combine an if-statement into the preceeding else-clause
The main purpose of this commit is to prepare for "soft-fp64/fadd: Move
common code out of both branches of an if-statement".
Results on the 308 shaders extracted from the fp64 portion of the OpenGL
CTS:
Tiger Lake and Ice Lake had similar results. (Tiger Lake shown)
total instructions in shared programs: 812590 -> 812094 (-0.06%)
instructions in affected programs: 672135 -> 671639 (-0.07%)
helped: 57
HURT: 0
helped stats (abs) min: 1 max: 32 x̄: 8.70 x̃: 7
helped stats (rel) min: <.01% max: 0.49% x̄: 0.12% x̃: 0.09%
95% mean confidence interval for instructions value: -10.46 -6.94
95% mean confidence interval for instructions %-change: -0.15% -0.09%
Instructions are helped.
total cycles in shared programs: 6798039 -> 6797157 (-0.01%)
cycles in affected programs: 5810059 -> 5809177 (-0.02%)
helped: 54
HURT: 2
helped stats (abs) min: 2 max: 68 x̄: 16.44 x̃: 12
helped stats (rel) min: <.01% max: 0.12% x̄: 0.03% x̃: 0.02%
HURT stats (abs) min: 2 max: 4 x̄: 3.00 x̃: 3
HURT stats (rel) min: <.01% max: <.01% x̄: <.01% x̃: <.01%
95% mean confidence interval for cycles value: -19.50 -12.00
95% mean confidence interval for cycles %-change: -0.03% -0.02%
Cycles are helped.
Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4142>
Ian Romanick [Wed, 4 Mar 2020 20:09:58 +0000 (12:09 -0800)]
soft-fp64/fadd: Reformat after previous commit
Convert
} else if (...) {
...
} else {
...
}
to
} else {
if (...) {
...
} else {
...
}
}
Not doing this reformatting in the previous commit makes the previous
commit easier to review, and doing it before the next commit makes the
next commit easier to review.
Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4142>
Ian Romanick [Wed, 4 Mar 2020 20:05:06 +0000 (12:05 -0800)]
soft-fp64/fadd: Delete a redundant condition check
Previous condition checks already guaranteen that expDiff != 0 and
!(expDiff > 0), so expDiff < 0 is the only option left.
The main purpose of this commit is to prepare for "soft-fp64/fadd: Move
common code out of both branches of an if-statement".
Results on the 308 shaders extracted from the fp64 portion of the OpenGL
CTS:
Tiger Lake and Ice Lake had similar results. (Tiger Lake shown)
total instructions in shared programs: 815491 -> 812590 (-0.36%)
instructions in affected programs: 753668 -> 750767 (-0.38%)
helped: 74
HURT: 0
helped stats (abs) min: 3 max: 281 x̄: 39.20 x̃: 25
helped stats (rel) min: 0.29% max: 0.73% x̄: 0.42% x̃: 0.40%
95% mean confidence interval for instructions value: -48.50 -29.91
95% mean confidence interval for instructions %-change: -0.45% -0.40%
Instructions are helped.
total cycles in shared programs: 6813681 -> 6798039 (-0.23%)
cycles in affected programs: 6381699 -> 6366057 (-0.25%)
helped: 74
HURT: 0
helped stats (abs) min: 24 max: 1488 x̄: 211.38 x̃: 149
helped stats (rel) min: 0.20% max: 0.44% x̄: 0.26% x̃: 0.25%
95% mean confidence interval for cycles value: -261.68 -161.08
95% mean confidence interval for cycles %-change: -0.28% -0.25%
Cycles are helped.
Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4142>
Ian Romanick [Wed, 4 Mar 2020 03:51:16 +0000 (19:51 -0800)]
soft-fp64/fadd: Just let the subtraction happen when the result will be zero
The main purpose of this commit is to prepare for "soft-fp64/fadd: Move
common code out of both branches of an if-statement".
Results on the 308 shaders extracted from the fp64 portion of the OpenGL
CTS:
Tiger Lake and Ice Lake had similar results. (Tiger Lake shown)
total instructions in shared programs: 815717 -> 815491 (-0.03%)
instructions in affected programs: 735489 -> 735263 (-0.03%)
helped: 39
HURT: 34
helped stats (abs) min: 2 max: 192 x̄: 20.79 x̃: 12
helped stats (rel) min: 0.01% max: 0.46% x̄: 0.26% x̃: 0.28%
HURT stats (abs) min: 1 max: 65 x̄: 17.21 x̃: 11
HURT stats (rel) min: <.01% max: 1.11% x̄: 0.35% x̃: 0.19%
95% mean confidence interval for instructions value: -10.40 4.21
95% mean confidence interval for instructions %-change: -0.07% 0.13%
Inconclusive result (value mean confidence interval includes 0).
total cycles in shared programs: 6820707 -> 6813681 (-0.10%)
cycles in affected programs: 6388725 -> 6381699 (-0.11%)
helped: 51
HURT: 23
helped stats (abs) min: 3 max: 1837 x̄: 184.76 x̃: 120
helped stats (rel) min: <.01% max: 0.48% x̄: 0.25% x̃: 0.25%
HURT stats (abs) min: 18 max: 216 x̄: 104.22 x̃: 98
HURT stats (rel) min: 0.06% max: 0.73% x̄: 0.31% x̃: 0.11%
95% mean confidence interval for cycles value: -154.67 -35.22
95% mean confidence interval for cycles %-change: -0.15% <.01%
Inconclusive result (%-change mean confidence interval includes 0).
total spills in shared programs: 702 -> 703 (0.14%)
spills in affected programs: 702 -> 703 (0.14%)
helped: 0
HURT: 1
total fills in shared programs: 1497 -> 1499 (0.13%)
fills in affected programs: 1497 -> 1499 (0.13%)
helped: 0
HURT: 1
Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4142>
Ian Romanick [Wed, 4 Mar 2020 03:49:13 +0000 (19:49 -0800)]
soft-fp64/fadd: Pick zero or non-zero result based on subtraction result
The main purpose of this commit is to prepare for "soft-fp64/fadd: Move
common code out of both branches of an if-statement".
