| Age | Commit message (Collapse) | Author |
|---|
|
src/__support/math folder. (#161297)
Part of #147386
in preparation for: https://discourse.llvm.org/t/rfc-make-clang-builtin-math-functions-constexpr-with-llvm-libc-to-support-c-23-constexpr-math-functions/86450
|
|
LIBC_MATH_SKIP_ACCURATE_PASS is set. (#130968)
|
|
(#130700)
|
|
|
|
This is a part of #97655.
|
|
declaration" (#98593)
Reverts llvm/llvm-project#98075
bots are broken
|
|
This is a part of #97655.
|
|
Part of #93566.
|
|
- reland #79113
- Fix aarch64 RISC-V build
|
|
Reverts llvm/llvm-project#79113
It broke aarch64 build bot machines.
|
|
This patch reduces the surface of `FPBits`.
|
|
Another patch is needed to cover `DyadicFloat` and `NormalFloat`
constructors.
|
|
|
|
all rounding modes. (#67048)
Implementing expm1 function for double precision based on exp function
algorithm:
- Reduced x = log2(e) * (hi + mid1 + mid2) + lo, where:
* hi is an integer
* mid1 * 2^-6 is an integer
* mid2 * 2^-12 is an integer
* |lo| < 2^-13 + 2^-30
- Then exp(x) - 1 = 2^hi * 2^mid1 * 2^mid2 * exp(lo) - 1 ~ 2^hi *
(2^mid1 * 2^mid2 * (1 + lo * P(lo)) - 2^(-hi) )
- We evaluate fast pass with P(lo) is a degree-3 Taylor polynomial of
(e^lo - 1) / lo in double precision
- If the Ziv accuracy test fails, we use degree-6 Taylor polynomial of
(e^lo - 1) / lo in double double precision
- If the Ziv accuracy test still fails, we re-evaluate everything in
128-bit precision.
|
|
This is step 4 of
https://discourse.llvm.org/t/rfc-customizable-namespace-to-allow-testing-the-libc-when-the-system-libc-is-also-llvms-libc/73079
|
|
Many math functions need to check for floating point rounding modes to
return correct values. Currently most of them use the internal implementation
of `fegetround`, which is platform-dependent and blocking math functions to be
enabled on platforms with unimplemented `fegetround`. In this change, we add
platform independent rounding mode checks and switching math functions to use
them instead. https://github.com/llvm/llvm-project/issues/63016
Reviewed By: sivachandra
Differential Revision: https://reviews.llvm.org/D152280
|
|
Set FE_OVERFLOW and FE_UNDERFLOW for expf, exp2f, exp10f, expm1f, sinhf
and coshf.
Reviewed By: sivachandra, renyichen
Differential Revision: https://reviews.llvm.org/D144340
|
|
|
|
|
|
|
|
|
|
|
|
This implements the same behavior as D141997 but makes sure that the same detection mechanism is used between CMake and source code.
Reviewed By: sivachandra, lntue
Differential Revision: https://reviews.llvm.org/D142108
|
|
try_compile instead of try_run."
The build bots are failing.
This reverts commit c84d74f5bfe810744de1268eb0516a6622e4aa73.
|
|
try_run.
This implements the same behavior as D141997 but makes sure that the same detection mechanism is used between CMake and source code.
Differential Revision: https://reviews.llvm.org/D142108
|
|
instead of try_run."
Build bots are failing.
https://lab.llvm.org/buildbot/#/builders/90/builds/44634
This reverts commit 9acc2f37bdfce08ca0c2faec03392db10d1bb7a9.
|
|
This implements the same behavior as D141997 but makes sure that the same detection mechanism is used between CMake and source code.
Differential Revision: https://reviews.llvm.org/D142108
|
|
Use nearest_integer instructions to improve expf performance.
Performance tests with CORE-MATH's perf tool:
Before the patch:
```
$ ./perf.sh expm1f
LIBC-location: /home/lnt/experiment/llvm/llvm-project/build/projects/libc/lib/libllvmlibc.a
GNU libc version: 2.31
GNU libc release: stable
CORE-MATH reciprocal throughput : 10.096
System LIBC reciprocal throughput : 44.036
LIBC reciprocal throughput : 11.575
$ ./perf.sh expm1f --latency
LIBC-location: /home/lnt/experiment/llvm/llvm-project/build/projects/libc/lib/libllvmlibc.a
GNU libc version: 2.31
GNU libc release: stable
CORE-MATH latency : 42.239
System LIBC latency : 122.815
LIBC latency : 50.122
```
After the patch:
```
$ ./perf.sh expm1f
LIBC-location: /home/lnt/experiment/llvm/llvm-project/build/projects/libc/lib/libllvmlibc.a
GNU libc version: 2.31
GNU libc release: stable
CORE-MATH reciprocal throughput : 10.046
System LIBC reciprocal throughput : 43.899
LIBC reciprocal throughput : 9.179
$ ./perf.sh expm1f --latency
LIBC-location: /home/lnt/experiment/llvm/llvm-project/build/projects/libc/lib/libllvmlibc.a
GNU libc version: 2.31
GNU libc release: stable
CORE-MATH latency : 42.078
System LIBC latency : 120.488
LIBC latency : 41.528
```
Reviewed By: zimmermann6
Differential Revision: https://reviews.llvm.org/D130502
|
|
Add another exceptional value and fix the case when |x| is small.
