[ptxdist] [PATCH] busybox: New patch to fix compilation with Linux v6.8-rc1
Signed-off-by: Uwe Kleine-König --- ...-Fix-compilation-with-Linux-v6.8-rc1.patch | 61 +++ patches/busybox-1.36.1/series | 3 +- 2 files changed, 63 insertions(+), 1 deletion(-) create mode 100644 patches/busybox-1.36.1/0203-tc-Fix-compilation-with-Linux-v6.8-rc1.patch diff --git a/patches/busybox-1.36.1/0203-tc-Fix-compilation-with-Linux-v6.8-rc1.patch b/patches/busybox-1.36.1/0203-tc-Fix-compilation-with-Linux-v6.8-rc1.patch new file mode 100644 index ..23715aa1a588 --- /dev/null +++ b/patches/busybox-1.36.1/0203-tc-Fix-compilation-with-Linux-v6.8-rc1.patch @@ -0,0 +1,61 @@ +From: =?UTF-8?q?Uwe=20Kleine-K=C3=B6nig?= +Date: Sat, 23 Mar 2024 13:28:56 +0100 +Subject: [PATCH] tc: Fix compilation with Linux v6.8-rc1 + +Linux v6.8-rc1 removed the definitions related to CBQ making tc fail to +build. Add some #ifdefs to handle this missing support. + +Forwarded: http://lists.busybox.net/pipermail/busybox/2024-March/090678.html +--- + networking/tc.c | 12 + 1 file changed, 12 insertions(+) + +diff --git a/networking/tc.c b/networking/tc.c +index 43187f7ee26c..1315b953e39f 100644 +--- a/networking/tc.c b/networking/tc.c +@@ -231,6 +231,13 @@ static int cbq_parse_opt(int argc, char **argv, struct nlmsghdr *n) + return 0; + } + #endif ++ ++#ifndef TCA_CBQ_MAX ++/* ++ * Linux v6.8-rc1~131^2~60^2^2 removed the uapi definitions for CBQ. ++ * See https://git.kernel.org/linus/33241dca48626 ++ */ ++#else + static int cbq_print_opt(struct rtattr *opt) + { + struct rtattr *tb[TCA_CBQ_MAX+1]; +@@ -322,6 +329,7 @@ static int cbq_print_opt(struct rtattr *opt) + done: + return 0; + } ++#endif + + static FAST_FUNC int print_qdisc( + const struct sockaddr_nl *who UNUSED_PARAM, +@@ -372,8 +380,10 @@ static FAST_FUNC int print_qdisc( + int qqq = index_in_strings(_q_, name); + if (qqq == 0) { /* pfifo_fast aka prio */ + prio_print_opt(tb[TCA_OPTIONS]); ++#ifdef TCA_CBQ_MAX + } else if (qqq == 1) { /* class based queuing */ + cbq_print_opt(tb[TCA_OPTIONS]); ++#endif + } else { + /* don't know how to print options for this qdisc */ + printf("(options for %s)", name); +@@ -442,9 +452,11 @@ static FAST_FUNC int print_class( + int qqq = index_in_strings(_q_, name); + if (qqq == 0) { /* pfifo_fast aka prio */ + /* nothing. */ /*prio_print_opt(tb[TCA_OPTIONS]);*/ ++#ifdef TCA_CBQ_MAX + } else if (qqq == 1) { /* class based queuing */ + /* cbq_print_copt() is identical to cbq_print_opt(). */ + cbq_print_opt(tb[TCA_OPTIONS]); ++#endif + } else { + /* don't know how to print options for this class */ + printf("(options for %s)", name); diff --git a/patches/busybox-1.36.1/series b/patches/busybox-1.36.1/series index ee536088790b..87fee50916c0 100644 --- a/patches/busybox-1.36.1/series +++ b/patches/busybox-1.36.1/series @@ -6,4 +6,5 @@ 0200-reactivate-check-for-tty.patch 0201-build-system-only-pass-real-libs-to-SELINUX_LIBS.patch 0202-scripts-trylink-honour-SKIP_STRIP-and-don-t-strip-if.patch -# 9c0cc4baa8090165b429198c9a10e02c - git-ptx-patches magic +0203-tc-Fix-compilation-with-Linux-v6.8-rc1.patch +# e781b8c6838e27fd021f9c1a9f87654d - git-ptx-patches magic base-commit: 30ecf336507b8ebc4742161ac00b4eb7edcc -- 2.39.2
Re: [ptxdist] [PATCH v2] mailmap: Add a few entries for improved commit statistics
On Wed, Dec 13, 2023 at 07:30:47PM +0100, Roland Hieber wrote: > On Fri, Nov 03, 2023 at 02:02:47PM +0100, Uwe Kleine-König wrote: > > When a single person uses different variants of their real name and > > different email addresses, they appear as different persons in the > > hall of fame in the category "Commits". > > > > With this .mailmap the number of individual committers is reduced from > > 148 to 134 (as of ptxdist-2023.11.0) as produced by > > > > git shortlog -sn > > > > . > > > > Signed-off-by: Uwe Kleine-König > > --- > > Hello, > > > > (implicit) v1 was sent with Message-Id: > > <20230607135129.189660-1-u.kleine-koe...@pengutronix.de> based on > > 2023.05.0. > > > > I added a few more entries pointed out by Alexander Dahl to merge some > > more entries. I was unsure if I should really merge Jürgen Beisert and > > Juergen Borleis. While this is only one person, his name actually > > changed, so I'm hesitant to merge. > > > > Best regards > > Uwe > > > > .mailmap | 16 > > 1 file changed, 16 insertions(+) > > create mode 100644 .mailmap > > > > diff --git a/.mailmap b/.mailmap > > new file mode 100644 > > index ..fb334866ce21 > > --- /dev/null > > +++ b/.mailmap > > @@ -0,0 +1,16 @@ > > +Andrej Gantvorg > > +Bjørn Bürger > > +Christian Gieseler > > +Christoph Rüdiger > > +Denis Osterland-Heim > > +Dennis Herbrich > > +Enrico Jorns > > +Enrico Jorns > > Is there a reason besides oversight why you put the "wrong" name in the > file, so now Enrico is always credited as "Enrico Jorns" instead of > "Enrico Jörns"? (The syntax according to the gitmailmap man page is > "Proper Name ".) "Enrico Jorns" is the spelling that Enrico uses in his contributions despite his official lastname being "Jörns". So I guess sticking to "Jorns" is right. > Otherwise I would suggest the following patch: > > diff --git a/.mailmap b/.mailmap > index fb334866ce21..f5f8dc7419a2 100644 > --- a/.mailmap > +++ b/.mailmap > @@ -4,12 +4,12 @@ Christian Gieseler > Christoph Rüdiger > Denis Osterland-Heim > Dennis Herbrich > -Enrico Jorns > -Enrico Jorns > +Enrico Jörns > +Enrico Jörns > Florian Bäuerle > Guillermo Rodríguez > -Jürgen Beisert > -Jürgen Beisert > +Jürgen Borleis > +Jürgen Borleis > Ladislav Michl > Ladislav Michl > Michael Olbrich At the time these commits were created the lastname was Beisert. So even knowing that Jürgen Borleis and Jürgen Beisert are the same guy, I didn't adapt that on purpose. Best regards Uwe -- Pengutronix e.K. | Uwe Kleine-König| Industrial Linux Solutions | https://www.pengutronix.de/ | signature.asc Description: PGP signature
[ptxdist] [PATCH 2/3] git-ptx-patches: Open .ptxdist/series less often for writing
Signed-off-by: Uwe Kleine-König --- scripts/git-ptx-patches | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/scripts/git-ptx-patches b/scripts/git-ptx-patches index df004fb8a8a5..56e82eb2ee56 100755 --- a/scripts/git-ptx-patches +++ b/scripts/git-ptx-patches @@ -204,7 +204,6 @@ fi GIT_EXTRA_ARGS="$GIT_EXTRA_ARGS --summary --stat=80" -cat .ptxdist/series.0 > .ptxdist/series ${GIT} format-patch -N $GIT_EXTRA_ARGS ${tagopt} -o .ptxdist/patches/ ${range} | while read patch; do if "$numbered_patches"; then patchname="${patch#.ptxdist/patches/}" @@ -215,8 +214,7 @@ ${GIT} format-patch -N $GIT_EXTRA_ARGS ${tagopt} -o .ptxdist/patches/ ${range} | echo "$patchname" done > .ptxdist/series.auto -cat .ptxdist/series.auto >> .ptxdist/series -cat .ptxdist/series.1 >> .ptxdist/series +cat .ptxdist/series.0 .ptxdist/series.auto .ptxdist/series.1 > .ptxdist/series cat .ptxdist/series | _md5sum >> .ptxdist/series # The first line of the patch is 'From ...' -- 2.42.0
[ptxdist] [PATCH 3/3] git-ptx-patches: Simplify nested if construct
Instead of if A; then ... else if B; then ... else ... fi fi use the more compact if A; then ... elif B; then ... else ... fi Signed-off-by: Uwe Kleine-König --- scripts/git-ptx-patches | 8 +++- 1 file changed, 3 insertions(+), 5 deletions(-) diff --git a/scripts/git-ptx-patches b/scripts/git-ptx-patches index 56e82eb2ee56..b27dbe7905d9 100755 --- a/scripts/git-ptx-patches +++ b/scripts/git-ptx-patches @@ -235,11 +235,9 @@ while read patch para; do if grep -q "$patch" .ptxdist/series.auto; then # ok, this is one of the patches we just touched : + elif grep -q "$patch" .ptxdist/series.{0,1}; then + echo "Base patch \"$patch\"!" else - if grep -q "$patch" .ptxdist/series.{0,1}; then - echo "Base patch \"$patch\"!" - else - echo "Old patch \"$patch\"!" - fi + echo "Old patch \"$patch\"!" fi done | sort -- 2.42.0
[ptxdist] [PATCH 1/3] git-ptx-patches: Fix a race condition
As the loop over the output of find moves files around, it can happen that find lists files more than once. In that case not only the "From " line is removed, but also the next one (usually the "From:" line). To process each patch in series.auto exactly once, iterate over that file to remove the "From " line instead. Still be a bit conservative and make sure that only the intended line is dropped. Reported-by: Thorsten Scherer Signed-off-by: Uwe Kleine-König --- scripts/git-ptx-patches | 12 +--- 1 file changed, 9 insertions(+), 3 deletions(-) diff --git a/scripts/git-ptx-patches b/scripts/git-ptx-patches index a063cb73a875..df004fb8a8a5 100755 --- a/scripts/git-ptx-patches +++ b/scripts/git-ptx-patches @@ -221,6 +221,13 @@ cat .ptxdist/series | _md5sum >> .ptxdist/series # The first line of the patch is 'From ...' # remove it to avoid unnecessary changes in the patch files. +while read patch para; do + # There are no comments or empty lines in series.auto, so no need to + # handle these. Also be a bit cautious to only remove lines matching + # "^From ". + sed -i '1{/^From /d}' ".ptxdist/patches/$patch" +done < .ptxdist/series.auto + find .ptxdist/patches/ ! -type d | sed -e 's,^.ptxdist/patches/,,' | \ while read patch para; do case "$patch" in @@ -228,9 +235,8 @@ while read patch para; do *) ;; esac if grep -q "$patch" .ptxdist/series.auto; then - p=".ptxdist/patches/$patch" - tail -n+2 "$p" > ".$patch.ptx-patches" - mv ".$patch.ptx-patches" "$p" + # ok, this is one of the patches we just touched + : else if grep -q "$patch" .ptxdist/series.{0,1}; then echo "Base patch \"$patch\"!" -- 2.42.0
[ptxdist] [PATCH 0/3] git-ptx-patches: A fix and two minor improvements
Hello, occasionally I hit the problem that git-ptx-patches removes the From: line from patches. I cannot reproduce that at will, but I think the issue is fixed in the first patch. I hit that sometimes in the past with no good idea how it could happen. Today Thorsten also hit it, which made me look into the problem again and then I spotted the issue. The other two patches are just minor improvements that I noticed while researching for the race. Best regards Uwe Uwe Kleine-König (3): git-ptx-patches: Fix a race condition git-ptx-patches: Open .ptxdist/series less often for writing git-ptx-patches: Simplify nested if construct scripts/git-ptx-patches | 24 +--- 1 file changed, 13 insertions(+), 11 deletions(-) base-commit: 8d8f9a47d135920da7b70a5d71b763c457f4bcaf -- 2.42.0
[ptxdist] [PATCH] ptxd_make_nested: Fix parameter passing for --dirty, --force and --pedantic
Some variables corresponding to ptxdist parameters use "true" as value when given, others use "1". PTXDIST_DIRTY, PTXDIST_FORCE and PTXDIST_PEDANTIC are in the first group according to git grep -E 'PTXDIST_(DIRTY|FORCE|FEDANTIC)=' . Adapt ptxd_make_nested_ptxdist() accordingly. Fixes: 5169ae99daf8 ("add support for nested PTXdist execution") Signed-off-by: Uwe Kleine-König --- scripts/lib/ptxd_make_nested.sh | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/scripts/lib/ptxd_make_nested.sh b/scripts/lib/ptxd_make_nested.sh index cae12bceb377..2c4af5337207 100644 --- a/scripts/lib/ptxd_make_nested.sh +++ b/scripts/lib/ptxd_make_nested.sh @@ -46,13 +46,13 @@ ptxd_make_nested_ptxdist() { fi args=( "cd" "${pkg_workspace}" "&&" "${PTXDIST_TOPDIR}/bin/ptxdist" ) -if [ "${PTXDIST_DIRTY}" = 1 ]; then +if [ "${PTXDIST_DIRTY}" = true ]; then args[${#args[*]}]="--dirty" fi -if [ "${PTXDIST_FORCE}" = 1 ]; then +if [ "${PTXDIST_FORCE}" = true ]; then args[${#args[*]}]="--force" fi -if [ "${PTXDIST_PEDANTIC}" = 1 ]; then +if [ "${PTXDIST_PEDANTIC}" = true ]; then args[${#args[*]}]="--pedantic" fi if [ "${PTXDIST_QUIET}" = 1 ]; then -- 2.42.0
[ptxdist] [PATCH v2] mailmap: Add a few entries for improved commit statistics
When a single person uses different variants of their real name and different email addresses, they appear as different persons in the hall of fame in the category "Commits". With this .mailmap the number of individual committers is reduced from 148 to 134 (as of ptxdist-2023.11.0) as produced by git shortlog -sn . Signed-off-by: Uwe Kleine-König --- Hello, (implicit) v1 was sent with Message-Id: <20230607135129.189660-1-u.kleine-koe...@pengutronix.de> based on 2023.05.0. I added a few more entries pointed out by Alexander Dahl to merge some more entries. I was unsure if I should really merge Jürgen Beisert and Juergen Borleis. While this is only one person, his name actually changed, so I'm hesitant to merge. Best regards Uwe .mailmap | 16 1 file changed, 16 insertions(+) create mode 100644 .mailmap diff --git a/.mailmap b/.mailmap new file mode 100644 index ..fb334866ce21 --- /dev/null +++ b/.mailmap @@ -0,0 +1,16 @@ +Andrej Gantvorg +Bjørn Bürger +Christian Gieseler +Christoph Rüdiger +Denis Osterland-Heim +Dennis Herbrich +Enrico Jorns +Enrico Jorns +Florian Bäuerle +Guillermo Rodríguez +Jürgen Beisert +Jürgen Beisert +Ladislav Michl +Ladislav Michl +Michael Olbrich +Thorsten Liepert -- 2.42.0
[ptxdist] [PATCH] Templates/src-autoconf-*: Fix a few typos in the configure template
Signed-off-by: Uwe Kleine-König --- rules/templates/src-autoconf-lib/configure.ac | 8 rules/templates/src-autoconf-prog/configure.ac| 6 +++--- rules/templates/src-autoconf-proglib/configure.ac | 8 3 files changed, 11 insertions(+), 11 deletions(-) diff --git a/rules/templates/src-autoconf-lib/configure.ac b/rules/templates/src-autoconf-lib/configure.ac index f062e6be5e57..b15a67945eec 100644 --- a/rules/templates/src-autoconf-lib/configure.ac +++ b/rules/templates/src-autoconf-lib/configure.ac @@ -65,11 +65,11 @@ LT_INIT # # Please note: 'configure' defaults the CFLAGS to "-O2 -g" if it detects a -# GCC. If the user overwrites the CFLAGS externallly, there will be no default! +# GCC. If the user overwrites the CFLAGS externally, there will be no default! # Keep the user informed about this behaviour. # -# TODO Check here for mandadory external libraries +# TODO Check here for mandatory external libraries # # Rule of thumb: when your library needs an external library, add it # to the 'LIBS' variable. Do _not_ add it manually into the Makefile.am files. @@ -96,7 +96,7 @@ LT_INIT # the 'autoscan' tool to get an idea what is still missing in your 'configure.ac' # -# where to install pgk info files +# where to install pkg info files PKG_INSTALLDIR PTX_COMMON_CHECKS @@ -149,7 +149,7 @@ AC_SUBST(LIBS) # add required lib information to the pkg-config file (on demand) AC_SUBST(REQUIRES) -# add lib information which may conflicts with this libarry +# add lib information which may conflict with this library AC_SUBST(CONFLICTS) # add special compile flags on demand diff --git a/rules/templates/src-autoconf-prog/configure.ac b/rules/templates/src-autoconf-prog/configure.ac index 8dcb7cf573bc..7618103c5280 100644 --- a/rules/templates/src-autoconf-prog/configure.ac +++ b/rules/templates/src-autoconf-prog/configure.ac @@ -43,11 +43,11 @@ AC_PROG_CC_C99 # # Please note: 'configure' defaults the CFLAGS to "-O2 -g" if it detects a -# GCC. If the user overwrites the CFLAGS externallly, there will be no default! +# GCC. If the user overwrites the CFLAGS externally, there will be no default! # Keep the user informed about this behaviour. # -# TODO Check here for mandadory external libraries +# TODO Check here for mandatory external libraries # # Rule of thumb: when your library needs an external library, add it # to the 'LIBS' variable. Do _not_ add it manually into the Makefile.am files. @@ -124,7 +124,7 @@ AC_SUBST(LIBS) # add required lib information to the pkg-config file (on demand) AC_SUBST(REQUIRES) -# add lib information which may conflicts with this libarry +# add lib information which may conflict with this libarry AC_SUBST(CONFLICTS) # add special compile flags on demand diff --git a/rules/templates/src-autoconf-proglib/configure.ac b/rules/templates/src-autoconf-proglib/configure.ac index ad873c9caca5..67738fad5a83 100644 --- a/rules/templates/src-autoconf-proglib/configure.ac +++ b/rules/templates/src-autoconf-proglib/configure.ac @@ -65,11 +65,11 @@ LT_INIT # # Please note: 'configure' defaults the CFLAGS to "-O2 -g" if it detects a -# GCC. If the user overwrites the CFLAGS externallly, there will be no default! +# GCC. If the user overwrites the CFLAGS externally, there will be no default! # Keep the user informed about this behaviour. # -# TODO Check here for mandadory external libraries +# TODO Check here for mandatory external libraries # # Rule of thumb: when your library needs an external library, add it # to the 'LIBS' variable. Do _not_ add it manually into the Makefile.am files. @@ -96,7 +96,7 @@ LT_INIT # the 'autoscan' tool to get an idea what is still missing in your 'configure.ac' # -# where to install pgk info files +# where to install pkg info files PKG_INSTALLDIR PTX_COMMON_CHECKS @@ -150,7 +150,7 @@ AC_SUBST(LIBS) # add required lib information to the pkg-config file (on demand) AC_SUBST(REQUIRES) -# add lib information which may conflicts with this libarry +# add lib information which may conflict with this library AC_SUBST(CONFLICTS) # add special compile flags on demand base-commit: dbf616b650552733786ff6b96f7b98a42128c398 -- 2.39.2
[ptxdist] [PATCH] mailmap: Add a few entries for improved commit statistics
When a single person uses different variants of their real name and different email addresses, they appear as different persons in the hall of fame in the category "Commits". With this .mailmap the number of individual committers is reduced from 148 to 141 (as of ptxdist-2023.05.0) as produced by git shortlog -sn . Signed-off-by: Uwe Kleine-König --- .mailmap | 7 +++ 1 file changed, 7 insertions(+) create mode 100644 .mailmap diff --git a/.mailmap b/.mailmap new file mode 100644 index ..c1c31f0237a0 --- /dev/null +++ b/.mailmap @@ -0,0 +1,7 @@ +Christian Gieseler +Denis Osterland-Heim +Florian Bäuerle +Ladislav Michl +Ladislav Michl +Michael Olbrich +Thorsten Liepert base-commit: 5015bfbd0347abfc20bcf0bfedf5b02670353924 -- 2.39.2
[ptxdist] [PATCH] ptxd_make_world_compile_commands_filter: Fix grep for uptodate check
The intend here is to check the return code of grep, no need to call [ (i.e. test). This fixes an error message: environment: line 10: [: syntax error: `-q' unexpected Additionally add -e to force interpretation of "${PTXDIST_CROSS_CPPFLAGS}" as pattern. For me ${PTXDIST_CROSS_CPPFLAGS} starts with -isystem; without -e this is interpreted as option to grep instead of pattern which makes grep emit grep: invalid option -- 't' .) Also add -F to make grep use a fixed-string compare (instead of interpreting ${PTXDIST_CROSS_CPPFLAGS} as regexp). Fixes: ac0aac7fc727 ("impove compile_commands.json handling") Signed-off-by: Uwe Kleine-König --- scripts/lib/ptxd_make_world_compile_commands.sh | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/scripts/lib/ptxd_make_world_compile_commands.sh b/scripts/lib/ptxd_make_world_compile_commands.sh index 56d7378fa28b..2703b5497799 100644 --- a/scripts/lib/ptxd_make_world_compile_commands.sh +++ b/scripts/lib/ptxd_make_world_compile_commands.sh @@ -21,7 +21,7 @@ ptxd_make_world_compile_commands_filter() { fi if [ "${src_cmds}" = "${dst_cmds}" ]; then - if [ grep -q "${PTXDIST_CROSS_CPPFLAGS}" "${src_cmds}" ]; then + if grep -q -F -e "${PTXDIST_CROSS_CPPFLAGS}" "${src_cmds}"; then ptxd_warning "compile_commands.json is already up to date!" return fi -- 2.39.0
[ptxdist] [PATCH] doc: Files are added to _LICENSE_FILES
This fixes a copy-n-paste error. _LICENSE_FILES is supposed to contain a list of files, not _LICENSE. Signed-off-by: Uwe Kleine-König --- doc/dev_licenses.rst | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/doc/dev_licenses.rst b/doc/dev_licenses.rst index 0bb1c8d77c5e..6aed9daaca97 100644 --- a/doc/dev_licenses.rst +++ b/doc/dev_licenses.rst @@ -195,7 +195,7 @@ Adding license files to PTXdist packages The SPDX license identifier of the package goes into the ``_LICENSE`` variable in the respective package rule file. -All relevant files identified in the steps above are then added to the variable ``_LICENSE``, +All relevant files identified in the steps above are then added to the variable ``_LICENSE_FILES``, including a checksum so that PTXdist complains when they change. Example: -- 2.39.0
[ptxdist] [PATCH] treewide: Consistently use ':' instead to separate user and group for chown
Since coreutils 9.1 chown warns when a '.' is used as separator. Fix this consistently also for documentation and scripts that are expected to use busybox's chown which (up to now) has no problem with '.'. Signed-off-by: Uwe Kleine-König --- config/busybox/Config.in | 4 ++-- config/busybox/networking/Config.in| 2 +- projectroot/usr/sbin/mysqlinit | 2 +- scripts/lib/ptxd_lib_dopermissions.awk | 4 ++-- 4 files changed, 6 insertions(+), 6 deletions(-) diff --git a/config/busybox/Config.in b/config/busybox/Config.in index 199cc6ab272f..329e2f8e8a1b 100644 --- a/config/busybox/Config.in +++ b/config/busybox/Config.in @@ -267,10 +267,10 @@ config BUSYBOX_FEATURE_SUID_CONFIG The file has to be owned by user root, group root and has to be writeable only by root: - (chown 0.0 /etc/busybox.conf; chmod 600 /etc/busybox.conf) + (chown 0:0 /etc/busybox.conf; chmod 600 /etc/busybox.conf) The busybox executable has to be owned by user root, group root and has to be setuid root for this to work: - (chown 0.0 /bin/busybox; chmod 4755 /bin/busybox) + (chown 0:0 /bin/busybox; chmod 4755 /bin/busybox) Robert 'sandman' Griebl has more information here: http://www.softforge.de/bb/suid.html >. diff --git a/config/busybox/networking/Config.in b/config/busybox/networking/Config.in index 3a3ad56d93f7..27e0f44ba811 100644 --- a/config/busybox/networking/Config.in +++ b/config/busybox/networking/Config.in @@ -959,7 +959,7 @@ config BUSYBOX_TELNETD FEATURE_SUID enabled. And finally, you should make certain that busybox has been installed setuid root: - chown root.root /bin/busybox + chown root:root /bin/busybox chmod 4755 /bin/busybox with all that done, telnetd _should_ work diff --git a/projectroot/usr/sbin/mysqlinit b/projectroot/usr/sbin/mysqlinit index 72b6d1f10cf5..c98946025264 100644 --- a/projectroot/usr/sbin/mysqlinit +++ b/projectroot/usr/sbin/mysqlinit @@ -5,7 +5,7 @@ MYSQLD_OPTS= # since mySQL runs as a regular user, we need to give its working directory # to it. -chown mysql.mysql "$1" +chown mysql:mysql "$1" chmod 0755 "$1" # now let mySQL do its initializing job diff --git a/scripts/lib/ptxd_lib_dopermissions.awk b/scripts/lib/ptxd_lib_dopermissions.awk index 56e951b45b75..9ab45d4979dd 100755 --- a/scripts/lib/ptxd_lib_dopermissions.awk +++ b/scripts/lib/ptxd_lib_dopermissions.awk @@ -7,14 +7,14 @@ BEGIN { $1 ~ "[df]" { printf("chmod %s'.%s'\n" \ - "chown %s.%s '.%s'\n", \ + "chown %s:%s '.%s'\n", \ $5, $2, $3, $4, $2); } $1 ~ "n" { printf("if [ ! \\( -b '.%s' -o -c '.%s' -o -p '.%s' \\) ]; then " \ " mknod -m %s '.%s' %s %s %s\n" \ - " chown %s.%s '.%s'\n" \ + " chown %s:%s '.%s'\n" \ "fi\n", \ $2, $2, $2, $5, $2, $6, $7, $8, $3, $4, $2); } -- 2.38.1
[ptxdist] [PATCH] ptxd_make_get_svn: Sort members of resulting archive by name
Without this option the order of the files in the created archive depends on things like the used filesystem. To make the order (and so the resulting md5sum) reproducible, use --sort=name. This option is supported in GNU tar since version 1.28 from 2014, which is available since Debian 7 (aka wheezy). Signed-off-by: Uwe Kleine-König --- scripts/lib/ptxd_make_get.sh | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/scripts/lib/ptxd_make_get.sh b/scripts/lib/ptxd_make_get.sh index 2b0ad9307476..e78c417278e4 100644 --- a/scripts/lib/ptxd_make_get.sh +++ b/scripts/lib/ptxd_make_get.sh @@ -268,7 +268,7 @@ ptxd_make_get_svn() { echo "${PROMPT}svn: last modification time '${lmtime}'" && GZIP=-n tar --exclude-vcs --show-stored-names ${tarcomp} \ --mtime="${lmtime}" --transform "s|^\.|${prefix}|g" \ - --create --file "${path}" -C "${mirror}" . + --create --sort=name --file "${path}" -C "${mirror}" . ptxd_make_serialize_put } export -f ptxd_make_get_svn base-commit: 9fc1adb2c35e681c7eb10107481a2e5023862088 -- 2.38.1
[ptxdist] [PATCH] clean: Fix a double whitespace
Fixes: c98e2ba9594c ("detect patch changes and reextract when needed") --- scripts/lib/ptxd_make_clean.sh | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/scripts/lib/ptxd_make_clean.sh b/scripts/lib/ptxd_make_clean.sh index a3a7cde797bc..4bfce603c94a 100644 --- a/scripts/lib/ptxd_make_clean.sh +++ b/scripts/lib/ptxd_make_clean.sh @@ -18,7 +18,7 @@ ptxd_make_clean() { if [ -z "${PTXCONF_SETUP_AUTO_EXTRACT}" ]; then ptxd_bailout "Refusing to delete existing source" \ "$(ptxd_print_path "${directory}")" \ - "Delete anyways with --force or skip extract with --dirty." + "Delete anyways with --force or skip extract with --dirty." elif [ -z "${PTXDIST_FORCE}" -a -d "${directory}/.git" ]; then ptxd_bailout "Refusing to delete existing source with git" \ "$(ptxd_print_path "${directory}")" \ -- 2.38.1
[ptxdist] [PATCH] ptxd_lib_code_signing: Check for errors in cs_import_cert_from_pem()
cs_import_cert_from_pem() uses a pipe command to do the necessary things. If the first command in that pipe fails the whole command should be considered failing. So add a call to check_pipe_status as is done in cs_import_privkey_from_pem(). Fixes: 8f41183e0afe ("Add initial code signing support") Signed-off-by: Uwe Kleine-König --- scripts/lib/ptxd_lib_code_signing.sh | 1 + 1 file changed, 1 insertion(+) diff --git a/scripts/lib/ptxd_lib_code_signing.sh b/scripts/lib/ptxd_lib_code_signing.sh index 5ba1a4666af4..bafdc16544d3 100644 --- a/scripts/lib/ptxd_lib_code_signing.sh +++ b/scripts/lib/ptxd_lib_code_signing.sh @@ -213,6 +213,7 @@ cs_import_cert_from_pem() { "${openssl_keyopt[@]}" \ -in "${pem}" -inform pem -outform der | softhsm_pkcs11_tool --type cert --write-object /dev/stdin --label "${role}" +check_pipe_status } export -f cs_import_cert_from_pem -- 2.37.2
[ptxdist] [PATCH] ptxdist: Add --silent and -s as aliases for --quiet
Many options that ptxdist parses are the same for make(1) (e.g. -j, -k, -d). However the option to make make silent is -s, add that to reduce the time I spend on thinking which option is for which command. While at it also add --silent in accordance to make (which also knows about --quiet by the way). Signed-off-by: Uwe Kleine-König --- bin/ptxdist | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/bin/ptxdist b/bin/ptxdist index af0e5253e89f..740d77321688 100755 --- a/bin/ptxdist +++ b/bin/ptxdist @@ -812,6 +812,7 @@ Options: --debug, -d print out additional info (like make decisions) --quiet, -q suppress output, show only stderr + --silent, -s alias for --quiet --verbose, -vbe more verbose, print command before execute them --output-syncImprove output readability for parallel building. Disabled by default except for quiet builds. @@ -1148,7 +1149,7 @@ parse_first() --progress) PTXDIST_PROGRESS=1 ;; - -q|--quiet) + -q|--quiet|-s|--silent) PTXDIST_QUIET=1 ;; --git) base-commit: bdb6f5f98505a13dbeffda3dfdea4a4f3bf586c6 -- 2.37.2
Re: [ptxdist] [PATCH] tf-a: Select openssl in host package
On Wed, Jul 13, 2022 at 08:49:53AM +0200, Uwe Kleine-König wrote: > tf-a needs openssl, otherwise: > > HOSTCC tbbr_config.c > In file included from fiptool.h:16, >from fiptool.c:19: > fiptool_platform.h:19:11: fatal error: openssl/sha.h: No such file or > directory > 19 | # include > | ^~~ > compilation terminated. > > (This only happens if the openssl development package isn't installed on > the host. Missing ), assuming you apply, can you fix that up please? Best regards Uwe -- Pengutronix e.K. | Uwe Kleine-König| Industrial Linux Solutions | https://www.pengutronix.de/ | signature.asc Description: PGP signature
[ptxdist] [PATCH] tf-a: Select openssl in host package
tf-a needs openssl, otherwise: HOSTCC tbbr_config.c In file included from fiptool.h:16, from fiptool.c:19: fiptool_platform.h:19:11: fatal error: openssl/sha.h: No such file or directory 19 | # include | ^~~ compilation terminated. (This only happens if the openssl development package isn't installed on the host. Signed-off-by: Uwe Kleine-König --- platforms/host-tf-a.in | 1 + 1 file changed, 1 insertion(+) diff --git a/platforms/host-tf-a.in b/platforms/host-tf-a.in index 1a299f042724..f8fd15c48937 100644 --- a/platforms/host-tf-a.in +++ b/platforms/host-tf-a.in @@ -3,5 +3,6 @@ config HOST_TF_A tristate default y if ALLYES && TF_A + select HOST_OPENSSL help This provides fiptool base-commit: 02f8e5e7cb2ae088b69bfd2d75ba9e702eeaac5f -- 2.36.1
Re: [ptxdist] [Kernel.org Helpdesk #86171] [linuxfoundation.org #86171] Re: Archive ptxdist@pengutronix.de on lore.kernel.org?
