[PATCH 1/4] Makefile: rules for printing kernel architecture and localversion

[PATCH 1/4] Makefile: rules for printing kernel architecture and localversion

[Enrico Weigelt, metux IT consult]

trivial rule to print out the kernel arch and localversion, so
external tools, like distro packagers, can easily get it.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
---
 Makefile | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/Makefile b/Makefile
index 3e4868a..5afc3de 100644
--- a/Makefile
+++ b/Makefile
@@ -1706,6 +1706,12 @@ kernelrelease:
 kernelversion:
 	@echo $(KERNELVERSION)
 
+kernellocalversion:
+	@$(CONFIG_SHELL) $(srctree)/scripts/setlocalversion $(srctree) | sed -e 's~^\-~~'
+
+kernelarch:
+	@echo $(ARCH)
+
 image_name:
 	@echo $(KBUILD_IMAGE)
 
-- 
1.9.1


_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

[PATCH 2/4] scripts: mkdebian: allow renaming generated debian/rules via env

[Enrico Weigelt, metux IT consult]

Add new environment variable KDEB_RULES for controlling where the
generated debian rules are written to. By defaults, it's debian/rules,
but packagers might override it for providing their own rules file.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
---
 scripts/package/mkdebian | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/scripts/package/mkdebian b/scripts/package/mkdebian
index 8351584..7a9ca4e 100755
--- a/scripts/package/mkdebian
+++ b/scripts/package/mkdebian
@@ -99,6 +99,9 @@ kernel_headers_packagename=linux-headers-$version
 dbg_packagename=$packagename-dbg
 debarch=
 set_debarch
+if [ -z "$KDEB_RULES" ]; then
+    KDEB_RULES=debian/rules
+fi
 
 if [ "$ARCH" = "um" ] ; then
 	packagename=user-mode-linux-$version
@@ -206,7 +209,7 @@ Description: Linux kernel debugging symbols for $version
  all the necessary debug symbols for the kernel and its modules.
 EOF
 
-cat <<EOF > debian/rules
+cat <<EOF > $KDEB_RULES
 #!$(command -v $MAKE) -f
 
 srctree ?= .
-- 
1.9.1


_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

[PATCH 3/4] scripts: checkpatch.pl: don't complain that debian/rules is executable

[Enrico Weigelt, metux IT consult]

checkpatch.pl complains when adding executable "debian/rules",
obviously a false alarm. Therefore add an exception for that.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
---
 scripts/checkpatch.pl | 1 +
 1 file changed, 1 insertion(+)

diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
index 342c7c7..caf0146 100755
--- a/scripts/checkpatch.pl
+++ b/scripts/checkpatch.pl
@@ -2587,6 +2587,7 @@ sub process {
 		if ($line =~ /^new (file )?mode.*[7531]\d{0,2}$/) {
 			my $permhere = $here . "FILE: $realfile\n";
 			if ($realfile !~ m@scripts/@ &&
+			    $realfile !~ "debian/rules" &&
 			    $realfile !~ /\.(py|pl|awk|sh)$/) {
 				ERROR("EXECUTE_PERMISSIONS",
 				      "do not set execute permissions for source files\n" . $permhere);
-- 
1.9.1


_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

Re: [PATCH 3/4] scripts: checkpatch.pl: don't complain that debian/rules is executable

[Joe Perches]

On Tue, 2019-07-09 at 11:32 +0200, Enrico Weigelt, metux IT consult
wrote:
> checkpatch.pl complains when adding executable "debian/rules",
> obviously a false alarm. Therefore add an exception for that.
> 
> Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
> ---
>  scripts/checkpatch.pl | 1 +
>  1 file changed, 1 insertion(+)
> 
> diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
> index 342c7c7..caf0146 100755
> --- a/scripts/checkpatch.pl
> +++ b/scripts/checkpatch.pl
> @@ -2587,6 +2587,7 @@ sub process {
>  		if ($line =~ /^new (file )?mode.*[7531]\d{0,2}$/) {
>  			my $permhere = $here . "FILE: $realfile\n";
>  			if ($realfile !~ m@scripts/@ &&
> +			    $realfile !~ "debian/rules" &&
>  			    $realfile !~ /\.(py|pl|awk|sh)$/) {
>  				ERROR("EXECUTE_PERMISSIONS",
>  				      "do not set execute permissions for source files\n" . $permhere);

There are several of these scripts that have
no filename extension type and I suspect just
adding debian/rules isn't all that valuable.

