Date: 1 Nov 1997 08:23:22 GMT From: nnd@itfs.nsk.su To: hackers@freebsd.org Subject: Re: Cyclades :( Message-ID: <63eotq$37v@news.itfs.nsk.su> References: <Pine.BSF.3.96.971029134805.2423A-100000@trojanhorse.ml.org> <199710300212.SAA01513@implode.root.com> <19971030102615.09637@grape.carrier.kiev.ua>
next in thread | previous in thread | raw e-mail | index | archive | help
Alexander Litvin <archer@lucky.net> wrote:
> On Wed, Oct 29, 1997 at 06:12:59PM -0800, David Greenman wrote:
> > >
> > >... Maybye you should let Cyclades do the drivers. They might work then.
> >
> > Jamil, this comment is without merit and does absolutely nothing to help
> > with the problem. Cyclades is quite familiar with the driver in FreeBSD and
> > has done extensive testing with both the ISA and PCI product under FreeBSD.
> > They haven't seen any problems in their testing, but of course that doesn't
> > mean that a problem doesn't exist. The device driver for the PCI version is
> > the same driver that the ISA card uses, the only difference is that the PCI
> > front-end memory-maps the registers into the address space at device attach
> > time. There might be a hardware problem with the PCI interface that is causing
> > the reboots, but at the moment, there is insufficient information to diagnose
> > the problem further. I have not been able to reproduce the reboots here. I
> > suspect that it is going to take a PCI bus analyzer to figure this one out.
> Well, I forgot to say: reboots doesn't happen when you just plag in
> Cyclades. All four boxes -- quite busy access servers, from 10 to
> 15 dialup modems.
> As for comment about using Cyclades for connecting serial terminals
> -- I mentioned that reboots happen with "getty" and don't with
> "mgetty", so callout devices are ok.
> What ditails can I provide? That's not panics, I don't have crashdumps.
> May be exact hardware configurations? If so -- I'll test it once again
> and write exactly all hardware details.
I can confirm that there IS the REAL problem
with Cyclades Cyclom 16YeP and some number of MBoards and
FreeBSD versions. My last test case was sufficiently simple -
Cyclades Cyclom-16YeP card + one modem on any of 16 ports
+ FreeBSD-2.2-stable (of some september 1997 version)
To crash (reboot) the system I must some ( ~10 ) times
switch the power of the modem off/on quickly.
After discussing this problem in some of Relcom's
newsgroups I've received patched version of FreeBSD's 'cy'
driver. Message's author (Andrey Ovodenko - ovod@dipt.donetsk.ua)
says that they've successfully used this patched version wtih
PCI Cyclades cards at least five month now.
Driver's text inspection shows that there is ONLY ONE
significant difference with 'standard FreeBSD's' cy-driver:
(this is a 'diff -bu' result for /sys/i386/isa/cy.c version
1.55 of 1997/09/21)
--- cy.c.ORIG Sat Nov 1 13:33:19 1997
+++ cy.c Sat Nov 1 13:36:55 1997
@@ -410,7 +404,7 @@
#endif
static int cy_chip_offset[] = {
- 0x0000, 0x0200, 0x0400, 0x0600, 0x0100, 0x0300, 0x0500, 0x0700,
+ 0x0000, 0x0400, 0x0800, 0x0c00, 0x0200, 0x0600, 0x0a00, 0x0e00
};
static int cy_nr_cd1400s[NCY];
static int cy_total_devices;
Can someone with Cyclom-8/16YeP (and problems ;-)
test this patch ? I myself can not test it because my only ONE
Cyclom-16YeP now busily serves 12 modems on Linux box.
Partial "correctnes proof" for this patch can be found
in (working) Linux 'cy' (or 'cz') driver which uses the same
chip_offset addresses as in "patched" driver, but not as in
original FreeBSD's 'cy' driver.
N.Dudorov
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?63eotq$37v>