;s no
additional stuff like init script or systemd service file for the broker
yet.
Signed-off-by: Alexander Dahl
---
rules/mosquitto.in | 36 +
rules/mosquitto.make | 88
2 files changed, 124 insertions(+)
create mode 1
20 booting from nand flash and it boots successfully.
Greets
Alex
Alexander Dahl (3):
at91bootstrap: Add mirror URL for legacy bootstrap
at91bootstrap: Add patch to work with EABI toolchains
at91bootstrap: Add wrapper blacklist
.../0001-crt0_gnu.S-fix-image-size.patch | 10 +-
...
This was the only working mirror URL I could find, all other sources
seem to be gone.
Signed-off-by: Alexander Dahl
---
rules/at91bootstrap.make | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/rules/at91bootstrap.make b/rules/at91bootstrap.make
index 5402c4b94d
This is basically the same as for at91bootstrap2 recently, however this
one has its own compile stage, so some additional lines are necessary
which I borrowed from the barebox package.
Signed-off-by: Alexander Dahl
---
rules/at91bootstrap.make | 11 ++-
1 file changed, 10 insertions
introducing the patch over there.
Also recreated whole patch series.
Signed-off-by: Alexander Dahl
---
.../0001-crt0_gnu.S-fix-image-size.patch | 10 +-
...32-littlearm.lds-fix-overlap-linker-error.patch | 6 +-
.../0003-Fix-build-with-EABI-toolchain.patch | 338
Hei hei,
On Tue, Dec 05, 2017 at 11:18:54PM +0100, Roland Hieber wrote:
> I'm not quite sure if this is the right place ("The content on
> public.pengutronix.de is mostly temporary, so it might vanish into thin
> air any time"), but the original link is now dead and this one works.
> Otherwise I d
Hei hei,
On Tue, Jan 16, 2018 at 12:54:05PM +0100, Fritz Emboli wrote:
> i'm trying to remote debug an application.
> Host is Ubuntu 16.04 with gdb-multiarch (7.11.1)
> Client is an embedded ARM device running gdbserver (7.6.1 from ptxdist).
Did you try the gdb coming with your toolchain? You can
Hei hei,
On Tue, Jan 16, 2018 at 02:33:29PM +0100, Fritz Emboli wrote:
> I tried the debugger from the toolchain.
> But it dies, because it cannot find libncurses.so.5.
What about installing this lib?
> It is OSELAS.Toolchain-2013.12.2. I don't know where
> it comes from (inherited, not self com
Hei hei,
On Fri, Jan 26, 2018 at 08:58:52AM +0100, Guillermo Rodriguez Garcia wrote:
> I am creating a ptxdist BSP and have a series. patchset for a
> given kernel version. I have noticed that this patchset is ignored when the
> KERNEL_HEADER package is built -- instead, a "series" file is automat
u-boot had the same problem as the other bootloaders (barebox,
at91bootstrap, …): much too big images were generated. Use the same
wrapper blacklist approach to prevent this like in barebox package.
Signed-off-by: Alexander Dahl
---
Out of curiosity: if I wanted to find out which of those
Hello Michael,
thanks for your explanations, I comment below.
Am Mittwoch, 31. Januar 2018, 16:09:40 CET schrieb Michael Olbrich:
> Build/runtime failures or by knowing what the options mean :-).
>
> TARGET_HARDEN_RELRO/TARGET_HARDEN_BINDNOW only makes sense for elf
> binaries.
> TARGET_HARDEN_P
be fixed by just creating the empty folder.
Note: users may have overridden the path to the state file by passing
another path together with the -s option of logrotate.
Fixes: 878707bd943a04fc8e3a55a151499359e81ce780
Signed-off-by: Alexander Dahl
---
Meanwhile upstream moved to https://github.com
t;,
although htop is running as root. :-/
So while I can not say anything about the added patch, the tool itself
in version 2.1.0 is running fine, so:
Tested-by: Alexander Dahl
Greets
Alex
--
»With the first link, the chain is forged. The first speech censured,
the first thought forbidd
Hei hei,
On Thu, Feb 08, 2018 at 10:17:31AM +0100, Marc Kleine-Budde wrote:
> Have you enabled CONFIG_TASK_XACCT and CONFIG_TASK_IO_ACCOUNTING?
