| Age | Commit message (Collapse) | Author |
|---|
|
<stdint.h> includes. (#150303)
https://github.com/llvm/llvm-project/issues/149993
|
|
- Merges `BinaryOpSingleOutputPerf.h` and
`SingleInputSingleOutputPerf.h` files into a unified `PerfTest.h` and
update all performance tests to use this.
- Improve the output printed to log file for tests.
- Removes unused `run_diff` method and redundant `run_perf` call in
`BINARY_INPUT_SINGLE_OUTPUT_PERF_EX` (previously
`BINARY_OP_SINGLE_OUTPUT_PERF_EX`)
- Change `BINARY_INPUT_SINGLE_OUTPUT_PERF_EX` and
`SINGLE_INPUT_SINGLE_OUTPUT_PERF` to not define `main`
|
|
Implement hypot for Float16 along with tests.
|
|
Use a combination of polynomial approximation and Newton-Raphson
iterations in 64-bit and 128-bit integers to improve the performance of
sqrtf128. The correct rounding is provided by squaring the result and
comparing it with the argument.
Performance improvement using the newly added perf test:
- My function = the improved implementation from this PR
- Other function = current implementation using
`libc/src/__support/FPUtil/generic/sqrt.h`
```
Performance tests with inputs in denormal range:
-- My function --
Total time : 1260765265 ns
Average runtime : 125.951 ns/op
Ops per second : 7939623 op/s
-- Other function --
Total time : 7160726518 ns
Average runtime : 715.357 ns/op
Ops per second : 1397902 op/s
-- Average runtime ratio --
Mine / Other's : 0.176067
Performance tests with inputs in normal range:
-- My function --
Total time : 373003808 ns
Average runtime : 37.2631 ns/op
Ops per second : 26836189 op/s
-- Other function --
Total time : 7353398916 ns
Average runtime : 734.605 ns/op
Ops per second : 1361275 op/s
-- Average runtime ratio --
Mine / Other's : 0.0507254
```
---------
Co-authored-by: Alexei Sibidanov <sibid@uvic.ca>
|
|
(#107517)
```
Performance tests with inputs in denormal range:
-- My function --
Total time : 2731072304 ns
Average runtime : 68.2767 ns/op
Ops per second : 14646276 op/s
-- Other function --
Total time : 3259744268 ns
Average runtime : 81.4935 ns/op
Ops per second : 12270913 op/s
-- Average runtime ratio --
Mine / Other's : 0.837818
Performance tests with inputs in normal range:
-- My function --
Total time : 93467258 ns
Average runtime : 2.33668 ns/op
Ops per second : 427957777 op/s
-- Other function --
Total time : 637295452 ns
Average runtime : 15.9324 ns/op
Ops per second : 62765299 op/s
-- Average runtime ratio --
Mine / Other's : 0.146662
Performance tests with inputs in normal range with exponents close to each other:
-- My function --
Total time : 95764894 ns
Average runtime : 2.39412 ns/op
Ops per second : 417690014 op/s
-- Other function --
Total time : 639866770 ns
Average runtime : 15.9967 ns/op
Ops per second : 62513075 op/s
-- Average runtime ratio --
Mine / Other's : 0.149664
```
---------
Co-authored-by: Tue Ly <lntue@google.com>
|
|
|
|
Part of #95250.
|
|
Part of #95250.
|
|
Also updates and sorts CMake target dependencies, and corrects the smoke
test that expected expf16(sNaN) to return sNaN instead of aNaN, although
the test still passed, as FPMatcher only checks whether both sides are
NaN, not whether they're the same NaN value.
|
|
Part of #95250.
|
|
available (#100002)
|
|
available (#99037)
|
|
Fixes linker errors due to hidden symbols when running CMake with
-DBUILD_SHARED_LIBS=ON.
|
|
available (#98376)
|
|
This is a part of #97655.
|
|
declaration" (#98593)
Reverts llvm/llvm-project#98075
bots are broken
|
|
We automatically inject this dependency into all object libraries that
use the libc CMake functions, but `libc_diff_test_utils` uses `add_library`
so we need to add this dependency manually.
|
|
This is a part of #97655.
|
|
|
|
Part of #93566.
|
|
- Allow `FMod` template to have different computational types and make
it work for 80-bit long double.
- Switch to use `uint64_t` as the intermediate computational types for
`float`, significantly reduce the latency of `fmodf` when the exponent
difference is large.
|
|
to performance_testing. (#84646)
Removing all the diff tests.
|
