On Fri, Feb 22, 2013 at 03:05:39PM +0100, Peter Zijlstra wrote:
On Fri, 2013-02-22 at 13:50 +0100, Frederic Weisbecker wrote:
Which is a problem how?
So here is a possible scenario, CPU 0 reads a kcpustat value, and CPU 1 writes it at the same time:
//Initial value of "cpustat" is 0xffffffff == CPU 0 == == CPU 1 == //load low part mov %eax, [cpustat] inc [cpustat] //Update the high part if necessary jnc 1f inc [cpustat + 4] 1: //load high part mov %edx, [cpustat + 4]Afterward, CPU 0 will think the value is 0x1ffffffff while it's actually 0x100000000.
atomic64_read() and atomic64_set() are supposed to take care of that, without even the need for _inc() or _add() parts that use LOCK.
Sure I get that, but again, why is that a problem,.. who relies on these statistics that makes it a problem?
I guess we want to provide at least some minimal reliability in /proc/stat I mean we don't mind if the read is slightly off, reading stats from userspace is inherently racy anyway, but if it suddenly shows a wrong increase of 4 billions which disappear soon after, it looks like a bug to me.