Didn't have that before. I get some values now, although I'm not sure
how to interpret those. I should probably look in the htop
documentation again. But
Hello Michael,
On Thu, Feb 08, 2018 at 02:53:18PM +, Alm, Michael wrote:
> What is best-practice for the LICENSE info in the rule file in that case?
> Would it be possible to exclude these packages from the license report?
Did you try setting those packages to ignore? Some ptxdist packages
li
Building v2018.02.0 of the toolchain failed on first try, maybe because
some mirrors were not yet synced with the new glibc version.
Signed-off-by: Alexander Dahl
---
rules/glibc.make | 1 +
1 file changed, 1 insertion(+)
diff --git a/rules/glibc.make b/rules/glibc.make
index 9d55312..7f62e62
This is a patch for oselas toolchain. Should have mentioned it in the first
place. O:-)
Greets
Alex
Am Dienstag, 13. Februar 2018, 09:29:42 CET schrieb Alexander Dahl:
> Building v2018.02.0 of the toolchain failed on first try, maybe because
> some mirrors were not yet synced with the new
Hei hei,
I had several problems compiling the new toolchain, the main one like in the
first example, see below.
Am Montag, 12. Februar 2018, 16:51:54 CET schrieb Michael Olbrich:
> Download the release here:
> https://public.pengutronix.de/oselas/toolchain/
I didn't download a prebuilt release,
Hei hei,
Am Dienstag, 13. Februar 2018, 14:30:38 CET schrieb Michael Olbrich:
> > Host is on Debian 9 (stretch), amd64, using icecc:
^^
> Hmm, generating Debian packages is not widely tested outside my local
> setup, but it should work. Can you add 'set -x' at the beginning of the
> main() funct
Hei hei,
On Sun, Feb 18, 2018 at 03:00:22PM -0500, Jon Ringle wrote:
> I built the arm-v5te toolchain, and it built:
> oselas.toolchain-2018.02.0-arm-v5te-linux-gnueabi-gcc-7.3.1-glibc-2.27-binutils-2.30-kernel-4.15-sanitized_2018.02.0-0ptx1~bpo14.04+1_amd64.deb
So you probably used Ubuntu 14.04?
When globally enabling -Wcast-align, which is useful on older arm
targets, libelf compile fails due to -Werror set.
Signed-off-by: Alexander Dahl
---
patches/elfutils-0.170/0001-Remove-Werror.patch | 23 +++
patches/elfutils-0.170/autogen.sh | 1 +
patches
Hei hei,
what about this patch? Did it get lost?
Greets
Alex
Am Freitag, 23. Februar 2018, 09:24:17 CET schrieb Alexander Dahl:
> When globally enabling -Wcast-align, which is useful on older arm
> targets, libelf compile fails due to -Werror set.
>
> Signed-off-by: Al
Signed-off-by: Alexander Dahl
---
rules/util-linux-ng.in | 7 +++
rules/util-linux-ng.make | 3 +++
2 files changed, 10 insertions(+)
diff --git a/rules/util-linux-ng.in b/rules/util-linux-ng.in
index c6b2e89899..8e89370f68 100644
--- a/rules/util-linux-ng.in
+++ b/rules/util-linux-ng.in
The previous sed regex for creating the basename for an intermediate
folder in the download URL failed on rc tarballs. Now URLs like this
work, too:
https://www.kernel.org/pub/linux/utils/util-linux/v2.32/util-linux-2.32-rc1.tar.gz
Signed-off-by: Alexander Dahl
---
rules/util-linux-ng.make | 2
Disable option for compilation with Address Sanitizer.
Signed-off-by: Alexander Dahl
---
rules/util-linux-ng.make | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/rules/util-linux-ng.make b/rules/util-linux-ng.make
index ed0f85c02e..0961c4c657 100644
--- a/rules/util
, but I'm not sure about all things configure_helper.py
complained about. So I marked the whole patch series as RFC.
