Subject: Re: panic: DMAWAIT1
To: der Mouse <mouse@Collatz.McRCIM.McGill.EDU>
From: Rolf Grossmann <grossman@informatik.tu-muenchen.de>
List: port-sparc
Date: 05/16/1995 12:01:34
Hi,
on Tue, 16 May 1995 00:25:00 +0200 der Mouse wrote
concerning "panic: DMAWAIT1" something like this:
> That upgrade from 1.0 to -current now seems to be complete; I have a
> system that is capable of rebuilding itself successfully. There's only
> one minor fly in the ointment: every time I halt the system, I get
> syncing disks... CSR = 80001118
> panic: DMAWAIT1
> Stopped at 0xf80d4b2c: jmpl [%o7 + 0x8], %g0
> db>
Oh, I thought I was the only one seeing this problem due to my old and
slow ss1. This small patch solved the problem, but I have no idea whether
it is correct.
diff -c -r1.1.1.2 dma.c
*** 1.1.1.2 1995/04/03 18:31:33
--- dma.c 1995/05/08 16:53:07
***************
*** 276,282 ****
/* load the count in */
ESPCMD(sc->sc_esp, ESPCMD_NOP|ESPCMD_DMA);
! DMAWAIT1(sc);
/* clear errors and D_TC flag */
DMACSR(sc) |= D_INVALIDATE;
--- 276,282 ----
/* load the count in */
ESPCMD(sc->sc_esp, ESPCMD_NOP|ESPCMD_DMA);
! /* DMAWAIT1(sc); ### */
/* clear errors and D_TC flag */
DMACSR(sc) |= D_INVALIDATE;
> Last time I
> tried for a kernel coredump with the modern scsi subsystem, it said
> sddump() wasn't implemented....
Remove the #undef SCSIDUMP from sd.c, but I guess there is a good reason
why it's in there (I guess it could trash your disk's data) ;))
Bye, Rolf