public inbox for archer@sourceware.org
 help / color / mirror / Atom feed
* Problem merging with git
@ 2009-02-18 15:49 Rick Moseley
  2009-02-18 16:11 ` Jan Kratochvil
  2009-02-18 16:44 ` Rick Moseley
  0 siblings, 2 replies; 5+ messages in thread
From: Rick Moseley @ 2009-02-18 15:49 UTC (permalink / raw)
  To: Project Archer

Hi all,

My git foo is obviously not what it should be.  I have been merging 
branches using the same basic commands the past few weeks.  First, I 
position myself in the local branch on my system that I want to merge 
with a remote branch.  Then I do a git-pull of the remote branch which 
then merges it with the local branch.  I am used to seeing messages like 
this:

Auto-merged gdb/windows-tdep.h
CONFLICT (rename/modify): Merge conflict in gdb/windows-tdep.h
Auto-merged COPYING.NEWLIB
CONFLICT (content): Merge conflict in COPYING.NEWLIB
Auto-merged ChangeLog
CONFLICT (content): Merge conflict in ChangeLog
 .
 .
 .


Then I go through and fix the conflicts and check them in after making 
sure it all builds and runs "make check" successfully.

Now I am trying to merge Jan's archer-jankratochvil-vla branch and I am 
seeing something different and I am wondering what the heck I am doing 
wrong.  Here is the sequence:

git-pull ssh://sources.redhat.com/git/archer.git archer-jankratochvil-vla
Enter passphrase for RSA key '/home/rmoseley/.ssh/identity':
 From ssh://sources.redhat.com/git/archer
 * branch            archer-jankratochvil-vla -> FETCH_HEAD
Auto-merged gdb/dwarf2read.c
Merge made by recursive.
 gdb/ada-lang.c                                  |   23 +-
 gdb/c-typeprint.c                               |    7 +-
 gdb/coffread.c                                  |    2 +-
 gdb/dwarf2expr.c                                |    7 +
 gdb/dwarf2expr.h                                |    2 +-
 gdb/dwarf2loc.c                                 |  213 ++++++-
 gdb/dwarf2loc.h                                 |    6 +
 gdb/dwarf2read.c                                |  419 ++++++++++----
 gdb/eval.c                                      |   51 ++-
 gdb/f-lang.h                                    |    4 +
 gdb/f-typeprint.c                               |   36 +-
 gdb/f-valprint.c                                |   59 ++-
 gdb/findvar.c                                   |  124 +++--
 gdb/gdbtypes.c                                  |  755 
+++++++++++++++++++----
 gdb/gdbtypes.h                                  |  150 ++++-
 gdb/jv-lang.c                                   |    2 +-
 gdb/mdebugread.c                                |    2 +-
 gdb/mi/mi-main.c                                |    1 +
 gdb/printcmd.c                                  |    5 +
 gdb/stabsread.c                                 |    4 +-
 gdb/testsuite/gdb.arch/x86_64-vla-typedef-foo.S |  455 ++++++++++++++
 gdb/testsuite/gdb.arch/x86_64-vla-typedef.c     |   43 ++
 gdb/testsuite/gdb.arch/x86_64-vla-typedef.exp   |   64 ++
 gdb/testsuite/gdb.base/vla-overflow.c           |   30 +
 gdb/testsuite/gdb.base/vla-overflow.exp         |  108 ++++
 gdb/testsuite/gdb.base/vla.c                    |   55 ++
 gdb/testsuite/gdb.base/vla.exp                  |   62 ++
 gdb/testsuite/gdb.dwarf2/dw2-stripped.c         |   42 ++
 gdb/testsuite/gdb.dwarf2/dw2-stripped.exp       |   79 +++
 gdb/testsuite/gdb.fortran/dynamic.exp           |  156 +++++
 gdb/testsuite/gdb.fortran/dynamic.f90           |   98 +++
 gdb/testsuite/gdb.fortran/string.exp            |   72 +++
 gdb/testsuite/gdb.fortran/string.f90            |   37 ++
 gdb/top.c                                       |    1 +
 gdb/typeprint.c                                 |   13 +-
 gdb/valarith.c                                  |   62 ++-
 gdb/valops.c                                    |   89 +++-
 gdb/value.c                                     |   58 ++-
 gdb/value.h                                     |    8 +-
 39 files changed, 2966 insertions(+), 438 deletions(-)
 create mode 100644 gdb/testsuite/gdb.arch/x86_64-vla-typedef-foo.S
 create mode 100644 gdb/testsuite/gdb.arch/x86_64-vla-typedef.c
 create mode 100644 gdb/testsuite/gdb.arch/x86_64-vla-typedef.exp
 create mode 100644 gdb/testsuite/gdb.base/vla-overflow.c
 create mode 100644 gdb/testsuite/gdb.base/vla-overflow.exp
 create mode 100644 gdb/testsuite/gdb.base/vla.c
 create mode 100644 gdb/testsuite/gdb.base/vla.exp
 create mode 100644 gdb/testsuite/gdb.dwarf2/dw2-stripped.c
 create mode 100644 gdb/testsuite/gdb.dwarf2/dw2-stripped.exp
 create mode 100644 gdb/testsuite/gdb.fortran/dynamic.exp
 create mode 100644 gdb/testsuite/gdb.fortran/dynamic.f90
 create mode 100644 gdb/testsuite/gdb.fortran/string.exp
 create mode 100644 gdb/testsuite/gdb.fortran/string.f90


