linux-kernel - Re: [tip:sched/core] sched: Lower chances of cputime scaling overflow

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] [day] [month] [year] [list]

Date:	Tue, 30 Apr 2013 16:03:30 +0200
From:	Stanislaw Gruszka <sgruszka@...hat.com>
To:	Linus Torvalds <torvalds@...ux-foundation.org>
Cc:	Peter Zijlstra <peterz@...radead.org>,
	Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
	Ingo Molnar <mingo@...nel.org>,
	"H. Peter Anvin" <hpa@...or.com>,
	Frédéric Weisbecker <fweisbec@...il.com>,
	Steven Rostedt <rostedt@...dmis.org>,
	Andrew Morton <akpm@...ux-foundation.org>,
	Thomas Gleixner <tglx@...utronix.de>,
	linux-tip-commits@...r.kernel.org
Subject: Re: [tip:sched/core] sched: Lower chances of cputime scaling overflow

On Sat, Apr 13, 2013 at 11:44:54AM -0700, Linus Torvalds wrote:
> PS. This is the "Make sure 'total' fits in 32 bits first" version. Not
> really tested, but it's just changing the order of operations a bit.
> 
>     /* We know one of the values has a bit set in the high 32 bits */
>     for (;;) {
>         /* Make sure "rtime" is the bigger of stime/rtime */
>         if (stime > rtime) {
>             u64 tmp = rtime; rtime = stime; stime = tmp;
>         }
> 
>         /* Make sure 'total' fits in 32 bits */
>         if (total >> 32)
>                 goto drop_precision;
> 
>         /* Does rtime (and thus stime) fit in 32 bits? */
>         if (!(rtime >> 32))
>                 break;
> 
>         /* Can we just balance rtime/stime rather than dropping bits? */
>         if (stime >> 31)
>             goto drop_precision;
> 
>         /* We can grow stime and shrink rtime and try to make them both fit */
>         stime <<= 1;
>         rtime >>= 1;
>         continue;
> 
> drop_precision:
>         /* We drop from rtime, it has more bits than stime */
>         rtime >>= 1;
>         total >>= 1;
>     }

For reference I'm attaching test program and script I used to validate
algorithm.

It generates lot of (possibly real) rtime, total, stime values for 4096
threads running for 10 years. Then compare scaled stime values caluclated
by above algorithm with precise python values. 

For all values generated, scaled stime relative error was less than 0.05%

Stanislaw

View attachment "scale_stime.c" of type "text/plain" (1625 bytes)

View attachment "scale_stime_test.py" of type "text/plain" (999 bytes)