Hello Konstantin, On Tue, May 03, 2022 at 09:04:39AM +0200, Ahmad Fatoum wrote: > On 26.11.21 11:28, Uwe Kleine-König wrote: > > On Tue, Apr 14, 2020 at 05:31:22PM -0400, Konstantin Ryabitsev via RT wrote: > >> I hope that we'll be able to roll out our aggregated mirroring/indexing > >> service closer to mid-year -- and when that happens, we'll be happy to > >> mirror your lists and offer them alongside all others. > > > > I'd like to come back to this offer. We have setup our own public-inbox > > instances now for a while and mirroring them on lore.kernel.org would be > > great. > > > > These are the links to our instances: > > > > https://lore.barebox.org/barebox > > https://lore.ptxdist.org/ptxdist > > https://lore.distrokit.org/distrokit > > https://lore.pengutronix.de/oss-tools > > > > Is there anything else you need, some meta-data maybe? > > Did you have time to look into mirroring these instances? It doesn't seem you had, at least these are not listed on https://lore.kernel.org/ yet :-\ Apart from that, I expected all vger lists to be on lore, but I miss devicetree-compi...@vger.kernel.org, too. Best regards Uwe -- Pengutronix e.K. | Uwe Kleine-König| Industrial Linux Solutions | https://www.pengutronix.de/ | signature.asc Description: PGP signature
[ptxdist] [PATCH] git-ptx-patches: Remove useless assignment
range is overwritten unconditionally after option parsing, so drop the assignment. Its only effect was irritating a human reader. Signed-off-by: Uwe Kleine-König --- scripts/git-ptx-patches | 1 - 1 file changed, 1 deletion(-) diff --git a/scripts/git-ptx-patches b/scripts/git-ptx-patches index 8bf1731aeb53..fe3d1b619d21 100755 --- a/scripts/git-ptx-patches +++ b/scripts/git-ptx-patches @@ -64,7 +64,6 @@ while getopts "aft:n" opt; do ;; t) tag="${OPTARG}" - range="${tag}" ;; n) numbered_patches=false base-commit: 5b07b31b850e69493e4fd7bd4fa46501ee92684a prerequisite-patch-id: 0d2b85af86e0b2a7844ba3fcb2d66d84228ab8a1 -- 2.35.1
[ptxdist] [PATCH] git-ptx-patches: Fix corner case when searching for a tag in series file
Require a space after the tag name, otherwise searching for tag=foo might match a line #tag:foobar ... Fixes: 957c6c663cdd ("git-ptx-patches: Allow patches before base tag") Signed-off-by: Uwe Kleine-König --- scripts/git-ptx-patches | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/scripts/git-ptx-patches b/scripts/git-ptx-patches index b52817399d45..8bf1731aeb53 100755 --- a/scripts/git-ptx-patches +++ b/scripts/git-ptx-patches @@ -73,7 +73,7 @@ while getopts "aft:n" opt; do done shift $((${OPTIND} - 1)) -tag2=$(grep "#tag:" .ptxdist/series | awk "/#tag:${tag}/{if (getline) print \$1}" | sed 's,#tag:,,') +tag2=$(grep "#tag:" .ptxdist/series | awk "/#tag:${tag} /{if (getline) print \$1}" | sed 's,#tag:,,') range="${tag}..${tag2}" if "${all_tags}" ; then base-commit: 5b07b31b850e69493e4fd7bd4fa46501ee92684a -- 2.35.1
Re: [ptxdist] [PATCH] microcom: correct SPDX license identifier GPL-2.0-only -> GPL-2.0-or-later
On Thu, Apr 07, 2022 at 12:58:15PM +0200, Bastian Krause wrote: > The license header in the source files includes the "or (at your option) > any later version" clause. Looks right, the only irritating thing is that the COPYRIGHT file says All files included in microcom are distributable under the "GNU GENERAL PUBLIC LICENSE version 2". (Some might have the "or any later version" phrase; this doesn't affect the previous statement though.) If the file doesn't specify a license explicitly, assume "version 2 only". The "GNU GENERAL PUBLIC LICENSE version 2" is appended below. but all *.c and *.h files have the "or later" annotation. Hmm. This is an upstream thing though, to this patch looks fine. Acked-by: Uwe Kleine-König Best regards Uwe -- Pengutronix e.K. | Uwe Kleine-König| Industrial Linux Solutions | https://www.pengutronix.de/ | signature.asc Description: PGP signature ___ ptxdist mailing list ptxdist@pengutronix.de To unsubscribe, send a mail with subject "unsubscribe" to ptxdist-requ...@pengutronix.de
[ptxdist] [PATCH v2] tf-a: Support enabling DEBUG via TF_A_EXTRA_ARGS
With DEBUG=1 the resulting images are located in a different folder, so adapt the install target accordingly. Before this adaption the build either failed when DEBUG=1 was contained in TF_A_EXTRA_ARGS (because no images were found in the release folder) or (IMHO worse) ptxdist picked up the build results of an earlier attempt where DEBUG wasn't set yet in TF_A_EXTRA_ARGS. Also remove DEBUG=0 from the hardcoded set of options. This is the default in the tf-a Makefiles anyhow, and passing both DEBUG=0 and DEBUG=1 only results in confusion. Signed-off-by: Uwe Kleine-König --- Hello, compared to (implicit) v1 this patch also works with DEBUG=1 in TF_A_EXTRA_ARGS. *sigh* Best regards Uwe rules/tf-a.make | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/rules/tf-a.make b/rules/tf-a.make index 25fdb0cd213a..2caa68d97054 100644 --- a/rules/tf-a.make +++ b/rules/tf-a.make @@ -45,7 +45,6 @@ TF_A_MAKE_OPT := \ -C $(TF_A_DIR) \ CROSS_COMPILE=$(BOOTLOADER_CROSS_COMPILE) \ HOSTCC=$(HOSTCC) \ - DEBUG=0 \ ARCH=$(PTXCONF_TF_A_ARCH_STRING) \ ARM_ARCH_MAJOR=$(PTXCONF_TF_A_ARM_ARCH_MAJOR) \ BUILD_STRING=$(PTXCONF_TF_A_VERSION) \ @@ -90,7 +89,7 @@ $(STATEDIR)/tf-a.compile: # tf-a/inst_plat = $(foreach artifact, \ - $(wildcard $(TF_A_BUILDDIR)/$(1)/release/$(TF_A_ARTIFACTS)), \ + $(wildcard $(TF_A_BUILDDIR)/$(1)/$(if $(filter DEBUG=1,$(call remove_quotes,$(PTXCONF_TF_A_EXTRA_ARGS))),debug,release)/$(TF_A_ARTIFACTS)), \ install -v -D -m 644 $(artifact) \ $(2)/$(1)-$(notdir $(artifact))$(ptx/nl)) -- 2.34.1 ___ ptxdist mailing list ptxdist@pengutronix.de To unsubscribe, send a mail with subject "unsubscribe" to ptxdist-requ...@pengutronix.de
[ptxdist] [PATCH] tf-a: Support enabling DEBUG via TF_A_EXTRA_ARGS
With DEBUG=1 the resulting images are located in a different folder, so adapt the install target accordingly. Before this adaption the build either failed when DEBUG=1 was contained in TF_A_EXTRA_ARGS (because no images were found in the release folder) or (IMHO worse) ptxdist picked up the build results of an earlier attempt where DEBUG wasn't set yet in TF_A_EXTRA_ARGS. Also remove DEBUG=0 from the hardcoded set of options. This is the default in the tf-a Makefiles anyhow, and passing both DEBUG=0 and DEBUG=1 only results in confusion. Signed-off-by: Uwe Kleine-König --- rules/tf-a.make | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/rules/tf-a.make b/rules/tf-a.make index 25fdb0cd213a..72e1bab69371 100644 --- a/rules/tf-a.make +++ b/rules/tf-a.make @@ -45,7 +45,6 @@ TF_A_MAKE_OPT := \ -C $(TF_A_DIR) \ CROSS_COMPILE=$(BOOTLOADER_CROSS_COMPILE) \ HOSTCC=$(HOSTCC) \ - DEBUG=0 \ ARCH=$(PTXCONF_TF_A_ARCH_STRING) \ ARM_ARCH_MAJOR=$(PTXCONF_TF_A_ARM_ARCH_MAJOR) \ BUILD_STRING=$(PTXCONF_TF_A_VERSION) \ @@ -90,7 +89,7 @@ $(STATEDIR)/tf-a.compile: # tf-a/inst_plat = $(foreach artifact, \ - $(wildcard $(TF_A_BUILDDIR)/$(1)/release/$(TF_A_ARTIFACTS)), \ + $(wildcard $(TF_A_BUILDDIR)/$(1)/$(if $(find DEBUG=1,$(PTXCONF_TF_A_EXTRA_ARGS)),debug,release)/$(TF_A_ARTIFACTS)), \ install -v -D -m 644 $(artifact) \ $(2)/$(1)-$(notdir $(artifact))$(ptx/nl)) -- 2.34.1 ___ ptxdist mailing list ptxdist@pengutronix.de To unsubscribe, send a mail with subject "unsubscribe" to ptxdist-requ...@pengutronix.de
Re: [ptxdist] [Kernel.org Helpdesk #86171] [linuxfoundation.org #86171] Re: Archive ptxdist@pengutronix.de on lore.kernel.org?
Hello Konstantin, On Fri, Nov 26, 2021 at 11:28:30AM +0100, Uwe Kleine-König wrote: > On Tue, Apr 14, 2020 at 05:31:22PM -0400, Konstantin Ryabitsev via RT wrote: > > On Tue, Apr 14, 2020 at 05:52:53AM -0400, Uwe Kleine-König via RT wrote: > > > > > > > > I didn't hear back from you in reply to my request from three weeks ago. > > > > Is this still on your radar and I just need some more patience? > > > > > > I still didn't hear anything back from you. The ticket on > > > rt.linuxfoundation.org is still marked as "new". > > > > > > Following https://korg.wiki.kernel.org/userdoc/lore I'm currently in the > > > step "Wait till you receive a response", but since I'm there since more > > > than eight weeks I wonder if there is a problem on your side? > > > > Hello, and sorry for radio silence. I wasn't deliberately trying to > > stall -- I was just trying to figure out the best way to approach this. > > While it's true that there are some things on lore.kernel.org that > > aren't *directly* related to the kernel, we nevertheless try to stick to > > lists that are either about low-level kernel subsystems, or about > > various tools used for kernel development, or at least have predominant > > topics that are close to the above two categories. The trouble with > > ptxdist is that it doesn't really fit inside those boundaries, so I am > > not sure whether it makes sense for it to live on lore.kernel.org. > > > > Above all, I'd like to avoid turning lore.kernel.org into a central > > resource that becomes a hard dependency for projects the same way gmane > > was -- so much so, that when it disappeared, it caused serious ripple > > across the open-source fabric. > > > > We are encouraging others to set up their own instances of public-inbox, > > and would prefer to mirror the resulting git repositories instead (and > > offer indexing/searching capabilities) instead of doing archival on our > > end. This is the approach we're taking with Yocto, and I would encourage > > you to investigate the same option, as this would avoid making > > lore.kernel.org a hard dependency for your project. > > > > I hope that we'll be able to roll out our aggregated mirroring/indexing > > service closer to mid-year -- and when that happens, we'll be happy to > > mirror your lists and offer them alongside all others. > > I'd like to come back to this offer. We have setup our own public-inbox > instances now for a while and mirroring them on lore.kernel.org would be > great. > > These are the links to our instances: > > https://lore.barebox.org/barebox > https://lore.ptxdist.org/ptxdist > https://lore.distrokit.org/distrokit > https://lore.pengutronix.de/oss-tools > > Is there anything else you need, some meta-data maybe? I didn't hear back from you. I wonder if this mail didn't make it to a mailbox that is read, or if it just fell through the (human) cracks. It would be great to get these lists indexed on lore.kernel.org. happy to get feedback Uwe -- Pengutronix e.K. | Uwe Kleine-König| Industrial Linux Solutions | https://www.pengutronix.de/ | signature.asc Description: PGP signature ___ ptxdist mailing list ptxdist@pengutronix.de To unsubscribe, send a mail with subject "unsubscribe" to ptxdist-requ...@pengutronix.de
Re: [ptxdist] [Kernel.org Helpdesk #86171] [linuxfoundation.org #86171] Re: Archive ptxdist@pengutronix.de on lore.kernel.org?