$ git ls-files -s | grep '^100755' | grep -vP '\.(py|pl|awk|sh)$' | grep -v 'scripts/'
100755 b803f4f886b5e5e4046daedbd516e894f0f7e897 0	Documentation/target/target-export-device
100755 532d458333965b80f8f3ea149ad705208893b58f 0	arch/powerpc/boot/wrapper
100755 fbac1732f874e8d117b9c4aaf0a66b203370b933 0	arch/s390/kernel/syscalls/syscalltbl
100755 47c481239e986c35ba4818cb3e96f9ba7a813c2e 0	drivers/staging/greybus/tools/lbtest
100755 d7fc32ea8ac224f0fb6f28c58b196c665e510992 0	lib/build_OID_registry
100755 030a60b0404632ed598e5f5b56267e2a6af4b25f 0	samples/mic/mpssd/micctrl
100755 248ac7313c714d7acb29ceaaadd8fc8ee484e85e 0	samples/mic/mpssd/mpss
100755 2240cb56e6e5e6ed30a75913d4d5fecbb8f193c3 0	tools/debugging/kernel-chktaint
100755 ad1b9e646c4913b31aab9696245bc94b661647ff 0	tools/kvm/kvm_stat/kvm_stat
100755 49af9fe19f5b0533b64c117a363fc1c3785fa1da 0	tools/lib/lockdep/lockdep
100755 459469b7222c1410f7a701561eec004d86568369 0	tools/perf/arch/arm64/entry/syscalls/mksyscalltbl
100755 6c58060aa03beae7456e4f7cb0588e70207314be 0	tools/perf/arch/powerpc/entry/syscalls/mksyscalltbl
100755 72ecbb67637079f709262298981162b6cbb12c9f 0	tools/perf/arch/s390/entry/syscalls/mksyscalltbl
100755 fae26b1cf08f50a8c0caacf7bc0715e948ef0c03 0	tools/perf/tests/perf-targz-src-pkg
100755 3802cee5e1887f5b406324484832305e4a1b29b6 0	tools/perf/util/PERF-VERSION-GEN
100755 6d5e9e87c4b7f51f5904d17fa016a134563ba0d0 0	tools/testing/selftests/ftrace/ftracetest
100755 ea5938ec009a5eb9e28cb1778e081a568e66fd65 0	tools/testing/selftests/net/run_afpackettests
100755 14e41faf2c5740633f9dd30e500543647b736734 0	tools/testing/selftests/net/run_netsocktests
100755 4509f0cc9c918d33c39caf3372f2bcdfc81bceff 0	tools/testing/selftests/pstore/common_tests
100755 2a329bbb4aca3ad391a5aca9a67f554b060fdca2 0	tools/testing/selftests/pstore/pstore_crash_test
100755 d6da5e86efbf89c7f4d7454d9a2c6a11ef2ea9e4 0	tools/testing/selftests/pstore/pstore_post_reboot_tests
100755 1cef54458aff916b2f00af33deef703cb3101ab0 0	tools/testing/selftests/pstore/pstore_tests
100755 951c507a27f7df05f82ce2de78abba2e0d9848f4 0	tools/testing/selftests/vm/run_vmtests

Maybe setting an internal flag and looking at the file content
for a shebang might be better.

