Surge 2015 - Lightning Talks

Channel:OmniTI · Length:1:19:43 · Published:2015-11-29 · Views:11550

Watch on YouTube

Summary

A collection of lightning talks from the Surge 2015 conference on scalable systems, hosted by Geir Magnusson. HN mostly points to one segment, a Bryan Cantrill talk touching on madvise semantics.

Why HN recommends it

Cited as a classic Cantrill talk when threads discuss madvise, malloc and Dirty COW behavior.

This instantly reminded me of this classic Bryan Cantrill talk — HN, 2020

Operating systems & kernels · Cloud, DevOps & reliability

Subtitles (excerpt)

these are the lightning talks for surge of 2015 it's um an event that has had a long history here of of being I think a lot of fun um we don't do the standard format of 2 minutes 2 minutes 2 minutes we've opened it up to 5 minutes thank you Brian for correct me again and um they can be funny they can be serious they can be just about anything um and sometimes they're very surprising um I want to thank Robert and the rest of the uh Team for for allowing me to do this uh and putting together such a great con uh conference I want to thank the sponsor app Nexus for making this possible with the uh support they gave and um without any further Ado we will start with our first talk it's a gentleman named Dan McDonald and he's speaking about two future internet directions um my name is Dan McDonald I'm with omniti um I'm the OS I is my day job and I'd like to talk about two upcoming directions that the Academia and research sectors of the internet Community are working on um most of you know what the internet architecture looks like it's the protocol hourglass where you have your apps on top your transport protocols and your are your presentation protocols your transport protocols and then you have IP in the middle

On Hacker News

Although this is true on Linux, it's worth noting that this behavior doesn't match the specified behavior of posix_madvise.[1] It was also a subject of Bryan Cantril's hilarious "A crime against common sense" Surge 2015 talk (at ~1:06:50).[2] [1]: [LINK] [2]: [VIDEO] — kevinoid, 2020-03
From the paper. > and subsequently allow the kernel to immediately reuse the associated physical memory. On Linux, this can be achieved by passing the MADV_DONTNEED flag to the the madvise system call. Shouldn't this be MADV_FREE? This instantly reminded me of this classic Bryan Cantrill talk [VIDEO] Edit: It seems that the Linux behavior is relied upon? From later in the paper. > Note that it is… — gbrown_, 2020-01
Perhaps the talk[1] given by Bryan Cantrill in 2015, and mentioned in footnote 1 of the article, may have alerted someone to the possibility that there might be something exploitable here? In addition, any of the other online discussions of madvise(,,MADV_DONTNEED), mentioned in that talk, may have attracted someone's interest. This exploit seems to lie at the intersection of several things that … — mannykannot, 2017-05

Related videos