Re: [OE-core] [oe-core] QEMU with ARM Cortex A9 with hard float configuration - Kernel panic
On Aug 29, 2013, at 11:16 PM, Bruce Ashfield wrote: > linux-yocto-3.10 is pretty much the standard kernel, unless you want to use > the extra features I've merged, it is vanilla. > > You are going overboard with the definitions. What the build really needs, > it'll > generate on it's own. If you want to make an official/controlled BSP > ... only then > do you actually need .scc files, etc. > > If you add compatbility with the linux-yocto recipe, put a defconfig in the > SRC_URI and you are done. Just adding the defconfig in the SRC_URI is not enough. The qemuarm launches with the versatilepb machine emulation, not vexpress. For me to get vexpress support, I've had to create a local copy of the linux-yocto repo, update the meta-branch and then do the adaptations. So, it's not just a simple one line fix as simple as adding a defconfig for linux-yocto. the runqemu scripts have to be modified, etc. I've shared all that I've done, with you, but you've apparently done this before, and I've just spend days to find out just that!! Thanks for the run around!! Elvis Dowson ___ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-core
Re: [OE-core] [oe-core] QEMU with ARM Cortex A9 with hard float configuration - Kernel panic
On Aug 29, 2013, at 10:53 PM, wrote: > I actually set up a build to get a vexpress-a15 booting in qemu with > neon support. However, I didn't use the yocto kernel and just pointed at > my own defconfig and kernel version 3.10.7. If it would be helpful for > you, I could put my changes up on github. > > The weirdest thing was if I just used the defconfig that is in the > kernel for vexpress it wouldn't boot, I had to tweak it a bit. This is the thing, I was using this as an excuse to learn linux-yocto kernel. I actually put off using linux-yocto for a while, and tended to use the official linux kernel, or vendor specific kernels (Xilinx, TI Arago, Freescale, etc) for actual target devices. If you could share your defconfig, I'll try it with the standard linux kernel. I'm doing this to get the process of using Yocto to build a representative QEMU emulator, and then using LaunchPad and Soyuz locally to build Ubuntu on a virtual machine, rather than just re-use the official Ubuntu binary packages. This way, I have more control over the toolchain, and options, rather than use the existing armel or armhf machine configurations supplied by Debian/Ubuntu. Best regards, Elvis Dowson ___ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-core
Re: [OE-core] [oe-core] QEMU with ARM Cortex A9 with hard float configuration - Kernel panic
On Aug 29, 2013, at 8:23 PM, Bruce Ashfield wrote: > On 13-08-29 10:02 AM, Elvis Dowson wrote: >> Hi Bruce, >> >> On Aug 29, 2013, at 4:35 PM, Bruce Ashfield >> wrote: >> >>> With that, it's always been a manual setup for me in the past. The typical >>> workflow is to test them via the usermode NFS boot. Have you given >>> that a go ? >> >> Does yocto have some wrapper script/command to setup nfs boot automatically >> without too much manual configuration? The runqemu and runqemu-internal >> scripts seems to indicate that, but haven't tried this before. >> >> Let me know if it exists! If not, if you could point me to a link that >> describes how >> to set NFS boot, and use it to load a beagleboard image it would be great! > > This is supported right out of the box, and it should be documented in > the project docs. > > Search for users-space NFS in > http://www.yoctoproject.org/docs/current/adt-manual/adt-manual.html I saw this, its only for the rootfs. what I've read on the net alluded to the fact that you need to have xload, uboot and the kernel image on a specific partition inside an sdcard image. Those were the boot procedures used for the beagleboard using qemu. This is for -M beagle, with qemu-system-arm. Anyway, I just need a way to boot vexpress-a9 using QEMU. I'll send a private email, with the patch to you, so that you can start off from where I left, in terms of modifications to poky-master and be able to debug the issue directly at your end. Best regards, Elvis Dowson ___ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-core
Re: [OE-core] [oe-core] QEMU with ARM Cortex A9 with hard float configuration - Kernel panic
On Aug 29, 2013, at 8:17 PM, Khem Raj wrote: > > On Aug 29, 2013, at 1:50 AM, Elvis Dowson wrote: > >>>> also take a look at meta-linaro, they test armv7+ qemu regularly. >> >> I built core-image-minimal using meta-linaro, using genericarmv7 machine but >> it just generates an Image file for the kernel and a tar.gz rootfs. It >> doesn't look like its useful out of the box for a QEMU image. Additionally >> you have to do something extra for the image to be usable for a beagle QEMU >> machine, related to x-load and u-boot, to get it to work. > > x-load, u-boot are not needed when booting a qemu machine. You just need a > kernel and rfs. So make sure that you use a kernel thats configured correctly > to boot on say vexpress emulation on qemu and that should be it. Real thing > is to figure out the kernel and any patches if needed. Tried that with the current poky master, it doesn't work. The only two qemu arm based machine configurations that work, at the moment are qemuarm (versatilepb) and qemuzynq from the meta-xilinx layer (xilinx-zynq-a9). The others don't work, it crashes upon booting. If you've got -M vexpress-a9 working, please let me know how you got it to run against yocto master. Elvis Dowson ___ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-core
Re: [OE-core] [oe-core] QEMU with ARM Cortex A9 with hard float configuration - Kernel panic
Hi Bruce, On Aug 29, 2013, at 4:35 PM, Bruce Ashfield wrote: > With that, it's always been a manual setup for me in the past. The typical > workflow is to test them via the usermode NFS boot. Have you given > that a go ? Does yocto have some wrapper script/command to setup nfs boot automatically without too much manual configuration? The runqemu and runqemu-internal scripts seems to indicate that, but haven't tried this before. Let me know if it exists! If not, if you could point me to a link that describes how to set NFS boot, and use it to load a beagleboard image it would be great! > Once I'm over the 3.10 introduction curve, I'm happy to dust off my > old A9 support and get the right qemu support for a full system boot. That sounds good. We should aim for adding support for both A8 and A9, since that represent the current bulk of last-gen and current-gen platforms out there. Best regards, Elvis Dowson ___ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-core
Re: [OE-core] [oe-core] QEMU with ARM Cortex A9 with hard float configuration - Kernel panic
>> also take a look at meta-linaro, they test armv7+ qemu regularly. I built core-image-minimal using meta-linaro, using genericarmv7 machine but it just generates an Image file for the kernel and a tar.gz rootfs. It doesn't look like its useful out of the box for a QEMU image. Additionally you have to do something extra for the image to be usable for a beagle QEMU machine, related to x-load and u-boot, to get it to work. Are you aware of the extract steps? Most of the docs I've seen use pre built sd card images. Best regards, Elvis ___ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-core
Re: [OE-core] [oe-core] QEMU with ARM Cortex A9 with hard float configuration - Kernel panic
On Aug 29, 2013, at 2:43 AM, Bruce Ashfield wrote: > On 13-08-28 11:16 AM, Elvis Dowson wrote: >> Hi, >>I get the following error while running the bitbake linux-yocto >> command for a qemu machine conf that I'm building. The build system is >> unable to find the dtc binary: >> >> | make: Leaving directory >> `/tool/yocto/poky/build/tmp/work/qemuarma9-poky-linux-gnueabi/linux-yocto/3.4.52+gitAUTOINC+7250de4d4a_4122d6cf3f-r4.5/image/usr/src/kernel' >> | NOTE: make -j 6 vexpress-v2p-ca9.dtb >> | DTC arch/arm/boot/vexpress-v2p-ca9.dtb >> | /bin/sh: >> /tool/yocto/poky/build/tmp/work/qemuarma9-poky-linux-gnueabi/linux-yocto/3.4.52+gitAUTOINC+7250de4d4a_4122d6cf3f-r4.5/linux-qemuarma9-standard-build/scripts/dtc/dtc: >> No such file or directory >> | make[3]: *** [arch/arm/boot/vexpress-v2p-ca9.dtb] Error 1 >> | make[2]: *** [vexpress-v2p-ca9.dtb] Error 2 >> | make[1]: *** [sub-make] Error 2 >> | make: *** [all] Error 2 >> | ERROR: oe_runmake failed >> | WARNING: >> /tool/yocto/poky/build/tmp/work/qemuarma9-poky-linux-gnueabi/linux-yocto/3.4.52+gitAUTOINC+7250de4d4a_4122d6cf3f-r4.5/temp/run.do_install.10606:1 >> exit 1 from >> | exit 1 >> | ERROR: Function failed: do_install (log file is located at >> /tool/yocto/poky/build/tmp/work/qemuarma9-poky-linux-gnueabi/linux-yocto/3.4.52+gitAUTOINC+7250de4d4a_4122d6cf3f-r4.5/temp/log.do_install.10606) >> ERROR: Task 4 >> (/tool/yocto/poky/meta/recipes-kernel/linux/linux-yocto_3.4.bb, >> do_install) failed with exit code '1' >> NOTE: Tasks Summary: Attempted 598 tasks of which 597 didn't need to be >> rerun and 1 failed. > > Did you ever get past this ? It looks like the recent device tree > changes where we use the kernel dtc versus an external dtc are causing > you a problem. I did get past this, after I updated the SRCREV_meta with the correct commit, which resulted in the correct kconfig fragment being picked up, with enabled the device tree option in the kernel. However, I'm unable to get qemu to boot correctly, for both ARM Cortex A9 and ARM Cortex A* (beagle). Maybe this has to do with the fact that it needs xload+uboot in addition to the kernel, using an SD card image. Any suggestions on how I can do this? I'd like to know how qemuarmv6, qemuarmv9 and qemuarma9 were tested earlier. I think its important to get atleast the A8 and A( QEMU platforms up and running in oe-core, since they represent a bunch of the current targets used for actual development. Best regards, Elvis Dowson ___ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-core
Re: [OE-core] [oe-core] QEMU with ARM Cortex A9 with hard float configuration - Kernel panic
Hi, I get the following error while running the bitbake linux-yocto command for a qemu machine conf that I'm building. The build system is unable to find the dtc binary: | make: Leaving directory `/tool/yocto/poky/build/tmp/work/qemuarma9-poky-linux-gnueabi/linux-yocto/3.4.52+gitAUTOINC+7250de4d4a_4122d6cf3f-r4.5/image/usr/src/kernel' | NOTE: make -j 6 vexpress-v2p-ca9.dtb | DTC arch/arm/boot/vexpress-v2p-ca9.dtb | /bin/sh: /tool/yocto/poky/build/tmp/work/qemuarma9-poky-linux-gnueabi/linux-yocto/3.4.52+gitAUTOINC+7250de4d4a_4122d6cf3f-r4.5/linux-qemuarma9-standard-build/scripts/dtc/dtc: No such file or directory | make[3]: *** [arch/arm/boot/vexpress-v2p-ca9.dtb] Error 1 | make[2]: *** [vexpress-v2p-ca9.dtb] Error 2 | make[1]: *** [sub-make] Error 2 | make: *** [all] Error 2 | ERROR: oe_runmake failed | WARNING: /tool/yocto/poky/build/tmp/work/qemuarma9-poky-linux-gnueabi/linux-yocto/3.4.52+gitAUTOINC+7250de4d4a_4122d6cf3f-r4.5/temp/run.do_install.10606:1 exit 1 from | exit 1 | ERROR: Function failed: do_install (log file is located at /tool/yocto/poky/build/tmp/work/qemuarma9-poky-linux-gnueabi/linux-yocto/3.4.52+gitAUTOINC+7250de4d4a_4122d6cf3f-r4.5/temp/log.do_install.10606) ERROR: Task 4 (/tool/yocto/poky/meta/recipes-kernel/linux/linux-yocto_3.4.bb, do_install) failed with exit code '1' NOTE: Tasks Summary: Attempted 598 tasks of which 597 didn't need to be rerun and 1 failed. File : qemuarmhf.conf #@TYPE: Machine #@NAME: qemuarma9 #@DESCRIPTION: Machine configuration for QEMU ARM Cortex A9. # Specify the default tune. DEFAULTTUNE = "armv7a-neon" require conf/machine/include/qemu.inc require conf/machine/include/tune-cortexa9.inc # Specify the kernel devicetree. KERNEL_DEVICETREE = "vexpress-v2p-ca9.dtb" # Specify the machine name used by the kernel. KMACHINE_qemuarma9 = "vexpress" # Specify the kernel image type. KERNEL_IMAGETYPE = "uImage" # Set default load address. # Override with KERNEL_EXTRA_ARGS_ += "..." in machine file if required #KERNEL_EXTRA_ARGS_qemuarma9 += "UIMAGE_LOADADDR=0x8000" # Specify the serial console parameters. SERIAL_CONSOLE = "115200 ttyAMA0" Best regards, Elvis Dowson___ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-core
Re: [OE-core] [oe-core] QEMU with ARM Cortex A9 with hard float configuration - Kernel panic
Hi Bruce,
On Aug 26, 2013, at 11:08 PM, Bruce Ashfield
wrote:
> Are you interested in something from scratch, versus something that
> isn't already partially in the tree (like qemuarma9) ?