$ git ls-files -s  | grep '^100755' | grep -vP '\.(py|pl|awk|sh)$'|grep -v 'scripts/'|awk '{print $4}'|while read file ; do echo "$file       $(head -2 $file | grep '#!')" ; done
Documentation/target/target-export-device	#!/bin/sh
arch/powerpc/boot/wrapper	#!/bin/sh
arch/s390/kernel/syscalls/syscalltbl	#!/bin/sh
drivers/staging/greybus/tools/lbtest	#!/usr/bin/env python
lib/build_OID_registry	#!/usr/bin/perl -w
samples/mic/mpssd/micctrl	#!/bin/bash
samples/mic/mpssd/mpss	#!/bin/bash
tools/debugging/kernel-chktaint	#! /bin/sh
tools/kvm/kvm_stat/kvm_stat	#!/usr/bin/env python3
tools/lib/lockdep/lockdep	#!/bin/bash
tools/perf/arch/arm64/entry/syscalls/mksyscalltbl	#!/bin/sh
tools/perf/arch/powerpc/entry/syscalls/mksyscalltbl	#!/bin/sh
tools/perf/arch/s390/entry/syscalls/mksyscalltbl	#!/bin/sh
tools/perf/tests/perf-targz-src-pkg	#!/bin/sh
tools/perf/util/PERF-VERSION-GEN	#!/bin/sh
tools/testing/selftests/ftrace/ftracetest	#!/bin/sh
tools/testing/selftests/net/run_afpackettests	#!/bin/sh
tools/testing/selftests/net/run_netsocktests	#!/bin/sh
tools/testing/selftests/pstore/common_tests	#!/bin/sh
tools/testing/selftests/pstore/pstore_crash_test	#!/bin/sh
tools/testing/selftests/pstore/pstore_post_reboot_tests	#!/bin/sh
tools/testing/selftests/pstore/pstore_tests	#!/bin/sh
tools/testing/selftests/vm/run_vmtests	#!/bin/bash


_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

[PATCH 4/4] debian: add generic rule file

[Enrico Weigelt, metux IT consult]

Adding a generic debian rule file, so we can build the directly
via usual Debian package build tools (eg. git-buildpackage,
dck-buildpackage, etc). It expects the .config file already
placed in the source tree.

The rule file contains a rule for creating debian/control and
other metadata - this is done similar to the 'deb-pkg' make rule,
scripts/packaging/mkdebian.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
---
 .gitignore   |  1 +
 MAINTAINERS  |  6 ++++++
 debian/rules | 27 +++++++++++++++++++++++++++
 3 files changed, 34 insertions(+)
 create mode 100755 debian/rules

diff --git a/.gitignore b/.gitignore
index 7587ef56..01d742c 100644
--- a/.gitignore
+++ b/.gitignore
@@ -69,6 +69,7 @@ modules.builtin
 # Debian directory (make deb-pkg)
 #
 /debian/
+!/debian/rules
 
 #
 # Snap directory (make snap-pkg)
diff --git a/MAINTAINERS b/MAINTAINERS
index 558acf2..56e034c 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -4516,6 +4516,12 @@ F:	include/uapi/linux/dccp.h
 F:	include/linux/tfrc.h
 F:	net/dccp/
 
