Re: [RFC PATCH 1/3] Unified trace buffer

[Linus Torvalds]

On Wed, 24 Sep 2008, Linus Torvalds wrote:
> 
>  - case 1: TSCExtend marker
> 	data = extended TSC (bits 28..59)
> 	size = 8
> 
>  - case 2: TimeStamp marker
> 	data = tv_nsec
> 	array[0] = tv_sec
> 	size = 16

Btw, in case it wasn't clear, those two are totally different things.

The "case 1" thing is the thing that gets inserted automatically by the 
trace code when it's needed because the 27-bit TSC is too limited.

The "case 2" thing is to allow us to occasionally synchronize with some 
global known wall-time clock like the HPET + xtime. IOW, it would be 
something that on demand creates a mapping from wall clock to TSC for that 
particular CPU. 

I guess I should perhaps have put the TSC frequency in there in that "case 
2" thing too. Maybe that should be in "data" (in kHz) and tv_sec/tv_nsec 
should be in array[0..1], and the time sync packet would be 24 bytes.

			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/