Re: 2.6.18-rc6-mm1: GPF loop on early boot

[Jeremy Fitzhardinge]

Ingo Molnar wrote:
> the ratio between the number of syscalls vs. the number of context 
> switches is 1-2 orders of magnitude. So a loss of 9 cycles in the 
> syscall path is roughly equal to a loss of 90-900 cycles in switch_to() 
> costs ...
>   

I agree.

> the TLS ABI is just a gcc stupidity. Why did they pick the _second_ 
> extra selector, instead of the first one ...?

Well, it doesn't matter which they chose if its the same for user and 
kernel space. Why does x86-64 have swapgs rather than swapfs?

>  Anyway, perhaps this could 
> be solved by extending gcc with a switch to also generate __thread code 
> off %fs. Probably not worth the pain though ...

If there's gcc hacking to be done, it would be trying to get it make the 
TLS offsets from the PDA/TCB positive rather than negative...

   J
-
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/