+DEBIAN PACKAGING FILES
+M:	Enrico Weigelt <info@metux.net>
+L:	linux-kbuild@vger.kernel.org
+S:	Maintained
+F:	debian/
+
 DECnet NETWORK LAYER
 W:	http://linux-decnet.sourceforge.net
 L:	linux-decnet-user@lists.sourceforge.net
diff --git a/debian/rules b/debian/rules
new file mode 100755
index 0000000..c2f0319
--- /dev/null
+++ b/debian/rules
@@ -0,0 +1,27 @@
+#!/usr/bin/make -f
+# SPDX-License-Identifier: GPL-2.0
+
+export MAKE
+export KERNELARCH           = $(shell $(MAKE) kernelarch)
+export KERNELRELEASE        = $(shell $(MAKE) kernelrelease)
+export KBUILD_DEBARCH       = $(shell dpkg-architecture -qDEB_HOST_ARCH)
+export KBUILD_BUILD_VERSION = $(shell $(MAKE) kernellocalversion)
+export KDEB_RULES           = debian/rules.auto
+export KDEB_SOURCENAME      = linux-source
+export ARCH                 = $(KERNELARCH)
+
+debian/control debian/changelong debian/arch debian/copyright:
+debian/control:
+	./scripts/package/mkdebian
+
+build:	debian/control
+	$(MAKE) KERNELRELEASE=$(KERNELRELEASE) ARCH=$(KERNELARCH) KBUILD_BUILD_VERSION=$(KBUILD_BUILD_VERSION) KBUILD_SRC=
+
+binary-arch:	debian/control
+	$(MAKE) KERNELRELEASE=$(KERNELRELEASE) ARCH=$(KERNELARCH) KBUILD_BUILD_VERSION=$(KBUILD_BUILD_VERSION) KBUILD_SRC= intdeb-pkg
+
+clean:
+	rm -rf debian/*tmp debian/files debian/changelog debian/control debian/copyright debian/rules.auto debian/arch
+	$(MAKE) clean
+
+binary: binary-arch
-- 
1.9.1


_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

Re: [PATCH 4/4] debian: add generic rule file

[Masahiro Yamada]

On Tue, Jul 9, 2019 at 6:33 PM Enrico Weigelt, metux IT consult
<info@metux.net> wrote:
>
> Adding a generic debian rule file, so we can build the directly
> via usual Debian package build tools (eg. git-buildpackage,
> dck-buildpackage, etc). It expects the .config file already
> placed in the source tree.
>
> The rule file contains a rule for creating debian/control and
> other metadata - this is done similar to the 'deb-pkg' make rule,
> scripts/packaging/mkdebian.

I saw a similar patch submission before, and negative feedback about it.

Debian maintains its own debian/rules, and it is fine.
I do not like to check-in the one in the kernel tree.



> Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
> ---
>  .gitignore   |  1 +
>  MAINTAINERS  |  6 ++++++
>  debian/rules | 27 +++++++++++++++++++++++++++
>  3 files changed, 34 insertions(+)
>  create mode 100755 debian/rules
>
> diff --git a/.gitignore b/.gitignore
> index 7587ef56..01d742c 100644
> --- a/.gitignore
> +++ b/.gitignore
> @@ -69,6 +69,7 @@ modules.builtin
>  # Debian directory (make deb-pkg)
>  #
>  /debian/
> +!/debian/rules
>
>  #
>  # Snap directory (make snap-pkg)
> diff --git a/MAINTAINERS b/MAINTAINERS
> index 558acf2..56e034c 100644
> --- a/MAINTAINERS
> +++ b/MAINTAINERS
> @@ -4516,6 +4516,12 @@ F:       include/uapi/linux/dccp.h
>  F:     include/linux/tfrc.h
>  F:     net/dccp/
>
> +DEBIAN PACKAGING FILES
> +M:     Enrico Weigelt <info@metux.net>
> +L:     linux-kbuild@vger.kernel.org
> +S:     Maintained
> +F:     debian/
> +
>  DECnet NETWORK LAYER
>  W:     http://linux-decnet.sourceforge.net
>  L:     linux-decnet-user@lists.sourceforge.net
> diff --git a/debian/rules b/debian/rules
> new file mode 100755
> index 0000000..c2f0319
> --- /dev/null
> +++ b/debian/rules
> @@ -0,0 +1,27 @@
> +#!/usr/bin/make -f
> +# SPDX-License-Identifier: GPL-2.0
> +
> +export MAKE
> +export KERNELARCH           = $(shell $(MAKE) kernelarch)
> +export KERNELRELEASE        = $(shell $(MAKE) kernelrelease)
> +export KBUILD_DEBARCH       = $(shell dpkg-architecture -qDEB_HOST_ARCH)
> +export KBUILD_BUILD_VERSION = $(shell $(MAKE) kernellocalversion)
> +export KDEB_RULES           = debian/rules.auto
> +export KDEB_SOURCENAME      = linux-source
> +export ARCH                 = $(KERNELARCH)
> +
> +debian/control debian/changelong debian/arch debian/copyright:
> +debian/control:
> +       ./scripts/package/mkdebian
> +
> +build: debian/control
> +       $(MAKE) KERNELRELEASE=$(KERNELRELEASE) ARCH=$(KERNELARCH) KBUILD_BUILD_VERSION=$(KBUILD_BUILD_VERSION) KBUILD_SRC=
> +
> +binary-arch:   debian/control
> +       $(MAKE) KERNELRELEASE=$(KERNELRELEASE) ARCH=$(KERNELARCH) KBUILD_BUILD_VERSION=$(KBUILD_BUILD_VERSION) KBUILD_SRC= intdeb-pkg
> +
> +clean:
> +       rm -rf debian/*tmp debian/files debian/changelog debian/control debian/copyright debian/rules.auto debian/arch
> +       $(MAKE) clean
> +
> +binary: binary-arch
> --
> 1.9.1
>