Greets
Alex
[1]
https://mirrors.edge.kernel.org/pub/linux/utils/util-linux/v2.32/v2.32-ReleaseNotes
Alexander Dahl (3):
util-linux-ng: Add lscpu to installable tools
util-
Hello,
from the Git repository and some hints in the ptxdist source code I assume
ptxdist can generate some kind of license report, based on the FOO_LICENSE
variables in each packages make rule. However I could not find any hint on how
to generate this report in:
* documentation (ptxdist docs-
Hei hei,
the are old patches for v2.31.1 which I forgot to check. :-/
Just to let you know, this should not be applied as is.
Greets
Alex
Am Donnerstag, 22. März 2018, 09:18:38 CET schrieb Alexander Dahl:
> Disable option for compilation with Address Sanitizer.
>
> Signed-off-by:
Signed-off-by: Alexander Dahl
---
rules/lsuio.make | 16 +---
1 file changed, 9 insertions(+), 7 deletions(-)
diff --git a/rules/lsuio.make b/rules/lsuio.make
index f6d4aa3f80..f8bd970db9 100644
--- a/rules/lsuio.make
+++ b/rules/lsuio.make
@@ -16,13 +16,15 @@ PACKAGES
Disable option for compilation with Address Sanitizer. Moved patch
series.
Signed-off-by: Alexander Dahl
---
v1 -> v2: also move patch series
---
.../0001-__progname-redefinition-fix.patch | 2 +-
patches/{util-linux-2.31.1 => util-linux-2.32}/series
Hello Roland,
Am Montag, 26. März 2018, 10:04:03 CEST schrieb Roland Hieber:
> On 26.03.2018 08:21, Alexander Dahl wrote:
> > Disable option for compilation with Address Sanitizer.
>
> Hmm, why did you disable it? Seems a rather useful feature to me to have
> by default :)
Tw
Hello,
On Tue, Mar 27, 2018 at 09:43:35AM +0200, Michael Olbrich wrote:
> I want to move this stuff out of rules/post/ and do something similar to
> the image rules. And then hook up these new packages these packages
> somewhere. I'm not sure 'world' is the correct place, so a new command
> 'ptxdi
Hello,
On Tue, Mar 27, 2018 at 09:01:58AM +0200, Michael Olbrich wrote:
> > +LSUIO_VERSION := 0.2.0
> > +LSUIO_MD5 := 8da82b681c24e6c5b3518c99e7f961d7
> > +LSUIO := lsuio-$(LSUIO_VERSION)
> > +LSUIO_SUFFIX := tar.gz
> > +LSUIO_URL :
Signed-off-by: Alexander Dahl
---
Notes:
v1 -> v2: Fix whitespace
rules/lsuio.make | 2 ++
1 file changed, 2 insertions(+)
diff --git a/rules/lsuio.make b/rules/lsuio.make
index f6d4aa3f80..91678ea4fb 100644
--- a/rules/lsuio.make
+++ b/rules/lsuio.make
@@ -23,6 +23,8 @@ LSUIO_SUF
Tested with OSELAS.Toolchain-2014.12.3/arm-v5te-linux-gnueabi and
ptxdist 2017.09.0.
Signed-off-by: Alexander Dahl
---
.../jq-1.5/0001-Support-without-oniguruma.patch| 60 +++
patches/jq-1.5/autogen.sh | 1 +
patches/jq-1.5/series
ain-2014.12.3/arm-v5te-linux-gnueabi and
ptxdist 2017.09.0 on an at91sam9g20 based board.
Greets
Alex
Alexander Dahl (2):
libnftnl: Upgrade from 1.0.6 to 1.0.9
nftables: Upgrade from v0.6 to v0.8.3
...on-t-enable-xtables-when-without-xtables-.patch | 32 ++
patches/nftab
/2017/000224.html
* 1.0.9:
https://lists.netfilter.org/pipermail/netfilter-announce/2018/000226.html
Signed-off-by: Alexander Dahl
---
rules/libnftnl.in | 6 --
rules/libnftnl.make | 5 ++---
2 files changed, 2 insertions(+), 9 deletions(-)
diff --git a/rules/libnftnl.in b/rules/libnftnl.in
://lists.netfilter.org/pipermail/netfilter-announce/2018/000227.html
* v0.8.2:
http://lists.netfilter.org/pipermail/netfilter-announce/2018/000229.html
* v0.8.3: https://marc.info/?l=netfilter&m=152009279821556
Signed-off-by: Alexander Dahl
---
...on-t-enable-xtables-when-without-xtables-.patch
/2017/000224.html
* 1.0.9:
https://lists.netfilter.org/pipermail/netfilter-announce/2018/000226.html
Signed-off-by: Alexander Dahl
---
rules/libnftnl.in | 6 --
rules/libnftnl.make | 5 ++---
2 files changed, 2 insertions(+), 9 deletions(-)
diff --git a/rules/libnftnl.in b/rules/libnftnl.in
This adds only empty chains, but it will be needed for the bbinit
startscript not to fail. User will probably overwrite it with something
more useful.
