ifconfig ppp0 shows apprimately 10 RX-Errors per second and also a
excessive high rate of TX overruns - which may be misleading. May be, that the
wrong counter is incremented, so these overruns should be read as RX-overruns.
hm
On Thu, 05 Nov 1998, Theodore Y. Ts'o wrote:
>Date: Thu, 5 Nov 1998 07:26:49 -0500 (EST)
> From: David G Hamblen <dave@afr-olt.com>
>
> I've been following the kernel digest, and searching for solutions to this
> problem, and several recent posts are suggesting that the NIC driver is at
> fault, or that its just a debugging message which will go away in 2.2. I
> get 1/3 the ftp transfer rate with 2.1.125 and 2.1.127-pre7 compared to
> 2.0.33. My problem occurs using ppp and serial drivers; so I would think
> the only common thing between my problem and the others is the tcp_ipv4
> code. Since I've got an old tired serial interface, my guess is that
> interrupts aren't being serviced fast enough?
>
>Umm.... I'm using a Linux 2.1.126 kernel with a V.90 56kbps modem card,
>and I'm getting 4-4.5kbps FTP transfer rates using PPP and the serial
>driver via the PCMCIA interface. So it's not quite that simple....
>
>Are you using autoirq with setserial? That broke for some people when
>we regularized the IRQ interface, but testing for that is very hard,
>becuase it's dependent on the vagrancies of random UART's chips,
>particularly the non-standard (i.e., non-National Semiconductor) UARTs.
>
>So if you're using autoirq, and it failed, the irq would be set to 0,
>and that would certainly explain the lousy performance. I very, very,
>very, strongly recommend that people NOT using autoirq. Trying to do
>automatic IRQ detection with the ISA bus is dicecy at best, because the
>ISA bus really isn't designed for it.....
>
>I recommend that people use setserial and manually set the irq to the
>correct value. In the latest (just released) setserial, there's a much
>nicer way of doin this which uses a file /etc/serial.conf as a place for
>configuring the IRQ information.
>
> - Ted
>
>-
>To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
>the body of a message to majordomo@vger.rutgers.edu
>Please read the FAQ at http://www.tux.org/lkml/
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/