-- 
Best Regards
Masahiro Yamada

_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

Re: [PATCH 4/4] debian: add generic rule file

[Enrico Weigelt, metux IT consult]

On 15.07.19 14:28, Masahiro Yamada wrote:

>> The rule file contains a rule for creating debian/control and
>> other metadata - this is done similar to the 'deb-pkg' make rule,
>> scripts/packaging/mkdebian.
> 
> I saw a similar patch submission before, and negative feedback about it.

Do you recall what negative feedback exactly ?

> Debian maintains its own debian/rules, and it is fine.

Not for me, I don't use it - given up trying to make anything useful
out of it. It's extremly complex, practically undebuggable and doesn't
even work w/o lots of external preparations.


--mtx

-- 
Enrico Weigelt, metux IT consult
Free software and Linux embedded engineering
info@metux.net -- +49-151-27565287

_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

Re: [PATCH 4/4] debian: add generic rule file

[Theodore Y. Ts'o]

On Mon, Jul 15, 2019 at 08:56:25PM +0200, Enrico Weigelt, metux IT consult wrote:
> On 15.07.19 14:28, Masahiro Yamada wrote:
> 
> >> The rule file contains a rule for creating debian/control and
> >> other metadata - this is done similar to the 'deb-pkg' make rule,
> >> scripts/packaging/mkdebian.
> > 
> > I saw a similar patch submission before, and negative feedback about it.
> 
> Do you recall what negative feedback exactly ?

It's possible I'm not remembering some of the feedback, but the only
thing I recall was the comment I made that I'd really like this use
case:

make O=/build/linux-build bindeb-pkg

to not break.  And as far as I can tell from the proposed patch series
(I haven't had a chance to experimentally verify it yet), I don't
think it should break anything --- I'm assuming that we will still
have a way of creating the debian/rules file in
/build/linux-build/debian/rules when doing a O= build, and that the
intdeb-pkg rule remains the same.  At least, it appears to be the case
from my doing a quick look at the patches.

> > Debian maintains its own debian/rules, and it is fine.
> 
> Not for me, I don't use it - given up trying to make anything useful
> out of it. It's extremly complex, practically undebuggable and doesn't
> even work w/o lots of external preparations.

Yeah, the official Debian debian/rules is optimized for doing a
distribution release, and in addition to the issues Enrico has raised,
last time I tried it, it was S-L-O-W since it was building a fully
generic kernel.  It's not at all useable for general developer use.

It sounds like what Enrico is trying to do is to enable running
"dpkg-buildpackage -us -uc -b" from the the top-level kernel package
as being easier than running "make bindeb-pkg".  I suspect this might
be because his goal is to integrate individual kernel builds from
using Debian's hermetic build / chroot systems (e.g., sbuild, pbuilder)?

