Re: PATCH 2.5.x disable BAR when sizing

David Mosberger (davidm@napali.hpl.hp.com)
Fri, 20 Dec 2002 01:42:33 -0800


>>>>> On Fri, 20 Dec 2002 05:57:23 +0000 (UTC), torvalds@transmeta.com (Linus Torvalds) said:

Linus> DO NOT DO THIS. It locks up some machines at
Linus> bootup. Hard. Total bus lockup if you have legacy USB enabled
Linus> (or anything else that does DMA, for that matter) at the same
Linus> time as probing the northbridge with this.

Linus> Trust me. If you have some new silly ia64-specific bug, the
Linus> fix is _not_ to break real and existing hardware out there.

Could you please stop this ia64 paranoia and instead explain to me why
it's OK to relocate a PCI device to (0x100000000-PCI_dev_size)
temporarily? That just seems horribly unsafe to me. The PCI spec
seems to say the same as it says pretty clearly that memory decoding
should be disabled during BAR-sizing. If certain bridges cause
problems, perhaps those need to be special-cased?

--david
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/