First cut at git port to Cygwin

First cut at git port to Cygwin

[H. Peter Anvin]

I have made a first cut at a git port to Cygwin.  It looks like the 
"git-diff-tree -p" problem has been resolved independently, or at least 
I can't reproduce it on a fresh Cygwin install (running on XP Home), but 
I have added support for running without the IPv6 and the getaddrinfo() API.

There are still funnies.  In particular, Cygwin and Samba handle 
symlinks differently, so you can't trivially share a repository via 
Samba.  Linus' "symbolic refs" changes should eventually take care of that.

Another funny which I haven't been able to figure out yet is that 'gitk' 
scrunches all its output up into a few pixels at the top of the window. 
  If I maximize the window, I can manually resize most of the panes and 
the output looks correct, but the highlighted text in the top panes show 
up in black on a really really dark blue background and is thus illegible.

I have set up a git-on-Cygwin temporary tree at:

http://www.kernel.org/pub/scm/git/git-cygwin.git

	-hpa

Re: First cut at git port to Cygwin

[Junio C Hamano]

"H. Peter Anvin" <hpa@zytor.com> writes:

> There are still funnies.  In particular, Cygwin and Samba handle 
> symlinks differently, so you can't trivially share a repository via 
> Samba.  Linus' "symbolic refs" changes should eventually take care of that.

I just sent out "The other side of Linus' symbolic refs" patch,
saying that Cygwin capable of doing symlink would probably made
it irrelevant.  But it may not be a waste after all, considering
what you said above.

Re: First cut at git port to Cygwin

[Martin Langhoff]

On 9/29/05, H. Peter Anvin <hpa@zytor.com> wrote:
> Another funny which I haven't been able to figure out yet is that 'gitk'
> scrunches all its output up into a few pixels at the top of the window.
>   If I maximize the window, I can manually resize most of the panes and
> the output looks correct

This is visible on OSX too, and someone's mentioned it's a Tk oddity
with rootless X. Does it do the same if you run X with a root window?

> I have set up a git-on-Cygwin temporary tree at:
>
> http://www.kernel.org/pub/scm/git/git-cygwin.git

Getting a 404 on that. Doesn't show up on gitweb either. I guess I
have to wait...

Is there a way to get gitweb to "compare" branches using git-cherry? I
often look at branches via git-web and it's impossible to tell what
makes them unique...

cheers,


martin

Re: First cut at git port to Cygwin

[H. Peter Anvin]

Junio C Hamano wrote:
> "H. Peter Anvin" <hpa@zytor.com> writes:
> 
> 
>>There are still funnies.  In particular, Cygwin and Samba handle 
>>symlinks differently, so you can't trivially share a repository via 
>>Samba.  Linus' "symbolic refs" changes should eventually take care of that.
> 
> 
> I just sent out "The other side of Linus' symbolic refs" patch,
> saying that Cygwin capable of doing symlink would probably made
> it irrelevant.  But it may not be a waste after all, considering
> what you said above.
> 

After looking at it some more, what Samba does when talking to a host 
that doesn't support Unix extensions is that it simply resolves the 
symlink, in effect turning it into a hard link.  That might be all git 
needs.  The reverse still doesn't work, though.

	-hpa

Re: First cut at git port to Cygwin

[Junio C Hamano]

Martin Langhoff <martin.langhoff@gmail.com> writes:

> Is there a way to get gitweb to "compare" branches using git-cherry? I
> often look at branches via git-web and it's impossible to tell what
> makes them unique...

Now that you mention it, I felt that too.  Maybe git-show-branch
output could help somehow?

Re: First cut at git port to Cygwin

[H. Peter Anvin]

Martin Langhoff wrote:
> 
>>I have set up a git-on-Cygwin temporary tree at:
>>
>>http://www.kernel.org/pub/scm/git/git-cygwin.git
> 
> Getting a 404 on that. Doesn't show up on gitweb either. I guess I
> have to wait...
> 

Well, it's up there now.

	-hpa

Re: First cut at git port to Cygwin

[Johannes Schindelin]

Hi,

On Wed, 28 Sep 2005, H. Peter Anvin wrote:

> Another funny which I haven't been able to figure out yet is that 'gitk'
> scrunches all its output up into a few pixels at the top of the window.

See my mail about rootless X11. I went about working around that 
particular Tk bug by specifying the dimensions of the panes explicitely. 
However, I was not especially happy with my workaround, since it did not 
reproduce the layout exactly after a restart. Maybe you can figure it out 
how to do that.

Ciao,
Dscho

Re: First cut at git port to Cygwin

[H. Peter Anvin]

Johannes Schindelin wrote:
> Hi,
> 
> On Wed, 28 Sep 2005, H. Peter Anvin wrote:
> 
>>Another funny which I haven't been able to figure out yet is that 'gitk'
>>scrunches all its output up into a few pixels at the top of the window.
> 
> See my mail about rootless X11. I went about working around that 
> particular Tk bug by specifying the dimensions of the panes explicitely. 
> However, I was not especially happy with my workaround, since it did not 
> reproduce the layout exactly after a restart. Maybe you can figure it out 
> how to do that.
> 

Unlikely, since I'm a complete Tcl/Tk illiterate.

	-hpa

Re: First cut at git port to Cygwin

[H. Peter Anvin]

Johannes Schindelin wrote:
> Hi,
> 
> On Wed, 28 Sep 2005, H. Peter Anvin wrote:
> 
>>Another funny which I haven't been able to figure out yet is that 'gitk'
>>scrunches all its output up into a few pixels at the top of the window.
> 
> See my mail about rootless X11. I went about working around that 
> particular Tk bug by specifying the dimensions of the panes explicitely. 
> However, I was not especially happy with my workaround, since it did not 
> reproduce the layout exactly after a restart. Maybe you can figure it out 
> how to do that.
> 

It looks like this isn't a rootless *X* thing; it looks like the wish 
that is included with Cygwin actually opens native Win32 windows; even 
when run from inside a rooted X session it still opens an external 
window.  I also tried using the wish from the latest ActiveState 
distribution; it exhibits the same problem although with slightly 
different geometries.

	-hpa

Re: First cut at git port to Cygwin

[Junio C Hamano]

"H. Peter Anvin" <hpa@zytor.com> writes:

> I have set up a git-on-Cygwin temporary tree at:
>
> http://www.kernel.org/pub/scm/git/git-cygwin.git

: siamese; git clone http://kernel.org/pub/scm/git/git-cygwin.git/ git-cygwin
defaulting to local storage area
Cannot get remote repository information.
Perhaps git-update-server-info needs to be run there?

Could you do update-server-info there, please?

hera$ cd /pub/scm/git/git-cygwin.git
hera$ GIT_DIR=. git-update-server-info



Knowing nothing about Cygwin environment, here are some
comments.

        +# Define NO_IPV6 if you lack IPv6 support and getaddrinfo().

This part probably is applicable outside Cygwin.  At some point,
can we have it in the mainline please?

         # The ones that do not have to link with lcrypto nor lz.
         SIMPLE_PROGRAMS = \
        -	git-get-tar-commit-id git-mailinfo git-mailsplit git-stripspace \
        -	git-daemon git-var
        +	git-get-tar-commit-id$(X) git-mailinfo$(X) git-mailsplit$(X) \
        +	git-stripspace$(X) git-var$(X) git-daemon$(X)
 
I have seen these $(X) in other programs' ports and found them
quite distasteful.  Since I do not have immediate suggestions
for improvements, I do not have rights to complain, though.

Spelling it $X is a bit less distracting but not that much
better.  Maybe "SIMPLE_PROGRAM_NAMES = git-foo git-bar" and
"SIMPLE_PROGRAMS = $(patsubst %,%$X,$(SIMPLE_PROGRAM_NAMES))"...
but that would not help bits like this:

        -	PROGRAMS += git-http-fetch
        +	PROGRAMS += git-http-fetch$(X)

or this: 

        -git-%: %.o $(LIB_FILE)
        +git-%$(X): %.o $(LIB_FILE)

... so I'd shut up about this part.

        diff --git a/daemon.c b/daemon.c
        --- a/daemon.c
        +++ b/daemon.c
        @@ -1,9 +1,11 @@
         #include "cache.h"
         #include "pkt-line.h"
        +#include <alloca.h>

Why?  I do not see any use of alloca in the added code...

        +#include <sys/poll.h>

Is poll preferrable over select in general?  Some may have only
select available and others may have only poll available,
perhaps?  In any case, this is probably relevant to wider
audience than just Cygwin; please give it to mainline at some
point, perhaps conditionally allowing either/both.

        +	*socklist_p = malloc(sizeof(int));
        +	pfd = calloc(socknum, sizeof(struct pollfd));

Please use xmalloc and xcalloc just for consistency.

                test -x $path/git-$cmd && exec $path/git-$cmd "$@" ;;
        +
        +	# In case we're running on Cygwin...
        +	test -x $path/git-$cmd.exe && exec $path/git-$cmd.exe "$@" ;;
         esac
 
Hmph, I think you forgot to drop double semicolon there.

The git.sh script is munged by Makefile so presumably we could
fix this part up there, like:

        git: git.sh Makefile
                rm -f $@+ $@
                sed -e '1s|#!.*/sh|#!$(SHELL_PATH)|' \
                    -e 's/@@GIT_VERSION@@/$(GIT_VERSION)/g' \
                    -e 's/@@X@@/$X/g' <$@.sh >$@+
                chmod +x $@+
                mv $@+ $@

And then (a patch on top of your "master"):

diff --git a/git.sh b/git.sh
--- a/git.sh
+++ b/git.sh
@@ -12,10 +12,14 @@ case "$#" in
 		exit 0 ;;
 	esac
 
-	test -x $path/git-$cmd && exec $path/git-$cmd "$@" ;;
+	test -x $path/git-$cmd && exec $path/git-$cmd "$@"
 
-	# In case we're running on Cygwin...
-	test -x $path/git-$cmd.exe && exec $path/git-$cmd.exe "$@" ;;
+	case '@@X@@' in
+	'')
+		;;
+	*)
+		test -x $path/git-$cmd@@X@@ && exec $path/git-$cmd@@X@@ "$@" ;;
+	esac		
 esac
 
 echo "Usage: git COMMAND [OPTIONS] [TARGET]"

Re: First cut at git port to Cygwin

[H. Peter Anvin]

Junio C Hamano wrote:
> 
> Could you do update-server-info there, please?
> 

Done...

> 
> Knowing nothing about Cygwin environment, here are some
> comments.
> 
>         +# Define NO_IPV6 if you lack IPv6 support and getaddrinfo().
> 
> This part probably is applicable outside Cygwin.  At some point,
> can we have it in the mainline please?
> 

Well, I would hope that all the changes could eventually be merged.

>          # The ones that do not have to link with lcrypto nor lz.
>          SIMPLE_PROGRAMS = \
>         -	git-get-tar-commit-id git-mailinfo git-mailsplit git-stripspace \
>         -	git-daemon git-var
>         +	git-get-tar-commit-id$(X) git-mailinfo$(X) git-mailsplit$(X) \
>         +	git-stripspace$(X) git-var$(X) git-daemon$(X)
>  
> I have seen these $(X) in other programs' ports and found them
> quite distasteful.  Since I not have immediate suggestions
> for improvements, I do not have rights to complain, though.
> 
> Spelling it $X is a bit less distracting but not that much
> better.  Maybe "SIMPLE_PROGRAM_NAMES = git-foo git-bar" and
> "SIMPLE_PROGRAMS = $(patsubst %,%$X,$(SIMPLE_PROGRAM_NAMES))"...
> but that would not help bits like this:
> 
>         -	PROGRAMS += git-http-fetch
>         +	PROGRAMS += git-http-fetch$(X)
> 
> or this: 
> 
>         -git-%: %.o $(LIB_FILE)
>         +git-%$(X): %.o $(LIB_FILE)
> 
> ... so I'd shut up about this part.

My first cut had PROGRAMS_X and SIMPLE_PROGRAMS_X being patsubst of the 
original versions, but in the end I decided it was even uglier, because 
these patterns were needed elsewhere.  I'll change them to $X except 
where the parens are needed.

>         diff --git a/daemon.c b/daemon.c
>         --- a/daemon.c
>         +++ b/daemon.c
>         @@ -1,9 +1,11 @@
>          #include "cache.h"
>          #include "pkt-line.h"
>         +#include <alloca.h>
> 
> Why?  I do not see any use of alloca in the added code...

I originally used alloca() before changing my mind and using calloc(); I 
think there might be platforms without alloca out there.

>         +#include <sys/poll.h>
> 
> Is poll preferrable over select in general?  Some may have only
> select available and others may have only poll available,
> perhaps?  In any case, this is probably relevant to wider
> audience than just Cygwin; please give it to mainline at some
> point, perhaps conditionally allowing either/both.

The main reason I switched to poll() is that I believe all platforms 
that are even remotely relevant have both these days, and forming a poll 
list is so much cleaner than forming a select set.  What makes forming a 
select set even remotely bearable is the invalid assumption that the 
number of file descriptors is bounded at compile time and therefore that 
fdset_t can be statically allocated.  We've had problems in the past 
with that assumption on Linux, and I've tried to avoid select since then.

>         +	*socklist_p = malloc(sizeof(int));
>         +	pfd = calloc(socknum, sizeof(struct pollfd));
> 
> Please use xmalloc and xcalloc just for consistency.

Check.

>                 test -x $path/git-$cmd && exec $path/git-$cmd "$@" ;;
>         +
>         +	# In case we're running on Cygwin...
>         +	test -x $path/git-$cmd.exe && exec $path/git-$cmd.exe "$@" ;;
>          esac
>  
> Hmph, I think you forgot to drop double semicolon there.

D'oh!

> The git.sh script is munged by Makefile so presumably we could
> fix this part up there, like:
> 
>         git: git.sh Makefile
>                 rm -f $@+ $@
>                 sed -e '1s|#!.*/sh|#!$(SHELL_PATH)|' \
>                     -e 's/@@GIT_VERSION@@/$(GIT_VERSION)/g' \
>                     -e 's/@@X@@/$X/g' <$@.sh >$@+
>                 chmod +x $@+
>                 mv $@+ $@
> 
> And then (a patch on top of your "master"):
> 
> diff --git a/git.sh b/git.sh
> --- a/git.sh
> +++ b/git.sh
> @@ -12,10 +12,14 @@ case "$#" in
>  		exit 0 ;;
>  	esac
>  
> -	test -x $path/git-$cmd && exec $path/git-$cmd "$@" ;;
> +	test -x $path/git-$cmd && exec $path/git-$cmd "$@"
>  
> -	# In case we're running on Cygwin...
> -	test -x $path/git-$cmd.exe && exec $path/git-$cmd.exe "$@" ;;
> +	case '@@X@@' in
> +	'')
> +		;;
> +	*)
> +		test -x $path/git-$cmd@@X@@ && exec $path/git-$cmd@@X@@ "$@" ;;
> +	esac		
>  esac
>  
>  echo "Usage: git COMMAND [OPTIONS] [TARGET]"

That wouldn't work, because the shell scripts don't get the .exe 
extension.  However, I can figure out something equivalent.

	-hpa

Re: First cut at git port to Cygwin

[H. Peter Anvin]

Okay, I have updated the git-cygwin repository with the changes 
proposed, and they seem to work.

	-hpa

Re: First cut at git port to Cygwin

[Alex Riesen]

On 9/29/05, H. Peter Anvin <hpa@zytor.com> wrote:
> I have made a first cut at a git port to Cygwin.  It looks like the
> "git-diff-tree -p" problem has been resolved independently, or at least
> I can't reproduce it on a fresh Cygwin install (running on XP Home), but
> I have added support for running without the IPv6 and the getaddrinfo() API.
>
> There are still funnies.  In particular, Cygwin and Samba handle
> symlinks differently, so you can't trivially share a repository via
> Samba.  Linus' "symbolic refs" changes should eventually take care of that.

I noticed that rename(2) in my copy of cygwin (1.5.18-1) does not remove the
target and returns an error (probably EPERM, but I have reasons not to trust
strerror on that thing).
The repository was on FAT.
Taking "rename(2)" from cygwin's libiberty solved this (they unlink if link(2)
returns EEXIST).

PS: Does broken rename(2) qualify a system "not worthy to support"?

Re: First cut at git port to Cygwin

[Alex Riesen]

On 10/4/05, Alex Riesen <raa.lkml@gmail.com> wrote:
> On 9/29/05, H. Peter Anvin <hpa@zytor.com> wrote:
> > I have made a first cut at a git port to Cygwin.  It looks like the
> > "git-diff-tree -p" problem has been resolved independently, or at least
> > I can't reproduce it on a fresh Cygwin install (running on XP Home), but
> > I have added support for running without the IPv6 and the getaddrinfo() API.
> >
> > There are still funnies.  In particular, Cygwin and Samba handle
> > symlinks differently, so you can't trivially share a repository via
> > Samba.  Linus' "symbolic refs" changes should eventually take care of that.
>
> I noticed that rename(2) in my copy of cygwin (1.5.18-1) does not remove the
> target and returns an error (probably EPERM, but I have reasons not to trust
> strerror on that thing).
> The repository was on FAT.
> Taking "rename(2)" from cygwin's libiberty solved this (they unlink if link(2)
> returns EEXIST).

I think I have to clarify: I copied the function (like in strcasestr
case) into compat/

Re: First cut at git port to Cygwin

[H. Peter Anvin]

Alex Riesen wrote:
> 
> I noticed that rename(2) in my copy of cygwin (1.5.18-1) does not remove the
> target and returns an error (probably EPERM, but I have reasons not to trust
> strerror on that thing).
> The repository was on FAT.
> Taking "rename(2)" from cygwin's libiberty solved this (they unlink if link(2)
> returns EEXIST).
> 
> PS: Does broken rename(2) qualify a system "not worthy to support"?

In this case a better way would be to just add -liberty to all link 
lines if necessary, but I would expect the core cygwin code to do this.

	-hpa

Re: First cut at git port to Cygwin

[H. Peter Anvin]

Alex Riesen wrote:
> 
> I noticed that rename(2) in my copy of cygwin (1.5.18-1) does not remove the
> target and returns an error (probably EPERM, but I have reasons not to trust
> strerror on that thing).
> The repository was on FAT.
> Taking "rename(2)" from cygwin's libiberty solved this (they unlink if link(2)
> returns EEXIST).
> 
> PS: Does broken rename(2) qualify a system "not worthy to support"?

I just tried this with Cygwin 1.5.18-1 and didn't have any such 
problems.  I tried it on NTFS, FAT and Samba, using WinXP.

	-hpa

Re: First cut at git port to Cygwin

[Christopher Faylor]

On Tue, Oct 04, 2005 at 07:06:29AM -0700, H. Peter Anvin wrote:
>Alex Riesen wrote:
>>
>>I noticed that rename(2) in my copy of cygwin (1.5.18-1) does not remove 
>>the
>>target and returns an error (probably EPERM, but I have reasons not to 
>>trust
>>strerror on that thing).
>>The repository was on FAT.
>>Taking "rename(2)" from cygwin's libiberty solved this (they unlink if 
>>link(2)
>>returns EEXIST).
>>
>>PS: Does broken rename(2) qualify a system "not worthy to support"?
>
>In this case a better way would be to just add -liberty to all link 
>lines if necessary, but I would expect the core cygwin code to do this.