Hello Konstantin, On Tue, Apr 14, 2020 at 05:31:22PM -0400, Konstantin Ryabitsev via RT wrote: > On Tue, Apr 14, 2020 at 05:52:53AM -0400, Uwe Kleine-König via RT wrote: > > > > > > I didn't hear back from you in reply to my request from three weeks ago. > > > Is this still on your radar and I just need some more patience? > > > > I still didn't hear anything back from you. The ticket on > > rt.linuxfoundation.org is still marked as "new". > > > > Following https://korg.wiki.kernel.org/userdoc/lore I'm currently in the > > step "Wait till you receive a response", but since I'm there since more > > than eight weeks I wonder if there is a problem on your side? > > Hello, and sorry for radio silence. I wasn't deliberately trying to > stall -- I was just trying to figure out the best way to approach this. > While it's true that there are some things on lore.kernel.org that > aren't *directly* related to the kernel, we nevertheless try to stick to > lists that are either about low-level kernel subsystems, or about > various tools used for kernel development, or at least have predominant > topics that are close to the above two categories. The trouble with > ptxdist is that it doesn't really fit inside those boundaries, so I am > not sure whether it makes sense for it to live on lore.kernel.org. > > Above all, I'd like to avoid turning lore.kernel.org into a central > resource that becomes a hard dependency for projects the same way gmane > was -- so much so, that when it disappeared, it caused serious ripple > across the open-source fabric. > > We are encouraging others to set up their own instances of public-inbox, > and would prefer to mirror the resulting git repositories instead (and > offer indexing/searching capabilities) instead of doing archival on our > end. This is the approach we're taking with Yocto, and I would encourage > you to investigate the same option, as this would avoid making > lore.kernel.org a hard dependency for your project. > > I hope that we'll be able to roll out our aggregated mirroring/indexing > service closer to mid-year -- and when that happens, we'll be happy to > mirror your lists and offer them alongside all others. I'd like to come back to this offer. We have setup our own public-inbox instances now for a while and mirroring them on lore.kernel.org would be great. These are the links to our instances: https://lore.barebox.org/barebox https://lore.ptxdist.org/ptxdist https://lore.distrokit.org/distrokit https://lore.pengutronix.de/oss-tools Is there anything else you need, some meta-data maybe? Thanks in advance Uwe -- Pengutronix e.K. | Uwe Kleine-König| Industrial Linux Solutions | https://www.pengutronix.de/ | signature.asc Description: PGP signature ___ ptxdist mailing list ptxdist@pengutronix.de To unsubscribe, send a mail with subject "unsubscribe" to ptxdist-requ...@pengutronix.de
[ptxdist] [PATCH] sdbus-cpp: version bump 0.8.2 -> 1.0.0
From: Ladislav Michl Signed-off-by: Ladislav Michl Signed-off-by: Uwe Kleine-König --- rules/sdbus-cpp.make | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/rules/sdbus-cpp.make b/rules/sdbus-cpp.make index 618873bbe90d..238030c31fe7 100644 --- a/rules/sdbus-cpp.make +++ b/rules/sdbus-cpp.make @@ -14,8 +14,8 @@ PACKAGES-$(PTXCONF_SDBUS_CPP) += sdbus-cpp # # Paths and names # -SDBUS_CPP_VERSION := 0.8.2 -SDBUS_CPP_MD5 := 29b02ecce98adbb34d8dcd22bc26cc53 +SDBUS_CPP_VERSION := 1.0.0 +SDBUS_CPP_MD5 := cf53be4342605419ba8d60d261a89466 SDBUS_CPP := sdbus-cpp-$(SDBUS_CPP_VERSION) SDBUS_CPP_SUFFIX := tar.gz SDBUS_CPP_URL := https://github.com/Kistler-Group/sdbus-cpp/archive/v$(SDBUS_CPP_VERSION).$(SDBUS_CPP_SUFFIX) -- 2.30.2 ___ ptxdist mailing list ptxdist@pengutronix.de To unsubscribe, send a mail with subject "unsubscribe" to ptxdist-requ...@pengutronix.de
[ptxdist] [PATCH] libmodbus: version bump 2.0.4 -> 3.0.8
From: Ladislav Michl Update to the latest stable version. As tarballs released on github are lacking generated configure script use the one provided on project pages (these are indeed different archives) Signed-off-by: Ladislav Michl Signed-off-by: Uwe Kleine-König --- rules/libmodbus.make | 16 ++-- 1 file changed, 6 insertions(+), 10 deletions(-) diff --git a/rules/libmodbus.make b/rules/libmodbus.make index 1335c85d07fd..3a709aa65f12 100644 --- a/rules/libmodbus.make +++ b/rules/libmodbus.make @@ -14,11 +14,11 @@ PACKAGES-$(PTXCONF_LIBMODBUS) += libmodbus # # Paths and names # -LIBMODBUS_VERSION := 2.0.4 -LIBMODBUS_MD5 := 6b3aa500ab441a953eeb73a8c58cdf76 +LIBMODBUS_VERSION := 3.0.8 +LIBMODBUS_MD5 := 69cb3ebe134d1f1e2768c2127ed9b5d6 LIBMODBUS := libmodbus-$(LIBMODBUS_VERSION) LIBMODBUS_SUFFIX := tar.gz -LIBMODBUS_URL := http://github.com/downloads/stephane/libmodbus/$(LIBMODBUS).$(LIBMODBUS_SUFFIX) +LIBMODBUS_URL := https://libmodbus.org/releases/$(LIBMODBUS).$(LIBMODBUS_SUFFIX) LIBMODBUS_SOURCE := $(SRCDIR)/$(LIBMODBUS).$(LIBMODBUS_SUFFIX) LIBMODBUS_DIR := $(BUILDDIR)/$(LIBMODBUS) LIBMODBUS_LICENSE := LGPL-3.0-only @@ -28,13 +28,9 @@ LIBMODBUS_LICENSE:= LGPL-3.0-only # -LIBMODBUS_PATH := PATH=$(CROSS_PATH) -LIBMODBUS_ENV := $(CROSS_ENV) - -# -# autoconf -# -LIBMODBUS_AUTOCONF := $(CROSS_AUTOCONF_USR) +LIBMODBUS_CONF_TOOL := autoconf +LIBMODBUS_CONF_OPT := \ + $(CROSS_AUTOCONF_USR) # # Target-Install -- 2.30.2 ___ ptxdist mailing list ptxdist@pengutronix.de To unsubscribe, send a mail with subject "unsubscribe" to ptxdist-requ...@pengutronix.de
Re: [ptxdist] [PATCH 2/2] chrony: version bump 3.5 -> 4.1
On Wed, Jul 07, 2021 at 04:52:23PM +0200, Bruno Thomsen wrote: > Den ons. 7. jul. 2021 kl. 16.31 skrev Michael Olbrich > : > > > > On Mon, Jul 05, 2021 at 09:49:25AM +0200, Uwe Kleine-König wrote: > > > --- > > > ...ke-unrecognized-option-a-fatal-error.patch | 25 - > > > patches/chrony-3.5/series | 4 --- > > > ...t-install-docs-to-not-need-additiona.patch | 21 +++ > > > ...vice-Make-this-a-non-forking-service.patch | 27 +++ > > > patches/chrony-4.1/series | 5 > > > rules/chrony.make | 4 +-- > > > 6 files changed, 55 insertions(+), 31 deletions(-) > > > delete mode 100644 > > > patches/chrony-3.5/0001-configure-make-unrecognized-option-a-fatal-error.patch > > > delete mode 100644 patches/chrony-3.5/series > > > create mode 100644 > > > patches/chrony-4.1/0001-Makefile.in-Don-t-install-docs-to-not-need-additiona.patch > > > create mode 100644 > > > patches/chrony-4.1/0002-chronyd.service-Make-this-a-non-forking-service.patch > > > create mode 100644 patches/chrony-4.1/series > > > > > > diff --git > > > a/patches/chrony-3.5/0001-configure-make-unrecognized-option-a-fatal-error.patch > > > > > > b/patches/chrony-3.5/0001-configure-make-unrecognized-option-a-fatal-error.patch > > > deleted file mode 100644 > > > index 01f2136338b9.. > > > --- > > > a/patches/chrony-3.5/0001-configure-make-unrecognized-option-a-fatal-error.patch > > > +++ /dev/null > > > @@ -1,25 +0,0 @@ > > > -From: Bruno Thomsen > > > -Date: Fri, 31 Jul 2020 09:07:49 + > > > -Subject: [PATCH] configure: make unrecognized option a fatal error > > > - > > > -This helps catch cases where options has been removed > > > -and are no longer valid. Including additional cross-build > > > -options that cannot be used. > > > - > > > -Signed-off-by: Bruno Thomsen > > Hi, > > This patch was added last year to ptxdist, so it's possible to detect > unrecognized options in new chrony versions as they use a custom > hand crafted configure script. So unless that situation has changed > this patch should be kept. I dropped it because it is only important when you do a version bump. As you usually go through all of the configure flags anyhow I didn't consider it useful enough to keep it. However I don't feel strong here, so if you do, I can keep it. Best regards Uwe -- Pengutronix e.K. | Uwe Kleine-König| Industrial Linux Solutions | https://www.pengutronix.de/ | signature.asc Description: PGP signature ___ ptxdist mailing list ptxdist@pengutronix.de To unsubscribe, send a mail with subject "unsubscribe" to ptxdist-requ...@pengutronix.de
[ptxdist] [PATCH 1/2] chrony: unify location of config files
The upstream default location of the config file is directly in /etc instead of /etc/chrony/ and also the systemd unit uses /etc. So switch the init script and ptxdist rules accordingly. Signed-off-by: Uwe Kleine-König --- projectroot/etc/init.d/chrony | 4 ++-- rules/chrony.make | 8 2 files changed, 6 insertions(+), 6 deletions(-) diff --git a/projectroot/etc/init.d/chrony b/projectroot/etc/init.d/chrony index d112927f825e..1602eec6dcb9 100644 --- a/projectroot/etc/init.d/chrony +++ b/projectroot/etc/init.d/chrony @@ -9,14 +9,14 @@ PATH=/sbin:/bin:/usr/bin:/usr/sbin BINARY="/usr/sbin/chronyd" -CONFIG="/etc/chrony/chrony.conf" +CONFIG="/etc/chrony.conf" PREFIX="chrony: " # This system doesn�t have full rtc ioctl support for # chrony statistic functions RTC_IOCTL="incomplete" # some chronyc commands need prior autentication: extract keys from config -KEY=$(awk '$1 ~ /^commandkey$/ { print $2; exit}' /etc/chrony/chrony.conf) +KEY=$(awk '$1 ~ /^commandkey$/ { print $2; exit}' "$CONFIG") PASSWORD=`awk '$1 ~ /^'$KEY'$/ {print $2; exit}' /etc/chrony/chrony.keys` # convenience functions diff --git a/rules/chrony.make b/rules/chrony.make index cbdf9314b52b..bdceb454c175 100644 --- a/rules/chrony.make +++ b/rules/chrony.make @@ -100,14 +100,14 @@ endif # generic one ifdef PTXCONF_CHRONY_INSTALL_CONFIG - @$(call install_alternative, chrony, 0, 0, 0644, /etc/chrony/chrony.conf) - @$(call install_alternative, chrony, 0, 0, 0600, /etc/chrony/chrony.keys) + @$(call install_alternative, chrony, 0, 0, 0644, /etc/chrony.conf) + @$(call install_alternative, chrony, 0, 0, 0600, /etc/chrony.keys) # modify placeholders with data from configuration - @$(call install_replace, chrony, /etc/chrony/chrony.conf, \ + @$(call install_replace, chrony, /etc/chrony.conf, \ @UNCONFIGURED_CHRONY_SERVER_IP@, $(PTXCONF_CHRONY_DEFAULT_NTP_SERVER)) - @$(call install_replace, chrony, /etc/chrony/chrony.keys, \ + @$(call install_replace, chrony, /etc/chrony.keys, \ @UNCONFIGURED_CHRONY_ACCESS_KEY@, $(PTXCONF_CHRONY_DEFAULT_ACCESS_KEY)) endif -- 2.30.2 ___ ptxdist mailing list ptxdist@pengutronix.de To unsubscribe, send a mail with subject "unsubscribe" to ptxdist-requ...@pengutronix.de
[ptxdist] [PATCH 2/2] chrony: version bump 3.5 -> 4.1
--- ...ke-unrecognized-option-a-fatal-error.patch | 25 - patches/chrony-3.5/series | 4 --- ...t-install-docs-to-not-need-additiona.patch | 21 +++ ...vice-Make-this-a-non-forking-service.patch | 27 +++ patches/chrony-4.1/series | 5 rules/chrony.make | 4 +-- 6 files changed, 55 insertions(+), 31 deletions(-) delete mode 100644 patches/chrony-3.5/0001-configure-make-unrecognized-option-a-fatal-error.patch delete mode 100644 patches/chrony-3.5/series create mode 100644 patches/chrony-4.1/0001-Makefile.in-Don-t-install-docs-to-not-need-additiona.patch create mode 100644 patches/chrony-4.1/0002-chronyd.service-Make-this-a-non-forking-service.patch create mode 100644 patches/chrony-4.1/series diff --git a/patches/chrony-3.5/0001-configure-make-unrecognized-option-a-fatal-error.patch b/patches/chrony-3.5/0001-configure-make-unrecognized-option-a-fatal-error.patch deleted file mode 100644 index 01f2136338b9.. --- a/patches/chrony-3.5/0001-configure-make-unrecognized-option-a-fatal-error.patch +++ /dev/null @@ -1,25 +0,0 @@ -From: Bruno Thomsen -Date: Fri, 31 Jul 2020 09:07:49 + -Subject: [PATCH] configure: make unrecognized option a fatal error - -This helps catch cases where options has been removed -and are no longer valid. Including additional cross-build -options that cannot be used. - -Signed-off-by: Bruno Thomsen - configure | 1 + - 1 file changed, 1 insertion(+) - -diff --git a/configure b/configure -index 323f5743cd57..080f7bce746d 100755 a/configure -+++ b/configure -@@ -388,6 +388,7 @@ do - ;; - * ) - echo "Unrecognized option : " $option -+exit 1 - esac - done - diff --git a/patches/chrony-3.5/series b/patches/chrony-3.5/series deleted file mode 100644 index 5019725586fd.. --- a/patches/chrony-3.5/series +++ /dev/null @@ -1,4 +0,0 @@ -# generated by git-ptx-patches -#tag:base --start-number 1 -0001-configure-make-unrecognized-option-a-fatal-error.patch -# c182e03aeda3f8c148ae51d52f839a6a - git-ptx-patches magic diff --git a/patches/chrony-4.1/0001-Makefile.in-Don-t-install-docs-to-not-need-additiona.patch b/patches/chrony-4.1/0001-Makefile.in-Don-t-install-docs-to-not-need-additiona.patch new file mode 100644 index ..13d3df8c7dcb --- /dev/null +++ b/patches/chrony-4.1/0001-Makefile.in-Don-t-install-docs-to-not-need-additiona.patch @@ -0,0 +1,21 @@ +From: =?UTF-8?q?Uwe=20Kleine-K=C3=B6nig?= +Date: Mon, 19 Apr 2021 17:33:41 +0200 +Subject: [PATCH] Makefile.in: Don't install docs to not need additional build + dependencies + +--- + Makefile.in | 1 - + 1 file changed, 1 deletion(-) + +diff --git a/Makefile.in b/Makefile.in +index 9a51df20074a..758e1e87771d 100644 +--- a/Makefile.in b/Makefile.in +@@ -95,7 +95,6 @@ install: chronyd chronyc + chmod 755 $(DESTDIR)$(SBINDIR)/chronyd + cp chronyc $(DESTDIR)$(BINDIR)/chronyc + chmod 755 $(DESTDIR)$(BINDIR)/chronyc +- $(MAKE) -C doc install + + docs : + $(MAKE) -C doc docs diff --git a/patches/chrony-4.1/0002-chronyd.service-Make-this-a-non-forking-service.patch b/patches/chrony-4.1/0002-chronyd.service-Make-this-a-non-forking-service.patch new file mode 100644 index ..39c1e0e79f16 --- /dev/null +++ b/patches/chrony-4.1/0002-chronyd.service-Make-this-a-non-forking-service.patch @@ -0,0 +1,27 @@ +From: =?UTF-8?q?Uwe=20Kleine-K=C3=B6nig?= +Date: Mon, 19 Apr 2021 17:34:34 +0200 +Subject: [PATCH] chronyd.service: Make this a non-forking service + +This allows to add -d to $OPTIONS without systemd killing the daemon +after a short uptime. +--- + examples/chronyd.service | 4 ++-- + 1 file changed, 2 insertions(+), 2 deletions(-) + +diff --git a/examples/chronyd.service b/examples/chronyd.service +index 289548cbab87..7a7c693d9e63 100644 +--- a/examples/chronyd.service b/examples/chronyd.service +@@ -6,10 +6,10 @@ Conflicts=ntpd.service systemd-timesyncd.service + ConditionCapability=CAP_SYS_TIME + + [Service] +-Type=forking ++Type=exec + PIDFile=/run/chrony/chronyd.pid + EnvironmentFile=-/etc/sysconfig/chronyd +-ExecStart=/usr/sbin/chronyd $OPTIONS ++ExecStart=/usr/sbin/chronyd -d $OPTIONS + PrivateTmp=yes + ProtectHome=yes + ProtectSystem=full diff --git a/patches/chrony-4.1/series b/patches/chrony-4.1/series new file mode 100644 index ..edccbe20d778 --- /dev/null +++ b/patches/chrony-4.1/series @@ -0,0 +1,5 @@ +# generated by git-ptx-patches +#tag:base --start-number 1 +0001-Makefile.in-Don-t-install-docs-to-not-need-additiona.patch +0002-chronyd.service-Make-this-a-non-forking-service.patch +# 615ea3ced1c5eafb46beca36157cbc23 - git-ptx-patches magic diff --git a/rules/chrony.make b/rules/chrony.make index bdceb454c175..6ba7da89da28 100644 --- a/rules/chrony.make +++ b/rules/chrony.make @@ -15,8 +15,8 @@ PACKAGES-$(PTXCONF_CHRONY) += chrony # # Paths and names # -CHRONY_VERSION :=
[ptxdist] [PATCH v3] chrony: Fix location of config files
chrony uses /etc/chrony.conf since (at least) the start of its git history in 2006. Signed-off-by: Uwe Kleine-König --- rules/chrony.make | 8 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/rules/chrony.make b/rules/chrony.make index 7cf59f30560c..49177e96fcb7 100644 --- a/rules/chrony.make +++ b/rules/chrony.make @@ -100,14 +100,14 @@ endif # generic one ifdef PTXCONF_CHRONY_INSTALL_CONFIG - @$(call install_alternative, chrony, 0, 0, 0644, /etc/chrony/chrony.conf) - @$(call install_alternative, chrony, 0, 0, 0600, /etc/chrony/chrony.keys) + @$(call install_alternative, chrony, 0, 0, 0644, /etc/chrony.conf) + @$(call install_alternative, chrony, 0, 0, 0600, /etc/chrony.keys) # modify placeholders with data from configuration - @$(call install_replace, chrony, /etc/chrony/chrony.conf, \ + @$(call install_replace, chrony, /etc/chrony.conf, \ @UNCONFIGURED_CHRONY_SERVER_IP@, $(PTXCONF_CHRONY_DEFAULT_NTP_SERVER)) - @$(call install_replace, chrony, /etc/chrony/chrony.keys, \ + @$(call install_replace, chrony, /etc/chrony.keys, \ @UNCONFIGURED_CHRONY_ACCESS_KEY@, $(PTXCONF_CHRONY_DEFAULT_ACCESS_KEY)) endif -- 2.30.2 ___ ptxdist mailing list ptxdist@pengutronix.de To unsubscribe, send a mail with subject "unsubscribe" to ptxdist-requ...@pengutronix.de
[ptxdist] [PATCH v2] chrony: Fix location of config files
chrony uses /etc/chrony.conf since (at least) the start of its git history in 2006. --- rules/chrony.make | 8 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/rules/chrony.make b/rules/chrony.make index 7cf59f30560c..49177e96fcb7 100644 --- a/rules/chrony.make +++ b/rules/chrony.make @@ -100,14 +100,14 @@ endif # generic one ifdef PTXCONF_CHRONY_INSTALL_CONFIG - @$(call install_alternative, chrony, 0, 0, 0644, /etc/chrony/chrony.conf) - @$(call install_alternative, chrony, 0, 0, 0600, /etc/chrony/chrony.keys) + @$(call install_alternative, chrony, 0, 0, 0644, /etc/chrony.conf) + @$(call install_alternative, chrony, 0, 0, 0600, /etc/chrony.keys) # modify placeholders with data from configuration - @$(call install_replace, chrony, /etc/chrony/chrony.conf, \ + @$(call install_replace, chrony, /etc/chrony.conf, \ @UNCONFIGURED_CHRONY_SERVER_IP@, $(PTXCONF_CHRONY_DEFAULT_NTP_SERVER)) - @$(call install_replace, chrony, /etc/chrony/chrony.keys, \ + @$(call install_replace, chrony, /etc/chrony.keys, \ @UNCONFIGURED_CHRONY_ACCESS_KEY@, $(PTXCONF_CHRONY_DEFAULT_ACCESS_KEY)) endif -- 2.30.2 ___ ptxdist mailing list ptxdist@pengutronix.de To unsubscribe, send a mail with subject "unsubscribe" to ptxdist-requ...@pengutronix.de
[ptxdist] [PATCH] chrony: Fix location of config files
chrony uses /etc/chrony.conf since (at least) the start of its git history in 2006. --- rules/chrony.make | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/rules/chrony.make b/rules/chrony.make index 7cf59f30560c..b8e5d694de2a 100644 --- a/rules/chrony.make +++ b/rules/chrony.make @@ -100,8 +100,8 @@ endif # generic one ifdef PTXCONF_CHRONY_INSTALL_CONFIG - @$(call install_alternative, chrony, 0, 0, 0644, /etc/chrony/chrony.conf) - @$(call install_alternative, chrony, 0, 0, 0600, /etc/chrony/chrony.keys) + @$(call install_alternative, chrony, 0, 0, 0644, /etc/chrony.conf) + @$(call install_alternative, chrony, 0, 0, 0600, /etc/chrony.keys) # modify placeholders with data from configuration @$(call install_replace, chrony, /etc/chrony/chrony.conf, \ -- 2.30.2 ___ ptxdist mailing list ptxdist@pengutronix.de To unsubscribe, send a mail with subject "unsubscribe" to ptxdist-requ...@pengutronix.de
[ptxdist] [PATCH] efax: New patch to fix build failure with gcc-10
Signed-off-by: Uwe Kleine-König --- .../0005-Fix-linker-error-with-gcc-10.patch | 45 +++ patches/efax-0.9/series | 3 +- 2 files changed, 47 insertions(+), 1 deletion(-) create mode 100644 patches/efax-0.9/0005-Fix-linker-error-with-gcc-10.patch diff --git a/patches/efax-0.9/0005-Fix-linker-error-with-gcc-10.patch b/patches/efax-0.9/0005-Fix-linker-error-with-gcc-10.patch new file mode 100644 index ..08f1d0a1a97a --- /dev/null +++ b/patches/efax-0.9/0005-Fix-linker-error-with-gcc-10.patch @@ -0,0 +1,45 @@ +From: =?UTF-8?q?Uwe=20Kleine-K=C3=B6nig?= +Date: Tue, 16 Mar 2021 12:33:05 +0100 +Subject: [PATCH] Fix linker error with gcc-10 + +This fixes: + + ld: efaxlib.o:(.bss+0x0): multiple definition of `reversebits'; efix.o:(.bss+0x0): first defined here + ld: efaxlib.o:(.bss+0x100): multiple definition of `normalbits'; efix.o:(.bss+0x100): first defined here + ld: efaxlib.o:(.bss+0x0): multiple definition of `reversebits'; efax.o:(.bss+0x0): first defined here + ld: efaxlib.o:(.bss+0x100): multiple definition of `normalbits'; efax.o:(.bss+0x100): first defined here + ld: efaxio.o:(.bss+0x0): multiple definition of `reversebits'; efax.o:(.bss+0x0): first defined here + ld: efaxio.o:(.bss+0x100): multiple definition of `normalbits'; efax.o:(.bss+0x100): first defined here + ld: efaxos.o:(.bss+0x0): multiple definition of `reversebits'; efax.o:(.bss+0x0): first defined here + ld: efaxos.o:(.bss+0x100): multiple definition of `normalbits'; efax.o:(.bss+0x100): first defined here +--- + efaxlib.c | 2 ++ + efaxlib.h | 2 +- + 2 files changed, 3 insertions(+), 1 deletion(-) + +diff --git a/efaxlib.c b/efaxlib.c +index 4887cd9e5d72..347e8bd326e2 100644 +--- a/efaxlib.c b/efaxlib.c +@@ -20,6 +20,8 @@ + #define DEFWIDTH 1728/* 215x297 mm image at fax resolution */ + #define DEFHEIGHT 2287 + ++uchar reversebits [ 256 ], normalbits [ 256 ] ; ++ + extern t4tab wtab [ ( 64 + 27 + 13 ) + 1 ] ; /* T.4 coding tables */ + extern t4tab btab [ ( 64 + 27 + 13 ) + 1 ] ; + +diff --git a/efaxlib.h b/efaxlib.h +index 9c725ae9a934..88505694dc0e 100644 +--- a/efaxlib.h b/efaxlib.h +@@ -207,7 +207,7 @@ int runor ( short *a, int na, short *b, int nb, short *c, int *pels ) ; + /* Bit reversal lookup tables (note that the `normalbits' array +is the one actually used for the bit reversal. */ + +-uchar reversebits [ 256 ], normalbits [ 256 ] ; ++extern uchar reversebits [ 256 ], normalbits [ 256 ] ; + + void initbittab(void) ; + diff --git a/patches/efax-0.9/series b/patches/efax-0.9/series index a67ca2181e0a..d4edeefb40ce 100644 --- a/patches/efax-0.9/series +++ b/patches/efax-0.9/series @@ -4,4 +4,5 @@ 0002-add-nullpointer-check.patch 0003-fix-l-option-in-efix.patch 0004-add-missing-include-file.patch -# 8338636f38d9624eaaf8c3530518be3d - git-ptx-patches magic +0005-Fix-linker-error-with-gcc-10.patch +# dc070db4a15b5fc5df6ab930fcf4bfdc - git-ptx-patches magic -- 2.29.2 ___ ptxdist mailing list ptxdist@pengutronix.de To unsubscribe, send a mail with subject "unsubscribe" to ptxdist-requ...@pengutronix.de
Re: [ptxdist] [PATCH] ssh: Don't terminate ssh connections on systemd target changes
Hello, On Mon, Mar 15, 2021 at 02:33:38PM +0100, Uwe Kleine-König wrote: > When switching the systemd target (e.g. by systemctl isolate > application.target) it is surprising when ssh sessions die. It's also > inconvenient if the target change is done via ssh ... > > So add IgnoreOnIsolate=true to sshd@.service. > > Signed-off-by: Uwe Kleine-König I just noticed I sent the very same patch with a different commit log already 10 days ago. I suffered some confusion about multiple git repositories and archived mail and so resent even though I thought I already sent this :-\ Pick whatever patch you consider the commit log better :-) Best regards Uwe -- Pengutronix e.K. | Uwe Kleine-König| Industrial Linux Solutions | https://www.pengutronix.de/ | signature.asc Description: PGP signature ___ ptxdist mailing list ptxdist@pengutronix.de To unsubscribe, send a mail with subject "unsubscribe" to ptxdist-requ...@pengutronix.de
[ptxdist] [PATCH] ssh: Don't terminate ssh connections on systemd target changes
