Messages in this thread | | | Subject | Re: [RFC][PATCH] TTY flip buffer SMP changes | From | Paul Fulghum <> | Date | Fri, 08 Oct 2004 08:35:06 -0500 |
| |
On Fri, 2004-10-08 at 01:26, Theodore Ts'o wrote: > Even if kmalloc() isn't as fast using two ring buffers which we flip > back and forth, CPU's have gotten a lot faster since when I > implemented the flip buffers some 12 years ago (i.e., 8 Moore law's > doublings ago).
The sk_buff solution does look attractive, particularly for high data rates.
It does seem to carry serious overhead (in relation to ring buffers) for devices with small FIFOs.
At 115200bps, I saw the 16550 driver accumulate ~8 bytes per interrupt. Using 2 sk_buffs per interrupt means 256 sk_buff allocations to push 1KiB (71ms) of data to the line discipline. This amounts to ~3600 sk_buff allocations per second at 115200bps.
-- Paul Fulghum paulkf@microgate.com
- 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/
| |