Re:An optimization guide for assembly programmers and compil...
by caf on Monday February 11, 2019 @09:51PM
(#19140588)
The XOR is still idiomatic because it is smaller than the MOV, because the immediate zero has to be encoded as the same size as the destination register.
Re:An optimization guide for assembly programmers and compil...
by kijiki on Monday February 11, 2019 @10:42PM
(#19140838)
Probably not directly faster, but "xor %rax, %rax" is still smaller (and thus faster if you're icache constrained) than "mov $0x0, %rax"
Re:An optimization guide for assembly programmers and compil...
by userbinator on Monday February 11, 2019 @11:10PM
(#19140981)
It is still faster because the register renamer handles it --- ditto for "sub reg, reg":
https://randomascii.wordpress.com/2012/12/29/the-surprising-...
Re:An optimization guide for assembly programmers and compil...
by jabl on Tuesday February 12, 2019 @02:35AM
(#19141734)
"xor %eax, %eax" is even smaller (2 bytes vs. 3) and does the same thing since operations on the 32-bit subregisters are defined to zero the upper half [1]. This is what compilers tend to do when they want to zero a register.
[1] This is not true for 8- or 16-bit subregisters, causing all kinds of fun issues (see "partial register stalls", also discussed in Agner Fog's materials).
Re:An optimization guide for assembly programmers and compil...
by bogomipz on Monday February 11, 2019 @10:58PM
(#19140924)
>"and Abrash'es book was the magnum opus of the pre-P6 era."
What book is that?
Re:An optimization guide for assembly programmers and compil...
by vnorilo on Tuesday February 12, 2019 @12:08AM
(#19141199)
Probably Zen of Code Optimization. Good times.
Re:An optimization guide for assembly programmers and compil...
by bogomipz on Tuesday February 12, 2019 @01:02AM
(#19141416)
The Zen of Assembler? Does that sound correct? That's the closest I could find.
Re:An optimization guide for assembly programmers and compil...
by markedathome on Tuesday February 12, 2019 @04:10AM
(#19142108)
Zen of Code Optimization 1994 (
https://www.amazon.com/Zen-Code-Optimization-Ultimate-Softwa...) is the follow up to Zen of Assembly Language 1990 (https://www.amazon.com/Zen-Assembly-Language-Knowledge-Progr...)
Re:An optimization guide for assembly programmers and compil...
by xsgsdfgsdg on Tuesday February 12, 2019 @06:04AM
(#19142553)
XOR was never faster, it is 2 bytes-vs 3 bytes instruction size.
I guess I better throw most of those assumptions out for modern microarchitectures.