Differential D4731
We assume uint8_t is an alias of unsigned char. Authored by fpelliccioni on Dec 16 2019, 18:12.
Details
char, unsigned char, and std::byte (C++17) are the only byte types according to the C++ Standard. References: ninja check
Diff Detail
Event TimelineComment Actions Thanks for this! I was wondering about it. I have seen some discussions that people want compilers to not count uint8_t as an alias for char. This has the advantage of giving optimization, but, apparently it breaks ABI on existing defined architectures: https://stackoverflow.com/questions/26297571/how-to-create-an-uint8-t-array-that-does-not-undermine-strict-aliasing By the way, in which parts of code do we assume uint8_t is a byte type for alasing purposes? For many parts of the code it seems to be used because the code specifically wants an octet, which makes sense. Comment Actions (BTW another assumption I don't see documented here is that CHAR_BIT == 8. Just because sizeof(int) == 4, doesn't necessarily mean that it's 32 bits long.) This isn't just a theoretical concern or thinking about PDP with its CHAR_BIT of 9 -- I wouldn't be surprised if some more future architectures choose CHAR_BIT=16 or even 32. (https://stackoverflow.com/questions/32091992/is-char-bit-ever-8) Comment Actions @markblundeberg see https://reviews.bitcoinabc.org/D4680 for a case that uses this assumption (not entirely true since there is a memcpy). I found some other places as well by grepping for (uint8_t *) to find the suspicious casts. With your rational it makes sense to assert CHAR_BIT==8 as well, or we might want to switch to int32_t everywhere it's needed. Comment Actions Yeah I've seen a huge number of places make this assumption, which is not going to be violated any time soon (currently only some DSP chips use CHAR_BIT != 8, and posix currently guarantees CHAR_BIT == 8), but doesn't hurt to make a static assert. |