     	       		      	       	       	      - Ted

_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

Re: [PATCH 4/4] debian: add generic rule file

[Masahiro Yamada]

On Tue, Jul 16, 2019 at 4:13 AM Theodore Y. Ts'o <tytso@mit.edu> wrote:
>
> On Mon, Jul 15, 2019 at 08:56:25PM +0200, Enrico Weigelt, metux IT consult wrote:
> > On 15.07.19 14:28, Masahiro Yamada wrote:
> >
> > >> The rule file contains a rule for creating debian/control and
> > >> other metadata - this is done similar to the 'deb-pkg' make rule,
> > >> scripts/packaging/mkdebian.
> > >
> > > I saw a similar patch submission before, and negative feedback about it.
> >
> > Do you recall what negative feedback exactly ?

Sorry, my memory was broken.

I did not like this patch set from the beginning,
but missed to express my opinion strongly.

I want debian/ to be kept as a drop-in directory
for packagers, without replacing the upstream debian/rules.

If a check-in source file is modified in anyway,
scripts/setlocalversion would set -dirty flag,
which I want to avoid.



> It's possible I'm not remembering some of the feedback, but the only
> thing I recall was the comment I made that I'd really like this use
> case:
>
> make O=/build/linux-build bindeb-pkg
>
> to not break.  And as far as I can tell from the proposed patch series
> (I haven't had a chance to experimentally verify it yet), I don't
> think it should break anything --- I'm assuming that we will still
> have a way of creating the debian/rules file in
> /build/linux-build/debian/rules when doing a O= build, and that the
> intdeb-pkg rule remains the same.  At least, it appears to be the case
> from my doing a quick look at the patches.
>
> > > Debian maintains its own debian/rules, and it is fine.
> >
> > Not for me, I don't use it - given up trying to make anything useful
> > out of it. It's extremly complex, practically undebuggable and doesn't
> > even work w/o lots of external preparations.
>
> Yeah, the official Debian debian/rules is optimized for doing a
> distribution release, and in addition to the issues Enrico has raised,
> last time I tried it, it was S-L-O-W since it was building a fully
> generic kernel.  It's not at all useable for general developer use.

It is OK if the package is targeting normal users instead of
kernel developers.


> It sounds like what Enrico is trying to do is to enable running
> "dpkg-buildpackage -us -uc -b" from the the top-level kernel package
> as being easier than running "make bindeb-pkg".  I suspect this might
> be because his goal is to integrate individual kernel builds from
> using Debian's hermetic build / chroot systems (e.g., sbuild, pbuilder)?

I am OK with generating debian/rules with 'make bindeb-pkg', a shell scripts
or whatever, but I dislike to commit it in upstream git tree.

debian/rules is a hook for packagers to do their jobs in downstream.
"We kindly committed a generic one for you" sounds weird to me.

-- 
Best Regards
Masahiro Yamada

_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

Re: [PATCH 4/4] debian: add generic rule file

[Theodore Y. Ts'o]

On Tue, Jul 16, 2019 at 05:58:49PM +0900, Masahiro Yamada wrote:
> I want debian/ to be kept as a drop-in directory
> for packagers, without replacing the upstream debian/rules.
> 
> If a check-in source file is modified in anyway,
> scripts/setlocalversion would set -dirty flag,
> which I want to avoid.

In practice, that's not going to be a problem for most distributions.
The traditional way Debian-derived systems have done builds is
completely outside of git.  So there will be a linux_5.2.orig.tar.gz
and a linux_5.2-1.debian.tar.xz.  dpkg_source -x will first unpackage
the orig.tar.gz, and then the debian.tar.xz, and if the second
overwrites the first, it's no big deal.

More modern Debian package maintainer workflows may be using git, but
in that case, all of the "Debianizations" are reflected in a separate
branch.  So it's not going to set the -dirty flag.

