Re: What's cooking in git.git (Apr 2016, #07; Mon, 25)

[Johannes Schindelin <Johannes.Schindelin@gmx.de>]

Hi Junio,

On Wed, 27 Apr 2016, Junio C Hamano wrote:

> Johannes Schindelin <Johannes.Schindelin@gmx.de> writes:
> 
> > Hi Junio,
> >
> > On Mon, 25 Apr 2016, Junio C Hamano wrote:
> >
> >> * js/win32-mmap (2016-04-22) 3 commits
> >>   (merged to 'next' on 2016-04-22 at cd39c60)
> >>  + mmap(win32): avoid expensive fstat() call
> >>  + mmap(win32): avoid copy-on-write when it is unnecessary
> >>  + win32mmap: set errno appropriately
> >> 
> >>  mmap emulation on Windows has been optimized.
> >
> > Please note that it is not purely an optimization. It is also a bug fix in
> > case of a pretty full disk: on Windows, mmap() is backed by the page file
> > if it is in copy-on-write mode, and that can fail when the free space on
> > the drive that has the page file drops below a certain threshold.
> 
> Thanks.  How does this look then?
> 
>     * js/win32-mmap (2016-04-22) 3 commits
>       (merged to 'next' on 2016-04-22 at cd39c60)
>      + mmap(win32): avoid expensive fstat() call
>      + mmap(win32): avoid copy-on-write when it is unnecessary
>      + win32mmap: set errno appropriately
> 
>      mmap emulation on Windows has been optimized and work better without
>      consuming paging store when not needed.
> 
>      Will merge to 'master'.

Perfect!

> Reducing paging file consumption still falls into optimization in my
> dictionary, though ;-)

Well, I considered this a bug fix when I worked on it: the command should
not have failed, and with my fix, it didn't ;-)

Ciao,
Dscho