On Fri, May 20, 2011 at 4:28 AM, Michael Olbrich
<[email protected]>wrote:

> On Mon, May 16, 2011 at 12:55:18PM -0400, Jon Ringle wrote:
> > -From db9bd703a9a696bd73cb7d41e98c66854a604477 Mon Sep 17 00:00:00 2001
> > +From 1f2c7532788626e668f02a5783388839e1ea97d6 Mon Sep 17 00:00:00 2001
> >  From: Karel Zak <[email protected]>
> >  Date: Fri, 11 Feb 2011 10:45:09 +0100
> > -Subject: [PATCH] libmount: add fallbacks for old systems without
> umount2() syscall
> > +Subject: [PATCH 1/5] libmount: add fallbacks for old systems without
> umount2() syscall
> >
>
> Don't change this patch. And use 'git format-patch -N' to avoid x/y in the
> subject. It doesn't match anyways.
>
> I've now used git ptx-patches which using -N


> > +From: Jon Ringle <[email protected]>
> > +Date: Tue, 26 Apr 2011 02:15:50 -0400
> > +Subject: [PATCH 4/5] replace index with strchr
> > +
>
> add a comment (see my other mails).
>
> Ok.


> > +From 4b6cf4859a0798eef6f59933101f08f950a0aa9b Mon Sep 17 00:00:00 2001
> > +From: Karel Zak <[email protected]>
> > +Date: Fri, 11 Feb 2011 10:18:00 +0100
> > +Subject: [PATCH] libmount: replace mkostmps() with more portable
> mkstemp()
> > +
> > +Reported-by: Peter Breitenlohner <[email protected]>
> > +Signed-off-by: Karel Zak <[email protected]>
> > +---
> > + shlibs/mount/src/mountP.h     |    2 +-
> > + shlibs/mount/src/tab_update.c |    2 +-
> > + shlibs/mount/src/utils.c      |    4 ++--
> > + 3 files changed, 4 insertions(+), 4 deletions(-)
> > +
> > +diff --git a/shlibs/mount/src/mountP.h b/shlibs/mount/src/mountP.h
> > +index aeceb7f..71314a5 100644
> > +--- a/shlibs/mount/src/mountP.h
> > ++++ b/shlibs/mount/src/mountP.h
> > +@@ -118,7 +118,7 @@ extern int mnt_in_group(gid_t gid);
> > +
> > + extern char *mnt_get_mountpoint(const char *path);
> > + extern char *mnt_get_fs_root(const char *path, const char *mountpoint);
> > +-extern int mnt_open_uniq_filename(const char *filename, char **name,
> int flags);
> > ++extern int mnt_open_uniq_filename(const char *filename, char **name);
> > + extern int mnt_has_regular_utab(const char **utab, int *writable);
> > + extern const char *mnt_get_utab_path(void);
> > +
> > +diff --git a/shlibs/mount/src/tab_update.c
> b/shlibs/mount/src/tab_update.c
> > +index a0f99f1..6b7b4be 100644
> > +--- a/shlibs/mount/src/tab_update.c
> > ++++ b/shlibs/mount/src/tab_update.c
> > +@@ -561,7 +561,7 @@ static int update_table(struct libmnt_update *upd,
> struct libmnt_table *tb)
> > +
> > +     DBG(UPDATE, mnt_debug_h(upd, "%s: updating", upd->filename));
> > +
> > +-    fd = mnt_open_uniq_filename(upd->filename, &uq, O_WRONLY);
> > ++    fd = mnt_open_uniq_filename(upd->filename, &uq);
> > +     if (fd < 0)
> > +             return fd;      /* error */
> > +
> > +diff --git a/shlibs/mount/src/utils.c b/shlibs/mount/src/utils.c
> > +index cb0ed1c..b02b6fe 100644
> > +--- a/shlibs/mount/src/utils.c
> > ++++ b/shlibs/mount/src/utils.c
> > +@@ -642,7 +642,7 @@ const char *mnt_get_utab_path(void)
> > +
> > + /* returns file descriptor or -errno, @name returns uniques filename
> > +  */
> > +-int mnt_open_uniq_filename(const char *filename, char **name, int
> flags)
> > ++int mnt_open_uniq_filename(const char *filename, char **name)
> > + {
> > +     int rc, fd;
> > +     char *n;
> > +@@ -656,7 +656,7 @@ int mnt_open_uniq_filename(const char *filename,
> char **name, int flags)
> > +     if (rc <= 0)
> > +             return -errno;
> > +
> > +-    fd = mkostemp(n, flags | O_EXCL);
> > ++    fd = mkstemp(n);
>
> I'm not sure what the consequences of this are. Please keep the function
> parameters, and maybe do something like this:
>
> #ifdef <this-is-glibc>
>         fd = mkostemp(n, flags | O_EXCL);
> #else
>        fd = mkstemp(n);
> #endif
>
> not sure what the correct macro would be.
>
> This is an upstream patch and would rather not change it.


> > +From 0198bb456b9a35a8b973f2978883cadc443499a3 Mon Sep 17 00:00:00 2001
> > +From: Karel Zak <[email protected]>
> > +Date: Fri, 11 Feb 2011 09:06:27 +0100
> > +Subject: [PATCH] partx, lsblk: fix gettext calls
> > +
>
> this is from upstream, right?
>
> Yes.

New patch coming.

Jon
-- 
ptxdist mailing list
[email protected]

Reply via email to