When switching the systemd target (e.g. by systemctl isolate application.target) it is surprising when ssh sessions die. It's also inconvenient if the target change is done via ssh ... So add IgnoreOnIsolate=true to sshd@.service. Signed-off-by: Uwe Kleine-König --- projectroot/usr/lib/systemd/system/sshd@.service | 1 + 1 file changed, 1 insertion(+) diff --git a/projectroot/usr/lib/systemd/system/sshd@.service b/projectroot/usr/lib/systemd/system/sshd@.service index 3814e9f7b095..269127c8f3ac 100644 --- a/projectroot/usr/lib/systemd/system/sshd@.service +++ b/projectroot/usr/lib/systemd/system/sshd@.service @@ -1,6 +1,7 @@ [Unit] Description=SSH Per-Connection Server DefaultDependencies=no +IgnoreOnIsolate=true [Service] ExecStart=/usr/sbin/sshd -i -- 2.29.2 ___ ptxdist mailing list ptxdist@pengutronix.de To unsubscribe, send a mail with subject "unsubscribe" to ptxdist-requ...@pengutronix.de
[ptxdist] [PATCH] ifupdown: Drop default dependencies of ifupdown.service
ifupdown can usually be started quite early. There might be some additional dependencies (e.g. on devices or kernel modules) but these are not formalized with the default dependencies either. So in this case you have to overwrite the service for your device anyhow. Signed-off-by: Uwe Kleine-König --- projectroot/usr/lib/systemd/system/ifupdown.service | 1 + 1 file changed, 1 insertion(+) diff --git a/projectroot/usr/lib/systemd/system/ifupdown.service b/projectroot/usr/lib/systemd/system/ifupdown.service index f9dd5b473b63..4531ef44326c 100644 --- a/projectroot/usr/lib/systemd/system/ifupdown.service +++ b/projectroot/usr/lib/systemd/system/ifupdown.service @@ -1,6 +1,7 @@ [Unit] Description=Handle network interface with ifup/ifdown Before=network.target +DefaultDependencies=no [Service] Type=oneshot -- 2.29.2 ___ ptxdist mailing list ptxdist@pengutronix.de To unsubscribe, send a mail with subject "unsubscribe" to ptxdist-requ...@pengutronix.de
[ptxdist] [PATCH] rtl8188eu: New package for an out-of-tree wifi driver
The hardware is supported in Linux mainline, but has no support for AP mode. So you have to stick to the externally maintained driver added here for some use cases. Signed-off-by: Uwe Kleine-König --- rules/rtl8188eu.in | 9 ++ rules/rtl8188eu.make | 65 2 files changed, 74 insertions(+) create mode 100644 rules/rtl8188eu.in create mode 100644 rules/rtl8188eu.make diff --git a/rules/rtl8188eu.in b/rules/rtl8188eu.in new file mode 100644 index ..1284c0cdd36e --- /dev/null +++ b/rules/rtl8188eu.in @@ -0,0 +1,9 @@ +## SECTION=kernel + +config RTL8188EU + tristate + select KERNEL + prompt "rtl8188eu" + help + Kernel module to support some Realtek USB wifi dongles; + taken from https://github.com/lwfinger/rtl8188eu . diff --git a/rules/rtl8188eu.make b/rules/rtl8188eu.make new file mode 100644 index ..25d6288e7079 --- /dev/null +++ b/rules/rtl8188eu.make @@ -0,0 +1,65 @@ +# -*-makefile-*- +# +# Copyright (C) 2021 by Uwe Kleine-Koenig +# +# For further information about the PTXdist project and license conditions +# see the README file. +# + +# +# We provide this package +# +PACKAGES-$(PTXCONF_RTL8188EU) += rtl8188eu + +# +# Paths and names and versions +# +RTL8188EU_VERSION := 20210304-780-g6d74135f34d3 +RTL8188EU_MD5 := 64c077f1b7855708369654a1d519a3b5 +RTL8188EU := rtl8188eu-$(RTL8188EU_VERSION) +RTL8188EU_URL := https://github.com/lwfinger/rtl8188eu/archive/$(RTL8188EU_VERSION).tar.gz +RTL8188EU_SOURCE := $(SRCDIR)/$(RTL8188EU).tar.gz +RTL8188EU_DIR := $(BUILDDIR)/$(RTL8188EU) +RTL8188EU_LICENSE := GPL-2.0 +RTL8188EU_LICENSE_FILES:= file://COPYING;md5=4fbd65380cdd255951079008b364516c + +ifneq ($(filter $(if $(PTX_COLLECTION),y,y m),$(PTXCONF_RTL8188EU)),) +$(STATEDIR)/kernel.targetinstall.post: $(STATEDIR)/rtl8188eu.targetinstall +endif + +# +# Prepare +# + +RTL8188EU_WRAPPER_BLACKLIST = $(KERNEL_WRAPPER_BLACKLIST) + +RTL8188EU_CONF_TOOL := NO + +# +# Compile +# + +RTL8188EU_MAKE_OPT = \ + $(KERNEL_MODULE_OPT) \ + -C $(KERNEL_DIR) \ + M=$(RTL8188EU_DIR) \ + modules + +# +# Install +# + +$(STATEDIR)/rtl8188eu.install: + @$(call targetinfo) + @$(call touch) + +# +# Target-Install +# + +$(STATEDIR)/rtl8188eu.targetinstall: + @$(call targetinfo) + @$(call compile, RTL8188EU, $(RTL8188EU_MAKE_OPT)_install) + @$(call touch) + +# vim: syntax=make -- 2.29.2 ___ ptxdist mailing list ptxdist@pengutronix.de To unsubscribe, send a mail with subject "unsubscribe" to ptxdist-requ...@pengutronix.de
[ptxdist] [PATCH] openssh: Don't kill ssh sessions when isolating a unit
When system state is switched using systemctl isolate ssh sessions usually shouldn't be killed. Make this the default. Signed-off-by: Uwe Kleine-König --- projectroot/usr/lib/systemd/system/sshd@.service | 1 + 1 file changed, 1 insertion(+) diff --git a/projectroot/usr/lib/systemd/system/sshd@.service b/projectroot/usr/lib/systemd/system/sshd@.service index 3814e9f7b095..269127c8f3ac 100644 --- a/projectroot/usr/lib/systemd/system/sshd@.service +++ b/projectroot/usr/lib/systemd/system/sshd@.service @@ -1,6 +1,7 @@ [Unit] Description=SSH Per-Connection Server DefaultDependencies=no +IgnoreOnIsolate=true [Service] ExecStart=/usr/sbin/sshd -i -- 2.29.2 ___ ptxdist mailing list ptxdist@pengutronix.de To unsubscribe, send a mail with subject "unsubscribe" to ptxdist-requ...@pengutronix.de
Re: [ptxdist] [PATCH] /etc/hosts: add local ipv6 definitions
On Mon, Mar 01, 2021 at 09:45:35AM +0100, Michael Olbrich wrote: > Hi, > > On Mon, Mar 01, 2021 at 06:43:40AM +0100, Alexander Dahl wrote: > > > Roland Hieber hat am 26.02.2021 14:24 geschrieben: > > > Hmmm, I wonder if this works even if GLOBAL_IPV6 is turned off. Or > > I just tried this: > > root@DistroKit:~ ping ip6-localhost > PING ip6-localhost (127.0.0.1): 56 data bytes > 64 bytes from 127.0.0.1: seq=0 ttl=64 time=9.578 ms Does it really say 127.0.0.1 if you ping ip6-localhost? On my distrokit build I cannot reproduce that: root@DistroKit:~ ping -c3 ip6-localhost PING ip6-localhost (::1): 56 data bytes 64 bytes from ::1: seq=0 ttl=64 time=1.209 ms 64 bytes from ::1: seq=1 ttl=64 time=4.836 ms 64 bytes from ::1: seq=2 ttl=64 time=3.668 ms --- ip6-localhost ping statistics --- 3 packets transmitted, 3 packets received, 0% packet loss round-trip min/avg/max = 1.209/3.237/4.836 ms Best regards Uwe -- Pengutronix e.K. | Uwe Kleine-König| Industrial Linux Solutions | https://www.pengutronix.de/ | signature.asc Description: PGP signature ___ ptxdist mailing list ptxdist@pengutronix.de To unsubscribe, send a mail with subject "unsubscribe" to ptxdist-requ...@pengutronix.de
[ptxdist] [PATCH] /etc/hosts: add local ipv6 definitions
It's 2021! Signed-off-by: Uwe Kleine-König --- projectroot/etc/hosts | 4 1 file changed, 4 insertions(+) diff --git a/projectroot/etc/hosts b/projectroot/etc/hosts index ba712fe03341..4389b1ebc0df 100644 --- a/projectroot/etc/hosts +++ b/projectroot/etc/hosts @@ -1 +1,5 @@ 127.0.0.1 localhost + +::1localhost ip6-localhost ip6-loopback +ff02::1ip6-allnodes +ff02::2ip6-allrouters -- 2.30.0 ___ ptxdist mailing list ptxdist@pengutronix.de To unsubscribe, send a mail with subject "unsubscribe" to ptxdist-requ...@pengutronix.de
[ptxdist] [PATCH] canfestival: Two new patches that make canfestival headers more robust
Signed-off-by: Uwe Kleine-König --- Hello, I already sent this patch approx. a year ago, for reasons I didn't try to understand the added patches failed to apply for Michael Olbrich. Since then they were used unmodified in a BSP without issues, so I'm resending for another try. Best regards Uwe ...4-let-canfestival.h-include-config.h.patch | 39 ++ -instead-of-include-.-for-own-files.patch | 118 ++ patches/canfestival-3+hg20180126.794/series | 4 +- 3 files changed, 160 insertions(+), 1 deletion(-) create mode 100644 patches/canfestival-3+hg20180126.794/0004-let-canfestival.h-include-config.h.patch create mode 100644 patches/canfestival-3+hg20180126.794/0005-Use-include-.-instead-of-include-.-for-own-files.patch diff --git a/patches/canfestival-3+hg20180126.794/0004-let-canfestival.h-include-config.h.patch b/patches/canfestival-3+hg20180126.794/0004-let-canfestival.h-include-config.h.patch new file mode 100644 index ..c6b005703e03 --- /dev/null +++ b/patches/canfestival-3+hg20180126.794/0004-let-canfestival.h-include-config.h.patch @@ -0,0 +1,39 @@ +From: =?UTF-8?q?Uwe=20Kleine-K=C3=B6nig?= +Date: Mon, 29 Jan 2018 14:07:18 +0100 +Subject: [PATCH] let canfestival.h include config.h + +This is necessary to make the header self-contained. Otherwise it is necessary to do + + #include + #include + +in the API user which might pull in a local config.h. +--- + include/sdo.h | 1 + + include/unix/canfestival.h | 1 + + 2 files changed, 2 insertions(+) + +diff --git a/include/sdo.h b/include/sdo.h +index 28b8c5c414eb..7863bb5c05a0 100644 +--- a/include/sdo.h b/include/sdo.h +@@ -36,6 +36,7 @@ Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + + struct struct_s_transfer; + ++#include "config.h" + #include "timer.h" + + /* Block mode : Data consumer receive step +diff --git a/include/unix/canfestival.h b/include/unix/canfestival.h +index f20e593dcf89..ee2a0bc0c4ae 100755 +--- a/include/unix/canfestival.h b/include/unix/canfestival.h +@@ -5,6 +5,7 @@ + extern "C" { + #endif + ++#include "config.h" + #include "timerscfg.h" + #include "can_driver.h" + #include "data.h" diff --git a/patches/canfestival-3+hg20180126.794/0005-Use-include-.-instead-of-include-.-for-own-files.patch b/patches/canfestival-3+hg20180126.794/0005-Use-include-.-instead-of-include-.-for-own-files.patch new file mode 100644 index ..8feaa48a8416 --- /dev/null +++ b/patches/canfestival-3+hg20180126.794/0005-Use-include-.-instead-of-include-.-for-own-files.patch @@ -0,0 +1,118 @@ +From: =?UTF-8?q?Uwe=20Kleine-K=C3=B6nig?= +Date: Wed, 29 Apr 2020 22:54:55 +0200 +Subject: [PATCH] Use #include "..." instead of #include <...> for own files + +This fixes inclusion if the consumer used + + #include + +(and not -I/usr/include/canfestival) as with + + #include <...> + +the local files are not found in this case. +--- + include/emcy.h | 2 +- + include/lifegrd.h | 2 +- + include/nmtSlave.h | 2 +- + include/objacces.h | 2 +- + include/pdo.h | 4 ++-- + include/states.h | 2 +- + include/timer.h| 4 ++-- + 7 files changed, 9 insertions(+), 9 deletions(-) + +diff --git a/include/emcy.h b/include/emcy.h +index d3a19eb6d335..ad7ce933eebd 100644 +--- a/include/emcy.h b/include/emcy.h +@@ -39,7 +39,7 @@ Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + #define __emcy_h__ + + +-#include ++#include "applicfg.h" + + /* The error states + * - */ +diff --git a/include/lifegrd.h b/include/lifegrd.h +index a0600a717038..222a3dd18f73 100644 +--- a/include/lifegrd.h b/include/lifegrd.h +@@ -53,7 +53,7 @@ Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + #define __lifegrd_h__ + + +-#include ++#include "applicfg.h" + + typedef void (*heartbeatError_t)(CO_Data*, UNS8); + void _heartbeatError(CO_Data* d, UNS8 heartbeatID); +diff --git a/include/nmtSlave.h b/include/nmtSlave.h +index 0a57ae878364..ffb720194e15 100755 +--- a/include/nmtSlave.h b/include/nmtSlave.h +@@ -28,7 +28,7 @@ Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + #ifndef __nmtSlave_h__ + #define __nmtSlave_h__ + +-#include ++#include "applicfg.h" + #include "data.h" + + /** +diff --git a/include/objacces.h b/include/objacces.h +index 26a96542a4ca..05dd46d313d7 100755 +--- a/include/objacces.h b/include/objacces.h +@@ -39,7 +39,7 @@ Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + #ifndef __objacces_h__ + #define __objacces_h__ + +-#include ++#include "applicfg.h" + + + #ifdef __cplusplus +diff --git a/include/pdo.h b/include/pdo.h +index d276d51734fc..713167f50071 100755 +--- a/include/pdo.h b/include/pdo.h +@@ -29,8 +29,8 @@ Foundation
Re: [ptxdist] [PATCH 1/3] mtd-utils: spring cleanup
Hello Michael, On Tue, Jan 19, 2021 at 03:46:24PM +0100, Uwe Kleine-König wrote: > Drop unused config item MTD_UTILS_JITTERTEST which is unused since more > than 15 years. > > Reactivate MTD_UTILS_JFFS2READER to install jffs2reader. This tool was > claimed to be removed upstream in commit ba0a8f624d02 (" * mtd-utils: > removed tools which are not in upstream any more"), which I failed to > find in the upstream git history. > > mtdpart is a tool that was introduced in mtd-utils 1.5.2, add an > installation option for it. you applied patches 2 and 3 but not this one and also didn't comment. Did you miss this patch? Best regards Uwe -- Pengutronix e.K. | Uwe Kleine-König| Industrial Linux Solutions | https://www.pengutronix.de/ | signature.asc Description: PGP signature ___ ptxdist mailing list ptxdist@pengutronix.de To unsubscribe, send a mail with subject "unsubscribe" to ptxdist-requ...@pengutronix.de
[ptxdist] [PATCH 3/3] mtd-utils: Add support for ubifs mount helper
This helper simplifies mounting ubifs filesystems via /etc/fstab. This helper already exists upstream, is not released yet, though. Signed-off-by: Uwe Kleine-König --- .../0002-Add-an-ubifs-mount-helper.patch | 142 ++ patches/mtd-utils-2.1.2/series| 3 +- rules/mtd-utils.in| 6 + rules/mtd-utils.make | 4 + 4 files changed, 154 insertions(+), 1 deletion(-) create mode 100644 patches/mtd-utils-2.1.2/0002-Add-an-ubifs-mount-helper.patch diff --git a/patches/mtd-utils-2.1.2/0002-Add-an-ubifs-mount-helper.patch b/patches/mtd-utils-2.1.2/0002-Add-an-ubifs-mount-helper.patch new file mode 100644 index ..f45b660ee2d1 --- /dev/null +++ b/patches/mtd-utils-2.1.2/0002-Add-an-ubifs-mount-helper.patch @@ -0,0 +1,142 @@ +From: =?UTF-8?q?Uwe=20Kleine-K=C3=B6nig?= +Date: Tue, 6 Oct 2020 11:19:13 +0200 +Subject: [PATCH] Add an ubifs mount helper +MIME-Version: 1.0 +Content-Type: text/plain; charset=UTF-8 +Content-Transfer-Encoding: 8bit + +This abstracts away attaching of the right ubi and then selecting the right +ubi device and volume to mount. + +As described in the comment at the top this allows to mount ubifs volumes +directly from /etc/fstab without having to use hardcoded numbers (which +depend on mount order and so are unreliable) and extra magic to care for +attaching. + +Signed-off-by: Uwe Kleine-König +Signed-off-by: David Oberhollenzer +Origin: upstream, commit:efeba0875ed181e7c1c7915742a3868799604d0c +--- + ubifs-utils/Makemodule.am | 2 + + ubifs-utils/mount.ubifs | 101 ++ + 2 files changed, 103 insertions(+) + create mode 100755 ubifs-utils/mount.ubifs + +diff --git a/ubifs-utils/Makemodule.am b/ubifs-utils/Makemodule.am +index 59109ccd613c..5c5d99f7572b 100644 +--- a/ubifs-utils/Makemodule.am b/ubifs-utils/Makemodule.am +@@ -47,4 +47,6 @@ UBIFS_EXTRA = \ + + EXTRA_DIST += $(UBIFS_HEADER) $(UBIFS_EXTRA) + ++dist_sbin_SCRIPTS = ubifs-utils/mount.ubifs ++ + sbin_PROGRAMS += $(UBIFS_BINS) +diff --git a/ubifs-utils/mount.ubifs b/ubifs-utils/mount.ubifs +new file mode 100755 +index ..b94ddc5649f4 +--- /dev/null b/ubifs-utils/mount.ubifs +@@ -0,0 +1,101 @@ ++#!/bin/sh ++ ++# This script should be installed as /sbin/mount.ubifs. The benefit is that an ++# fstab entry like: ++# ++# mtd=mtddev:home /homeubifs defaults 0 0 ++# ++# results in the ubi contained in the mtd named "mtddev" to be attached (if not ++# already done) and then the volume named "home" being mounted to /home. ++ ++# This is called by mount with the following options: ++# /sbin/mount.ubifs spec dir [-sfnv] [-N namespace] [-o options] [-t type.subtype] ++ ++spec="$1" ++shift ++ ++mtdname2num() { ++ local name ++ ++ name="$1" ++ ++ for d in $(find /sys/class/mtd/ -regex '.*/mtd[0-9]*'); do ++ case "$d" in ++ *ro) ++ continue ++ ;; ++ esac ++ ++ if test "$name" = "$(cat "$d/name")"; then ++ local dev mtdnum ++ ++ dev="$(basename "$d")" ++ mtdnum="${dev#mtd}" ++ echo "$mtdnum" ++ return ++ fi ++ done ++ ++ return 1 ++} ++ ++mtdnum2ubi() { ++ local mtdnum ++ ++ mtdnum="$1" ++ ++ for d in $(find /sys/class/ubi/ -regex '.*/ubi[0-9]*'); do ++ case "$d" in ++ *_[0-9]*) ++ continue ++ ;; ++ esac ++ ++ if test "$mtdnum" = "$(cat "$d/mtd_num")"; then ++ local ubi ++ ++ ubi="$(basename "$d")" ++ echo "$ubi" ++ return; ++ fi ++ done ++ ++ return 1 ++} ++ ++mtdnum2ubi_autoattach() { ++ local mtdnum ubi ++ ++ mtdnum="$1" ++ ++ ubi="$(mtdnum2ubi "$mtdnum")" && { echo "$ubi"; return; } ++ ++ # ubiattach might fail with "mtdX is already attached to ubiY" if there ++ # is more than one mount to do in the same mtd partition. So ignore errors. ++ ubiattach -m "$mtdnum" >&2 || true ++ ++ mtdnum2ubi "$mtdnum" ++} ++ ++case "$spec" in ++ mtd=*:*) ++ spec="${spec#mtd=}" ++ mtd="${spec%:*}" ++ rspec="${spec#*:}" ++ ++ mtdnum="$(mtdname2num "$mtd")" || { ++ echo "Failed to find mtdnum for mtd \"$mtd\"" ++
[ptxdist] [PATCH 2/3] mtd-utils: version bump 2.1.1 -> 2.1.2
This brings a new tool ubihealthd. Signed-off-by: Uwe Kleine-König --- .../0001-nandmarkbad-new-util-to-mark-blocks-as-bad.patch | 0 patches/{mtd-utils-2.1.1 => mtd-utils-2.1.2}/autogen.sh | 0 patches/{mtd-utils-2.1.1 => mtd-utils-2.1.2}/series | 0 rules/mtd-utils.in| 7 +++ rules/mtd-utils.make | 8 ++-- 5 files changed, 13 insertions(+), 2 deletions(-) rename patches/{mtd-utils-2.1.1 => mtd-utils-2.1.2}/0001-nandmarkbad-new-util-to-mark-blocks-as-bad.patch (100%) rename patches/{mtd-utils-2.1.1 => mtd-utils-2.1.2}/autogen.sh (100%) rename patches/{mtd-utils-2.1.1 => mtd-utils-2.1.2}/series (100%) diff --git a/patches/mtd-utils-2.1.1/0001-nandmarkbad-new-util-to-mark-blocks-as-bad.patch b/patches/mtd-utils-2.1.2/0001-nandmarkbad-new-util-to-mark-blocks-as-bad.patch similarity index 100% rename from patches/mtd-utils-2.1.1/0001-nandmarkbad-new-util-to-mark-blocks-as-bad.patch rename to patches/mtd-utils-2.1.2/0001-nandmarkbad-new-util-to-mark-blocks-as-bad.patch diff --git a/patches/mtd-utils-2.1.1/autogen.sh b/patches/mtd-utils-2.1.2/autogen.sh similarity index 100% rename from patches/mtd-utils-2.1.1/autogen.sh rename to patches/mtd-utils-2.1.2/autogen.sh diff --git a/patches/mtd-utils-2.1.1/series b/patches/mtd-utils-2.1.2/series similarity index 100% rename from patches/mtd-utils-2.1.1/series rename to patches/mtd-utils-2.1.2/series diff --git a/rules/mtd-utils.in b/rules/mtd-utils.in index 1a56d12dccef..cdf8b15adfac 100644 --- a/rules/mtd-utils.in +++ b/rules/mtd-utils.in @@ -288,6 +288,13 @@ config MTD_UTILS_UBIFORMAT help Format an MTD device +config MTD_UTILS_UBIHEALTHD + bool + prompt "ubihealthd" + help + Daemon that randomly scans each PEB of a UBI device to ensure that + filesystems with little reading do enough wear leveling. + config MTD_UTILS_UBIMKVOL bool prompt "ubimkvol" diff --git a/rules/mtd-utils.make b/rules/mtd-utils.make index df7db047624f..a5a12b3d14a9 100644 --- a/rules/mtd-utils.make +++ b/rules/mtd-utils.make @@ -15,8 +15,8 @@ PACKAGES-$(PTXCONF_MTD_UTILS) += mtd-utils # # Paths and names # -MTD_UTILS_VERSION := 2.1.1 -MTD_UTILS_MD5 := 94bbd31b217a5169ae26ab8c0442f691 +MTD_UTILS_VERSION := 2.1.2 +MTD_UTILS_MD5 := 19191bc0195a779c0bd1284c886084ab MTD_UTILS := mtd-utils-$(MTD_UTILS_VERSION) MTD_UTILS_SUFFIX := tar.bz2 MTD_UTILS_URL := ftp://ftp.infradead.org/pub/mtd-utils/$(MTD_UTILS).$(MTD_UTILS_SUFFIX) @@ -197,6 +197,10 @@ ifdef PTXCONF_MTD_UTILS_UBICRC32 @$(call install_copy, mtd-utils, 0, 0, 0755, -, \ /usr/sbin/ubicrc32) endif +ifdef PTXCONF_MTD_UTILS_UBIHEALTHD + @$(call install_copy, mtd-utils, 0, 0, 0755, -, \ + /usr/sbin/ubihealthd) +endif ifdef PTXCONF_MTD_UTILS_UBIMKVOL @$(call install_copy, mtd-utils, 0, 0, 0755, -, \ /usr/sbin/ubimkvol) -- 2.30.0 ___ ptxdist mailing list ptxdist@pengutronix.de To unsubscribe, send a mail with subject "unsubscribe" to ptxdist-requ...@pengutronix.de
[ptxdist] [PATCH 1/3] mtd-utils: spring cleanup
Drop unused config item MTD_UTILS_JITTERTEST which is unused since more than 15 years. Reactivate MTD_UTILS_JFFS2READER to install jffs2reader. This tool was claimed to be removed upstream in commit ba0a8f624d02 (" * mtd-utils: removed tools which are not in upstream any more"), which I failed to find in the upstream git history. mtdpart is a tool that was introduced in mtd-utils 1.5.2, add an installation option for it. Signed-off-by: Uwe Kleine-König --- rules/mtd-utils.in | 25 ++--- rules/mtd-utils.make | 11 +++ 2 files changed, 21 insertions(+), 15 deletions(-) diff --git a/rules/mtd-utils.in b/rules/mtd-utils.in index 8988ba116bdc..1a56d12dccef 100644 --- a/rules/mtd-utils.in +++ b/rules/mtd-utils.in @@ -134,23 +134,12 @@ config MTD_UTILS_JFFS2_DUMP help This utility dumps the contents of a binary JFFS2 image. -#config MTD_UTILS_JFFS2READER -# bool -# prompt "jffs2reader" -# select MTD_UTILS_JFFS -# help -#A jffs2 image reader - -config MTD_UTILS_JITTERTEST +config MTD_UTILS_JFFS2READER bool - prompt "jittertest" + prompt "jffs2reader" + select MTD_UTILS_JFFS help - This program is used to measure what the jitter of a - real time task would be under "standard" Linux. - - More particularly, what is the effect of running - a real time task under Linux with background - JFFS file system activity. + A jffs2 image reader config MTD_UTILS_MTDDEBUG bool @@ -158,6 +147,12 @@ config MTD_UTILS_MTDDEBUG help Does info, read, write and erase on mtd devices +config MTD_UTILS_MTDPART + bool + prompt "mtdpart" + help + Add or remove a partition from an MTD device + config MTD_UTILS_NANDDUMP bool prompt "nanddump" diff --git a/rules/mtd-utils.make b/rules/mtd-utils.make index 52dddae8e567..df7db047624f 100644 --- a/rules/mtd-utils.make +++ b/rules/mtd-utils.make @@ -118,10 +118,21 @@ ifdef PTXCONF_MTD_UTILS_JFFS2_DUMP @$(call install_copy, mtd-utils, 0, 0, 0755, -, \ /usr/sbin/jffs2dump) endif +ifdef PTXCONF_MTD_UTILS_JFFS2READER + @$(call install_copy, mtd-utils, 0, 0, 0755, -, \ + /usr/sbin/jffs2reader) +endif + ifdef PTXCONF_MTD_UTILS_MTDDEBUG @$(call install_copy, mtd-utils, 0, 0, 0755, -, \ /usr/sbin/mtd_debug) endif + +ifdef PTXCONF_MTD_UTILS_MTDPART + @$(call install_copy, mtd-utils, 0, 0, 0755, -, \ + /usr/sbin/mtdpart) +endif + ifdef PTXCONF_MTD_UTILS_NANDDUMP @$(call install_copy, mtd-utils, 0, 0, 0755, -, \ /usr/sbin/nanddump) -- 2.30.0 ___ ptxdist mailing list ptxdist@pengutronix.de To unsubscribe, send a mail with subject "unsubscribe" to ptxdist-requ...@pengutronix.de
Re: [ptxdist] [PATCH v3] rootfs: Fix creation of the /media directory
On Thu, Oct 08, 2020 at 09:32:53AM +0200, Gavin Schenk wrote: > Ensure that with PTXCONF_ROOTFS_MEDIA set directory '/media' is created > even if none of PTXCONF_ROOTFS_MEDIA[1-6] is set. > > On our read-only system we mount a tmpfs on directory '/media' in order > to dynamically create additional mount points there. > > Fixes: 63e2d8696 (rootfs: don't create empty /media subdirs) > Signed-off-by: Gavin Schenk > --- > Hi, > > the "messed whitespaces" of changes in our mail server software. > Our IT department implemented a workaround, I hope this > mail looks better. It seems the workaround worked fine. I didn't try to apply, but at least now there are tabs in your patch. Acked-by: Uwe Kleine-König Best regards Uwe -- Pengutronix e.K. | Uwe Kleine-König| Industrial Linux Solutions | https://www.pengutronix.de/ | signature.asc Description: PGP signature ___ ptxdist mailing list ptxdist@pengutronix.de To unsubscribe, send a mail with subject "unsubscribe" to ptxdist-requ...@pengutronix.de
