Linked by snydeq on Tue 7th Oct 2008 17:04 UTC
Thread beginning with comment 332895
To view parent comment, click here.
To read all comments associated with this story, please click here.
To view parent comment, click here.
To read all comments associated with this story, please click here.
Actually it depended on which version of Linux 2.4 and what drivers were running. Many things didn't handle rollover well and if your SCSI controller decided command reset timeout was now 390 days in the future, the system may as well have been locked up.
At any rate, my original point (that I forgot to write out) was that the Windows bug wasn't a memory corruption bug, but this timer bug.






Member since:
2005-07-20
I believe that was a 32-bit timer count rollover bug, just like the one Linux 2.4 had for a while, except that Linux used a slower count and therefore lasted 400+ days.
That bug is the reason that the Linux 2.6 kernel now initializes the jiffies count to -5 minutes, forcing a rollover in order to catch that bug.