| Message ID | 20141022194408.GA20989@pobox.suse.cz (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
On Wed, 22 Oct 2014 21:44:08 +0200 Michal Marek <mmarek@suse.cz> wrote: > Dne 22.10.2014 v 16:19 Steven Rostedt napsal(a): > > > > Commit 7ff525712acf "kbuild: fake the "Entering directory ..." message > > more simply" changed the output of "make kernelrelease" such that the > > kernel release version was not the last line printed. This broke various > > tools that would find the kernel release with "make kernelrelease | tail -1". > > The cleaner and recommended (see recent make help) way is to use make -s: > > $ make O=build -s kernelrelease > 3.18.0-rc1+ Hmm, I see the help index was recently updated to include that. Does this work with older kernels too? That's very important. -- Steve -- To unsubscribe from this list: send the line "unsubscribe linux-kbuild" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On Wed, 22 Oct 2014 21:44:08 +0200 Michal Marek <mmarek@suse.cz> wrote: > Dne 22.10.2014 v 16:19 Steven Rostedt napsal(a): > > > > Commit 7ff525712acf "kbuild: fake the "Entering directory ..." message > > more simply" changed the output of "make kernelrelease" such that the > > kernel release version was not the last line printed. This broke various > > tools that would find the kernel release with "make kernelrelease | tail -1". > > The cleaner and recommended (see recent make help) way is to use make -s: > > $ make O=build -s kernelrelease > 3.18.0-rc1+ > > no further processing is needed. I don't mind changing my script with your patch. But this does break other scripts of mine that I need to hunt down and change. My fear is why do we need the '-s' when there may be other methods to preserve the old functionality. If this is breaking my scripts, I wonder if it is also breaking other scripts out there too? As you stated, it wasn't until recently that the help message recommended the '-s'. -- Steve > > > > One of those tools that broke was ktest.pl which resides in the kernel. > > Can you please apply this patch? > > Thanks, > Michal > > >From c660b235e25eee053337e0e6c952e87f39839c63 Mon Sep 17 00:00:00 2001 > From: Michal Marek <mmarek@suse.cz> > Date: Wed, 22 Oct 2014 21:25:39 +0200 > Subject: [PATCH] ktest: Use make -s kernelrelease > > The previous tail -1 broke with commit 7ff525712acf ("kbuild: fake the > "Entering directory ..." message more simply") > > Reported-by: Steven Rostedt <rostedt@goodmis.org> > Signed-off-by: Michal Marek <mmarek@suse.cz> > --- > tools/testing/ktest/ktest.pl | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/tools/testing/ktest/ktest.pl b/tools/testing/ktest/ktest.pl > index bf13981..60fe020 100755 > --- a/tools/testing/ktest/ktest.pl > +++ b/tools/testing/ktest/ktest.pl > @@ -2005,7 +2005,7 @@ sub get_version { > # get the release name > return if ($have_version); > doprint "$make kernelrelease ... "; > - $version = `$make kernelrelease | tail -1`; > + $version = `$make -s kernelrelease`; > chomp($version); > doprint "$version\n"; > $have_version = 1; -- To unsubscribe from this list: send the line "unsubscribe linux-kbuild" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On 2014-10-22 22:35, Steven Rostedt wrote: > On Wed, 22 Oct 2014 21:44:08 +0200 > Michal Marek <mmarek@suse.cz> wrote: > >> Dne 22.10.2014 v 16:19 Steven Rostedt napsal(a): >>> >>> Commit 7ff525712acf "kbuild: fake the "Entering directory ..." message >>> more simply" changed the output of "make kernelrelease" such that the >>> kernel release version was not the last line printed. This broke various >>> tools that would find the kernel release with "make kernelrelease | tail -1". >> >> The cleaner and recommended (see recent make help) way is to use make -s: >> >> $ make O=build -s kernelrelease >> 3.18.0-rc1+ > > Hmm, I see the help index was recently updated to include that. > > Does this work with older kernels too? That's very important. Yes. Except for a few bugs in the history, make -s has always been silent. Michal -- To unsubscribe from this list: send the line "unsubscribe linux-kbuild" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On 2014-10-23 03:11, Steven Rostedt wrote: > On Wed, 22 Oct 2014 21:44:08 +0200 > Michal Marek <mmarek@suse.cz> wrote: > >> Dne 22.10.2014 v 16:19 Steven Rostedt napsal(a): >>> >>> Commit 7ff525712acf "kbuild: fake the "Entering directory ..." message >>> more simply" changed the output of "make kernelrelease" such that the >>> kernel release version was not the last line printed. This broke various >>> tools that would find the kernel release with "make kernelrelease | tail -1". >> >> The cleaner and recommended (see recent make help) way is to use make -s: >> >> $ make O=build -s kernelrelease >> 3.18.0-rc1+ >> >> no further processing is needed. > > > I don't mind changing my script with your patch. But this does break > other scripts of mine that I need to hunt down and change. My fear is > why do we need the '-s' when there may be other methods to preserve the > old functionality. If you have scripts that only do x=$(make kernelrelease), then you should definitely fix them. You are right that the make kernelrelease | tail -1 trick had always been working until 7ff525712acf, so I can apply your patch. But please fix ktest.pl so that it's consistent with what the help text says. Also, the Makefile patch should cover kernelversion and image_name as well. Michal -- To unsubscribe from this list: send the line "unsubscribe linux-kbuild" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On Thu, 23 Oct 2014 10:57:57 +0200 Michal Marek <mmarek@suse.cz> wrote: > On 2014-10-23 03:11, Steven Rostedt wrote: > > On Wed, 22 Oct 2014 21:44:08 +0200 > > Michal Marek <mmarek@suse.cz> wrote: > > > >> Dne 22.10.2014 v 16:19 Steven Rostedt napsal(a): > >>> > >>> Commit 7ff525712acf "kbuild: fake the "Entering directory ..." message > >>> more simply" changed the output of "make kernelrelease" such that the > >>> kernel release version was not the last line printed. This broke various > >>> tools that would find the kernel release with "make kernelrelease | tail -1". > >> > >> The cleaner and recommended (see recent make help) way is to use make -s: > >> > >> $ make O=build -s kernelrelease > >> 3.18.0-rc1+ > >> > >> no further processing is needed. > > > > > > I don't mind changing my script with your patch. But this does break > > other scripts of mine that I need to hunt down and change. My fear is > > why do we need the '-s' when there may be other methods to preserve the > > old functionality. > > If you have scripts that only do x=$(make kernelrelease), then you > should definitely fix them. You are right that the make kernelrelease | > tail -1 trick had always been working until 7ff525712acf, so I can apply > your patch. Thanks, because that's what my scripts indeed do ;-) > But please fix ktest.pl so that it's consistent with what > the help text says. Yeah, regardless I was planning on applying your patch as it needs to handle bisects, and without it, ktest will fail to bisect after 7ff525712acf. > Also, the Makefile patch should cover kernelversion > and image_name as well. OK, I'll make the update. Thanks! -- Steve -- To unsubscribe from this list: send the line "unsubscribe linux-kbuild" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
diff --git a/tools/testing/ktest/ktest.pl b/tools/testing/ktest/ktest.pl index bf13981..60fe020 100755 --- a/tools/testing/ktest/ktest.pl +++ b/tools/testing/ktest/ktest.pl @@ -2005,7 +2005,7 @@ sub get_version { # get the release name return if ($have_version); doprint "$make kernelrelease ... "; - $version = `$make kernelrelease | tail -1`; + $version = `$make -s kernelrelease`; chomp($version); doprint "$version\n"; $have_version = 1;