Signed-off-by: Alexander Dahl
---
projectroot/etc/nftables.conf | 15 +++
rules/nftables.make | 1 +
2 files changed, 16
://lists.netfilter.org/pipermail/netfilter-announce/2018/000227.html
* v0.8.2:
http://lists.netfilter.org/pipermail/netfilter-announce/2018/000229.html
* v0.8.3: https://marc.info/?l=netfilter&m=152009279821556
Signed-off-by: Alexander Dahl
---
...on-t-enable-xtables-when-without-xtables-.patch
t; v2:
* Added three additional patches not included in v1 of the series
Alexander Dahl (5):
libnftnl: Upgrade from 1.0.6 to 1.0.9
nftables: Upgrade from v0.6 to v0.8.3
nftables: Fix Vim modelines
nftables: Add example config file
nftables: Add bbinit startscript
...on-t-enable-xtab
* Add line in kconfig rule
* Fix line in make rule
Signed-off-by: Alexander Dahl
---
rules/nftables.in | 3 ++-
rules/nftables.make | 2 +-
2 files changed, 3 insertions(+), 2 deletions(-)
diff --git a/rules/nftables.in b/rules/nftables.in
index 0377e731c1..be7fca1fa0 100644
--- a/rules
For loading your packet filter rules on system startup.
Signed-off-by: Alexander Dahl
---
projectroot/etc/init.d/nftables | 103
rules/nftables-bbinit.in| 9
rules/nftables.in | 4 ++
rules/nftables.make | 11
Hello,
Am Freitag, 6. April 2018, 11:07:27 CEST schrieb Alexander Dahl:
> Support for the new option --with-xtables is not included. The
> additional patch is necessary to successfully call configure with
> --without-xtables for a setup without iptables. (Patch is currently
>
Hei hei,
On Mon, Apr 09, 2018 at 10:30:56AM +0200, Michael Olbrich wrote:
> On Fri, Apr 06, 2018 at 11:07:30AM +0200, Alexander Dahl wrote:
> > For loading your packet filter rules on system startup.
> >
> > Signed-off-by: Alexander Dahl
>
> This patch does
es from the mails by myself on two systems and got no
error. Strange.
> Please resent. Maybe as an attachment.
I attach it to this mail. You may need to apply other patches of this series
first?
Greets
Alex
>From 29a2513f241e3be9428d20756ac2929c764b4115 Mon Sep 17 00:00:00 2001
From: Al
Hei hei,
Am Dienstag, 10. April 2018, 10:47:32 CEST schrieb Michael Olbrich:
> No. If we change the modelines, then we do this for all files. I don't
> really care about this. Feel free to start a discussion to figure out the
> 'correct' modelines. I see several different ones right now. Then we c
Hei hei,
Am Dienstag, 10. April 2018, 17:34:09 CEST schrieb Bastian Stender:
> There are patches on top of htop 2.1.0. You should port them or drop
> them as needed.
FWIW …
0001-linux-LinuxProcessList-fix-reading-of-number-of-read.patch was applied
upstream.
0002-configure-check-for-ncurses-w
Hello Jon,
On Tue, Jun 05, 2018 at 01:18:06PM -0400, j...@ringle.org wrote:
> rules/libelf.make | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
There's a patch in patches/elfutils-0.170 you didn't cover. ;-)
Greets
Alex
--
/"\ ASCII RIBBON | »With the first link, the chain is forg
Hei hei,
On Tue, Jun 05, 2018 at 02:20:37PM -0400, Jon Ringle wrote:
> The bump to 0.171 fixed the errors I was getting when building 0.170, so I
> think the patch is not needed
I already saw your v2 also removing the patch.