After that, I do this:

git status
# On branch archer-rmoseley-f11-merge
nothing to commit (working directory clean)


git does not seem to acknowledge that files have changed.  I know they 
have because I have a pristine copy of archer-rmoseley-f11-merge checked 
out in a different directory and I do a diff and see the chanages.

Any ideas why git is not recognizing files have changed after the merge?

Thanks,

Rick

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: Problem merging with git
  2009-02-18 15:49 Problem merging with git Rick Moseley
@ 2009-02-18 16:11 ` Jan Kratochvil
  2009-02-18 16:15   ` Rick Moseley
  2009-02-18 16:17   ` Phil Muldoon
  2009-02-18 16:44 ` Rick Moseley
  1 sibling, 2 replies; 5+ messages in thread
From: Jan Kratochvil @ 2009-02-18 16:11 UTC (permalink / raw)
  To: Rick Moseley; +Cc: Project Archer, Phil Muldoon, Tom Tromey

On Wed, 18 Feb 2009 16:49:28 +0100, Rick Moseley wrote:
> I am used to seeing messages like this:
>
> Auto-merged COPYING.NEWLIB
> CONFLICT (content): Merge conflict in COPYING.NEWLIB

This should never happen as AFAIK no branch in the Archer project should
modify files like COPYING.NEWLIB.  If you see this message someone (who you
merge from) had to do something unclean regarding the branching/versioning.
IMO you should postpone such merge and ask the author to fixup his branch
first otherwise you start cluttering by needless conflict resolutions the
merge branch.

According to
	for i in `git branch -r|grep archer-`;do git diff origin/master...$i|grep -q COPYING.NEWLIB && echo $i;done
these are
	origin/archer-pmuldoon-pretty-printers-lookup
	origin/archer-tromey-python
	(sure just a consequence) origin/archer-rmoseley-fedora-merge

I am a GIT newbie so please correct me if diff
origin/master...origin/archer-BRANCH is not appropriate way how to show the
branch-specific changes against upstream ignoring any differences due to
possibly newer origin/master.


> Now I am trying to merge Jan's archer-jankratochvil-vla branch and I am  
> seeing something different and I am wondering what the heck I am doing  
> wrong.  Here is the sequence:

Just the merge with archer-jankratochvil-vla happened the way how every other
merge should happen in most cases.

...
> Auto-merged gdb/dwarf2read.c
> Merge made by recursive.
...
> After that, I do this:
>
> git status
> # On branch archer-rmoseley-f11-merge
> nothing to commit (working directory clean)
>
> git does not seem to acknowledge that files have changed.

The last merge is visible in `git log', isn't it?



