4 ms·

C is portable assembly until it isn't: - no access to carry/overflow flag in registers making it a chore to write bigint libraries - no way to guarantee emitt

C is portable assembly until it isn't: - no access to carry/overflow flag in registers making it a chore to write bigint libraries - no way to guarantee emitting add with carry or sub with borrow even though those are simple instructions and some architecture (6502) don't even provide a normal add/sub - need to drop down to assembly to implement resumable function/fibers to avoid the syscall costs of ucontext/setjmp/longjmp - no access to hardware counters like RDTSC - no way to get the CPU frequency in a portable and reliable way - no portable threadID function and then, those that are available (pthread_self, and Windows') rely on expensive syscalls. - no way to do CPU feature detection (SSE, AVX, ARM Neon, ...) - no way to control emission of common intrinsics like popcount, bit-scan-reverse, lowest bit isolation in a portable way. The portable assembly narrative breaks down rapidly when you actually need specific code to be emitted.

baq 7y ago
Precisely my point - learning both together as they compliment each other instead of C abstracting the underlying computer away.
sigjuice 7y ago
need to drop down to assembly to implement resumable function/fibers to avoid the syscall costs of ucontext/setjmp/longjmp Which syscalls would be involved in setjmp/longjmp?
- mratsim 7y ago
  See http://www.1024cores.net/home/lock-free-algorithms/tricks/fibers http://www.1024cores.net/home/lock-free-algorithms/tricks/fi... on ucontext And for longjmp, Apple source for example https://opensource.apple.com/source/Libc/Libc-262/i386/sys/setjmp.s.auto.html https://opensource.apple.com/source/Libc/Libc-262/i386/sys/s.... It does seem like glibc impl doesn't involve syscalls: https://groups.google.com/forum/#!topic/comp.unix.programmer/LvUP9H-7Aes https://groups.google.com/forum/#!topic/comp.unix.programmer...
- loeg 7y ago
  For inexplicable reasons, ucontext/setjmp/longjmp include signal masks, which, for lack of VDSO implementation, must be obtained by invoking syscalls. (FreeBSD provides the non-portable _setjmp/_longjmp, which do not preserve signal mask state and avoid the syscall. There are also the POSIX sigsetjmp/siglongjmp, which with savesigs=0 may bypass the syscalls as well. Both FreeBSD and Linux provide the POSIX routines.)
throwaway17_17 7y ago
The following is a genuine question (not a flippant remark): Is there any way to have those concepts be portable across ISAs/architectures. I don’t think those types of ops are exposed in LLVM-IR or GCC’s various intermediate levels (but could be wrong). I’d love to be able to investigate a language that enables a level of hardware access possible in a portable manner.
renox 7y ago
> C is portable assembly until it isn't: > - no access to carry/overflow flag in registers making it a chore to write bigint libraries C is portable assembly: so it provides access to a minimum common set of features which allow the code to be portable: good luck trying to use the carry flag on the RISC-V..

baq 7y ago

Precisely my point - learning both together as they compliment each other instead of C abstracting the underlying computer away.

sigjuice 7y ago

need to drop down to assembly to implement resumable function/fibers to avoid the syscall costs of ucontext/setjmp/longjmp Which syscalls would be involved in setjmp/longjmp?

mratsim 7y ago
See http://www.1024cores.net/home/lock-free-algorithms/tricks/fibers http://www.1024cores.net/home/lock-free-algorithms/tricks/fi... on ucontext And for longjmp, Apple source for example https://opensource.apple.com/source/Libc/Libc-262/i386/sys/setjmp.s.auto.html https://opensource.apple.com/source/Libc/Libc-262/i386/sys/s.... It does seem like glibc impl doesn't involve syscalls: https://groups.google.com/forum/#!topic/comp.unix.programmer/LvUP9H-7Aes https://groups.google.com/forum/#!topic/comp.unix.programmer...
loeg 7y ago
For inexplicable reasons, ucontext/setjmp/longjmp include signal masks, which, for lack of VDSO implementation, must be obtained by invoking syscalls. (FreeBSD provides the non-portable _setjmp/_longjmp, which do not preserve signal mask state and avoid the syscall. There are also the POSIX sigsetjmp/siglongjmp, which with savesigs=0 may bypass the syscalls as well. Both FreeBSD and Linux provide the POSIX routines.)

throwaway17_17 7y ago

The following is a genuine question (not a flippant remark): Is there any way to have those concepts be portable across ISAs/architectures. I don’t think those types of ops are exposed in LLVM-IR or GCC’s various intermediate levels (but could be wrong). I’d love to be able to investigate a language that enables a level of hardware access possible in a portable manner.

renox 7y ago

> C is portable assembly until it isn't: > - no access to carry/overflow flag in registers making it a chore to write bigint libraries C is portable assembly: so it provides access to a minimum common set of features which allow the code to be portable: good luck trying to use the carry flag on the RISC-V..