Messages in this thread | | | Subject | RE: My vote against eepro* removal | From | Steven Rostedt <> | Date | Thu, 19 Jan 2006 17:57:25 -0500 |
| |
On Thu, 2006-01-19 at 14:09 -0500, Steven Rostedt wrote: > On Thu, 2006-01-19 at 11:26 +0100, kus Kusche Klaus wrote: > > > From: Lee Revell > > > On Thu, 2006-01-19 at 08:19 +0100, kus Kusche Klaus wrote: > > > > Last time I tested (around 2.6.12), eepro100 worked much better > > > > in -rt kernels w.r.t. latencies than e100: > > Try the latest -rt kernel with e100 to see if it still is a delay. You > can also run in PREEMPT_DESKTOP so that the interrupt handlers are not > threads and see if that shows up in the latency.
I just booted up 2.6.15-rt6 (PREEMPT_DESKTOP, regular soft and hard irqs) on a 366 MHz UP machine with init=/bin/bash. Loaded the e100 driver, setup the network. Then started to ping it from another box. I had a 80 usec latency, and that wasn't even from the network card.
So, e100 should not be a problem. I did see the interrupts go off every 2 seconds too.
Check to see if you still get the latencies with e100 and the latest kernel.
As Lee already said. You notice something fishy __PLEASE__ report it. Arjan's response was that this shows that we should only have one driver for a certain task, otherwise people wont report a problem with one, if the other satisfies their needs. And thus, the problem remains.
-- Steve
- 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/
| |