@@ -4,14 +4,15 @@ Optional macros to predefine to tune performance
4
4
There are a number of macros you can optionally predefine to tune the
5
5
performance on your system for the modular arithmetic functions. It is
6
6
generally recommended not to do so, but in some cases you may find it useful.
7
- You would predefine one or more of these macros when compiling the sources. For
8
- example, with CMake you would add the command "target_compile_definitions" to
9
- the CMakeLists.txt, like this:
10
- target_compile_features(hurchalla_modular_arithmetic INTERFACE HURCHALLA_ALLOW_INLINE_ASM_ALL)
7
+ You would predefine one or more of these macros when compiling * your* sources,
8
+ given that Clockwork is a header-only library.
11
9
12
- If you are not using CMake, then with clang or gcc you would compile with the -D
13
- compilation flag like this:
14
- clang++ -DHURCHALLA_ALLOW_INLINE_ASM_ALL
10
+ For example, if you are compiling using clang or gcc from the command line, you would
11
+ specify the -D compilation flag, similarly to the following:
12
+ clang++ -DHURCHALLA_ALLOW_INLINE_ASM_ALL ...more arguments...
13
+ As another example, if you are using CMake you would add the command "target_compile_definitions"
14
+ to your CMakeLists.txt, similarly to the following:
15
+ target_compile_definitions(< ; your_target_name> ; PRIVATE HURCHALLA_ALLOW_INLINE_ASM_ALL)
15
16
\
16
17
\
17
18
HURCHALLA_TARGET_ISA_HAS_NO_DIVIDE - predefine this macro if your target system's
0 commit comments