Results on the 308 shaders extracted from the fp64 portion of the OpenGL
CTS:
Tiger Lake and Ice Lake had similar results. (Tiger Lake shown)
total instructions in shared programs: 817327 -> 815717 (-0.20%)
instructions in affected programs: 755504 -> 753894 (-0.21%)
helped: 73
HURT: 1
helped stats (abs) min: 1 max: 159 x̄: 22.12 x̃: 14
helped stats (rel) min: 0.05% max: 0.40% x̄: 0.22% x̃: 0.23%
HURT stats (abs) min: 5 max: 5 x̄: 5.00 x̃: 5
HURT stats (rel) min: 0.07% max: 0.07% x̄: 0.07% x̃: 0.07%
95% mean confidence interval for instructions value: -27.27 -16.24
95% mean confidence interval for instructions %-change: -0.24% -0.20%
Instructions are helped.
total cycles in shared programs: 6822826 -> 6820707 (-0.03%)
cycles in affected programs: 6390844 -> 6388725 (-0.03%)
helped: 71
HURT: 3
helped stats (abs) min: 2 max: 537 x̄: 30.72 x̃: 18
helped stats (rel) min: <.01% max: 0.08% x̄: 0.03% x̃: 0.03%
HURT stats (abs) min: 10 max: 32 x̄: 20.67 x̃: 20
HURT stats (rel) min: 0.01% max: 0.02% x̄: 0.02% x̃: 0.02%
95% mean confidence interval for cycles value: -43.41 -13.86
95% mean confidence interval for cycles %-change: -0.04% -0.03%
Cycles are helped.
Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4142>
Ian Romanick [Wed, 4 Mar 2020 03:44:13 +0000 (19:44 -0800)]
soft-fp64/fadd: Massively split the live range of zFrac0 and zFrac1
The main purpose of this commit is to prepare for "soft-fp64/fadd: Move
common code out of both branches of an if-statement".
Results on the 308 shaders extracted from the fp64 portion of the OpenGL
CTS:
Tiger Lake and Ice Lake had similar results. (Tiger Lake shown)
total instructions in shared programs: 822766 -> 817327 (-0.66%)
instructions in affected programs: 760943 -> 755504 (-0.71%)
helped: 74
HURT: 0
helped stats (abs) min: 8 max: 515 x̄: 73.50 x̃: 51
helped stats (rel) min: 0.58% max: 1.10% x̄: 0.77% x̃: 0.73%
95% mean confidence interval for instructions value: -91.17 -55.83
95% mean confidence interval for instructions %-change: -0.81% -0.74%
Instructions are helped.
total cycles in shared programs: 6816791 -> 6822826 (0.09%)
cycles in affected programs: 6384809 -> 6390844 (0.09%)
helped: 0
HURT: 74
HURT stats (abs) min: 6 max: 1179 x̄: 81.55 x̃: 50
HURT stats (rel) min: 0.02% max: 0.17% x̄: 0.09% x̃: 0.09%
95% mean confidence interval for cycles value: 48.99 114.12
95% mean confidence interval for cycles %-change: 0.09% 0.10%
Cycles are HURT.
Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4142>
Ian Romanick [Wed, 4 Mar 2020 03:38:13 +0000 (19:38 -0800)]
soft-fp64/fadd: Instead of tracking "b < a", track sign of the difference
Results on the 308 shaders extracted from the fp64 portion of the OpenGL
CTS:
Tiger Lake and Ice Lake had similar results. (Tiger Lake shown)
total instructions in shared programs: 824403 -> 822766 (-0.20%)
instructions in affected programs: 756260 -> 754623 (-0.22%)
helped: 68
HURT: 1
helped stats (abs) min: 1 max: 118 x̄: 26.26 x̃: 18
helped stats (rel) min: 0.02% max: 0.97% x̄: 0.31% x̃: 0.23%
HURT stats (abs) min: 149 max: 149 x̄: 149.00 x̃: 149
HURT stats (rel) min: 0.17% max: 0.17% x̄: 0.17% x̃: 0.17%
95% mean confidence interval for instructions value: -31.94 -15.51
95% mean confidence interval for instructions %-change: -0.37% -0.23%
Instructions are helped.
total cycles in shared programs: 6828935 -> 6816791 (-0.18%)
cycles in affected programs: 6385191 -> 6373047 (-0.19%)
helped: 73
HURT: 0
helped stats (abs) min: 2 max: 852 x̄: 166.36 x̃: 120
helped stats (rel) min: <.01% max: 0.80% x̄: 0.22% x̃: 0.17%
95% mean confidence interval for cycles value: -210.80 -121.91
95% mean confidence interval for cycles %-change: -0.27% -0.17%
Cycles are helped.
total fills in shared programs: 1442 -> 1497 (3.81%)
fills in affected programs: 1442 -> 1497 (3.81%)
helped: 0
HURT: 1
Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4142>
Ian Romanick [Tue, 3 Mar 2020 18:21:18 +0000 (10:21 -0800)]
soft-fp64: Optimize __fmin64 and __fmax64 by using different evaluation order [v2]
v2: Go to extra effort to avoid flow control inserted to implement
short-circuit evaluation rules.
Results on the 308 shaders extracted from the fp64 portion of the OpenGL
CTS:
Tiger Lake and Ice Lake had similar results. (Tiger Lake shown)
total instructions in shared programs: 797779 -> 796849 (-0.12%)
instructions in affected programs: 3499 -> 2569 (-26.58%)
helped: 21
HURT: 0
helped stats (abs) min: 8 max: 112 x̄: 44.29 x̃: 44
helped stats (rel) min: 16.09% max: 33.15% x̄: 25.72% x̃: 24.62%
95% mean confidence interval for instructions value: -55.94 -32.63
95% mean confidence interval for instructions %-change: -28.14% -23.30%
Instructions are helped.
