git.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Dmitry Potapov <dpotapov@gmail.com>
To: Alex Riesen <raa.lkml@gmail.com>
Cc: git@vger.kernel.org, Johannes Sixt <johannes.sixt@telecom.at>,
	Junio C Hamano <gitster@pobox.com>,
	Steffen Prohaska <prohaska@zib.de>
Subject: Re: [PATCH] add GIT_FAST_STAT mode for Cygwin
Date: Tue, 23 Sep 2008 20:52:47 +0400	[thread overview]
Message-ID: <20080923165247.GO21650@dpotapov.dyndns.org> (raw)
In-Reply-To: <81b0412b0809230737s7498e214w4c58991e79f76507@mail.gmail.com>

On Tue, Sep 23, 2008 at 04:37:14PM +0200, Alex Riesen wrote:
> 2008/9/23 Dmitry Potapov <dpotapov@gmail.com>:
> >
> > This fast mode works only for relative paths. It is assumed that the
> > whole repository is located inside one directory without using Cygwin
> > mount to bind external paths inside of the current tree.
> 
> Why runtime conditional? Why conditional at all?

I thought that in rather unusual circumstances (such as using Cygwin
mount to connect separately directories in one tree), this fast version
may not work. So, I made it conditional. It is runtime conditional,
because most users do not build Git themselves but install a ready
Cygwin package.

> Why not fallback
> to cygwin's slow stat on absolute pathnames like you do for symlinks?

Of course, I do:

+       if (file_name[0] == '/')
+               return cygstat (file_name, buf);

Sorry, if it was not clear from my above comment.

> 
> > +/*
> > + * This are startup stubs, which choose what implementation of lstat/stat
> 
> why do you need two of them? Isn't one not enough?

I did not want to give people reasons to say that I broke lstat :)
You can opt for the standard Cygwin version of it if for some reason,
this new function does not work. Now, I know only one case -- it is
when you use Cygwin mount inside of Git repo. Yet, I don't know enough
about Cygwin to be sure that there is no other cases. So, I just wanted
to be extra careful and not to break anything.

> 
> > +stat_fn_t cygwin_stat_fn = cygwin_stat_choice;
> > +stat_fn_t cygwin_lstat_fn = cygwin_lstat_choice;
> ...
> > +typedef int (*stat_fn_t)(const char*, struct stat*);
> > +extern stat_fn_t cygwin_stat_fn;
> > +extern stat_fn_t cygwin_lstat_fn;
> 
>     extern int (*cygwin_stat_fn)(const char *, struct stat *);
> 
> Is shorter, easier to read and easier to understand (for a C person).
> You don't even use the type anywhere else, it is just for the declaration sake!

I use it in description of a parameter of another function:

static int do_stat(const char *file_name, struct stat *buf, stat_fn_t cygstat)

Of course, you can avoid it here too, but the declaration will become
somewhat longer:

static int do_stat(const char *file_name, struct stat *buf,
       int (*cygstat)(const char *, struct stat *));

so I am not sure that removing stat_fn_t improves readability, but if
there are other people who think so, I will correct that.


Dmitry

  reply	other threads:[~2008-09-23 16:54 UTC|newest]

Thread overview: 25+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-09-23 14:06 [PATCH] add GIT_FAST_STAT mode for Cygwin Dmitry Potapov
2008-09-23 14:37 ` Alex Riesen
2008-09-23 16:52   ` Dmitry Potapov [this message]
2008-09-23 17:51     ` Jakub Narebski
2008-09-24 11:25     ` Alex Riesen
2008-09-24 14:03       ` Dmitry Potapov
2008-09-24 14:42         ` Alex Riesen
2008-09-24 15:02           ` Shawn O. Pearce
2008-09-24 15:09             ` Alex Riesen
2008-09-24 15:16               ` Shawn O. Pearce
2008-09-24 15:32                 ` Alex Riesen
2008-09-23 15:31 ` Shawn O. Pearce
2008-09-23 17:12   ` Dmitry Potapov
2008-09-23 19:06     ` Shawn O. Pearce
2008-09-23 20:04       ` Dmitry Potapov
2008-09-23 20:17         ` Shawn O. Pearce
2008-09-23 21:28           ` Dmitry Potapov
2008-09-23 21:58             ` Shawn O. Pearce
2008-09-23 19:03 ` Johannes Sixt
2008-09-23 19:48   ` Dmitry Potapov
2008-09-23 20:41     ` Johannes Sixt
2008-09-23 21:11       ` Dmitry Potapov
2008-09-27  7:02 ` [PATCH v2] Add a "fast stat" " Marcus Griep
2008-09-27  8:35   ` Alex Riesen
2008-09-27 10:39   ` Dmitry Potapov

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20080923165247.GO21650@dpotapov.dyndns.org \
    --to=dpotapov@gmail.com \
    --cc=git@vger.kernel.org \
    --cc=gitster@pobox.com \
    --cc=johannes.sixt@telecom.at \
    --cc=prohaska@zib.de \
    --cc=raa.lkml@gmail.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).