又再次看到了 Spectre Mitigation 的效能損失...

Hacker News 首頁上看到的文章,講 Spectre Mitigation 的效能損失:「Spectre Mitigations Murder *Userspace* Performance In The Presence Of Frequent Syscalls」,對應的討論串在「Spectre Mitigations Murder Userspace Performance (ocallahan.org)」。

看起來作者是在調校 rr 時遇到的問題,幾年前有提到過 rr:「Microsoft 的 TTD 與 Mozilla 的 RR」。

對此作者對 rr 上了一個 patch,減少了 mitigation code 會在 syscall 時清掉 cache 與 TLB,這個 patch 讓執行的速度大幅提昇:「Cache access() calls to avoid syscalls」。

另外作者提到了他的硬體是 IntelSkylake,他又再跑一次 pre-patch 與 post-patch 的速度,可以看到在 pre-patch 前,mitigation 會讓系統慢超多 (從 2m5.776s 到 3m19.648s),而 post-patch 後大幅降低 syscall 的使用,就不會影響那麼多 (從 0m33.422s 到 0m36.160s)。

就目前知道的 mitigation 方式來說,這個猜測應該是對的...

cURL 與 Travis CI 的事件

前幾天 Daniel Stenberg (cURL 的發明人與現在的維護者) 發表一篇從 Travis CI 搬出來的文章,換到 Zuul CICircle CI 上:「Bye bye Travis CI」,對應的 Hacker News 的討論可以在「Bye Bye Travis CI (haxx.se)」這邊翻到。

文章裡提到主要的兩個點是,Travis CI 當初有承諾會提供免費服務給 open source project:

今年的時候整個 Travis CI 的商業收費的機制改了,另外也嚴格增加了對 open source project 的限制,包括了你不能收到任何商業公司或是任何組織的贊助:

Project must not be sponsored by a commercial company or organization (monetary or with employees paid to work on the project)

另外 Daniel Stenberg 在文章裡也表明目前不打算付錢,要找市場上便宜可用的方案 (而目前看起來還有,至少 Zuul CI 與 Circle CI 都在選項內),所以就從 Travis CI 搬離了:

Lots of people have commented and think I’m “whining” about Travis CI charging for something that is useful and that I should rather just pay up. I could probably have gone with that but I dislike their broken promise and that they don’t consider us Open source anymore and I feel I have a responsibility to use the funds we get from gracious donors as wisely and economically as possible, and that includes using no-cost or cheap services rather than services charging thousands of dollars per year.

If there really were no other available and viable options, then paying could’ve been an alternative. Now, moving on to something else was the right choice for us.

然後 Travis CI 過了兩天丟出「Open Source Terms at Travis CI – An Update and Clarification」,雖然沒有表明是在講 cURL,但放在一起看其實也都大概知道發生什麼事情。

看了一下自己的小專案 (更新頻率不高,test case 也不多),丟 Circle CI 還算是夠用,另外自己也有弄個 GitLab,需要的時候也可以在上面跑 CI。

這邊另外看到 cURL 這種大型專案,因為 test case 數量很多,丟到不同的 CI vendor 上跑,看起來是個還不錯的架構...

用圖片表示 SHA-256 的值

Hacker News 上看到的用法,把 SHA-256 的值表示成 avatar:「Representing SHA-256 Hashes As Avatars」,Hacker News 上的討論在「Representing SHA-256 Hashes as Avatars (francoisbest.com)」這邊可以看到。

做出來大概是這樣,這邊的 e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855 是空字串的 SHA-256 值:

裡面有提到一些演算法,讓圖片看起來不要那麼「奇怪」,還蠻有趣的...

在文章最後面還有一些其他的圖片類型,如果要自己搞 avatar service 的人可以參考演算法的部份,自己搞一套出來玩。

Chromium 看起來正在實做 CSS4 的 :has()

先前一直都有在盯著「Issue 669058: CSS selectors Level 4: support :has()」這個,主要是有這個功能會對於很多應用很方便,無論是對 designer 在設計上,或是對 javascript developer 的操作。

Chromium 上對應的 commit 在「Supports :has() pseudo class matching」這邊,可以看到目前只實做了一部分:

So this CL only supports :has argument selectors starting with
descendant combinator. Argument selectors starting with other
combinators are not supported yet.
 - .a:has(.b)    : Supported
 - .a:has(> .b)  : Not supported yet
 - .a:has(~ .b)  : Not supported yet
 - .a:has(+ .b)  : Not supported yet

但至少是個進展了...

Brendan Gregg 遇到的 An Unbelievable Demo

Hacker News Daily 上看到的熱門話題,Brendan Gregg 是效能分析領域的大老,現在在 Netflix 工作,在維基百科的條目「Brendan Gregg」上也有提到他的一些知名發明,像是 Flame Graphs:

He has also created visualization types to aid performance analysis, including latency heat maps, utilization heat maps, subsecond offset heat maps, and flame graphs.