total cycles in shared programs: 6601355 -> 6588351 (-0.20%)
cycles in affected programs: 25376 -> 12372 (-51.25%)
helped: 21
HURT: 0
helped stats (abs) min: 156 max: 1410 x̄: 619.24 x̃: 526
helped stats (rel) min: 42.39% max: 53.98% x̄: 50.12% x̃: 50.75%
95% mean confidence interval for cycles value: -776.58 -461.89
95% mean confidence interval for cycles %-change: -51.57% -48.67%
Cycles are helped.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net> [v1]
Reviewed-by: Matt Turner <mattst88@gmail.com> [v1]
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4142>
Ian Romanick [Wed, 4 Mar 2020 22:21:02 +0000 (14:21 -0800)]
soft-fp64/ffloor: Simplify the >= 0 comparison
Results on the 308 shaders extracted from the fp64 portion of the OpenGL
CTS:
Tiger Lake and Ice Lake had similar results. (Tiger Lake shown)
total instructions in shared programs: 797951 -> 797779 (-0.02%)
instructions in affected programs: 126482 -> 126310 (-0.14%)
helped: 15
HURT: 0
helped stats (abs) min: 1 max: 20 x̄: 11.47 x̃: 10
helped stats (rel) min: <.01% max: 0.60% x̄: 0.28% x̃: 0.29%
95% mean confidence interval for instructions value: -14.79 -8.14
95% mean confidence interval for instructions %-change: -0.40% -0.16%
Instructions are helped.
total cycles in shared programs: 6601437 -> 6601355 (<.01%)
cycles in affected programs: 1089336 -> 1089254 (<.01%)
helped: 15
HURT: 0
helped stats (abs) min: 2 max: 12 x̄: 5.47 x̃: 6
helped stats (rel) min: <.01% max: 0.04% x̄: 0.01% x̃: 0.01%
95% mean confidence interval for cycles value: -7.06 -3.87
95% mean confidence interval for cycles %-change: -0.02% <.01%
Cycles are helped.
Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4142>
Ian Romanick [Tue, 3 Mar 2020 03:20:42 +0000 (19:20 -0800)]
soft-fp64: Relax the way NaN is propagated
Also reassociate a couple expressions to encourage some CSE.
Results on the 308 shaders extracted from the fp64 portion of the OpenGL
CTS:
Tiger Lake and Ice Lake had similar results. (Tiger Lake shown)
total instructions in shared programs: 813599 -> 797951 (-1.92%)
instructions in affected programs: 796110 -> 780462 (-1.97%)
helped: 92
HURT: 0
helped stats (abs) min: 3 max: 5198 x̄: 170.09 x̃: 83
helped stats (rel) min: 0.36% max: 5.50% x̄: 1.57% x̃: 1.40%
95% mean confidence interval for instructions value: -282.42 -57.75
95% mean confidence interval for instructions %-change: -1.71% -1.42%
Instructions are helped.
total cycles in shared programs: 6687128 -> 6601437 (-1.28%)
cycles in affected programs: 6582246 -> 6496555 (-1.30%)
helped: 92
HURT: 0
helped stats (abs) min: 36 max: 14442 x̄: 931.42 x̃: 592
helped stats (rel) min: 0.45% max: 3.16% x̄: 1.44% x̃: 1.23%
95% mean confidence interval for cycles value: -1257.58 -605.27
95% mean confidence interval for cycles %-change: -1.58% -1.30%
Cycles are helped.
total spills in shared programs: 759 -> 702 (-7.51%)
spills in affected programs: 759 -> 702 (-7.51%)
helped: 3
HURT: 0
total fills in shared programs: 2412 -> 1442 (-40.22%)
fills in affected programs: 2412 -> 1442 (-40.22%)
helped: 3
HURT: 0
Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4142>
Ian Romanick [Tue, 3 Mar 2020 02:50:44 +0000 (18:50 -0800)]
soft-fp64/fsat: Micro-optimize x >= 1 test
Results on the 308 shaders extracted from the fp64 portion of the OpenGL
CTS:
Tiger Lake and Ice Lake had similar results. (Tiger Lake shown)
total instructions in shared programs: 841590 -> 841332 (-0.03%)
instructions in affected programs: 121957 -> 121699 (-0.21%)
helped: 7
HURT: 0
helped stats (abs) min: 15 max: 54 x̄: 36.86 x̃: 41
helped stats (rel) min: 0.16% max: 0.33% x̄: 0.23% x̃: 0.18%
95% mean confidence interval for instructions value: -49.73 -23.98
95% mean confidence interval for instructions %-change: -0.29% -0.16%
Instructions are helped.
total cycles in shared programs: 6926828 -> 6923967 (-0.04%)
cycles in affected programs: 1038569 -> 1035708 (-0.28%)
helped: 7
HURT: 0
helped stats (abs) min: 128 max: 616 x̄: 408.71 x̃: 446
helped stats (rel) min: 0.18% max: 0.44% x̄: 0.29% x̃: 0.22%
95% mean confidence interval for cycles value: -571.72 -245.70
95% mean confidence interval for cycles %-change: -0.38% -0.19%
Cycles are helped.
Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4142>
Ian Romanick [Tue, 3 Mar 2020 02:42:04 +0000 (18:42 -0800)]
soft-fp64/fsat: Micro-optimize x < 0 test
Results on the 308 shaders extracted from the fp64 portion of the OpenGL
CTS:
Tiger Lake and Ice Lake had similar results. (Tiger Lake shown)
total instructions in shared programs: 841647 -> 841590 (<.01%)
instructions in affected programs: 122014 -> 121957 (-0.05%)
helped: 7
HURT: 0
helped stats (abs) min: 3 max: 12 x̄: 8.14 x̃: 9
helped stats (rel) min: 0.04% max: 0.07% x̄: 0.05% x̃: 0.04%
95% mean confidence interval for instructions value: -11.23 -5.06
95% mean confidence interval for instructions %-change: -0.06% -0.03%
Instructions are helped.
total cycles in shared programs: 6926904 -> 6926828 (<.01%)
cycles in affected programs: 1038645 -> 1038569 (<.01%)
helped: 7
HURT: 0
helped stats (abs) min: 4 max: 16 x̄: 10.86 x̃: 12
helped stats (rel) min: <.01% max: 0.01% x̄: <.01% x̃: <.01%
95% mean confidence interval for cycles value: -14.97 -6.74
95% mean confidence interval for cycles %-change: -0.01% <.01%
Cycles are helped.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4142>
Ian Romanick [Tue, 3 Mar 2020 02:38:11 +0000 (18:38 -0800)]
soft-fp64/fsat: Correctly handle NaN
fsat is defined as min(max(a, 0.0), 1.0), and IEEE defines both min and
max to return the non-NaN value when one value is NaN. Based on this,
fsat should definitely return 0.0 for NaN.
