Re: 2.5.50 + e100 benchmarking

Serge Kuznetsov (serge@wcom.ca)
Mon, 9 Dec 2002 13:08:56 -0500

Messages sorted by: [ date ][ thread ][ subject ][ author ]
Next message: Linus Torvalds: "Re: /proc/pci deprecation?"
Previous message: Martin Schwidefsky: "Re: [PATCH] compatibility syscall layer (lets try again)"
Maybe in reply to: Anton Blanchard: "2.5.50 + e100 benchmarking"

> e100_intr was the worst function in a profile. PCI reads are very costly

I profiled this part and I found what the main issue is the e100_alloc_skbs.
It takes from 10000 to ~19000 processor's tacts.
At the same time eepro100's speedo_iterrupt is pretty fast.

All the Best!
Serge.
-
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/

Next message: Linus Torvalds: "Re: /proc/pci deprecation?"
Previous message: Martin Schwidefsky: "Re: [PATCH] compatibility syscall layer (lets try again)"
Maybe in reply to: Anton Blanchard: "2.5.50 + e100 benchmarking"