cxykevin/jslinux
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

document CPUID, with one puzzling exception

Browse Source
This commit is contained in:
Dan Luu 2013-03-19 15:09:51 -04:00
parent b52e4ead18
commit 7df7cedbf1
2 changed files with 10 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

16
cpux86-ta.js
View File

@ -4862,21 +4862,21 @@ CPU_X86.prototype.exec_internal = function(N_cycles, interrupt) {
_op = 24;
}
function op_CPUID() {
var Rb;
Rb = regs[0];
switch (Rb) {
case 0:
var eax;
eax = regs[0];
switch (eax) {
case 0: // eax == 0: vendor ID
regs[0] = 1;
regs[3] = 0x756e6547 & -1;
regs[2] = 0x49656e69 & -1;
regs[1] = 0x6c65746e & -1;
break;
case 1:
case 1: // eax == 1: processor info and feature flags
default:
regs[0] = (5 << 8) | (4 << 4) | 3;
regs[3] = 8 << 8;
regs[0] = (5 << 8) | (4 << 4) | 3; // family | model | stepping
regs[3] = 8 << 8; // danluu: This is a mystery to me. This bit now indicates clflush line size, but must have meant something else in the past.
regs[1] = 0;
regs[2] = (1 << 4);
regs[2] = (1 << 4); // rdtsc support
break;
}
}

2
readme.md
View File

@ -25,6 +25,8 @@ have been added.
The core opcode execution loop has been autocommented to indicate what
instruction operation the opcode refers to.
One mystery is, why does CPUID(1) return 8 << 8 in EBX? EBX[15:8] is now used to indicate CLFLUSH line size, but that field must have been used for something else in the past.
### ETC
I highly recommend, by the way, the excellent [JSShaper][2] library for transforming large javascript code bases. The hacks I made from it are in this repo: a little symbol-name-transformer node.js script and an emacs function for doing this in live buffers.