Regards,
Jan

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: Problem merging with git
  2009-02-18 16:11 ` Jan Kratochvil
@ 2009-02-18 16:15   ` Rick Moseley
  2009-02-18 16:17   ` Phil Muldoon
  1 sibling, 0 replies; 5+ messages in thread
From: Rick Moseley @ 2009-02-18 16:15 UTC (permalink / raw)
  To: Jan Kratochvil; +Cc: Project Archer

Jan Kratochvil wrote:
> On Wed, 18 Feb 2009 16:49:28 +0100, Rick Moseley wrote:
>   
>> I am used to seeing messages like this:
>>
>> Auto-merged COPYING.NEWLIB
>> CONFLICT (content): Merge conflict in COPYING.NEWLIB
>>     
>
> This should never happen as AFAIK no branch in the Archer project should
> modify files like COPYING.NEWLIB.  If you see this message someone (who you
> merge from) had to do something unclean regarding the branching/versioning.
> IMO you should postpone such merge and ask the author to fixup his branch
> first otherwise you start cluttering by needless conflict resolutions the
> merge branch.
>
> According to
> 	for i in `git branch -r|grep archer-`;do git diff origin/master...$i|grep -q COPYING.NEWLIB && echo $i;done
> these are
> 	origin/archer-pmuldoon-pretty-printers-lookup
> 	origin/archer-tromey-python
> 	(sure just a consequence) origin/archer-rmoseley-fedora-merge
>
> I am a GIT newbie so please correct me if diff
> origin/master...origin/archer-BRANCH is not appropriate way how to show the
> branch-specific changes against upstream ignoring any differences due to
> possibly newer origin/master.
>
>
>   
>> Now I am trying to merge Jan's archer-jankratochvil-vla branch and I am  
>> seeing something different and I am wondering what the heck I am doing  
>> wrong.  Here is the sequence:
>>     
>
> Just the merge with archer-jankratochvil-vla happened the way how every other
> merge should happen in most cases.
>
> ...
>   
>> Auto-merged gdb/dwarf2read.c
>> Merge made by recursive.
>>     
> ...
>   
>> After that, I do this:
>>
>> git status
>> # On branch archer-rmoseley-f11-merge
>> nothing to commit (working directory clean)
>>
>> git does not seem to acknowledge that files have changed.
>>     
>
> The last merge is visible in `git log', isn't it?
>   

Yes, it appears to be:

 git log
commit 396b49611247265f167ebdfc78ff09aa896d1e3c
Merge: e947ab6... 9d03935...
Author: Rick Moseley <rmoseley@localhost.localdomain>
Date:   Wed Feb 18 09:18:30 2009 -0600

    Merge branch 'archer-jankratochvil-vla' of 
ssh://sources.redhat.com/git/archer into archer-rmoseley-f11-merge

commit e947ab663936483dc67424c153f67e78a910b96b
Author: Nick Clifton <nickc@redhat.com>
Date:   Thu Feb 12 15:06:54 2009 +0000

            PR 9827
            * elflink.c (bfd_elf_final_link): When counting the relocations,
            if the header size has not been set yet then assume that it will
            match the output section's reloc type.

commit ad88187efac2686ec0b15c558e05262a003ba504
Author: Jan Kratochvil <jan.kratochvil@redhat.com>
Date:   Thu Feb 12 09:15:06 2009 +0000

    gdb/
        PR fortran/9806
        * dwarf2read.c (process_die <DW_TAG_module>, read_module)
        (scan_partial_symbols <DW_TAG_module>, add_partial_module): New.
   
    gdb/testsuite/
        PR fortran/9806
        * gdb.fortran/module.exp, gdb.fortran/module.f90: New.

commit 57f9b3e3aca4b09da7dee45bba37a60653de27ee
Author: Nathan Sidwell <nathan@codesourcery.com>
Date:   Thu Feb 12 08:31:02 2009 +0000

        gas/
        * config/tc-m68k.c (mcf51qe_ctrl): Add CPUCR.
        (mcf52259_ctrl, mcf52277_ctrl, mcf53017_ctrl): New.
        (mcf5307_ctrl): Add VBR.
        (no_mac): New variable.
        (m68k_extensions): Refer to no_mac mask.
 .
 .
 .
>
>
> Regards,
> Jan
>   

