Messages in this thread | | | Date | Fri, 07 Feb 2003 17:57:46 -0800 | From | "Martin J. Bligh" <> | Subject | Re: [RFC][PATCH] linux-2.5.59_getcycles_A0 |
| |
>> However this doesn't work on systems w/o a synced TSC, so by simply > > Why not? This shouldn't be performance critical and you can make > it monotonous with an additional variable + lock if backwards jumps > should be a problem. > > Also the variations between non synced TSCs should be far below > any watchdog's radar screen.
Not true. They'll drift further and further apart over time. Even a 0.01% crystal difference will eventually kill you.
And if that isn't bad enough think about what happens when I run 180 MHz processors in one node, and 900MHz in another.
You really can't make any assumptions about TSC sync on boxes where they're not synced.
M.
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |