Re: [RFC/PATCH, 2/4] readX_check() performance evaluation

[Linus Torvalds]

On Wed, 28 Jan 2004, Russell King wrote:
> 
> What if the failing PCI access happened in an interrupt routine?
> (I'm thinking of the situation where you may need to read the PCI
> status registers to find out whether an error occurred.)
> 
> Also, for that matter, what if a network device receives an abort
> while performing BM-DMA?
> 
> Do we even care about either of these two scenarios?

We do, and the people who care about readX_check() had better be careful.
Quite possibly the "clear_pcix_error()" has to get a lock and disable
interrupts, and the "read_pcix_error()" routine would release the lock.
But that depends on the hardware - details like whether hardware can track
individual errors on multiple CPU's or not.

And keep in mind, 99% of all people won't ever care.

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