linux-kernel - Re: [Linux] Linux PID algorithm is BRAINDEAD!

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <CAOCW0DhJGC1jCjoG69gjvHG5uwkjxb-GB9N7=S=XWsHL5z+fNA@mail.gmail.com>
Date:	Sat, 10 Oct 2015 23:51:48 -0400
From:	Dave Goel <deego3@...il.com>
To:	"Theodore Ts'o" <tytso@....edu>, Dave Goel <deego3@...il.com>,
	linux-kernel@...r.kernel.org
Subject: Re: [Linux] Linux PID algorithm is BRAINDEAD!

Hi Ted,

Thanks for responding. Fair points all of them.

I would like to take exception to one of them, the bottleneck part:

> The biggest problem is that accessing this free pid queue is now
> a locking bottleneck --- especially on a very large NUMA system

That was exactly what I was trying to say towards the end: the the
queue idea or implementation need not be strict. No one cares if
instead of grabbing the very firts aof
the queue, you grab, say, the third element. The /only/ real
requirement is that newly entered elements not go near the head of the
queue.

I would argue that bottleneck or resource locking doesn't exist at all:

If you have n cpu's, you can have n queues, albeit each now smaller.
(Or, equivalently, the cpu's deciding to divvy up the queue into a mod
n space.). And, the populator pushing into them in a round-robin
fashion.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/