>
> We've got a few docs and tutorials, but I can also help step you through
> the process to assist the existing material.
I'd like to adapt the existing linux-yocto standard/base branch, and build up
on existing qemuarma9 support. I've included detailed patch fragments below, so
that you can replicate the current state of my work at your end.
I've performed the following steps till date, but unfortunately get a qemu ram
/ rom out of memory address error after doing all this. Must be because of an
incorrectly mapped KMACHINE. The vexpress device doesn't support PCI, IDE or
SCSI, only microsd card or ram devices, so the boot method should use an initrd
file, for the root filesystem, as described here.
https://wiki.linaro.org/PeterMaydell/QemuVersatileExpress
Step 01.00: Clone linux-yocto-3.8 locally, create local meta branch, create
local standard/qemuarma9 branch, tracking the master branch.
Make the following modification to the qemuarma9-standard.scc file.
From 8e9b465f242941b70d7b61045c073e82700fcd91 Mon Sep 17 00:00:00 2001
From: Elvis Dowson
Date: Mon, 26 Aug 2013 22:18:24 +0400
Subject: [PATCH] qemuarma9: Update branch to qemuarma9.
Signed-off-by: Elvis Dowson
---
meta/cfg/kernel-cache/bsp/qemuarma9/qemuarma9-standard.scc |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/meta/cfg/kernel-cache/bsp/qemuarma9/qemuarma9-standard.scc
b/meta/cfg/kernel-cache/bsp/qemuarma9/qemuarma9-standard.scc
index c71fe1c..729125d 100644
--- a/meta/cfg/kernel-cache/bsp/qemuarma9/qemuarma9-standard.scc
+++ b/meta/cfg/kernel-cache/bsp/qemuarma9/qemuarma9-standard.scc
@@ -3,7 +3,7 @@ define KTYPE standard
define KARCH arm
include ktypes/standard/standard.scc
-branch arm-versatile-926ejs
+branch qemuarma9
include qemuarma9.scc
--
1.7.10.4
Step 02.00: Create a new qemuarma9.conf machine definition.
There is a dts file in arch/arm/boot/dts called vexpress-v2p-ca9.dts.
Q01: I guess the KMACHINE_qemuarma9 should be set to "vex press" ? instead of
qemuarma9 below?
#@TYPE: Machine
#@NAME: qemuarma9
#@DESCRIPTION: Machine configuration for QEMU ARM Cortex A9.
# Specify the default tune.
DEFAULTTUNE = "armv7a-neon"
require conf/machine/include/qemu.inc
require conf/machine/include/tune-cortexa9.inc
# Specify the preferred kernel recipe version.
PREFERRED_VERSION_linux-yocto ?= "3.8%"
# Specify the machine name used by the kernel.
KMACHINE_qemuarma9 = "qemuarma9" # Error?!!
# Specify the kernel image type.
KERNEL_IMAGETYPE = "uImage"
MACHINE_DEVICETREE := "qemuarma9.dts"
# Specify the serial console parameters.
SERIAL_CONSOLE = "115200 ttyAMA0"
Step 03.00: Patch the oe-core sources.
Step 03.01: Patch scripts/runqemu
@@ -37,10 +37,11 @@ usage() {
echo " qemuparams=\"xyz\" - specify custom parameters to QEMU"
echo " bootparams=\"xyz\" - specify custom kernel parameters during boot"
echo ""
echo "Examples:"
echo " $MYNAME qemuarm"
+echo " $MYNAME qemuarma9 ramfs"
echo " $MYNAME qemux86-64 core-image-sato ext3"
echo " $MYNAME path/to/bzImage-qemux86.bin path/to/nfsrootdir/ serial"
echo " $MYNAME qemux86 ramfs"
echo " $MYNAME qemux86 iso"
echo " $MYNAME qemux86 qemuparams=\"-m 256\""
@@ -105,11 +106,11 @@ process_filename() {
# bit more complex, but offers a great user experience.
KVM_ENABLED="no"
while true; do
arg=${1}
case "$arg" in
-"qemux86" | "qemux86-64" | "qemuarm" | "qemumips" | "qemumipsel" | \
+"qemux86" | "qemux86-64" | "qemuarm" | "qemuarma9" |"qemumips" |
"qemumipsel" | \
"qemumips64" | "qemush4" | "qemuppc" | "qemumicroblaze" | "qemuzynq")
[ -z "$MACHINE" ] && MACHINE=$arg || \
error "conflicting MACHINE types [$MACHINE] and [$arg]"
;;
"ext2" | "ext3" | "ext4" | "jffs2" | "nfs" | "btrfs")
@@ -222,17 +223,17 @@ if [ "$FSTYPE" = "nfs" -a -z "$ROOTFS" ]; then
error "NFS booting without an explicit ROOTFS path is not yet supported"
fi
if [ -z "$MACHINE" ]; then
if [ "x$FSTYPE" = "xvmdk" ]; then
-MACHINE=`basename $VM | sed
's/.*\(qemux86-64\|qemux86\|qemuarm\|qemumips64\|qemumips\|qemuppc\|qemush
Re: [OE-core] [oe-core] QEMU with ARM Cortex A9 with hard float configuration - Kernel panic
Hi Bruce, On Aug 26, 2013, at 8:24 AM, Bruce Ashfield wrote: > Perhaps the zynq machine definition is the right way to go, since nearly > enough support is mainline (in both qemu and the kernel), and we can > motivate the rest into their respective upstreams. > > If someone can come up the right conf files, and machine definition, I > can pretty easily support it in linux-yocto. Could you tell me what's steps I need to follow, to add a new machine definition and support to the linux-yocto kernel? I'd like to learn how to do this! Best regards, Elvis Dowson ___ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-core
Re: [OE-core] [oe-core] QEMU with ARM Cortex A9 with hard float configuration - Kernel panic
Hi Bruce, On Aug 26, 2013, at 5:03 AM, Bruce Ashfield wrote: > On 13-08-25 3:09 PM, Elvis Dowson wrote: >> I noticed that you've committed the changes for qemuarma9 KMACHINE >> configuration to the linux-yocto kernel. >> >> http://git.yoctoproject.org/cgit/cgit.cgi/linux-yocto-3.8/tree/meta/cfg/kernel-cache/bsp?h=meta >> >> I adapted the qemuarm.conf configuration and booted it up, but nothing shows >> up in the QEMU window, I just get a black screen. >> >> Has qemu with the qemuarma9 configuration been successfully built and tested >> before? > > It has been, but not with a machine definition that is part of oe-core > or Yocto. > > It was contributed as the output from some cortexA9 testing, and I've > been carrying the kernel parts ever since. Can you share the qemucortexa9.conf machine definition please? > >> >> This is how my qemuarm.conf looks like, locally at the moment: >> >> #@TYPE: Machine >> #@NAME: qemuarmhf >> #@DESCRIPTION: Machine configuration for QEMU ARM Cortex A9 hard float. >> >> # Specify the default tune. >> DEFAULTTUNE = "armv7a-neon" >> >> require conf/machine/include/qemu.inc >> require conf/machine/include/tune-cortexa9.inc >> >> # Specify the machine name used by the kernel. >> KMACHINE_qemuarm = "qemuarma9" > > This is a default, so you don't need to set it. The default for qemuarm, as documented in http://www.yoctoproject.org/docs/1.5/ref-manual/ref-manual.html#var-KFEATURE_DESCRIPTION is arm_versatile926ejs. So I had to modify it to force it to use qemuarma9. > >> >> # Specify the kernel image type. >> KERNEL_IMAGETYPE = "zImage" > > I recall booting it with a uImage, but maybe my memory is failing. > >> >> # Specify the serial console parameters. >> SERIAL_CONSOLE = "115200 ttyAMA0" > > All of the above looks reasonably sane, but it's more about how you > launch qemu than the kernel build that is probably resulting in your > issues. What machine subtype did you use ? I used -M versatilepb. I also passed the -cpu cortex-a9 parameter to qemu. I just got a blank screen. FYI, I was able to get the qemuzynq machine working with the meta-xilinx layer, and in that, there was also a blank screen, but the QEMU output and login prompt came on the console from where the rumqemu qemuzynq command was run. A different type of behaviour from that of qemuarm running arm_versatile926ejs. The qemuzynq machine uses the linux-xlnx kernel. I'd like to get a generic qemuarma9 machine configuration working with linux-yocto, if possible, and add it to the standard list of qemu targets. Best regards, Elvis Dowson ___ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-core
Re: [OE-core] [oe-core] QEMU with ARM Cortex A9 with hard float configuration - Kernel panic
Hi Bruce, I noticed that you've committed the changes for qemuarma9 KMACHINE configuration to the linux-yocto kernel. http://git.yoctoproject.org/cgit/cgit.cgi/linux-yocto-3.8/tree/meta/cfg/kernel-cache/bsp?h=meta I adapted the qemuarm.conf configuration and booted it up, but nothing shows up in the QEMU window, I just get a black screen. Has qemu with the qemuarma9 configuration been successfully built and tested before? This is how my qemuarm.conf looks like, locally at the moment: #@TYPE: Machine #@NAME: qemuarmhf #@DESCRIPTION: Machine configuration for QEMU ARM Cortex A9 hard float. # Specify the default tune. DEFAULTTUNE = "armv7a-neon" require conf/machine/include/qemu.inc require conf/machine/include/tune-cortexa9.inc # Specify the machine name used by the kernel. KMACHINE_qemuarm = "qemuarma9" # Specify the kernel image type. KERNEL_IMAGETYPE = "zImage" # Specify the serial console parameters. SERIAL_CONSOLE = "115200 ttyAMA0" Elvis Dowson ___ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-core
Re: [OE-core] [oe-core] QEMU with ARM Cortex A9 with hard float configuration - Kernel panic
On Aug 25, 2013, at 10:02 PM, Elvis Dowson wrote: > The thing is I don't see a defconfig anywhere, and I was wondering how the > linux-yocto kernel picks up the defconfig from the standard/base branch. I found the answer for this one: http://www.yoctoproject.org/docs/1.4/ref-manual/ref-manual.html#var-KFEATURE_DESCRIPTION Now what do I do if the standard linux-yocto kernel doesn't have an ARM Cortex A9 processor based machine? Elvis Dowson ___ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-core
Re: [OE-core] [oe-core] QEMU with ARM Cortex A9 with hard float configuration - Kernel panic
Hi, On Aug 25, 2013, at 9:43 PM, rew...@declaratino.com wrote: >> >> Thanks! Any idea why the qemu with hard float configuration has a kernel >> panic with linux-yocto? > > init is going away so the kernel panics. Usually this means it couldn't > run the init on the rootfilesystem. Try adding keep_bootcon to your > kernel command line and see if that gives anymore debug. I ran bitbake linux-yocto -c menuconfig and noticed that the wrong ARM architecture was selected (it was using the same one set using tune-arm926ejs.inc) The thing is I don't see a defconfig anywhere, and I was wondering how the linux-yocto kernel picks up the defconfig from the standard/base branch. A TI OMAP 4430 Pandaboard or i.MX6 quad-core SABRE development kit platform would do, but not sure how the linux-yocto kernel works, in terms of pre-selecting the correct defconfig file. Best regards, Elvis Dowson ___ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-core
Re: [OE-core] [oe-core] QEMU with ARM Cortex A9 with hard float configuration - Kernel panic
Hi Martin, On Aug 25, 2013, at 6:44 PM, Martin Jansa wrote: >> I find that the kernel panics upon boot. Thanks! Any idea why the qemu with hard float configuration has a kernel panic with linux-yocto? Elvis Dowson ___ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-core
[OE-core] [oe-core] QEMU with ARM Cortex A9 with hard float configuration - Kernel panic
Hi, I have modified the default qemuarm.conf file, to switch it over to using ARM Cortex A9, with hard float, vfp and neon support, by applying the following patch (which basically includes the tune-cortexa9.inc file).diff --git a/meta/conf/machine/qemuarm.conf b/meta/conf/machine/qemuarm.confindex d07084b..2ec57d8 100644--- a/meta/conf/machine/qemuarm.conf+++ b/meta/conf/machine/qemuarm.conf@@ -1,10 +1,9 @@ #@TYPE: Machine-#@NAME: arm_versatile_926ejs-#@DESCRIPTION: arm_versatile_926ejs+#@NAME: qemuarmhf+#@DESCRIPTION: Machine configuration for QEMU ARM Cortex A9 hard float. require conf/machine/include/qemu.inc-require conf/machine/include/tune-arm926ejs.inc-#require conf/machine/include/tune-arm1136jf-s.inc+require conf/machine/include/tune-cortexa9.inc KERNEL_IMAGETYPE = "zImage"I find that the kernel panics upon boot. Is it possible to have a QEMU build with support for hard float? I need this, so that I can bootstrap a Ubuntu build from scratch, using a QEMU arm hard float machine, using a oe-core generated QEMU and toolchain. I don't want to use the default armel or armhf emulator and toolchain provided by debian, and want to build my own version of the emulator and toolchain, specifically optimized for a particular target machine.I've attached a screenshot below.Best regards,Elvis Dowson___ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-core
[OE-core] meta-xilinx moved to meta-xilinx-community
Hi, The existing meta-xilinx repo has moved to meta-xilinx-community http://git.yoctoproject.org/cgit/cgit.cgi/meta-xilinx-community/ This layer will contain community support for the Xilinx platforms, including legacy boards (ML507) and other soft-processor architectures. This paves the way for an officially supported meta-xilinx layer, the details of which will be officially announced by Xilinx. Best regards, Elvis Dowson ___ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-core
Re: [OE-core] [PATCH] local.conf.sample: Have comments ready to go for quad-core (4 -> 8).
