This patch adds the recently-added tweak to split some SVE VL-based scalar
operations [1] to the generic tuning used for SVE, as enabled by adding +sve
to the -march flag, for example -march=armv8.2-a+sve.
The recommendation for best performance on a particular CPU remains unchanged:
use the -mcpu option for that CPU, where possible. -mcpu=native makes this
straightforward for native compilation.
The tweak to split out SVE VL-based scalar operations is a consistent win for
the Neoverse V1 CPU and should be neutral for the Fujitsu A64FX. A run of
SPEC2017 on A64FX with this tweak on didn't show any non-noise differences.
It is also expected to be neutral on SVE2 implementations.
Therefore, the patch enables the tweak for generic +sve tuning e.g.
-march=armv8.2-a+sve. No SVE2 CPUs are expected to benefit from it,
therefore the tweak is disabled for generic tuning when +sve2 is in
-march e.g. -march=armv8.2-a+sve2.
The implementation of this approach requires a bit of custom logic in
aarch64_override_options_internal to handle these kinds of
architecture-dependent decisions, but we do believe the user-facing principle
here is important to implement.
In general, for the generic target we're using a decision framework that looks
like:
* If all cores that are known to benefit from an optimization
are of architecture X, and all other cores that implement X or above
are not impacted, or have a very slight impact, we will consider it for
generic tuning for architecture X.
* We will not enable that optimisation for generic tuning for architecture X+1
if no known cores of architecture X+1 or above will benefit.
This framework allows us to improve generic tuning for CPUs of generation X
while avoiding accumulating tweaks for future CPUs of generation X+1, X+2...
that do not need them, and thus avoid even the slight negative effects of
these optimisations if the user is willing to tell us the desired architecture
accurately.
X above can mean either annual architecture updates (Armv8.2-a, Armv8.3-a etc)
or optional architecture extensions (like SVE, SVE2).
[1] http://gcc.gnu.org/g:
a65b9ad863c5fc0aea12db58557f4d286a1974d7
gcc/ChangeLog:
* config/aarch64/aarch64.c (aarch64_adjust_generic_arch_tuning): Define.
(aarch64_override_options_internal): Use it.
(generic_tunings): Add AARCH64_EXTRA_TUNE_CSE_SVE_VL_CONSTANTS to
tune_flags.
gcc/testsuite/ChangeLog:
* g++.target/aarch64/sve/aarch64-sve.exp: Add -moverride=tune=none to
sve_flags.
* g++.target/aarch64/sve/acle/aarch64-sve-acle-asm.exp: Likewise.
* g++.target/aarch64/sve/acle/aarch64-sve-acle.exp: Likewise.
* gcc.target/aarch64/sve/aarch64-sve.exp: Likewise.
* gcc.target/aarch64/sve/acle/aarch64-sve-acle-asm.exp: Likewise.
* gcc.target/aarch64/sve/acle/aarch64-sve-acle.exp: Likewise.
2, /* min_div_recip_mul_df. */
0, /* max_case_values. */
tune_params::AUTOPREFETCHER_WEAK, /* autoprefetcher_model. */
- (AARCH64_EXTRA_TUNE_NONE), /* tune_flags. */
+ /* Enabling AARCH64_EXTRA_TUNE_CSE_SVE_VL_CONSTANTS significantly benefits
+ Neoverse V1. It does not have a noticeable effect on A64FX and should
+ have at most a very minor effect on SVE2 cores. */
+ (AARCH64_EXTRA_TUNE_CSE_SVE_VL_CONSTANTS), /* tune_flags. */
&generic_prefetch_tune
};
free (string_root);
}
+/* Adjust CURRENT_TUNE (a generic tuning struct) with settings that
+ are best for a generic target with the currently-enabled architecture
+ extensions. */
+static void
+aarch64_adjust_generic_arch_tuning (struct tune_params ¤t_tune)
+{
+ /* Neoverse V1 is the only core that is known to benefit from
+ AARCH64_EXTRA_TUNE_CSE_SVE_VL_CONSTANTS. There is therefore no
+ point enabling it for SVE2 and above. */
+ if (TARGET_SVE2)
+ current_tune.extra_tuning_flags
+ &= ~AARCH64_EXTRA_TUNE_CSE_SVE_VL_CONSTANTS;
+}
static void
aarch64_override_options_after_change_1 (struct gcc_options *opts)
we may later overwrite. */
aarch64_tune_params = *(selected_tune->tune);
aarch64_architecture_version = selected_arch->architecture_version;
+ if (selected_tune->tune == &generic_tunings)
+ aarch64_adjust_generic_arch_tuning (aarch64_tune_params);
if (opts->x_aarch64_override_tune_string)
aarch64_parse_override_string (opts->x_aarch64_override_tune_string,
set sve_flags "-march=armv8.2-a+sve"
}
+# Turn off any codegen tweaks by default that may affect expected assembly.
+# Tests relying on those should turn them on explicitly.
+set sve_flags "$sve_flags -moverride=tune=none"
+
# Main loop.
dg-runtest [lsort [glob -nocomplain $srcdir/$subdir/*.C]] $sve_flags ""
set sve_flags "-march=armv8.2-a+sve"
}
+# Turn off any codegen tweaks by default that may affect expected assembly.
+# Tests relying on those should turn them on explicitly.
+set sve_flags "$sve_flags -moverride=tune=none"
+
global gcc_runtest_parallelize_limit_minor
if { [info exists gcc_runtest_parallelize_limit_minor] } {
set old_limit_minor $gcc_runtest_parallelize_limit_minor
set sve_flags "-march=armv8.2-a+sve"
}
+# Turn off any codegen tweaks by default that may affect expected assembly.
+# Tests relying on those should turn them on explicitly.
+set sve_flags "$sve_flags -moverride=tune=none"
+
# Main loop.
set gcc_subdir [string replace $subdir 0 2 gcc]
set files [glob -nocomplain \
set sve_flags "-march=armv8.2-a+sve"
}
+# Turn off any codegen tweaks by default that may affect expected assembly.
+# Tests relying on those should turn them on explicitly.
+set sve_flags "$sve_flags -moverride=tune=none"
+
# Most of the code-quality tests are written for LP64. Just do the
# correctness tests for ILP32.
if { [check_effective_target_ilp32] } {
set sve_flags "-march=armv8.2-a+sve"
}
+# Turn off any codegen tweaks by default that may affect expected assembly.
+# Tests relying on those should turn them on explicitly.
+set sve_flags "$sve_flags -moverride=tune=none"
+
global gcc_runtest_parallelize_limit_minor
if { [info exists gcc_runtest_parallelize_limit_minor] } {
set old_limit_minor $gcc_runtest_parallelize_limit_minor
set sve_flags "-march=armv8.2-a+sve"
}
+# Turn off any codegen tweaks by default that may affect expected assembly.
+# Tests relying on those should turn them on explicitly.
+set sve_flags "$sve_flags -moverride=tune=none"
+
# Main loop.
set files [glob -nocomplain \
"$srcdir/$subdir/general/*.c" \
projects / platform / upstream / gcc.git / commitdiff