Update to v101r18 release.

byuu says:

Changelog:

  - added 30 new PAL games to icarus (courtesy of Mikerochip)
  - new version of libco no longer requires mprotect nor W|X permissions
  - nall: default C compiler to -std=c11 instead of -std=c99
  - nall: use `-fno-strict-aliasing` during compilation
  - updated nall/certificates (hopefully for the last time)
  - updated nall/http to newer coding conventions
  - nall: improve handling of range() function

I didn't really work on higan at all, this is mostly just a release
because lots of other things have changed.

The most interesting is `-fno-strict-aliasing` ... basically, it joins
`-fwrapv` as being "stop the GCC developers from doing *really* evil
shit that could lead to security vulnerabilities or instabilities."

For the most part, it's a ~2% speed penalty for higan. Except for the
Sega Genesis, where it's a ~10% speedup. I have no idea how that's
possible, but clearly something's going very wrong with strict aliasing
on the Genesis core.

So ... it is what it is. If you need the performance for the non-Genesis
cores, you can turn it off in your builds. But I'm getting quite sick of
C++'s "surprises" and clever compiler developers, so I'm keeping it on
in all of my software going forward.
This commit is contained in:
Tim Allen
2016-09-14 21:55:53 +10:00
parent d6e9d94ec3
commit c6fc15f8d2
26 changed files with 980 additions and 514 deletions

View File

@@ -1,15 +1,12 @@
/*
libco.amd64 (2015-06-19)
libco.amd64 (2016-09-14)
author: byuu
license: public domain
*/
#define LIBCO_C
#include "libco.h"
//Win64 only: provides a substantial speed-up, but will thrash XMM regs
//do not use this unless you are certain your application won't use SSE
//#define LIBCO_AMD64_NO_SSE
#include "settings.h"
#include <assert.h>
#include <stdlib.h>
@@ -22,9 +19,14 @@ static thread_local long long co_active_buffer[64];
static thread_local cothread_t co_active_handle = 0;
static void (*co_swap)(cothread_t, cothread_t) = 0;
#ifdef LIBCO_MPROTECT
alignas(4096)
#else
section(text)
#endif
#ifdef _WIN32
/* ABI: Win64 */
static unsigned char co_swap_function[] = {
static const unsigned char co_swap_function[4096] = {
0x48, 0x89, 0x22, /* mov [rdx],rsp */
0x48, 0x8b, 0x21, /* mov rsp,[rcx] */
0x58, /* pop rax */
@@ -36,7 +38,7 @@ static void (*co_swap)(cothread_t, cothread_t) = 0;
0x4c, 0x89, 0x6a, 0x30, /* mov [rdx+48],r13 */
0x4c, 0x89, 0x72, 0x38, /* mov [rdx+56],r14 */
0x4c, 0x89, 0x7a, 0x40, /* mov [rdx+64],r15 */
#if !defined(LIBCO_AMD64_NO_SSE)
#if !defined(LIBCO_NO_SSE)
0x0f, 0x29, 0x72, 0x50, /* movaps [rdx+ 80],xmm6 */
0x0f, 0x29, 0x7a, 0x60, /* movaps [rdx+ 96],xmm7 */
0x44, 0x0f, 0x29, 0x42, 0x70, /* movaps [rdx+112],xmm8 */
@@ -57,7 +59,7 @@ static void (*co_swap)(cothread_t, cothread_t) = 0;
0x4c, 0x8b, 0x69, 0x30, /* mov r13,[rcx+48] */
0x4c, 0x8b, 0x71, 0x38, /* mov r14,[rcx+56] */
0x4c, 0x8b, 0x79, 0x40, /* mov r15,[rcx+64] */
#if !defined(LIBCO_AMD64_NO_SSE)
#if !defined(LIBCO_NO_SSE)
0x0f, 0x28, 0x71, 0x50, /* movaps xmm6, [rcx+ 80] */
0x0f, 0x28, 0x79, 0x60, /* movaps xmm7, [rcx+ 96] */
0x44, 0x0f, 0x28, 0x41, 0x70, /* movaps xmm8, [rcx+112] */
@@ -75,13 +77,15 @@ static void (*co_swap)(cothread_t, cothread_t) = 0;
#include <windows.h>
void co_init() {
static void co_init() {
#ifdef LIBCO_MPROTECT
DWORD old_privileges;
VirtualProtect(co_swap_function, sizeof co_swap_function, PAGE_EXECUTE_READWRITE, &old_privileges);
VirtualProtect((void*)co_swap_function, sizeof co_swap_function, PAGE_EXECUTE_READ, &old_privileges);
#endif
}
#else
/* ABI: SystemV */
static unsigned char co_swap_function[] = {
static const unsigned char co_swap_function[4096] = {
0x48, 0x89, 0x26, /* mov [rsi],rsp */
0x48, 0x8b, 0x27, /* mov rsp,[rdi] */
0x58, /* pop rax */
@@ -103,11 +107,13 @@ static void (*co_swap)(cothread_t, cothread_t) = 0;
#include <unistd.h>
#include <sys/mman.h>
void co_init() {
static void co_init() {
#ifdef LIBCO_MPROTECT
unsigned long long addr = (unsigned long long)co_swap_function;
unsigned long long base = addr - (addr % sysconf(_SC_PAGESIZE));
unsigned long long size = (addr - base) + sizeof co_swap_function;
mprotect((void*)base, size, PROT_READ | PROT_WRITE | PROT_EXEC);
mprotect((void*)base, size, PROT_READ | PROT_EXEC);
#endif
}
#endif