Human Folly about “amd64”

index

01 XOR eax eax (xania.org):
#Godbolt about why #amd64 #asm uses so many xor instructions
on 02026-04-22
(devblogs.microsoft.com):
no real reason people on #amd64 use xor instead of sub to clear a register in #asm; both of them require zero micro-ops, merely renaming a register. #performance
on 02026-04-22
TZCNT (felixcloutier.com):
#amd64 #asm instruction TZCNT or BSF counts the number of trailing zeroes. #bitmanip
on 02025-09-09
Linux 64 ABI calling convention (stackoverflow.com):
Linux’s system-call #ABI differs from the standard #amd64 function-call ABI. “This means system-call wrapper functions can just set eax and run syscall instead of needing to do something like...” #asm
on 02024-09-27
What registers are preserved through a Linux x86 64 function call (stackoverflow.com):
outline of #amd64 #ABI for #asm on Linux
on 02024-08-01
Acceptability of regular usage of r10 and r11 (stackoverflow.com):
some notes on the #amd64 SysV ABI register use
on 02024-06-30
Accessing the low 32 bits of r8 through r15 (stackoverflow.com):
In #amd64 #asm you call the low 32 bits of a numbered register rNd, e.g., r8d, r11d, r15d. Writing to it clears the upper 32 bits, same as ecx vs. rcx. The low byte is rNb, I forget if there’s a name for the low 16 bits.
on 02024-06-30
MOVZX (felixcloutier.com):
#x86 and #amd64 MOVZX instruction doesn't take an immediate argument. #asm
on 02023-07-03
What is the fPIE option for position independent executables in GCC and LD (stackoverflow.com):
notes on #GDB and #position-independent executables and PC-relative lea on #amd64
on 02023-07-01
Why we Cant move a 64 bit immediate value to memory (stackoverflow.com):
the movabsq instruction in #amd64 is needed to load a 64-bit immediate
on 02022-09-16
X86 calling conventions (en.wikipedia.org):
#amd64 and #i386 calling conventions
on 02022-09-16
Pmaddwd (felixcloutier.com):
an #SSE widening dot-product operation in #X86/#amd64. With 128-bit operands, it does 8 16-bit multiplies and adds pairs of the 8 32-bit products to get 4 32-bit sums, which fit into the destination register.
on 02016-08-08
Debugging on amd64 part one (blogs.oracle.com):
DTrace hacker unhappy that the #amd64 #ABI doesn’t have frame pointers, resulting in inefficiency in unwinding the stack.
on 02016-07-24
Deep wizardry stack unwinding (blog.reverberate.org):
How stack unwinding works on the #amd64 #ABI, with .eh_frame and .debug_frame and whatnot.
on 02016-07-24
Bare metal Rust custom target kernel space (randomhacks.net):
128 bytes below %rsp on #amd64 is a “red zone” of scratch space, making the standard amd64 #abi unusable for code that has to handle interrupts.
on 02015-11-11