Results on the 308 shaders extracted from the fp64 portion of the OpenGL
CTS:
Tiger Lake and Ice Lake had similar results. (Tiger Lake shown)
total instructions in shared programs: 841666 -> 841647 (<.01%)
instructions in affected programs: 122033 -> 122014 (-0.02%)
helped: 7
HURT: 0
helped stats (abs) min: 1 max: 4 x̄: 2.71 x̃: 3
helped stats (rel) min: 0.01% max: 0.02% x̄: 0.02% x̃: 0.01%
95% mean confidence interval for instructions value: -3.74 -1.69
95% mean confidence interval for instructions %-change: -0.02% -0.01%
Instructions are helped.
total cycles in shared programs: 6927246 -> 6926904 (<.01%)
cycles in affected programs: 1038987 -> 1038645 (-0.03%)
helped: 7
HURT: 0
helped stats (abs) min: 18 max: 72 x̄: 48.86 x̃: 54
helped stats (rel) min: 0.03% max: 0.05% x̄: 0.03% x̃: 0.03%
95% mean confidence interval for cycles value: -67.38 -30.33
95% mean confidence interval for cycles %-change: -0.05% -0.02%
Cycles are helped.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Fixes:
a42163cbbc1 ("compiler: Add lowering support for 64-bit saturate operations to software")
Closes: https://gitlab.freedesktop.org/mesa/mesa/issues/2585
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4142>
Ian Romanick [Tue, 3 Mar 2020 02:57:44 +0000 (18:57 -0800)]
soft-fp64/flt: Perform checks in a different order
The change to nir_opt_algebraic cleans up a pattern that was never
produced before the rest of this commit was added.
Results on the 308 shaders extracted from the fp64 portion of the OpenGL
CTS:
Tiger Lake and Ice Lake had similar results. (Tiger Lake shown)
total instructions in shared programs: 843005 -> 841666 (-0.16%)
instructions in affected programs: 460655 -> 459316 (-0.29%)
helped: 64
HURT: 17
helped stats (abs) min: 1 max: 72 x̄: 21.72 x̃: 20
helped stats (rel) min: 0.01% max: 28.07% x̄: 12.67% x̃: 16.07%
HURT stats (abs) min: 1 max: 7 x̄: 3.00 x̃: 2
HURT stats (rel) min: 0.01% max: 0.04% x̄: 0.02% x̃: 0.02%
95% mean confidence interval for instructions value: -20.87 -12.19
95% mean confidence interval for instructions %-change: -12.35% -7.66%
Instructions are helped.
total cycles in shared programs: 6944998 -> 6927246 (-0.26%)
cycles in affected programs: 3891872 -> 3874120 (-0.46%)
helped: 71
HURT: 10
helped stats (abs) min: 2 max: 772 x̄: 254.21 x̃: 156
helped stats (rel) min: <.01% max: 66.44% x̄: 21.72% x̃: 18.40%
HURT stats (abs) min: 18 max: 69 x̄: 29.70 x̃: 20
HURT stats (rel) min: 0.02% max: 0.04% x̄: 0.03% x̃: 0.03%
95% mean confidence interval for cycles value: -270.82 -167.50
95% mean confidence interval for cycles %-change: -24.41% -13.65%
Cycles are helped.
Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4142>
Ian Romanick [Tue, 3 Mar 2020 02:31:01 +0000 (18:31 -0800)]
soft-fp64/fneg: Don't treat NaN specially
__fabs64 doesn't do anything special, and the value is still NaN
regardless of the value of the MSB. In a strict sense, it's possible
that both functions should set the "signal" bit.
lts on the 308 shaders extracted from the fp64 portion of the OpenGL
CTS:
Tiger Lake and Ice Lake had similar results. (Tiger Lake shown)
total instructions in shared programs: 844558 -> 843005 (-0.18%)
instructions in affected programs: 725975 -> 724422 (-0.21%)
helped: 53
HURT: 4
helped stats (abs) min: 1 max: 313 x̄: 29.87 x̃: 21
helped stats (rel) min: 0.01% max: 0.94% x̄: 0.30% x̃: 0.22%
HURT stats (abs) min: 4 max: 11 x̄: 7.50 x̃: 7
HURT stats (rel) min: 0.03% max: 0.09% x̄: 0.05% x̃: 0.04%
95% mean confidence interval for instructions value: -39.02 -15.47
95% mean confidence interval for instructions %-change: -0.34% -0.21%
Instructions are helped.
total cycles in shared programs: 6962024 -> 6944998 (-0.24%)
cycles in affected programs: 6185470 -> 6168444 (-0.28%)
helped: 59
HURT: 0
helped stats (abs) min: 64 max: 2863 x̄: 288.58 x̃: 208
helped stats (rel) min: 0.11% max: 0.87% x̄: 0.33% x̃: 0.27%
95% mean confidence interval for cycles value: -387.15 -190.00
95% mean confidence interval for cycles %-change: -0.38% -0.28%
Cycles are helped.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4142>
Ian Romanick [Tue, 3 Mar 2020 00:37:58 +0000 (16:37 -0800)]
soft-fp64: Store sign value as 0 or 0x80000000
...instead of 0 or 1. Many places the sign bit is extracted, then later
put back in the same position. This saves some left-shift operations.
Results on the 308 shaders extracted from the fp64 portion of the OpenGL
CTS:
Tiger Lake and Ice Lake had similar results. (Tiger Lake shown)
total instructions in shared programs: 848106 -> 844558 (-0.42%)
instructions in affected programs: 833480 -> 829932 (-0.43%)
helped: 106
HURT: 1
helped stats (abs) min: 1 max: 995 x̄: 33.48 x̃: 12
helped stats (rel) min: 0.15% max: 2.20% x̄: 0.60% x̃: 0.35%
HURT stats (abs) min: 1 max: 1 x̄: 1.00 x̃: 1
HURT stats (rel) min: <.01% max: <.01% x̄: <.01% x̃: <.01%
95% mean confidence interval for instructions value: -51.88 -14.43
95% mean confidence interval for instructions %-change: -0.71% -0.47%
Instructions are helped.