The reason Clemens introduced the patch was compilation failed with
GCC
Hello Jakov
On Mon, Jun 11, 2018 at 07:34:04AM +, Jakov Simunic wrote:
> I am confused about rules/libblkid.in, there is no corresponding
> rules/libblkid.make file, I use this to build udev, but it fails
> this way when i do the following:
>
> ./ptxdist clean
> ./ptxdist compile udev
>
> It
Hei hei,
Am Montag, 11. Juni 2018, 09:29:01 CEST schrieb Jakov Simunic:
> Thanks for replying so fast.
> Could you give me a brief explanation of virtual packages (or a link)?
I'm sorry, I can't. Virtual packages seem to be missing in the documentation.
:-/
However, if you are new to ptxdist yo
t optional, if you don't like it.
Greets
Alex
Alexander Dahl (2):
coreutils: Fix missing busybox guard comments
coreutils: Unify case in comments
rules/coreutils.in | 35 ++-
1 file changed, 22 insertions(+), 13 deletions
Signed-off-by: Alexander Dahl
---
rules/coreutils.in | 26 +-
1 file changed, 13 insertions(+), 13 deletions(-)
diff --git a/rules/coreutils.in b/rules/coreutils.in
index 73cd094b0..19a93758f 100644
--- a/rules/coreutils.in
+++ b/rules/coreutils.in
@@ -27,7 +27,7
Signed-off-by: Alexander Dahl
---
rules/coreutils.in | 13 +++--
1 file changed, 11 insertions(+), 2 deletions(-)
diff --git a/rules/coreutils.in b/rules/coreutils.in
index 75d897440..73cd094b0 100644
--- a/rules/coreutils.in
+++ b/rules/coreutils.in
@@ -327,6 +327,9 @@ config
Hei hei,
Am Freitag, 22. Juni 2018, 09:15:16 CEST schrieb Michael Olbrich:
> That's an upstream patch, right? It should come before the other patch in
> the 'upstream' section.
I checked yesterday, it's not in libxml2 master, yet. See the libxml2
bugtracker for details: https://bugzilla.gnome.or
Hei hei,
Am Freitag, 22. Juni 2018, 08:12:10 CEST schrieb Denis OSTERLAND:
> +#
> +# autoconf
> +#
> +YAJL_CONF_TOOL := cmake
Comment does not match assignment. ;-)
Greets
Alex
___
ptxdist mailing list
ptxdist@pengutronix.de
Hello Denis,
> > > +#
> > > +# autoconf
> > > +#
> > > +YAJL_CONF_TOOL := cmake
> >
> > Comment does not match assignment. ;-)
>
> You are guess configtool as comment would be better.
From the 'src-cmake-prog' template you would get:
#
Hello Denis,
Am Freitag, 22. Juni 2018, 09:07:58 CEST schrieb Denis OSTERLAND:
> So I just used the wrong template.
I would not say so. You wanted a target package with external sources and you
picked the right template for that. I guess it just makes no sense to have
more target templates for
Hei hei,
Am Montag, 25. Juni 2018, 22:04:47 CEST schrieb Uwe Kleine-König:
> On Mon, Jun 25, 2018 at 09:28:10AM -0500, Crim, Jason wrote:
> > Is there good documentation that someone can point me to for generating
> > a patch and updating the series file?
>
> If there is already a patch stack, ju
version.
Let me know what you think.
Alex
Alexander Dahl (6):
timezone: Remove menu entry 'Mideast'
timezone: Cleanup timezone file removal
timezone: Order entries
timezone: Add missing zoneinfo entries
tz: version bump 2017c -> 2018e
Revert "host-tz-database: use snprin
This reverts commit ee06ca66eb191271c937e9b0325a85a3c19dafc8.
Quote from the 2018a NEWS:
> Builders no longer need to specify -DHAVE_SNPRINTF on platforms that
> have snprintf and use pre-C99 compilers. (Problem reported by Jon
> Skeet.)
