Differences From Artifact [f9c11c2712]:
- File src/macros.h — part of check-in [26ddd2e4e4] at 2024-01-02 17:17:25 on branch trunk — Update copyright (user: js, size: 19444) [annotate] [blame] [check-ins using] [more...]
To Artifact [1147a05908]:
- File
src/macros.h
— part of check-in
[c80075f6e4]
at
2024-02-13 22:06:12
on branch trunk
— Don't use max_align_t for OF_BIGGEST_ALIGNMENT
Now that we use SSE in some inline assembly, it needs to be at least 16.
However, _Alignof(max_align_t) can be just 8. (user: js, size: 19543) [annotate] [blame] [check-ins using] [more...]
| ︙ | ︙ | |||
110 111 112 113 114 115 116 | # define OF_ALIGNOF(type) _Alignof(type) # define OF_ALIGNAS(type) _Alignas(type) #else # define OF_ALIGNOF(type) __alignof__(type) # define OF_ALIGNAS(type) OF_ALIGN(OF_ALIGNOF(type)) #endif | < < < | | | | | | > > > > > > > | 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 | # define OF_ALIGNOF(type) _Alignof(type) # define OF_ALIGNAS(type) _Alignas(type) #else # define OF_ALIGNOF(type) __alignof__(type) # define OF_ALIGNAS(type) OF_ALIGN(OF_ALIGNOF(type)) #endif #ifdef __BIGGEST_ALIGNMENT__ # define OF_BIGGEST_ALIGNMENT __BIGGEST_ALIGNMENT__ #else /* Hopefully no arch needs more than 16 byte alignment */ # define OF_BIGGEST_ALIGNMENT 16 #endif /* * We use SSE inline assembly on AMD64 and x86, so it must never be smaller * than 16. */ #if (defined(OF_AMD64) || defined(OF_X86)) && OF_BIGGEST_ALIGNMENT < 16 # undef OF_BIGGEST_ALIGNMENT # define OF_BIGGEST_ALIGNMENT 16 #endif #define OF_PREPROCESSOR_CONCAT2(a, b) a##b #define OF_PREPROCESSOR_CONCAT(a, b) OF_PREPROCESSOR_CONCAT2(a, b) #if __OBJFW_RUNTIME_ABI__ || (defined(OF_APPLE_RUNTIME) && defined(__OBJC2__)) # define OF_HAVE_NONFRAGILE_IVARS |
| ︙ | ︙ |