Performance tests with CORE-MATH project scripts:
With FMA instructions on Ryzen 1700:
```
$ ./perf.sh expm1f
LIBC-location: /home/lnt/experiment/llvm/llvm-project/build/projects/libc/lib/libllvmlibc.a
CORE-MATH reciprocal throughput : 15.362
System LIBC reciprocal throughput : 53.194
LIBC reciprocal throughput : 14.595
$ ./perf.sh expm1f --latency
LIBC-location: /home/lnt/experiment/llvm/llvm-project/build/projects/libc/lib/libllvmlibc.a
CORE-MATH latency : 57.755
System LIBC latency : 147.020
LIBC latency : 60.269
```
Without FMA instructions:
```
$ ./perf.sh expm1f
LIBC-location: /home/lnt/experiment/llvm/llvm-project/build/projects/libc/lib/libllvmlibc.a
CORE-MATH reciprocal throughput : 15.362
System LIBC reciprocal throughput : 53.300
LIBC reciprocal throughput : 18.020
$ ./perf.sh expm1f --latency
LIBC-location: /home/lnt/experiment/llvm/llvm-project/build/projects/libc/lib/libllvmlibc.a
CORE-MATH latency : 57.758
System LIBC latency : 147.025
LIBC latency : 70.304
```
Reviewed By: michaelrj
Differential Revision: https://reviews.llvm.org/D123440
|
|
Detect if the architecture supports FMA instructions and if
the targets depend on fma.
Reviewed By: gchatelet
Differential Revision: https://reviews.llvm.org/D123615
|
|
Make FMA flag checks more accurate for x86-64 targets, and refactor
polyeval to use multiply and add instead when FMA instructions are not
available.
Reviewed By: michaelrj, sivachandra
Differential Revision: https://reviews.llvm.org/D123335
|
|
Improve the performance of expm1f:
- Rearrange the selection logic for different cases to improve the overall
throughput.
- Use the same degree-4 polynomial for large inputs as `expf`
(https://reviews.llvm.org/D122418), reduced from a degree-7 polynomial.
Performance benchmark using perf tool from CORE-MATH project
(https://gitlab.inria.fr/core-math/core-math/-/tree/master):
Before this patch:
```
$ ./perf.sh expm1f
CORE-MATH reciprocal throughput : 15.362
System LIBC reciprocal throughput : 53.288
LIBC reciprocal throughput : 54.572
$ ./perf.sh expm1f --latency
CORE-MATH latency : 57.759
System LIBC latency : 147.146
LIBC latency : 118.057
```
After this patch:
```
$ ./perf.sh expm1f
CORE-MATH reciprocal throughput : 15.359
System LIBC reciprocal throughput : 53.188
LIBC reciprocal throughput : 14.600
$ ./perf.sh expm1f --latency
CORE-MATH latency : 57.774
System LIBC latency : 147.119
LIBC latency : 60.280
```
Reviewed By: michaelrj, santoshn, zimmermann6
Differential Revision: https://reviews.llvm.org/D122538
|
|
modes.
Implement expm1f function that is correctly rounded for all rounding modes. This is based on expf implementation.
From exhaustive testings, using expf implementation, and subtract 1.0 before rounding the final result to single precision
gives correctly rounded results for all |x| > 2^-4 with 1 exception. When |x| < 2^-25, we use x + x^2 (implemented with a
single fma). And for 2^-25 <= |x| <= 2^-4, we use a single degree-8 minimax polynomial generated by Sollya.
Reviewed By: sivachandra, zimmermann6
Differential Revision: https://reviews.llvm.org/D121574
|
|
Add ADD_FMA_FLAG macro to add -mfma flag to functions that requires it.
Reviewed By: sivachandra
Differential Revision: https://reviews.llvm.org/D115572
|
|
This patch applies the lint rules described in the previous patch. There
was also a significant amount of effort put into manually fixing things,
since all of the templated functions, or structs defined in /spec, were
not updated and had to be handled manually.
Reviewed By: sivachandra, lntue
Differential Revision: https://reviews.llvm.org/D114302
|
|
Some ctype functions are called from other libc functions (e.g. isspace
is used in atoi). By moving ctype_utils.h to __support it becomes easier
to include just the implementations of these functions. For these
reasons the implementation for isspace was moved into
ctype_utils as well.
FPUtils was moved to simplify the build order, and to clarify which
files are a part of the actual libc.
Many files were modified to accomodate these changes, mostly changing
the #include paths.
Reviewed By: sivachandra
Differential Revision: https://reviews.llvm.org/D107600
|
|
Use expm1f(x) = exp(x) - 1 for |x| > ln(2).
For |x| <= ln(2), divide it into 3 subintervals: [-ln2, -1/8], [-1/8, 1/8], [1/8, ln2]
and use a degree-6 polynomial approximation generated by Sollya's fpminmax for each interval.
Errors < 1.5 ULPs when we use fma to evaluate the polynomials.
Differential Revision: https://reviews.llvm.org/D101134
|