Cc: Roland Hieber
Signed-off-by: Alexa
The targetinstall of the file was removed ptxdist 2017.10.0, but not the
file itself. Also the help text in menu was not updated.
Fixes: 3ee522b06e41b3b4da9490b06859a62a23fca18e
Signed-off-by: Alexander Dahl
---
projectroot/etc/timezone | 1 -
rules/timezone.in| 6 ++
2 files
set and use TZDIR and that should still be
sufficient for the ptxdist use case.
* Since 2018 parallelized builds with -j should be safe.
Signed-off-by: Alexander Dahl
---
rules/host-tz-database.make | 8
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/rules/host-tz
This will make it easier in the future to check for updates and
add/remove updated entries.
Signed-off-by: Alexander Dahl
---
rules/timezone.in | 216 ++--
rules/timezone.make | 76 +-
2 files changed, 146 insertions(+), 146
In ptxdist 2016.08.0 only the line in timezone.make causing trouble was
removed.
Fixes: 76060f6d5f11009fbb964acfa1c504e6b4df376a
Signed-off-by: Alexander Dahl
---
rules/timezone.in | 4
1 file changed, 4 deletions(-)
diff --git a/rules/timezone.in b/rules/timezone.in
index 941e56127
d approach.
Note: if the user selects e.g. 'Poland' and 'Europe/Warsaw' that would
waste space on the target, because the same file is copied twice without
symbolic or hard link.
Signed-off-by: Alexander Dahl
---
rules/timezone.in | 12
rules/timezone.make
Hei hei,
Am Dienstag, 26. Juni 2018, 15:50:53 CEST schrieb Roland Hieber:
> On Tue, Jun 26, 2018 at 08:56:48AM +0200, Alexander Dahl wrote:
> > config TIMEZONE_LOCALTIME
> >
> > - prompt "Timezone for /etc/localtime and /etc/timezone"
> > +
Hello,
Am Dienstag, 26. Juni 2018, 16:01:03 CEST schrieb Alexander Dahl:
> You never know where the device ends up, and I think UTC is the only
> reasonable default. Should we make installation of the UTC zoneinfo file
> mandatory then?
Wait, this can be set to an empty string (see
Hei hei,
Am Dienstag, 19. Juni 2018, 15:51:09 CEST schrieb Alexander Dahl:
> I just stumbled over a vanishing menu entry without comment for the
> activated busybox tool 'readlink' so I thought I check on more missing
> comments. One was actually even in the wrong place.
Maybe
Hei hei,
Am Dienstag, 24. Juli 2018, 13:46:47 CEST schrieb Jakov Simunic:
> 2. Is systemd really a good option for embedded systems?
You don't need to use it. We still use busybox-init with plain old init
scripts and monit for monitoring.
> 4. If I have config_devtmpfs and config_devtmpfs
This includes a lot of fixes and new features, including a fix for
CVE-2016-7067 in monit 5.20.0.
See https://mmonit.com/monit/changes/ for complete release notes.
Signed-off-by: Alexander Dahl
---
rules/monit.in | 5 +
rules/monit.make | 6 --
2 files changed, 9 insertions(+), 2
Signed-off-by: Alexander Dahl
---
rules/monit.make | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/rules/monit.make b/rules/monit.make
index a72c8e77d..99266622d 100644
--- a/rules/monit.make
+++ b/rules/monit.make
@@ -1,6 +1,6 @@
# -*-makefile-*-
#
-# Copyright (C
Hei hei,
since 5.17.1 from March 2016 there were some new releases for monit,
including bug fixes, security fixes, build fixes, and new features.
This series does a little cleanup on the old version and then
upgrades to 5.25.2, introducing one new option.
