Messages in this thread | | | Date | Sun, 21 Jul 2002 23:56:11 +0200 (CEST) | From | Ingo Molnar <> | Subject | Re: [patch] "big IRQ lock" removal, 2.5.27-A9 |
| |
On Sun, 21 Jul 2002, Christoph Hellwig wrote:
> > the genhd.c bit is safe as well, removed the comment. > > Is there any reason the sti is there at all? In -dj almost all drivers > use module_init() now so it becomes increasingly useless..
well, indeed. While the sti() can be understood to a certain degree - we used to boot with the IRQ lock on and accidentally leaving it enabled can cause problems - but otherwise preceeding code should not disable interrupts in an unbalanced way. I've removed the __sti() from my tree.
there's even more ancient code in the block driver init path, eg. in drivers/block/ll_rw_blk.c:blk_dev_init():
outb_p(0xc, 0x3f2);
i suspect this is ancient Linux code. 0x3f2 is one of the floppy controller ports - many modern x86 boxes do not even have a floppy controller! I've removed this from my tree as well - if this is needed at all then it belongs into the floppy driver. Latest patch is at:
http://redhat.com/~mingo/remove-irqlock-patches/remove-irqlock-2.5.27-B0
Ingo
- 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/
| |