Messages in this thread | | | Date | Thu, 29 Jan 2004 13:02:51 +0000 | From | Russell King <> | Subject | Re: [PATCH] PC300 update |
| |
On Thu, Jan 29, 2004 at 09:02:22AM +0000, Russell King wrote: > If _any_ PCI ID table which is part registered as part of a driver is > marked using __devinitdata or __initdata, this will either cause the > kernel to read invalid data (possibly entering a long loop) or oops.
After doing some more digging, I don't think __devinitdata is a problem anymore.
There seem to be two scenarios where we look at the PCI device ID tables:
- when a new PCI device is added - when the drivers newid file is written to
The first case should only ever occur if CONFIG_HOTPLUG is set (and indeed we only compile PCMCIA/Cardbus if it is.)
The second case is disabled if CONFIG_HOTPLUG is not set.
Therefore, I think marking PCI device ID tables with __devinitdata should theoretically be fine, but marking them with __initdata is most definitely unsafe.
-- Russell King Linux kernel 2.6 ARM Linux - http://www.arm.linux.org.uk/ maintainer of: 2.6 PCMCIA - http://pcmcia.arm.linux.org.uk/ 2.6 Serial core - 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/
| |