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: Problem merging with git
  2009-02-18 16:11 ` Jan Kratochvil
  2009-02-18 16:15   ` Rick Moseley
@ 2009-02-18 16:17   ` Phil Muldoon
  1 sibling, 0 replies; 5+ messages in thread
From: Phil Muldoon @ 2009-02-18 16:17 UTC (permalink / raw)
  To: Jan Kratochvil; +Cc: Rick Moseley, Project Archer, Tom Tromey

Jan Kratochvil wrote:

Answering as I was CC'd:
> According to
> 	for i in `git branch -r|grep archer-`;do git diff origin/master...$i|grep -q COPYING.NEWLIB && echo $i;done
> these are
> 	origin/archer-pmuldoon-pretty-printers-lookup
>   

The parent of: archer-pmuldoon-pretty-printers-lookup  is 
archer-tromey-python so I suspect that is why you see the 
archer-pmuldoon-pretty-printers-lookup here..

Anyway it (archer-pmuldoon-pretty...) is not included in the current 
merge with the patch-set that Rick is working with (this branch will 
eventually die and the changes merge into the archer-tromey-python branch).

Regards

Phil

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: Problem merging with git
  2009-02-18 15:49 Problem merging with git Rick Moseley
  2009-02-18 16:11 ` Jan Kratochvil
@ 2009-02-18 16:44 ` Rick Moseley
  1 sibling, 0 replies; 5+ messages in thread
From: Rick Moseley @ 2009-02-18 16:44 UTC (permalink / raw)
  To: Project Archer

