Re: RFC: avoid asmlinkage on x86 traps/interrupts

[Andrea Arcangeli]

On Tue, Nov 02, 2004 at 12:56:54PM -0800, Linus Torvalds wrote:
>  	if (curctx != irqctx) {
> +		int arg1, arg2, ebx;
> +
>  		/* build the stack frame on the IRQ stack */
>  		isp = (u32*) ((char*)irqctx + sizeof(*irqctx));
>  		irqctx->tinfo.task = curctx->tinfo.task;
>  		irqctx->tinfo.previous_esp = current_stack_pointer();
>  
> -		*--isp = (u32) &regs;
> -		*--isp = (u32) irq;
> -
>  		asm volatile(
>  			"       xchgl   %%ebx,%%esp      \n"
>  			"       call    __do_IRQ         \n"
> -			"       xchgl   %%ebx,%%esp      \n"
> -			: : "b"(isp)
> -			: "memory", "cc", "eax", "edx", "ecx"
> +			"       movl   %%ebx,%%esp      \n"
> +			: "=a" (arg1), "=d" (arg2), "=b" (ebx)
> +			:  "0" (irq),   "1" (regs),  "2" (isp)
> +			: "memory", "cc", "ecx"
>  		);

why do you restore the parameters into arg1/arg2/ebx? since it's
volatile I doubt gcc can optimize them away despite they're clearly
going to be discarded.

I guess it'd be nicer to simply move the output into the input with "a",
"d", "b", and the not add any output at all, and put "eax/edx" back into
the clobbers.
-
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/