Date: Thu, 23 Dec 2004 00:03:01 +0200 From: Sven Petai <hadara@bsd.ee> To: Wilko Bulte <wb@freebie.xs4all.nl> Cc: freebsd-alpha@freebsd.org Subject: Re: alpha/75317: ATA DMA broken on PCalpha Message-ID: <200412230003.02588.hadara@bsd.ee> In-Reply-To: <20041222184436.GA50553@freebie.xs4all.nl> References: <200412221730.iBMHUJC6097807@freefall.freebsd.org> <20041222184436.GA50553@freebie.xs4all.nl>
next in thread | previous in thread | raw e-mail | index | archive | help
On Wednesday 22 December 2004 20:44, Wilko Bulte wrote: > > anyway this still doesn't make things work correctly for me, > > because the real problem seems to be the Pyxis > > page crossing bug. Basically it comes down to corrupting > > DMA transfers larger than 8k. It didn't cause problems before, > > since we never did larger than PAGE_SIZE transfers before > > the ATA dma change mentioned in the original report. > > There's a detection code for the buggy chip @ > > src/sys/alpha/pci/cia.c > > but it's little too naive, since it assumes only DEC_ST550 can > > have it, in reality it seems to be used in some very early > > revisions of 164LX(SX too?). But there doesn't seem to be a > > Yes, the SX too: > what I actually meant was that sure enough ALL the SX and LX boxes have PYXIS chips but _maybe_ some of the early ones don't have the fix for 8k problem, since the fix consists of adding some off-chip logic, it's not detectable in the version of Pyxis chip itself. This seems to be only reasonable explanation to why everything works fine when I hack PAGE_SIZE to be the max segment size for ata dma and causes machine checks, unknown signals and other anomalies for anything larger. [snip] > cia0: Pyxis, pass 1 <<<<<---- pass 1 > cia0: extended capabilities: 1<BWEN> > > Interestingly enough the pass 1 thing is not fully reliable as a trouble > indicator: yes, since the fix is implemented off-chip... [snip] > so without the warning. In Miatas using a 32bit slot instead of a 64bit > PCI slot avoided the problem. but that doesn't help you in the case of onboard ata... actually it would be nice to know if the early Miatas that are detected as having faulty Pyxis, have the same issue when using onboard ata in the CURRENT, they certainly should... > > My 164SX does not report any warning. yeah since the code currently only checks if it's early version of Miata, it doesn't try to trigger the bug to see if it exists or do anything else fancy like that...
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200412230003.02588.hadara>