On Apr 18, 2013, at 8:51 PM, "Robert P. J. Day" wrote:
> On Thu, 18 Apr 2013, Enrico Scholz wrote:
>
>> "Robert P. J. Day"
>> writes:
>>
>>> Have default, commented values for parallelism reflect quad-core
>>> hosts.
>>
>> fwiw, I use
>>
>> PARALLEL_MAKE = "\
>> -j ${@int(os.sysconf(os.sysconf_names['SC_NPROCESSORS_ONLN'])) * 2} \
>> -l ${@int(os.sysconf(os.sysconf_names['SC_NPROCESSORS_ONLN'])) * 150/100} \
>> "
>>
>> BB_NUMBER_THREADS ?= "\
>> ${@int(os.sysconf(os.sysconf_names['SC_NPROCESSORS_ONLN'])) * 150/100}"
>>
>>
>> which adjusts parallelism automatically.
>
> i'd be perfectly fine with that, too, if someone wants to put that
> in instead.
I've noticed that on a quad-core machine (i7) the best setting is to leave it
at 6 for
BB_NUMBER_THREADS and PARALLEL_MAKE.
8, 16, etc don't have any discernable impact, perhaps save 30 to 40 seconds off
a build. The 3.5Ghz standard CPU frequency can be over-clocked to as high
as 4.5GHz, but I tested it out at 4.2Ghz and 3.8Ghz, and the difference was
only 2 to 3 minutes between 3.8GHz and 4.2GHz over-clocked, so I currently
leave it at 3.8GHz over-clocked.
I'm using two SSDs in a RAID 0 striped configuration, and build times is
around 22 minutes for a Zynq-7-ZC702 build for core-image-minimal.
Elvis Dowson
___
Openembedded-core mailing list
Openembedded-core@lists.openembedded.org
http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
Re: [OE-core] gcc-cross-initial Failure
On Apr 5, 2013, at 8:36 PM, Jack Mitchell wrote: > On 05/04/2013 17:15, Khem Raj wrote: >> On Apr 5, 2013, at 7:03 AM, Jack Mitchell wrote: >> >>> Doing my usual git master pull and rebuild today, resulted in >>> gcc-cross-initial failing to build. >>> >>> Please find the log attached, does anyone know what could have caused it? I >>> recently changed the TUNES, so could something not be being re-built when >>> it should be? >> >> it could be gcc is crashing on you. What TUNEs did you change > > I build for the beaglebone and I changed them in line with your default > beaglebone build patch you posted a week or so ago. I think it moved it form > soft to hard float possibly… This is the same type of error that I'm currently stuck with, while attempting to build for microblaze hard-float using both gcc-4.7 and gcc-4.8 recipes with the current poky master branch. I haven't tried building for soft-float, but will do so now. If you look at the errors, it is related to libgcc. | /tool/yocto/poky/build/tmp/sysroots/x86_64-linux/usr/bin/microblazeel-v830-bs-cmp-re-ml-div-fb-poky-linux/microblazeel-poky-linux-as: unrecognized option '-EL' | make[1]: *** [divsi3_s.o] Error 1 | make[1]: *** Waiting for unfinished jobs | /tool/yocto/poky/build/tmp/sysroots/x86_64-linux/usr/bin/microblazeel-v830-bs-cmp-re-ml-div-fb-poky-linux/microblazeel-poky-linux-as: unrecognized option '-EL' | make[1]: *** [moddi3_s.o] Error 1 | /tool/yocto/poky/build/tmp/sysroots/x86_64-linux/usr/bin/microblazeel-v830-bs-cmp-re-ml-div-fb-poky-linux/microblazeel-poky-linux-as: unrecognized option '-EL' | make[1]: *** [modsi3_s.o] Error 1 | /tool/yocto/poky/build/tmp/sysroots/x86_64-linux/usr/bin/microblazeel-v830-bs-cmp-re-ml-div-fb-poky-linux/microblazeel-poky-linux-as: unrecognized option '-EL' | make[1]: *** [muldi3_hard_s.o] Error 1 | /tool/yocto/poky/build/tmp/sysroots/x86_64-linux/usr/bin/microblazeel-v830-bs-cmp-re-ml-div-fb-poky-linux/microblazeel-poky-linux-as: unrecognized option '-EL' | make[1]: *** [stack_overflow_exit_s.o] Error 1 | /tool/yocto/poky/build/tmp/sysroots/x86_64-linux/usr/bin/microblazeel-v830-bs-cmp-re-ml-div-fb-poky-linux/microblazeel-poky-linux-as: unrecognized option '-EL' | make[1]: *** [mulsi3_s.o] Error 1 | make[1]: Leaving directory `/tool/yocto/poky/build/tmp/work/microblazeel-v830-bs-cmp-re-ml-div-fb-poky-linux/gcc-cross-initial/4.8.0-r0/gcc-4.8.0/build.x86_64-linux.microblazeel-poky-linux/microblazeel-poky-linux/libgcc' | make: *** [all-target-libgcc] Error 2 | ERROR: oe_runmake failed | ERROR: Function failed: do_compile (see /tool/yocto/poky/build/tmp/work/microblazeel-v830-bs-cmp-re-ml-div-fb-poky-linux/gcc-cross-initial/4.8.0-r0/temp/log.do_compile.8602 for further information) Best regards, Elvis Dowson ___ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
Re: [OE-core] [RFT] GCC 4.8 recipes
Hi Khem, On Mar 28, 2013, at 9:51 AM, Khem Raj wrote: > All, > > I have added gcc 4.8 recipes and the branch is here > > git://git.openembedded.org/openembedded-core-contrib kraj/gcc-4.8 > > cgit > > http://git.openembedded.org/openembedded-core-contrib/log/?h=kraj/gcc-4.8 > > Please test it out on your respective environments if you can. > > To enable gcc 4.8 with OE-Core you would > > in local.conf set > > GCCVERSION = "4.8%" > > and if you use angstrom. Please checkout 'next' branch of setup-scripts > and then in conf/local.conf > > set > > ANGSTROM_GCC_VERSION = "4.8%" > > My testing so far has been building systems-image for all qemu machines and > they > seem to boot as good as 4.7 did. But I am hoping for wider testing and > finding issues > or very least improve the existing recipes for now and once the 1.4 release > is done > we can then propose it for inclusion in OE-Core > > Thanks for your time I've built and tested the new gcc-4.8.0 recipe for the Xilinx ZC702 Zynq-7020 platform (dual ARM Cortex A9). The kernel (3.6) and corresponding rootfilesystem boots fine. Best regards, Elvis Dowson ___ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
[OE-core] Enabling microblaze support for gcc-4.7.2
Hi, I'm trying to build the toolchain for the microblaze processor, but I get an error with gcc-cross-initial_4.7.bb: opening microblaze-c.o: No such file or directory. If I look in the build/tmp/work/microblaze-poky-linux/gcc-cross-initial/4.7.2-r18/gcc-4.7.2/gcc/config folder I can only see the following processor architectures i386 mips rs6000 s390 sparc If I search for microblaze-c.o, it hasn't been built. I using the meta-xilinx tree, and the yocto/poky master branch for this build. How do I go about enabling support in gcc to get it to build a toolchain for the microblaze processor? Elvis Dowson ___ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
Re: [OE-core] for a quad-core machine, shouldn't parallelism be set to 8?
