-
The test in `gambit.h`
```
#if __has_builtin(__builtin_ia32_rdtscp)
```
tests only whether the compiler knows about `rdtscp`, not that the architecture actually has the `rdtscp` instruction.
I…
-
https://windowbr.top/2024/09/17/fnos-mihomo/
近日飞牛OS（fnOS）开启了公测，我也第一时间开始了尝鲜，但很快便遇到一个问题：系统的网络设置中并未提供代理设置的选项。 没有代理设置怎么行，这可是直接与世界脱离连接了啊，于是我便开始努力寻找解决方案，最后得到了一个并不完美但可行的方案：使用虚空终端 mihomo(clash meta) 及对应的 W…
-
What's the difference between,
mpv-x86_64-20241112-git-43d8966.7z
mpv-x86_64-gcc-20241112-git-43d8966.7z
mpv-x86_64-v3-20241112-git-43d8966.7z
Thanks
-
Bun does not support x86_64 CPUs without SSE4.2 (older than Intel Nehalem or AMD Bulldozer) at this time, even in our x64-baseline builds. On unsupported CPUs, currently Bun crashes with "illegal inst…
-
In package [`golang.org/x/sys/cpu`](https://pkg.go.dev/golang.org/x/sys/cpu) add an `int` symbol [0-4] (0 if not AMD64) that exposes the AMD64 microarchitecture level of the machine running the progra…
-
The last line of fig. 13-2 is as follows. (In both of English and Japanese version.)
```
b.ne loop // Reload LBSY to x2 if LBSY = BST
```
The branch condition on the comment should be *LBSY…
-
See also:
https://www.phoronix.com/scan.php?page=news_item&px=GCC-11-x86-64-Feature-Levels
Basically, if you provide: --march=x86_64-v2 or --march=x86_64-v3 or --march=x86_64-v4 to recent versions…
-
Just like an earlier issue discusses address range CMOs vs per-cache-line CMOs... but this time for operations that are typically used for things like "flush the entire I$ or D$".
Such "cache micro…
-
According to Intel SOM, different microarchitectures have different limitations in the use of LEA instructions, aka slow LEA in some cases. But none of current Intel schedule model describes it. It wo…
-
I've observed a fairly nasty performance regression when switching to 1.82 related to compression performance that appears to boil down to some very weird/suboptimal machine code being generated.
Del…