AFAIK, cygwin has a working rename().  Many packages rely on it.

If rename() is not working then a bug report with a test case would be
appreciated.
--
Christopher Faylor			spammer? ->	aaaspam@sourceware.org
Cygwin Co-Project Leader				aaaspam@duffek.com
TimeSys, Inc.

Re: First cut at git port to Cygwin

[Christopher Faylor]

On Tue, Oct 04, 2005 at 08:03:55AM -0700, H. Peter Anvin wrote:
>Alex Riesen wrote:
>>
>>I noticed that rename(2) in my copy of cygwin (1.5.18-1) does not remove 
>>the
>>target and returns an error (probably EPERM, but I have reasons not to 
>>trust
>>strerror on that thing).
>>The repository was on FAT.
>>Taking "rename(2)" from cygwin's libiberty solved this (they unlink if 
>>link(2)
>>returns EEXIST).
>>
>>PS: Does broken rename(2) qualify a system "not worthy to support"?
>
>I just tried this with Cygwin 1.5.18-1 and didn't have any such 
>problems.  I tried it on NTFS, FAT and Samba, using WinXP.

That's a relief.  Btw, AFAIK, strerror is working correctly under
Cygwin also.
--
Christopher Faylor			spammer? ->	aaaspam@sourceware.org
Cygwin Co-Project Leader				aaaspam@duffek.com
TimeSys, Inc.

Re: First cut at git port to Cygwin

[H. Peter Anvin]

Christopher Faylor wrote:
> That's a relief.  Btw, AFAIK, strerror is working correctly under
> Cygwin also.

Now if we can only figure out why gitk is messed up...

	-hpa

Re: First cut at git port to Cygwin

[Alex Riesen]

On 10/4/05, H. Peter Anvin <hpa@zytor.com> wrote:
> > I noticed that rename(2) in my copy of cygwin (1.5.18-1) does not remove the
> > target and returns an error (probably EPERM, but I have reasons not to trust
> > strerror on that thing).
> > The repository was on FAT.
> > Taking "rename(2)" from cygwin's libiberty solved this (they unlink if link(2)
> > returns EEXIST).
> >
> > PS: Does broken rename(2) qualify a system "not worthy to support"?
>
> I just tried this with Cygwin 1.5.18-1 and didn't have any such
> problems.  I tried it on NTFS, FAT and Samba, using WinXP.

It's on Win2k, there was multiple cygwin installations in path, the other one
supposedly is 1.5.5 (it's from QNX Momentics installation).
I had that old "cygwin1.dll" renamed into "cygwin1.dll-disabled" long
ago, though...
I can't reproduce this out of GIT context, and the error is not
reproducable after
I removed the other cygwin installation out of PATH.
Anyway, sorry, I should have tried this before posting.

Re: First cut at git port to Cygwin

[Jonas Fonseca]

I have a few things I experienced with the merged cygwin stuff. Sorry I
haven't investigated it further, but there should be enough for a few
fixes.

When I ...

  user@machine /usr/local/dev/git/git
  $ make prefix=/usr/local install
  install -d -m755 /usr/local/bin
  install git-apply.exe [...]
  sh ./cmd-rename.sh /usr/local/bin
  ln: creating symbolic link `/usr/local/bin/git-http-pull.exe' to `git-http-fetch.exe': File exists
  make: *** [install] Error 1

Can be fixed by the patch below. I don't know if it would be cleaner to
pass cmd-rename.sh "$X" as a second argument from the Makefile.

--- cmd-rename.sh	2005-10-05 14:42:00.000000000 +0200
+++ cmd-rename.sh-orig	2005-10-05 14:43:48.000000000 +0200
@@ -3,7 +3,7 @@
 test -d "$d" || exit
 while read old new
 do
-	rm -f "$d/$old" "$d/$old.exe" 
+	rm -f "$d/$old"
 	if test -f "$d/$new"
 	then
 		ln -s "$new" "$d/$old" || exit


Some other obscurities ...

  user@machine /usr/local/dev/git/git
  $ git-log
  fatal: Not a git repository

  user@machine /usr/local/dev/git/git
  $ GIT_DIR=.git git-log | wc -l
  26094

and I cannot rebuild the index file with git-reset. First time I run it
it creates the index.lock file and errors out when writing. The second
time it errors out because the lock file was not removed in the first
case.

  user@machine /usr/local/dev/git/git
  $ GIT_DIR=.git git-reset
  fatal: unable to write new index file
  
  user@machine /usr/local/dev/git/git
  $ GIT_DIR=.git git-reset
  fatal: unable to create new cachefile
  
  user@machine /usr/local/dev/git/git
  $ uname -a
  CYGWIN_NT-5.1 antimatter 1.5.18(0.132/4/2) 2005-07-02 20:30 i686 unknown unknown Cygwin

-- 
Jonas Fonseca

Re: First cut at git port to Cygwin

[Johannes Schindelin]

[-- Attachment #1: Type: TEXT/PLAIN, Size: 468 bytes --]

Hi,

On Wed, 5 Oct 2005, Jonas Fonseca wrote:

>   user@machine /usr/local/dev/git/git
>   $ git-log
>   fatal: Not a git repository
> 
>   user@machine /usr/local/dev/git/git
>   $ GIT_DIR=.git git-log | wc -l
>   26094

That could have its cause in your .git/HEAD being no symlink. That happens 
when rsync´ing the .git directory.

The other errors could also stem from the fact that quite a few places 
expect HEAD to be a symlink.

Ciao,
Dscho

Re: First cut at git port to Cygwin

[Alex Riesen]

[-- Attachment #1: Type: text/plain, Size: 1392 bytes --]

On 10/5/05, Alex Riesen <raa.lkml@gmail.com> wrote:
> On 10/4/05, H. Peter Anvin <hpa@zytor.com> wrote:
> > > I noticed that rename(2) in my copy of cygwin (1.5.18-1) does not remove the
> > > target and returns an error (probably EPERM, but I have reasons not to trust
> > > strerror on that thing).
> > > The repository was on FAT.
> > > Taking "rename(2)" from cygwin's libiberty solved this (they unlink if link(2)
> > > returns EEXIST).
> > >
> > > PS: Does broken rename(2) qualify a system "not worthy to support"?
> >
> > I just tried this with Cygwin 1.5.18-1 and didn't have any such
> > problems.  I tried it on NTFS, FAT and Samba, using WinXP.
>
> It's on Win2k, there was multiple cygwin installations in path, the other one
> supposedly is 1.5.5 (it's from QNX Momentics installation).
> I had that old "cygwin1.dll" renamed into "cygwin1.dll-disabled" long
> ago, though...
> I can't reproduce this out of GIT context, and the error is not
> reproducable after
> I removed the other cygwin installation out of PATH.
> Anyway, sorry, I should have tried this before posting.

Still does not work for me. I cannot isolate the problem out of git,
but at the moment the only way for me to make commit_index_file to work
is to put unlink(indexfile) before rename(cf->lockfile, indexfile).

For everyone interested, I attach cygwin's strace output here.

[-- Attachment #2: log --]
[-- Type: application/octet-stream, Size: 65558 bytes --]

