VDSO support for 32bit time functions

VDSO support for 32bit time functions

[Greg KH]

Hi Stefani,

About a year ago you posted a big patch to implement VDSO support for
32bit functions, and the response was a request to clean it up a bit by
breaking up the generic bits into a series to make it easier to review /
apply.

The patch I'm referring to can be found here:
	http://thread.gmane.org/gmane.linux.kernel/1411713

Did that ever happen?

If not, any specific reason why?  Do you have a newer version somewhere
for "modern" kernel versions?

If you're not interested in this anymore, mind if I take it up based on
your last version?

I'm getting some complaints that this type of thing would really be good
as 32bit gettimeofday() on 64bit kernels is really slow (65 nanoseconds
on 32bit vs. 17 nanoseconds on 64bit on a high-end i7 processor.)

thanks,

greg k-h

Re: VDSO support for 32bit time functions

[Stefani Seibold]

Hi Grek,

On Monday, the 20.01.2014, 23:51 -0800 wrote Greg KH:
> Hi Stefani,
> 
> About a year ago you posted a big patch to implement VDSO support for
> 32bit functions, and the response was a request to clean it up a bit by
> breaking up the generic bits into a series to make it easier to review /
> apply.
> 
> The patch I'm referring to can be found here:
> 	http://thread.gmane.org/gmane.linux.kernel/1411713
> 
> Did that ever happen?
> 

I have revamped the patch, but the request for the IA32_EMULATION is
still pending. I have now a solution for this which i had already
implemented, but i need to do some tests.

> If not, any specific reason why?  Do you have a newer version somewhere
> for "modern" kernel versions?
> 

Currently my latest version is for 3.10, but i think i can bring it to
3.13. I will do this next weekend and post the patch.

> If you're not interested in this anymore, mind if I take it up based on
> your last version?
> 

I am interested since i must port again and again to the next kernel
version for my devices. 

Your support will be welcome.

> I'm getting some complaints that this type of thing would really be good
> as 32bit gettimeofday() on 64bit kernels is really slow (65 nanoseconds
> on 32bit vs. 17 nanoseconds on 64bit on a high-end i7 processor.)
> 

Stay tune until next week.

- Stefani

Re: VDSO support for 32bit time functions

[Greg KH]

On Tue, Jan 21, 2014 at 09:54:06PM +0100, Stefani Seibold wrote:
> Hi Grek,
> 
> On Monday, the 20.01.2014, 23:51 -0800 wrote Greg KH:
> > Hi Stefani,
> > 
> > About a year ago you posted a big patch to implement VDSO support for
> > 32bit functions, and the response was a request to clean it up a bit by
> > breaking up the generic bits into a series to make it easier to review /
> > apply.
> > 
> > The patch I'm referring to can be found here:
> > 	http://thread.gmane.org/gmane.linux.kernel/1411713
> > 
> > Did that ever happen?
> > 
> 
> I have revamped the patch, but the request for the IA32_EMULATION is
> still pending. I have now a solution for this which i had already
> implemented, but i need to do some tests.
> 
> > If not, any specific reason why?  Do you have a newer version somewhere
> > for "modern" kernel versions?
> > 
> 
> Currently my latest version is for 3.10, but i think i can bring it to
> 3.13. I will do this next weekend and post the patch.

Very cool, thanks, if you remember, can you cc: me when you post it?

I'll be glad to test it out on some systems here that I have.

thanks,

greg k-h