Re: [ptxdist] [PATCH] rootfs: Fix creation of the /media directory
Hallo Gavin, On Mon, Oct 05, 2020 at 11:16:16AM +0200, Gavin Schenk wrote: > The directory /media is created depending on PTXCONF_ROOTFS_MEDIA only. I'd write: Ensure that with PTXCONF_ROOTFS_MEDIA set /media is created even if none of PTXCONF_ROOTFS_MEDIA[1-6] is set. (Is that what you intended to say?) > On our read-only system we mount a tmpfs on /media, so that we can > dynamically create additional mount points. With PTXCONF_ROOTFS_MEDIA set, > but none of PTXCONF_ROOTFS_MEDIA[1-6], the /media directory was not created > and mounting failed. > > fixes: 63e2d8696124083fc The usual format here (at least in the kernel) is: Fixes: 63e2d8696124 ("rootfs: don't create empty /media subdirs") (I have in my ~/.gitconfig: [alias] oneq = show -s --pretty='format:%h (\"%s\")' for that.) Other than that the change looks good. Best regards Uwe -- Pengutronix e.K. | Uwe Kleine-König| Industrial Linux Solutions | https://www.pengutronix.de/ | signature.asc Description: PGP signature ___ ptxdist mailing list ptxdist@pengutronix.de To unsubscribe, send a mail with subject "unsubscribe" to ptxdist-requ...@pengutronix.de
[ptxdist] [PATCH v2] git-ptx-patches: support flag -n to get unnumbered patches
When passing -n to git-ptx-patches the filenames in the patch stack don't get a number as prefix. The obvious downside is that the order of the patches isn't obvious without the series file and you cannot create a patch stack if two patches share have the same shortlog (which ideally shouldn't happen ...) The obvious advantage is that when a patch is added or removed in the middle of the series there are no renames for the later patches necessary and the changes to series are smaller accordingly. So if the patchstack of openssl used unnumbered patches the commit c45d66cdae4b ("openssl: remove engines-path patch") would only have to touch two lines in series (i.e. the removal of the patch and the changed checksum) and contain no renames. Signed-off-by: Uwe Kleine-König --- Hello, On Tue, Sep 22, 2020 at 09:48:44AM +0200, Michael Olbrich wrote: > Hmmm, I prefer numbers for upstream, but if someone want's patches like > this in a BSP, then I don't mind. I actually expected that and so kept the default as is. > However, this patch does not apply on master. Please rebase. Done with this patch. Best regards Uwe scripts/git-ptx-patches | 17 +++-- 1 file changed, 15 insertions(+), 2 deletions(-) diff --git a/scripts/git-ptx-patches b/scripts/git-ptx-patches index fe680273085b..772d20992838 100755 --- a/scripts/git-ptx-patches +++ b/scripts/git-ptx-patches @@ -34,6 +34,7 @@ fi remove_old=no tag=base +numbered_patches=true if grep -q "$PTX_PATCHES_HEADER" .ptxdist/series; then echo "Found series file generated by git-ptx-patches." @@ -52,7 +53,7 @@ if [ "x$1" = "x--force-remove" ]; then shift fi -while getopts "ft:n:" opt; do +while getopts "ft:n" opt; do case "${opt}" in f) remove_old="force" @@ -61,6 +62,9 @@ while getopts "ft:n:" opt; do tag="${OPTARG}" range="${tag}" ;; + n) + numbered_patches=false + ;; esac done shift $((${OPTIND} - 1)) @@ -144,7 +148,16 @@ fi GIT_EXTRA_ARGS="$GIT_EXTRA_ARGS --summary --stat=80" cat .ptxdist/series.0 > .ptxdist/series -${GIT} format-patch -N $GIT_EXTRA_ARGS ${tagopt} -o .ptxdist/patches/ ${range} | sed -e 's,^.ptxdist/patches/,,' > .ptxdist/series.auto +${GIT} format-patch -N $GIT_EXTRA_ARGS ${tagopt} -o .ptxdist/patches/ ${range} | while read patch; do + if "$numbered_patches"; then + patchname="${patch#.ptxdist/patches/}" + else + patchname="${patch#.ptxdist/patches/[0-9][0-9][0-9][0-9]-}" + mv -n "$patch" ".ptxdist/patches/$patchname" + fi + echo "$patchname" +done > .ptxdist/series.auto + cat .ptxdist/series.auto >> .ptxdist/series cat .ptxdist/series.1 >> .ptxdist/series cat .ptxdist/series | _md5sum >> .ptxdist/series -- 2.28.0 ___ ptxdist mailing list ptxdist@pengutronix.de To unsubscribe, send a mail with subject "unsubscribe" to ptxdist-requ...@pengutronix.de
[ptxdist] [PATCH] rsyslog: version bump 8.26.0 -> 8.2008.0
This fixes a compile error with gcc-10 en passant. Signed-off-by: Uwe Kleine-König --- rules/rsyslog.in | 1 + rules/rsyslog.make | 48 -- 2 files changed, 39 insertions(+), 10 deletions(-) diff --git a/rules/rsyslog.in b/rules/rsyslog.in index 74f6395679f1..157c60e3bae4 100644 --- a/rules/rsyslog.in +++ b/rules/rsyslog.in @@ -7,6 +7,7 @@ menuconfig RSYSLOG select LIBC_PTHREAD select GCCLIBS_GCC_S select ZLIB + select LIBCURL select LIBESTR select LIBFASTJSON select LIBNET if RSYSLOG_OMUDPSPOOF diff --git a/rules/rsyslog.make b/rules/rsyslog.make index 28c7a8901dbb..0c7f15c5be5b 100644 --- a/rules/rsyslog.make +++ b/rules/rsyslog.make @@ -14,8 +14,8 @@ PACKAGES-$(PTXCONF_RSYSLOG) += rsyslog # # Paths and names # -RSYSLOG_VERSION:= 8.26.0 -RSYSLOG_MD5:= abe20d1621d1e73326c08b964a556ed7 +RSYSLOG_VERSION:= 8.2008.0 +RSYSLOG_MD5:= 9b6ae1a517231764ad4fbd68181cc23e RSYSLOG:= rsyslog-$(RSYSLOG_VERSION) RSYSLOG_SUFFIX := tar.gz RSYSLOG_URL:= http://www.rsyslog.com/files/download/rsyslog/$(RSYSLOG).$(RSYSLOG_SUFFIX) @@ -37,31 +37,41 @@ RSYSLOG_LICENSE_FILES := \ RSYSLOG_CONF_TOOL := autoconf RSYSLOG_CONF_OPT := \ $(CROSS_AUTOCONF_USR) \ + --enable-compile-warnings=yes \ + --disable-Werror \ + --enable-atomic-operations \ $(GLOBAL_LARGE_FILE_OPTION) \ --$(call ptx/endis, PTXCONF_RSYSLOG_REGEXP)-regexp \ + --disable-fmhash \ --disable-gssapi-krb5 \ + --disable-root-tests \ --$(call ptx/endis, PTXCONF_RSYSLOG_IMKLOG)-klog \ --$(call ptx/endis, PTXCONF_RSYSLOG_IMKMSG)-kmsg \ --$(call ptx/endis, PTXCONF_RSYSLOG_SYSTEMD)-imjournal \ + --$(call ptx/endis, PTXCONF_RSYSLOG_SYSTEMD)-libsystemd \ --$(call ptx/endis, PTXCONF_RSYSLOG_INET)-inet \ --disable-jemalloc \ --disable-unlimited-select \ --disable-debug \ --disable-debug-symbols \ - --disable-rtinst \ --disable-debugless \ --disable-valgrind \ - --disable-memcheck \ --disable-diagtools \ --disable-usertools \ --disable-mysql \ --disable-pgsql \ --disable-libdbi \ --disable-snmp \ + --disable-snmp-tests \ --disable-uuid \ --disable-elasticsearch \ + --disable-clickhouse \ + --disable-omhttp \ --disable-elasticsearch-tests \ + --disable-clickhouse-tests \ + --disable-openssl \ --disable-gnutls \ + --disable-gnutls-tests \ --disable-libgcrypt \ --enable-rsyslogrt \ --enable-rsyslogd \ @@ -69,6 +79,7 @@ RSYSLOG_CONF_OPT := \ --disable-mysql-tests \ --disable-pgsql-tests \ --disable-mail \ + --enable-fmhttp \ --$(call ptx/endis, PTXCONF_RSYSLOG_IMDIAG)-imdiag \ --disable-mmnormalize \ --$(call ptx/endis, PTXCONF_RSYSLOG_MMJSONPARSE)-mmjsonparse \ @@ -80,27 +91,41 @@ RSYSLOG_CONF_OPT:= \ --disable-mmcount \ --disable-mmsequence \ --disable-mmdblookup \ + --disable-mmdarwin \ --disable-mmfields \ --disable-mmpstrucdata \ --disable-mmrfc5424addhmac \ + --disable-omfile-hardened \ --disable-relp \ - --disable-guardtime \ - --disable-gt-ksi \ + --disable-ksi-ls12 \ --disable-liblogging-stdlog \ --disable-rfc3195 \ --disable-testbench \ + --disable-libfaketime \ + --disable-default-tests \ + --disable-helgrind \ + --disable-imbatchreport \ + --disable-pmdb2diag \ --$(call ptx/endis, PTXCONF_RSYSLOG_IMFILE)-imfile \ + --disable-imfile-tests \ + --disable-imdocker \ + --disable-imdocker-tests \ + --disable-imtuxedoulog \ + --disable-improg \ --disable-imsolaris \ --$(call ptx/endis, PTXCONF_RSYSLOG_IMPTCP)-imptcp \ --$(call ptx/endis, PTXCONF_RSYSLOG_IMPSTATS)-impstats \ + --disable-impcap \ --$(call ptx/endis, PTXCONF_RSYSLOG_OMPROG)-omprog \ --$(call ptx/endis, PTXCONF_RSYSLOG_OMUDPSPOOF)-omudpspoof \ --$(call ptx/endis, PTXCONF_RSYSLOG_OMSTDOUT)-omstdout \ --$(call ptx/endis, PTXCONF_RSYSLOG_SYSTEMD)-omjournal \ + --disable-journal-tests \ --$(call ptx/endis, PTXCONF_RSYSLOG_PMLASTMSG)-pmlastmsg \ --disable-pmcisconames \ --$(call ptx/endis, PTXCONF_RSYSLOG_PMCISCOIOS)-pmciscoios \ --disable-pmnull \ + --disable-pmnormalize \ --disable-pmaixforwardedfrom \ --disable-pmsnare \ --disable-pmpanngfw \ @@ -109,19 +134,23 @@ RSYSLOG_CONF_OPT := \ --disable-mmsnmptrapd \ --disable-omhdfs \ --disable-omkafka \ + --disable-imkafka \ --disable-kafka-tests \ + --disable-kafka-static \ --disable-ommongodb
[ptxdist] [PATCH] git-ptx-patches: support flag -n to get unnumbered patches
When passing -n to git-ptx-patches the filenames in the patch stack don't get a number as prefix. The obvious downside is that the order of the patches isn't obvious without the series file and you cannot create a patch stack if two patches share have the same shortlog (which ideally shouldn't happen ...) The obvious advantage is that when a patch is added or removed in the middle of the series there are no renames for the later patches necessary and the changes to series are smaller accordingly. So if the patchstack of openssl used unnumbered patches the commit c45d66cdae4b ("openssl: remove engines-path patch") would only have to touch two lines in series (i.e. the removal of the patch and the changed checksum) and contain no renames. Signed-off-by: Uwe Kleine-König --- scripts/git-ptx-patches | 17 +++-- 1 file changed, 15 insertions(+), 2 deletions(-) diff --git a/scripts/git-ptx-patches b/scripts/git-ptx-patches index 721aa78ba31c..b2bed9664fca 100755 --- a/scripts/git-ptx-patches +++ b/scripts/git-ptx-patches @@ -19,6 +19,7 @@ fi remove_old=no tag=base +numbered_patches=true if grep -q "$PTX_PATCHES_HEADER" .ptxdist/series; then echo "Found series file generated by git-ptx-patches." @@ -37,7 +38,7 @@ if [ "x$1" = "x--force-remove" ]; then shift fi -while getopts "ft:n:" opt; do +while getopts "ft:n" opt; do case "${opt}" in f) remove_old="force" @@ -46,6 +47,9 @@ while getopts "ft:n:" opt; do tag="${OPTARG}" range="${tag}" ;; + n) + numbered_patches=false + ;; esac done shift $((${OPTIND} - 1)) @@ -129,7 +133,16 @@ fi GIT_EXTRA_ARGS="$GIT_EXTRA_ARGS --summary --stat=80" cat .ptxdist/series.0 > .ptxdist/series -git format-patch -N $GIT_EXTRA_ARGS ${tagopt} -o .ptxdist/patches/ ${range} | sed -e 's,^.ptxdist/patches/,,' > .ptxdist/series.auto +git format-patch -N $GIT_EXTRA_ARGS ${tagopt} -o .ptxdist/patches/ ${range} | while read patch; do + if "$numbered_patches"; then + patchname="${patch#.ptxdist/patches/}" + else + patchname="${patch#.ptxdist/patches/[0-9][0-9][0-9][0-9]-}" + mv -n "$patch" ".ptxdist/patches/$patchname" + fi + echo "$patchname" +done > .ptxdist/series.auto + cat .ptxdist/series.auto >> .ptxdist/series cat .ptxdist/series.1 >> .ptxdist/series cat .ptxdist/series | _md5sum >> .ptxdist/series -- 2.27.0 ___ ptxdist mailing list ptxdist@pengutronix.de To unsubscribe, send a mail with subject "unsubscribe" to ptxdist-requ...@pengutronix.de
Re: [ptxdist] [PATCH] rc-once: make sure data makes it to the medium
On Mon, Aug 10, 2020 at 12:17:13PM +0200, Robert Schwebel wrote: > We have seen that, if the first boot runs under a testsuite that powers > the board off after rc-once is finished, data hasn't found its way to > the medium. This has been observed to result in an empty ssh key. > > Adding a sync helps. > > Signed-off-by: Robert Schwebel > --- > projectroot/usr/lib/systemd/systemd-rc-once | 2 ++ > 1 file changed, 2 insertions(+) > > diff --git a/projectroot/usr/lib/systemd/systemd-rc-once > b/projectroot/usr/lib/systemd/systemd-rc-once > index e14936e40..db3d84e3c 100644 > --- a/projectroot/usr/lib/systemd/systemd-rc-once > +++ b/projectroot/usr/lib/systemd/systemd-rc-once > @@ -21,6 +21,8 @@ start() { > } > > umount() { > + # make sure rc-once generated data makes it to the medium > + sync > if ! mount_root_restore; then > # remounting rw/ro during the second boot will flush anything > # left in the filesystem journal I would expect that the mount -o remount,ro that is done (sometimes) as part of mount_root_restore should handle this just fine. How is your use case? Are you muddying the waters of a bug somewhere else here? Best regards Uwe -- Pengutronix e.K. | Uwe Kleine-König| Industrial Linux Solutions | https://www.pengutronix.de/ | signature.asc Description: PGP signature ___ ptxdist mailing list ptxdist@pengutronix.de To unsubscribe, send a mail with subject "unsubscribe" to ptxdist-requ...@pengutronix.de
[ptxdist] [PATCH v2 1/2] openssh/rc-once: deduplicate some data
The create_keys() function passed the key type three times. Now it's only passed once. Signed-off-by: Uwe Kleine-König --- projectroot/etc/rc.once.d/openssh | 25 +++-- 1 file changed, 11 insertions(+), 14 deletions(-) diff --git a/projectroot/etc/rc.once.d/openssh b/projectroot/etc/rc.once.d/openssh index 4a3c594cc3ae..bfec7064181a 100644 --- a/projectroot/etc/rc.once.d/openssh +++ b/projectroot/etc/rc.once.d/openssh @@ -18,32 +18,29 @@ host_keys_required() { } create_key() { - msg="$1" + keytype="$1" + prettykeytype="$(echo $_type | tr a-z A-Z)" shift hostkeys="$1" shift - file="$1" - shift + + file="/etc/ssh/ssh_host_${keytype}_key" if echo "$hostkeys" | grep -x "$file" >/dev/null; then - echo "$msg; this may take some time ..." + echo "Create $prettykeytype key; this may take some time ..." rm -f $file && - ssh-keygen -q -f "$file" -N '' "$@" || return - echo "$msg; done." + ssh-keygen -q -f "$file" -N '' -t "$keytype" "$@" || return + echo "Created $prettykeytype key." fi } create_keys() { hostkeys="$(host_keys_required)" - create_key "Creating DSA key" \ - "$hostkeys" /etc/ssh/ssh_host_dsa_key -t dsa && - create_key "Creating ECDSA key" \ - "$hostkeys" /etc/ssh/ssh_host_ecdsa_key -t ecdsa && - create_key "Creating ED25519 key" \ - "$hostkeys" /etc/ssh/ssh_host_ed25519_key -t ed25519 && - create_key "Creating RSA key" \ - "$hostkeys" /etc/ssh/ssh_host_rsa_key -t rsa -b 4096 + create_key "dsa" "$hostkeys" && + create_key "ecdsa" "$hostkeys" && + create_key "ed25519" "$hostkeys" && + create_key "rsa" "$hostkeys" -b 4096 } if ! create_keys; then -- 2.28.0 ___ ptxdist mailing list ptxdist@pengutronix.de To unsubscribe, send a mail with subject "unsubscribe" to ptxdist-requ...@pengutronix.de
[ptxdist] [PATCH v2 2/2] openssh/rc-once: use fixed string grep to match filename
This is a (very) minor optimisation. There is no semantical change as the fixed list of possible filenames doesn't contain anything that has a different meaning when interpreted as a regex, still I consider it better style to interpret the filename as a fixed string to match. Both busybox and the "big" grep support -F unconditionally so there is no problem in using -F. Signed-off-by: Uwe Kleine-König --- projectroot/etc/rc.once.d/openssh | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/projectroot/etc/rc.once.d/openssh b/projectroot/etc/rc.once.d/openssh index bfec7064181a..66cfa06df12a 100644 --- a/projectroot/etc/rc.once.d/openssh +++ b/projectroot/etc/rc.once.d/openssh @@ -26,7 +26,7 @@ create_key() { file="/etc/ssh/ssh_host_${keytype}_key" - if echo "$hostkeys" | grep -x "$file" >/dev/null; then + if echo "$hostkeys" | grep -x -F "$file" >/dev/null; then echo "Create $prettykeytype key; this may take some time ..." rm -f $file && ssh-keygen -q -f "$file" -N '' -t "$keytype" "$@" || return -- 2.28.0 ___ ptxdist mailing list ptxdist@pengutronix.de To unsubscribe, send a mail with subject "unsubscribe" to ptxdist-requ...@pengutronix.de
[ptxdist] [PATCH] openssh/rc-once: deduplicate some function parameters
The create_keys() function passed the key type three times. Now it's only passed once. While at it also use the shell builtin case instead of echo | grep to find the needed keys. The only visible change is that the end message message changes from Creating RSA key; done. to Created RSA key. Signed-off-by: Uwe Kleine-König --- Hello, I noticed this "triplication" while updating a BSP and merging in some BSP specific features. I guess it is subjective if you consider this an improvement, so judge freely if you want it in ptxdist or not. Best regards Uwe projectroot/etc/rc.once.d/openssh | 35 +-- 1 file changed, 19 insertions(+), 16 deletions(-) diff --git a/projectroot/etc/rc.once.d/openssh b/projectroot/etc/rc.once.d/openssh index 4a3c594cc3ae..813e7c620e49 100644 --- a/projectroot/etc/rc.once.d/openssh +++ b/projectroot/etc/rc.once.d/openssh @@ -18,32 +18,35 @@ host_keys_required() { } create_key() { - msg="$1" + _type="$1" + prettytype="$(echo $_type | tr a-z A-Z)" shift hostkeys="$1" shift - file="$1" - shift - if echo "$hostkeys" | grep -x "$file" >/dev/null; then - echo "$msg; this may take some time ..." + file="/etc/ssh/ssh_host_${_type}_key" + + case " +$hostkeys +" in + *" +$file +"*) + echo "Create $prettytype key; this may take some time ..." rm -f $file && - ssh-keygen -q -f "$file" -N '' "$@" || return - echo "$msg; done." - fi + ssh-keygen -q -f "$file" -N '' -t "$_type" "$@" || return + echo "Created $prettytype key." + ;; + esac } create_keys() { hostkeys="$(host_keys_required)" - create_key "Creating DSA key" \ - "$hostkeys" /etc/ssh/ssh_host_dsa_key -t dsa && - create_key "Creating ECDSA key" \ - "$hostkeys" /etc/ssh/ssh_host_ecdsa_key -t ecdsa && - create_key "Creating ED25519 key" \ - "$hostkeys" /etc/ssh/ssh_host_ed25519_key -t ed25519 && - create_key "Creating RSA key" \ - "$hostkeys" /etc/ssh/ssh_host_rsa_key -t rsa -b 4096 + create_key "dsa" "$hostkeys" && + create_key "ecdsa" "$hostkeys" && + create_key "ed25519" "$hostkeys" && + create_key "rsa" "$hostkeys" -b 4096 } if ! create_keys; then -- 2.28.0 ___ ptxdist mailing list ptxdist@pengutronix.de To unsubscribe, send a mail with subject "unsubscribe" to ptxdist-requ...@pengutronix.de
[ptxdist] [PATCH] net-snmp: unbreak AES support
Commit c300c3973143 ("net-snmp: disable broken crypto support") introduced this and noted that net-snmp 5.8 would be needed to fix this. We have 5.8 since fda39d3fd0d5 ("net-snmp: version bump v5.7.3 -> v5.8"). Signed-off-by: Uwe Kleine-König --- rules/net-snmp.in | 2 -- 1 file changed, 2 deletions(-) diff --git a/rules/net-snmp.in b/rules/net-snmp.in index ec4bf71bed23..573ed9e73a4a 100644 --- a/rules/net-snmp.in +++ b/rules/net-snmp.in @@ -227,8 +227,6 @@ config NET_SNMP_MD5 config NET_SNMP_SHA_AES bool - # 5.7.3 fails to build with OpenSSL 1.1.x - depends on BROKEN prompt "SHA authentication and AES encryption" config NET_SNMP_DOM_SOCK_ONLY -- 2.27.0 ___ ptxdist mailing list ptxdist@pengutronix.de To unsubscribe, send a mail with subject "unsubscribe" to ptxdist-requ...@pengutronix.de
[ptxdist] [PATCH] microcom: version bump 2017.03.0 -> 2019.01.0
The project moved to github, so adapt the download URL accordingly. Signed-off-by: Uwe Kleine-König --- rules/microcom.make | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/rules/microcom.make b/rules/microcom.make index 0aeadaadff1a..d8ff76bb8e6a 100644 --- a/rules/microcom.make +++ b/rules/microcom.make @@ -14,11 +14,11 @@ PACKAGES-$(PTXCONF_MICROCOM) += microcom # # Paths and names # -MICROCOM_VERSION := 2017.03.0 -MICROCOM_MD5 := 801d99dcb4b2ef7e8b77b51bba193066 +MICROCOM_VERSION := 2019.01.0 +MICROCOM_MD5 := 8e8c58b4d95d4466097dec3c7ed03ccd MICROCOM := microcom-$(MICROCOM_VERSION) MICROCOM_SUFFIX:= tar.xz -MICROCOM_URL := http://www.pengutronix.de/software/microcom/download/$(MICROCOM).$(MICROCOM_SUFFIX) +MICROCOM_URL := https://github.com/pengutronix/microcom/releases/download/v$(MICROCOM_VERSION)/microcom-$(MICROCOM_VERSION).$(MICROCOM_SUFFIX) MICROCOM_SOURCE:= $(SRCDIR)/$(MICROCOM).$(MICROCOM_SUFFIX) MICROCOM_DIR := $(BUILDDIR)/$(MICROCOM) MICROCOM_LICENSE := GPL-2.0-only -- 2.27.0 ___ ptxdist mailing list ptxdist@pengutronix.de To unsubscribe, send a mail with subject "unsubscribe" to ptxdist-requ...@pengutronix.de
[ptxdist] [PATCH 2/2] tpm2-tss: make installation of libtss2-tcti-{device, mssim}.so.0 conditional
Not all users of tpm2-tss make use of the libraries, so introduce a config symbol to make it possible to not install them. Signed-off-by: Uwe Kleine-König --- rules/tpm2-tss.in | 10 ++ rules/tpm2-tss.make | 4 2 files changed, 14 insertions(+) diff --git a/rules/tpm2-tss.in b/rules/tpm2-tss.in index dbb31007276b..bf1506da67c4 100644 --- a/rules/tpm2-tss.in +++ b/rules/tpm2-tss.in @@ -20,3 +20,13 @@ config TPM2_TSS This package contains the TSS and TCTI libraries that client applications will link against when they require accessing the TPM. + +if TPM2_TSS + +config TPM2_TSS_LIBTSS2_TCTI_DEVICE + bool "install libtss2-tcti-device.so.0" + +config TPM2_TSS_LIBTSS2_TCTI_MSSIM + bool "install libtss2-tcti-mssim.so.0" + +endif diff --git a/rules/tpm2-tss.make b/rules/tpm2-tss.make index 768c64426cbb..ed49a7d56c3d 100644 --- a/rules/tpm2-tss.make +++ b/rules/tpm2-tss.make @@ -83,8 +83,12 @@ $(STATEDIR)/tpm2-tss.targetinstall: @$(call install_lib, tpm2-tss, 0, 0, 0644, libtss2-esys) @$(call install_lib, tpm2-tss, 0, 0, 0644, libtss2-mu) @$(call install_lib, tpm2-tss, 0, 0, 0644, libtss2-sys) +ifdef PTXCONF_TPM2_TSS_LIBTSS2_TCTI_DEVICE @$(call install_lib, tpm2-tss, 0, 0, 0644, libtss2-tcti-device) +endif +ifdef PTXCONF_TPM2_TSS_LIBTSS2_TCTI_MSSIM @$(call install_lib, tpm2-tss, 0, 0, 0644, libtss2-tcti-mssim) +endif @$(call install_lib, tpm2-tss, 0, 0, 0644, libtss2-tctildr) @$(call install_lib, tpm2-tss, 0, 0, 0644, libtss2-rc) -- 2.27.0 ___ ptxdist mailing list ptxdist@pengutronix.de To unsubscribe, send a mail with subject "unsubscribe" to ptxdist-requ...@pengutronix.de
[ptxdist] [PATCH 1/2] tpm2-abrmd: make installation of libtss2-tcti-tabrmd.so.0 conditional
Not all users of tpm2-abrmd make use of the library, so introduce a config symbol to make it possible to not install it. Signed-off-by: Uwe Kleine-König --- rules/tpm2-abrmd.in | 7 +++ rules/tpm2-abrmd.make | 2 ++ 2 files changed, 9 insertions(+) diff --git a/rules/tpm2-abrmd.in b/rules/tpm2-abrmd.in index 1ab0a6a7fe0e..61bba8a02030 100644 --- a/rules/tpm2-abrmd.in +++ b/rules/tpm2-abrmd.in @@ -16,3 +16,10 @@ config TPM2_ABRMD (TAB) & Resource Manager (RM) spec from the TCG. The daemon (tpm2-abrmd) is implemented using Glib and the GObject system. + +if TPM2_ABRMD + +config TPM2_ABRMD_LIBTSS2_TCTI_TABRMD + bool "install libtss2-tcti-tabrmd.so.0" + +endif diff --git a/rules/tpm2-abrmd.make b/rules/tpm2-abrmd.make index bbf3344126f7..edc1fcfc0e6e 100644 --- a/rules/tpm2-abrmd.make +++ b/rules/tpm2-abrmd.make @@ -63,7 +63,9 @@ $(STATEDIR)/tpm2-abrmd.targetinstall: @$(call install_fixup, tpm2-abrmd,AUTHOR,"Marc Kleine-Budde ") @$(call install_fixup, tpm2-abrmd,DESCRIPTION,missing) +ifdef PTXCONF_TPM2_ABRMD_LIBTSS2_TCTI_TABRMD @$(call install_lib, tpm2-abrmd, 0, 0, 0644, libtss2-tcti-tabrmd) +endif @$(call install_copy, tpm2-abrmd, 0, 0, 0755, -, /usr/sbin/tpm2-abrmd) @$(call install_alternative, tpm2-abrmd, 0, 0, 0644, \ -- 2.27.0 ___ ptxdist mailing list ptxdist@pengutronix.de To unsubscribe, send a mail with subject "unsubscribe" to ptxdist-requ...@pengutronix.de
[ptxdist] [PATCH] hd.img: use efi type also for GPT partitions
Similar to using 0xef for efi partitions with MBR partitioning, there is a dedicated type for these when GPT partitioning is to be used. Signed-off-by: Uwe Kleine-König --- config/images/vfat_partitions.config | 1 + rules/image-hdimg.make | 1 + 2 files changed, 2 insertions(+) diff --git a/config/images/vfat_partitions.config b/config/images/vfat_partitions.config index 799c19d5763d..0aac561de8e0 100644 --- a/config/images/vfat_partitions.config +++ b/config/images/vfat_partitions.config @@ -1,5 +1,6 @@ partition boot-vfat { image = "boot.vfat" partition-type = @VFAT_PARTITION_TYPE@ + partition-type-uuid = @VFAT_PARTITION_TYPE_UUID@ bootable = true } diff --git a/rules/image-hdimg.make b/rules/image-hdimg.make index e8fd01f1a890..b0f6af77f154 100644 --- a/rules/image-hdimg.make +++ b/rules/image-hdimg.make @@ -42,6 +42,7 @@ endif ifdef PTXCONF_IMAGE_HDIMG_VFAT IMAGE_HDIMG_BOOTLOADER_ENV := \ VFAT_PARTITION_TYPE=$(call ptx/ifdef, PTXCONF_IMAGE_BOOT_VFAT_EFI_BAREBOX,0xef,0x0c) \ + VFAT_PARTITION_TYPE_UUID=$(call ptx/ifdef, PTXCONF_IMAGE_BOOT_VFAT_EFI_BAREBOX,U,L) \ BOOTLOADER_IMAGES='' \ BOOTLOADER_PARTITIONS='include("vfat_partitions.config")' endif -- 2.26.2 ___ ptxdist mailing list ptxdist@pengutronix.de
[ptxdist] [PATCH] util-linux-ng: make wdctl tool selectable
Signed-off-by: Uwe Kleine-König --- rules/util-linux-ng.in | 6 ++ rules/util-linux-ng.make | 5 - 2 files changed, 10 insertions(+), 1 deletion(-) diff --git a/rules/util-linux-ng.in b/rules/util-linux-ng.in index a9735e13c910..34fcd5a47271 100644 --- a/rules/util-linux-ng.in +++ b/rules/util-linux-ng.in @@ -342,6 +342,12 @@ config UTIL_LINUX_NG_HWCLOCK comment "BusyBox' hwclock is selected!" depends on BUSYBOX_HWCLOCK +config UTIL_LINUX_NG_WDCTL + bool + prompt "wdctl" + help + wdctl - Utility to show hardware watchdog status. + config UTIL_LINUX_NG_IONICE bool prompt "ionice" diff --git a/rules/util-linux-ng.make b/rules/util-linux-ng.make index 88abca85f15e..ff583267f8ed 100644 --- a/rules/util-linux-ng.make +++ b/rules/util-linux-ng.make @@ -92,7 +92,7 @@ UTIL_LINUX_NG_CONF_OPT:= \ --disable-fdformat \ --$(call ptx/endis, PTXCONF_UTIL_LINUX_NG_HWCLOCK)-hwclock \ --disable-lslogins \ - --disable-wdctl \ + --$(call ptx/endis, PTXCONF_UTIL_LINUX_NG_WDCTL)-wdctl \ --disable-cal \ --disable-logger \ --disable-switch_root \ @@ -279,6 +279,9 @@ endif ifdef PTXCONF_UTIL_LINUX_NG_HWCLOCK @$(call install_copy, util-linux-ng, 0, 0, 0755, -, /usr/sbin/hwclock) endif +ifdef PTXCONF_UTIL_LINUX_NG_WDCTL + @$(call install_copy, util-linux-ng, 0, 0, 0755, -, /usr/bin/wdctl) +endif ifdef PTXCONF_UTIL_LINUX_NG_IONICE @$(call install_copy, util-linux-ng, 0, 0, 0755, -, /usr/bin/ionice) endif -- 2.26.2 ___ ptxdist mailing list ptxdist@pengutronix.de
[ptxdist] [PATCH] canfestival: Two new patches that make canfestival headers more robust
Signed-off-by: Uwe Kleine-König --- ...4-let-canfestival.h-include-config.h.patch | 39 ++ -instead-of-include-.-for-own-files.patch | 118 ++ patches/canfestival-3+hg20180126.794/series | 4 +- 3 files changed, 160 insertions(+), 1 deletion(-) create mode 100644 patches/canfestival-3+hg20180126.794/0004-let-canfestival.h-include-config.h.patch create mode 100644 patches/canfestival-3+hg20180126.794/0005-Use-include-.-instead-of-include-.-for-own-files.patch diff --git a/patches/canfestival-3+hg20180126.794/0004-let-canfestival.h-include-config.h.patch b/patches/canfestival-3+hg20180126.794/0004-let-canfestival.h-include-config.h.patch new file mode 100644 index ..c6b005703e03 --- /dev/null +++ b/patches/canfestival-3+hg20180126.794/0004-let-canfestival.h-include-config.h.patch @@ -0,0 +1,39 @@ +From: =?UTF-8?q?Uwe=20Kleine-K=C3=B6nig?= +Date: Mon, 29 Jan 2018 14:07:18 +0100 +Subject: [PATCH] let canfestival.h include config.h + +This is necessary to make the header self-contained. Otherwise it is necessary to do + + #include + #include + +in the API user which might pull in a local config.h. +--- + include/sdo.h | 1 + + include/unix/canfestival.h | 1 + + 2 files changed, 2 insertions(+) + +diff --git a/include/sdo.h b/include/sdo.h +index 28b8c5c414eb..7863bb5c05a0 100644 +--- a/include/sdo.h b/include/sdo.h +@@ -36,6 +36,7 @@ Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + + struct struct_s_transfer; + ++#include "config.h" + #include "timer.h" + + /* Block mode : Data consumer receive step +diff --git a/include/unix/canfestival.h b/include/unix/canfestival.h +index f20e593dcf89..ee2a0bc0c4ae 100755 +--- a/include/unix/canfestival.h b/include/unix/canfestival.h +@@ -5,6 +5,7 @@ + extern "C" { + #endif + ++#include "config.h" + #include "timerscfg.h" + #include "can_driver.h" + #include "data.h" diff --git a/patches/canfestival-3+hg20180126.794/0005-Use-include-.-instead-of-include-.-for-own-files.patch b/patches/canfestival-3+hg20180126.794/0005-Use-include-.-instead-of-include-.-for-own-files.patch new file mode 100644 index ..8feaa48a8416 --- /dev/null +++ b/patches/canfestival-3+hg20180126.794/0005-Use-include-.-instead-of-include-.-for-own-files.patch @@ -0,0 +1,118 @@ +From: =?UTF-8?q?Uwe=20Kleine-K=C3=B6nig?= +Date: Wed, 29 Apr 2020 22:54:55 +0200 +Subject: [PATCH] Use #include "..." instead of #include <...> for own files + +This fixes inclusion if the consumer used + + #include + +(and not -I/usr/include/canfestival) as with + + #include <...> + +the local files are not found in this case. +--- + include/emcy.h | 2 +- + include/lifegrd.h | 2 +- + include/nmtSlave.h | 2 +- + include/objacces.h | 2 +- + include/pdo.h | 4 ++-- + include/states.h | 2 +- + include/timer.h| 4 ++-- + 7 files changed, 9 insertions(+), 9 deletions(-) + +diff --git a/include/emcy.h b/include/emcy.h +index d3a19eb6d335..ad7ce933eebd 100644 +--- a/include/emcy.h b/include/emcy.h +@@ -39,7 +39,7 @@ Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + #define __emcy_h__ + + +-#include ++#include "applicfg.h" + + /* The error states + * - */ +diff --git a/include/lifegrd.h b/include/lifegrd.h +index a0600a717038..222a3dd18f73 100644 +--- a/include/lifegrd.h b/include/lifegrd.h +@@ -53,7 +53,7 @@ Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + #define __lifegrd_h__ + + +-#include ++#include "applicfg.h" + + typedef void (*heartbeatError_t)(CO_Data*, UNS8); + void _heartbeatError(CO_Data* d, UNS8 heartbeatID); +diff --git a/include/nmtSlave.h b/include/nmtSlave.h +index 0a57ae878364..ffb720194e15 100755 +--- a/include/nmtSlave.h b/include/nmtSlave.h +@@ -28,7 +28,7 @@ Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + #ifndef __nmtSlave_h__ + #define __nmtSlave_h__ + +-#include ++#include "applicfg.h" + #include "data.h" + + /** +diff --git a/include/objacces.h b/include/objacces.h +index 26a96542a4ca..05dd46d313d7 100755 +--- a/include/objacces.h b/include/objacces.h +@@ -39,7 +39,7 @@ Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + #ifndef __objacces_h__ + #define __objacces_h__ + +-#include ++#include "applicfg.h" + + + #ifdef __cplusplus +diff --git a/include/pdo.h b/include/pdo.h +index d276d51734fc..713167f50071 100755 +--- a/include/pdo.h b/include/pdo.h +@@ -29,8 +29,8 @@ Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + #ifndef __pdo_h__ + #define __pdo_h__ + +-#include +-#include ++#include "applicfg.h" ++#include "def.h" + + #include "can.h" + +diff --git a/include/states.h b/include/s
Re: [ptxdist] [Kernel.org Helpdesk #86171] [linuxfoundation.org #86171] Re: Archive ptxdist@pengutronix.de on lore.kernel.org?
Hello Konstantin, On Tue, Apr 14, 2020 at 05:31:22PM -0400, Konstantin Ryabitsev via RT wrote: > On Tue, Apr 14, 2020 at 05:52:53AM -0400, Uwe Kleine-König via RT wrote: > > > > > > I didn't hear back from you in reply to my request from three weeks ago. > > > Is this still on your radar and I just need some more patience? > > > > I still didn't hear anything back from you. The ticket on > > rt.linuxfoundation.org is still marked as "new". > > > > Following https://korg.wiki.kernel.org/userdoc/lore I'm currently in the > > step "Wait till you receive a response", but since I'm there since more > > than eight weeks I wonder if there is a problem on your side? > > Hello, and sorry for radio silence. I wasn't deliberately trying to > stall -- I was just trying to figure out the best way to approach this. That's fine, I didn't expect ill intend. On the receiving end of such a silence it is however really hard to know what is going on. Did the other side receive the mail at all? Did I do something wrong? Did I get on their nerve? Are they just drowning in other work? Did they reply to the moderated list only and the reply was dropped by mistake? When is the right time to ping this request, or does this just increase annoyance? So a quick and small heads up à la "We got your request, it doesn't completely fit our intend, give us some time to think about how to handle it" would have been great. (I'm aware about the cost on your end for such a reply, I'm on the silence-sending side at times, too. So I just want to make you (more) aware of my situation here, without ranting.) > While it's true that there are some things on lore.kernel.org that > aren't *directly* related to the kernel, we nevertheless try to stick to > lists that are either about low-level kernel subsystems, or about > various tools used for kernel development, or at least have predominant > topics that are close to the above two categories. The trouble with > ptxdist is that it doesn't really fit inside those boundaries, so I am > not sure whether it makes sense for it to live on lore.kernel.org. I think that is OK. I'll have to talk to the people on our side about our options and about how to proceed. > Above all, I'd like to avoid turning lore.kernel.org into a central > resource that becomes a hard dependency for projects the same way gmane > was -- so much so, that when it disappeared, it caused serious ripple > across the open-source fabric. Indeed I also already thought that having links to lore.kernel.org in ptxdist commit messages is inferior to an URI that is related to the project. (Having said that, just putting something like Message-id: msg.id@here and have the tooling to convert that to something clickable depending on the project would be great.) > We are encouraging others to set up their own instances of public-inbox, > and would prefer to mirror the resulting git repositories instead (and > offer indexing/searching capabilities) instead of doing archival on our > end. This is the approach we're taking with Yocto, and I would encourage > you to investigate the same option, as this would avoid making > lore.kernel.org a hard dependency for your project. This sounds like a great option. Thanks for your feedback, Uwe -- Pengutronix e.K. | Uwe Kleine-König| Industrial Linux Solutions | https://www.pengutronix.de/ | ___ ptxdist mailing list ptxdist@pengutronix.de
Re: [ptxdist] [Kernel.org Helpdesk #86171] [linuxfoundation.org #86171] Re: Archive ptxdist@pengutronix.de on lore.kernel.org?
Hello Konstantin, On Tue, Apr 14, 2020 at 05:31:22PM -0400, Konstantin Ryabitsev via RT wrote: > On Tue, Apr 14, 2020 at 05:52:53AM -0400, Uwe Kleine-König via RT wrote: > > > > > > I didn't hear back from you in reply to my request from three weeks ago. > > > Is this still on your radar and I just need some more patience? > > > > I still didn't hear anything back from you. The ticket on > > rt.linuxfoundation.org is still marked as "new". > > > > Following https://korg.wiki.kernel.org/userdoc/lore I'm currently in the > > step "Wait till you receive a response", but since I'm there since more > > than eight weeks I wonder if there is a problem on your side? > > Hello, and sorry for radio silence. I wasn't deliberately trying to > stall -- I was just trying to figure out the best way to approach this. That's fine, I didn't expect ill intend. On the receiving end of such a silence it is however really hard to know what is going on. Did the other side receive the mail at all? Did I do something wrong? Did I get on their nerve? Are they just drowning in other work? Did they reply to the moderated list only and the reply was dropped by mistake? When is the right time to ping this request, or does this just increase annoyance? So a quick and small heads up à la "We got your request, it doesn't completely fit our intend, give us some time to think about how to handle it" would have been great. (I'm aware about the cost on your end for such a reply, I'm on the silence-sending side at times, too. So I just want to make you (more) aware of my situation here, without ranting.) > While it's true that there are some things on lore.kernel.org that > aren't *directly* related to the kernel, we nevertheless try to stick to > lists that are either about low-level kernel subsystems, or about > various tools used for kernel development, or at least have predominant > topics that are close to the above two categories. The trouble with > ptxdist is that it doesn't really fit inside those boundaries, so I am > not sure whether it makes sense for it to live on lore.kernel.org. I think that is OK. I'll have to talk to the people on our side about our options and about how to proceed. > Above all, I'd like to avoid turning lore.kernel.org into a central > resource that becomes a hard dependency for projects the same way gmane > was -- so much so, that when it disappeared, it caused serious ripple > across the open-source fabric. Indeed I also already thought that having links to lore.kernel.org in ptxdist commit messages is inferior to an URI that is related to the project. (Having said that, just putting something like Message-id: msg.id@here and have the tooling to convert that to something clickable depending on the project would be great.) > We are encouraging others to set up their own instances of public-inbox, > and would prefer to mirror the resulting git repositories instead (and > offer indexing/searching capabilities) instead of doing archival on our > end. This is the approach we're taking with Yocto, and I would encourage > you to investigate the same option, as this would avoid making > lore.kernel.org a hard dependency for your project. This sounds like a great option. Thanks for your feedback, Uwe -- Pengutronix e.K. | Uwe Kleine-König| Industrial Linux Solutions | https://www.pengutronix.de/ | ___ ptxdist mailing list ptxdist@pengutronix.de
Re: [ptxdist] [Kernel.org Helpdesk #86171] [linuxfoundation.org #86171] Re: Archive ptxdist@pengutronix.de on lore.kernel.org?
Hello, On Tue, Mar 10, 2020 at 05:40:56AM -0400, Uwe Kleine-König via RT wrote: > On Thu, Feb 20, 2020 at 10:36:48AM +0100, Uwe Kleine-König wrote: > > I and others at Pengutronix would consider it very useful to have a > > public-inbox archive of the ptxdist mailing list. Given that there are > > already a few lists archived on lore.kernel.org that (I think) are not > > strongly related to the Linux kernel I wondered if it was possible to > > add an archive for ptxdist there, too. (ptxdist has a similar purpose as > > buildroot and yocto, it's an environment that allows to build Linux > > images for embedded systems.) > > I didn't hear back from you in reply to my request from three weeks ago. > Is this still on your radar and I just need some more patience? I still didn't hear anything back from you. The ticket on rt.linuxfoundation.org is still marked as "new". Following https://korg.wiki.kernel.org/userdoc/lore I'm currently in the step "Wait till you receive a response", but since I'm there since more than eight weeks I wonder if there is a problem on your side? Best regards Uwe -- Pengutronix e.K. | Uwe Kleine-König| Industrial Linux Solutions | https://www.pengutronix.de/ | ___ ptxdist mailing list ptxdist@pengutronix.de
Re: [ptxdist] [APPLIED] Python handling was changes in the past, so remove a leftover as well
Hallo Michael, On Fri, Apr 03, 2020 at 08:53:52PM +0200, Michael Olbrich wrote: > So maybe like this. > > On Fri, Apr 03, 2020 at 08:52:04PM +0200, Michael Olbrich wrote: > > Thanks, applied as 70e95734668542ac638d5cb8a759ad402d678ff4. > > > > Michael > > > > [sent from post-receive hook] > > > > On Fri, 03 Apr 2020 20:52:04 +0200, Juergen Borleis > > wrote: > > > In commit a1a8045112e06d5b15869c60ce4b70056c020666 > > > "remove hard dependency on Python2" > > > > > > the handling of the external Python component was changed and a > > > substitution > > > removed from configure.ac - but the 'Makefile.in' still uses the PYTHON > > PYTHON is fixed here. > > > > symbol and creates a strange softlink now. Remove that. > > > > > > Signed-off-by: Juergen Borleis > > > > > > Message-Id: <202003300907.01952@pengutronix.de> > > > Signed-off-by: Michael Olbrich > > And the extra tags are there as well. While improving here: The empty line between Jürgen's S-o-b and the Message-Id could be dropped. Best regards Uwe -- Pengutronix e.K. | Uwe Kleine-König| Industrial Linux Solutions | https://www.pengutronix.de/ | ___ ptxdist mailing list ptxdist@pengutronix.de
Re: [ptxdist] Python handling was changes in the past, so remove a leftover as well
On Thu, Apr 02, 2020 at 05:03:29PM +0200, Michael Olbrich wrote: > Thanks, applied as 70e95734668542ac638d5cb8a759ad402d678ff4. Sad that the two typos I indicated were not fixed :-| Best regards Uwe -- Pengutronix e.K. | Uwe Kleine-König| Industrial Linux Solutions | https://www.pengutronix.de/ | ___ ptxdist mailing list ptxdist@pengutronix.de
Re: [ptxdist] [PATCH] ncdu: store options only when package selected
Just a typo alert: $Subject ~= s/package selected/package is selected/ Best regards Uwe -- Pengutronix e.K. | Uwe Kleine-König| Industrial Linux Solutions | https://www.pengutronix.de/ | ___ ptxdist mailing list ptxdist@pengutronix.de
[ptxdist] [PATCH] json-glib: add dependency on host-glib
json-glib requires glib-mkenums in its compile stage. Without this dependency you either need glib-mkenums on your host machine or luck that json-glib is build only after host-glib (assuming host-glib is enabled). Signed-off-by: Uwe Kleine-König --- rules/json-glib.in | 1 + 1 file changed, 1 insertion(+) diff --git a/rules/json-glib.in b/rules/json-glib.in index b87a49a1849b..03e3743697ce 100644 --- a/rules/json-glib.in +++ b/rules/json-glib.in @@ -4,6 +4,7 @@ config JSON_GLIB tristate prompt "json-glib" select GLIB + select HOST_GLIB # for glib-mkenums select GOBJECT_INTROSPECTIONif JSON_GLIB_INTROSPECTION help A library providing (de)serialization support for the JSON format. -- 2.25.1 ___ ptxdist mailing list ptxdist@pengutronix.de
[ptxdist] [Kernel.org Helpdesk #86171] [linuxfoundation.org #86171] Re: Archive ptxdist@pengutronix.de on lore.kernel.org?
Hello, On Thu, Feb 20, 2020 at 10:36:48AM +0100, Uwe Kleine-König wrote: > I and others at Pengutronix would consider it very useful to have a > public-inbox archive of the ptxdist mailing list. Given that there are > already a few lists archived on lore.kernel.org that (I think) are not > strongly related to the Linux kernel I wondered if it was possible to > add an archive for ptxdist there, too. (ptxdist has a similar purpose as > buildroot and yocto, it's an environment that allows to build Linux > images for embedded systems.) I didn't hear back from you in reply to my request from three weeks ago. Is this still on your radar and I just need some more patience? Thanks Uwe > If yes, subscription is handled by mailman via > ptxdist-requ...@pengutronix.de (with "subscribe" in the Subject). -- Pengutronix e.K. | Uwe Kleine-König| Industrial Linux Solutions | https://www.pengutronix.de/ | ___ ptxdist mailing list ptxdist@pengutronix.de
Re: [ptxdist] Archive ptxdist@pengutronix.de on lore.kernel.org?
Hello, On Thu, Feb 20, 2020 at 10:36:48AM +0100, Uwe Kleine-König wrote: > I and others at Pengutronix would consider it very useful to have a > public-inbox archive of the ptxdist mailing list. Given that there are > already a few lists archived on lore.kernel.org that (I think) are not > strongly related to the Linux kernel I wondered if it was possible to > add an archive for ptxdist there, too. (ptxdist has a similar purpose as > buildroot and yocto, it's an environment that allows to build Linux > images for embedded systems.) I didn't hear back from you in reply to my request from three weeks ago. Is this still on your radar and I just need some more patience? Thanks Uwe > If yes, subscription is handled by mailman via > ptxdist-requ...@pengutronix.de (with "subscribe" in the Subject). -- Pengutronix e.K. | Uwe Kleine-König| Industrial Linux Solutions | https://www.pengutronix.de/ | ___ ptxdist mailing list ptxdist@pengutronix.de
Re: [ptxdist] [PATCH] unfs3: fix attribute setting for symlinks
On Tue, Feb 18, 2020 at 03:57:20PM +0100, Uwe Kleine-König wrote: > unfs makes use of utime() which is unsuitable to set last access and > modification times for symlinks. So make use of the O_PATH flag to > open() and use the futimens() function. > > As a side effect this is less racy and improves resolution from seconds > to nanoseconds. > > Note this will break builds of unfs on Windows. The futimens() function > is specified in POSIX.1-2008, so this shouldn't be a problem for other > platforms. > > Signed-off-by: Uwe Kleine-König Here is something fishy. After deploying this patch in a local BSP I see some errors that were not there before. So please hold up merging until I found the time to look into these issues please. Best regards Uwe -- Pengutronix e.K. | Uwe Kleine-König| Industrial Linux Solutions | https://www.pengutronix.de/ | ___ ptxdist mailing list ptxdist@pengutronix.de
Re: [ptxdist] [PATCH] sqlite: enable foreign keys by default
On Fri, Feb 21, 2020 at 12:20:33PM +0100, Bruno Thomsen wrote: > Hi Michael > > Den fre. 21. feb. 2020 kl. 08.23 skrev Michael Olbrich > : > > > > On Sat, Feb 15, 2020 at 01:52:05PM +0100, Bruno Thomsen wrote: > > > Support for foreign_keys pragma was already enabled at > > > compile-time, but it was not enabled by default on new > > > database connections. It's still possible to turn off > > > enforcement at runtime. > > > This makes it easier to work on databases that uses > > > this feature without causing them to end up in an > > > invalid state. > > > > Hmm, what's the impact of this? What are the performance implications? And > > what's the default in other distros? I mean, if the default is generally > > off, then I don't expect any existing software to disable this. > > Fedora 31 has =0, I don't know about other distros, > but you can test it fast with an in-memory database. > > $ sqlite3 > sqlite> pragma foreign_keys; > 1 > sqlite> .exit FTR: on Debian the pragma command yields 0. (Tested with sqlite3 3.31.1-2 that is currently in testing/unstable) Best regards Uwe -- Pengutronix e.K. | Uwe Kleine-König| Industrial Linux Solutions | https://www.pengutronix.de/ | ___ ptxdist mailing list ptxdist@pengutronix.de
[ptxdist] Archive ptxdist@pengutronix.de on lore.kernel.org?