There will be potential merge conflicts between Enrico's proposed
"upstream default debian/rules" file and the Debian/Ubuntu
debian/rules file on their distro branch.  However, I don't think
that's a big issue, for two reasons.

First, once it's checked in, I expect changes to the default
debian/rules file will be relatively rare.  Secondly, it's easy enough
to use gitattributes and defining a custom merge driver so that a
distribution can configure things so that they always use the version
of debian/rules from their branch, so the merge conflict resolution
can be set up to always do the right thing.

There are certainly other upstreams which ship their own debian/
directories.  E2fsprogs is one such example, but in that case I'm
cheating because I'm both the Debian package maintainer as well as the
upstream maintainer.  :-)   However, it's never been an issue for Ubuntu
when they choose to ship their own customized debian/rules file.

> debian/rules is a hook for packagers to do their jobs in downstream.
> "We kindly committed a generic one for you" sounds weird to me.

It is weird, and it's not common for upstream packages (which are not
native Debian packages) to ship their own debian directory.  But it
certainly does happen, and it won't cause any problems in actual
practice.

Regards,

					- Ted

_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

Re: [PATCH 4/4] debian: add generic rule file

[Enrico Weigelt, metux IT consult]

On 16.07.19 14:34, Theodore Y. Ts'o wrote:

Hi,

> In practice, that's not going to be a problem for most distributions.
> The traditional way Debian-derived systems have done builds is
> completely outside of git.  So there will be a linux_5.2.orig.tar.gz
> and a linux_5.2-1.debian.tar.xz.  dpkg_source -x will first unpackage
> the orig.tar.gz, and then the debian.tar.xz, and if the second
> overwrites the first, it's no big deal.

ACK. IIRC they already filter out debian/ directories when generating
upstream tarballs - other upstreams already provide their debian/
stuff, too.

> First, once it's checked in, I expect changes to the default
> debian/rules file will be relatively rare.  

ACK. I currently don't see much reasons for future changes. If anybody
sees something missing, just let me know, and I'll take it up for the
next review round.

--mtx

-- 
Enrico Weigelt, metux IT consult
Free software and Linux embedded engineering
info@metux.net -- +49-151-27565287

_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

Re: [PATCH 4/4] debian: add generic rule file

[Theodore Y. Ts'o]

On Wed, Jul 17, 2019 at 04:16:39PM +0200, Enrico Weigelt, metux IT consult wrote:
> 
> > In practice, that's not going to be a problem for most distributions.
> > The traditional way Debian-derived systems have done builds is
> > completely outside of git.  So there will be a linux_5.2.orig.tar.gz
> > and a linux_5.2-1.debian.tar.xz.  dpkg_source -x will first unpackage
> > the orig.tar.gz, and then the debian.tar.xz, and if the second
> > overwrites the first, it's no big deal.
> 
> ACK. IIRC they already filter out debian/ directories when generating
> upstream tarballs - other upstreams already provide their debian/
> stuff, too.

Well, no, actually they don't.  That's because as much as possible
they want the upstream tarball to be bit-for-bit identical to the one
published on the official upstream distribution site.  That allows
them to include the detached PGP signature from the upstream
maintainer, if one is provided.

If there are files in the upstream debian/ directory that they don't
need, they can delete in the distro's debian/rules file.  Ideally, so
we shouldn't include files in the Linux kernel's debian/ directory
willy-nilly.  But the debian/rules file will *always* be present, and
so it will be overwritten by the <package>_<ver>.debian.tar.xz file,
and so it's no big deal.

