Sent from my BlackBerry 10 smartphone on the Rogers network.
Original Message
From: René Scharfe
Sent: Tuesday, October 28, 2014 16:59
To: Git Mailing List
Cc: Junio C Hamano
Subject: [PATCH] use child_process_init() to initialize struct child_process
variables
Call child_process_init() ins
> If I remove the call to pthread_create, it doesn't output anything and
> exits successfully.
I see. Most resource managers use procmgr_daemon(), which has no such
limitation. Anyway, as far as I can see current git sources do not use
fork together with pthread, except for daemonize() function.
-
> The test does _not_ fail. That if condition does return -1 on Linux
> and BSD, and making tm_to_time_t() return a failure, but the caller
> goes on, ending up with the right values in year/month/date in the
> tm struct, which is the primary thing the function is used for.
I said it wrong, test
> I don't think it's a good idea to just enable thread support. On QNX,
> once a process creates a thread, fork stops working. This breaks
> commands that create threads and then try to run other programs, such
> as "git fetch" with an https remote. If threads are enabled, I think
> that the use
> I saw Matt has comment on this patch, so I'll keep the patch out of
> 'next' for now and let you two figure it out.
Anyway in current form this patch is broken. Junio, may I ask you
about the rest patches in a separate posts, have you applied any of
them?
Thanks.
--
To unsubscribe from this lis
> Is there a point to the version checking? I don't know that anyone
> has tried to build Git on QNX 4, so adding a case for it seems
> misleading.
getpagesize() was introduced in QNX 6.4.1, it is present in QNX 6.5.0
also. So at least for this version checking is requied.
> I didn't realize tha
Fix time offset calculation expression in case if time_t
is unsigned. This code works fine for signed and
unsigned time_t.
Signed-off-by: Mike Gorchak
---
date.c | 11 +--
1 file changed, 9 insertions(+), 2 deletions(-)
diff --git a/date.c b/date.c
index 1ac28e5..11ee7b4 100644
--- a
Do not compare time_t (less comparison) with -1. If time_t
is unsigned this leads to always true comparison.
Signed-off-by: Mike Gorchak
---
date.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/date.c b/date.c
index 57331ed..1ac28e5 100644
--- a/date.c
+++ b/date.c
>> So is_date() always return negative result for the text string where
>> date is placed before time like '2008-02-14 20:30:45'.
> Yes, it returns this -1 on other platforms, but...
>> It must fail on
>> other platforms as well.
It also fails under Linux, but real problem is not here, it is just
> The thing that puzzles me is that nobody reported that the following
> fail on their platforms (and they do not fail for me on platforms I
> have to test in my real/virtual boxes).
Ok, check_parse calls function parse_date(), it calls
parse_date_basic(), where following code is present:
>> Fix is_date() function failings in detection of correct date in case
>> if time was not properly initialized.
>
> Please explain why this patch is needed and what problem this patch
> is trying to fix (if any) a bit better in the proposed log message.
> For example, on what input do we call this
Add detection of getpagesize() function in libc. Declare
empty "NO_GETPAGESIZE" macro in case if getpagesize()
exists and "NO_GETPAGESIZE=YesPlease" if no.
Signed-off-by: Mike Gorchak
---
configure.ac | 5 +
1 file changed, 5 insertions(+)
diff --git a/configure.ac
> Also, please leave the "autodetection" out. If it is common to have
> strcasestr (or any other) on a newer QNX, then not defining the
> symbol NO_STRCASESTR in this file may still be the right thing to
> do, but the justification for such a change should not be because we
> rely on "autodetectio
Add pthread support in QNX. Do not declare NO_ macros if they can be
autodetected.
From: Mike Gorchak
Signed-off-by: Mike Gorchak
---
config.mak.uname | 22 +++---
1 file changed, 15 insertions(+), 7 deletions(-)
diff --git a/config.mak.uname b/config.mak.uname
index 8743a6d
Fix is_date() function failings in detection of correct date in case
if time was not properly initialized.
From: Mike Gorchak
Signed-off-by: Mike Gorchak
---
date.c | 12 +++-
1 file changed, 11 insertions(+), 1 deletion(-)
diff --git a/date.c b/date.c
index 57331ed..ec758f4 100644
Hi Junio,
> Swapping the order between CFLAGS and BASIC_CFLAGS in ALL_CFLAGS may
> be a good change for that reason as well.
This sounds very reasonable.
> In any case, I won't take a patch to rename source files left and
> right only to work around name collisions with random system header
> fi
Fix git compilation without available libiconv library.
From: Mike Gorchak
Signed-off-by: Mike Gorchak
---
configure.ac | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/configure.ac b/configure.ac
index 1991258..d0e82c1 100644
--- a/configure.ac
+++ b/configure.ac
>>> 2) Rename fnmatch.h to fnmatch_gnu.h and fnmatch.c to fnmatch_gnu.c to
>>> avoid inclusion of system fnmatch.h header in case if -I/usr/include
>>> path is specified before -Icompat/fnmatch. Which is common situation.
> As to the substance, I am fairly negative about the approach this
> patch t
Hello,
> Here is a small patch with QNX build improvements:
>
> 1) Rename tar.h to tar_git.h. Latest QNX versions have system tar.h
> header according to
> http://pubs.opengroup.org/onlinepubs/009696699/basedefs/tar.h.html ,
> to avoid inclusion of another tar.h, original header was renamed.
> 2)
Hi,
Here is a small patch with QNX build improvements:
1) Rename tar.h to tar_git.h. Latest QNX versions have system tar.h
header according to
http://pubs.opengroup.org/onlinepubs/009696699/basedefs/tar.h.html ,
to avoid inclusion of another tar.h, original header was renamed.
2) Rename fnmatch.h
20 matches
Mail list logo