Hi Robert, On my i7 quad core machine, I've set both values to 6. Setting it to 6 appears to be nearly the same as setting it to 8, I dare say a marginally faster by a minute. I can build core image console in around 22 minutes. Elvis Sent from my iPad On Jan 9, 2013, at 4:55 PM, "Robert P. J. Day" wrote: > > just noticed this in local.conf.sample: > > # For a quad-core machine, BB_NUMBER_THREADS = "4", PARALLEL_MAKE = "-j 4" > would > # be appropriate for example. > > i thought the general rule was that parallelism should be *twice* > the number of cores. that's certainly what's suggested in the yocto > QS guide: > > "Another couple of variables of interest are the BB_NUMBER_THREADS and > the PARALLEL_MAKE variables. By default, these variables are commented > out. However, if you have a multi-core CPU you might want to uncomment > the lines and set both variables equal to twice the number of your > host's processor cores. " > > thoughts? > > rday > > -- > > > Robert P. J. Day Ottawa, Ontario, CANADA >http://crashcourse.ca > > Twitter: http://twitter.com/rpjday > LinkedIn: http://ca.linkedin.com/in/rpjday > > > ___ > Openembedded-core mailing list > Openembedded-core@lists.openembedded.org > http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core ___ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
Re: [OE-core] [PATCH 3/9] gcc-cross: Make gcc-cross-initial as the only intermediate gcc stage
Hi Khem, On Oct 9, 2012, at 5:48 PM, Elvis Dowson wrote: > So, now I've got gcc-4.5.4 working with eglibc-2.13, upto the point where the > kconfig infrastructure was added. > > Next, to add the kconfig infrastructure to eglibc-2.13, and fast forward to > the tip of the poky master branch, and > see if things are still ok, with the PowerPC440. Good news, I was able to get the backported patches for gcc-4.5.4 and eglibc-2.13 working with the latest poky and meta-openembedded master branch commits. This includes the gcc toolchain rework and the kconfig support to eglibc. Let me now try to switch to gcc-4.7.2 and eglibc-2.16, and see if things work. I noticed that there was a recent patch to fix a core-image.bb class, that prevented root login if debug tweaks was enabled. Perhaps that is why I didn't see the boot prompt on the target console with the recent gcc versions? Best regards, Elvis Dowson ___ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
Re: [OE-core] [PATCH 3/9] gcc-cross: Make gcc-cross-initial as the only intermediate gcc stage
Hi Khem, On Oct 6, 2012, at 7:41 PM, Khem Raj wrote: > On Fri, Oct 5, 2012 at 11:59 PM, Elvis Dowson wrote: >> >> http://git.yoctoproject.org/cgit/cgit.cgi/poky/commit/?id=48df3fe947c4d900aa6b66d9d8352c0d9e59ebe4 eglibc: Fix eglibc-initial and let eglibc depend on it >> I've narrowed down the init problem, to the application of the above patch, >> but I'm unable to figure out why it is failing. Any suggestions? >> > > is it eglibc 2.13 ? or 2.16 one thing I see is that it is disabling > ssp but otherwise it looks not to so much than changing eglibc to > use gcc-initial I was able to get past the error, with back porting the following patch: http://git.yoctoproject.org/cgit/cgit.cgi/poky/commit/?id=42f0f96903ac312881ef841ae4380b4d1c72980f gcc-4.7: Backport libgcc fixes to appease the new build sequence disabling ssp was not the problem, tested with both ssp enabled and disabled. So, now I've got gcc-4.5.4 working with eglibc-2.13, upto the point where the kconfig infrastructure was added. Next, to add the kconfig infrastructure to eglibc-2.13, and fast forward to the tip of the poky master branch, and see if things are still ok, with the PowerPC440. Best regards, Elvis Dowson___ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
Re: [OE-core] [PATCH 3/9] gcc-cross: Make gcc-cross-initial as the only intermediate gcc stage
Hi Khem, On Oct 6, 2012, at 7:41 PM, Khem Raj wrote: > On Fri, Oct 5, 2012 at 11:59 PM, Elvis Dowson wrote: >> >> I've narrowed down the init problem, to the application of the above patch, >> but I'm unable to figure out why it is failing. Any suggestions? >> > > is it eglibc 2.13 ? or 2.16 one thing I see is that it is disabling > ssp but otherwise it looks not to so much than changing eglibc to > use gcc-initial It's eglibc-2.13. I will try without disabling ssp. I also cherry-picked this patch http://git.yoctoproject.org/cgit/cgit.cgi/poky/commit/?id=800ce0f1d09f2d4fc489c98ecca04cf695d707fa gcc-cross-initial: Reinstate limits.h generation fix and I think I'll need to backport this patch to gcc-4.5.1, http://git.yoctoproject.org/cgit/cgit.cgi/poky/commit/?id=42f0f96903ac312881ef841ae4380b4d1c72980f gcc-4.7: Backport libgcc fixes to appease the new build sequence FYI, I also briefly tested the latest poky master as of today, for the powerpc440 target, but no change from the previous sitatuation, with gcc-4.7.x and eglibc-2.16, which is init=/bin/sh gives no bash prompt. It just finishes loading the kernel and stops with no prompt. If I click on the enter key, the display scrolls up, but that's it. With gcc-4.5.1, and specific commits, it varies with things working correctly and giving a bash prompt; to a kernel panic when attempting to run the bash prompt. Best regards, Elvis Dowson ___ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
Re: [OE-core] [PATCH 3/9] gcc-cross: Make gcc-cross-initial as the only intermediate gcc stage
Hi Khem, On Aug 25, 2012, at 1:59 AM, Khem Raj wrote: > On Friday, August 24, 2012, Elvis Dowson wrote: > > > > On Aug 25, 2012, at 1:10 AM, Khem Raj wrote: > > > >> On Fri, Aug 24, 2012 at 2:08 PM, Elvis Dowson > >> wrote: > >>> > >>> I'm adapting my earlier gcc-4.5 recipes, to try and get them to build with > >>> the latest poky master. > >>> > >>> I'm using eglibc-2.13, so it is possible for this version to drop the > >>> gcc-cross-intermediate stage, or is it only with newer eglibc versions? > >>> > >> > >> well it wont work with anything older than eglibc 2.16 > > > > What should I do to get the gcc-4.5 recipes to continue to build, with > > eglib-2.13, > > against the poky master branch? Should I retain the gcc-intermediate > > recipes? > > Back port some patches I did for eglibc 2.16 and drop intermediate recipes I've back ported the patches you did for eglibc-2.16 for eglibc-2.13, to use with my gcc-4.5 recipe, for the PowerPC 440 target. With that patch applied, in a local copy of my meta-openembedded/toolchain-layer, containing both the eglibc-2.13 and gcc-4.5, recipe, building against the poky/master commit id works: http://git.yoctoproject.org/cgit/cgit.cgi/poky/commit/?id=b0faebd1f07e1616004bd19664395932e7c2c48f gcc-cross: Make gcc-cross-initial as the only intermediate gcc stage I am able to get to the shell prompt, and get no crashes upto this point. The root file system seems to be working fine, and I am able to ping the host machine. It is with the application of the next patch, that I get a an init crash on the console, as I have been seeing earlier. http://git.yoctoproject.org/cgit/cgit.cgi/poky/commit/?id=48df3fe947c4d900aa6b66d9d8352c0d9e59ebe4 eglibc: Fix eglibc-initial and let eglibc depend on it I've narrowed down the init problem, to the application of the above patch, but I'm unable to figure out why it is failing. Any suggestions? Best regards, Elvis Dowson ___ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
Re: [OE-core] [PATCH 3/9] gcc-cross: Make gcc-cross-initial as the only intermediate gcc stage
On Aug 25, 2012, at 1:10 AM, Khem Raj wrote: > On Fri, Aug 24, 2012 at 2:08 PM, Elvis Dowson wrote: >> >> I'm adapting my earlier gcc-4.5 recipes, to try and get them to build with >> the latest poky master. >> >> I'm using eglibc-2.13, so it is possible for this version to drop the >> gcc-cross-intermediate stage, or is it only with newer eglibc versions? >> > > well it wont work with anything older than eglibc 2.16 What should I do to get the gcc-4.5 recipes to continue to build, with eglib-2.13, against the poky master branch? Should I retain the gcc-intermediate recipes? Best regards, Elvis ___ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
Re: [OE-core] [PATCH 3/9] gcc-cross: Make gcc-cross-initial as the only intermediate gcc stage
Hi Khem, On Aug 21, 2012, at 7:07 AM, Khem Raj wrote: > Now glibc can be compiled with gcc-cross-initial therefore prepare > the stage to drop gcc-cross-intermediate I'm adapting my earlier gcc-4.5 recipes, to try and get them to build with the latest poky master. I'm using eglibc-2.13, so it is possible for this version to drop the gcc-cross-intermediate stage, or is it only with newer eglibc versions? At the moment, I've made the required changes to the gcc-4.5 recipes, to match that of the current gcc-4.7 templates, and deleted the gcc-4.5 intermediate recipes. Right now I get a compile error with eglibc-2.13, as follows: | powerpc-poky-linux-gcc -m32 -msoft-float -mcpu=440 --sysroot=/tool/yocto/poky/build/tmp/sysroots/virtex5-tcbootstrap -nostdlib -nostartfiles -o /tool/yocto/poky/build/tmp/work/ppc440-poky-linux/eglibc-2.13-r28+svnr15508/build-powerpc-poky-linux/iconv/iconv_prog -msoft-float -Wl,-dynamic-linker=/lib/ld.so.1 -Wl,-z,combreloc -Wl,-z,relro -Wl,--hash-style=both /tool/yocto/poky/build/tmp/work/ppc440-poky-linux/eglibc-2.13-r28+svnr15508/build-powerpc-poky-linux/csu/crt1.o /tool/yocto/poky/build/tmp/work/ppc440-poky-linux/eglibc-2.13-r28+svnr15508/build-powerpc-poky-linux/csu/crti.o `powerpc-poky-linux-gcc -m32 -msoft-float -mcpu=440 --sysroot=/tool/yocto/poky/build/tmp/sysroots/virtex5-tcbootstrap -msoft-float --print-file-name=crtbegin.o` /tool/yocto/poky/build/tmp/work/ppc440-poky-linux/eglibc-2.13-r28+svnr15508/build-powerpc-poky-linux/iconv/iconv_prog.o /tool/yocto/poky/build/tmp/work/ppc440-poky-linux/eglibc-2.13-r28+svnr15508/build-powerpc-poky-linux/iconv/iconv_charmap.o /tool/yocto/poky/build/tmp/work/ppc440-poky-linux/eglibc-2.13-r28+svnr15508/build-powerpc-poky-linux/iconv/charmap.o /tool/yocto/poky/build/tmp/work/ppc440-poky-linux/eglibc-2.13-r28+svnr15508/build-powerpc-poky-linux/iconv/charmap-dir.o /tool/yocto/poky/build/tmp/work/ppc440-poky-linux/eglibc-2.13-r28+svnr15508/build-powerpc-poky-linux/iconv/linereader.o /tool/yocto/poky/build/tmp/work/ppc440-poky-linux/eglibc-2.13-r28+svnr15508/build-powerpc-poky-linux/iconv/dummy-repertoire.o /tool/yocto/poky/build/tmp/work/ppc440-poky-linux/eglibc-2.13-r28+svnr15508/build-powerpc-poky-linux/iconv/simple-hash.o /tool/yocto/poky/build/tmp/work/ppc440-poky-linux/eglibc-2.13-r28+svnr15508/build-powerpc-poky-linux/iconv/xstrdup.o /tool/yocto/poky/build/tmp/work/ppc440-poky-linux/eglibc-2.13-r28+svnr15508/build-powerpc-poky-linux/iconv/xmalloc.o -Wl,-rpath-link=/tool/yocto/poky/build/tmp/work/ppc440-poky-linux/eglibc-2.13-r28+svnr15508/build-powerpc-poky-linux:/tool/yocto/poky/build/tmp/work/ppc440-poky-linux/eglibc-2.13-r28+svnr15508/build-powerpc-poky-linux/math:/tool/yocto/poky/build/tmp/work/ppc440-poky-linux/eglibc-2.13-r28+svnr15508/build-powerpc-poky-linux/elf:/tool/yocto/poky/build/tmp/work/ppc440-poky-linux/eglibc-2.13-r28+svnr15508/build-powerpc-poky-linux/dlfcn:/tool/yocto/poky/build/tmp/work/ppc440-poky-linux/eglibc-2.13-r28+svnr15508/build-powerpc-poky-linux/nss:/tool/yocto/poky/build/tmp/work/ppc440-poky-linux/eglibc-2.13-r28+svnr15508/build-powerpc-poky-linux/nis:/tool/yocto/poky/build/tmp/work/ppc440-poky-linux/eglibc-2.13-r28+svnr15508/build-powerpc-poky-linux/rt:/tool/yocto/poky/build/tmp/work/ppc440-poky-linux/eglibc-2.13-r28+svnr15508/build-powerpc-poky-linux/resolv:/tool/yocto/poky/build/tmp/work/ppc440-poky-linux/eglibc-2.13-r28+svnr15508/build-powerpc-poky-linux/crypt:/tool/yocto/poky/build/tmp/work/ppc440-poky-linux/eglibc-2.13-r28+svnr15508/build-powerpc-poky-linux/nptl /tool/yocto/poky/build/tmp/work/ppc440-poky-linux/eglibc-2.13-r28+svnr15508/build-powerpc-poky-linux/libc.so.6 /tool/yocto/poky/build/tmp/work/ppc440-poky-linux/eglibc-2.13-r28+svnr15508/build-powerpc-poky-linux/libc_nonshared.a /tool/yocto/poky/build/tmp/work/ppc440-poky-linux/eglibc-2.13-r28+svnr15508/build-powerpc-poky-linux/elf/ld.so -lgcc -lgcc_eh `powerpc-poky-linux-gcc -m32 -msoft-float -mcpu=440 --sysroot=/tool/yocto/poky/build/tmp/sysroots/virtex5-tcbootstrap -msoft-float --print-file-name=crtend.o` /tool/yocto/poky/build/tmp/work/ppc440-poky-linux/eglibc-2.13-r28+svnr15508/build-powerpc-poky-linux/csu/crtn.o | /tool/yocto/poky/build/tmp/sysroots/x86_64-linux/usr/libexec/ppc440-poky-linux.gcc-cross-initial/gcc/powerpc-poky-linux/4.5.1/ld: cannot find -lgcc_eh | collect2: ld returned 1 exit status | make[2]: *** [/tool/yocto/poky/build/tmp/work/ppc440-poky-linux/eglibc-2.13-r28+svnr15508/build-powerpc-poky-linux/iconv/iconvconfig] Error 1 | make[2]: *** Waiting for unfinished jobs | /tool/yocto/poky/build/tmp/sysroots/x86_64-linux/usr/libexec/ppc440-poky-linux.gcc-cross-initial/gcc/powerpc-poky-linux/4.5.1/ld: cannot find -lgcc_eh | collect2: ld returned 1 exit status | make[2]: *** [/tool/yocto/poky/build/tmp/work/ppc440-poky-linux/eglibc-2.13-r28+svnr15508/build-powerpc-poky-linux/iconv/iconv_prog] Error 1 | make[2]: Leaving directory `/tool/yocto/poky/build
Re: [OE-core] can't login??
Hi, On Aug 20, 2012, at 2:34 PM, Paul Eggleton wrote: > > Is "debug-tweaks" in IMAGE_FEATURES? > I too have the same problem, I've got debug-tweaks in IMAGE_FEATURES, but it doesn't allow me to login. If I type root as the username, it does nothing and brings up the login prompt again. Starting syslogd/klogd: Yocto (Built by Poky 7.0) 1.2+snapshot-20120820 (none) ttyS0 (none) login: root Yocto (Built by Poky 7.0) 1.2+snapshot-20120820 (none) ttyS0 (none) login: Yocto (Built by Poky 7.0) 1.2+snapshot-20120820 (none) ttyS0 (none) login: Yocto (Built by Poky 7.0) 1.2+snapshot-20120820 (none) ttyS0 (none) login: Elvis Dowson___ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
Re: [OE-core] number of tasks increased since last update
Hi, On Aug 1, 2012, at 11:08 PM, Chris Larson wrote: > On Wed, Aug 1, 2012 at 8:29 AM, Andreas Müller > wrote: >> today I updated and found the the number of tasks has increased about >> 400. Another strange behaviour: in my build dir there are two new >> folders: >> >> x86_64-nativesdk-angstromsdk-linux >> x86_64-angstromsdk-linux >> >> but up to now I was not interested in sdk stuff. Is that another bug >> or a feature? I too found some new extra directories in my sysroots folder and work folders sysroots virtex5 virtex5-tcbootstrap x86_64-linux x86_64-nativesdk-pokysdk-linux x86_64-nativesdk-poky-linux-tcbootstrap work all-poky-linux ppc440-poky-linux virtex5-poky-linux x86_64-linux x86_64-nativesdk-pokysdk-linux x86_64-pokysdk-linux Elvis Dowson ___ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
Re: [OE-core] The Mythical Sato Replacement
Hi, How about adding a meta-android layer, and creating a core-image-minimal-android build? The android kernels are available, they have a qemu target for android, and the android sdk will need a bbclass to download, build and create an android root filesystem. A couple of targets already are compatible, mostly the TI OMAP/AM series. Elvis Dowson ___ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
Re: [OE-core] ERROR: ExpansionError during parsing /tool/yocto/poky/meta/recipes-kernel/linux/linux-yocto_3.2.bb: Failure expanding variable SRCPV
Hi, On Jun 8, 2012, at 4:05 PM, Tom Zanussi wrote: > I just posted a patch to fix this - it was a SRCREV typo. > > Building fine here now with that simple fix... Yes, removing the stray 'l' character from the SRCREV fixed it, the build is progressing fine now. Best regards, Elvis Dowson ___ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
[OE-core] ERROR: ExpansionError during parsing /tool/yocto/poky/meta/recipes-kernel/linux/linux-yocto_3.2.bb: Failure expanding variable SRCPV
Hi,
On Jun 8, 2012, at 12:58 PM, Richard Purdie wrote:
> On Thu, 2012-06-07 at 15:59 -0400, Bruce Ashfield wrote:
>> Richard/Saul,
>>
>> Here's a 9 patch series that is a collection of work that's been
>> brewing for about three weeks.
>>
>> Highlights include:
>>
>> - removal of the 2.6.37 recipe, it's time is now gone
>> - addition of the 3.4 kernel recipe
>> - configuration fixups and audits within that 3.4 kernel
>> - kernel *only* support for mips64 mti emulation
>> - -stable updates for supported kernels 3.2.18, 3.0.32
>> - machine fixes/configurations for the 3.2 and 3.0 kernels
>> - kern tool updates, streamlining and fixes
I just updated my poky installation, and tried to build the meta-toolchain, and
I got a bunch of errors. This is related to the recent updates to the
linux-yocto_3.2.bb kernel.
elvis@eos:/tool/yocto/poky/build$ bitbake meta-toolchain
Pseudo is not present but is required, building this first before the main build
WARNING: Unable to get checksum for python-numpy SRC_URI entry config.h: file
could not be found | ETA:
00:00:43
WARNING: Unable to get checksum for python-numpy SRC_URI entry numpyconfig.h:
file could not be found
ERROR: ExpansionError during parsing
/tool/yocto/poky/meta/recipes-kernel/linux/linux-yocto_3.2.bb: Failure
expanding variable SRCPV, expression was ${@bb.fetch2.get_srcrev(d)} which
triggered exception FetchError: Fetcher failure for URL:
'git://git.yoctoproject.org/linux-yocto-3.2;protocol=git;bareclone=1;branch=standard/default/base,meta;name=machine,meta'.
The command git ls-remote git://git.yoctoproject.org/linux-yocto-3.2
ee78519365bdb25287703bbc31c06b193263c654l gave empty output unexpectedly
ERROR: No recipes available for:
/tool/yocto/poky/meta-yocto/recipes-core/uclibc/uclibc_git.bbappend
/tool/yocto/poky/meta-yocto/recipes-kernel/linux/linux-yocto_3.2.bbappend
/tool/yocto/meta-xilinx/recipes-core/util-linux/util-linux_2.21.1.bbappend
/tool/yocto/poky/meta-yocto/recipes-core/psplash/psplash_git.bbappend
/tool/yocto/poky/meta-yocto/recipes-core/netbase/netbase_4.47.bbappend
/tool/yocto/meta-openembedded/meta-oe/recipes-multimedia/gstreamer/gst-ffmpeg_0.10.13.bbappend
/tool/yocto/meta-openembedded/meta-oe/recipes-core/dropbear/dropbear_2012.55.bbappend
/tool/yocto/poky/meta-yocto/recipes-core/tasks/task-core-tools-profile.bbappend
/tool/yocto/meta-xilinx/recipes-core/initscripts/initscripts_1.0.bbappend
/tool/yocto/meta-openembedded/meta-oe/recipes-core/busybox/busybox_1.19.4.bbappend
ERROR: Command execution failed: Exited with 1
Summary: There were 2 WARNING messages shown.
Summary: There were 3 ERROR messages shown, returning a non-zero exit code.
best regards,
Elvis Dowson
___
Openembedded-core mailing list
Openembedded-core@lists.openembedded.org
http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
[OE-core] qemuppc does not respond to mouse input when running inside Ubuntu12.04 VMware image
Hi, I'm running all my Yocto builds using an Ubuntu 12.04 VMware image using VMware Fusion. My host operating system is Mac OS X 10.7.4. I just built a core-image-sato image, and ran it. It brings up the sato desktop, but I cannot interact with QEMU using the mouse. If I press ctrl+alt it releases the mouse grab, but if I click on the QEMU window, I can see the mouse cursor moving around, but it doesn't interact with the sato desktop. Is there a fix or workaround for this issue? Elvis Dowson ___ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
[OE-core] [meta-xilinx] Creating a QEMU for Xilinx ARM for Zynq 7020 device
Hi, I'd like to add support for qemuxarm and build a core-image-sato image, so that I can emulate the ARM processor on the Xilinx Zynq 7020 device using Yocto. The sources for qemuxarm are located below: http://git.xilinx.com/?p=qemu-xarm.git;a=summary I haven't done this before, so if someone could guide me as to what I should do, and the recipes I should take a look at, it would be great! Best regards, Elvis Dowson ___ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
[OE-core] [meta-xilinx] WARNING: QA Issue: systemd
lib/ld.so.1 (0xdead8000) libdl.so.2 => /tool/yocto/poky/build/tmp/sysroots/virtex5/lib/libdl.so.2 (0xdead9000) NOTE: package gdk-pixbuf-2.24.1-r3: task do_populate_lic: Succeeded NOTE: Running task 1422 of 2269 (ID: 1695, /tool/yocto/poky/meta/recipes-gnome/gdk-pixbuf/gdk-pixbuf_2.24.1.bb, do_package) WARNING: QA Issue: systemd: /tool/yocto/poky/build/tmp/work/ppc440-poky-linux/systemd-git-r27/packages-split/libsystemd-journal/lib/libsystemd-journal.so.0.0.3 links to something under exec_prefix WARNING: QA Issue: ldd reports: libsystemd-id128.so.0 => /tool/yocto/poky/build/tmp/sysroots/virtex5/lib/libsystemd-id128.so.0 (0xdead1000) libcap.so.2 => /tool/yocto/poky/build/tmp/sysroots/virtex5/lib/libcap.so.2 (0xdead2000) liblzma.so.5 => /tool/yocto/poky/build/tmp/sysroots/virtex5/usr/lib/liblzma.so.5 (0xdead3000) libdl.so.2 => /tool/yocto/poky/build/tmp/sysroots/virtex5/lib/libdl.so.2 (0xdead4000) librt.so.1 => /tool/yocto/poky/build/tmp/sysroots/virtex5/lib/librt.so.1 (0xdead5000) libpthread.so.0 => /tool/yocto/poky/build/tmp/sysroots/virtex5/lib/libpthread.so.0 (0xdead6000) libc.so.6 => /tool/yocto/poky/build/tmp/sysroots/virtex5/lib/libc.so.6 (0xdead7000) /lib/ld.so.1 => /tool/yocto/poky/build/tmp/sysroots/virtex5/lib/ld.so.1 (0xdead8000) Best regards, Elvis Dowson ___ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
[OE-core] [meta-xilinx] How to create a u-boot-xilinx_git.bb recipe for virtex5 and microblaze machine architectures
Hi,
I'm trying to make the
meta-xilinx/recipes-bsp/u-boot/u-boot-xilinx_git.bb recipe work for both the
virtex5 and microblaze machine architectures.
So far, I find that you can start off by simply prefixing the machine
architecture for a bunch of variables like
PV_virtex5
SRC_URI_virtex5
and so on.
I have two basic questions:
01. When setting PV_virtext5, should I be using SRCPV or SRCREV? What is the
difference between SRCPV and SRCREV?
PV_virtex5 = "v2012.04.01+git${SRCPV}"
02. When setting the SRCREV, which one should I use SRCREV_${PN}_virtex5 or
SRCREV_pn-${PN}_virtex5
SRCREV_${PN}_virtex5 = "415d386877df49eb051b85ef74fa59a16dc17c7d"
or
SRCREV_pn-${PN}_virtex5 = "415d386877df49eb051b85ef74fa59a16dc17c7d"
I'm looking at meta-ti/recipes-bsp/u-boot/u-boot_git.bb, and I see inconsistent
application, e.g. SRCREV_${PN} for omap3evm, and and SRCREV_pn-${PN} for the
rest like am37x-evm, etc, in that recipe. It's got me confused as to what the
proper way to set SRCREV is.
Best regards,
Elvis Dowson___
Openembedded-core mailing list
Openembedded-core@lists.openembedded.org
http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
Re: [OE-core] [meta-xilinx] How to debug the value of ${TARGET_ARCH} in meta-xilinx/classes/xilinx-boot.class
Hi Khem, On Jun 3, 2012, at 2:48 AM, Khem Raj wrote: > What's your default she I guess its not bash I have left it as the default Dash shell, as indicated in the yocto 1.2 setup guidelines, for Ubuntu-12.04 LTS. Best regards, Elvis Dowson ___ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
Re: [OE-core] [meta-xilinx] How to debug the value of ${TARGET_ARCH} in meta-xilinx/classes/xilinx-boot.class
Hi,
> On Jun 2, 2012, at 10:36 PM, Elvis Dowson wrote:
>
>> How can I debug the value of ${TARGET_ARCH} in
>> meta-xilinx/classes/xilinx-boot.class, for the following function:
>>
>> do_export_xparam() {
>> bbnote "Replacing xparameters header to match hardware model"
>> xparam=$1
>> if [ "${TARGET_ARCH}" == "powerpc" ]; then
>> cpu="PPC`echo ${TARGET_CPU} | tr '[:lower:]' '[:upper:]'`"
>> else
>> cpu=`echo ${TARGET_CPU} | tr '[:lower:]' '[:upper:]'`
>> fi
>> cp ${xparam} ${S}/board/xilinx/${XILINX_BOARD}
>> echo "/*** Cannonical definitions ***/
>> #define XPAR_PLB_CLOCK_FREQ_HZ XPAR_PROC_BUS_0_FREQ_HZ
>> #define XPAR_CORE_CLOCK_FREQ_HZ XPAR_CPU_${cpu}_CORE_CLOCK_FREQ_HZ
>> #ifndef XPAR_DDR2_SDRAM_MEM_BASEADDR
>> # define XPAR_DDR2_SDRAM_MEM_BASEADDR XPAR_DDR_SDRAM_MPMC_BASEADDR
>> #endif
>> #define XPAR_PCI_0_CLOCK_FREQ_HZ0" >>
>> ${S}/board/xilinx/${XILINX_BOARD}/xparameters.h
>> }
>
> In the above code ${TARGET_ARCH} always evaluates to "powerpc". I just
> inserted ${TARGET_ARCH} in the echo statements, and saw the value in the file.
>
> Now what I can't seem to figure out is why, for the following if condition,
> the first branch isn't being take; the else branch is always evaluated, and
> the value for cpu is always set to "440" instead of "PPC440"
>
> if [ "${TARGET_ARCH}" == "powerpc" ]; then
> cpu="PPC`echo ${TARGET_CPU} | tr '[:lower:]' '[:upper:]'`"
> else
> cpu="`echo ${TARGET_CPU} | tr '[:lower:]' '[:upper:]'`"
> fi
Changing it to use a single = , instead of == solved it, but it would help if
someone could explain why. From the link below
http://tldp.org/LDP/Bash-Beginners-Guide/html/sect_07_01.html
apparently, using a "=" is for strict posix compilance.
but I fail to see why when ${TARGET_ARCH} evaluates to "powerpc", and I do a
string comparision if [ "${TARGET_ARCH}" == "powerpc" ] it evaluates to false,
and if [ "${TARGET_ARCH}" = "powerpc" ] evaluates to true.
Best regards,
Elvis Dowson
___
Openembedded-core mailing list
Openembedded-core@lists.openembedded.org
http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
Re: [OE-core] [meta-xilinx] How to debug the value of ${TARGET_ARCH} in meta-xilinx/classes/xilinx-boot.class
Hi,
On Jun 2, 2012, at 10:36 PM, Elvis Dowson wrote:
> How can I debug the value of ${TARGET_ARCH} in
> meta-xilinx/classes/xilinx-boot.class, for the following function:
>
> do_export_xparam() {
> bbnote "Replacing xparameters header to match hardware model"
> xparam=$1
> if [ "${TARGET_ARCH}" == "powerpc" ]; then
> cpu="PPC`echo ${TARGET_CPU} | tr '[:lower:]' '[:upper:]'`"
> else
> cpu=`echo ${TARGET_CPU} | tr '[:lower:]' '[:upper:]'`
> fi
> cp ${xparam} ${S}/board/xilinx/${XILINX_BOARD}
> echo "/*** Cannonical definitions ***/
> #define XPAR_PLB_CLOCK_FREQ_HZ XPAR_PROC_BUS_0_FREQ_HZ
> #define XPAR_CORE_CLOCK_FREQ_HZ XPAR_CPU_${cpu}_CORE_CLOCK_FREQ_HZ
> #ifndef XPAR_DDR2_SDRAM_MEM_BASEADDR
> # define XPAR_DDR2_SDRAM_MEM_BASEADDR XPAR_DDR_SDRAM_MPMC_BASEADDR
> #endif
> #define XPAR_PCI_0_CLOCK_FREQ_HZ0" >>
> ${S}/board/xilinx/${XILINX_BOARD}/xparameters.h
> }
In the above code ${TARGET_ARCH} always evaluates to "powerpc". I just inserted
${TARGET_ARCH} in the echo statements, and saw the value in the file.
Now what I can't seem to figure out is why, for the following if condition, the
first branch isn't being take; the else branch is always evaluated, and the
value for cpu is always set to "440" instead of "PPC440"
if [ "${TARGET_ARCH}" == "powerpc" ]; then
cpu="PPC`echo ${TARGET_CPU} | tr '[:lower:]' '[:upper:]'`"
else
cpu="`echo ${TARGET_CPU} | tr '[:lower:]' '[:upper:]'`"
fi
Best regards,
Elvis Dowson
___
Openembedded-core mailing list
Openembedded-core@lists.openembedded.org
http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
[OE-core] [meta-xilinx] How to debug the value of ${TARGET_ARCH} in meta-xilinx/classes/xilinx-boot.class
Hi,
How can I debug the value of ${TARGET_ARCH} in
meta-xilinx/classes/xilinx-boot.class, for the following function:
do_export_xparam() {
bbnote "Replacing xparameters header to match hardware model"
xparam=$1
if [ "${TARGET_ARCH}" == "powerpc" ]; then
cpu="PPC`echo ${TARGET_CPU} | tr '[:lower:]' '[:upper:]'`"
else
cpu=`echo ${TARGET_CPU} | tr '[:lower:]' '[:upper:]'`
fi
cp ${xparam} ${S}/board/xilinx/${XILINX_BOARD}
echo "/*** Cannonical definitions ***/
#define XPAR_PLB_CLOCK_FREQ_HZ XPAR_PROC_BUS_0_FREQ_HZ
#define XPAR_CORE_CLOCK_FREQ_HZ XPAR_CPU_${cpu}_CORE_CLOCK_FREQ_HZ
#ifndef XPAR_DDR2_SDRAM_MEM_BASEADDR
# define XPAR_DDR2_SDRAM_MEM_BASEADDR XPAR_DDR_SDRAM_MPMC_BASEADDR
#endif
#define XPAR_PCI_0_CLOCK_FREQ_HZ0" >>
${S}/board/xilinx/${XILINX_BOARD}/xparameters.h
}
My machine is configured as virtex-5, and meta-xilinx/conf/machine/virtex5.conf
doesn't specify TARGET_ARCH, it just defines the following
TARGET_CPU = "440"
#tune for the 440 cpu
require conf/machine/include/tune-ppc440.inc
So is that a mistake?
At the moment, the value for cpu=440, instead of PPC440 for do_export_xparam().
Best regards,
Elvis Dowson
___
Openembedded-core mailing list
Openembedded-core@lists.openembedded.org
http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
Re: [OE-core] [meta-xilinx] Generating device tree for building linux-xilinx kernel
Hi Adrian, On Jun 1, 2012, at 11:18 PM, Elvis Dowson wrote: > I've created a hardware bsp project using Xilinx SDK 14.1, and used the > xilinx-ml507-update project, (after updating it to XPS 14.1 format). > > Now the script in meta-xilinx/classes/xilinx-kernel.bbclass is looking for a > *.dts file, but I don't see that being created anywhere, in my bsp projects. Ok, got past that error, the *.dts file doesn't get created if I choose Wind River GPP 2.0. If I switch the distro to Monta Vista 5.0, it generates a ml507.dts file. Now when I execute bitbake linux-xilinx, it picks up the ml507.dts file correctly and builds the linux image. NOTE: Running task 521 of 867 (ID: 6, /tool/yocto/meta-xilinx/recipes-kernel/linux/linux-xilinx_git.bb, do_configure) NOTE: package linux-xilinx-2.6.37.1+git1+86c81ccd7c8e895904bfff95bd4f87c9615f7fba-r15: task do_configure: Started NOTE: package linux-xilinx-2.6.37.1+git1+86c81ccd7c8e895904bfff95bd4f87c9615f7fba-r15: task do_configure: Succeeded NOTE: Running task 857 of 867 (ID: 7, /tool/yocto/meta-xilinx/recipes-kernel/linux/linux-xilinx_git.bb, do_compile) NOTE: package linux-xilinx-2.6.37.1+git1+86c81ccd7c8e895904bfff95bd4f87c9615f7fba-r15: task do_compile: Started NOTE: package linux-xilinx-2.6.37.1+git1+86c81ccd7c8e895904bfff95bd4f87c9615f7fba-r15: task do_compile: Succeeded NOTE: Running task 858 of 867 (ID: 10, /tool/yocto/meta-xilinx/recipes-kernel/linux/linux-xilinx_git.bb, do_compile_kernelmodules) NOTE: Running task 859 of 867 (ID: 9, /tool/yocto/meta-xilinx/recipes-kernel/linux/linux-xilinx_git.bb, do_sizecheck) NOTE: Running task 860 of 867 (ID: 8, /tool/yocto/meta-xilinx/recipes-kernel/linux/linux-xilinx_git.bb, do_uboot_mkimage) NOTE: package linux-xilinx-2.6.37.1+git1+86c81ccd7c8e895904bfff95bd4f87c9615f7fba-r15: task do_compile_kernelmodules: Started NOTE: package linux-xilinx-2.6.37.1+git1+86c81ccd7c8e895904bfff95bd4f87c9615f7fba-r15: task do_sizecheck: Started NOTE: package linux-xilinx-2.6.37.1+git1+86c81ccd7c8e895904bfff95bd4f87c9615f7fba-r15: task do_sizecheck: Succeeded NOTE: package linux-xilinx-2.6.37.1+git1+86c81ccd7c8e895904bfff95bd4f87c9615f7fba-r15: task do_uboot_mkimage: Started NOTE: package linux-xilinx-2.6.37.1+git1+86c81ccd7c8e895904bfff95bd4f87c9615f7fba-r15: task do_uboot_mkimage: Succeeded NOTE: package linux-xilinx-2.6.37.1+git1+86c81ccd7c8e895904bfff95bd4f87c9615f7fba-r15: task do_compile_kernelmodules: Succeeded NOTE: Running task 861 of 867 (ID: 2, /tool/yocto/meta-xilinx/recipes-kernel/linux/linux-xilinx_git.bb, do_install) NOTE: package linux-xilinx-2.6.37.1+git1+86c81ccd7c8e895904bfff95bd4f87c9615f7fba-r15: task do_install: Started NOTE: package linux-xilinx-2.6.37.1+git1+86c81ccd7c8e895904bfff95bd4f87c9615f7fba-r15: task do_install: Succeeded NOTE: Running task 862 of 867 (ID: 13, /tool/yocto/meta-xilinx/recipes-kernel/linux/linux-xilinx_git.bb, do_package) NOTE: Running task 863 of 867 (ID: 11, /tool/yocto/meta-xilinx/recipes-kernel/linux/linux-xilinx_git.bb, do_deploy) NOTE: Running task 864 of 867 (ID: 3, /tool/yocto/meta-xilinx/recipes-kernel/linux/linux-xilinx_git.bb, do_populate_sysroot) NOTE: package linux-xilinx-2.6.37.1+git1+86c81ccd7c8e895904bfff95bd4f87c9615f7fba-r15: task do_package: Started NOTE: package linux-xilinx-2.6.37.1+git1+86c81ccd7c8e895904bfff95bd4f87c9615f7fba-r15: task do_deploy: Started NOTE: package linux-xilinx-2.6.37.1+git1+86c81ccd7c8e895904bfff95bd4f87c9615f7fba-r15: task do_populate_sysroot: Started NOTE: package linux-xilinx-2.6.37.1+git1+86c81ccd7c8e895904bfff95bd4f87c9615f7fba-r15: task do_deploy: Succeeded NOTE: package linux-xilinx-2.6.37.1+git1+86c81ccd7c8e895904bfff95bd4f87c9615f7fba-r15: task do_populate_sysroot: Succeeded NOTE: package linux-xilinx-2.6.37.1+git1+86c81ccd7c8e895904bfff95bd4f87c9615f7fba-r15: task do_package: Succeeded NOTE: Running task 865 of 867 (ID: 15, /tool/yocto/meta-xilinx/recipes-kernel/linux/linux-xilinx_git.bb, do_package_write_rpm) NOTE: package linux-xilinx-2.6.37.1+git1+86c81ccd7c8e895904bfff95bd4f87c9615f7fba-r15: task do_package_write_rpm: Started NOTE: package linux-xilinx-2.6.37.1+git1+86c81ccd7c8e895904bfff95bd4f87c9615f7fba-r15: task do_package_write_rpm: Succeeded NOTE: Running noexec task 866 of 867 (ID: 12, /tool/yocto/meta-xilinx/recipes-kernel/linux/linux-xilinx_git.bb, do_package_write) NOTE: Running noexec task 867 of 867 (ID: 5, /tool/yocto/meta-xilinx/recipes-kernel/linux/linux-xilinx_git.bb, do_build) NOTE: Tasks Summary: Attempted 867 tasks of which 855 didn't need to be rerun and all succeeded. Best regards, Elvis Dowson ___ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
Re: [OE-core] [meta-xilinx] Generating device tree for building linux-xilinx kernel
Hi Adrian, On May 31, 2012, at 11:13 PM, Adrian Alonso wrote: >>> That's a second process that you need todo with xilinx EDK to export >>> and launch xilinx SDK >>> from there create a new device tree project which will parse the "mhs" >>> hardware descriptor file >>> and from there generate the device tree according to the hwd settings... I've created a hardware bsp project using Xilinx SDK 14.1, and used the xilinx-ml507-update project, (after updating it to XPS 14.1 format). Now the script in meta-xilinx/classes/xilinx-kernel.bbclass is looking for a *.dts file, but I don't see that being created anywhere, in my bsp projects. The XPS SDK project is generated into the following location from your XPS system.mhs project: /project/xilinx-sdk-workspace (which contains two subfolders linux_2_6_bsp_0, and xilinx-ml507_hw_platform) Your XPS project is located in /project/xilinx-ml507-updated and my local.conf settings are XILINX_BSP_PATH ?= "/project/xilinx-ml507-updated" is that the correct BSP path location? When I try to build the linux-xilinx recipe, I get the following errors: NOTE: package linux-xilinx-2.6.37.1+git1+86c81ccd7c8e895904bfff95bd4f87c9615f7fba-r15: task do_unpack: Started NOTE: package linux-xilinx-2.6.37.1+git1+86c81ccd7c8e895904bfff95bd4f87c9615f7fba-r15: task do_unpack: Succeeded NOTE: Running task 854 of 867 (ID: 1, /tool/yocto/meta-xilinx/recipes-kernel/linux/linux-xilinx_git.bb, do_patch) NOTE: package linux-xilinx-2.6.37.1+git1+86c81ccd7c8e895904bfff95bd4f87c9615f7fba-r15: task do_patch: Started NOTE: package linux-xilinx-2.6.37.1+git1+86c81ccd7c8e895904bfff95bd4f87c9615f7fba-r15: task do_patch: Succeeded NOTE: Running task 855 of 867 (ID: 6, /tool/yocto/meta-xilinx/recipes-kernel/linux/linux-xilinx_git.bb, do_configure) NOTE: package linux-xilinx-2.6.37.1+git1+86c81ccd7c8e895904bfff95bd4f87c9615f7fba-r15: task do_configure: Started ERROR: Function failed: do_configure (see /tool/yocto/poky/build/tmp/work/virtex5-poky-linux/linux-xilinx-2.6.37.1+git1+86c81ccd7c8e895904bfff95bd4f87c9615f7fba-r15/temp/log.do_configure.30845 for further information) ERROR: Logfile of failure stored in: /tool/yocto/poky/build/tmp/work/virtex5-poky-linux/linux-xilinx-2.6.37.1+git1+86c81ccd7c8e895904bfff95bd4f87c9615f7fba-r15/temp/log.do_configure.30845 Log data follows: | DEBUG: Executing python function sysroot_cleansstate | DEBUG: Python function sysroot_cleansstate finished | DEBUG: Executing shell function do_configure | ERROR: Function failed: do_configure (see /tool/yocto/poky/build/tmp/work/virtex5-poky-linux/linux-xilinx-2.6.37.1+git1+86c81ccd7c8e895904bfff95bd4f87c9615f7fba-r15/temp/log.do_configure.30845 for further information) | ERROR: No device tree found, missing hardware ref design? NOTE: package linux-xilinx-2.6.37.1+git1+86c81ccd7c8e895904bfff95bd4f87c9615f7fba-r15: task do_configure: Failed ERROR: Task 6 (/tool/yocto/meta-xilinx/recipes-kernel/linux/linux-xilinx_git.bb, do_configure) failed with exit code '1' NOTE: Tasks Summary: Attempted 856 tasks of which 852 didn't need to be rerun and 1 failed. Summary: 1 task failed: /tool/yocto/meta-xilinx/recipes-kernel/linux/linux-xilinx_git.bb, do_configure Summary: There was 1 ERROR message shown, returning a non-zero exit code. Best regards, Elvis Dowson ___ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
Re: [OE-core] [meta-xilinx] Generating device tree for building linux-xilinx kernel
Hi Adrian, On May 31, 2012, at 10:49 PM, Adrian Alonso wrote: > That's a second process that you need todo with xilinx EDK to export > and launch xilinx SDK > from there create a new device tree project which will parse the "mhs" > hardware descriptor file > and from there generate the device tree according to the hwd settings... Perhaps these steps could be automated by modifying the meta-xilinx/classes/xilinx-kernel.bbclass file, so that the EDK gets invoked from the command line to parse the mhs file and create the dts file? Best regards, Elvis Dowson ___ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
[OE-core] [meta-xilinx] Generating device tree for building linux-xilinx kernel
Hi Adrian, When I attempt to build the linux-xilinx kernel, it doesn't find the device tree (*.dts) in the xilinx-ml507-update BSP project. Is the device tree supposed to be generated manually? If so, could you please tell me how to generate it? Best regards, Elvis Dowson ___ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
[OE-core] [meta-xilinx] WARNING: linux-xilinx: No generic license file exists for: GPL in any provider
Hi, While attempting to build linux-xilinx, I get the following warning: WARNING: linux-xilinx: No generic license file exists for: GPL in any provider What should I do to fix this warning? Best regards, Elvis Dowson ___ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
[OE-core] [meta-xilinx] libpcre_8.21.bb: IOError: invalid Python installation
Hi, I noticed a situation where, when I try to bitbake linux-xilinx, it fails a libpcre, because of an invalid python instalation. However, after doing a bitbake -c clean linux-xilinx libpcre, and then rebuilding libpcre, python-native gets staged correctly and the build succeeds. Shouldn't python-native get built correctly the first time around? My parallelism options as follows: BB_NUMBER_THREADS = "6" PARALLEL_MAKE = "-j 6" Best regards, Elvis Dowson ___ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
Re: [OE-core] [meta-xilinx] WARNING: QA Issue: bash & shell scripts
Hi, On May 31, 2012, at 7:38 PM, Elvis Dowson wrote: > I've found the following warnings while attempting to build linux-xilinx. How > can I trace which package it's originating from and what can I do to fix it? > > WARNING: QA Issue: bash: Found a reference to /usr/ in > /tool/yocto/poky/build/tmp/work/ppc440-poky-linux/bash-4.2-r2/packages-split/bash/bin/bashbug > WARNING: QA Issue: Shell scripts in base_bindir and base_sbindir should not > reference anything in exec_prefix Sorry about that silly question, its originating from the bash-4.2 package. What should I do to fix it? Elvis Dowson ___ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
[OE-core] [meta-xilinx] WARNING: QA Issue: bash & shell scripts
Hi, I've found the following warnings while attempting to build linux-xilinx. How can I trace which package it's originating from and what can I do to fix it? WARNING: QA Issue: bash: Found a reference to /usr/ in /tool/yocto/poky/build/tmp/work/ppc440-poky-linux/bash-4.2-r2/packages-split/bash/bin/bashbug WARNING: QA Issue: Shell scripts in base_bindir and base_sbindir should not reference anything in exec_prefix Best regards, Elvis Dowson ___ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
Re: [OE-core] meta-xilinx: ERROR: Failure expanding variable TUNE_PKGARCH ppc440
Hi, On May 30, 2012, at 5:00 PM, Khem Raj wrote: > On Wed, May 30, 2012 at 7:47 AM, Elvis Dowson wrote: >> >> What does the -nf suffix do ? >> > > indicates no float What should I call the tune configuration file: tune-ppc440.inc or tune-ppc440-nf.inc ? For the initial build tests, I will first try to get this running without a hardware FPU. By default the embedded PowerPC 440 processor on the Xilinx Virtex-5 doesn't have a hardware FPU. But one can configure a LogiCORE IP Virtex-5 APU Floating-Point Unit (v1.01a) to get single or double precision FPU support. How would I handle file naming and configuration for these 3 possible variants (i.e. software floating point emulation, hardware FPU single precision & hardware FPU double precision)? Best regards, Elvis Dowson ___ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
Re: [OE-core] meta-xilinx: ERROR: Failure expanding variable TUNE_PKGARCH ppc440
Hi Khem,
> On Tue, May 29, 2012 at 9:56 PM, Khem Raj wrote:
>> On Tue, May 29, 2012 at 12:46 PM, Elvis Dowson
>> wrote:
>>> PACKAGE_EXTRA_ARCHS_tune-ppc440 = "${PACKAGE_EXTRA_ARCHS_tune-powerpc}
>>> ppc440"
>>
>> may be this should be be
>>
>> PACKAGE_EXTRA_ARCHS_tune-ppc440 = "${PACKAGE_EXTRA_ARCHS_tune-powerpc-nf}
>> ppc440"
>>
>> other pieces look good
What does the -nf suffix do ?
Best regards,
Elvis Dowson
___
Openembedded-core mailing list
Openembedded-core@lists.openembedded.org
http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
[OE-core] meta-xilinx: Which version of the linux kernel?
Hi Adrian,
Could you please tell me which version of the linux kernel,
the meta-xilinx layer was tested against, using ISE 12.1?
At the moment I get an error because no version number is specified for the
kernel recipe, so I think I have to rename it to something like say
linux-xilinx-3.2.bb or linux-xilinx-2.6.38.bb.
Here is an output of the errors that I get when trying to bitbake
meta-toolchain:
$ bitbake meta-toolchain
Pseudo is not present but is required, building this first before the main build
NOTE: Error expanding variable pkg_postinst_kernel-base
| ETA: --:--:--
NOTE: Error during finalise of
/tool/yocto/meta-xilinx/recipes-kernel/linux/linux-xilinx_git.bb
ERROR: Failure expanding variable KERNEL_PRIORITY, expression was
${@int(d.getVar('PV',1).split('-')[0].split('+')[0].split('.')[0]) * 1 +
int(d.getVar('PV',1).split('-')[0].split('+')[0].split('.')[1]) * 100 +
int(d.getVar('PV',1).split('-')[0].split('+')[0].split('.')[-1])} which
triggered exception ValueError: invalid literal for int() with base 10: 'git'
ERROR: Command execution failed: Traceback (most recent call last):
File "/tool/yocto/poky/bitbake/lib/bb/command.py", line 84, in runAsyncCommand
self.cooker.updateCache()
File "/tool/yocto/poky/bitbake/lib/bb/cooker.py", line 1202, in updateCache
if not self.parser.parse_next():
File "/tool/yocto/poky/bitbake/lib/bb/cooker.py", line 1671, in parse_next
self.virtuals += len(result)
UnboundLocalError: local variable 'result' referenced before assignment
Best regards,
Elvis Dowson
___
Openembedded-core mailing list
Openembedded-core@lists.openembedded.org
http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
[OE-core] meta-ti: ERROR: Unable to parse /tool/yocto/meta-ti/recipes-bsp/powervr-drivers/omap3-sgx-modules_4.05.00.03.bb
Hi,
When building with the latest meta-ti layer, I get the following error:
ERROR: Unable to analyse format of PR variable: b
| ETA: 00:01:18
ERROR: Unable to parse
/tool/yocto/meta-ti/recipes-bsp/powervr-drivers/omap3-sgx-modules_4.05.00.03.bb
Traceback (most recent call last):
File "/tool/yocto/poky/bitbake/lib/bb/parse/ast.py", line 381, in
multi_finalize(fn='/tool/yocto/meta-ti/recipes-bsp/powervr-drivers/omap3-sgx-modules_4.05.00.03.bb',
d=):
try:
>finalize(fn, d)
except bb.parse.SkipPackage as e:
File "/tool/yocto/poky/bitbake/lib/bb/parse/ast.py", line 323, in
finalize(fn='/tool/yocto/meta-ti/recipes-bsp/powervr-drivers/omap3-sgx-modules_4.05.00.03.bb',
d=, variant=None):
code.append("%s(d)" % funcname)
>bb.utils.simple_exec("\n".join(code), {"d": d})
bb.data.update_data(d)
File "/tool/yocto/poky/bitbake/lib/bb/utils.py", line 274, in
simple_exec(code='__anon_119__tool_yocto_poky_meta_classes_staging_bbclass(d)\n__anon_540__tool_yocto_poky_meta_classes_base_bbclass(d)\n__anon_176__tool_yocto_poky_meta_classes_package_bbclass(d)\n__anon_1106__tool_yocto_poky_meta_classes_package_rpm_bbclass(d)\n__anon_736__tool_yocto_poky_meta_classes_insane_bbclass(d)\n__anon_20__tool_yocto_poky_meta_classes_debian_bbclass(d)\n__anon_52__tool_yocto_poky_meta_classes_sstate_bbclass(d)',
context={'d': }):
def simple_exec(code, context):
>exec(code, _context, context)
File "", line 2, in
File "__anon_540__tool_yocto_poky_meta_classes_base_bbclass", line 75, in
__anon_540__tool_yocto_poky_meta_classes_base_bbclass(d=)
AttributeError: 'NoneType' object has no attribute 'group'
ERROR: Command execution failed: Traceback (most recent call last):
File "/tool/yocto/poky/bitbake/lib/bb/command.py", line 84, in runAsyncCommand
self.cooker.updateCache()
File "/tool/yocto/poky/bitbake/lib/bb/cooker.py", line 1202, in updateCache
if not self.parser.parse_next():
File "/tool/yocto/poky/bitbake/lib/bb/cooker.py", line 1671, in parse_next
self.virtuals += len(result)
UnboundLocalError: local variable 'result' referenced before assignment
best regards,
Elvis Dowson
___
Openembedded-core mailing list
Openembedded-core@lists.openembedded.org
http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
Re: [OE-core] meta-xilinx: ERROR: Failure expanding variable TUNE_PKGARCH ppc440
Hi Mark,
On May 29, 2012, at 8:36 PM, Mark Hatle wrote:
>
> The error above is an issue w/ the tune configuration. In stock oe-core
> there is no 405/440 configuration.