Hello, I and others at Pengutronix would consider it very useful to have a public-inbox archive of the ptxdist mailing list. Given that there are already a few lists archived on lore.kernel.org that (I think) are not strongly related to the Linux kernel I wondered if it was possible to add an archive for ptxdist there, too. (ptxdist has a similar purpose as buildroot and yocto, it's an environment that allows to build Linux images for embedded systems.) If yes, subscription is handled by mailman via ptxdist-requ...@pengutronix.de (with "subscribe" in the Subject). Thanks Uwe -- Pengutronix e.K. | Uwe Kleine-König| Industrial Linux Solutions | https://www.pengutronix.de/ | ___ ptxdist mailing list ptxdist@pengutronix.de
[ptxdist] [PATCH] unfs3: fix attribute setting for symlinks
unfs makes use of utime() which is unsuitable to set last access and modification times for symlinks. So make use of the O_PATH flag to open() and use the futimens() function. As a side effect this is less racy and improves resolution from seconds to nanoseconds. Note this will break builds of unfs on Windows. The futimens() function is specified in POSIX.1-2008, so this shouldn't be a problem for other platforms. Signed-off-by: Uwe Kleine-König --- Hello, note I didn't try to contact upstream. Best regards Uwe ...5-attr-use-futimens-instead-of-utime.patch | 102 ++ ...-O_PATH-to-set-attributes-of-symlink.patch | 81 ++ patches/unfs3-0.9.22/series | 4 +- 3 files changed, 186 insertions(+), 1 deletion(-) create mode 100644 patches/unfs3-0.9.22/0005-attr-use-futimens-instead-of-utime.patch create mode 100644 patches/unfs3-0.9.22/0006-attr-make-use-of-O_PATH-to-set-attributes-of-symlink.patch diff --git a/patches/unfs3-0.9.22/0005-attr-use-futimens-instead-of-utime.patch b/patches/unfs3-0.9.22/0005-attr-use-futimens-instead-of-utime.patch new file mode 100644 index ..401142fb1b16 --- /dev/null +++ b/patches/unfs3-0.9.22/0005-attr-use-futimens-instead-of-utime.patch @@ -0,0 +1,102 @@ +From: =?UTF-8?q?Uwe=20Kleine-K=C3=B6nig?= +Date: Tue, 18 Feb 2020 15:20:12 +0100 +Subject: [PATCH] attr: use futimens() instead of utime() + +The former has two relevant advantages: It works on file descriptors +instead of a path name and it has nano second resolution (instead of only +seconds). + +Note, I don't know about Windows which is probably broken by this commit +as I assume it doesn't support futimens(). +--- + attr.c | 53 - + 1 file changed, 48 insertions(+), 5 deletions(-) + +diff --git a/attr.c b/attr.c +index 2653ed0a16bb..08c9d24cb446 100644 +--- a/attr.c b/attr.c +@@ -265,12 +265,51 @@ post_op_attr get_post_cached(struct svc_req * req) + return get_post_buf(st_cache, req); + } + ++static nfsstat3 set_time(int fd, sattr3 new) ++{ ++if (new.atime.set_it != DONT_CHANGE || new.mtime.set_it != DONT_CHANGE) { ++ ++ /* atime in t[0], mtime in t[1] */ ++ struct timespec t[2]; ++ int res; ++ ++ /* compute atime to set */ ++ if (new.atime.set_it == SET_TO_SERVER_TIME) { ++ t[0].tv_sec = UTIME_NOW; ++ t[0].tv_nsec = UTIME_NOW; ++ } else if (new.atime.set_it == SET_TO_CLIENT_TIME) { ++ t[0].tv_sec = new.atime.set_atime_u.atime.seconds; ++ t[0].tv_nsec = new.atime.set_atime_u.atime.nseconds; ++ } else { /* DONT_CHANGE */ ++ t[0].tv_sec = UTIME_OMIT; ++ t[0].tv_nsec = UTIME_OMIT; ++ } ++ ++ /* compute mtime to set */ ++ if (new.mtime.set_it == SET_TO_SERVER_TIME) { ++ t[1].tv_sec = UTIME_NOW; ++ t[1].tv_nsec = UTIME_NOW; ++ } else if (new.mtime.set_it == SET_TO_CLIENT_TIME) { ++ t[1].tv_sec = new.mtime.set_mtime_u.mtime.seconds; ++ t[1].tv_nsec = new.mtime.set_mtime_u.mtime.nseconds; ++ } else { /* DONT_CHANGE */ ++ t[1].tv_sec = UTIME_OMIT; ++ t[1].tv_nsec = UTIME_OMIT; ++ } ++ ++ res = futimens(fd, ); ++ if (res == -1) ++ return setattr_err(); ++} ++return NFS3_OK; ++} ++ + /* + * setting of time, races with local filesystem + * + * there is no futimes() function in POSIX or Linux + */ +-static nfsstat3 set_time(const char *path, backend_statstruct buf, sattr3 new) ++static nfsstat3 set_time_path(const char *path, backend_statstruct buf, sattr3 new) + { + time_t new_atime, new_mtime; + struct utimbuf utim; +@@ -358,7 +397,7 @@ static nfsstat3 set_attr_unsafe(const char *path, nfs_fh3 nfh, sattr3 new) + return setattr_err(); + } + +-return set_time(path, buf, new); ++return set_time_path(path, buf, new); + } + + /* +@@ -454,14 +493,18 @@ nfsstat3 set_attr(const char *path, nfs_fh3 nfh, sattr3 new) + } + } + ++/* finally, set times */ ++res = set_time(fd, new); ++if (res != NFS3_OK) { ++ backend_close(fd); ++ return res; ++} ++ + res = backend_close(fd); + if (res == -1) { + /* error on close probably means attributes didn't make it */ + return NFS3ERR_IO; + } +- +-/* finally, set times */ +-return set_time(path, buf, new); + } + + /* diff --git a/patches/unfs3-0.9.22/0006-attr-make-use-of-O_PATH-to-set-attributes-of-symlink.patch b/patches/unfs3-0.9.22/0006-attr-make-use-of-O_PATH-to-set-attributes-of-symlink.patch new file mode 100644 index ..798927a520e6 --- /dev/null +++ b/patches/unfs3-0.9.22/0006-attr-make-use-of-O_PATH-to-set-attributes-of-symlink.patch @@ -0,0 +1,81 @@ +From: =?UTF-8?q?Uwe=20Kleine-K=C3=B6nig?= +Date: Tue, 18 Feb 2020 15:23:52 +0100 +Subject: [PATCH] attr: make use of O_PATH to set attributes of symlinks and + devices + +This fixes setting
Re: [ptxdist] [PATCH] networkmanager: version bump 1.20.8-> 1.22.6
On Wed, Feb 12, 2020 at 11:11:12AM +0100, Ulrich Ölmann wrote: > Two new config options have been added in the meantime (see NetworkManager > commits [1] & [2]) which are both set to their reasonable default values. > > The license did not change, but its text was adjusted - citing from > NetworkManagers's commit [3]: > > The update to the GPL text is purely cosmetic. However, shipping the exact > same file as GNU publishes may help distros that deduplicate the license > texts > or hardlink duplicates. > > [1] > https://gitlab.freedesktop.org/NetworkManager/NetworkManager/commit/69f048bf0ca387d2dc4683cfdfe9d170bfceb52b > [2] > https://gitlab.freedesktop.org/NetworkManager/NetworkManager/commit/d27fcd07541ae6f524115d5b0f36e14673135ca3 > [3] > https://gitlab.freedesktop.org/NetworkManager/NetworkManager/commit/e9f2ea6c22f36cb7986d2228763629ed44b9e76b > > Signed-off-by: Ulrich Ölmann > --- > rules/networkmanager.make | 10 ++ > 1 file changed, 6 insertions(+), 4 deletions(-) > > diff --git a/rules/networkmanager.make b/rules/networkmanager.make > index de0e755b3a5d..86542b6f74e1 100644 > --- a/rules/networkmanager.make > +++ b/rules/networkmanager.make > @@ -15,15 +15,15 @@ PACKAGES-$(PTXCONF_NETWORKMANAGER) += networkmanager > # > # Paths and names > # > -NETWORKMANAGER_VERSION := 1.20.8 > -NETWORKMANAGER_MD5 := c0ceb5ab14bfdfeee07536d94cc5c548 > +NETWORKMANAGER_VERSION := 1.22.6 > +NETWORKMANAGER_MD5 := 0f4b493cc0c67f94a2ad3573363eb3b2 > NETWORKMANAGER := NetworkManager-$(NETWORKMANAGER_VERSION) > NETWORKMANAGER_SUFFIX:= tar.xz > -NETWORKMANAGER_URL := > https://ftp.gnome.org/pub/GNOME/sources/NetworkManager/1.20/$(NETWORKMANAGER).$(NETWORKMANAGER_SUFFIX) > +NETWORKMANAGER_URL := > https://ftp.gnome.org/pub/GNOME/sources/NetworkManager/1.22/$(NETWORKMANAGER).$(NETWORKMANAGER_SUFFIX) I wonder if it would make sense to use some make magic to derive the directory name (1.22) from NETWORKMANAGER_VERSION. Hmm, this is more complicated than I thought, the following works: EMPTY = SPACE = $(EMPTY) $(EMPTY) NETWORKMANAGER_MAJOR := $(subst $(SPACE),.,$(wordlist 1, 2, $(subst ., ,$(NETWORKMANAGER_VERSION or NETWORKMANAGER_V1 = $(word 1, $(subst ., ,$(NETWORKMANAGER_VERSION))) NETWORKMANAGER_V2 = $(word 2, $(subst ., ,$(NETWORKMANAGER_VERSION))) NETWORKMANAGER_MAJOR = $(NETWORKMANAGER_V1).$(NETWORKMANAGER_V2) Unless someone can come up with something easier, probably not worth the effort? Best regards Uwe -- Pengutronix e.K. | Uwe Kleine-König| Industrial Linux Solutions | https://www.pengutronix.de/ | ___ ptxdist mailing list ptxdist@pengutronix.de
[ptxdist] [PATCH] gnutls: switch to https URL
This tends to work better than ftp URLs in combination with corporate firewall rules. Signed-off-by: Uwe Kleine-König --- rules/gnutls.make | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/rules/gnutls.make b/rules/gnutls.make index f07ad8ddb517..9319887d3405 100644 --- a/rules/gnutls.make +++ b/rules/gnutls.make @@ -18,7 +18,7 @@ GNUTLS_VERSION:= 3.6.10 GNUTLS_MD5 := 254e756d882a38ce9ad6f47589330d87 GNUTLS := gnutls-$(GNUTLS_VERSION) GNUTLS_SUFFIX := tar.xz -GNUTLS_URL := ftp://ftp.gnutls.org/gcrypt/gnutls/v3.6/$(GNUTLS).$(GNUTLS_SUFFIX) +GNUTLS_URL := https://www.gnupg.org/ftp/gcrypt/gnutls/v3.6/$(GNUTLS).$(GNUTLS_SUFFIX) GNUTLS_SOURCE := $(SRCDIR)/$(GNUTLS).$(GNUTLS_SUFFIX) GNUTLS_DIR := $(BUILDDIR)/$(GNUTLS) GNUTLS_LICENSE := LGPL-3.0-or-later -- 2.25.0 ___ ptxdist mailing list ptxdist@pengutronix.de
Re: [ptxdist] [PATCH] perl: Add patch for compatibility with recent libcrypt implementations.
On Thu, Oct 10, 2019 at 10:44:45AM +0200, Björn Esser wrote: > Signed-off-by: Björn Esser > --- > .../0001-It-s-lpthread-not-lpthreads.patch| 7 +++- > ...-fix-build-failure-with-recent-glibc.patch | 39 +++ > patches/perl-5.18.2/series| 3 +- > 3 files changed, 47 insertions(+), 2 deletions(-) > create mode 100644 > patches/perl-5.18.2/0002-fix-build-failure-with-recent-glibc.patch > > diff --git a/patches/perl-5.18.2/0001-It-s-lpthread-not-lpthreads.patch > b/patches/perl-5.18.2/0001-It-s-lpthread-not-lpthreads.patch > index 319cefb2e..7a7bb1beb 100644 > --- a/patches/perl-5.18.2/0001-It-s-lpthread-not-lpthreads.patch > +++ b/patches/perl-5.18.2/0001-It-s-lpthread-not-lpthreads.patch > @@ -1,15 +1,20 @@ > From: Michael Olbrich > Date: Wed, 23 Apr 2014 09:02:25 +0200 > Subject: [PATCH] It's -lpthread not -lpthreads > +MIME-Version: 1.0 > +Content-Type: text/plain; charset=UTF-8 > +Content-Transfer-Encoding: 8bit > > Otherwise linking fails with missing pthread symbols when building with > with -Dusethreads > + > +Signed-off-by: Björn Esser This seems unrelated. Does Perl use the S-o-b at all? Best regards Uwe -- Pengutronix e.K. | Uwe Kleine-König| Industrial Linux Solutions | http://www.pengutronix.de/ | ___ ptxdist mailing list ptxdist@pengutronix.de
[ptxdist] [PATCH] mmc-utils: use https:// URL instead of git://
In contrast to git:// https:// is usually not blocked in corporate firewalls and provides some anonymity. Signed-off-by: Uwe Kleine-König --- rules/mmc-utils.make | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/rules/mmc-utils.make b/rules/mmc-utils.make index a2501b19a2f2..ba1a10f42d2b 100644 --- a/rules/mmc-utils.make +++ b/rules/mmc-utils.make @@ -19,7 +19,7 @@ MMC_UTILS_VERSION := 2018-03-27-gb4fe0c8c0e57 MMC_UTILS_MD5 := 1018cdca94530054cc11e4bcb97d4b11 MMC_UTILS := mmc-utils-$(MMC_UTILS_VERSION) MMC_UTILS_SUFFIX := tar.gz -MMC_UTILS_URL := git://git.kernel.org/pub/scm/linux/kernel/git/cjb/mmc-utils.git;tag=$(MMC_UTILS_VERSION) +MMC_UTILS_URL := https://git.kernel.org/pub/scm/linux/kernel/git/cjb/mmc-utils.git;tag=$(MMC_UTILS_VERSION) MMC_UTILS_SOURCE := $(SRCDIR)/$(MMC_UTILS).$(MMC_UTILS_SUFFIX) MMC_UTILS_DIR := $(BUILDDIR)/$(MMC_UTILS) MMC_UTILS_LICENSE := GPL-2.0-only AND BSD-3-Clause -- 2.20.1 ___ ptxdist mailing list ptxdist@pengutronix.de
Re: [ptxdist] [RFC 3/4] libptxdist: add a function to find unreferenced source directories
$Subject ~= s/source/patch/ ? Best regards Uwe -- Pengutronix e.K. | Uwe Kleine-König| Industrial Linux Solutions | http://www.pengutronix.de/ | ___ ptxdist mailing list ptxdist@pengutronix.de
[ptxdist] [PATCH v2] nfsutils: don't install nfsiostat without ensuring python is available
nfsiostat is a python script. Introduce a new suboption with appropriate selects to assert /usr/bin/python being available and don't install nfsiostat if this is disabled. Signed-off-by: Uwe Kleine-König --- Hello, On Wed, Aug 07, 2019 at 01:45:35PM +0200, Michael Olbrich wrote: > On Wed, Aug 07, 2019 at 11:27:33AM +0200, Uwe Kleine-König wrote: > > --- a/rules/nfsutils.in > > +++ b/rules/nfsutils.in > > @@ -9,6 +9,7 @@ menuconfig NFSUTILS > > select RPCBIND if RUNTIME > > select LIBTIRPC if GLOBAL_IPV6 > > select ROOTFS_PROTOCOLS if NFSUTILS_CLIENT > > + select PYTHON if NFSUTILS_PYTHON > > Hmm, I don't like this. The scripts works with python3 as well if > PTXCONF_PYTHON3_SYMLINK is enabled. > > Maybe like this: > > select PYTHON3 if NFSUTILS_PYTHON && !PYTHON && RUNTIME > select PYTHON3_SYMLINK if NFSUTILS_PYTHON && !PYTHON && RUNTIME > > It's not a build-time dependency, and I think we should prefer Python3 > here. This is fine for me, too, and implemented in this v2. Thanks for the feed-back Uwe rules/nfsutils.in | 8 rules/nfsutils.make | 5 - 2 files changed, 12 insertions(+), 1 deletion(-) diff --git a/rules/nfsutils.in b/rules/nfsutils.in index ef7eb7b98d60..77ea1a2ca877 100644 --- a/rules/nfsutils.in +++ b/rules/nfsutils.in @@ -9,6 +9,8 @@ menuconfig NFSUTILS select RPCBIND if RUNTIME select LIBTIRPC if GLOBAL_IPV6 select ROOTFS_PROTOCOLS if NFSUTILS_CLIENT + select PYTHON3 if NFSUTILS_PYTHON && !PYTHON && RUNTIME + select PYTHON3_SYMLINK if NFSUTILS_PYTHON && !PYTHON && RUNTIME select ROOTFS_VAR_LIB if RUNTIME help The Sun Network Filesystem (NFS) protocol provides transparent remote @@ -32,6 +34,12 @@ config NFSUTILS_SERVER help Install all tools and daemons to run an NFS server. +config NFSUTILS_PYTHON + bool + prompt "install python utilities" + help + Install tools depending on Python. Currently this affects nfsiostat only + config NFSUTILS_NFSD_STARTSCRIPT bool depends on NFSUTILS_SERVER diff --git a/rules/nfsutils.make b/rules/nfsutils.make index 5986d51fd716..ff1f4cae0560 100644 --- a/rules/nfsutils.make +++ b/rules/nfsutils.make @@ -77,7 +77,6 @@ $(STATEDIR)/nfsutils.targetinstall: @$(call install_fixup, nfsutils,DESCRIPTION,"Network Filesystem Support") @$(call install_copy, nfsutils, 0, 0, 0755, -, /usr/sbin/nfsstat) - @$(call install_copy, nfsutils, 0, 0, 0755, -, /usr/sbin/nfsiostat) @$(call install_copy, nfsutils, 0, 0, 0755, -, /usr/sbin/showmount) @$(call install_copy, nfsutils, 0, 0, 0755, -, /usr/sbin/rpc.statd) @@ -119,6 +118,10 @@ ifdef PTXCONF_NFSUTILS_SERVER /etc/exports) endif +ifdef PTXCONF_NFSUTILS_PYTHON + @$(call install_copy, nfsutils, 0, 0, 0755, -, /usr/sbin/nfsiostat) +endif + # # # # busybox init: start scripts # # -- 2.20.1 ___ ptxdist mailing list ptxdist@pengutronix.de
[ptxdist] [PATCH] nfsutils: don't install nfsiostat without ensuring python is available
nfsiostat is a python script. Introduce a new suboption to select PYTHON and don't install nfsiostat if this is disabled. Signed-off-by: Uwe Kleine-König --- rules/nfsutils.in | 7 +++ rules/nfsutils.make | 5 - 2 files changed, 11 insertions(+), 1 deletion(-) diff --git a/rules/nfsutils.in b/rules/nfsutils.in index ef7eb7b98d60..34d237737024 100644 --- a/rules/nfsutils.in +++ b/rules/nfsutils.in @@ -9,6 +9,7 @@ menuconfig NFSUTILS select RPCBIND if RUNTIME select LIBTIRPC if GLOBAL_IPV6 select ROOTFS_PROTOCOLS if NFSUTILS_CLIENT + select PYTHON if NFSUTILS_PYTHON select ROOTFS_VAR_LIB if RUNTIME help The Sun Network Filesystem (NFS) protocol provides transparent remote @@ -32,6 +33,12 @@ config NFSUTILS_SERVER help Install all tools and daemons to run an NFS server. +config NFSUTILS_PYTHON + bool + prompt "install python utilities" + help + Install tools depending on Python. Currently this affects nfsiostat only + config NFSUTILS_NFSD_STARTSCRIPT bool depends on NFSUTILS_SERVER diff --git a/rules/nfsutils.make b/rules/nfsutils.make index 5986d51fd716..ff1f4cae0560 100644 --- a/rules/nfsutils.make +++ b/rules/nfsutils.make @@ -77,7 +77,6 @@ $(STATEDIR)/nfsutils.targetinstall: @$(call install_fixup, nfsutils,DESCRIPTION,"Network Filesystem Support") @$(call install_copy, nfsutils, 0, 0, 0755, -, /usr/sbin/nfsstat) - @$(call install_copy, nfsutils, 0, 0, 0755, -, /usr/sbin/nfsiostat) @$(call install_copy, nfsutils, 0, 0, 0755, -, /usr/sbin/showmount) @$(call install_copy, nfsutils, 0, 0, 0755, -, /usr/sbin/rpc.statd) @@ -119,6 +118,10 @@ ifdef PTXCONF_NFSUTILS_SERVER /etc/exports) endif +ifdef PTXCONF_NFSUTILS_PYTHON + @$(call install_copy, nfsutils, 0, 0, 0755, -, /usr/sbin/nfsiostat) +endif + # # # # busybox init: start scripts # # -- 2.20.1 ___ ptxdist mailing list ptxdist@pengutronix.de
Re: [ptxdist] some problems with start ptxdist
Hello Petr, On Wed, Jul 24, 2019 at 07:23:34AM +0200, Petr Lajtkep wrote: > petr@debian:~$ ptxdist > /usr/local/lib/ptxdist-2019.07.0/bin/ptxdist: line 2261: > /home/petr/.ptxdist/ptxdistrc-2010.01: No such file or directory > /usr/local/lib/ptxdist-2019.07.0/scripts/libptxdist.sh: line 105: > /home/petr/.ptxdist/ptxdistrc-2010.01: No such file or directory > cp: cannot stat '/home/petr/.ptxdist/ptxdistrc-2010.01': No such file or > directory > PTXdist: fatal error ... cannot start, sorry! > petr@debian:~$ This is a problem in v2019.07.0 that is fixed in master. See commit a653df5d3f35. Best regards Uwe -- Pengutronix e.K. | Uwe Kleine-König| Industrial Linux Solutions | http://www.pengutronix.de/ | ___ ptxdist mailing list ptxdist@pengutronix.de
Re: [ptxdist] [PATCH] patchin: Don't trigger git garbage collection
Hello On Thu, Apr 04, 2019 at 09:18:09PM +0200, Uwe Kleine-König wrote: > When extracting the sources for a big package (e.g. kernel) git forks a > process to repack the repository "for optimum performance" (unless > configured otherwise for example in ~/.gitconfig). However this is time > that is not well spend when waiting for a build to complete. > > So ensure that ptxdist doesn't trigger this housekeeping. As this > cleanup is sensible as soon as you start modifying the git repository, > suppress it only via the commandline instead of modifying the > repository's configuration. > > Signed-off-by: Uwe Kleine-König The problem was noticed and reported to me by Tim Knecht. As he is currently not subscribed he cannot answer himself. So please take Reported-by: Tim Knecht from me. Best regards Uwe -- Pengutronix e.K. | Uwe Kleine-König| Industrial Linux Solutions | http://www.pengutronix.de/ | ___ ptxdist mailing list ptxdist@pengutronix.de
[ptxdist] [PATCH] patchin: Don't trigger git garbage collection
When extracting the sources for a big package (e.g. kernel) git forks a process to repack the repository "for optimum performance" (unless configured otherwise for example in ~/.gitconfig). However this is time that is not well spend when waiting for a build to complete. So ensure that ptxdist doesn't trigger this housekeeping. As this cleanup is sensible as soon as you start modifying the git repository, suppress it only via the commandline instead of modifying the repository's configuration. Signed-off-by: Uwe Kleine-König --- scripts/lib/ptxd_make_world_patchin.sh | 35 +- 1 file changed, 23 insertions(+), 12 deletions(-) diff --git a/scripts/lib/ptxd_make_world_patchin.sh b/scripts/lib/ptxd_make_world_patchin.sh index 606d9ebc028b..021989e27e06 100644 --- a/scripts/lib/ptxd_make_world_patchin.sh +++ b/scripts/lib/ptxd_make_world_patchin.sh @@ -45,6 +45,17 @@ ptxd_make_world_patchin_apply_init() export -f ptxd_make_world_patchin_apply_init +# +# Wrapper that prevents running automatic repack during ptxdist runs to not +# loose more time on patch stacks that maybe are not modified anyhow. +# +__git() +{ + git -c gc.auto=0 "$@" +} +export -f __git + + # # initialize git database in $pkg_patchin_dir and do initial commit # @@ -56,14 +67,14 @@ ptxd_make_world_patchin_apply_git_init() # is already git repo? if [ "${git_dir}" != ".git" ]; then echo "patchin: git: initializing repository" - git init -q && - git add -f . && - git commit -q -m "initial commit" --author="ptxdist-${PTXDIST_VERSION_FULL} " && - git tag "${pkg_pkg//\~/-}" && - git tag base && - git config alias.ptx-patches "!${PTXDIST_TOPDIR}/scripts/git-ptx-patches \"\${@}\"" && - git config diff.renames false && - git config core.abbrev 12 && + __git init -q && + __git add -f . && + __git commit -q -m "initial commit" --author="ptxdist-${PTXDIST_VERSION_FULL} " && + __git tag "${pkg_pkg//\~/-}" && + __git tag base && + __git config alias.ptx-patches "!${PTXDIST_TOPDIR}/scripts/git-ptx-patches \"\${@}\"" && + __git config diff.renames false && + __git config core.abbrev 12 && echo "patchin: git: done" fi } @@ -139,24 +150,24 @@ ptxd_make_world_patchin_apply_git() mv ${pkg_patchin_dir}/.ptxdist/git-patches/series{,.1} || return mv ${pkg_patchin_dir}/.ptxdist/git-patches/series{.${tag},} || return - git quiltimport \ + __git quiltimport \ --patches "${pkg_patchin_dir}/.ptxdist/git-patches" \ --author "unknown author " && echo "tagging -> ${tag}" && - git tag -f ${tag} && + __git tag -f ${tag} && mv ${pkg_patchin_dir}/.ptxdist/git-patches/series{,.${tag}} || return mv ${pkg_patchin_dir}/.ptxdist/git-patches/series{.1,} || return fi done -git quiltimport \ +__git quiltimport \ --patches "${pkg_patchin_dir}/.ptxdist/git-patches" \ --author "unknown author " && # skip patches generated by other tools with '# -end' as end marker if tail -n1 "${pkg_patchin_dir}/.ptxdist/series" | grep -q '# [^ ]*-end$'; then local last="$(sed -n 's/^0*\([1-9][0-9]*\)-.*/\1/p' "${pkg_patchin_dir}/.ptxdist/series" | tail -n1)" - git tag -f base || return + __git tag -f base || return local next=1000 if [ -n "${last}" ]; then next="$[(1+${last}/1000)*1000]" -- 2.20.1 ___ ptxdist mailing list ptxdist@pengutronix.de
[ptxdist] [PATCH] fstab: don't explicitly mount debugfs
If the kernel in use doesn't support it having an explicit entry results in a failure. On systems with systemd it is mounted anyways by /lib/systemd/system/sys-kernel-debug.mount iff DEBUG_FS is enabled in the kernel. Signed-off-by: Uwe Kleine-König --- projectroot/etc/fstab | 1 - 1 file changed, 1 deletion(-) diff --git a/projectroot/etc/fstab b/projectroot/etc/fstab index 269e35f66859..88099aacc17e 100644 --- a/projectroot/etc/fstab +++ b/projectroot/etc/fstab @@ -6,7 +6,6 @@ proc /proc procnosuid,nodev,noexec 0 0 devpts /dev/ptsdevpts nosuid,noexec,mode=620,gid=112 0 0 sysfs /syssysfs nosuid,nodev,noexec 0 0 -debugfs/sys/kernel/debug debugfs noauto 0 0 # ramdisks tmpfs /tmptmpfs nosuid,nodev,mode=1777,size=20% 0 0 -- 2.19.1 ___ ptxdist mailing list ptxdist@pengutronix.de
