Date: Sat, 6 Jan 2001 18:25:45 -0500 (EST) From: Andrew Gallatin <gallatin@cs.duke.edu> To: Wilko Bulte <wkb@freebie.demon.nl> Cc: freebsd-alpha@freebsd.org Subject: Re: debugging fpa / FDDI panic Message-ID: <14935.42278.147202.426911@grasshopper.cs.duke.edu> In-Reply-To: <20010106234244.A78789@freebie.demon.nl> References: <20010106214357.X77275@freebie.demon.nl> <14935.34600.806565.787237@grasshopper.cs.duke.edu> <20010106224542.A78582@freebie.demon.nl> <14935.37795.408651.78296@grasshopper.cs.duke.edu> <20010106234244.A78789@freebie.demon.nl>
next in thread | previous in thread | raw e-mail | index | archive | help
Yes. that's what I meant. Sorry for the delay. The maze of #ifdefs in the driver is making me dizzy.. I wonder if there's an unifdef that's smart enough to deal with #if defined().. Anyway, try changing the typdef of the pdq_bus_memaddr_t from typedef volatile pdq_uint32_t *pdq_bus_memaddr_t; to typedef volatile vm_offset_t *pdq_bus_memaddr_t; in the FreeBSD section of ifdefs. Also, printf (with %lx) the value of sc->sc_membase prior to calling pdq_initialize(). If this works, we're still going to need a DMA hack. It will be something like this: #if defined(__FreeBSD__) && defined(__alpha__) #define PDQ_OS_VA_TO_PA(pdq, p) (vtophys((vm_offset_t)p) | (pdq->pdq_type == PDQ_DEFTA ? 0 : alpha_XXX_dmamap_or)) #endif Ick.. Drew To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?14935.42278.147202.426911>