Messages in this thread | | | Date | Mon, 11 Aug 2008 10:55:42 -0400 (EDT) | From | Steven Rostedt <> | Subject | Re: [PATCH] printk: robustify printk |
| |
On Mon, 11 Aug 2008, Andi Kleen wrote:
> $INSERT-ANY-OTHER-PRINTK-USE$ > > The point was that printk may have been perfectly adequate for > them with its existing buffer sizes and dynamic wake up frequency, > and might not anymore with the timer wakeup change. Essentially > it is making a widely used kernel facility more fragile.
I've measured several latencies in the kernel that would cause klogd to take serveral jiffies to wake up. I highly doubt that adding one jiffy will break anything. And if it did, then it would show a bug in their system. If adding one jiffy causes lost data, then the system administrators are relying on a utility (klogd) that can easily fail them without these patches.
If we are making a highly fragile setup a bit more fragile, maybe it will wake the IT guys up to change their setup.
-- Steve
| |