Re: large files and low memory
- From
Nguyen Thai Ngoc Duy <pclouds@gmail.com>
- Date
- Oct 5, 2010, 11:55 UTC
- Message-ID
- <AANLkTi=ni_21uP7oPrmB3obMscg0o9Z=uawkM2PoxWxT@mail.gmail.com>
- In-Reply-To
- <vpqbp78x418.fsf@bauges.imag.fr>
On Tue, Oct 5, 2010 at 6:38 PM, Matthieu Moy <Matthieu.Moy@grenoble-inp.fr> wrote:
Show 24 quoted lines
> Alex Riesen <raa.lkml@gmail.com> writes: > >> On Tue, Oct 5, 2010 at 10:17, Enrico Weigelt <weigelt@metux.de> wrote: >>> * Matthieu Moy <Matthieu.Moy@grenoble-inp.fr> wrote: >>> >>>> git commit will show what's being commited (the output of "git commit >>>> --dry-run") in your editor, hence it needs to compute that. >>> >>> hmm, is there any way to get around this ? >>> >> >> Try "git commit -q -uno". This should skip creation of summary in the >> commit message and lookup for untracked files. > > To avoid including the summary, the option would be --no-status (-q > makes commit less verbose in stdout, not in COMMIT_EDITMSG). > > But > > strace -fe lstat64 git commit -uno --no-status -q > > still shows lstat64 for each tracked file in my working tree (even > when using -m to avoid launching the editor). I don't know if this is > intended, or just that nobody cared enough to optimize this.
I assume you do git-commit with no index modification at all. The index refresh part dated back in 2888605 (builtin-commit: fix partial-commit support - 2007-11-18). Commit message does not tell why index refresh is needed (for summary stat maybe?). If so, then we can skip refreshing if -q is given.
-- Duy