| commit | e7af9751d9f9fd3b47911f51a5cfd08af256a9ab | [log] [tgz] |
|---|---|---|
| author | Olaf Hering <olaf@aepfle.de> | Fri Jul 21 07:16:11 2023 +0000 |
| committer | Olaf Hering <olaf@aepfle.de> | Fri Jul 21 15:36:33 2023 +0200 |
| tree | 9e166ce5ee90e0cb45f975a89c877bc4114de601 | |
| parent | 40619cbb3bf32872df8c53cc457039229428a263 [diff] |
Fix -Wreturn-type errors in fNNRandom In case global CFLAGS contains -Werror=return-type, compilation fails as shown below. For some reason gcc is unable to recognize the possible range of values in the switch statements, which can only ever be between zero and seven. Assist the compiler by adding a default statement. This fixes compilation in openSUSE Tumbleweed, which adds -Werror=return-type to global CFLAGS. source/genCases_f64.c: In function 'f64Random': source/genCases_f64.c:559:1: error: control reaches end of non-void function [-Werror=return-type] source/genCases_f32.c: In function 'f32Random': source/genCases_f32.c:417:1: error: control reaches end of non-void function [-Werror=return-type] source/genCases_f16.c: In function 'f16Random': source/genCases_f16.c:336:1: error: control reaches end of non-void function [-Werror=return-type] Signed-off-by: Olaf Hering <olaf@aepfle.de>
John R. Hauser
2018 January 20
Berkeley TestFloat is a small collection of programs for testing that an implementation of binary floating-point conforms to the IEEE Standard for Floating-Point Arithmetic. TestFloat is distributed in the form of C source code.
The TestFloat package is documented in the following files in the doc subdirectory:
testfloat_gen program.testfloat_ver program.testfloat program.testsoftfloat program.timesoftfloat program.Other files in the package comprise the source code for TestFloat.