经典的 WebSocket.bufferedAmount，在服务端把 TCP recv wnd 置零，调用 WebSocket.send() 会使得数据在代理也就是 Fiddler 而是浏览器中堆积。

现在 Chromium Project Fugu 的目的更是把所有能塞到浏览器里的 API 都塞进去...可以供给使用的侧信道只会越来越多。

https://developer.mozilla.org/en-US/docs/Web/API/WebSocket/bufferedAmount
https://www.chromium.org/teams/web-capabilities-fugu

靠网络侧信道还是能探测出一些的，Fiddler 作为一个 HTTP/HTTPS 代理，

《 let/const for loop 的 v8 实现》
https://zhuanlan.zhihu.com/p/340068236

// {
// let/const x = i;
// temp_x = x;
// first = 1;
// undefined;
// outer: for (;;) {
// let/const x = temp_x;
// {{ if (first == 1) {
// first = 0;
// } else {
// next;
// }
// flag = 1;
// if (!cond) break;
// }}
// labels: for (; flag == 1; flag = 0, temp_x = x) {
// body
// }
// {{ if (flag == 1) // Body used break.
// break;
// }}
// }
// }

啊这

小米卖的是视频广告位投放（ to B ）+ 大屏广告牌租赁（ to C ），播放广告的间隙用户可以使用其它服务。

少想多读书

相比之下，我们这种埋头蹲 https://source.chromium.org/ 看源码的，真的好傻好天真。。。

学习了，

围观学习

@jackrelative 原来是 webview 套壳啊，我终于知道 v 站网友说的副业是啥了。。。

@chengkai 合议庭如果认为不需要开庭，直接就书面审理了（驳回上诉

你还和法官讲这些，法官一是可能根本不懂，二是也根本不在乎。。二审可能庭都不用开，直接就维持原判了

@cht1995 真巧，我也在某个论坛用过 Herb Sutter 的头像...

@meganut 感谢，随便摘抄的

@dadupi
@Gandum

并未持有设备，不清楚。

## Native FP16 on Apple M1, 来自 Steve Canon (Apple Swift team)

A couple random fun numerics things to play with on M1 macs:

- native half-precision arithmetic
- ARMv8.3 complex arithmetic SIMD instructions

https://twitter.com/stephentyrone/status/1328861121382666240

## ARM x86 关于 Memory model 的区别，来自 Bruce Dawson (Google Chrome Team)

mov register-> a
mov register-> b

On x86/x64 the store to "b" is guaranteed to not be visible to other processors before the store to "a". On ARM that is not guaranteed.

Many benign x86/x64 data races will becomes real bugs on ARM, and it appears many programmers don't know this

The emulator _must_ emulate the x86/x64 memory model. It is as important as properly emulating the CPU status flags and arithmetic instructions. So, yes. If Rosetta didn't emulate the memory model it would not be an x86/x64 emulator.

model name : Intel(R) Core(TM) i5-4300U CPU @ 1.90GHz

Detailed Results
Iteration 1 30.08 runs/min
Iteration 2 28.98 runs/min
Iteration 3 31.18 runs/min
Iteration 4 31.04 runs/min
Iteration 5 30.52 runs/min
Iteration 6 30.80 runs/min
Iteration 7 30.52 runs/min
Iteration 8 30.65 runs/min
Iteration 9 30.53 runs/min
Iteration 10 30.71 runs/min
Arithmetic Mean:30.5 ± 0.44 (1.4%)