昨天他發了一篇文章在講之間他遇到的事情,原文把過程寫的很有戲劇性,值得去看一看:「An Unbelievable Demo」,而 Hacker News 上的討論也很精彩:「An Unbelievable Demo (brendangregg.com)」,還引出了 Colin Percival 也分享他的經驗。

快速講 Brendan Gregg 遇到的事情,2005 年時 Brendan Gregg 因為業務上的需要 (他當年是效能分析的顧問),幫 Sun 推出的 DTrace 寫了一包工具,叫做 DTraceToolkit,用 GPLv2 或是 CDDL 釋出。

這包工具被 Sun 的人拿去用,並且拔掉作者與授權資訊,然後還被拿去「世界巡迴」介紹這個工具,最後在雪梨的時候居然是拿來介紹給 Brendan Gregg,然後被原作者打臉。

不過他後來還是加入了 Sun... XDDD

Colin Percival 的故事則沒有牽扯到 copyright issue,不過也很有趣,這邊提到的是 bsdiff,也是個經典的工具:

Reminds me of when Apple started providing "smaller size updates" to OS X. I was curious about the details since my doctorate had touched on the topic, so I worked my contacts (I had a few in Apple engineering from the FreeBSD / OS X relationship) and after a few months I got back as answer: "We're using a tool called bsdiff, are you familiar with it?" I was indeed, since I was the author of said tool.

(Just to be clear, there was no license violation involved in this case; just a lack of awareness of the provenance of the open source software they were using.)

另外在其他的 thread 裡面,可以看到 Brendan Gregg 也有浮上來回應 (可以直接字串搜尋 brendangregg),裡面也提到了有趣的故事,像是他另外一個發明 latency heat map 在一些會場上的交流,以及自己也有遇到其他工具的作者:

Thanks. There was a time when many observability products were adding latency heat maps, and at one conference expo floor there were three companies with latency heat maps on their screen at the same time, pitching them as a flagship feature. If I walked near them they'd start trying to explain them to me, and I never figured out an appropriate response. If I said "hey, great to see you added them, I invented these back at Sun" I'd get funny looks.

I think it's a small world, and everything is software, so the chance you'll bump into someone who wrote software you are using I think is pretty high. I was once trying to get my head around Andi Kleen's pmu-tools, and I had the github repo open in my browser on my laptop I was carrying, when the guy sitting next to me on a bus says he's Andi Kleen. (Ok, it was a bus taking Linux conference attendees to an event, not a random bus, but I still found it remarkable timing -- I was studying pmu-tools at that exact time!)

拿來配啤酒的文章 XD

Stack Overflow 被 Prosus 併購

Hacker News 首頁上看到的消息,Stack OverflowProsus 併購,直接站上第一名:「Stack Overflow Sold to Tech Giant Prosus for $1.8 Billion」,Prosus 的官方新聞稿也已經出來了:「Prosus to acquire Stack Overflow for US$1.8 billion」,Hacker News 上的討論:「Stack Overflow sold to Prosus for $1.8B (wsj.com)」也很熱鬧。

算是大消息沒錯,不過不知道後續會有什麼樣的影響... Prosus 算是投資集團,不知道會對 Stack Overflow 有什麼想法,怎麼介入或是整合手上的資源。

原來 Sentry 有支援 Bash...

Sentry 可以拿來蒐集自己開發軟體的錯誤事件,剛剛突發奇想用搜尋引擎找了一下,發現 Sentry 是有支援 Bash 的:「Sending Sentry Events from Bash」。

文章裡面也直接提到了 hook 的用法,在裝了 sentry-cli 後可以掛進來:

#!/bin/bash
export SENTRY_DSN=
eval "$(sentry-cli bash-hook)"

自己架 Sentry 的步驟我有丟在「Sentry」這邊,或者也可以用雲端的版本。

丟給同事參考看看...

CloudFront 把本來的 Lambda@Edge 產品線拆細,推出 CloudFront Functions

Amazon CloudFront 本來的 Lambda@Edge 產品線拆細,多出一個 CloudFront Functions:「Introducing CloudFront Functions – Run Your Code at the Edge with Low Latency at Any Scale」。

就產品面的角度就是限制比 Lambda@Edge 多,但價錢變便宜很多。

先看價錢的部份,CloudFront Functions 的價錢只有 request:

Invocation pricing is $0.10 per 1 million invocations ($0.0000001 per request).

而 Lambda@Edge 則是兩筆費用,光是 request 費用就是六倍:

Request pricing is $0.60 per 1 million requests ($0.0000006 per request).

Duration is calculated from the time your code begins executing until it returns or otherwise terminates. You are charged $0.00005001 for every GB-second used.

當然,CloudFront Functions 便宜帶來的限制也不少,最主要的限制可以從最大執行時間只有 1ms,以及記憶體只能用 2MB 就可以看出來:

但這對於輕量的操作來說已經夠用了,主要就是對 HTTP header 的操作...

另外比較表上看到個有趣的點「JavaScript (ECMAScript 5.1 compliant)」,這樣應該就不會是 Node.js (V8 engine),而是其他的 JS engine?