Rick Moseley wrote:
> Hi all,
>
> My git foo is obviously not what it should be.  I have been merging 
> branches using the same basic commands the past few weeks.  First, I 
> position myself in the local branch on my system that I want to merge 
> with a remote branch.  Then I do a git-pull of the remote branch which 
> then merges it with the local branch.  I am used to seeing messages 
> like this:
>
> Auto-merged gdb/windows-tdep.h
> CONFLICT (rename/modify): Merge conflict in gdb/windows-tdep.h
> Auto-merged COPYING.NEWLIB
> CONFLICT (content): Merge conflict in COPYING.NEWLIB
> Auto-merged ChangeLog
> CONFLICT (content): Merge conflict in ChangeLog
> .
> .
> .
>
>
> Then I go through and fix the conflicts and check them in after making 
> sure it all builds and runs "make check" successfully.
>
> Now I am trying to merge Jan's archer-jankratochvil-vla branch and I 
> am seeing something different and I am wondering what the heck I am 
> doing wrong.  Here is the sequence:
>
> git-pull ssh://sources.redhat.com/git/archer.git archer-jankratochvil-vla
> Enter passphrase for RSA key '/home/rmoseley/.ssh/identity':
> From ssh://sources.redhat.com/git/archer
> * branch            archer-jankratochvil-vla -> FETCH_HEAD
> Auto-merged gdb/dwarf2read.c
> Merge made by recursive.
> gdb/ada-lang.c                                  |   23 +-
> gdb/c-typeprint.c                               |    7 +-
> gdb/coffread.c                                  |    2 +-
> gdb/dwarf2expr.c                                |    7 +
> gdb/dwarf2expr.h                                |    2 +-
> gdb/dwarf2loc.c                                 |  213 ++++++-
> gdb/dwarf2loc.h                                 |    6 +
> gdb/dwarf2read.c                                |  419 ++++++++++----
> gdb/eval.c                                      |   51 ++-
> gdb/f-lang.h                                    |    4 +
> gdb/f-typeprint.c                               |   36 +-
> gdb/f-valprint.c                                |   59 ++-
> gdb/findvar.c                                   |  124 +++--
> gdb/gdbtypes.c                                  |  755 
> +++++++++++++++++++----
> gdb/gdbtypes.h                                  |  150 ++++-
> gdb/jv-lang.c                                   |    2 +-
> gdb/mdebugread.c                                |    2 +-
> gdb/mi/mi-main.c                                |    1 +
> gdb/printcmd.c                                  |    5 +
> gdb/stabsread.c                                 |    4 +-
> gdb/testsuite/gdb.arch/x86_64-vla-typedef-foo.S |  455 ++++++++++++++
> gdb/testsuite/gdb.arch/x86_64-vla-typedef.c     |   43 ++
> gdb/testsuite/gdb.arch/x86_64-vla-typedef.exp   |   64 ++
> gdb/testsuite/gdb.base/vla-overflow.c           |   30 +
> gdb/testsuite/gdb.base/vla-overflow.exp         |  108 ++++
> gdb/testsuite/gdb.base/vla.c                    |   55 ++
> gdb/testsuite/gdb.base/vla.exp                  |   62 ++
> gdb/testsuite/gdb.dwarf2/dw2-stripped.c         |   42 ++
> gdb/testsuite/gdb.dwarf2/dw2-stripped.exp       |   79 +++
> gdb/testsuite/gdb.fortran/dynamic.exp           |  156 +++++
> gdb/testsuite/gdb.fortran/dynamic.f90           |   98 +++
> gdb/testsuite/gdb.fortran/string.exp            |   72 +++
> gdb/testsuite/gdb.fortran/string.f90            |   37 ++
> gdb/top.c                                       |    1 +
> gdb/typeprint.c                                 |   13 +-
> gdb/valarith.c                                  |   62 ++-
> gdb/valops.c                                    |   89 +++-
> gdb/value.c                                     |   58 ++-
> gdb/value.h                                     |    8 +-
> 39 files changed, 2966 insertions(+), 438 deletions(-)
> create mode 100644 gdb/testsuite/gdb.arch/x86_64-vla-typedef-foo.S
> create mode 100644 gdb/testsuite/gdb.arch/x86_64-vla-typedef.c
> create mode 100644 gdb/testsuite/gdb.arch/x86_64-vla-typedef.exp
> create mode 100644 gdb/testsuite/gdb.base/vla-overflow.c
> create mode 100644 gdb/testsuite/gdb.base/vla-overflow.exp
> create mode 100644 gdb/testsuite/gdb.base/vla.c
> create mode 100644 gdb/testsuite/gdb.base/vla.exp
> create mode 100644 gdb/testsuite/gdb.dwarf2/dw2-stripped.c
> create mode 100644 gdb/testsuite/gdb.dwarf2/dw2-stripped.exp
> create mode 100644 gdb/testsuite/gdb.fortran/dynamic.exp
> create mode 100644 gdb/testsuite/gdb.fortran/dynamic.f90
> create mode 100644 gdb/testsuite/gdb.fortran/string.exp
> create mode 100644 gdb/testsuite/gdb.fortran/string.f90
>
>
> After that, I do this:
>
> git status
> # On branch archer-rmoseley-f11-merge
> nothing to commit (working directory clean)
>
>
> git does not seem to acknowledge that files have changed.  I know they 
> have because I have a pristine copy of archer-rmoseley-f11-merge 
> checked out in a different directory and I do a diff and see the 
> chanages.
>
> Any ideas why git is not recognizing files have changed after the merge?
>
> Thanks,
>
> Rick
OK, I got the answer from Tom and Jan on our weekly call.  All of my 
previous commits had merge conflicts, so git did not commit the 
changes.  Since this time there were no conflicts, git went ahead and 
committed the merges so no "git commit" was necessary.

Thanks,

Rick

PS:  Thanks Jan for the great branch with no merge conflicts.  I needed 
to experience that at least once.   :)

^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2009-02-18 16:44 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2009-02-18 15:49 Problem merging with git Rick Moseley
2009-02-18 16:11 ` Jan Kratochvil
2009-02-18 16:15   ` Rick Moseley
2009-02-18 16:17   ` Phil Muldoon
2009-02-18 16:44 ` Rick Moseley

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).