diff options
| author | Todd C. Miller <millert@cvs.openbsd.org> | 2000-05-04 19:42:54 +0000 |
|---|---|---|
| committer | Todd C. Miller <millert@cvs.openbsd.org> | 2000-05-04 19:42:54 +0000 |
| commit | 0d99ffb220193247e21c174e5e18cb337d025e99 (patch) | |
| tree | aec8acb46e3e6150a057a03e87d0b44f67c16a3b /sys | |
| parent | ba5109fe82aeb027bef7a1ecff625ce209b2b0b7 (diff) | |
Don't attach a PCTECH RZ1000 on the PCI bus--it has severe problems.
It will attach at wdc and be less problematical, though I still
wouldn't trust my data to it.
Diffstat (limited to 'sys')
| -rw-r--r-- | sys/dev/pci/pciide.c | 10 |
1 files changed, 9 insertions, 1 deletions
diff --git a/sys/dev/pci/pciide.c b/sys/dev/pci/pciide.c index e62efc2287b..3acd59e9360 100644 --- a/sys/dev/pci/pciide.c +++ b/sys/dev/pci/pciide.c @@ -1,4 +1,4 @@ -/* $OpenBSD: pciide.c,v 1.22 2000/04/10 07:06:17 csapuntz Exp $ */ +/* $OpenBSD: pciide.c,v 1.23 2000/05/04 19:42:53 millert Exp $ */ /* $NetBSD: pciide.c,v 1.48 1999/11/28 20:05:18 bouyer Exp $ */ /* @@ -448,6 +448,14 @@ pciide_match(parent, match, aux) const struct pciide_product_desc *pp; /* + * Some IDE controllers have severe bugs when used in PCI mode. + * We punt and attach them to the ISA bus instead. + */ + if (PCI_VENDOR(pa->pa_id) == PCI_VENDOR_PCTECH && + PCI_PRODUCT(pa->pa_id) == PCI_PRODUCT_PCTECH_RZ1000) + return (0); + + /* * Check the ID register to see that it's a PCI IDE controller. * If it is, we assume that we can deal with it; it _should_ * work in a standardized way... |