					- Ted

_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

Re: [PATCH 4/4] debian: add generic rule file

[Enrico Weigelt, metux IT consult]

On 15.07.19 21:12, Theodore Y. Ts'o wrote:

Hi,

> It's possible I'm not remembering some of the feedback, but the only> thing I recall was the comment I made that I'd really like this use>
case:> > make O=/build/linux-build bindeb-pkg
ah, I yet wanted to test that - thx for reminding me. > to not break.
And as far as I can tell from the proposed patch series> (I haven't had
a chance to experimentally verify it yet), I don't> think it should
break anything --- I'm assuming that we will still> have a way of
creating the debian/rules file in> /build/linux-build/debian/rules when
doing a O= build, and that the> intdeb-pkg rule remains the same.  At
least, it appears to be the case> from my doing a quick look at the patches.
Yes (unless i've missed something), everything should remain as it is.
One thing that could happen (not checked yet) is that running good old
'make bindeb-pkg' without O=... could overwrite the now already existing
debian/rules file.

If that's really a problem, we could tweak the machinery to use a
different name for the rule file (for now, one the preceeding patch
just allows giving a different name for just *generating* the rules
file). Another idea could be rewriting the whole process so that no
rules file needs to be generated at all.

> Yeah, the official Debian debian/rules is optimized for doing a
> distribution release, and in addition to the issues Enrico has raised,
> last time I tried it, it was S-L-O-W since it was building a fully
> generic kernel.  It's not at all useable for general developer use.

I'm a bit reluctant calling this 'optimized' :p

The strangest aspect (IMHO) is they're building several different trees
(w/ different huge patch queues) from only one source package. Instead
I'd rather:
* try to get as much as possible in one tree
* have separate source packages if there really need to be separate
  patche queues (IMHO, these things, like RT stuff, just need proper
  Kconfig's)
* do all the patching in git and skip the text-based patches entirely

Haven't found out, why they're actually doing it that complicated way
(didn't get any useful answers from debian kernel folks)

> It sounds like what Enrico is trying to do is to enable running
> "dpkg-buildpackage -us -uc -b" from the the top-level kernel package
> as being easier than running "make bindeb-pkg".  I suspect this might
> be because his goal is to integrate individual kernel builds from
> using Debian's hermetic build / chroot systems (e.g., sbuild, pbuilder)?

Yes, I'm building all deb's by the same process / infrastructure.
In my case it's dck-buildpackage (*1) which runs the build in a docker
container (kinda pbuilder w/ docker). It always starts with a fresh
(minimal) base image, calls debian/rules to create debian/control
(if necessary) deploys the dependencies found in the control file
and finally fire's up dpkg-buildpackage - the output is collected
in an ready-to-use apt repo.

The goal of this is having a canonical build process for all deb
packages, not having to care of any special cases anymore. I also
have another tool ontop of that, which runs the whole show for dozens
of packages and targets (*2).

My first approach was trying to use Debian source packages with new
kernel trees, but had to give up after a few days. Then I've found out
that the kernel already has *almost* what I needed. The difference
between almost and fine is this patch queue (minus local .config files)


--mtx

-- 
Enrico Weigelt, metux IT consult
Free software and Linux embedded engineering
info@metux.net -- +49-151-27565287

_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

Re: [PATCH 1/4] Makefile: rules for printing kernel architecture and localversion

[Randy Dunlap]

On 7/9/19 2:32 AM, Enrico Weigelt, metux IT consult wrote:
> trivial rule to print out the kernel arch and localversion, so
> external tools, like distro packagers, can easily get it.
> 
> Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>

Hi,
If accepted, these targets should be added to the top level Makefile's
"make help" output also.

Thanks.

> ---
>  Makefile | 6 ++++++
>  1 file changed, 6 insertions(+)
> 
> diff --git a/Makefile b/Makefile
> index 3e4868a..5afc3de 100644
> --- a/Makefile
> +++ b/Makefile
> @@ -1706,6 +1706,12 @@ kernelrelease:
>  kernelversion:
>  	@echo $(KERNELVERSION)
>  
> +kernellocalversion:
> +	@$(CONFIG_SHELL) $(srctree)/scripts/setlocalversion $(srctree) | sed -e 's~^\-~~'
> +
> +kernelarch:
> +	@echo $(ARCH)
> +
>  image_name:
>  	@echo $(KBUILD_IMAGE)
>  
> 


-- 
~Randy

_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv