Re: [oe] [PATCH] redis: fix hiredis build so it's less failure-prone and links properly

True,, but it still needs the ranlib hunk; do I really need to redo it, or can we take it as-is? Steve On Wed, Jun 22, 2016 at 9:56 AM, Khem Raj <raj.k...@gmail.com> wrote: > >> On Jun 7, 2016, at 5:14 PM, Stephen Arnold <sarn...@vctlabs.com> wrote: >> >> This p

[oe] [PATCH] redis: fix hiredis build so it's less failure-prone and links properly

This patch resets the default OPT flags to -O2 and adds the missing ranlib command to index the static library. Signed-off-by: Stephen Arnold <sarn...@vctlabs.com> --- ...-update-Makefile-to-add-symbols-to-staticlib.patch | 19 +++ meta-oe/recipes-extended/redis/redis_3.

Re: [oe] [meta-networking][PATCH v4] dhcpcd: add recipe for dhcpcd

Sorry, in answer to previous question I'm fine with Roy's dhcp client package (if I still need wifi-radar I'll port the stupid thing to work with it). Steve On Thu, Jan 28, 2016 at 5:21 PM, Karstens, Nate wrote: > Martin, > > It looks like the recipe that was merged in

[oe] [meta-networking][PATCH] net-snmp: fix cross-compile breakage, add libnl pkgconfig option

From: Stephen Arnold <stephen.arnol...@gmail.com> This patch enhances a previously unapplied patch on jethro, plus adds some flexibility in terms of required deps and a few cleanups. Signed-off-by: Stephen Arnold <stephen.arnol...@gmail.com> --- meta-networking/recipes-protocols/

Re: [oe] [meta-oe][PATCH] dhcpcd: add recipe for dhcpcd

On Tue, 5 Jan 2016 13:37:35 + "Karstens, Nate" wrote: Me too, I sent this earlier from a non-subscribed address... Steve > Sorry for the delay. I contacted the author and he pointed me to > the following page: > >

Re: [oe] [meta-oe][PATCH] dhcpcd: add recipe for dhcpcd

ven't seen one yet with a bitbake-layers add-layer step to add networking... Steve On Tue, Jan 5, 2016 at 2:04 PM, Trevor Woerner <twoer...@gmail.com> wrote: > On 01/05/16 16:55, Stephen Arnold wrote: >> My vote would be whatever is the most stable/reliable should be in >&

Re: [oe] [PATCH][meta-fsl-arm] 2.6.35 kernel build error with perl 5.22

it in either case above. Steve On Mon, Aug 17, 2015 at 2:59 AM, Otavio Salvador otavio.salva...@ossystems.com.br wrote: Hello Stephen, (Adding meta-freescale mailing list as it is the proper place for this discussion) On Sat, Aug 15, 2015 at 3:05 AM, Stephen Arnold stephen.arnol

[oe] [PATCH][meta-fsl-arm] 2.6.35 kernel build error with perl 5.22

When building fsl-image-machine-test for imx28-evk with n00b defaults using the fsl-community-bsp repo manifest thing, it fails on the old kernel in meta-fsl-arm/recipes-kernel/linux/linux-imx_2.6.35.3.bb. There are a few things questionable/wrong here: 1) It uses the build host perl, which is

Re: [oe] Doubts regarding the final Image file

I'm not the TI guru-dude, but the beaglebone machines (white or black) don't seem to use the image bbclass, but should at least produce the rootfs tarballs and kernel/u-boot files to make a card (in your deploy directory). The TI sdcard and u-boot stuff is documented all over, but essentially you

Re: [oe] Fwd: Re: Fwd: Re: open embedded build issues with beagle board

Yes, check the layer index. I have a patched meta-beagleboard on github, with an older kernel, but meta-beagleboard is not as current as meta-ti (and not quite as official I believe). The latest kernel works reasonably well (the patches maintained by RCN) if you want to check the LinuxOnArm

Re: [oe] Fwd: Re: open embedded build issues with beagle board

Just add meta-ti to your bblayers and set your machine. The latest kernel *should* support video; there's also the LinuxOnArm wiki with a working kernel. Steve On Mon, Jul 14, 2014 at 9:29 PM, Bryan just4tec...@gmail.com wrote: Hi Cliff, thanks for your suggestion, i tried links

Re: [oe] [meta-oe][MASTER][PATCH] lxdm: add libpam DEPENDS

what I can. I'll let you test the systemd version... ;) Steve On Thu, Jul 10, 2014 at 8:09 AM, Trevor Woerner trevor.woer...@linaro.org wrote: Hi Steve, On 25 June 2014 00:16, Stephen Arnold stephen.arnol...@gmail.com wrote: Yeah, it would certainly make more sense with a packageconfig

Re: [oe] [meta-oe][MASTER][PATCH] lxdm: add libpam DEPENDS

this weekend and see what makes sense. I guess I need to warp my brain away from risk management and think about OE stuff again... Steve On Thu, Jul 10, 2014 at 8:09 AM, Trevor Woerner trevor.woer...@linaro.org wrote: Hi Steve, On 25 June 2014 00:16, Stephen Arnold stephen.arnol...@gmail.com wrote

Re: [oe] [meta-oe][MASTER][PATCH] lxdm: add libpam DEPENDS

Hey Trevor: I was just working on this one the other day (I have a bbappend for my xorg image) and added some pam deps, but not as hard deps (since pam is an optional feature). One issue I just hit in my master branch build was missing pam plugins for lxdm login, even with the pam feature

Re: [oe] [meta-oe][MASTER][PATCH] lxdm: add libpam DEPENDS

option to the flexible depends sounds good. I still need an init script anyway, so... Steve On Tue, Jun 24, 2014 at 1:49 PM, Trevor Woerner trevor.woer...@linaro.org wrote: On 06/24/14 16:02, Stephen Arnold wrote: but not as hard deps (since pam is an optional feature). I wasn't 100% before

Re: [oe] [meta-oe][WIP][PATCH 04/25] imlib2, mcnavi: Blacklist because of broken imlib2 after freetype2-config was removed

This one doesn't build with the pkg-config trick; it needs more help than that... The attached bbappend is what works for me, after removing the blacklist. Steve On Thu, Jun 19, 2014 at 2:59 AM, Martin Jansa martin.ja...@gmail.com wrote: Signed-off-by: Martin Jansa martin.ja...@gmail.com

Re: [oe] [meta-oe][WIP][PATCH 04/25] imlib2, mcnavi: Blacklist because of broken imlib2 after freetype2-config was removed

}-loaders = 1 DEBIAN_NOAUTONAME_${PN}-filters-dbg = 1 DEBIAN_NOAUTONAME_${PN}-loaders-dbg = 1 On Sun, Jun 22, 2014 at 7:05 PM, Stephen Arnold stephen.arnol...@gmail.com wrote: This one doesn't build with the pkg-config trick; it needs more help than that... The attached bbappend is what works

Re: [oe] meta-oe patch queue

Hi Martin: From what I can tell, it looks like the gnuplot patches got in but not all of the gd changes? I also don't see the gpm DESTDIR patch. Can you check if the gd license patch made it into your queue? There was also gpm DESTDIR patch sent on 4 April; let me know if I need to resend.

Re: [oe] meta-oe patch queue

Never mind, I hadn't plowed through the backlog yet... Steve On Mon, Apr 21, 2014 at 10:06 AM, Stephen Arnold stephen.arnol...@gmail.com wrote: Hi Martin: From what I can tell, it looks like the gnuplot patches got in but not all of the gd changes? I also don't see the gpm DESTDIR patch

Re: [oe] [meta-oe][PATCH v2 2/2] gnuplot: added upstream license file and updated recipe to match

This is 2/2 as requested by Martin, et al. Is there pending feedback, or did I miss an email? Thanks, Steve On Tue, Apr 1, 2014 at 8:47 PM, Stephen Arnold stephen.arnol...@gmail.comwrote: This fixes QA warnings about missing license file and updates the license type to match custom license

[oe] [meta-oe][PATCH] gpm: sets D equal to DESTDIR instead of ROOT in do_install

This corrects a gpm install failure due to ROOT being used instead of DESTDIR, resulting in DESTDIR being undefined. Upstream-Status: Inappropriate (distribution/packaging fix) Signed-off-by: Stephen Arnold stephen.arnol...@gmail.com --- meta-oe/recipes-support/gpm/gpm_1.99.7.bb | 2 +- 1 file

Re: [oe] [oe-core][PATCH] added upstream license files for gd and gnuplot

Kooi k...@dominion.thruhere.netwrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Stephen Arnold schreef op 01-04-14 23:49: Subject: [PATCH] added upstream license files for gd and gnuplot The general rule is that the first line says what the patch does, not what you did, so: [PATCH

Re: [oe] [meta-oe][PATCH v2 1/2] gd: added upstream license file and fixed configure warnings

Both license files are as-is from their respective source trees. md5sums match the LIC_FILES_CHKSUM in each of the recipes. Steve On Wed, Apr 2, 2014 at 5:50 AM, Martin Jansa martin.ja...@gmail.com wrote: On Tue, Apr 01, 2014 at 08:47:44PM -0700, Stephen Arnold wrote: This fixes QA

Re: [oe] USB wifi configuration

It pretty much depends on your image recipe (and/or kernel recipe if you've got a custom one). It sounds like your custom kernel config is being used, which is the first thing to check (there should be a kernel module package with your driver). Second is what gets installed to the image , so you

Re: [oe] ntpd fails to start if libcap not installed

I would agree, as well as making sure the ntp user owns the drift file and its run and log directories (although I don't believe the latter is configured in the current recipe). Moving the drift file (and keys) to /var/lib/ntp is a pretty standard config. Just my $.02 ... Steve On Sat, Mar

[oe] [meta-oe][PATCH 1/1] upgrade gnuplot to version 4.6.5

Applied changes in patch from Tim Orling, updated license to gnuplot. Signed-off-by: Stephen Arnold stephen.arnol...@gmail.com --- .../gnuplot/gnuplot-4.4.4/automake-1.12.x.patch| 44 -- .../lua-loadlibs-configure-in-fix.patch| 16 .../{gnuplot-4.4.4

[oe] [meta-oe][PATCH 1/1] fixed QA configure warnings

Removed stale confgure option, added staging path to freetype option, and added --disable-rpath (hardcoded rpaths being bad juju). Signed-off-by: Stephen Arnold stephen.arnol...@gmail.com --- meta-oe/recipes-support/gd/gd_2.0.35+2.0.36rc1.bb | 4 ++-- 1 file changed, 2 insertions(+), 2

[oe] [oe-core][PATCH] added upstream license files for gd and gnuplot

Fixes QA warnings about missing license files and supports gnuplot recipe change in license classification. Signed-off-by: Stephen Arnold stephen.arnol...@gmail.com --- meta/files/common-licenses/GD | 53 ++ meta/files/common-licenses/gnuplot | 29

Re: [oe] [oe-core][PATCH] added upstream license files for gd and gnuplot

Okay, resending this as a patch to meta-oe/licenses instead. Thanks for the tip... Steve On Tue, Apr 1, 2014 at 2:49 PM, Stephen Arnold stephen.arnol...@gmail.comwrote: Fixes QA warnings about missing license files and supports gnuplot recipe change in license classification. Signed-off

Re: [oe] [meta-oe][PATCH v2] changed gnuplot recipe license to gnuplot

good on this one... Steve On Tue, Apr 1, 2014 at 4:14 PM, Paul Barker p...@paulbarker.me.uk wrote: On 1 April 2014 23:36, Stephen Arnold stephen.arnol...@gmail.com wrote: Also sent license file patch for oe-core. This supercedes previous gnuplot patch that incorrectly included

[oe] [meta-oe][PATCH v2 1/2] gd: added upstream license file and fixed configure warnings

This fixes QA warnings about missing license file and stale configure options, (includes updates for freetype libpath and disables hardcoded rpaths). Upstream-Status: Inappropriate (distribution/packaging fix) Signed-off-by: Stephen Arnold stephen.arnol...@gmail.com --- meta-oe/licenses/GD

[oe] [meta-oe][PATCH v2 2/2] gnuplot: added upstream license file and updated recipe to match

This fixes QA warnings about missing license file and updates the license type to match custom license file. Upstream-Status: Inappropriate (distribution/packaging fix) Signed-off-by: Stephen Arnold stephen.arnol...@gmail.com --- meta-oe/licenses/gnuplot | 29

Re: [oe] [meta-oe][PATCH v2] changed gnuplot recipe license to gnuplot

Okay, restructured and resent as requested. Steve On Tue, Apr 1, 2014 at 5:41 PM, Martin Jansa martin.ja...@gmail.com wrote: On Tue, Apr 01, 2014 at 05:09:02PM -0700, Stephen Arnold wrote: Was working with Martin on this one; the original patch referred to is Tim's gnuplot upgrade patch

Re: [oe] [oe-commits] Cristian Iorga : musicpd: upgrade to 0.18.9

-openembedded.gita=commit;h=dde222cef9bb8ef20d139ab58760b27c4d45c93a Author: Stephen Arnold stephen.arnol...@gmail.com Date: Sun Mar 16 17:52:47 2014 -0700 musicpd/mpd: bump to current upstream stable Bump music player daemon to the latest upstream stable release. Same as previous 0.18.4

Re: [oe] [meta-xfce][PATCH v3] recipes-xfce/libxfce4ui: depends and autotools fixes

ERROR message shown, returning a non-zero exit code. Poky [raspberrypi] On Tue, Mar 18, 2014 at 2:19 AM, Stephen Arnold stephen.arnol...@gmail.comwrote: Because it only works in a separate task (see previous email). On Tue, Mar 18, 2014 at 1:46 AM, Andreas Müller schnitzelt

[oe] [meta-graphics][PATCH v3] recipes-graphics/lxdm: autotools and config updates

) and some configure options were added to ensure the desired X/gtk2 config and enable (optional) auto-login. Signed-off-by: Stephen Arnold stephen.arnol...@gmail.com --- meta-oe/recipes-graphics/lxdm/lxdm_git.bb | 13 ++--- 1 file changed, 10 insertions(+), 3 deletions(-) diff --git

Re: [oe] [meta-xfce][PATCH v3] recipes-xfce/libxfce4ui: depends and autotools fixes

Because it only works in a separate task (see previous email). On Tue, Mar 18, 2014 at 1:46 AM, Andreas Müller schnitzelt...@googlemail.com wrote: On Tue, Mar 18, 2014 at 5:34 AM, Stephen Arnold stephen.arnol...@gmail.com wrote: Note: this patch (v3) obseletes previous versions

Re: [oe] [meta-graphics][PATCH v3] recipes-graphics/lxdm: autotools and config updates

in getting to the true error. But as I said, it should be okay to remove them now. Do you want a new version of the patch minus the above? Steve On Tue, Mar 18, 2014 at 2:32 AM, Andreas Müller schnitzelt...@googlemail.com wrote: On Tue, Mar 18, 2014 at 10:09 AM, Stephen Arnold stephen.arnol

Re: [oe] [meta-xfce][PATCH v2 1/1] recipes-xfce/libxfce4ui: build fixes

17, 2014 at 1:39 AM, Andreas Müller schnitzelt...@googlemail.com wrote: On Mon, Mar 17, 2014 at 1:43 AM, Stephen Arnold stephen.arnol...@gmail.com wrote: From: Stephen Arnold stephen.arnol...@gmail.com libxfce4ui build fixes: - updated depends - set SRCREV to 4.10.0 tag AFAICR

[oe] [meta-xfce][PATCH v3] recipes-xfce/libxfce4ui: depends and autotools fixes

option. Signed-off-by: Stephen Arnold stephen.arnol...@gmail.com --- meta-xfce/recipes-xfce/libxfce4ui/libxfce4ui_4.10.0.bb | 14 ++ 1 file changed, 10 insertions(+), 4 deletions(-) diff --git a/meta-xfce/recipes-xfce/libxfce4ui/libxfce4ui_4.10.0.bb b/meta-xfce/recipes-xfce/libxfce4ui

Re: [oe] [meta-oe][PATCH] updated lxdm recipe depends and build config

Okay, apparently git-send-email doesn't suck any from info from the file in outgoing/. Good to know... On Sun, Mar 16, 2014 at 3:58 PM, stephen.arnold42 stephen.arnol...@gmail.com wrote: --- meta-oe/recipes-graphics/lxdm/lxdm_git.bb | 13 +++-- 1 file changed, 11 insertions(+), 2

Re: [oe] GDM alternative?

will be appreciated on my embedded boards. Adam On Thu, Feb 27, 2014 at 2:29 PM, Stephen Arnold stephen.arnol...@gmail.comwrote: It's a bit more subtle than just obvious advantage since the deps are different, config options are different, etc. Slim is probably lighter than lxdm, in both

[oe] [meta-oe][PATCH v2 1/1] recipes-graphics/lxdm: add missing automake file, update build config

From: Stephen Arnold stephen.arnol...@gmail.com PATCH recipes-graphics/lxdm: - add missing depends and inherit pkgconfig for autogen/configure - silence confgure warnings and update configure options - provide missing input file for automake Built and tested on master branches using meta

[oe] [meta-multimedia][PATCH v2 1/1] musicpd/mpd: bump to current upstream stable

From: Stephen Arnold stephen.arnol...@gmail.com Bump music player daemon to the latest upstream stable release. Same as previous 0.18.4 recipe with trivial updates. Signed-off-by: Stephen Arnold stephen.arnol...@gmail.com --- .../recipes-multimedia/musicpd/mpd_0.18.8.bb | 59

Re: [oe] [meta-multimedia][PATCH] updated mpd to current upstream stable

Not sure I understand; there should be only a new 0.18.8 recipe file in the commit. I have an updated one with a more standard message body, signed-off-by, etc. I'll resend a v2 in a minute, so please be specific if something looks funky. Thanks, Steve On Sun, Mar 16, 2014 at 5:02 PM, Martin

[oe] [meta-xfce][PATCH v2 1/1] recipes-xfce/libxfce4ui: build fixes

From: Stephen Arnold stephen.arnol...@gmail.com libxfce4ui build fixes: - updated depends - set SRCREV to 4.10.0 tag - added missing inherit - added pre-config task Signed-off-by: Stephen Arnold stephen.arnol...@gmail.com --- .../recipes-xfce/libxfce4ui/libxfce4ui_4.10.0.bb | 23

[oe] [meta-multimedia][PATCH v3 1/1] musicpd/mpd: bump to current upstream stable

From: Stephen Arnold stephen.arnol...@gmail.com Bump music player daemon to the latest upstream stable release. Same as previous 0.18.4 recipe with trivial updates. Signed-off-by: Stephen Arnold stephen.arnol...@gmail.com --- .../musicpd/{mpd_0.18.4.bb = mpd_0.18.8.bb} | 10

Re: [oe] [OE-devel] [PATCH] ntp: Resolve some abnormal behaviors

Sorry I missed the previous traffic, but does it make more sense to set the ntp user's home to something like /var/lib/ntp instead? For me it's always been that way as the typical place for the drift file (written to by the ntpd process running as ntp) as well as a *.keys file it needed. IMHO

Re: [oe] GDM alternative?

And I think I just convinced myself I need to get slim working... On Thu, Feb 27, 2014 at 2:29 PM, Stephen Arnold stephen.arnol...@gmail.comwrote: It's a bit more subtle than just obvious advantage since the deps are different, config options are different, etc. Slim is probably lighter

Re: [oe] GDM alternative?

LXDM and Slim? I love how LXDM doesn't require complicated authentication framework like GDM. Adam On Tue, Feb 25, 2014 at 7:49 PM, Stephen Arnold stephen.arnol...@gmail.comwrote: Is there any interest in reviving slim? I'm using it on gentoo-arm on several devices; works just like x86

Re: [oe] GDM alternative?

Is there any interest in reviving slim? I'm using it on gentoo-arm on several devices; works just like x86. Since I haven't tried it or otherwise looked at it in oe yet, I have no idea why it's in a non-working state. That said, it was on my todo list until I noticed lxdm, so I'll probably

Re: [oe] [meta-openembedded/meta-webserver][PATCH] nginx: new recipe, updated from aging original in rpi layer (has not been scrubbed for any potential policy issues).

Paul: Okay, I have a cleaner and more compliant recipe and a few questions for your comments. See below... On Wed, Oct 9, 2013 at 2:55 AM, Paul Eggleton paul.eggle...@linux.intel.com wrote: Hi Stephen, Thanks for sending this. A few comments below. On Tuesday 08 October 2013 22:33:44

Re: [oe] [meta-openembedded/meta-webserver][PATCH] nginx: new recipe, updated from aging original in rpi layer (has not been scrubbed for any potential policy issues).

Hmm, fun with gmail... Anyway, I was (somewhat blindly) following the submission/patch instructions here: http://www.openembedded.org/wiki/How_to_submit_a_patch_to_OpenEmbedded I can certainly provide more info (as well as clean up the recipe) but your first few points don't seem to be