total cycles in shared programs: 6969125 -> 6962024 (-0.10%)
cycles in affected programs: 6717689 -> 6710588 (-0.11%)
helped: 78
HURT: 7
helped stats (abs) min: 2 max: 2083 x̄: 110.27 x̃: 56
helped stats (rel) min: <.01% max: 0.30% x̄: 0.11% x̃: 0.11%
HURT stats (abs) min: 2 max: 1340 x̄: 214.29 x̃: 4
HURT stats (rel) min: 0.01% max: 0.71% x̄: 0.13% x̃: 0.02%
95% mean confidence interval for cycles value: -144.02 -23.06
95% mean confidence interval for cycles %-change: -0.12% -0.07%
Cycles are helped.
total spills in shared programs: 814 -> 759 (-6.76%)
spills in affected programs: 814 -> 759 (-6.76%)
helped: 2
HURT: 1
total fills in shared programs: 2488 -> 2412 (-3.05%)
fills in affected programs: 2488 -> 2412 (-3.05%)
helped: 2
HURT: 1
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4142>
Ian Romanick [Tue, 3 Mar 2020 00:33:16 +0000 (16:33 -0800)]
soft-fp64: Pick a single idiom for treating sign value as a Boolean
Replace all of the bool(qSign) with qSign != 0u. Remove unnecessary
parenthesis from around most of the existing qSign != 0u.
This dramatically simplifies the next commit.
Results on the 308 shaders extracted from the fp64 portion of the OpenGL
CTS:
Tiger Lake and Ice Lake had similar results. (Tiger Lake shown)
total instructions in shared programs: 848109 -> 848106 (<.01%)
instructions in affected programs: 53 -> 50 (-5.66%)
helped: 1
HURT: 0
total cycles in shared programs: 6969145 -> 6969125 (<.01%)
cycles in affected programs: 396 -> 376 (-5.05%)
helped: 1
HURT: 0
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4142>
Ian Romanick [Tue, 3 Mar 2020 00:43:11 +0000 (16:43 -0800)]
soft-fp64: Simplify __countLeadingZeros32 function
findMSB returns -1 for an input of zero, so 31 - findMSB(a) is
sufficient on its own.
There's only one user of findMSB in shader-db, and it does not match
this pattern.
TODO: Add a pattern in the backend code generator that emits 31 -
nir_op_ufind_msb(a) as if it were nir_op_uclz. That should save a couple
instructions.
Results on the 308 shaders extracted from the fp64 portion of the OpenGL
CTS:
Tiger Lake and Ice Lake had similar results. (Tiger Lake shown)
total instructions in shared programs: 859509 -> 848109 (-1.33%)
instructions in affected programs: 841058 -> 829658 (-1.36%)
helped: 97
HURT: 0
helped stats (abs) min: 3 max: 1161 x̄: 117.53 x̃: 72
helped stats (rel) min: 0.98% max: 6.74% x̄: 1.70% x̃: 1.35%
95% mean confidence interval for instructions value: -147.21 -87.84
95% mean confidence interval for instructions %-change: -1.94% -1.46%
Instructions are helped.
total cycles in shared programs: 7072275 -> 6969145 (-1.46%)
cycles in affected programs: 6955767 -> 6852637 (-1.48%)
helped: 97
HURT: 0
helped stats (abs) min: 32 max: 10900 x̄: 1063.20 x̃: 560
helped stats (rel) min: 1.18% max: 7.58% x̄: 1.84% x̃: 1.45%
95% mean confidence interval for cycles value: -1339.43 -786.96
95% mean confidence interval for cycles %-change: -2.11% -1.57%
Cycles are helped.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4142>
Ian Romanick [Tue, 3 Mar 2020 20:26:37 +0000 (12:26 -0800)]
soft-fp64: Don't open-code umulExtended
Results on the 308 shaders extracted from the fp64 portion of the OpenGL
CTS:
Tiger Lake and Ice Lake had similar results. (Tiger Lake shown)
total instructions in shared programs: 928859 -> 859509 (-7.47%)
instructions in affected programs: 866293 -> 796943 (-8.01%)
helped: 76
HURT: 0
helped stats (abs) min: 75 max: 8042 x̄: 912.50 x̃: 688
helped stats (rel) min: 5.35% max: 21.02% x̄: 10.35% x̃: 7.58%
95% mean confidence interval for instructions value: -1138.37 -686.63
95% mean confidence interval for instructions %-change: -11.69% -9.00%
Instructions are helped.
total cycles in shared programs: 7272912 -> 7072275 (-2.76%)
cycles in affected programs: 6763486 -> 6562849 (-2.97%)
helped: 76
HURT: 0
helped stats (abs) min: 214 max: 30136 x̄: 2639.96 x̃: 1923
helped stats (rel) min: 1.75% max: 9.20% x̄: 4.04% x̃: 2.41%
95% mean confidence interval for cycles value: -3455.29 -1824.63
95% mean confidence interval for cycles %-change: -4.69% -3.39%
Cycles are helped.
total spills in shared programs: 817 -> 814 (-0.37%)
spills in affected programs: 791 -> 788 (-0.38%)
helped: 2
HURT: 0
total fills in shared programs: 2438 -> 2488 (2.05%)
fills in affected programs: 2392 -> 2442 (2.09%)
helped: 0
HURT: 2
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4142>
Ian Romanick [Thu, 5 Mar 2020 00:53:36 +0000 (16:53 -0800)]
soft-fp64/b2f: Reimplement using bitwise logic ops
This doesn't help a lot of shaders, but it helps those few a LOT.
This could also be implemented using bcsel. That version is very
slightly worse because the generated SEL instruction wants to have two
immediate sources, so one of them usually needs an extra MOV instruction
to load.
Results on the 308 shaders extracted from the fp64 portion of the OpenGL
CTS:
Tiger Lake and Ice Lake had similar results. (Tiger Lake shown)
total instructions in shared programs: 929619 -> 928859 (-0.08%)
instructions in affected programs: 1651 -> 891 (-46.03%)
helped: 8
HURT: 0
helped stats (abs) min: 38 max: 152 x̄: 95.00 x̃: 95
helped stats (rel) min: 42.70% max: 86.36% x̄: 49.88% x̃: 44.66%
95% mean confidence interval for instructions value: -132.97 -57.03
95% mean confidence interval for instructions %-change: -62.28% -37.49%
Instructions are helped.
total cycles in shared programs: 7280180 -> 7272912 (-0.10%)
cycles in affected programs: 12960 -> 5692 (-56.08%)
helped: 8
HURT: 0
helped stats (abs) min: 352 max: 1456 x̄: 908.50 x̃: 910
helped stats (rel) min: 52.45% max: 91.19% x̄: 59.24% x̃: 55.15%
95% mean confidence interval for cycles value: -1274.03 -542.97
95% mean confidence interval for cycles %-change: -70.06% -48.41%
Cycles are helped.
Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4142>
Ian Romanick [Fri, 6 Mar 2020 02:03:35 +0000 (18:03 -0800)]
nir/algebraic: Simplify a contradiction that can occur in __flt64_nonnan
The pattern is added to opt_algebraic because, for example, comparisons
with constant 0.0 will produce (a1 < 0).
Even with a pass that optimized Boolean expressions, I think this would
be very difficult to automatically recognize and optimize.
Results on the 308 shaders extracted from the fp64 portion of the OpenGL
CTS:
Tiger Lake and Ice Lake had similar results. (Tiger Lake shown)
total instructions in shared programs: 933054 -> 929619 (-0.37%)
instructions in affected programs: 784041 -> 780606 (-0.44%)
helped: 59
HURT: 0
helped stats (abs) min: 2 max: 213 x̄: 58.22 x̃: 44
helped stats (rel) min: 0.02% max: 2.51% x̄: 0.72% x̃: 0.46%
95% mean confidence interval for instructions value: -70.80 -45.64
95% mean confidence interval for instructions %-change: -0.92% -0.53%
Instructions are helped.
total cycles in shared programs: 7304712 -> 7280180 (-0.34%)
cycles in affected programs: 7176260 -> 7151728 (-0.34%)
helped: 92
HURT: 0
helped stats (abs) min: 8 max: 1414 x̄: 266.65 x̃: 166
helped stats (rel) min: 0.04% max: 2.34% x̄: 0.43% x̃: 0.22%
95% mean confidence interval for cycles value: -333.05 -200.26
95% mean confidence interval for cycles %-change: -0.54% -0.31%
Cycles are helped.
Regular shader-db changes:
No changes on any Intel platform.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4142>
Ian Romanick [Sat, 7 Mar 2020 00:05:27 +0000 (16:05 -0800)]
nir/algebraic: Constant reassociation for bitwise operations too
Like
5886cd79a0e, but for iand, ior, and ixor.
Results on the 308 shaders extracted from the fp64 portion of the OpenGL
CTS:
Tiger Lake
total instructions in shared programs: 903108 -> 902830 (-0.03%)
instructions in affected programs: 654910 -> 654632 (-0.04%)
helped: 31
HURT: 5
helped stats (abs) min: 2 max: 31 x̄: 9.58 x̃: 7
helped stats (rel) min: 0.01% max: 0.23% x̄: 0.06% x̃: 0.04%
HURT stats (abs) min: 1 max: 10 x̄: 3.80 x̃: 3
HURT stats (rel) min: 0.01% max: 0.10% x̄: 0.03% x̃: 0.02%
95% mean confidence interval for instructions value: -10.55 -4.89
95% mean confidence interval for instructions %-change: -0.07% -0.03%
Instructions are helped.
total cycles in shared programs: 7059681 -> 7058006 (-0.02%)
cycles in affected programs: 5081309 -> 5079634 (-0.03%)
helped: 33
HURT: 12
helped stats (abs) min: 1 max: 444 x̄: 60.91 x̃: 18
helped stats (rel) min: <.01% max: 2.17% x̄: 0.25% x̃: 0.05%
HURT stats (abs) min: 1 max: 288 x̄: 27.92 x̃: 2
HURT stats (rel) min: <.01% max: 1.00% x̄: 0.23% x̃: 0.02%
95% mean confidence interval for cycles value: -68.32 -6.12
95% mean confidence interval for cycles %-change: -0.28% 0.03%
Inconclusive result (%-change mean confidence interval includes 0).
Ice Lake
total instructions in shared programs: 895384 -> 895159 (-0.03%)
instructions in affected programs: 658678 -> 658453 (-0.03%)
helped: 37
HURT: 0
helped stats (abs) min: 3 max: 16 x̄: 6.08 x̃: 4
helped stats (rel) min: <.01% max: 0.07% x̄: 0.04% x̃: 0.04%
95% mean confidence interval for instructions value: -7.46 -4.70
95% mean confidence interval for instructions %-change: -0.04% -0.03%
Instructions are helped.
total cycles in shared programs: 7092224 -> 7091195 (-0.01%)
cycles in affected programs: 5221666 -> 5220637 (-0.02%)
helped: 35
HURT: 11
helped stats (abs) min: 1 max: 247 x̄: 43.46 x̃: 12
helped stats (rel) min: <.01% max: 2.17% x̄: 0.23% x̃: 0.05%
HURT stats (abs) min: 2 max: 432 x̄: 44.73 x̃: 5
HURT stats (rel) min: <.01% max: 1.00% x̄: 0.25% x̃: 0.02%
95% mean confidence interval for cycles value: -49.00 4.26
95% mean confidence interval for cycles %-change: -0.27% 0.03%
Inconclusive result (value mean confidence interval includes 0).
Regular shader-db results:
All Haswell+ platforms had similar results. (Tiger Lake shown)
total instructions in shared programs:
17611408 ->
17611398 (<.01%)
instructions in affected programs: 1648 -> 1638 (-0.61%)
helped: 2
HURT: 0
total cycles in shared programs:
338366148 ->
338366124 (<.01%)
cycles in affected programs: 124048 -> 124024 (-0.02%)
helped: 2
HURT: 0
No changes on any earlier Intel platforms.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4142>
Ian Romanick [Sat, 7 Mar 2020 00:23:29 +0000 (16:23 -0800)]
nir/algebraic: Generalize some and-of-shift-right patterns [v2]
Generalizes some of the patterns from
76289fbfa84a and
905ff8619824. In
particular, some of the soft-fp64 code generates (a & 0x7fffffff) << 1
when constant 0.0 is compared (flt or feq).
v2: Reduce the set of added patterns to those that actually help
something. This reduces the size of the state transition tables by
about 29k. Suggested by Jason. Remove the existing patterns that this
commit subsumes.