**********************************************
Program name: d:\download\git-mine\git-update-index.exe (pid 1476, ppid 1)
App version:  1005.18, api: 0.132
DLL version:  1005.18, api: 0.132
DLL build:    2005-07-02 20:30
OS version:   Windows NT-5.0
Heap size:    536870912
Date/Time:    2005-10-05 17:27:37
**********************************************
   55     474 [main] git-update-index 1476 set_myself: myself->dwProcessId 1476
   57     531 [main] git-update-index 1476 time: 1128526057 = time (0)
  404     935 [main] git-update-index 1476 environ_init: 0x10020238: !F:=F:\
  100    1035 [main] git-update-index 1476 environ_init: 0x10020248: !H:=H:\
  100    1135 [main] git-update-index 1476 environ_init: 0x10020258: !I:=I:\
  100    1235 [main] git-update-index 1476 environ_init: 0x10020268: !J:=J:\
   99    1334 [main] git-update-index 1476 environ_init: 0x10020278: !N:=N:\
   99    1433 [main] git-update-index 1476 environ_init: 0x10020288: !X:=X:\
  100    1533 [main] git-update-index 1476 environ_init: 0x10020298: !Y:=Y:\
   99    1632 [main] git-update-index 1476 environ_init: 0x100202A8: !Z:=Z:\
  100    1732 [main] git-update-index 1476 environ_init: 0x100202B8: ALLUSERSPROFILE=C:\Documents and Settings\All Users
  100    1832 [main] git-update-index 1476 environ_init: 0x100202F0: APPDATA=C:\Documents and Settings\ARiesen\Application Data
  100    1932 [main] git-update-index 1476 environ_init: 0x10020330: APR_ICONV_PATH=d:\Programs\Subversion\iconv
  100    2032 [main] git-update-index 1476 environ_init: 0x10020360: CLASSPATH="C:\WINNT\system32\QTJava.zip"
  101    2133 [main] git-update-index 1476 environ_init: 0x10020390: COMMONPROGRAMFILES=C:\Program Files\Common Files
  100    2233 [main] git-update-index 1476 environ_init: 0x100203C8: COMPUTERNAME=OEKAP852
  100    2333 [main] git-update-index 1476 environ_init: 0x100203E8: COMSPEC=C:\WINNT\system32\cmd.exe
   99    2432 [main] git-update-index 1476 environ_init: 0x10020410: CVS_RSH=/bin/ssh
  100    2532 [main] git-update-index 1476 environ_init: 0x10020428: CYGWIN_BASH_HOSTNAME=oekap852
  100    2632 [main] git-update-index 1476 environ_init: 0x10020450: EDITOR=gvim -f
  100    2732 [main] git-update-index 1476 environ_init: 0x10020468: EXSERVER=OEKAEX01
   99    2831 [main] git-update-index 1476 environ_init: 0x10020480: GREP_OPTIONS=--exclude=*.svn-base
  104    2935 [main] git-update-index 1476 environ_init: 0x100204A8: HB_TOOLS_DIR=D:/p4scm/SCM/Release/BuildProcess/tools
  101    3036 [main] git-update-index 1476 environ_init: 0x100204E8: HGEDITOR=ee
  100    3136 [main] git-update-index 1476 environ_init: 0x100204F8: HGMERGE=ee -d
  100    3236 [main] git-update-index 1476 environ_init: 0x10020510: HGUSER=ariesen@harmanbecker.com
  100    3336 [main] git-update-index 1476 environ_init: 0x10020538: HISTFILESIZE=20000
  100    3436 [main] git-update-index 1476 environ_init: 0x10020550: HISTSIZE=1000
  100    3536 [main] git-update-index 1476 getwinenv: can't set native for HOME= since no environ yet
  106    3642 [main] git-update-index 1476 mount_info::conv_to_posix_path: conv_to_posix_path (h:, no-keep-rel, no-add-slash)
   59    3701 [main] git-update-index 1476 normalize_win32_path: h: = normalize_win32_path (h:)
   62    3763 [main] git-update-index 1476 mount_info::conv_to_posix_path: /cygdrive/h = conv_to_posix_path (h:)
  150    3913 [main] git-update-index 1476 win_env::add_cache: posix /cygdrive/h
   51    3964 [main] git-update-index 1476 win_env::add_cache: native HOME=h:
   51    4015 [main] git-update-index 1476 posify: env var converted to HOME=/cygdrive/h
   99    4114 [main] git-update-index 1476 environ_init: 0x10020578: HOME=/cygdrive/h
  139    4253 [main] git-update-index 1476 environ_init: 0x100206A8: HOMEDRIVE=C:
  101    4354 [main] git-update-index 1476 environ_init: 0x10020568: HOMEPATH=\
  101    4455 [main] git-update-index 1476 environ_init: 0x100206C0: HOSTNAME=oekap852
  100    4555 [main] git-update-index 1476 environ_init: 0x100206D8: INCLUDE=C:\Program Files\Microsoft Visual Studio\VC98\atl\include;C:\Program Files\Microsoft Visual Studio\VC98\mfc\include;C:\Program Files\Microsoft Visual Studio\VC98\include
  107    4662 [main] git-update-index 1476 environ_init: 0x10020790: INFOPATH=/usr/local/info:/usr/info:/usr/share/info:/usr/autotool/devel/info:/usr/autotool/stable/info:
  102    4764 [main] git-update-index 1476 environ_init: 0x10020800: LANG=en_US
  100    4864 [main] git-update-index 1476 environ_init: 0x10020810: LIB=C:\Program Files\Microsoft Visual Studio\VC98\mfc\lib;C:\Program Files\Microsoft Visual Studio\VC98\lib
  101    4965 [main] git-update-index 1476 environ_init: 0x10020880: LM_LICENSE_FILE=7192@oekaw2do01
  101    5066 [main] git-update-index 1476 environ_init: 0x100208A8: LOGIN_CONTEXT=PDE-SW.Karlsbad
  100    5166 [main] git-update-index 1476 environ_init: 0x100208D0: LOGNAME=%USERNAME.PDE-SW.Karlsbad
  100    5266 [main] git-update-index 1476 environ_init: 0x100208F8: LOGONSERVER=\\OEKAW2DC03
  100    5366 [main] git-update-index 1476 environ_init: 0x10020918: MAKEFLAGS=-Id:/QNX630/target/qnx6/usr/include
  167    5533 [main] git-update-index 1476 environ_init: 0x10020950: MAKE_MODE=unix
  101    5634 [main] git-update-index 1476 environ_init: 0x10020968: MANPATH=/usr/local/man:/usr/man:/usr/share/man:/usr/autotool/devel/man:
  128    5762 [main] git-update-index 1476 environ_init: 0x100209B8: MSDEVDIR=C:\Program Files\Microsoft Visual Studio\Common\MSDev98
  102    5864 [main] git-update-index 1476 environ_init: 0x10020A00: NAME=Riesen, Alexander
  100    5964 [main] git-update-index 1476 environ_init: 0x10020A20: NUMBER_OF_PROCESSORS=1
  100    6064 [main] git-update-index 1476 environ_init: 0x10020A40: NWLANGUAGE=ENGLISH
   99    6163 [main] git-update-index 1476 environ_init: 0x10020A58: NWUSERNAME=ARiesen
  100    6263 [main] git-update-index 1476 environ_init: 0x10020A70: OLDPWD=/d/download/git-cygwin
  100    6363 [main] git-update-index 1476 environ_init: 0x10020A98: OS2LIBPATH=C:\WINNT\system32\os2\dll;
   99    6462 [main] git-update-index 1476 environ_init: 0x10020AC8: OS=WINNT
  100    6562 [main] git-update-index 1476 environ_init: 0x10020AD8: OS_VERSION=V5.00
  100    6662 [main] git-update-index 1476 environ_init: 0x10020AF0: P4_QNX_OS_ROOT=D:/p4qnx/qnx/release/os
  101    6763 [main] git-update-index 1476 getwinenv: can't set native for PATH= since no environ yet
   60    6823 [main] git-update-index 1476 normalize_posix_path: src .
   54    6877 [main] git-update-index 1476 mount_info::conv_to_posix_path: conv_to_posix_path (d:\download\git-mine, no-keep-rel, no-add-slash)
   52    6929 [main] git-update-index 1476 normalize_win32_path: d:\download\git-mine = normalize_win32_path (d:\download\git-mine)
   53    6982 [main] git-update-index 1476 mount_info::conv_to_posix_path: /d/download/git-mine = conv_to_posix_path (d:\download\git-mine)
   52    7034 [main] git-update-index 1476 cwdstuff::get: posix /d/download/git-mine
   51    7085 [main] git-update-index 1476 cwdstuff::get: (/d/download/git-mine) = cwdstuff::get (0x22EBC0, 260, 1, 0), errno 0
   52    7137 [main] git-update-index 1476 normalize_posix_path: /d/download/git-mine/ = normalize_posix_path (.)
   52    7189 [main] git-update-index 1476 mount_info::conv_to_win32_path: conv_to_win32_path (/d/download/git-mine)
   57    7246 [main] git-update-index 1476 set_flags: flags: binary (0x2)
   52    7298 [main] git-update-index 1476 mount_info::conv_to_win32_path: src_path /d/download/git-mine, dst d:\download\git-mine, flags 0xA, rc 0
   98    7396 [main] git-update-index 1476 symlink_info::check: not a symlink
   53    7449 [main] git-update-index 1476 symlink_info::check: 0 = symlink.check (d:\download\git-mine, 0x22E880) (0xA)
  441    7890 [main] git-update-index 1476 set_privilege: 1 = set_privilege ((token 330) SeChangeNotifyPrivilege, 1)
  149    8039 [main] git-update-index 1476 path_conv::check: this->path(d:\download\git-mine), has_acls(0)
  117    8156 [main] git-update-index 1476 mount_info::conv_to_posix_path: conv_to_posix_path (d:\scripts\git, keep-rel, no-add-slash)
   58    8214 [main] git-update-index 1476 normalize_win32_path: d:\scripts\git = normalize_win32_path (d:\scripts\git)
   53    8267 [main] git-update-index 1476 mount_info::conv_to_posix_path: /d/scripts/git = conv_to_posix_path (d:\scripts\git)
   52    8319 [main] git-update-index 1476 mount_info::conv_to_posix_path: conv_to_posix_path (D:\Programs\cygwin\usr\local\bin, keep-rel, no-add-slash)
   52    8371 [main] git-update-index 1476 normalize_win32_path: D:\Programs\cygwin\usr\local\bin = normalize_win32_path (D:\Programs\cygwin\usr\local\bin)
   52    8423 [main] git-update-index 1476 mount_info::conv_to_posix_path: /usr/local/bin = conv_to_posix_path (D:\Programs\cygwin\usr\local\bin)
   53    8476 [main] git-update-index 1476 mount_info::conv_to_posix_path: conv_to_posix_path (D:\Programs\cygwin\bin, keep-rel, no-add-slash)
   51    8527 [main] git-update-index 1476 normalize_win32_path: D:\Programs\cygwin\bin = normalize_win32_path (D:\Programs\cygwin\bin)
   52    8579 [main] git-update-index 1476 mount_info::conv_to_posix_path: /usr/bin = conv_to_posix_path (D:\Programs\cygwin\bin)
   52    8631 [main] git-update-index 1476 mount_info::conv_to_posix_path: conv_to_posix_path (D:\Programs\cygwin\bin, keep-rel, no-add-slash)
   51    8682 [main] git-update-index 1476 normalize_win32_path: D:\Programs\cygwin\bin = normalize_win32_path (D:\Programs\cygwin\bin)
   52    8734 [main] git-update-index 1476 mount_info::conv_to_posix_path: /usr/bin = conv_to_posix_path (D:\Programs\cygwin\bin)
   52    8786 [main] git-update-index 1476 mount_info::conv_to_posix_path: conv_to_posix_path (D:\Programs\cygwin\usr\X11R6\bin, keep-rel, no-add-slash)
   52    8838 [main] git-update-index 1476 normalize_win32_path: D:\Programs\cygwin\usr\X11R6\bin = normalize_win32_path (D:\Programs\cygwin\usr\X11R6\bin)
   52    8890 [main] git-update-index 1476 mount_info::conv_to_posix_path: /usr/X11R6/bin = conv_to_posix_path (D:\Programs\cygwin\usr\X11R6\bin)
   52    8942 [main] git-update-index 1476 mount_info::conv_to_posix_path: conv_to_posix_path (c:\WINNT\system32, keep-rel, no-add-slash)
   51    8993 [main] git-update-index 1476 normalize_win32_path: c:\WINNT\system32 = normalize_win32_path (c:\WINNT\system32)
   52    9045 [main] git-update-index 1476 mount_info::conv_to_posix_path: /c/WINNT/system32 = conv_to_posix_path (c:\WINNT\system32)
   52    9097 [main] git-update-index 1476 mount_info::conv_to_posix_path: conv_to_posix_path (c:\WINNT, keep-rel, no-add-slash)
   51    9148 [main] git-update-index 1476 normalize_win32_path: c:\WINNT = normalize_win32_path (c:\WINNT)
   52    9200 [main] git-update-index 1476 mount_info::conv_to_posix_path: /c/WINNT = conv_to_posix_path (c:\WINNT)
   52    9252 [main] git-update-index 1476 mount_info::conv_to_posix_path: conv_to_posix_path (c:\WINNT\System32\Wbem, keep-rel, no-add-slash)
   51    9303 [main] git-update-index 1476 normalize_win32_path: c:\WINNT\System32\Wbem = normalize_win32_path (c:\WINNT\System32\Wbem)
   52    9355 [main] git-update-index 1476 mount_info::conv_to_posix_path: /c/WINNT/System32/Wbem = conv_to_posix_path (c:\WINNT\System32\Wbem)
   52    9407 [main] git-update-index 1476 mount_info::conv_to_posix_path: conv_to_posix_path (c:\WINNT\system32\nls, keep-rel, no-add-slash)
   52    9459 [main] git-update-index 1476 normalize_win32_path: c:\WINNT\system32\nls = normalize_win32_path (c:\WINNT\system32\nls)
   51    9510 [main] git-update-index 1476 mount_info::conv_to_posix_path: /c/WINNT/system32/nls = conv_to_posix_path (c:\WINNT\system32\nls)
   52    9562 [main] git-update-index 1476 mount_info::conv_to_posix_path: conv_to_posix_path (c:\WINNT\system32\nls\ENGLISH, keep-rel, no-add-slash)
   52    9614 [main] git-update-index 1476 normalize_win32_path: c:\WINNT\system32\nls\ENGLISH = normalize_win32_path (c:\WINNT\system32\nls\ENGLISH)
   52    9666 [main] git-update-index 1476 mount_info::conv_to_posix_path: /c/WINNT/system32/nls/ENGLISH = conv_to_posix_path (c:\WINNT\system32\nls\ENGLISH)
   52    9718 [main] git-update-index 1476 mount_info::conv_to_posix_path: conv_to_posix_path (c:\Program Files\Perforce, keep-rel, no-add-slash)
   55    9773 [main] git-update-index 1476 normalize_win32_path: c:\Program Files\Perforce = normalize_win32_path (c:\Program Files\Perforce)
   51    9824 [main] git-update-index 1476 mount_info::conv_to_posix_path: /c/Program Files/Perforce = conv_to_posix_path (c:\Program Files\Perforce)
   53    9877 [main] git-update-index 1476 mount_info::conv_to_posix_path: conv_to_posix_path (c:\Program Files\QNX Software Systems\bin, keep-rel, no-add-slash)
   52    9929 [main] git-update-index 1476 normalize_win32_path: c:\Program Files\QNX Software Systems\bin = normalize_win32_path (c:\Program Files\QNX Software Systems\bin)
   52    9981 [main] git-update-index 1476 mount_info::conv_to_posix_path: /c/Program Files/QNX Software Systems/bin = conv_to_posix_path (c:\Program Files\QNX Software Systems\bin)
   53   10034 [main] git-update-index 1476 mount_info::conv_to_posix_path: conv_to_posix_path (D:\Programs\cygwin\bin, keep-rel, no-add-slash)
   51   10085 [main] git-update-index 1476 normalize_win32_path: D:\Programs\cygwin\bin = normalize_win32_path (D:\Programs\cygwin\bin)
   52   10137 [main] git-update-index 1476 mount_info::conv_to_posix_path: /usr/bin = conv_to_posix_path (D:\Programs\cygwin\bin)
   52   10189 [main] git-update-index 1476 mount_info::conv_to_posix_path: conv_to_posix_path (c:\Program Files\doxygen\bin, keep-rel, no-add-slash)
   51   10240 [main] git-update-index 1476 normalize_win32_path: c:\Program Files\doxygen\bin = normalize_win32_path (c:\Program Files\doxygen\bin)
   52   10292 [main] git-update-index 1476 mount_info::conv_to_posix_path: /c/Program Files/doxygen/bin = conv_to_posix_path (c:\Program Files\doxygen\bin)
   53   10345 [main] git-update-index 1476 mount_info::conv_to_posix_path: conv_to_posix_path (d:\Programs\Perl\bin, keep-rel, no-add-slash)
   51   10396 [main] git-update-index 1476 normalize_win32_path: d:\Programs\Perl\bin = normalize_win32_path (d:\Programs\Perl\bin)
   51   10447 [main] git-update-index 1476 mount_info::conv_to_posix_path: /d/Programs/Perl/bin = conv_to_posix_path (d:\Programs\Perl\bin)
   53   10500 [main] git-update-index 1476 mount_info::conv_to_posix_path: conv_to_posix_path (d:\Programs\Vim\vim63, keep-rel, no-add-slash)
   51   10551 [main] git-update-index 1476 normalize_win32_path: d:\Programs\Vim\vim63 = normalize_win32_path (d:\Programs\Vim\vim63)
  100   10651 [main] git-update-index 1476 mount_info::conv_to_posix_path: /d/Programs/Vim/vim63 = conv_to_posix_path (d:\Programs\Vim\vim63)
   53   10704 [main] git-update-index 1476 mount_info::conv_to_posix_path: conv_to_posix_path (d:\scripts, keep-rel, no-add-slash)
   88   10792 [main] git-update-index 1476 normalize_win32_path: d:\scripts = normalize_win32_path (d:\scripts)
   52   10844 [main] git-update-index 1476 mount_info::conv_to_posix_path: /d/scripts = conv_to_posix_path (d:\scripts)
   52   10896 [main] git-update-index 1476 mount_info::conv_to_posix_path: conv_to_posix_path (d:\Programs\Subversion\bin, keep-rel, no-add-slash)
   52   10948 [main] git-update-index 1476 normalize_win32_path: d:\Programs\Subversion\bin = normalize_win32_path (d:\Programs\Subversion\bin)
   53   11001 [main] git-update-index 1476 mount_info::conv_to_posix_path: /d/Programs/Subversion/bin = conv_to_posix_path (d:\Programs\Subversion\bin)
   52   11053 [main] git-update-index 1476 mount_info::conv_to_posix_path: conv_to_posix_path (X, keep-rel, no-add-slash)
   51   11104 [main] git-update-index 1476 mount_info::conv_to_posix_path: X = conv_to_posix_path (X)
   52   11156 [main] git-update-index 1476 mount_info::conv_to_posix_path: conv_to_posix_path (.\, keep-rel, add-slash)
   51   11207 [main] git-update-index 1476 mount_info::conv_to_posix_path: ./ = conv_to_posix_path (.\)
   51   11258 [main] git-update-index 1476 mount_info::conv_to_posix_path: conv_to_posix_path (Y, keep-rel, no-add-slash)
   52   11310 [main] git-update-index 1476 mount_info::conv_to_posix_path: Y = conv_to_posix_path (Y)
   55   11365 [main] git-update-index 1476 mount_info::conv_to_posix_path: conv_to_posix_path (.\, keep-rel, add-slash)
   51   11416 [main] git-update-index 1476 mount_info::conv_to_posix_path: ./ = conv_to_posix_path (.\)
   52   11468 [main] git-update-index 1476 mount_info::conv_to_posix_path: conv_to_posix_path (.\, keep-rel, add-slash)
   51   11519 [main] git-update-index 1476 mount_info::conv_to_posix_path: ./ = conv_to_posix_path (.\)
  168   11687 [main] git-update-index 1476 win_env::add_cache: posix /d/scripts/git:/usr/local/bin:/usr/bin:/usr/bin:/usr/X11R6/bin:/c/WINNT/system32:/c/WINNT:/c/WINNT/System32/Wbem:/c/WINNT/system32/nls:/c/WINNT/system32/nls/ENGLISH:/c/Program Files/Perforce:/c/Program Files/QNX Software Systems/bin:/usr/bin:/c/Program Files/doxygen/bin:/d/Programs/Perl/bin:/d/Programs/Vim/vim63:/d/scripts:/d/Programs/Subversion/bin:X:./:Y:./:./
   56   11743 [main] git-update-index 1476 win_env::add_cache: native PATH=d:\scripts\git;D:\Programs\cygwin\usr\local\bin;D:\Programs\cygwin\bin;D:\Programs\cygwin\bin;D:\Programs\cygwin\usr\X11R6\bin;c:\WINNT\system32;c:\WINNT;c:\WINNT\System32\Wbem;c:\WINNT\system32\nls;c:\WINNT\system32\nls\ENGLISH;c:\Program Files\Perforce;c:\Program Files\QNX Software Systems\bin;D:\Programs\cygwin\bin;c:\Program Files\doxygen\bin;d:\Programs\Perl\bin;d:\Programs\Vim\vim63;d:\scripts;d:\Programs\Subversion\bin;X;.\;Y;.\;.\
   60   11803 [main] git-update-index 1476 posify: env var converted to PATH=/d/scripts/git:/usr/local/bin:/usr/bin:/usr/bin:/usr/X11R6/bin:/c/WINNT/system32:/c/WINNT:/c/WINNT/System32/Wbem:/c/WINNT/system32/nls:/c/WINNT/system32/nls/ENGLISH:/c/Program Files/Perforce:/c/Program Files/QNX Software Systems/bin:/usr/bin:/c/Program Files/doxygen/bin:/d/Programs/Perl/bin:/d/Programs/Vim/vim63:/d/scripts:/d/Programs/Subversion/bin:X:./:Y:./:./
  103   11906 [main] git-update-index 1476 environ_init: 0x10020CE8: PATH=/d/scripts/git:/usr/local/bin:/usr/bin:/usr/bin:/usr/X11R6/bin:/c/WINNT/system32:/c/WINNT:/c/WINNT/System32/Wbem:/c/WINNT/system32/nls:/c/WINNT/system32/nls/ENGLISH:/c/Program Files/Perforce:/c/Program Files/QNX Software Systems/bin:/usr/bin:/c/Program Files/doxygen/bin:/d/Programs/Perl/bin:/d/Programs/Vim/vim63:/d/scripts:/d/Programs/Subversion/bin:X:./:Y:./:./
  108   12014 [main] git-update-index 1476 environ_init: 0x10020B20: PATHEXT=.COM;.EXE;.BAT;.CMD;.VBS;.VBE;.JS;.JSE;.WSF;.WSH
  100   12114 [main] git-update-index 1476 environ_init: 0x10020B60: PLATFORM=WNT
  100   12214 [main] git-update-index 1476 environ_init: 0x10020B78: PRINTER=
  100   12314 [main] git-update-index 1476 environ_init: 0x10020B88: PROCESSOR_ARCHITECTURE=x86
  100   12414 [main] git-update-index 1476 environ_init: 0x10020BA8: PROCESSOR_IDENTIFIER=x86 Family 15 Model 3 Stepping 4, GenuineIntel
  100   12514 [main] git-update-index 1476 environ_init: 0x10020BF0: PROCESSOR_LEVEL=15
  101   12615 [main] git-update-index 1476 environ_init: 0x10020C08: PROCESSOR_REVISION=0304
   99   12714 [main] git-update-index 1476 environ_init: 0x10020C28: PROGRAMFILES=C:\Program Files
  100   12814 [main] git-update-index 1476 environ_init: 0x10020C50: PWD=/d/download/git-mine
  100   12914 [main] git-update-index 1476 environ_init: 0x10020C70: PYTHONPATH=/d/scripts/lib/python
  100   13014 [main] git-update-index 1476 environ_init: 0x10020C98: QNX_CONFIGURATION=C:\Program Files\QNX Software Systems
  100   13114 [main] git-update-index 1476 environ_init: 0x10021520: QNX_HOST=d:/QNX630/host/win32/x86
  100   13214 [main] git-update-index 1476 environ_init: 0x10021548: QNX_TARGET=d:/QNX630/target/qnx6
  100   13314 [main] git-update-index 1476 environ_init: 0x10021570: QTJAVA="C:\WINNT\system32\QTJava.zip"
   99   13413 [main] git-update-index 1476 environ_init: 0x100215A0: ROOTDIR=Z:\SYS
  100   13513 [main] git-update-index 1476 environ_init: 0x10020CD8: SHLVL=1
   99   13612 [main] git-update-index 1476 environ_init: 0x100215B8: SYSTEMDRIVE=C:
  100   13712 [main] git-update-index 1476 environ_init: 0x100215D0: SYSTEMROOT=C:\WINNT
  104   13816 [main] git-update-index 1476 getwinenv: can't set native for TEMP= since no environ yet
   99   13915 [main] git-update-index 1476 mount_info::conv_to_posix_path: conv_to_posix_path (d:\tmp, no-keep-rel, no-add-slash)
   51   13966 [main] git-update-index 1476 normalize_win32_path: d:\tmp = normalize_win32_path (d:\tmp)
   52   14018 [main] git-update-index 1476 mount_info::conv_to_posix_path: /tmp = conv_to_posix_path (d:\tmp)
  148   14166 [main] git-update-index 1476 win_env::add_cache: posix /tmp
   50   14216 [main] git-update-index 1476 win_env::add_cache: native TEMP=d:\tmp
   51   14267 [main] git-update-index 1476 posify: env var converted to TEMP=/tmp
   98   14365 [main] git-update-index 1476 environ_init: 0x100215F8: TEMP=/tmp
  100   14465 [main] git-update-index 1476 environ_init: 0x100215E8: TERM=cygwin
  100   14565 [main] git-update-index 1476 getwinenv: can't set native for TMP= since no environ yet
  100   14665 [main] git-update-index 1476 mount_info::conv_to_posix_path: conv_to_posix_path (d:\tmp, no-keep-rel, no-add-slash)
   51   14716 [main] git-update-index 1476 normalize_win32_path: d:\tmp = normalize_win32_path (d:\tmp)
   52   14768 [main] git-update-index 1476 mount_info::conv_to_posix_path: /tmp = conv_to_posix_path (d:\tmp)
  147   14915 [main] git-update-index 1476 win_env::add_cache: posix /tmp
   50   14965 [main] git-update-index 1476 win_env::add_cache: native TMP=d:\tmp
   51   15016 [main] git-update-index 1476 posify: env var converted to TMP=/tmp