>
> You will need to create a new tune file, see
> meta/conf/machine/include/tune-ppc603e.inc as a reasonable example..
I modified the poky/meta/conf/machine/include/tune-ppc440.inc as follows:
DEFAULTTUNE ?= "ppc440"
require conf/machine/include/powerpc/arch-powerpc.inc
TUNEVALID[ppc440] = "Enable ppc440 specific processor optimizations"
TUNE_CCARGS += "${@bb.utils.contains("TUNE_FEATURES", "ppc440", "-mcpu=440",
"", d)}"
AVAILTUNES += "ppc440"
TUNE_FEATURES_tune-ppc440 = "m32 fpu-soft ppc440"
TUNE_PKGARCH_tune-ppc603e = "ppc440"
PACKAGE_EXTRA_ARCHS_tune-ppc440 = "${PACKAGE_EXTRA_ARCHS_tune-powerpc} ppc440"
# glibc configure options to get 440 specific library (for sqrt)
GLIBC_EXTRA_OECONF += "${@bb.utils.contains("TUNE_FEATURES", "ppc440",
"-with-cpu=440", "", d)}"
This is for the embedded PowerPC440 processor on the Virtex-5 FX70T FPGA chip,
which doesn't have a hardware FPU. Do the options look okay: "m32 fpu-soft
ppc440" ?
Is there a resource I can refer to see a full list of applicable PowerPC440
gcc-4.x compiler options?
Best regards,
Elvis Dowson
___
Openembedded-core mailing list
Openembedded-core@lists.openembedded.org
http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
[OE-core] meta-xilinx: ERROR: Failure expanding variable TUNE_PKGARCH ppc440
Hi,
I've added the meta-xilinx layer, and modified my build/conf/local.conf
file as follows:
# Set the target machine details
MACHINE ?= "virtex5"
# Set Xilinx Platform Studio hardware project path
XILINX_BSP_PATH ?= "/project/xilinx-ml507"
# Set target board
XILINX_BOARD ?= "ml507"
and when I try to build the meta-toolchain
$ cd /tool/yocto/poky
$ source oe-init-build-env build
$ bitbake meta-toolchain
I get the following errors:
ERROR: Execution of event handler 'check_sanity_eventhandler' failed
Traceback (most recent call last):
File "/tool/yocto/poky/bitbake/lib/bb/utils.py", line 397, in
better_eval(source= at 0x344b6b0, file "MACHINE_ARCH",
line 1>, locals={'d': }):
def better_eval(source, locals):
>return eval(source, _context, locals)
File "MACHINE_ARCH", line 1, in
File "/tool/yocto/poky/bitbake/lib/bb/data_smart.py", line 282, in
DataSmart.getVar(var='TUNE_PKGARCH', expand=True, noweakdefault=False):
if expand and value:
>return self.expand(value, var)
return value
File "/tool/yocto/poky/bitbake/lib/bb/data_smart.py", line 151, in
DataSmart.expand(s='${@bb.utils.contains("TUNE_FEATURES", "ppc440", "ppc440",
"${PPCPKGARCH}", d)}', varname='TUNE_PKGARCH'):
def expand(self, s, varname = None):
>return self.expandWithRefs(s, varname).value
File "/tool/yocto/poky/bitbake/lib/bb/data_smart.py", line 141, in
DataSmart.expandWithRefs(s='${@bb.utils.contains("TUNE_FEATURES", "ppc440",
"ppc440", "${PPCPKGARCH}", d)}', varname='TUNE_PKGARCH'):
except Exception as exc:
>raise ExpansionError(varname, s, exc)
ExpansionError: Failure expanding variable TUNE_PKGARCH, expression was
${@bb.utils.contains("TUNE_FEATURES", "ppc440", "ppc440", "${PPCPKGARCH}", d)}
which triggered exception SyntaxError: EOL while scanning string literal
(TUNE_PKGARCH, line 1)
ERROR: Failure expanding variable TUNE_PKGARCH, expression was
${@bb.utils.contains("TUNE_FEATURES", "ppc440", "ppc440", "${PPCPKGARCH}", d)}
which triggered exception SyntaxError: EOL while scanning string literal
(TUNE_PKGARCH, line 1)
ERROR: Command execution failed: Exited with 1
Summary: There were 3 ERROR messages shown, returning a non-zero exit code.
Has support for the PowerPC440 processor (present in the Virtex-5 FPGA) been
removed from the newer gcc-versions?
Best regards,
Elvis Dowson___
Openembedded-core mailing list
Openembedded-core@lists.openembedded.org
http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
Re: [OE-core] How to fix warning: set SYSTEMD_PACKAGES as -systemd
Hi Denys, On May 4, 2012, at 8:27 PM, Denys Dmytriyenko wrote: > You are trying to fix it in oe-core/poky layer, where it doesn't use systemd. > There is a .bbappend in meta-oe for that. Yes, you're right! :-) Ok, I'm slowly starting to see how its important to trace any additional .bbappend recipes. > For these 2 please report them on openembedded-devel mailing list, where > meta-oe things are being discussed. The last warning is confisuing, as it > reports the recipe is in poky/oe-core, but it actually comes from the > corresponding .bbappend in meta-oe. I'll submit these two patches to the openembedded-devel mailing list, once I go through the sutmitting patches README. :-) God, the things you have to do just to submit a simple patch! ;-) Best regards, Elvis ___ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
Re: [OE-core] How to fix warning: set SYSTEMD_PACKAGES as -systemd
Hi Koen,
On May 4, 2012, at 11:46 AM, Koen Kooi wrote:
> When doing things like that you need to bump PR and fixup all dependents like
> tasks and images to ensure upgrade paths.
I bumped the PR, but still get the warning message. How is the warning for this
specific package linked to its dependents. The warning looks localized to the
lighttpd_1.4.30.bb recipe.
If you can tell me how to fix it, I'll do it and send a patch.
diff --git a/meta/recipes-extended/lighttpd/lighttpd_1.4.30.bb
b/meta/recipes-extended/lighttpd/lighttpd_1.4.30.
index 3ae3867..71ae64c 100644
--- a/meta/recipes-extended/lighttpd/lighttpd_1.4.30.bb
+++ b/meta/recipes-extended/lighttpd/lighttpd_1.4.30.bb
@@ -16,7 +16,7 @@ RDEPENDS_${PN} += " \
lighttpd-module-staticfile \
"
-PR = "r0"
+PR = "r1"
SRC_URI =
"http://download.lighttpd.net/lighttpd/releases-1.4.x/lighttpd-${PV}.tar.bz2 \
file://index.html.lighttpd \
@@ -51,6 +51,9 @@ do_install_append() {
install -m 0644 ${WORKDIR}/index.html.lighttpd ${D}/www/pages/index.html
}
+SYSTEMD_PACKAGES = "${PN}-systemd"
+SYSTEMD_SERVICE = "lighttpd.service"
+
FILES_${PN} += "${sysconfdir} /www"
CONFFILES_${PN} = "${sysconfdir}/lighttpd.conf"
Elvis Dowson
___
Openembedded-core mailing list
Openembedded-core@lists.openembedded.org
http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
Re: [OE-core] [meta-openembedded] [PATCH] networkmanager: silence systemd bbclass WARNING.
Hi Denys,
On May 4, 2012, at 7:33 PM, Denys Dmytriyenko wrote:
>>
>> -SYSTEMD_PACKAGES = "${PN}"
>> +SYSTEMD_PACKAGES = "${PN}-systemd"
>
> That might be an overkill. I'm personally not yet very familiar with all the
> systemd specifics and not clear when to create ${PN}-systemd subpackage and
> when to just silence the warning...
>
Me neither :-) ,... but this is what the warning message, suggested be done.
It fixed the warning for networkmanager, though.
Best regards,
Elvis
___
Openembedded-core mailing list
Openembedded-core@lists.openembedded.org
http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
Re: [OE-core] How to fix warning: set SYSTEMD_PACKAGES as -systemd
Hi,
I've submitted a patch to fix the warning for networkmanager_0.9.2.0.bb.
However, I still get the warning for lighttpd_1.4.30.bb
even after adding the following line:
SYSTEMD_PACKAGES = "${PN}-systemd"
or both these two lines in the recipe:
SYSTEMD_PACKAGES = "${PN}-systemd"
SYSTEMD_SERVICE = "lighttpd.service"
WARNING: /tool/yocto/poky/meta/recipes-extended/lighttpd/lighttpd_1.4.30.bb: it
is recommended to set SYSTEMD_PACKAGES as -systemd
Best regards,
Elvis Dowson
___
Openembedded-core mailing list
Openembedded-core@lists.openembedded.org
http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
[OE-core] [meta-openembedded] [PATCH] networkmanager: silence systemd bbclass WARNING.
This patch fixes WARNING: /tool/yocto/meta-openembedded/meta-oe/
recipes-connectivity/networkmanager/networkmanager_0.9.2.0.bb:
it is recommended to set SYSTEMD_PACKAGES as -systemd
Signed-off-by: Elvis Dowson
---
.../networkmanager/networkmanager_0.9.2.0.bb |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git
a/meta-oe/recipes-connectivity/networkmanager/networkmanager_0.9.2.0.bb
b/meta-oe/recipes-connectivity/networkmanager/networkmanager_0.9.2.0.bb
index 474d98a..478d61e 100644
--- a/meta-oe/recipes-connectivity/networkmanager/networkmanager_0.9.2.0.bb
+++ b/meta-oe/recipes-connectivity/networkmanager/networkmanager_0.9.2.0.bb
@@ -54,7 +54,7 @@ do_install_append () {
install -d ${D}/etc/NetworkManager/VPN
}
-SYSTEMD_PACKAGES = "${PN}"
+SYSTEMD_PACKAGES = "${PN}-systemd"
SYSTEMD_SERVICE = "NetworkManager.service"
PACKAGES =+ "libnmutil libnmglib libnmglib-vpn ${PN}-tests"
--
1.7.9.5
___
Openembedded-core mailing list
Openembedded-core@lists.openembedded.org
http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
Re: [OE-core] How to fix warning: set SYSTEMD_PACKAGES as -systemd
Hi Martin, On May 4, 2012, at 10:02 AM, Martin Jansa wrote: > > Check this thread > http://www.mail-archive.com/openembedded-devel@lists.openembedded.org/msg24963.html The above thread only discusses the warning, but no-one has fixed it yet. If you can tell me what needs to be done, I'll create a patch and post it back to the list. Best regards, Elvis Dowson___ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
[OE-core] How to fix warning: set SYSTEMD_PACKAGES as -systemd
Hi, I get the following warnings when I try to do bitbake core-image-minimal WARNING: /tool/yocto/meta-openembedded/meta-oe/recipes-connectivity/networkmanager/networkmanager_0.9.2.0.bb: it is recommended to set SYSTEMD_PACKAGES as -systemd WARNING: /tool/yocto/poky/meta/recipes-extended/lighttpd/lighttpd_1.4.30.bb: it is recommended to set SYSTEMD_PACKAGES as -systemd What is the appropriate change to make these two warnings go away? Best regards, Elvis Dowson___ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