Results on the 308 shaders extracted from the fp64 portion of the OpenGL
CTS:
Tiger Lake
total instructions in shared programs: 903171 -> 903108 (<.01%)
instructions in affected programs: 635903 -> 635840 (<.01%)
helped: 25
HURT: 11
helped stats (abs) min: 1 max: 16 x̄: 5.04 x̃: 3
helped stats (rel) min: <.01% max: 0.15% x̄: 0.04% x̃: 0.03%
HURT stats (abs) min: 2 max: 14 x̄: 5.73 x̃: 5
HURT stats (rel) min: <.01% max: 0.11% x̄: 0.04% x̃: 0.02%
95% mean confidence interval for instructions value: -3.91 0.41
95% mean confidence interval for instructions %-change: -0.03% <.01%
Inconclusive result (value mean confidence interval includes 0).
total cycles in shared programs: 7059527 -> 7059681 (<.01%)
cycles in affected programs: 5249401 -> 5249555 (<.01%)
helped: 41
HURT: 9
helped stats (abs) min: 2 max: 76 x̄: 11.90 x̃: 10
helped stats (rel) min: <.01% max: 11.86% x̄: 0.99% x̃: 0.01%
HURT stats (abs) min: 2 max: 380 x̄: 71.33 x̃: 12
HURT stats (rel) min: <.01% max: 0.22% x̄: 0.04% x̃: 0.01%
95% mean confidence interval for cycles value: -14.93 21.09
95% mean confidence interval for cycles %-change: -1.40% -0.20%
Inconclusive result (value mean confidence interval includes 0).
Ice Lake
total instructions in shared programs: 895506 -> 895384 (-0.01%)
instructions in affected programs: 658800 -> 658678 (-0.02%)
helped: 37
HURT: 0
helped stats (abs) min: 2 max: 8 x̄: 3.30 x̃: 2
helped stats (rel) min: <.01% max: 0.03% x̄: 0.02% x̃: 0.02%
95% mean confidence interval for instructions value: -4.00 -2.59
95% mean confidence interval for instructions %-change: -0.02% -0.02%
Instructions are helped.
total cycles in shared programs: 7092748 -> 7092224 (<.01%)
cycles in affected programs: 5272008 -> 5271484 (<.01%)
helped: 36
HURT: 14
helped stats (abs) min: 2 max: 440 x̄: 21.67 x̃: 8
helped stats (rel) min: <.01% max: 11.86% x̄: 1.12% x̃: 0.02%
HURT stats (abs) min: 2 max: 122 x̄: 18.29 x̃: 6
HURT stats (rel) min: <.01% max: 0.07% x̄: 0.01% x̃: <.01%
95% mean confidence interval for cycles value: -29.24 8.28
95% mean confidence interval for cycles %-change: -1.40% -0.21%
Inconclusive result (value mean confidence interval includes 0).
Regular shader-db results:
All Haswell+ platforms had similar results. (Tiger Lake shown)
total instructions in shared programs:
17611489 ->
17611408 (<.01%)
instructions in affected programs: 21188 -> 21107 (-0.38%)
helped: 23
HURT: 1
helped stats (abs) min: 1 max: 16 x̄: 3.78 x̃: 3
helped stats (rel) min: 0.03% max: 5.82% x̄: 1.13% x̃: 0.85%
HURT stats (abs) min: 6 max: 6 x̄: 6.00 x̃: 6
HURT stats (rel) min: 0.60% max: 0.60% x̄: 0.60% x̃: 0.60%
95% mean confidence interval for instructions value: -5.27 -1.48
95% mean confidence interval for instructions %-change: -1.70% -0.42%
Instructions are helped.
total cycles in shared programs:
338418502 ->
338366148 (-0.02%)
cycles in affected programs: 2289052 -> 2236698 (-2.29%)
helped: 18
HURT: 3
helped stats (abs) min: 4 max: 18000 x̄: 2909.67 x̃: 38
helped stats (rel) min: 0.09% max: 4.07% x̄: 0.96% x̃: 0.43%
HURT stats (abs) min: 2 max: 14 x̄: 6.67 x̃: 4
HURT stats (rel) min: 0.22% max: 1.13% x̄: 0.66% x̃: 0.64%
95% mean confidence interval for cycles value: -5204.00 217.91
95% mean confidence interval for cycles %-change: -1.31% -0.14%
Inconclusive result (value mean confidence interval includes 0).
Ivy Bridge
total instructions in shared programs:
11875617 ->
11875615 (<.01%)
instructions in affected programs: 1339 -> 1337 (-0.15%)
helped: 2
HURT: 0
No changes on any earlier Intel platforms.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Matt Turner <mattst88@gmail.com> [v1]
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4142>
Ian Romanick [Thu, 31 Oct 2019 00:41:41 +0000 (17:41 -0700)]
nir/algebraic: optimize ior(ine(a, 0), ine(b, 0)) to ine(ior(a, b), 0)
Like
70f9e2589e6b. Also scrub the unnecessary size qualifier in both
replacement patterns.
This occurs in a handful of places in the soft-fp64 code, and that is
the primary reason for the change.
Perhaps the patterns that generate umin should be conditioned on
something, but I'm not sure what. lower_bitops might cover the cases
that matter, but it seems ugly.
Results on the 308 shaders extracted from the fp64 portion of the OpenGL
CTS:
Tiger Lake and Ice Lake had similar results. (Tiger Lake shown)
total instructions in shared programs: 936505 -> 933388 (-0.33%)
instructions in affected programs: 925719 -> 922602 (-0.34%)
helped: 154
HURT: 1
helped stats (abs) min: 1 max: 211 x̄: 35.45 x̃: 16
helped stats (rel) min: 0.34% max: 9.30% x̄: 2.28% x̃: 0.96%
HURT stats (abs) min: 2342 max: 2342 x̄: 2342.00 x̃: 2342
HURT stats (rel) min: 2.28% max: 2.28% x̄: 2.28% x̃: 2.28%
95% mean confidence interval for instructions value: -51.21 10.99
95% mean confidence interval for instructions %-change: -2.61% -1.89%
Inconclusive result (value mean confidence interval includes 0).