   98   15114 [main] git-update-index 1476 environ_init: 0x10021738: TMP=/tmp
  100   15214 [main] git-update-index 1476 getwinenv: can't set native for TMPDIR= since no environ yet
  100   15314 [main] git-update-index 1476 mount_info::conv_to_posix_path: conv_to_posix_path (d:\tmp, no-keep-rel, no-add-slash)
   51   15365 [main] git-update-index 1476 normalize_win32_path: d:\tmp = normalize_win32_path (d:\tmp)
   51   15416 [main] git-update-index 1476 mount_info::conv_to_posix_path: /tmp = conv_to_posix_path (d:\tmp)
  148   15564 [main] git-update-index 1476 win_env::add_cache: posix /tmp
   51   15615 [main] git-update-index 1476 win_env::add_cache: native TMPDIR=d:\tmp
   50   15665 [main] git-update-index 1476 posify: env var converted to TMPDIR=/tmp
   98   15763 [main] git-update-index 1476 environ_init: 0x10021880: TMPDIR=/tmp
  100   15863 [main] git-update-index 1476 environ_init: 0x10021868: USER=ARiesen
   99   15962 [main] git-update-index 1476 environ_init: 0x100219A8: USERDOMAIN=HBI
  100   16062 [main] git-update-index 1476 environ_init: 0x100219C0: USERNAME=ARiesen
  100   16162 [main] git-update-index 1476 environ_init: 0x100219D8: USERPROFILE=C:\Documents and Settings\ARiesen
  100   16262 [main] git-update-index 1476 environ_init: 0x10021A10: WINDIR=C:\WINNT
   99   16361 [main] git-update-index 1476 environ_init: 0x10021A28: WINDOWS_LOGIN=0
  100   16461 [main] git-update-index 1476 environ_init: 0x10021A40: WINVER=5.0.2195  Service Pack 4
  100   16561 [main] git-update-index 1476 environ_init: 0x10021A68: X11=/usr/X11R6
   99   16660 [main] git-update-index 1476 environ_init: 0x10021A80: _=/usr/bin/strace
   58   16718 [main] git-update-index 1476 pinfo_init: Set nice to 0
   51   16769 [main] git-update-index 1476 pinfo_init: pid 1476, pgid 1476
  398   17167 [main] git-update-index 1476 set_privilege: 0 = set_privilege ((token 35C) SeRestorePrivilege, 1)
   64   17231 [main] git-update-index 1476 set_privilege: 1 = set_privilege ((token 35C) SeChangeNotifyPrivilege, 1)
   87   17318 [main] git-update-index 1476 dtable::extend: size 32, fds 0x611556B0
   63   17381 [main] git-update-index 1476 normalize_posix_path: src /etc/passwd
   51   17432 [main] git-update-index 1476 normalize_posix_path: /etc/passwd = normalize_posix_path (/etc/passwd)
   52   17484 [main] git-update-index 1476 mount_info::conv_to_win32_path: conv_to_win32_path (/etc/passwd)
   53   17537 [main] git-update-index 1476 set_flags: flags: binary (0x2)
   51   17588 [main] git-update-index 1476 mount_info::conv_to_win32_path: src_path /etc/passwd, dst D:\Programs\cygwin\etc\passwd, flags 0xA, rc 0
   98   17686 [main] git-update-index 1476 symlink_info::check: not a symlink
   53   17739 [main] git-update-index 1476 symlink_info::check: 0 = symlink.check (D:\Programs\cygwin\etc\passwd, 0x22E830) (0xA)
   54   17793 [main] git-update-index 1476 path_conv::check: this->path(D:\Programs\cygwin\etc\passwd), has_acls(0)
  107   17900 [main] git-update-index 1476 etc::test_file_change: FindFirstFile succeeded
  356   18256 [main] git-update-index 1476 pwdgrp::load: /etc/passwd curr_lines 5
   52   18308 [main] git-update-index 1476 pwdgrp::load: /etc/passwd load succeeded
   66   18374 [main] git-update-index 1476 normalize_posix_path: src /etc/group
   52   18426 [main] git-update-index 1476 normalize_posix_path: /etc/group = normalize_posix_path (/etc/group)
   51   18477 [main] git-update-index 1476 mount_info::conv_to_win32_path: conv_to_win32_path (/etc/group)
   52   18529 [main] git-update-index 1476 set_flags: flags: binary (0x2)
   51   18580 [main] git-update-index 1476 mount_info::conv_to_win32_path: src_path /etc/group, dst D:\Programs\cygwin\etc\group, flags 0xA, rc 0
   78   18658 [main] git-update-index 1476 symlink_info::check: not a symlink
   52   18710 [main] git-update-index 1476 symlink_info::check: 0 = symlink.check (D:\Programs\cygwin\etc\group, 0x22E7F0) (0xA)
   53   18763 [main] git-update-index 1476 path_conv::check: this->path(D:\Programs\cygwin\etc\group), has_acls(0)
   94   18857 [main] git-update-index 1476 etc::test_file_change: FindFirstFile succeeded
  343   19200 [main] git-update-index 1476 pwdgrp::load: /etc/group curr_lines 8
   52   19252 [main] git-update-index 1476 pwdgrp::load: /etc/group load succeeded
  184   19436 [main] git-update-index 1476 pwdgrp::read_group: Completing /etc/group: mkgroup:S-1-5-21-484763869-1275210071-682003330-513:10513:ARiesen
  108   19544 [main] git-update-index 1476 cygheap_user::ontherange: what 2, pw 0x10021CE8
   54   19598 [main] git-update-index 1476 cygheap_user::ontherange: HOME is already in the environment /cygdrive/h
  205   19803 [main] git-update-index 1476 sigproc_init: process/signal handling enabled(1)
  313   20116 [sig] git-update-index 1476 wait_sig: myself->dwProcessId 1476
   61   20177 [sig] git-update-index 1476 wait_sig: entering ReadFile loop, readsig 0x328, myself->sendsig 0x324
  154   20331 [main] git-update-index 1476 build_argv: argv[0] = 'git-update-index'
   54   20385 [main] git-update-index 1476 build_argv: argv[1] = 'index.c'
   56   20441 [main] git-update-index 1476 build_argv: argc 2
  116   20557 [main] git-update-index 1476 build_fh_pc: fh 0x611557E8
   72   20629 [main] git-update-index 1476 open_shared: name (null), shared 0x60020000 (wanted 0x60020000), h 0x318
   61   20690 [main] git-update-index 1476 _pinfo::set_ctty: checking if /dev/tty-1 changed
   53   20743 [main] git-update-index 1476 _pinfo::set_ctty: attached tty1073741824 sid 1476, pid 1476, tty->pgid 1476, tty->sid 1476
 3244   23987 [main] git-update-index 1476 fhandler_base::set_flags: flags 0x10002, supplied_bin 0x0
   74   24061 [main] git-update-index 1476 fhandler_base::set_flags: O_TEXT/O_BINARY set in flags 0x10000
   55   24116 [main] git-update-index 1476 fhandler_base::set_flags: filemode set to binary
  106   24222 [main] git-update-index 1476 fhandler_console::open: incremented open_fhs, now 1
   54   24276 [main] git-update-index 1476 fhandler_console::open: opened conin$ 0x2B, conout$ 0x2F
   68   24344 [main] git-update-index 1476 fhandler_console::output_tcsetattr: 0 = tcsetattr (,60020018) (ENABLE FLAGS 3) (lflag 107 oflag 9)
  137   24481 [main] git-update-index 1476 build_fh_pc: fh 0x611559F0
   54   24535 [main] git-update-index 1476 fhandler_base::set_flags: flags 0x10002, supplied_bin 0x0
   54   24589 [main] git-update-index 1476 fhandler_base::set_flags: O_TEXT/O_BINARY set in flags 0x10000
   53   24642 [main] git-update-index 1476 fhandler_base::set_flags: filemode set to binary
   95   24737 [main] git-update-index 1476 fhandler_console::open: incremented open_fhs, now 2
   56   24793 [main] git-update-index 1476 fhandler_console::open: opened conin$ 0xB, conout$ 0x33
   66   24859 [main] git-update-index 1476 fhandler_console::output_tcsetattr: 0 = tcsetattr (,60020018) (ENABLE FLAGS 3) (lflag 107 oflag 9)
  126   24985 [main] git-update-index 1476 build_fh_pc: fh 0x61155BF8
   53   25038 [main] git-update-index 1476 fhandler_base::set_flags: flags 0x10002, supplied_bin 0x0
   53   25091 [main] git-update-index 1476 fhandler_base::set_flags: O_TEXT/O_BINARY set in flags 0x10000
   53   25144 [main] git-update-index 1476 fhandler_base::set_flags: filemode set to binary
   88   25232 [main] git-update-index 1476 fhandler_console::open: incremented open_fhs, now 3
   53   25285 [main] git-update-index 1476 fhandler_console::open: opened conin$ 0x13, conout$ 0x37
   66   25351 [main] git-update-index 1476 fhandler_console::output_tcsetattr: 0 = tcsetattr (,60020018) (ENABLE FLAGS 3) (lflag 107 oflag 9)
  128   25479 [main] git-update-index 1476 dll_crt0_1: user_data->main 0x401580
   54   25533 [main] git-update-index 1476 __set_errno: void dll_crt0_1(char*):880 val 0
   53   25586 [main] git-update-index 1476 wait_for_sigthread: wait_sig_inited 0x320
  169   25755 [main] git-update-index 1476 cwdstuff::get: posix /d/download/git-mine
   53   25808 [main] git-update-index 1476 cwdstuff::get: (/d/download/git-mine) = cwdstuff::get (0x40C450, 261, 1, 1), errno 0
   60   25868 [main] git-update-index 1476 normalize_posix_path: src .git/refs/
   53   25921 [main] git-update-index 1476 cwdstuff::get: posix /d/download/git-mine
   52   25973 [main] git-update-index 1476 cwdstuff::get: (/d/download/git-mine) = cwdstuff::get (0x22EAA0, 260, 1, 0), errno 0
   54   26027 [main] git-update-index 1476 normalize_posix_path: /d/download/git-mine/.git/refs/ = normalize_posix_path (.git/refs/)
   54   26081 [main] git-update-index 1476 mount_info::conv_to_win32_path: conv_to_win32_path (/d/download/git-mine/.git/refs)
   55   26136 [main] git-update-index 1476 set_flags: flags: binary (0x2)
   53   26189 [main] git-update-index 1476 mount_info::conv_to_win32_path: src_path /d/download/git-mine/.git/refs, dst d:\download\git-mine\.git\refs, flags 0xA, rc 0
  104   26293 [main] git-update-index 1476 symlink_info::check: not a symlink
   55   26348 [main] git-update-index 1476 symlink_info::check: 0 = symlink.check (d:\download\git-mine\.git\refs, 0x22E760) (0xA)
   55   26403 [main] git-update-index 1476 path_conv::check: this->path(d:\download\git-mine\.git\refs), has_acls(0)
   55   26458 [main] git-update-index 1476 build_fh_pc: fh 0x61155E00
  231   26689 [main] git-update-index 1476 get_file_attribute: file: d:\download\git-mine\.git\refs
   60   26749 [main] git-update-index 1476 fhandler_base::fstat_helper: 0 = fstat (, 0x22ECF0) st_atime=4341A9E0 st_size=0, st_mode=0x41ED, st_ino=318799352, sizeof=96
   56   26805 [main] git-update-index 1476 fhandler_base::fhaccess: returning 0
   53   26858 [main] git-update-index 1476 access: returning 0
   55   26913 [main] git-update-index 1476 normalize_posix_path: src .git/objects/
   52   26965 [main] git-update-index 1476 cwdstuff::get: posix /d/download/git-mine
   53   27018 [main] git-update-index 1476 cwdstuff::get: (/d/download/git-mine) = cwdstuff::get (0x22EAA0, 260, 1, 0), errno 0
   54   27072 [main] git-update-index 1476 normalize_posix_path: /d/download/git-mine/.git/objects/ = normalize_posix_path (.git/objects/)
   53   27125 [main] git-update-index 1476 mount_info::conv_to_win32_path: conv_to_win32_path (/d/download/git-mine/.git/objects)
   54   27179 [main] git-update-index 1476 set_flags: flags: binary (0x2)
   53   27232 [main] git-update-index 1476 mount_info::conv_to_win32_path: src_path /d/download/git-mine/.git/objects, dst d:\download\git-mine\.git\objects, flags 0xA, rc 0
   82   27314 [main] git-update-index 1476 symlink_info::check: not a symlink
   53   27367 [main] git-update-index 1476 symlink_info::check: 0 = symlink.check (d:\download\git-mine\.git\objects, 0x22E760) (0xA)
  173   27540 [main] git-update-index 1476 path_conv::check: this->path(d:\download\git-mine\.git\objects), has_acls(0)
   57   27597 [main] git-update-index 1476 build_fh_pc: fh 0x61155E00
  400   27997 [main] git-update-index 1476 get_file_attribute: file: d:\download\git-mine\.git\objects
   57   28054 [main] git-update-index 1476 fhandler_base::fstat_helper: 0 = fstat (, 0x22ECF0) st_atime=4341A9E0 st_size=0, st_mode=0x41ED, st_ino=67615292, sizeof=96
   55   28109 [main] git-update-index 1476 fhandler_base::fhaccess: returning 0
   53   28162 [main] git-update-index 1476 access: returning 0
   53   28215 [main] git-update-index 1476 lstat64: entering
   53   28268 [main] git-update-index 1476 normalize_posix_path: src .git/HEAD
   52   28320 [main] git-update-index 1476 cwdstuff::get: posix /d/download/git-mine
   53   28373 [main] git-update-index 1476 cwdstuff::get: (/d/download/git-mine) = cwdstuff::get (0x22E8F0, 260, 1, 0), errno 0
   54   28427 [main] git-update-index 1476 normalize_posix_path: /d/download/git-mine/.git/HEAD = normalize_posix_path (.git/HEAD)
   53   28480 [main] git-update-index 1476 mount_info::conv_to_win32_path: conv_to_win32_path (/d/download/git-mine/.git/HEAD)
   55   28535 [main] git-update-index 1476 set_flags: flags: binary (0x2)
   52   28587 [main] git-update-index 1476 mount_info::conv_to_win32_path: src_path /d/download/git-mine/.git/HEAD, dst d:\download\git-mine\.git\HEAD, flags 0xA, rc 0
   81   28668 [main] git-update-index 1476 symlink_info::check: GetFileAttributes (d:\download\git-mine\.git\HEAD) failed
   55   28723 [main] git-update-index 1476 geterrno_from_win_error: windows error 2 == errno 2
   81   28804 [main] git-update-index 1476 symlink_info::check: GetFileAttributes (d:\download\git-mine\.git\HEAD.exe) failed
   54   28858 [main] git-update-index 1476 geterrno_from_win_error: windows error 2 == errno 2
   76   28934 [main] git-update-index 1476 symlink_info::check: GetFileAttributes (d:\download\git-mine\.git\HEAD.exe.lnk) failed
   54   28988 [main] git-update-index 1476 geterrno_from_win_error: windows error 2 == errno 2
  129   29117 [main] git-update-index 1476 symlink_info::check: 17 = symlink.check (d:\download\git-mine\.git\HEAD.lnk, 0x22E5B0) (0x1000001)
   59   29176 [main] git-update-index 1476 path_conv::check: this->path(d:\download\git-mine\.git\HEAD.lnk), has_acls(0)
   56   29232 [main] git-update-index 1476 build_fh_pc: fh 0x61155E00
   53   29285 [main] git-update-index 1476 stat_worker: (.git/HEAD, 0x22ED20, 1, 0x61155E00), file_attributes 1
   99   29384 [main] git-update-index 1476 get_file_attribute: file: d:\download\git-mine\.git\HEAD.lnk
   55   29439 [main] git-update-index 1476 fhandler_base::fstat_helper: 0 = fstat (, 0x22ED20) st_atime=4342FB60 st_size=17, st_mode=0xA1FF, st_ino=332219763, sizeof=96
   56   29495 [main] git-update-index 1476 stat_worker: 0 = (.git/HEAD, 0x22ED20)
   54   29549 [main] git-update-index 1476 normalize_posix_path: src .git/HEAD
   53   29602 [main] git-update-index 1476 cwdstuff::get: posix /d/download/git-mine
   52   29654 [main] git-update-index 1476 cwdstuff::get: (/d/download/git-mine) = cwdstuff::get (0x22E940, 260, 1, 0), errno 0
   54   29708 [main] git-update-index 1476 normalize_posix_path: /d/download/git-mine/.git/HEAD = normalize_posix_path (.git/HEAD)
   53   29761 [main] git-update-index 1476 mount_info::conv_to_win32_path: conv_to_win32_path (/d/download/git-mine/.git/HEAD)
   54   29815 [main] git-update-index 1476 set_flags: flags: binary (0x2)
   53   29868 [main] git-update-index 1476 mount_info::conv_to_win32_path: src_path /d/download/git-mine/.git/HEAD, dst d:\download\git-mine\.git\HEAD, flags 0xA, rc 0
   78   29946 [main] git-update-index 1476 symlink_info::check: GetFileAttributes (d:\download\git-mine\.git\HEAD) failed
   55   30001 [main] git-update-index 1476 geterrno_from_win_error: windows error 2 == errno 2
   75   30076 [main] git-update-index 1476 symlink_info::check: GetFileAttributes (d:\download\git-mine\.git\HEAD.exe) failed
   59   30135 [main] git-update-index 1476 geterrno_from_win_error: windows error 2 == errno 2
   75   30210 [main] git-update-index 1476 symlink_info::check: GetFileAttributes (d:\download\git-mine\.git\HEAD.exe.lnk) failed
   55   30265 [main] git-update-index 1476 geterrno_from_win_error: windows error 2 == errno 2
  110   30375 [main] git-update-index 1476 symlink_info::check: 17 = symlink.check (d:\download\git-mine\.git\HEAD.lnk, 0x22E600) (0x1000001)
   58   30433 [main] git-update-index 1476 fs_info::update: Cannot get root component of path refs/heads/master
  786   31219 [main] git-update-index 1476 open: open (.git/index.lock, 0xA02)
   54   31273 [main] git-update-index 1476 normalize_posix_path: src .git/index.lock
   53   31326 [main] git-update-index 1476 cwdstuff::get: posix /d/download/git-mine
   52   31378 [main] git-update-index 1476 cwdstuff::get: (/d/download/git-mine) = cwdstuff::get (0x22EA70, 260, 1, 0), errno 0
   55   31433 [main] git-update-index 1476 normalize_posix_path: /d/download/git-mine/.git/index.lock = normalize_posix_path (.git/index.lock)
   53   31486 [main] git-update-index 1476 mount_info::conv_to_win32_path: conv_to_win32_path (/d/download/git-mine/.git/index.lock)
   54   31540 [main] git-update-index 1476 set_flags: flags: binary (0x2)
   53   31593 [main] git-update-index 1476 mount_info::conv_to_win32_path: src_path /d/download/git-mine/.git/index.lock, dst d:\download\git-mine\.git\index.lock, flags 0xA, rc 0
   80   31673 [main] git-update-index 1476 symlink_info::check: GetFileAttributes (d:\download\git-mine\.git\index.lock) failed
   55   31728 [main] git-update-index 1476 geterrno_from_win_error: windows error 2 == errno 2
   75   31803 [main] git-update-index 1476 symlink_info::check: GetFileAttributes (d:\download\git-mine\.git\index.lock.lnk) failed
   55   31858 [main] git-update-index 1476 geterrno_from_win_error: windows error 2 == errno 2
   53   31911 [main] git-update-index 1476 symlink_info::check: 0 = symlink.check (d:\download\git-mine\.git\index.lock, 0x22E730) (0xA)
   54   31965 [main] git-update-index 1476 mount_info::conv_to_win32_path: conv_to_win32_path (/d/download/git-mine/.git)
   54   32019 [main] git-update-index 1476 set_flags: flags: binary (0x2)
   53   32072 [main] git-update-index 1476 mount_info::conv_to_win32_path: src_path /d/download/git-mine/.git, dst d:\download\git-mine\.git, flags 0xA, rc 0
   78   32150 [main] git-update-index 1476 symlink_info::check: not a symlink
   54   32204 [main] git-update-index 1476 symlink_info::check: 0 = symlink.check (d:\download\git-mine\.git, 0x22E730) (0xA)
   54   32258 [main] git-update-index 1476 path_conv::check: this->path(d:\download\git-mine\.git\index.lock), has_acls(0)
   56   32314 [main] git-update-index 1476 build_fh_pc: fh 0x61155E00
   95   32409 [main] git-update-index 1476 fhandler_base::open: (d:\download\git-mine\.git\index.lock, 0x100A02)
   98   32507 [main] git-update-index 1476 fhandler_base::set_flags: flags 0x100A02, supplied_bin 0x10000
   56   32563 [main] git-update-index 1476 fhandler_base::set_flags: filemode set to binary
   53   32616 [main] git-update-index 1476 fhandler_base::open: 0 = NtCreateFile (0x320, C0100000, d:\download\git-mine\.git\index.lock, io, NULL, 80, 7, 2, 20, NULL, 0)
   54   32670 [main] git-update-index 1476 fhandler_base::open: 1 = fhandler_base::open (d:\download\git-mine\.git\index.lock, 0x100A02)
   54   32724 [main] git-update-index 1476 fhandler_base::open_fs: 1 = fhandler_disk_file::open (d:\download\git-mine\.git\index.lock, 0xA02)
   55   32779 [main] git-update-index 1476 open: 3 = open (.git/index.lock, 0xA02)
   56   32835 [main] git-update-index 1476 void: 0x0 = signal (2, 0x402E20)
  157   32992 [main] git-update-index 1476 open: open (.git/index, 0x0)
   53   33045 [main] git-update-index 1476 normalize_posix_path: src .git/index
   53   33098 [main] git-update-index 1476 cwdstuff::get: posix /d/download/git-mine
   57   33155 [main] git-update-index 1476 cwdstuff::get: (/d/download/git-mine) = cwdstuff::get (0x22E960, 260, 1, 0), errno 2
   55   33210 [main] git-update-index 1476 normalize_posix_path: /d/download/git-mine/.git/index = normalize_posix_path (.git/index)
   53   33263 [main] git-update-index 1476 mount_info::conv_to_win32_path: conv_to_win32_path (/d/download/git-mine/.git/index)
   53   33316 [main] git-update-index 1476 set_flags: flags: binary (0x2)
   53   33369 [main] git-update-index 1476 mount_info::conv_to_win32_path: src_path /d/download/git-mine/.git/index, dst d:\download\git-mine\.git\index, flags 0xA, rc 0
   84   33453 [main] git-update-index 1476 symlink_info::check: not a symlink
   54   33507 [main] git-update-index 1476 symlink_info::check: 0 = symlink.check (d:\download\git-mine\.git\index, 0x22E620) (0xA)
   54   33561 [main] git-update-index 1476 path_conv::check: this->path(d:\download\git-mine\.git\index), has_acls(0)
   77   33638 [main] git-update-index 1476 build_fh_pc: fh 0x61156008
   55   33693 [main] git-update-index 1476 fhandler_base::open: (d:\download\git-mine\.git\index, 0x100000)
   66   33759 [main] git-update-index 1476 fhandler_base::set_flags: flags 0x100000, supplied_bin 0x10000
   54   33813 [main] git-update-index 1476 fhandler_base::set_flags: filemode set to binary
   53   33866 [main] git-update-index 1476 fhandler_base::open: 0 = NtCreateFile (0x2FC, 80100000, d:\download\git-mine\.git\index, io, NULL, 0, 7, 1, 20, NULL, 0)
   54   33920 [main] git-update-index 1476 fhandler_base::open: 1 = fhandler_base::open (d:\download\git-mine\.git\index, 0x100000)
   54   33974 [main] git-update-index 1476 fhandler_base::open_fs: 1 = fhandler_disk_file::open (d:\download\git-mine\.git\index, 0x0)
   53   34027 [main] git-update-index 1476 open: 4 = open (.git/index, 0x0)
   75   34102 [main] git-update-index 1476 get_file_attribute: file: d:\download\git-mine\.git\index
   79   34181 [main] git-update-index 1476 fhandler_base::fstat_helper: 0 = fstat (, 0x22ED40) st_atime=4342FB60 st_size=32616, st_mode=0x81A4, st_ino=1769659834, sizeof=96
   57   34238 [main] git-update-index 1476 fstat64: 0 = fstat (4, 0x22ED40)
  101   34339 [main] git-update-index 1476 mmap64: addr 0, len 32616, prot 3, flags 2, fd 4, off 0
   78   34417 [main] git-update-index 1476 fhandler_disk_file::mmap: 307C0000 = MapViewOfFileEx (h:2F8, access:1, 0, off:0, len:32616, addr:0)
   60   34477 [main] git-update-index 1476 mmap64: 307C0000 = mmap() succeeded
   54   34531 [main] git-update-index 1476 close: close (4)
   56   34587 [main] git-update-index 1476 fhandler_base::close: closing '/d/download/git-mine/.git/index' handle 0x2FC
   60   34647 [main] git-update-index 1476 close: 0 = close (4)
  244   34891 [main] git-update-index 1476 lstat64: entering
   53   34944 [main] git-update-index 1476 normalize_posix_path: src index.c
   52   34996 [main] git-update-index 1476 cwdstuff::get: posix /d/download/git-mine
   53   35049 [main] git-update-index 1476 cwdstuff::get: (/d/download/git-mine) = cwdstuff::get (0x22E9A0, 260, 1, 0), errno 22