Greets
Alex
Alexander Dahl (3):
monit
Signed-off-by: Alexander Dahl
---
rules/monit.make | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/rules/monit.make b/rules/monit.make
index 2870e8056..a72c8e77d 100644
--- a/rules/monit.make
+++ b/rules/monit.make
@@ -35,9 +35,10 @@ MONIT_CONF_ENV := $(CROSS_ENV
Hello Roland,
Am Freitag, 27. Juli 2018, 11:42:42 CEST schrieb Roland Hieber:
> On Fri, Jul 27, 2018 at 10:59:13AM +0200, Alexander Dahl wrote:
> > + $(GLOBAL_LARGE_FILE_OPTION) \
> >
> > --enable-optimized \
> > --disable-profiling \
> >
&
Hello Uwe,
found some minor issues, see below. Despite of that:
Tested-by: Alexander Dahl
Greets
Alex
Am Dienstag, 14. August 2018, 18:15:39 CEST schrieb Uwe Kleine-König:
> Signed-off-by: Uwe Kleine-König
> ---
> Changes since (implicit) v1, sent with Message-Id:
> 2018081314
.??.?-?-gcb7418a2299c-dirty when the last tag in the BSP was only a
lightweight one, no annotated tag.
Signed-off-by: Alexander Dahl
---
scripts/kernel/setlocalversion | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/scripts/kernel/setlocalversion b/scripts/kernel/setlocalversion
Hello,
Am Freitag, 7. September 2018, 12:01:05 CEST schrieb Michael Olbrich:
> On Thu, Sep 06, 2018 at 10:39:27AM +0200, Alexander Dahl wrote:
> > This fixes the case for PTXDIST_BSP_AUTOVERSION returning things like
> > .??.?-?-gcb7418a2299c-dirty when the last tag in the
on so far is: the default install stage installs to
$(FOO_PKGDIR) and that content is magically copied to $(SYSROOT) then?
I tried to find some help in the docs, there's not much, but I spotted
one thing I consider to be wrong at this time. See the patch.
Alexander Dahl (1):
doc: Fix wr
matching subfolder for the
package's targetinstall, $(FOO_PKGDIR) expands to that.
Signed-off-by: Alexander Dahl
---
doc/ref_manual.rst | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/doc/ref_manual.rst b/doc/ref_manual.rst
index 56034d8e7..c9e26ffab 100644
---
http://dfu-util.gnumonks.org/ resets connections and according to the
buildroot project is down for years. Also dfu-util is not listed as
project on http://gnumonks.org/ anymore.
http://dfu-util.sourceforge.net/ however is alive and even has more
recent downloads.
Signed-off-by: Alexander Dahl
Hei hei,
I needed to use dfu-util today and realized the current package got
its last change in 2014. Because the old download URL fails now, I
took the opportunity for a version bump. Tested with some NXP radio
modules connected to USB.
Greets
Alex
Alexander Dahl (2):
dfu-util: Fix download
Bugfixes and new features from over 4 years of development.
Signed-off-by: Alexander Dahl
---
rules/dfu-util.make | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/rules/dfu-util.make b/rules/dfu-util.make
index d9e7285c6..c7cfa3d23 100644
--- a/rules/dfu-util.make
+++ b
packages, bugfixes and version bumps.
>
> Thanks to all contributors and - as always - the shortlog below.
>
> Enjoy,
> Michael Olbrich
>
>
> Alexander Dahl (1):
> doc: Fix wrong use of PKGDIR
>
> Artur Wiebe (5):
> python3-lxml: new packag
Hello,
Am Donnerstag, 13. September 2018, 09:22:57 CEST schrieb Alexander Dahl:
> I'm currently writing package rules for some ugly Make only based
> sources. I could sort out my questions about how to use $(SYSROOT) and
> $(FOO_PKGDIR) in compile and install stages by looking at ot
Hello Michael,
Am Freitag, 5. Oktober 2018, 16:21:40 CEST schrieb Michael Olbrich:
> Note that the changes for the layer support where rather invasive. I've
> tested this here intensively, but it's possible that I missed some
> use-cases. Please test this release carefully and report any problems
Hei hei,
Am Montag, 8. Oktober 2018, 09:44:12 CEST schrieb Alexander Dahl:
> Am Freitag, 5. Oktober 2018, 16:21:40 CEST schrieb Michael Olbrich:
> > Note that the changes for the layer support where rather invasive. I've
> > tested this here intensively, but it's po
Hello Michael,
Am Dienstag, 9. Oktober 2018, 17:12:26 CEST schrieb Michael Olbrich:
> On Tue, Oct 09, 2018 at 12:55:03PM +0200, Michael Olbrich wrote:
> > I can reproduce this. I broke this with the opkg-utils update. I'm working
> > on a fix.