total cycles in shared programs: 7323502 -> 7306184 (-0.24%)
cycles in affected programs: 7220376 -> 7203058 (-0.24%)
helped: 126
HURT: 1
helped stats (abs) min: 2 max: 946 x̄: 159.10 x̃: 95
helped stats (rel) min: 0.01% max: 9.62% x̄: 0.80% x̃: 0.37%
HURT stats (abs) min: 2728 max: 2728 x̄: 2728.00 x̃: 2728
HURT stats (rel) min: 0.37% max: 0.37% x̄: 0.37% x̃: 0.37%
95% mean confidence interval for cycles value: -192.07 -80.66
95% mean confidence interval for cycles %-change: -1.07% -0.51%
Cycles are helped.
total spills in shared programs: 635 -> 817 (28.66%)
spills in affected programs: 635 -> 817 (28.66%)
helped: 0
HURT: 3
total fills in shared programs: 2065 -> 2438 (18.06%)
fills in affected programs: 2019 -> 2392 (18.47%)
helped: 0
HURT: 2
Regular shader-db results:
All Haswell+ platforms had similar results. (Tiger Lake shown)
total instructions in shared programs:
17611506 ->
17611489 (<.01%)
instructions in affected programs: 33442 -> 33425 (-0.05%)
helped: 32
HURT: 6
helped stats (abs) min: 1 max: 6 x̄: 1.69 x̃: 1
helped stats (rel) min: 0.08% max: 1.90% x̄: 0.27% x̃: 0.11%
HURT stats (abs) min: 1 max: 15 x̄: 6.17 x̃: 5
HURT stats (rel) min: 0.09% max: 1.50% x̄: 0.65% x̃: 0.55%
95% mean confidence interval for instructions value: -1.70 0.80
95% mean confidence interval for instructions %-change: -0.30% 0.05%
Inconclusive result (value mean confidence interval includes 0).
total cycles in shared programs:
338419218 ->
338418502 (<.01%)
cycles in affected programs: 385795 -> 385079 (-0.19%)
helped: 42
HURT: 3
helped stats (abs) min: 2 max: 192 x̄: 24.57 x̃: 16
helped stats (rel) min: 0.04% max: 2.09% x̄: 0.33% x̃: 0.22%
HURT stats (abs) min: 64 max: 164 x̄: 105.33 x̃: 88
HURT stats (rel) min: 0.77% max: 1.58% x̄: 1.09% x̃: 0.93%
95% mean confidence interval for cycles value: -29.76 -2.06
95% mean confidence interval for cycles %-change: -0.40% -0.07%
Cycles are helped.
Ivy Bridge and Sandy Bridge had similar results. (Ivy Bridge shown)
total instructions in shared programs:
11875620 ->
11875617 (<.01%)
instructions in affected programs: 421 -> 418 (-0.71%)
helped: 2
HURT: 0
total cycles in shared programs:
178245336 ->
178245326 (<.01%)
cycles in affected programs: 3425 -> 3415 (-0.29%)
helped: 2
HURT: 0
No changes on Gen4 or Gen5.
Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4142>
Ian Romanick [Tue, 3 Mar 2020 18:51:59 +0000 (10:51 -0800)]
nir/algebraic: Simplify logic to detect sign of an integer
This occurs in a handful of places in the soft-fp64 code, and that is
the primary reason for the change.
v2: Fix a typo in a comment. Noticed by Matt. Copy the correct fp64
shader-db results to the commit message. I realized that I used
accidentally used the results from the next commit.
Results on the 308 shaders extracted from the fp64 portion of the OpenGL
CTS:
Tiger Lake and Ice Lake had similar results. (Tiger Lake shown)
total instructions in shared programs: 906235 -> 906149 (<.01%)
instructions in affected programs: 353966 -> 353880 (-0.02%)
helped: 31
HURT: 2
helped stats (abs) min: 1 max: 8 x̄: 3.03 x̃: 3
helped stats (rel) min: 0.01% max: 1.59% x̄: 0.10% x̃: 0.04%
HURT stats (abs) min: 3 max: 5 x̄: 4.00 x̃: 4
HURT stats (rel) min: 0.02% max: 0.02% x̄: 0.02% x̃: 0.02%
95% mean confidence interval for instructions value: -3.51 -1.70
95% mean confidence interval for instructions %-change: -0.19% <.01%
Inconclusive result (%-change mean confidence interval includes 0).
total cycles in shared programs: 7076552 -> 7076173 (<.01%)
cycles in affected programs: 2878361 -> 2877982 (-0.01%)
helped: 37
HURT: 2
helped stats (abs) min: 2 max: 48 x̄: 10.81 x̃: 6
helped stats (rel) min: <.01% max: 2.17% x̄: 0.47% x̃: 0.01%
HURT stats (abs) min: 1 max: 20 x̄: 10.50 x̃: 10
HURT stats (rel) min: <.01% max: 0.01% x̄: <.01% x̃: <.01%
95% mean confidence interval for cycles value: -13.96 -5.48
95% mean confidence interval for cycles %-change: -0.72% -0.16%
Cycles are helped.
total fills in shared programs: 2064 -> 2065 (0.05%)
fills in affected programs: 45 -> 46 (2.22%)
helped: 0
HURT: 1
Regular shader-db results:
All Gen7+ platforms had similar results. (Tiger Lake shown)
total instructions in shared programs:
17611530 ->
17611506 (<.01%)
instructions in affected programs: 5934 -> 5910 (-0.40%)
helped: 10
HURT: 0
helped stats (abs) min: 1 max: 5 x̄: 2.40 x̃: 2
helped stats (rel) min: 0.14% max: 1.24% x̄: 0.47% x̃: 0.34%
95% mean confidence interval for instructions value: -3.53 -1.27
95% mean confidence interval for instructions %-change: -0.78% -0.17%
Instructions are helped.
total cycles in shared programs:
338419178 ->
338419218 (<.01%)
cycles in affected programs: 19244 -> 19284 (0.21%)
helped: 4
HURT: 2
helped stats (abs) min: 2 max: 4 x̄: 3.00 x̃: 3
helped stats (rel) min: 0.05% max: 0.11% x̄: 0.08% x̃: 0.08%
HURT stats (abs) min: 26 max: 26 x̄: 26.00 x̃: 26
HURT stats (rel) min: 1.20% max: 1.20% x̄: 1.20% x̃: 1.20%
95% mean confidence interval for cycles value: -9.08 22.41
95% mean confidence interval for cycles %-change: -0.35% 1.04%
Inconclusive result (value mean confidence interval includes 0).
No changes on any earlier Intel platform.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4142>