[ptxdist] [REGRESSION] a73b00a breaks ptxdist nfsroot
Hello, with a project using ptxdist-2018.11.0 I get the following: $ ptxdist nfsroot ptxdist: error: glib is not a package or not selected make: *** [ptxd_make_nfsd] Error 1 because ptx_packages_selected is empty. When I revert a73b00a71955 ("ptxd_do_xpkg_map: validate package list") it works fine. I guess world/image/env/impl must somehow be called before nfsroot calls ptxd_do_xpkg_map, but that's way to deep into ptxdist for me, so I let you work out the right fix. :-) Best regards Uwe -- Pengutronix e.K. | Uwe Kleine-König| Industrial Linux Solutions | http://www.pengutronix.de/ | ___ ptxdist mailing list ptxdist@pengutronix.de
Re: [ptxdist] [PATCH] host-autotools-automake: version bump 1.5.1 -> 1.6.1
Hello, On Tue, May 15, 2018 at 10:53:54AM +, Baeuerle, Florian wrote: > Am Dienstag, den 15.05.2018, 08:30 +0200 schrieb Uwe Kleine-König: > > Hello, > > > > On Mon, May 14, 2018 at 10:54:56AM +, Baeuerle, Florian wrote: > > > Removed patch: > > > > > > - 0001-disable-manpages.patch: > > > Build succeeds without this patch. > > > > the problem that was fixed with this patch maybe only happened on Arch, > > at least that's what reading through > > 5d107ecae118ac3dd49fdf10c54901eb0a4c3bdf suggests. Adding Clemens Gruber > > to the recipients, who authored the patch. Maybe he can test/comment? > > If the problem only occurred on Arch Linux, then I can confirm it's no > longer occurring - I've built my ptxdist project successfully on Arch Linux > with this patch (which removed the patch in question). > > Did some further investigation. > > Apparently, help2man was updated a while after the automake v1.15.1 release: > https://git.savannah.gnu.org/cgit/automake.git/commit/doc/help2man?id=9322f409a957f153b38ff37ba79ddf4c19cff6ca > > The previous help2man v1.40.8 was from 2012: > https://salsa.debian.org/bod/help2man/commit/4efbcca2dc796a5f9bf2797905fc9e6bee6a6afc FTR: I just got a failure to build on Centos 7.5.1804 which was fixed by the patch that got dropped here. Didn't debug further yet. Best regards Uwe -- Pengutronix e.K. | Uwe Kleine-König| Industrial Linux Solutions | http://www.pengutronix.de/ | ___ ptxdist mailing list ptxdist@pengutronix.de
Re: [ptxdist] [PATCH] setlocalversion: only use tags actually on current branch
On Wed, Dec 12, 2018 at 09:14:52AM +, Denis OSTERLAND wrote: > Consider following situation: > >tag C > v > A---B---C topic > / \ >D---E---F---G---H master >^ > tag E > > PTXDIST_BSP_AUTOVERSION at commit G will be E-2-g, > but at commit H it will be C-3-g. At H I'd expect C-1-g. I think in general using --first-parent is wrong. It's only in some workflows (i.e. when there is a single person that does "mainline merges") where the first parent is special. > With --first-parent it will be E-3-g. > This is more intuitive result, > especially when more than just one branch gets merged. > > Signed-off-by: Denis Osterland > --- > scripts/kernel/setlocalversion | 4 ++-- > 1 file changed, 2 insertions(+), 2 deletions(-) > > diff --git a/scripts/kernel/setlocalversion b/scripts/kernel/setlocalversion > index 3feb87dce..7eaaaf413 100755 > --- a/scripts/kernel/setlocalversion > +++ b/scripts/kernel/setlocalversion > @@ -55,12 +55,12 @@ if head=`git ${git_abbrev} rev-parse --verify --short > HEAD 2>/dev/null`; then > > # If we are at a tagged commit (like "ptxdist-2010.05.0"), we > # print it here > - if atag="`git describe --exact-match 2>/dev/null`"; then > + if atag="`git describe --first-parent --exact-match 2>/dev/null`"; then With --exact-match --first-parent is a noop, isn't it? > echo "$atag" | awk -F- '{printf("%s", $(NF))}' unrelated to this patch: if atag is "ELWMS-1.0-rc17" the result here is "rc17", isn't it? Best regards Uwe -- Pengutronix e.K. | Uwe Kleine-König| Industrial Linux Solutions | http://www.pengutronix.de/ | ___ ptxdist mailing list ptxdist@pengutronix.de
Re: [ptxdist] [PATCH 2/5] ptxdist: add tpm2-tss version 2.1.0
Hello Marc, On Wed, Nov 28, 2018 at 09:02:10AM +0100, Marc Kleine-Budde wrote: > On 11/26/18 11:23 AM, Thorsten K. Scherer wrote: > > Signed-off-by: Thorsten K. Scherer > > --- > > ...rom-the-list-of-libs-to-link-against.patch | 21 ++ > > patches/tpm2-tss-2.1.0/series | 4 ++ > > rules/tpm.in | 5 ++ > > rules/tpm2-tss.in | 22 +++ > > rules/tpm2-tss.make | 66 +++ > > 5 files changed, 118 insertions(+) > > create mode 100644 > > patches/tpm2-tss-2.1.0/0001-drop-lgcrypt-from-the-list-of-libs-to-link-against.patch > > create mode 100644 patches/tpm2-tss-2.1.0/series > > create mode 100644 rules/tpm.in > > create mode 100644 rules/tpm2-tss.in > > create mode 100644 rules/tpm2-tss.make > > > > diff --git > > a/patches/tpm2-tss-2.1.0/0001-drop-lgcrypt-from-the-list-of-libs-to-link-against.patch > > > > b/patches/tpm2-tss-2.1.0/0001-drop-lgcrypt-from-the-list-of-libs-to-link-against.patch > > new file mode 100644 > > index 000..0230570 > > --- /dev/null > > +++ > > b/patches/tpm2-tss-2.1.0/0001-drop-lgcrypt-from-the-list-of-libs-to-link-against.patch > > @@ -0,0 +1,21 @@ > > +From: =?UTF-8?q?Uwe=20Kleine-K=C3=B6nig?= > > +Date: Fri, 12 Oct 2018 11:13:42 +0200 > > +Subject: [PATCH] drop -lgcrypt from the list of libs to link against > > + > > +If --with-crypto=ossl is used -lgcrypt is wrong. There shouldn't be any > > need > > +to explicitly tell the linker because the library already depends the > > chosen > > +crypto lib > > S-o-b missing I'm unsure why you think a S-o-b is necessary here. The upstream project requires a S-o-b by the submitter (who is not necessarily the patch author) only when the patch is sent to them for application. And in my eyes the S-o-b on the ptxdist-patch already covers that the whole content (so including the added patch file) is compatible with ptxdist's license. Can you clearify? Best regards Uwe -- Pengutronix e.K. | Uwe Kleine-König| Industrial Linux Solutions | http://www.pengutronix.de/ | ___ ptxdist mailing list ptxdist@pengutronix.de
Re: [ptxdist] [PATCH v2] Makefile: don't install .done, .gitignore and vim swap files
Hello Roland, On Tue, Nov 13, 2018 at 09:10:56PM +0100, Roland Hieber wrote: > .gitignore is only resonable in git repositories, and we exclude .git/. s/resonable/reasonable/ Best regards Uwe -- Pengutronix e.K. | Uwe Kleine-König| Industrial Linux Solutions | http://www.pengutronix.de/ | ___ ptxdist mailing list ptxdist@pengutronix.de
[ptxdist] [PATCH v4] minicoredumper: New package
Tested-by: Roland Hieber Acked-by: Roland Hieber Tested-by: Alexander Dahl Signed-off-by: Uwe Kleine-König --- Hello, On Tue, Oct 23, 2018 at 03:37:44PM +0200, Alexander Dahl wrote: > what about the minor issues I reported with Message-Id > <4087030.zncrgpGN7O@ada> back in August? ;-) hupsi, I missed not only these, but also the feedback by Roland (even though he only provided his ack/test). So: Changes since v3, sent with Message-Id: 20181023132846.1473-1-u.kleine-koe...@pengutronix.de: - add Tested-by and Ack by Roland and Alexander - make PTXCONF_MINICOREDUMPER tristate (found by Alexander) - fix a tpyo (found by Alexander) rules/minicoredumper.in | 21 rules/minicoredumper.make | 68 +++ 2 files changed, 89 insertions(+) create mode 100644 rules/minicoredumper.in create mode 100644 rules/minicoredumper.make diff --git a/rules/minicoredumper.in b/rules/minicoredumper.in new file mode 100644 index ..cc8008d1c39f --- /dev/null +++ b/rules/minicoredumper.in @@ -0,0 +1,21 @@ +## SECTION=shell_and_console + +config MINICOREDUMPER + tristate + prompt "minicoredumper" + select LIBELF + select JSON_C + help + The minicoredumper project provides a program that handles the + creation of core dump files on Linux. It can produce much smaller + core dump files by making use of sparse files, compression, and + allowing the user to configure what parts of the process memory image + should be dumped. + + Note that there are currently no mechanisms integrated to + automatically register minicoredumper as kernel coredump helper. So + you might want to do + + echo '|/usr/sbin/minicoredumper %P %u %g %s %t %h %e' > /proc/sys/kernel/core_pattern + + by hand to actually make use of it. diff --git a/rules/minicoredumper.make b/rules/minicoredumper.make new file mode 100644 index ..af01ece49805 --- /dev/null +++ b/rules/minicoredumper.make @@ -0,0 +1,68 @@ +# -*-makefile-*- +# +# Copyright (C) 2018 by Uwe Kleine-Koenig +# +# See CREDITS for details about who has contributed to this project. +# +# For further information about the PTXdist project and license conditions +# see the README file. +# + +# +# We provide this package +# +PACKAGES-$(PTXCONF_MINICOREDUMPER) += minicoredumper + +# +# Paths and names +# +MINICOREDUMPER_VERSION := 2.0.1 +MINICOREDUMPER_MD5 := 813b864e0c6a833d14bab244723de6a4 +MINICOREDUMPER := minicoredumper-$(MINICOREDUMPER_VERSION) +MINICOREDUMPER_SUFFIX := tar.xz +MINICOREDUMPER_URL := https://linutronix.de/minicoredumper/files/$(MINICOREDUMPER).$(MINICOREDUMPER_SUFFIX) +MINICOREDUMPER_SOURCE := $(SRCDIR)/$(MINICOREDUMPER).$(MINICOREDUMPER_SUFFIX) +MINICOREDUMPER_DIR := $(BUILDDIR)/$(MINICOREDUMPER) +MINICOREDUMPER_LICENSE := BSD AND LGPL-2.1 +MINICOREDUMPER_LICENSE_FILES := file://COPYING;md5=71827c617ec7b45a0dd23658347cc1e9 + +# +# Prepare +# + +MINICOREDUMPER_CONF_TOOL := autoconf +MINICOREDUMPER_CONF_OPT := \ + $(CROSS_AUTOCONF_USR) \ + --disable-silent-rules \ + --without-coreinject \ + --with-minicoredumper \ + --without-minicoredumper_regd \ + --without-minicoredumper_trigger \ + --without-libminicoredumper \ + --without-minicoredumper_demo \ + --without-werror \ + +# +# Target-Install +# + +$(STATEDIR)/minicoredumper.targetinstall: + @$(call targetinfo) + + @$(call install_init, minicoredumper) + @$(call install_fixup, minicoredumper, PRIORITY, optional) + @$(call install_fixup, minicoredumper, SECTION, base) + @$(call install_fixup, minicoredumper, AUTHOR, "Uwe Kleine-Koenig ") + @$(call install_fixup, minicoredumper, DESCRIPTION, missing) + + @$(call install_copy, minicoredumper, 0, 0, 0755, -, /usr/sbin/minicoredumper) + + @$(call install_alternative_tree, minicoredumper, 0, 0, /etc/minicoredumper) + + @$(call install_copy, minicoredumper, 0, 0, 0755, /var/crash/minicoredumper) + + @$(call install_finish, minicoredumper) + + @$(call touch) + +# vim: syntax=make -- 2.19.0 ___ ptxdist mailing list ptxdist@pengutronix.de
[ptxdist] [PATCH v3] minicoredumper: New package
Signed-off-by: Uwe Kleine-König --- Changes since v2, sent with Message-Id: 20180814161539.16983-1-u.kleine-koe...@pengutronix.de: - drop $(call install_copy, minicoredumper, 0, 0, 0755, /var/crash) rules/minicoredumper.in | 21 rules/minicoredumper.make | 68 +++ 2 files changed, 89 insertions(+) create mode 100644 rules/minicoredumper.in create mode 100644 rules/minicoredumper.make diff --git a/rules/minicoredumper.in b/rules/minicoredumper.in new file mode 100644 index ..969be8210c88 --- /dev/null +++ b/rules/minicoredumper.in @@ -0,0 +1,21 @@ +## SECTION=shell_and_console + +config MINICOREDUMPER + bool + prompt "minicoredumper" + select LIBELF + select JSON_C + help + The minicoredumper project provides a program that handles the + creation of core dump files on Linux. It can produce much smaller + core dump files by making use of sparse files, compression, and + allowing the user to configure what parts of the process memory image + should be dumped. + + Note that there are currently no mechnisms integrated to + automatically register minicoredumper as kernel coredump helper. So + you might want to do + + echo '|/usr/sbin/minicoredumper %P %u %g %s %t %h %e' > /proc/sys/kernel/core_pattern + + by hand to actually make use of it. diff --git a/rules/minicoredumper.make b/rules/minicoredumper.make new file mode 100644 index ..af01ece49805 --- /dev/null +++ b/rules/minicoredumper.make @@ -0,0 +1,68 @@ +# -*-makefile-*- +# +# Copyright (C) 2018 by Uwe Kleine-Koenig +# +# See CREDITS for details about who has contributed to this project. +# +# For further information about the PTXdist project and license conditions +# see the README file. +# + +# +# We provide this package +# +PACKAGES-$(PTXCONF_MINICOREDUMPER) += minicoredumper + +# +# Paths and names +# +MINICOREDUMPER_VERSION := 2.0.1 +MINICOREDUMPER_MD5 := 813b864e0c6a833d14bab244723de6a4 +MINICOREDUMPER := minicoredumper-$(MINICOREDUMPER_VERSION) +MINICOREDUMPER_SUFFIX := tar.xz +MINICOREDUMPER_URL := https://linutronix.de/minicoredumper/files/$(MINICOREDUMPER).$(MINICOREDUMPER_SUFFIX) +MINICOREDUMPER_SOURCE := $(SRCDIR)/$(MINICOREDUMPER).$(MINICOREDUMPER_SUFFIX) +MINICOREDUMPER_DIR := $(BUILDDIR)/$(MINICOREDUMPER) +MINICOREDUMPER_LICENSE := BSD AND LGPL-2.1 +MINICOREDUMPER_LICENSE_FILES := file://COPYING;md5=71827c617ec7b45a0dd23658347cc1e9 + +# +# Prepare +# + +MINICOREDUMPER_CONF_TOOL := autoconf +MINICOREDUMPER_CONF_OPT := \ + $(CROSS_AUTOCONF_USR) \ + --disable-silent-rules \ + --without-coreinject \ + --with-minicoredumper \ + --without-minicoredumper_regd \ + --without-minicoredumper_trigger \ + --without-libminicoredumper \ + --without-minicoredumper_demo \ + --without-werror \ + +# +# Target-Install +# + +$(STATEDIR)/minicoredumper.targetinstall: + @$(call targetinfo) + + @$(call install_init, minicoredumper) + @$(call install_fixup, minicoredumper, PRIORITY, optional) + @$(call install_fixup, minicoredumper, SECTION, base) + @$(call install_fixup, minicoredumper, AUTHOR, "Uwe Kleine-Koenig ") + @$(call install_fixup, minicoredumper, DESCRIPTION, missing) + + @$(call install_copy, minicoredumper, 0, 0, 0755, -, /usr/sbin/minicoredumper) + + @$(call install_alternative_tree, minicoredumper, 0, 0, /etc/minicoredumper) + + @$(call install_copy, minicoredumper, 0, 0, 0755, /var/crash/minicoredumper) + + @$(call install_finish, minicoredumper) + + @$(call touch) + +# vim: syntax=make -- 2.19.0 ___ ptxdist mailing list ptxdist@pengutronix.de
Re: [ptxdist] [PATCH] imx-kobs: New package
On Tue, Oct 23, 2018 at 02:38:55PM +0200, Guillermo Rodriguez Garcia wrote: > > What about IMX_KOBS_LICENSE_FILES := file://COPYING;md5=… > > Never did this before, I thought it was only necessary for > not-well-known licenses. Should I add it? This has nothing to do with well-known or not. It's to catch license changes when the version is bumped. Best regards Uwe -- Pengutronix e.K. | Uwe Kleine-König| Industrial Linux Solutions | http://www.pengutronix.de/ | ___ ptxdist mailing list ptxdist@pengutronix.de
[ptxdist] [PATCH] ptxd_install_replace_figlet: replace all backslashes, not only the first
Fixes: f53889792197 ("ptxd_install_replace_figlet: Use sed instead of awk to quote backslashes") Signed-off-by: Uwe Kleine-König --- scripts/lib/ptxd_make_xpkg_pkg.sh | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/scripts/lib/ptxd_make_xpkg_pkg.sh b/scripts/lib/ptxd_make_xpkg_pkg.sh index b3f2f20f3e4e..766d46ac3238 100644 --- a/scripts/lib/ptxd_make_xpkg_pkg.sh +++ b/scripts/lib/ptxd_make_xpkg_pkg.sh @@ -587,7 +587,7 @@ install replace figlet: # figlet needs to be replaced by . As a \ in sed needs to be # quoted, too, this results in eight backslashes in the replacement # string. -etcissue) sed 's,\\,,';; +etcissue) sed 's,\\,,g';; *) ;; esac | \ awk '{ if ($0 !~ "^ *$") printf("%s\\n", $0) }' # newlines for sed -- 2.19.0 ___ ptxdist mailing list ptxdist@pengutronix.de
[ptxdist] [PATCH] ptxd_install_replace_figlet: Use sed instead of awk to quote backslashes
TL;DR: awk and backslashes in the replacement argument of gsub is ugly. With GNU Awk 4.0.2 we have: $ echo '\' | awk '{ gsub("", ""); print }' and with GNU Awk 4.1.4 (and newer) we have: $ echo '\' | awk '{ gsub("", ""); print }' This results in too much backslashes in /etc/issues for users of the former version. For the glory details I recommend reading through https://www.gnu.org/software/gawk/manual/html_node/Gory-Details.html https://bugs.debian.org/909147 As it's awkward to come up with an expression that does the same on both versions (e.g. using --posix) use sed instead. Signed-off-by: Uwe Kleine-König --- scripts/lib/ptxd_make_xpkg_pkg.sh | 14 +++--- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/scripts/lib/ptxd_make_xpkg_pkg.sh b/scripts/lib/ptxd_make_xpkg_pkg.sh index 5e69cefdc85d..685f4e549446 100644 --- a/scripts/lib/ptxd_make_xpkg_pkg.sh +++ b/scripts/lib/ptxd_make_xpkg_pkg.sh @@ -584,13 +584,13 @@ install replace figlet: local escapemode="$2" figlet -d "${PTXDIST_SYSROOT_HOST}/share/figlet" -- "${value}" | \ case "$escapemode" in -# a lot of leaning toothpicks because we need to escape a literal -# '\' with '\\' on multiple levels: -# - one level for the string inside awk: -> -# - one level for the shell string after sed -e: -> -# - one level for the s expression inside sed:-> \\ -# - and finally, one level for /etc/issue:-> \ -etcissue) awk '{ gsub("", ""); print }' ;; +# /etc/issue needs each backslash quoted by another backslash. As +# the string is interpreted by the shell once more below, another +# level of quoting is needed such that every \ in the output of +# figlet needs to be replaced by . As a \ in sed needs to be +# quoted, too, this results in eight backslashes in the replacement +# string. +etcissue) sed 's,\\,,';; *) ;; esac | \ awk '{ if ($0 !~ "^ *$") printf("%s\\n", $0) }' # newlines for sed -- 2.19.0 ___ ptxdist mailing list ptxdist@pengutronix.de
Re: [ptxdist] [PATCH v2 1/3] busybox: don't install /sbin/init link if busybox is not the initmethod
On Wed, Aug 22, 2018 at 10:27:49AM +0200, Michael Grzeschik wrote: > This patch fixes the case where we use other initmethod then busybox and s/then/than/ -- Pengutronix e.K. | Uwe Kleine-König| Industrial Linux Solutions | http://www.pengutronix.de/ | ___ ptxdist mailing list ptxdist@pengutronix.de
[ptxdist] [PATCH v2] minicoredumper: New package
Signed-off-by: Uwe Kleine-König --- Changes since (implicit) v1, sent with Message-Id: 20180813145748.21019-1-u.kleine-koe...@pengutronix.de: - s/LGPG-2.1/LGPG-2.1/ - Add "AND" to MINICOREDUMPER_LICENSE - Mention missing automatic registration as coredump helper in .in instead of the commit log - drop complete extract stage. The ptxdist defaults are fine enough. Most of this was noticed by Roland, thanks Best regards Uwe rules/minicoredumper.in | 21 rules/minicoredumper.make | 69 +++ 2 files changed, 90 insertions(+) create mode 100644 rules/minicoredumper.in create mode 100644 rules/minicoredumper.make diff --git a/rules/minicoredumper.in b/rules/minicoredumper.in new file mode 100644 index ..969be8210c88 --- /dev/null +++ b/rules/minicoredumper.in @@ -0,0 +1,21 @@ +## SECTION=shell_and_console + +config MINICOREDUMPER + bool + prompt "minicoredumper" + select LIBELF + select JSON_C + help + The minicoredumper project provides a program that handles the + creation of core dump files on Linux. It can produce much smaller + core dump files by making use of sparse files, compression, and + allowing the user to configure what parts of the process memory image + should be dumped. + + Note that there are currently no mechnisms integrated to + automatically register minicoredumper as kernel coredump helper. So + you might want to do + + echo '|/usr/sbin/minicoredumper %P %u %g %s %t %h %e' > /proc/sys/kernel/core_pattern + + by hand to actually make use of it. diff --git a/rules/minicoredumper.make b/rules/minicoredumper.make new file mode 100644 index ..948068b21c89 --- /dev/null +++ b/rules/minicoredumper.make @@ -0,0 +1,69 @@ +# -*-makefile-*- +# +# Copyright (C) 2018 by Uwe Kleine-Koenig +# +# See CREDITS for details about who has contributed to this project. +# +# For further information about the PTXdist project and license conditions +# see the README file. +# + +# +# We provide this package +# +PACKAGES-$(PTXCONF_MINICOREDUMPER) += minicoredumper + +# +# Paths and names +# +MINICOREDUMPER_VERSION := 2.0.1 +MINICOREDUMPER_MD5 := 813b864e0c6a833d14bab244723de6a4 +MINICOREDUMPER := minicoredumper-$(MINICOREDUMPER_VERSION) +MINICOREDUMPER_SUFFIX := tar.xz +MINICOREDUMPER_URL := https://linutronix.de/minicoredumper/files/$(MINICOREDUMPER).$(MINICOREDUMPER_SUFFIX) +MINICOREDUMPER_SOURCE := $(SRCDIR)/$(MINICOREDUMPER).$(MINICOREDUMPER_SUFFIX) +MINICOREDUMPER_DIR := $(BUILDDIR)/$(MINICOREDUMPER) +MINICOREDUMPER_LICENSE := BSD AND LGPL-2.1 +MINICOREDUMPER_LICENSE_FILES := file://COPYING;md5=71827c617ec7b45a0dd23658347cc1e9 + +# +# Prepare +# + +MINICOREDUMPER_CONF_TOOL := autoconf +MINICOREDUMPER_CONF_OPT := \ + $(CROSS_AUTOCONF_USR) \ + --disable-silent-rules \ + --without-coreinject \ + --with-minicoredumper \ + --without-minicoredumper_regd \ + --without-minicoredumper_trigger \ + --without-libminicoredumper \ + --without-minicoredumper_demo \ + --without-werror \ + +# +# Target-Install +# + +$(STATEDIR)/minicoredumper.targetinstall: + @$(call targetinfo) + + @$(call install_init, minicoredumper) + @$(call install_fixup, minicoredumper, PRIORITY, optional) + @$(call install_fixup, minicoredumper, SECTION, base) + @$(call install_fixup, minicoredumper, AUTHOR, "Uwe Kleine-Koenig ") + @$(call install_fixup, minicoredumper, DESCRIPTION, missing) + + @$(call install_copy, minicoredumper, 0, 0, 0755, -, /usr/sbin/minicoredumper) + + @$(call install_alternative_tree, minicoredumper, 0, 0, /etc/minicoredumper) + + @$(call install_copy, minicoredumper, 0, 0, 0755, /var/crash) + @$(call install_copy, minicoredumper, 0, 0, 0755, /var/crash/minicoredumper) + + @$(call install_finish, minicoredumper) + + @$(call touch) + +# vim: syntax=make -- 2.18.0 ___ ptxdist mailing list ptxdist@pengutronix.de