   54   35103 [main] git-update-index 1476 normalize_posix_path: /d/download/git-mine/index.c = normalize_posix_path (index.c)
   53   35156 [main] git-update-index 1476 mount_info::conv_to_win32_path: conv_to_win32_path (/d/download/git-mine/index.c)
   54   35210 [main] git-update-index 1476 set_flags: flags: binary (0x2)
   53   35263 [main] git-update-index 1476 mount_info::conv_to_win32_path: src_path /d/download/git-mine/index.c, dst d:\download\git-mine\index.c, flags 0xA, rc 0
   86   35349 [main] git-update-index 1476 symlink_info::check: not a symlink
   54   35403 [main] git-update-index 1476 symlink_info::check: 0 = symlink.check (d:\download\git-mine\index.c, 0x22E660) (0xA)
   55   35458 [main] git-update-index 1476 path_conv::check: this->path(d:\download\git-mine\index.c), has_acls(0)
   56   35514 [main] git-update-index 1476 build_fh_pc: fh 0x61156008
   52   35566 [main] git-update-index 1476 stat_worker: (index.c, 0x22ED40, 1, 0x61156008), file_attributes 32
   61   35627 [main] git-update-index 1476 fhandler_base::open: (d:\download\git-mine\index.c, 0x110000)
   65   35692 [main] git-update-index 1476 fhandler_base::set_flags: flags 0x110000, supplied_bin 0x10000
   54   35746 [main] git-update-index 1476 fhandler_base::set_flags: O_TEXT/O_BINARY set in flags 0x10000
   53   35799 [main] git-update-index 1476 fhandler_base::set_flags: filemode set to binary
   52   35851 [main] git-update-index 1476 fhandler_base::open: 0 = NtCreateFile (0x2FC, 80100000, d:\download\git-mine\index.c, io, NULL, 0, 7, 1, 20, NULL, 0)
   55   35906 [main] git-update-index 1476 fhandler_base::open: 1 = fhandler_base::open (d:\download\git-mine\index.c, 0x110000)
   53   35959 [main] git-update-index 1476 fhandler_base::open_fs: 1 = fhandler_disk_file::open (d:\download\git-mine\index.c, 0x10000)
   67   36026 [main] git-update-index 1476 get_file_attribute: file: d:\download\git-mine\index.c
   77   36103 [main] git-update-index 1476 fhandler_base::fstat_helper: 0 = fstat (, 0x22ED40) st_atime=4342FB60 st_size=1301, st_mode=0x81A4, st_ino=-792472185, sizeof=96
   57   36160 [main] git-update-index 1476 fhandler_base::close: closing '/d/download/git-mine/index.c' handle 0x2FC
   59   36219 [main] git-update-index 1476 stat_worker: 0 = (index.c, 0x22ED40)
  120   36339 [main] git-update-index 1476 open: open (index.c, 0x0)
   53   36392 [main] git-update-index 1476 normalize_posix_path: src index.c
   52   36444 [main] git-update-index 1476 cwdstuff::get: posix /d/download/git-mine
   53   36497 [main] git-update-index 1476 cwdstuff::get: (/d/download/git-mine) = cwdstuff::get (0x22E990, 260, 1, 0), errno 22
   54   36551 [main] git-update-index 1476 normalize_posix_path: /d/download/git-mine/index.c = normalize_posix_path (index.c)
   53   36604 [main] git-update-index 1476 mount_info::conv_to_win32_path: conv_to_win32_path (/d/download/git-mine/index.c)
   54   36658 [main] git-update-index 1476 set_flags: flags: binary (0x2)
   63   36721 [main] git-update-index 1476 mount_info::conv_to_win32_path: src_path /d/download/git-mine/index.c, dst d:\download\git-mine\index.c, flags 0xA, rc 0
   80   36801 [main] git-update-index 1476 symlink_info::check: not a symlink
   55   36856 [main] git-update-index 1476 symlink_info::check: 0 = symlink.check (d:\download\git-mine\index.c, 0x22E650) (0xA)
   54   36910 [main] git-update-index 1476 path_conv::check: this->path(d:\download\git-mine\index.c), has_acls(0)
   55   36965 [main] git-update-index 1476 build_fh_pc: fh 0x61156008
   54   37019 [main] git-update-index 1476 fhandler_base::open: (d:\download\git-mine\index.c, 0x100000)
   64   37083 [main] git-update-index 1476 fhandler_base::set_flags: flags 0x100000, supplied_bin 0x10000
   54   37137 [main] git-update-index 1476 fhandler_base::set_flags: filemode set to binary
   53   37190 [main] git-update-index 1476 fhandler_base::open: 0 = NtCreateFile (0x2FC, 80100000, d:\download\git-mine\index.c, io, NULL, 0, 7, 1, 20, NULL, 0)
   54   37244 [main] git-update-index 1476 fhandler_base::open: 1 = fhandler_base::open (d:\download\git-mine\index.c, 0x100000)
   54   37298 [main] git-update-index 1476 fhandler_base::open_fs: 1 = fhandler_disk_file::open (d:\download\git-mine\index.c, 0x0)
   53   37351 [main] git-update-index 1476 open: 4 = open (index.c, 0x0)
   53   37404 [main] git-update-index 1476 mmap64: addr 0, len 1301, prot 1, flags 2, fd 4, off 0
  256   37660 [main] git-update-index 1476 fhandler_disk_file::mmap: 307D0000 = MapViewOfFileEx (h:2F4, access:1, 0, off:0, len:1301, addr:0)
   64   37724 [main] git-update-index 1476 mmap64: 307D0000 = mmap() succeeded
   54   37778 [main] git-update-index 1476 close: close (4)
   53   37831 [main] git-update-index 1476 fhandler_base::close: closing '/d/download/git-mine/index.c' handle 0x2FC
   64   37895 [main] git-update-index 1476 close: 0 = close (4)
  329   38224 [main] git-update-index 1476 normalize_posix_path: src .git/objects/pack
   58   38282 [main] git-update-index 1476 cwdstuff::get: posix /d/download/git-mine
   53   38335 [main] git-update-index 1476 cwdstuff::get: (/d/download/git-mine) = cwdstuff::get (0x22E630, 260, 1, 0), errno 22
   54   38389 [main] git-update-index 1476 normalize_posix_path: /d/download/git-mine/.git/objects/pack = normalize_posix_path (.git/objects/pack)
   54   38443 [main] git-update-index 1476 mount_info::conv_to_win32_path: conv_to_win32_path (/d/download/git-mine/.git/objects/pack)
   54   38497 [main] git-update-index 1476 set_flags: flags: binary (0x2)
   53   38550 [main] git-update-index 1476 mount_info::conv_to_win32_path: src_path /d/download/git-mine/.git/objects/pack, dst d:\download\git-mine\.git\objects\pack, flags 0xA, rc 0
   92   38642 [main] git-update-index 1476 symlink_info::check: not a symlink
   54   38696 [main] git-update-index 1476 symlink_info::check: 0 = symlink.check (d:\download\git-mine\.git\objects\pack, 0x22E2F0) (0xA)
   55   38751 [main] git-update-index 1476 path_conv::check: this->path(d:\download\git-mine\.git\objects\pack), has_acls(0)
   56   38807 [main] git-update-index 1476 build_fh_pc: fh 0x61156008
  349   39156 [main] git-update-index 1476 get_file_attribute: file: d:\download\git-mine\.git\objects\pack
   59   39215 [main] git-update-index 1476 fhandler_base::fstat_helper: 0 = fstat (, 0x22E820) st_atime=4341A9E0 st_size=0, st_mode=0x41ED, st_ino=165413209, sizeof=96
   56   39271 [main] git-update-index 1476 fhandler_base::fhaccess: returning 0
   54   39325 [main] git-update-index 1476 fhandler_disk_file::opendir: 0x100229B8 = opendir (/d/download/git-mine/.git/objects/pack)
  105   39430 [main] git-update-index 1476 fhandler_disk_file::readdir: 0x100229BC = readdir (0x100229B8) (.)
   69   39499 [main] git-update-index 1476 fhandler_disk_file::readdir: 0x100229BC = readdir (0x100229B8) (..)
   81   39580 [main] git-update-index 1476 fhandler_disk_file::readdir: 0x0 = readdir (0x100229B8)
   55   39635 [main] git-update-index 1476 fhandler_disk_file::closedir: 0 = closedir (0x100229B8)
  207   39842 [main] git-update-index 1476 closedir: 0 = closedir (0x22E8D0)
  159   40001 [main] git-update-index 1476 open: open (.git/objects/info/alternates, 0x0)
   53   40054 [main] git-update-index 1476 normalize_posix_path: src .git/objects/info/alternates
   53   40107 [main] git-update-index 1476 cwdstuff::get: posix /d/download/git-mine
   52   40159 [main] git-update-index 1476 cwdstuff::get: (/d/download/git-mine) = cwdstuff::get (0x22E5B0, 260, 1, 0), errno 22
   55   40214 [main] git-update-index 1476 normalize_posix_path: /d/download/git-mine/.git/objects/info/alternates = normalize_posix_path (.git/objects/info/alternates)
   53   40267 [main] git-update-index 1476 mount_info::conv_to_win32_path: conv_to_win32_path (/d/download/git-mine/.git/objects/info/alternates)
   54   40321 [main] git-update-index 1476 set_flags: flags: binary (0x2)
   53   40374 [main] git-update-index 1476 mount_info::conv_to_win32_path: src_path /d/download/git-mine/.git/objects/info/alternates, dst d:\download\git-mine\.git\objects\info\alternates, flags 0xA, rc 0
  108   40482 [main] git-update-index 1476 symlink_info::check: GetFileAttributes (d:\download\git-mine\.git\objects\info\alternates) failed
   57   40539 [main] git-update-index 1476 geterrno_from_win_error: windows error 2 == errno 2
  100   40639 [main] git-update-index 1476 symlink_info::check: GetFileAttributes (d:\download\git-mine\.git\objects\info\alternates.lnk) failed
   56   40695 [main] git-update-index 1476 geterrno_from_win_error: windows error 2 == errno 2
   54   40749 [main] git-update-index 1476 symlink_info::check: 0 = symlink.check (d:\download\git-mine\.git\objects\info\alternates, 0x22E270) (0xA)
   54   40803 [main] git-update-index 1476 mount_info::conv_to_win32_path: conv_to_win32_path (/d/download/git-mine/.git/objects/info)
   54   40857 [main] git-update-index 1476 set_flags: flags: binary (0x2)
   52   40909 [main] git-update-index 1476 mount_info::conv_to_win32_path: src_path /d/download/git-mine/.git/objects/info, dst d:\download\git-mine\.git\objects\info, flags 0xA, rc 0
   85   40994 [main] git-update-index 1476 symlink_info::check: not a symlink
   54   41048 [main] git-update-index 1476 symlink_info::check: 0 = symlink.check (d:\download\git-mine\.git\objects\info, 0x22E270) (0xA)
   55   41103 [main] git-update-index 1476 path_conv::check: this->path(d:\download\git-mine\.git\objects\info\alternates), has_acls(0)
   56   41159 [main] git-update-index 1476 build_fh_pc: fh 0x61156008
   55   41214 [main] git-update-index 1476 fhandler_base::open: (d:\download\git-mine\.git\objects\info\alternates, 0x100000)
   95   41309 [main] git-update-index 1476 seterrno_from_win_error: /netrel/src/cygwin-1.5.18-1/winsup/cygwin/fhandler.cc:666 windows error 2
   56   41365 [main] git-update-index 1476 geterrno_from_win_error: windows error 2 == errno 2
   53   41418 [main] git-update-index 1476 __set_errno: void seterrno_from_win_error(const char*, int, DWORD):307 val 2
   53   41471 [main] git-update-index 1476 fhandler_base::open: C0000034 = NtCreateFile (0x0, 80100000, d:\download\git-mine\.git\objects\info\alternates, io, NULL, 0, 7, 1, 20, NULL, 0)
   55   41526 [main] git-update-index 1476 fhandler_base::open: 0 = fhandler_base::open (d:\download\git-mine\.git\objects\info\alternates, 0x100000)
   54   41580 [main] git-update-index 1476 fhandler_base::open_fs: 0 = fhandler_disk_file::open (d:\download\git-mine\.git\objects\info\alternates, 0x0)
   54   41634 [main] git-update-index 1476 open: -1 = open (.git/objects/info/alternates, 0x0)
   54   41688 [main] git-update-index 1476 stat64: entering
   53   41741 [main] git-update-index 1476 normalize_posix_path: src .git/objects/44/279b2481951e474e959a395d085327082700b0
   53   41794 [main] git-update-index 1476 cwdstuff::get: posix /d/download/git-mine
   52   41846 [main] git-update-index 1476 cwdstuff::get: (/d/download/git-mine) = cwdstuff::get (0x22E770, 260, 1, 0), errno 2
   54   41900 [main] git-update-index 1476 normalize_posix_path: /d/download/git-mine/.git/objects/44/279b2481951e474e959a395d085327082700b0 = normalize_posix_path (.git/objects/44/279b2481951e474e959a395d085327082700b0)
   55   41955 [main] git-update-index 1476 mount_info::conv_to_win32_path: conv_to_win32_path (/d/download/git-mine/.git/objects/44/279b2481951e474e959a395d085327082700b0)
   54   42009 [main] git-update-index 1476 set_flags: flags: binary (0x2)
   53   42062 [main] git-update-index 1476 mount_info::conv_to_win32_path: src_path /d/download/git-mine/.git/objects/44/279b2481951e474e959a395d085327082700b0, dst d:\download\git-mine\.git\objects\44\279b2481951e474e959a395d085327082700b0, flags 0xA, rc 0
   87   42149 [main] git-update-index 1476 symlink_info::check: not a symlink
   54   42203 [main] git-update-index 1476 symlink_info::check: 0 = symlink.check (d:\download\git-mine\.git\objects\44\279b2481951e474e959a395d085327082700b0, 0x22E430) (0xA)
   56   42259 [main] git-update-index 1476 path_conv::check: this->path(d:\download\git-mine\.git\objects\44\279b2481951e474e959a395d085327082700b0), has_acls(0)
   56   42315 [main] git-update-index 1476 build_fh_pc: fh 0x61156008
   53   42368 [main] git-update-index 1476 stat_worker: (.git/objects/44/279b2481951e474e959a395d085327082700b0, 0x22EAD0, 0, 0x61156008), file_attributes 33
   56   42424 [main] git-update-index 1476 fhandler_base::open: (d:\download\git-mine\.git\objects\44\279b2481951e474e959a395d085327082700b0, 0x110000)
   66   42490 [main] git-update-index 1476 fhandler_base::set_flags: flags 0x110000, supplied_bin 0x10000
   55   42545 [main] git-update-index 1476 fhandler_base::set_flags: O_TEXT/O_BINARY set in flags 0x10000
   97   42642 [main] git-update-index 1476 fhandler_base::set_flags: filemode set to binary
   53   42695 [main] git-update-index 1476 fhandler_base::open: 0 = NtCreateFile (0x2FC, 80100000, d:\download\git-mine\.git\objects\44\279b2481951e474e959a395d085327082700b0, io, NULL, 0, 7, 1, 20, NULL, 0)
   55   42750 [main] git-update-index 1476 fhandler_base::open: 1 = fhandler_base::open (d:\download\git-mine\.git\objects\44\279b2481951e474e959a395d085327082700b0, 0x110000)
   59   42809 [main] git-update-index 1476 fhandler_base::open_fs: 1 = fhandler_disk_file::open (d:\download\git-mine\.git\objects\44\279b2481951e474e959a395d085327082700b0, 0x10000)
   69   42878 [main] git-update-index 1476 get_file_attribute: file: d:\download\git-mine\.git\objects\44\279b2481951e474e959a395d085327082700b0
   78   42956 [main] git-update-index 1476 fhandler_base::fstat_helper: 0 = fstat (, 0x22EAD0) st_atime=4342FB60 st_size=562, st_mode=0x8124, st_ino=-1400112376, sizeof=96
   57   43013 [main] git-update-index 1476 fhandler_base::close: closing '/d/download/git-mine/.git/objects/44/279b2481951e474e959a395d085327082700b0' handle 0x2FC
   61   43074 [main] git-update-index 1476 stat_worker: 0 = (.git/objects/44/279b2481951e474e959a395d085327082700b0, 0x22EAD0)
   56   43130 [main] git-update-index 1476 munmap: munmap (addr 307D0000, len 1301)
   66   43196 [main] git-update-index 1476 build_fh_pc: fh 0x61156008
   61   43257 [main] git-update-index 1476 munmap: 0 = munmap(): 307D0000
  101   43358 [main] git-update-index 1476 writev: writev (3, 0x22ECE0, 1)
   53   43411 [main] git-update-index 1476 fhandler_base::write: binary write
   88   43499 [main] git-update-index 1476 writev: 8192 = write (3, 0x22ECE0, 1), errno 2
   87   43586 [main] git-update-index 1476 writev: writev (3, 0x22ECE0, 1)
   53   43639 [main] git-update-index 1476 fhandler_base::write: binary write
   59   43698 [main] git-update-index 1476 writev: 8192 = write (3, 0x22ECE0, 1), errno 2
   86   43784 [main] git-update-index 1476 writev: writev (3, 0x22ECE0, 1)
   53   43837 [main] git-update-index 1476 fhandler_base::write: binary write
   59   43896 [main] git-update-index 1476 writev: 8192 = write (3, 0x22ECE0, 1), errno 2
   86   43982 [main] git-update-index 1476 writev: writev (3, 0x22ED10, 1)
   53   44035 [main] git-update-index 1476 fhandler_base::write: binary write
   59   44094 [main] git-update-index 1476 writev: 8040 = write (3, 0x22ED10, 1), errno 2
  354   44448 [main] git-update-index 1476 fhandler_console::write: 22D140, 36
   52   44500 [main] git-update-index 1476 fhandler_console::write: at 114(r) state is 0
  189   44689 [main] git-update-index 1476 fhandler_console::write: 36 = write_console (,..36)
  160   44849 [main] git-update-index 1476 normalize_posix_path: src .git/index.lock
   54   44903 [main] git-update-index 1476 cwdstuff::get: posix /d/download/git-mine
   52   44955 [main] git-update-index 1476 cwdstuff::get: (/d/download/git-mine) = cwdstuff::get (0x22E730, 260, 1, 0), errno 2
   54   45009 [main] git-update-index 1476 normalize_posix_path: /d/download/git-mine/.git/index.lock = normalize_posix_path (.git/index.lock)
   54   45063 [main] git-update-index 1476 mount_info::conv_to_win32_path: conv_to_win32_path (/d/download/git-mine/.git/index.lock)
   55   45118 [main] git-update-index 1476 set_flags: flags: binary (0x2)
   53   45171 [main] git-update-index 1476 mount_info::conv_to_win32_path: src_path /d/download/git-mine/.git/index.lock, dst d:\download\git-mine\.git\index.lock, flags 0xA, rc 0
   94   45265 [main] git-update-index 1476 symlink_info::check: not a symlink
   54   45319 [main] git-update-index 1476 symlink_info::check: 0 = symlink.check (d:\download\git-mine\.git\index.lock, 0x22E3F0) (0xA)
   55   45374 [main] git-update-index 1476 path_conv::check: this->path(d:\download\git-mine\.git\index.lock), has_acls(0)
   54   45428 [main] git-update-index 1476 normalize_posix_path: src .git/index
   52   45480 [main] git-update-index 1476 cwdstuff::get: posix /d/download/git-mine
   53   45533 [main] git-update-index 1476 cwdstuff::get: (/d/download/git-mine) = cwdstuff::get (0x22E730, 260, 1, 0), errno 2
   54   45587 [main] git-update-index 1476 normalize_posix_path: /d/download/git-mine/.git/index = normalize_posix_path (.git/index)
   53   45640 [main] git-update-index 1476 mount_info::conv_to_win32_path: conv_to_win32_path (/d/download/git-mine/.git/index)
   58   45698 [main] git-update-index 1476 set_flags: flags: binary (0x2)
   53   45751 [main] git-update-index 1476 mount_info::conv_to_win32_path: src_path /d/download/git-mine/.git/index, dst d:\download\git-mine\.git\index, flags 0xA, rc 0
   80   45831 [main] git-update-index 1476 symlink_info::check: not a symlink
   54   45885 [main] git-update-index 1476 symlink_info::check: 0 = symlink.check (d:\download\git-mine\.git\index, 0x22E3F0) (0xA)
   54   45939 [main] git-update-index 1476 path_conv::check: this->path(d:\download\git-mine\.git\index), has_acls(0)
  275   46214 [main] git-update-index 1476 seterrno_from_win_error: /netrel/src/cygwin-1.5.18-1/winsup/cygwin/syscalls.cc:1262 windows error 5
   60   46274 [main] git-update-index 1476 geterrno_from_win_error: windows error 5 == errno 13
   53   46327 [main] git-update-index 1476 __set_errno: void seterrno_from_win_error(const char*, int, DWORD):307 val 13
   54   46381 [main] git-update-index 1476 rename: -1 = rename (d:\download\git-mine\.git\index.lock, d:\download\git-mine\.git\index)
  352   46733 [main] git-update-index 1476 fhandler_console::write: 22D140, 26
   52   46785 [main] git-update-index 1476 fhandler_console::write: at 114(r) state is 0
  142   46927 [main] git-update-index 1476 fhandler_console::write: 26 = write_console (,..26)
  451   47378 [main] git-update-index 1476 fhandler_console::write: 22D280, 38
   53   47431 [main] git-update-index 1476 fhandler_console::write: at 117(u) state is 0
   83   47514 [main] git-update-index 1476 fhandler_console::write: 38 = write_console (,..38)
  253   47767 [main] git-update-index 1476 fhandler_console::write: 409AEB, 7
   53   47820 [main] git-update-index 1476 fhandler_console::write: at 102(f) state is 0
   76   47896 [main] git-update-index 1476 fhandler_console::write: 7 = write_console (,..7)
  448   48344 [main] git-update-index 1476 fhandler_console::write: 22D270, 29
   52   48396 [main] git-update-index 1476 fhandler_console::write: at 85(U) state is 0
   78   48474 [main] git-update-index 1476 fhandler_console::write: 29 = write_console (,..29)
  359   48833 [main] git-update-index 1476 fhandler_console::write: 610E30E7, 1
   53   48886 [main] git-update-index 1476 fhandler_console::write: at 10(0x20) state is 0
   98   48984 [main] git-update-index 1476 fhandler_console::write: 1 = write_console (,..1)
  448   49432 [main] git-update-index 1476 close: close (0)
   68   49500 [main] git-update-index 1476 fhandler_console::close: decremented open_fhs, now 2
   54   49554 [main] git-update-index 1476 close: 0 = close (0)
  442   49996 [main] git-update-index 1476 close: close (1)
   65   50061 [main] git-update-index 1476 fhandler_console::close: decremented open_fhs, now 1
   53   50114 [main] git-update-index 1476 close: 0 = close (1)
  441   50555 [main] git-update-index 1476 close: close (2)
   66   50621 [main] git-update-index 1476 fhandler_console::close: decremented open_fhs, now 0
   53   50674 [main] git-update-index 1476 close: 0 = close (2)
  247   50921 [main] git-update-index 1476 do_exit: do_exit (32768), exit_state 0
   54   50975 [main] git-update-index 1476 void: 0x0 = signal (20, 0x1)
   52   51027 [main] git-update-index 1476 void: 0x0 = signal (1, 0x1)
   53   51080 [main] git-update-index 1476 void: 0x402E20 = signal (2, 0x1)
   53   51133 [main] git-update-index 1476 void: 0x0 = signal (3, 0x1)
   53   51186 [main] git-update-index 1476 fhandler_base::close: closing '/d/download/git-mine/.git/index.lock' handle 0x320
   69   51255 [main] git-update-index 1476 sigproc_terminate: entering
   67   51322 [sig] git-update-index 1476 wait_sig: done
   70   51392 [sig] git-update-index 1476 _cygtls::remove: wait 0x0
  133   51525 [main] git-update-index 1476 proc_terminate: nprocs 0
   52   51577 [main] git-update-index 1476 proc_terminate: leaving
  101   51678 [main] git-update-index 1476 sigproc_terminate: sigproc handling not active
   50   51728 [main] git-update-index 1476 proc_terminate: nprocs 0
   58   51786 [main] git-update-index 1476 proc_terminate: leaving
   55   51841 [main] git-update-index 1476 __to_clock_t: dwHighDateTime 0, dwLowDateTime 0
   50   51891 [main] git-update-index 1476 __to_clock_t: total 00000000 00000000
   51   51942 [main] git-update-index 1476 __to_clock_t: dwHighDateTime 0, dwLowDateTime 468750
   50   51992 [main] git-update-index 1476 __to_clock_t: total 00000000 0000002E
  540   52532 [main] git-update-index 1476 pinfo::maybe_set_exit_code_from_windows: pid 1476, exit value - old 0x80008000, windows 0xDEADBEEF, cygwin 0x80008000

[PATCH] Fix symbolic ref validation

[Jonas Fonseca]

Johannes Schindelin <Johannes.Schindelin@gmx.de> wrote Wed, Oct 05, 2005:
> Hi,

Hello,

> On Wed, 5 Oct 2005, Jonas Fonseca wrote:
> 
> >   user@machine /usr/local/dev/git/git
> >   $ git-log
> >   fatal: Not a git repository
> > 
> >   user@machine /usr/local/dev/git/git
> >   $ GIT_DIR=.git git-log | wc -l
> >   26094
> 
> That could have its cause in your .git/HEAD being no symlink. That happens 
> when rsync´ing the .git directory.

Yes, used rsync when I cloned. Seems validate_symref() was buggy. 

> The other errors could also stem from the fact that quite a few places 
> expect HEAD to be a symlink.

git-reset still error out ...

---

Use the correct buffer when validating 'ref: refs/...'

Signed-off-by: Jonas Fonseca <fonseca@diku.dk>

---
diff --git a/refs.c b/refs.c
--- a/refs.c
+++ b/refs.c
@@ -46,7 +46,7 @@ int validate_symref(const char *path)
 	len -= 4;
 	while (len && isspace(*buf))
 		buf++, len--;
-	if (len >= 5 && !memcmp("refs/", buffer, 5))
+	if (len >= 5 && !memcmp("refs/", buf, 5))
 		return 0;
 	return -1;
 }

-- 
Jonas Fonseca

Re: First cut at git port to Cygwin

[Christopher Faylor]

On Wed, Oct 05, 2005 at 05:46:08PM +0200, Alex Riesen wrote:
>On 10/5/05, Alex Riesen <raa.lkml@gmail.com> wrote:
>> On 10/4/05, H. Peter Anvin <hpa@zytor.com> wrote:
>> > > I noticed that rename(2) in my copy of cygwin (1.5.18-1) does not remove the
>> > > target and returns an error (probably EPERM, but I have reasons not to trust
>> > > strerror on that thing).
>> > > The repository was on FAT.
>> > > Taking "rename(2)" from cygwin's libiberty solved this (they unlink if link(2)
>> > > returns EEXIST).
>> > >
>> > > PS: Does broken rename(2) qualify a system "not worthy to support"?
>> >
>> > I just tried this with Cygwin 1.5.18-1 and didn't have any such
>> > problems.  I tried it on NTFS, FAT and Samba, using WinXP.
>>
>> It's on Win2k, there was multiple cygwin installations in path, the other one
>> supposedly is 1.5.5 (it's from QNX Momentics installation).
>> I had that old "cygwin1.dll" renamed into "cygwin1.dll-disabled" long
>> ago, though...
>> I can't reproduce this out of GIT context, and the error is not
>> reproducable after
>> I removed the other cygwin installation out of PATH.
>> Anyway, sorry, I should have tried this before posting.
>
>Still does not work for me. I cannot isolate the problem out of git,
>but at the moment the only way for me to make commit_index_file to work
>is to put unlink(indexfile) before rename(cf->lockfile, indexfile).
>
>For everyone interested, I attach cygwin's strace output here.

I'm sorry that I missed this thread.  I'm usually pretty alert to the word
"cygwin" showing up in a subject.

I'll go back and read the archives to catch up but, at the risk of
making an observation that has already been made, under windows you
can't always rename a file that is open.  Is that what's happening here?

--
Christopher Faylor			spammer? ->	aaaspam@sourceware.org
Cygwin Co-Project Leader				aaaspam@duffek.com
TimeSys, Inc.

Re: First cut at git port to Cygwin

[Davide Libenzi]

On 10/5/05, Alex Riesen <raa.lkml@gmail.com> wrote:

> Still does not work for me. I cannot isolate the problem out of git,
> but at the moment the only way for me to make commit_index_file to work
> is to put unlink(indexfile) before rename(cf->lockfile, indexfile).

I don't know how Cygwin implemented rename(), but if they used MoveFile() 
they broke the POSIX rename() since MoveFile() fails if destination 
already exists. They should have used MoveFileEx(MOVEFILE_REPLACE_EXISTING)
instead, to guarantee POSIX semantics. The symptoms you're experiencing 
make me think this might be the case (even if it is strange, since other 
Unix software would fail the same way).


- Davide

Re: First cut at git port to Cygwin

[Christopher Faylor]

On Wed, Oct 05, 2005 at 09:09:49AM -0700, Davide Libenzi wrote:
>On 10/5/05, Alex Riesen <raa.lkml@gmail.com> wrote:
>>Still does not work for me.  I cannot isolate the problem out of git,
>>but at the moment the only way for me to make commit_index_file to work
>>is to put unlink(indexfile) before rename(cf->lockfile, indexfile).
>
>I don't know how Cygwin implemented rename(), but if they used
>MoveFile() they broke the POSIX rename() since MoveFile() fails if
>destination already exists.  They should have used
>MoveFileEx(MOVEFILE_REPLACE_EXISTING) instead, to guarantee POSIX
>semantics.  The symptoms you're experiencing make me think this might
>be the case (even if it is strange, since other Unix software would
>fail the same way).

Cygwin's rename is much more than just a simple wrapper around MoveFile
or MoveFileEx.  It tries hard to guarantee POSIX semantics within the
strictures imposed by Windows.
--
Christopher Faylor			spammer? ->	aaaspam@sourceware.org
Cygwin Co-Project Leader				aaaspam@duffek.com
TimeSys, Inc.

Re: First cut at git port to Cygwin

[H. Peter Anvin]

Christopher Faylor wrote:
> Cygwin's rename is much more than just a simple wrapper around MoveFile
> or MoveFileEx.  It tries hard to guarantee POSIX semantics within the
> strictures imposed by Windows.

Certainly, but presumably different versions of Win32 have different 
limitations, no?

	-hpa

Re: First cut at git port to Cygwin

[Christopher Faylor]

On Wed, Oct 05, 2005 at 09:23:36AM -0700, H. Peter Anvin wrote:
>Christopher Faylor wrote:
>>Cygwin's rename is much more than just a simple wrapper around MoveFile
>>or MoveFileEx.  It tries hard to guarantee POSIX semantics within the
>>strictures imposed by Windows.
>
>Certainly, but presumably different versions of Win32 have different 
>limitations, no?

Yes, definitely.  That's why the Cygwin rename code (and unlink code for
that matter) is so headache-inducing.

FWIW, I just looked at the source and, AFAICT, if rename() is setting
EPERM on Windows NT/2000/XP, it should be the result of a failing
MoveFileEx (old, new, MOVEFILE_REPLACE_EXISTING).
--
Christopher Faylor			spammer? ->	aaaspam@sourceware.org
Cygwin Co-Project Leader				aaaspam@duffek.com
TimeSys, Inc.

Re: [PATCH] Fix symbolic ref validation

[Junio C Hamano]

Jonas Fonseca <fonseca@diku.dk> writes:

> Yes, used rsync when I cloned. Seems validate_symref() was buggy. 
>
>> The other errors could also stem from the fact that quite a few places 
>> expect HEAD to be a symlink.
>
> git-reset still error out ...
>
> ---
>
> Use the correct buffer when validating 'ref: refs/...'
>
> Signed-off-by: Jonas Fonseca <fonseca@diku.dk>
>
> ---
> diff --git a/refs.c b/refs.c

Thanks.

One request, not just to Jonas.  Please do not use '^---$' to
separate the introductory discussion and the real commit log
message.

Linus style (recently the kernel list had a thread on this as
well) is to have the commit log upfront with signoff, three-dash
line, optional discussion and diffstat, and then diff.

I do not mind seeing discussion upfront personally [*1*], but
the thing is the tool treats everything after the first '^---$' 
something to be fed to patch, and does not treat it as the
commit log message.


[Footnote]

*1* ...but remember, Linus does.

Re: First cut at git port to Cygwin

[Davide Libenzi]

On Wed, 5 Oct 2005, Christopher Faylor wrote:

> On Wed, Oct 05, 2005 at 09:09:49AM -0700, Davide Libenzi wrote:
>> On 10/5/05, Alex Riesen <raa.lkml@gmail.com> wrote:
>>> Still does not work for me.  I cannot isolate the problem out of git,
>>> but at the moment the only way for me to make commit_index_file to work
>>> is to put unlink(indexfile) before rename(cf->lockfile, indexfile).
>>
>> I don't know how Cygwin implemented rename(), but if they used
>> MoveFile() they broke the POSIX rename() since MoveFile() fails if
>> destination already exists.  They should have used
>> MoveFileEx(MOVEFILE_REPLACE_EXISTING) instead, to guarantee POSIX
>> semantics.  The symptoms you're experiencing make me think this might
>> be the case (even if it is strange, since other Unix software would
>> fail the same way).
>
> Cygwin's rename is much more than just a simple wrapper around MoveFile
> or MoveFileEx.  It tries hard to guarantee POSIX semantics within the
> strictures imposed by Windows.

Ouch, IC:

http://cygwin.com/cgi-bin/cvsweb.cgi/~checkout~/src/winsup/cygwin/syscalls.cc?rev=1.390&content-type=text/plain&cvsroot=src

That's quite some code.



- Davide

Re: First cut at git port to Cygwin

[Alex Riesen]

Christopher Faylor, Wed, Oct 05, 2005 17:54:57 +0200:
> >Still does not work for me. I cannot isolate the problem out of git,
> >but at the moment the only way for me to make commit_index_file to work
> >is to put unlink(indexfile) before rename(cf->lockfile, indexfile).
> >
> >For everyone interested, I attach cygwin's strace output here.
> 
> I'm sorry that I missed this thread.  I'm usually pretty alert to the word
> "cygwin" showing up in a subject.
> 
> I'll go back and read the archives to catch up but, at the risk of
> making an observation that has already been made, under windows you
> can't always rename a file that is open.  Is that what's happening here?
> 

Don't think so, but will check in about 10 hrs. The code in question
is in index.c, commit_index_file.

Re: First cut at git port to Cygwin

[Christopher Faylor]

On Wed, Oct 05, 2005 at 09:17:41PM +0200, Alex Riesen wrote:
>Christopher Faylor, Wed, Oct 05, 2005 17:54:57 +0200:
>> >Still does not work for me. I cannot isolate the problem out of git,
>> >but at the moment the only way for me to make commit_index_file to work
>> >is to put unlink(indexfile) before rename(cf->lockfile, indexfile).
>> >
>> >For everyone interested, I attach cygwin's strace output here.
>> 
>> I'm sorry that I missed this thread.  I'm usually pretty alert to the word
>> "cygwin" showing up in a subject.
>> 
>> I'll go back and read the archives to catch up but, at the risk of
>> making an observation that has already been made, under windows you
>> can't always rename a file that is open.  Is that what's happening here?
>> 
>
>Don't think so, but will check in about 10 hrs. The code in question
>is in index.c, commit_index_file.

Ok.  Looks pretty simple.  FWIW, I've just built git on windows and I
don't see this behavior.  For the most part, it "just works".

I do see the gitk behavior but I'm tk illiterate too, unfortunately, so
I can't help there.

Cygwin's tcl/tk is a funny beast.  It is primarily included (by me) to
allow for the operation of the insight debugger and is more windows than
POSIX.  As was noted, it doesn't interact with Cygwin/X but just draws
windows using the Windows API.

So, tcl/tk are just barely supported in Cygwin currently.  I'm actually
a little surprised that it works as well as it does with gitk.
--
Christopher Faylor			spammer? ->	aaaspam@sourceware.org
Cygwin Co-Project Leader				aaaspam@duffek.com
TimeSys, Inc.

Re: First cut at git port to Cygwin

[Alex Riesen]

[-- Attachment #1: Type: text/plain, Size: 924 bytes --]

On 10/5/05, Christopher Faylor <me@cgf.cx> wrote:
> >Don't think so, but will check in about 10 hrs. The code in question
> >is in index.c, commit_index_file.
>
> Ok.  Looks pretty simple.  FWIW, I've just built git on windows and I
> don't see this behavior.  For the most part, it "just works".

Thanks for the hint. There are open files involved (both index.lock and index).
I attach the patch which closes index.lock (this is not really needed, btw:
rename works even without closing index.lock) and unmaps the index
(a bit too intrusive). The patch fixes only update-index.c (the one I had
problems with), there probably are other places were the situation is alike.

I don't like the patch (and win32 at all; hence the offending comment),
so use it only unless there is no other possibility to workaround.
I specifically do not request its inclusion into official branch
(even though Junio is cc'ed).

[-- Attachment #2: unmap_active_cache.patch --]
[-- Type: application/xxxxx, Size: 1998 bytes --]

Re: First cut at git port to Cygwin

[Alex Riesen]

[-- Attachment #1: Type: text/plain, Size: 399 bytes --]

On 10/6/05, Alex Riesen <raa.lkml@gmail.com> wrote:
> (a bit too intrusive). The patch fixes only update-index.c (the one I had
> problems with), there probably are other places were the situation is alike.

of course there are "other places". Please, try the attached patch instead.

For the record: the patch is supposed to help people with
"Unable to write new cachefile" kind of errors.

[-- Attachment #2: unmap_active_cache.patch --]
[-- Type: application/xxxxx, Size: 3608 bytes --]

Re: First cut at git port to Cygwin

[Alex Riesen]

On 10/6/05, Alex Riesen <raa.lkml@gmail.com> wrote:
> > (a bit too intrusive). The patch fixes only update-index.c (the one I had
> > problems with), there probably are other places were the situation is alike.
>
> of course there are "other places". Please, try the attached patch instead.
>
> For the record: the patch is supposed to help people with
> "Unable to write new cachefile" kind of errors.

Just as I thought the situation improved (by closing index.lock and
unmapping index),
it suddenly get worse: now I'm stuck on git-pull.

git-merge-index (called at some point by git-pull) maps the index in, and starts
git-merge-one-file for each (or the given) entry in the index.
git-merge-one-file
calls git-update-index, which wants to update the index. Which doesn't work,
because it's locked by that piece of s$%^.

The only working walkaround for me atm is unlinking indexfile before rename in
commit_index_file :(

Re: First cut at git port to Cygwin

[Linus Torvalds]

On Fri, 7 Oct 2005, Alex Riesen wrote:
>
> it suddenly get worse: now I'm stuck on git-pull.
> 
> git-merge-index (called at some point by git-pull) maps the index in, and starts
> git-merge-one-file for each (or the given) entry in the index.
> git-merge-one-file
> calls git-update-index, which wants to update the index. Which doesn't work,
> because it's locked by that piece of s$%^.

NOTE! git doesn't use mmap() because it _needs_ to use mmap(), but because 
it was simple to do that way, and it's a total idiosyncracy of mine that I 
often try to mmap the data. I often also tend to do my own allocators 
instead of using malloc() (see my "sparse" project in case you're 
interested in other idiosyncracies of mine - macros to do list traversal 
etc).

The fact is, "mmap()" isn't really any better than "read()": it has some 
advantages wrt memory management for the kernel, which is probably one big 
reason why I do it, but quite frankly, if you were to change every single 
mmap() to be a "map_file()" instead, and made it optional whether it used 
mmap() or "malloc + read()", I personally don't think it would be 
horrible.

And it might make things much simpler for portability. The "use mmap" 
approach is very much a unixism, particularly the way unix people do it 
(mmap followed by close, making the file descriptor "go away"). Sure, 
other OS's have mmap too, but I think on them it tends to be less commonly 
used.

			Linus

Re: First cut at git port to Cygwin

[Alex Riesen]

Linus Torvalds, Fri, Oct 07, 2005 17:34:19 +0200:
> > it suddenly get worse: now I'm stuck on git-pull.
> > 
> > git-merge-index (called at some point by git-pull) maps the index
> > in, and starts git-merge-one-file for each (or the given) entry in
> > the index.  git-merge-one-file calls git-update-index, which wants
> > to update the index. Which doesn't work, because it's locked by
> > that piece of s$%^.
> 
> NOTE! git doesn't use mmap() because it _needs_ to use mmap(), but because 
> it was simple to do that way, and it's a total idiosyncracy of mine that I 
> often try to mmap the data. I often also tend to do my own allocators 
> instead of using malloc() (see my "sparse" project in case you're 
> interested in other idiosyncracies of mine - macros to do list traversal 
> etc).
> 
> The fact is, "mmap()" isn't really any better than "read()": it has some 
> advantages wrt memory management for the kernel, which is probably one big 
> reason why I do it, but quite frankly, if you were to change every single 
> mmap() to be a "map_file()" instead, and made it optional whether it used 
> mmap() or "malloc + read()", I personally don't think it would be 
> horrible.
> 
> And it might make things much simpler for portability. The "use mmap" 
> approach is very much a unixism, particularly the way unix people do it 
> (mmap followed by close, making the file descriptor "go away"). Sure, 
> other OS's have mmap too, but I think on them it tends to be less commonly 
> used.

"Sounds like a thinly veiled threat or a very effective prodding" 8)

---

Make read_cache copy the index into memory, to improve portability on
other OS's which have mmap too, tend to use it less commonly.

Signed-off-by: Alex Riesen <raa.lkml@gmail.com>

diff --git a/read-cache.c b/read-cache.c
--- a/read-cache.c
+++ b/read-cache.c
@@ -497,9 +497,11 @@ int read_cache(void)
 	offset = sizeof(*hdr);
 	for (i = 0; i < active_nr; i++) {
 		struct cache_entry *ce = map + offset;
-		offset = offset + ce_size(ce);
-		active_cache[i] = ce;
+		size_t size = ce_size(ce);
+		offset = offset + size;
+		active_cache[i] = malloc(ce, size);
 	}
+	munmap(map, size);
 	return active_nr;
 
 unmap:

Re: First cut at git port to Cygwin

[Alex Riesen]

Alex Riesen, Fri, Oct 07, 2005 22:54:50 +0200:
> Linus Torvalds, Fri, Oct 07, 2005 17:34:19 +0200:
> > > it suddenly get worse: now I'm stuck on git-pull.
> > > 
> > > git-merge-index (called at some point by git-pull) maps the index
> > > in, and starts git-merge-one-file for each (or the given) entry in
> > > the index.  git-merge-one-file calls git-update-index, which wants
> > > to update the index. Which doesn't work, because it's locked by
> > > that piece of s$%^.
> > 
> > NOTE! git doesn't use mmap() because it _needs_ to use mmap(), but because 
> > it was simple to do that way, and it's a total idiosyncracy of mine that I 
> > often try to mmap the data. I often also tend to do my own allocators 
> > instead of using malloc() (see my "sparse" project in case you're 
> > interested in other idiosyncracies of mine - macros to do list traversal 
> > etc).
> > 
> > The fact is, "mmap()" isn't really any better than "read()": it has some 
> > advantages wrt memory management for the kernel, which is probably one big 
> > reason why I do it, but quite frankly, if you were to change every single 
> > mmap() to be a "map_file()" instead, and made it optional whether it used 
> > mmap() or "malloc + read()", I personally don't think it would be 
> > horrible.
> > 
> > And it might make things much simpler for portability. The "use mmap" 
> > approach is very much a unixism, particularly the way unix people do it 
> > (mmap followed by close, making the file descriptor "go away"). Sure, 
> > other OS's have mmap too, but I think on them it tends to be less commonly 
> > used.
> 
> "Sounds like a thinly veiled threat or a very effective prodding" 8)
> 

Junio C Hamano, Fri, Oct 07, 2005 23:00:02 +0200:
> Huh?  where is your memcpy?

Unbelievable... I actually tested the change! But not _the_ patch.
Thanks. Next time, hit me :)

---

Make read_cache copy the index into memory, to improve portability on
other OS's which have mmap too, tend to use it less commonly.

Signed-off-by: Alex Riesen <raa.lkml@gmail.com>

diff --git a/read-cache.c b/read-cache.c
--- a/read-cache.c
+++ b/read-cache.c
@@ -497,9 +497,12 @@ int read_cache(void)
 	offset = sizeof(*hdr);
 	for (i = 0; i < active_nr; i++) {
 		struct cache_entry *ce = map + offset;
-		offset = offset + ce_size(ce);
-		active_cache[i] = ce;
+		size_t size = ce_size(ce);
+		struct cache_entry *newce = malloc(size);
+		offset = offset + size;
+		active_cache[i] = memcpy(newce, ce, size);
 	}
+	munmap(map, size);
 	return active_nr;
 
 unmap:

Re: First cut at git port to Cygwin

[Chuck Lever]

[-- Attachment #1: Type: text/plain, Size: 752 bytes --]

Alex Riesen wrote:
> Make read_cache copy the index into memory, to improve portability on
> other OS's which have mmap too, tend to use it less commonly.
> 
> Signed-off-by: Alex Riesen <raa.lkml@gmail.com>
> 
> diff --git a/read-cache.c b/read-cache.c
> --- a/read-cache.c
> +++ b/read-cache.c
> @@ -497,9 +497,12 @@ int read_cache(void)
>  	offset = sizeof(*hdr);
>  	for (i = 0; i < active_nr; i++) {
>  		struct cache_entry *ce = map + offset;
> -		offset = offset + ce_size(ce);
> -		active_cache[i] = ce;
> +		size_t size = ce_size(ce);
> +		struct cache_entry *newce = malloc(size);
> +		offset = offset + size;
> +		active_cache[i] = memcpy(newce, ce, size);
>  	}
> +	munmap(map, size);
>  	return active_nr;
>  
>  unmap:

s/malloc/xmalloc/

[-- Attachment #2: cel.vcf --]
[-- Type: text/x-vcard, Size: 439 bytes --]

begin:vcard
fn:Chuck Lever
n:Lever;Charles
org:Network Appliance, Incorporated;Linux NFS Client Development
adr:535 West William Street, Suite 3100;;Center for Information Technology Integration;Ann Arbor;MI;48103-4943;USA
email;internet:cel@citi.umich.edu
title:Member of Technical Staff
tel;work:+1 734 763 4415
tel;fax:+1 734 763 4434
tel;home:+1 734 668 1089
x-mozilla-html:FALSE
url:http://www.monkey.org/~cel/
version:2.1
end:vcard

Re: First cut at git port to Cygwin

[Alex Riesen]

Chuck Lever, Fri, Oct 07, 2005 23:29:16 +0200:
> s/malloc/xmalloc/

It's not that funny after second repost...

---

Make read_cache copy the index into memory, to improve portability on
other OS's which have mmap too, tend to use it less commonly.

Signed-off-by: Alex Riesen <raa.lkml@gmail.com>

diff --git a/read-cache.c b/read-cache.c
--- a/read-cache.c
+++ b/read-cache.c
@@ -497,9 +497,12 @@ int read_cache(void)
 	offset = sizeof(*hdr);
 	for (i = 0; i < active_nr; i++) {
 		struct cache_entry *ce = map + offset;
-		offset = offset + ce_size(ce);
-		active_cache[i] = ce;
+		size_t size = ce_size(ce);
+		struct cache_entry *newce = xmalloc(size);
+		offset = offset + size;
+		active_cache[i] = memcpy(newce, ce, size);
 	}
+	munmap(map, size);
 	return active_nr;
 
 unmap:

Re: First cut at git port to Cygwin

[Alex Riesen]

Junio C Hamano, Fri, Oct 07, 2005 23:00:02 +0200:
> > "Sounds like a thinly veiled threat or a very effective prodding" 8)
> > ---
> >
> > Make read_cache copy the index into memory, to improve portability on
> > other OS's which have mmap too, tend to use it less commonly.
> >
> 
> Huh?  where is your memcpy?
> 

Junio, unless there already are pressing reasons to put the patch in
GIT, could you postpone its inclusion (if you ever considered)? Or at
least put "#ifdef __cygwin" (I hope this is the define) around it?

It just so ugly... And besides, GIT reportedly works without problems
for many people even without it.

Anyway, the patch is out, so anyone with the problems can just patch
their copy to workaround this specific win2k problem.

Thanks,
Alex

Re: First cut at git port to Cygwin

[Elfyn McBratney]

[-- Attachment #1: Type: text/plain, Size: 1011 bytes --]

On Sat, Oct 08, 2005 at 01:45:47AM +0200, Alex Riesen wrote:
 > Junio C Hamano, Fri, Oct 07, 2005 23:00:02 +0200:
 > > > "Sounds like a thinly veiled threat or a very effective prodding" 8)
 > > > ---
 > > >
 > > > Make read_cache copy the index into memory, to improve portability on
 > > > other OS's which have mmap too, tend to use it less commonly.
 > > >
 > > 
 > > Huh?  where is your memcpy?
 > > 
 > 
 > Junio, unless there already are pressing reasons to put the patch in
 > GIT, could you postpone its inclusion (if you ever considered)? Or at
 > least put "#ifdef __cygwin" (I hope this is the define) around it?

Close ;) - the define is "__CYGWIN__".

Best,
Elfyn

-- 
Elfyn McBratney
Gentoo Developer/Perl Team Lead
beu/irc.freenode.net                            http://dev.gentoo.org/~beu/
+------------O.o--------------------- http://dev.gentoo.org/~beu/pubkey.asc

PGP Key ID: 0x69DF17AD
PGP Key Fingerprint:
  DBD3 B756 ED58 B1B4 47B9  B3BD 8D41 E597 69DF 17AD

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: First cut at git port to Cygwin

[Linus Torvalds]

On Fri, 7 Oct 2005, Alex Riesen wrote:
> 
> Make read_cache copy the index into memory, to improve portability on
> other OS's which have mmap too, tend to use it less commonly.

I really think that you should just get rid of the mmap.

As it is, you're just slowing the code down on sane architectures. That's 
not good.

So I'd suggest something like this instead.

Totally untested, of course.

		Linus

----
diff --git a/read-cache.c b/read-cache.c
--- a/read-cache.c
+++ b/read-cache.c
@@ -454,13 +454,39 @@ static int verify_hdr(struct cache_heade
 	return 0;
 }
 
+static void *map_index_file(int fd, size_t size)
+{
+	void *map;
+#ifdef NO_MMAP
+	map = malloc(size);
+	if (!map)
+		die("Unable to allocate index file mapping");
+	if (read(fd, map, size) != size)
+		die("Unable to read %z bytes from inde
+#else
+	map = mmap(NULL, size, PROT_READ | PROT_WRITE, MAP_PRIVATE, fd, 0);
+	if (map == MAP_FAILED)
+		die("index file mmap failed (%s)", strerror(errno));
+#endif
+	return map;
+}
+
+static void unmap_index_file(void *map, size_t size)
+{
+#ifdef NO_MMAP
+	free(map);
+#else
+	munmap(map, size);
+#endif
+}
+
 int read_cache(void)
 {
 	int fd, i;
 	struct stat st;
 	unsigned long size, offset;
-	void *map;
 	struct cache_header *hdr;
+	void *map;
 
 	errno = EBUSY;
 	if (active_cache)
@@ -475,16 +501,15 @@ int read_cache(void)
 	}
 
 	size = 0; // avoid gcc warning
-	map = MAP_FAILED;
-	if (!fstat(fd, &st)) {
-		size = st.st_size;
-		errno = EINVAL;
-		if (size >= sizeof(struct cache_header) + 20)
-			map = mmap(NULL, size, PROT_READ | PROT_WRITE, MAP_PRIVATE, fd, 0);
-	}
+	if (fstat(fd, &st))
+		die("unable to fstat index file");
+
+	size = st.st_size;
+	errno = EINVAL;
+	if (size < sizeof(struct cache_header) + 20)
+		goto corrupt;
+	map = map_index_file(fd, size);
 	close(fd);
-	if (map == MAP_FAILED)
-		die("index file mmap failed (%s)", strerror(errno));
 
 	hdr = map;
 	if (verify_hdr(hdr, size) < 0)
@@ -503,8 +528,9 @@ int read_cache(void)
 	return active_nr;
 
 unmap:
-	munmap(map, size);
+	unmap_index_file(map, size);
 	errno = EINVAL;
+corrupt:
 	die("index file corrupt");
 }
 

Re: First cut at git port to Cygwin

[Elfyn McBratney]

[-- Attachment #1: Type: text/plain, Size: 2779 bytes --]

On Sat, Oct 08, 2005 at 09:11:03AM -0700, Linus Torvalds wrote:
 > 
 > On Fri, 7 Oct 2005, Alex Riesen wrote:
 > > 
 > > Make read_cache copy the index into memory, to improve portability on
 > > other OS's which have mmap too, tend to use it less commonly.
 > 
 > I really think that you should just get rid of the mmap.
 > 
 > As it is, you're just slowing the code down on sane architectures. That's 
 > not good.
 > 
 > So I'd suggest something like this instead.
 > 
 > Totally untested, of course.
 > 
 > 		Linus

Slightly adjusted diff below so it compiles ;)  (Note: only the second
die() un hunk #1 was changed.)

Best,
Elfyn

----
diff --git a/read-cache.c b/read-cache.c
--- a/read-cache.c
+++ b/read-cache.c
@@ -454,13 +454,39 @@ static int verify_hdr(struct cache_heade
 	return 0;
 }
 
+static void *map_index_file(int fd, size_t size)
+{
+	void *map;
+#ifdef NO_MMAP
+	map = malloc(size);
+	if (!map)
+		die("Unable to allocate index file mapping");
+	if (read(fd, map, size) != size)
+		die("Unable to read %z bytes from index", size);
+#else
+	map = mmap(NULL, size, PROT_READ | PROT_WRITE, MAP_PRIVATE, fd, 0);
+	if (map == MAP_FAILED)
+		die("index file mmap failed (%s)", strerror(errno));
+#endif
+	return map;
+}
+
+static void unmap_index_file(void *map, size_t size)
+{
+#ifdef NO_MMAP
+	free(map);
+#else
+	munmap(map, size);
+#endif
+}
+
 int read_cache(void)
 {
 	int fd, i;
 	struct stat st;
 	unsigned long size, offset;
-	void *map;
 	struct cache_header *hdr;
+	void *map;
 
 	errno = EBUSY;
 	if (active_cache)
@@ -475,16 +501,15 @@ int read_cache(void)
 	}
 
 	size = 0; // avoid gcc warning
-	map = MAP_FAILED;
-	if (!fstat(fd, &st)) {
-		size = st.st_size;
-		errno = EINVAL;
-		if (size >= sizeof(struct cache_header) + 20)
-			map = mmap(NULL, size, PROT_READ | PROT_WRITE, MAP_PRIVATE, fd, 0);
-	}
+	if (fstat(fd, &st))
+		die("unable to fstat index file");
+
+	size = st.st_size;
+	errno = EINVAL;
+	if (size < sizeof(struct cache_header) + 20)
+		goto corrupt;
+	map = map_index_file(fd, size);
 	close(fd);
-	if (map == MAP_FAILED)
-		die("index file mmap failed (%s)", strerror(errno));
 
 	hdr = map;
 	if (verify_hdr(hdr, size) < 0)
@@ -503,8 +528,9 @@ int read_cache(void)
 	return active_nr;
 
 unmap:
-	munmap(map, size);
+	unmap_index_file(map, size);
 	errno = EINVAL;
+corrupt:
 	die("index file corrupt");
 }


-- 
Elfyn McBratney
Gentoo Developer/Perl Team Lead
beu/irc.freenode.net                            http://dev.gentoo.org/~beu/
+------------O.o--------------------- http://dev.gentoo.org/~beu/pubkey.asc

PGP Key ID: 0x69DF17AD
PGP Key Fingerprint:
  DBD3 B756 ED58 B1B4 47B9  B3BD 8D41 E597 69DF 17AD

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: First cut at git port to Cygwin

[Elfyn McBratney]

[-- Attachment #1: Type: text/plain, Size: 424 bytes --]

Er, apologies for the dups - postfix crapped itself :/

*goes and stands in the corner donning the 'D' hat*

-- 
Elfyn McBratney
Gentoo Developer/Perl Team Lead
beu/irc.freenode.net                            http://dev.gentoo.org/~beu/
+------------O.o--------------------- http://dev.gentoo.org/~beu/pubkey.asc

PGP Key ID: 0x69DF17AD
PGP Key Fingerprint:
  DBD3 B756 ED58 B1B4 47B9  B3BD 8D41 E597 69DF 17AD

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: First cut at git port to Cygwin

[Johannes Schindelin]

Hi,

On Sat, 8 Oct 2005, Linus Torvalds wrote:

> I really think that you should just get rid of the mmap.
> 
> As it is, you're just slowing the code down on sane architectures. That's 
> not good.
> 
> So I'd suggest something like this instead.
> 
> Totally untested, of course.

Am I missing something? I don't see where the changes are written back to 
the fd. After all, mmap() is called with PROT_WRITE...

*shameless plug* Of course, this problem does not come up with my NO_MMAP 
patch.

Ciao,
Dscho

Re: First cut at git port to Cygwin

[Junio C Hamano]

Johannes Schindelin <Johannes.Schindelin@gmx.de> writes:

> Am I missing something? I don't see where the changes are written back to 
> the fd. After all, mmap() is called with PROT_WRITE...

PROT_WRITE is true, but we do MAP_PRIVATE, and if I recall
correctly we do not write file via mmap -- at least we do not
intend to.

 - index file is mapped for reading, long ago it was mapped
   read-only but these days we do PROT_WRITE, but updates are
   done via opening a new file and writing afresh.

 - objects are mapped for reading, but, never updated once
   created.  Creation side is regular open - write - close.

 - diff reads original by mapping, but obviously has no business
   writing.

 - local-fetch reads original by mapping for copying.

> *shameless plug* Of course, this problem does not come up with
> my NO_MMAP patch.

Yes.  It might have been overkill that you supported writing
changes back, though.
.  

Re: First cut at git port to Cygwin

[Alex Riesen]

Johannes Schindelin, Sat, Oct 08, 2005 20:27:06 +0200:
> > I really think that you should just get rid of the mmap.
> > 
> > As it is, you're just slowing the code down on sane architectures. That's 
> > not good.
> > 
> > So I'd suggest something like this instead.
> > 
> > Totally untested, of course.
> 
> Am I missing something? I don't see where the changes are written back to 
> the fd. After all, mmap() is called with PROT_WRITE...

It's just becase the file is open for reading only.
Also, it is not an mmap/unmap implementation. Just reading cache in.

Re: First cut at git port to Cygwin

[Johannes Schindelin]

Hi,

On Sat, 8 Oct 2005, Junio C Hamano wrote:

> Johannes Schindelin <Johannes.Schindelin@gmx.de> writes:
> 
> > Am I missing something? I don't see where the changes are written back to 
> > the fd. After all, mmap() is called with PROT_WRITE...
> 
> PROT_WRITE is true, but we do MAP_PRIVATE, and if I recall
> correctly we do not write file via mmap -- at least we do not
> intend to.

Ahh! Reading the man page helps!

> Yes.  It might have been overkill that you supported writing
> changes back, though.

Sure. Something like this?

diff --git a/compat/mmap.c b/compat/mmap.c
index fca6321..fda39fc 100644
--- a/compat/mmap.c
+++ b/compat/mmap.c
@@ -49,7 +49,7 @@ void *gitfakemmap(void *start, size_t le
 		n += count;
 	}
 
-	if(prot & PROT_WRITE) {
+	if((prot & PROT_WRITE) && !(flags & MAP_PRIVATE)) {
 		fakemmapwritable *next = xmalloc(sizeof(fakemmapwritable));
 		next->start = start;
 		next->length = length;

Re: First cut at git port to Cygwin

[Junio C Hamano]

Johannes Schindelin <Johannes.Schindelin@gmx.de> writes:

>> PROT_WRITE is true, but we do MAP_PRIVATE, and if I recall
>> correctly we do not write file via mmap -- at least we do not
>> intend to.
>
> Ahh! Reading the man page helps!
>
>> Yes.  It might have been overkill that you supported writing
>> changes back, though.
>
> Sure. Something like this?

Not, really.  What I meant was to rip out the writing out
altogether, and perhaps making sure that the caller never calls
us without MAP_PRIVATE.

Re: First cut at git port to Cygwin

[Johannes Schindelin]

Hi,

On Sat, 8 Oct 2005, Junio C Hamano wrote:

> Johannes Schindelin <Johannes.Schindelin@gmx.de> writes:
> 
> > Sure. Something like this?
> 
> Not, really.  What I meant was to rip out the writing out
> altogether, and perhaps making sure that the caller never calls
> us without MAP_PRIVATE.

How about this, then?

[PATCH] If NO_MMAP is defined, fake mmap() and munmap()

Since some platforms do not support mmap() at all, and others do only just so,
this patch introduces the option to fake mmap() and munmap() by malloc()ing the
region explicitely.

Signed-off-by: Johannes Schindelin <Johannes.Schindelin@gmx.de>

---

 Makefile      |    6 ++++++
 cache.h       |   16 ++++++++++++++++
 compat/mmap.c |   51 +++++++++++++++++++++++++++++++++++++++++++++++++++
 mailsplit.c   |    1 -
 4 files changed, 73 insertions(+), 1 deletions(-)
 create mode 100644 compat/mmap.c

applies-to: 274542bcbc891cca353c2728ac4075df3d1d2c0d
ed334e3e2276fe9d41ed78917544ef6a3fa87eb7
diff --git a/Makefile b/Makefile
index 1bdf4de..7ca77cf 100644
--- a/Makefile
+++ b/Makefile
@@ -27,6 +27,8 @@
 # Define NEEDS_SOCKET if linking with libc is not enough (SunOS,
 # Patrick Mauritz).
 #
+# Define NO_MMAP if you want to avoid mmap.
+#
 # Define WITH_OWN_SUBPROCESS_PY if you want to use with python 2.3.
 #
 # Define NO_IPV6 if you lack IPv6 support and getaddrinfo().
@@ -258,6 +260,10 @@ ifdef NO_STRCASESTR
 	DEFINES += -Dstrcasestr=gitstrcasestr
 	LIB_OBJS += compat/strcasestr.o
 endif
+ifdef NO_MMAP
+	DEFINES += -Dmmap=gitfakemmap -Dmunmap=gitfakemunmap -DNO_MMAP
+	LIB_OBJS += compat/mmap.o
+endif
 ifdef NO_IPV6
 	DEFINES += -DNO_IPV6 -Dsockaddr_storage=sockaddr_in
 endif
diff --git a/cache.h b/cache.h
index 514adb8..5987d4c 100644
--- a/cache.h
+++ b/cache.h
@@ -11,7 +11,9 @@
 #include <string.h>
 #include <errno.h>
 #include <limits.h>
+#ifndef NO_MMAP
 #include <sys/mman.h>
+#endif
 #include <sys/param.h>
 #include <netinet/in.h>
 #include <sys/types.h>
@@ -356,4 +358,18 @@ extern void packed_object_info_detail(st
 /* Dumb servers support */
 extern int update_server_info(int);
 
+#ifdef NO_MMAP
+
+#ifndef PROT_READ
+#define PROT_READ 1
+#define PROT_WRITE 2
+#define MAP_PRIVATE 1
+#define MAP_FAILED ((void*)-1)
+#endif
+
+extern void *gitfakemmap(void *start, size_t length, int prot , int flags, int fd, off_t offset);
+extern int gitfakemunmap(void *start, size_t length);
+
+#endif
+
 #endif /* CACHE_H */
diff --git a/compat/mmap.c b/compat/mmap.c
new file mode 100644
index 0000000..3f035a0
--- /dev/null
+++ b/compat/mmap.c
@@ -0,0 +1,51 @@
+#include <stdio.h>
+#include <stdlib.h>
+#include <unistd.h>
+#include <errno.h>
+#include "../cache.h"
+
+void *gitfakemmap(void *start, size_t length, int prot , int flags, int fd, off_t offset)
+{
+	int n = 0;
+
+	if(start != NULL || !(flags & MAP_PRIVATE))
+		die("Invalid usage of gitfakemmap.");
+
+	if(lseek(fd, offset, SEEK_SET)<0) {
+		errno = EINVAL;
+		return MAP_FAILED;
+	}
+
+	start = xmalloc(length);
+	if(start == NULL) {
+		errno = ENOMEM;
+		return MAP_FAILED;
+	}
+
+	while(n < length) {
+		int count = read(fd, start+n, length-n);
+
+		if(count == 0) {
+			memset(start+n, 0, length-n);
+			break;
+		}
+
+		if(count < 0) {
+			free(start);
+			errno = EACCES;
+			return MAP_FAILED;
+		}
+
+		n += count;
+	}
+
+	return start;
+}
+
+int gitfakemunmap(void *start, size_t length)
+{
+	free(start);
+
+	return 0;
+}
+
diff --git a/mailsplit.c b/mailsplit.c
index 7981f87..0f8100d 100644
--- a/mailsplit.c
+++ b/mailsplit.c
@@ -9,7 +9,6 @@
 #include <fcntl.h>
 #include <sys/types.h>
 #include <sys/stat.h>
-#include <sys/mman.h>
 #include <string.h>
 #include <stdio.h>
 #include <ctype.h>
---
0.99.8.GIT

Commit text BEFORE the dashes (Re: First cut at git port to Cygwin)

[Matthias Urlichs]

Hi, Alex Riesen wrote:

> [ some text ]
> ---
> [ the actual commit text ]

REMINDER: These need to be swapped.

-- 
Matthias Urlichs   |   {M:U} IT Design @ m-u-it.de   |  smurf@smurf.noris.de

Re: First cut at git port to Cygwin

[H. Peter Anvin]

Junio C Hamano wrote:
> 
> PROT_WRITE is true, but we do MAP_PRIVATE, and if I recall
> correctly we do not write file via mmap -- at least we do not
> intend to.
> 

Then PROT_READ probably makes more sense?

> 
> Yes.  It might have been overkill that you supported writing
> changes back, though.

Not just overkill; if we do MAP_PRIVATE it's actively WRONG.

	-hpa

Re: First cut at git port to Cygwin

[H. Peter Anvin]

Elfyn McBratney wrote:
>  > 
>  > Junio, unless there already are pressing reasons to put the patch in
>  > GIT, could you postpone its inclusion (if you ever considered)? Or at
>  > least put "#ifdef __cygwin" (I hope this is the define) around it?
> 
> Close ;) - the define is "__CYGWIN__".
> 

This should be a feature-control macro in the Makefile.

Re: First cut at git port to Cygwin

[Johannes Schindelin]

Hi,

On Mon, 10 Oct 2005, H. Peter Anvin wrote:

> Junio C Hamano wrote:
> > 
> > PROT_WRITE is true, but we do MAP_PRIVATE, and if I recall
> > correctly we do not write file via mmap -- at least we do not
> > intend to.
> > 
> 
> Then PROT_READ probably makes more sense?

Not necessarily. Sometimes you need to annotate the data from the index, 
and this does not need to be written back to the index file.

> > Yes.  It might have been overkill that you supported writing
> > changes back, though.
> 
> Not just overkill; if we do MAP_PRIVATE it's actively WRONG.

See above.

BTW, is there a mechanism to make sure that the index file is locked 
between reading and writing?

Ciao,
Dscho

Re: First cut at git port to Cygwin

[H. Peter Anvin]

Johannes Schindelin wrote:
> 
>>Junio C Hamano wrote:
>>
>>>PROT_WRITE is true, but we do MAP_PRIVATE, and if I recall
>>>correctly we do not write file via mmap -- at least we do not
>>>intend to.
>>>
>>
>>Then PROT_READ probably makes more sense?
> 
> Not necessarily. Sometimes you need to annotate the data from the index, 
> and this does not need to be written back to the index file.
> 

In the above sentence, emphasis on "at least we do not intend to."  If 
writes are done legitimately then that's fine, but we shouldn't have 
"accidental writes" -- those would be program bugs!

> 
>>>Yes.  It might have been overkill that you supported writing
>>>changes back, though.
>>
>>Not just overkill; if we do MAP_PRIVATE it's actively WRONG.
> 
> See above.
> 

Eh?  If we MAP_PRIVATE, *and* we (intentionally) write to it, we 
*BETTER* not write anything back.

	-hpa

Re: First cut at git port to Cygwin

[Johannes Schindelin]

Hi,

On Mon, 10 Oct 2005, H. Peter Anvin wrote:

> Johannes Schindelin wrote:
> > 
> > > Junio C Hamano wrote:
> > > 
> > > > PROT_WRITE is true, but we do MAP_PRIVATE, and if I recall
> > > > correctly we do not write file via mmap -- at least we do not
> > > > intend to.
> > > > 
> > > 
> > > Then PROT_READ probably makes more sense?
> > 
> > Not necessarily. Sometimes you need to annotate the data from the index, and
> > this does not need to be written back to the index file.
> > 
> 
> In the above sentence, emphasis on "at least we do not intend to."  If writes
> are done legitimately then that's fine, but we shouldn't have "accidental
> writes" -- those would be program bugs!

Yes, those would be bugs. However, if I understood the man page for mmap() 
correctly, then PROT_WRITE && MAP_PRIVATE makes the data copy-on-write, 
which means that those bugs would have been found (because the changes 
would no longer be present when git was called the next time). And I 
checked: all mmap() calls in git are MAP_PRIVATE.

> > > > Yes.  It might have been overkill that you supported writing
> > > > changes back, though.
> > > 
> > > Not just overkill; if we do MAP_PRIVATE it's actively WRONG.
> > 
> > See above.
> > 
> 
> Eh?  If we MAP_PRIVATE, *and* we (intentionally) write to it, we *BETTER* not
> write anything back.

Yes. That was *my* mistake.

Ciao,
Dscho

Re: First cut at git port to Cygwin

[Junio C Hamano]

"H. Peter Anvin" <hpa@zytor.com> writes:

> Eh?  If we MAP_PRIVATE, *and* we (intentionally) write to it, we 
> *BETTER* not write anything back.

Correct.

It has already been fixed by Johannes last week and I merged it
over the weekend if not earlier if I recall correctly.

Re: First cut at git port to Cygwin

[Daniel Barkalow]

On Mon, 10 Oct 2005, Johannes Schindelin wrote:

> Hi,
> 
> On Mon, 10 Oct 2005, H. Peter Anvin wrote:
> 
> > Junio C Hamano wrote:
> > > 
> > > PROT_WRITE is true, but we do MAP_PRIVATE, and if I recall
> > > correctly we do not write file via mmap -- at least we do not
> > > intend to.
> > > 
> > 
> > Then PROT_READ probably makes more sense?
> 
> Not necessarily. Sometimes you need to annotate the data from the index, 
> and this does not need to be written back to the index file.

In fact, it is intentional that we open the file O_RDONLY, and mmap it 
PROT_READ | PROT_WRITE, MAP_PRIVATE. We prepare the next index in the 
memory where we mapped the old index, but we don't want to change what's 
on the disk using the mapping; we write that later to a different file 
using write().

> > > Yes.  It might have been overkill that you supported writing
> > > changes back, though.
> > 
> > Not just overkill; if we do MAP_PRIVATE it's actively WRONG.
> 
> See above.
> 
> BTW, is there a mechanism to make sure that the index file is locked 
> between reading and writing?

There's definitely locking; the new file is written to "(filename).lock", 
which is openned O_CREAT | O_EXCL, and is moved to the destination when 
it's complete. I believe everything that intends to write a new index gets 
the lock before reading the old index, although I haven't actually 
checked.

	-Daniel
*This .sig left intentionally blank*

Re: First cut at git port to Cygwin

[Junio C Hamano]

"H. Peter Anvin" <hpa@zytor.com> writes:

>>>Junio C Hamano wrote:
>>>
>>>>PROT_WRITE is true, but we do MAP_PRIVATE, and if I recall
>>>>correctly we do not write file via mmap -- at least we do not
>>>>intend to.
>>>>
>
> In the above sentence, emphasis on "at least we do not intend to."  If 
> writes are done legitimately then that's fine, but we shouldn't have 
> "accidental writes" -- those would be program bugs!

What I meant to say was "we do not intend to write back the
changes by expecting the modification on mapped area are written
back by mmap() mechanism -- the updates to index file is done by
creat - write - close - rename".  So your saying "the overkill
being actively wrong" was technically correct, but that wrongly
written data was renamed out anyway and no real harm was done.

Re: First cut at git port to Cygwin

[H. Peter Anvin]

Junio C Hamano wrote:
> "H. Peter Anvin" <hpa@zytor.com> writes:
> 
> 
>>>>Junio C Hamano wrote:
>>>>
>>>>
>>>>>PROT_WRITE is true, but we do MAP_PRIVATE, and if I recall
>>>>>correctly we do not write file via mmap -- at least we do not
>>>>>intend to.
>>>>>
>>
>>In the above sentence, emphasis on "at least we do not intend to."  If 
>>writes are done legitimately then that's fine, but we shouldn't have 
>>"accidental writes" -- those would be program bugs!
> 
> 
> What I meant to say was "we do not intend to write back the
> changes by expecting the modification on mapped area are written
> back by mmap() mechanism -- the updates to index file is done by
> creat - write - close - rename".  So your saying "the overkill
> being actively wrong" was technically correct, but that wrongly
> written data was renamed out anyway and no real harm was done.

Well, it broke the atomicity of an operation, which *is* a real problem.

Anyway, malloc+read is a dead ringer for MAP_PRIVATE with PROT_WRITE, so 
that makes it even easier to mimic.

	-hpa