>
> I've pushed some stuff to master. Can you verify
Hei hei,
while I could build current ptxdist-2018.10.0 based DistroKit on one
host, I have problems building the very same BSP (DistroKit @
DistroKit-2018.03.0-30-g78f0dbf) on another, quite similar host
(Debian GNU/Linux 9 (stretch), amd64). There seems to be a problem
with the host-coreutils pac
Hei hei,
On Wed, Oct 10, 2018 at 02:21:48PM +0200, Roland Hieber wrote:
> On Wed, Oct 10, 2018 at 01:04:01PM +0200, Roland Hieber wrote:
> > Ohai,
> >
> > I set my source download folder to ./src/ and I can reproduce this with
> > various other packages as well (kernel, coreutils, systemd,
> > ne
Hei hei,
Am Mittwoch, 10. Oktober 2018, 14:03:32 CEST schrieb Ahmad Fatoum:
> Signed-off-by: Ahmad Fatoum
> ---
> rules/u-boot.make | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/rules/u-boot.make b/rules/u-boot.make
> index 94aa7745809d..aa7a01ca8fea 100644
> --- a/rules/u-boot.make
MPILE)
> HOSTCC=$(HOSTCC) +U_BOOT_CONF_OPT := \
> + V=$(PTXDIST_VERBOSE) \
> + CROSS_COMPILE=$(BOOTLOADER_CROSS_COMPILE) \
> + HOSTCC=$(HOSTCC)
> U_BOOT_MAKE_PAR := NO
> U_BOOT_TAGS_OPT := ctags cscope etags
> U_BOOT_MAKE_OPT :
Hei hei,
Am Dienstag, 16. Oktober 2018, 09:52:55 CEST schrieb Michael Olbrich:
> On Wed, Oct 10, 2018 at 02:03:33PM +0200, Ahmad Fatoum wrote:
> > while at it, remove U_BOOT_PATH, because it's PATH=$(CROSS_PATH) already
> > by default.
> >
> > Signed-off-by: Ahmad Fatoum
> > ---
> >
> > rules/
Hello,
Am Mittwoch, 10. Oktober 2018, 14:03:37 CEST schrieb Ahmad Fatoum:
> U-Boot can be configured at compile-time to expect the environment at a
> set offset of e.g. the MMC.
This does not describe what the patch adds. You generate a binary image and
extend ptxdist to do that at compile time.
Hei hei,
Am Dienstag, 16. Oktober 2018, 10:58:50 CEST schrieb Denis OSTERLAND:
> Am Mittwoch, den 10.10.2018, 14:03 +0200 schrieb Ahmad Fatoum:
> > +ifdef PTXCONF_U_BOOT_ENV
> > + @$(U_BOOT_DIR)/tools/mkenvimage \
> > + $(U_BOOT_ENV_FLAGS) -s $(PTXCONF_U_BOOT_ENV_IMAGE_SIZE) \
> > +
Hello,
Am Dienstag, 16. Oktober 2018, 10:51:04 CEST schrieb Michael Olbrich:
> On Fri, Oct 12, 2018 at 07:47:35AM +, Baeuerle, Florian wrote:
> > Is there also a mechanism in ptxdist that allows leaving out system
> > packages if the provided tools are already present?
>
> No, and we prefer
Hello,
Am Dienstag, 16. Oktober 2018, 14:26:10 CEST schrieb Michael Olbrich:
> On Tue, Oct 16, 2018 at 01:27:13PM +0200, Alexander Dahl wrote:
> > Am Dienstag, 16. Oktober 2018, 10:58:50 CEST schrieb Denis OSTERLAND:
> > > Am Mittwoch, den 10.10.2018, 14:03 +0200 sc
Hello Florian,
On Tue, Oct 16, 2018 at 10:57:20AM +, Baeuerle, Florian wrote:
> SOURCE_DATE_EPOCH uses the ptxdist release number for the timestamp, thus the
> timestamp changes with every ptxdist release. I'd prefer to *really* get the
> same binary when the source code was not changed at all
301 - 400 of 1102 matches
Mail list logo
