Message ID | 1450978394-822-1-git-send-email-sergio.prado@e-labworks.com |
---|---|
State | Changes Requested |
Headers | show |
Dear Sergio Prado, On Thu, 24 Dec 2015 15:33:14 -0200, Sergio Prado wrote: > VFPv3 and NEON was enabled and a readme.txt file was added. > > Fixes bug #7580. > > Tested on Pandaboard ES. > > Signed-off-by: Sergio Prado <sergio.prado@e-labworks.com> Thanks a lot for this update. I have a number of comments, see below. > --- > diff --git a/board/pandaboard/readme.txt b/board/pandaboard/readme.txt > new file mode 100644 > index 000000000000..7a0e1660c0cb > --- /dev/null > +++ b/board/pandaboard/readme.txt > @@ -0,0 +1,74 @@ > +Pandaboard > +==================== Nit: excess "=====" signs. > + > +This file documents the Buildroot support for the Pandaboard, a > +low-power, low-cost single-board computer development platform based > +on the Texas Instruments OMAP4 system on a chip (SoC). > + > + One too many empty new line. > +Configuring and building Buildroot > +---------------------------------- > + > +Start from the defconfig: > + > + $ make pandaboard_defconfig > + > +You can edit build options the usual way: > + > + $ make menuconfig > + > +When you are happy with the setup, run: > + > + $ make > + > +The result of the build with the default settings should be these files: > + > + output/images > + ├── MLO > + ├── omap4-panda-a4.dtb > + ├── omap4-panda.dtb > + ├── omap4-panda-es.dtb > + ├── rootfs.ext2 > + ├── u-boot.img > + └── zImage > + > + Ditto. > +Setting up your SD card > +----------------------- > + > +*Important*: pay attention which partition you are modifying so you don't > +accidentally erase the wrong file system, e.g your host computer or your > +external storage! > + > +In the default setup you need to create two partitions on your SD card: > +a boot partition and a rootfs partition. > + > +The sdcard needs to be formatted in an specific way. You can use the > +shell script below to format the sdcard (this script was extracted > +from http://elinux.org/Panda_How_to_MLO_%26_u-boot). > + > +#!/bin/sh > +if [ ! "$1" = "/dev/sda" ] ; then I think this check is silly. I know it is in the original script, but it is still silly. > + DRIVE=$1 > + if [ -b "$DRIVE" ] ; then > + dd if=/dev/zero of=$DRIVE bs=1024 count=1024 > + SIZE=`fdisk -l $DRIVE | grep Disk | awk '{print $5}'` > + echo DISK SIZE - $SIZE bytes > + CYLINDERS=`echo $SIZE/255/63/512 | bc` > + echo CYLINDERS - $CYLINDERS > + { > + echo ,9,0x0C,* > + echo ,,,- > + } | sfdisk -D -H 255 -S 63 -C $CYLINDERS $DRIVE > + mkfs.vfat -F 32 -n "boot" ${DRIVE}1 > + mke2fs -j -L "rootfs" ${DRIVE}2 > + fi > +fi Could you instead replace this with a genimage configuration file ? You can look at the support for the following boards for examples: ./board/via/imx6_vab820/genimage.cfg ./board/intel/galileo/genimage.cfg ./board/wandboard/genimage.cfg > diff --git a/configs/pandaboard_defconfig b/configs/pandaboard_defconfig > index 50c3714d84d9..53870b0e1f8d 100644 > --- a/configs/pandaboard_defconfig > +++ b/configs/pandaboard_defconfig > @@ -1,39 +1,19 @@ > -# Architecture > BR2_arm=y > BR2_cortex_a9=y > - > -# system > -BR2_TARGET_GENERIC_GETTY=y > +BR2_ARM_ENABLE_NEON=y > +BR2_ARM_FPU_VFPV3=y > BR2_TARGET_GENERIC_GETTY_PORT="ttyO2" > -BR2_ROOTFS_DEVICE_CREATION_DYNAMIC_DEVTMPFS=y > - > -# filesystem > -BR2_TARGET_ROOTFS_EXT2=y > -# BR2_TARGET_ROOTFS_TAR is not set > - > -# Lock to 3.12 headers to avoid breaking with newer kernels > -BR2_KERNEL_HEADERS_VERSION=y > -BR2_DEFAULT_KERNEL_VERSION="3.12.2" > -BR2_PACKAGE_HOST_LINUX_HEADERS_CUSTOM_3_12=y You need to keep something like this to lock the kernel headers version to a version matching the kernel version being used. > - > -# Kernel > BR2_LINUX_KERNEL=y > -BR2_LINUX_KERNEL_CUSTOM_VERSION=y > -BR2_LINUX_KERNEL_CUSTOM_VERSION_VALUE="3.12.2" You also need to keep something like this to lock the kernel version being used. Right now, you're using the "latest" kernel, which is not what we want, as we want defconfigs to use a fixed kernel and bootloader version, which have been tested by the submitter of the defconfig. > BR2_LINUX_KERNEL_DEFCONFIG="omap2plus" > -BR2_LINUX_KERNEL_UIMAGE_LOADADDR="0x80008000" > -BR2_LINUX_KERNEL_APPENDED_UIMAGE=y > -BR2_LINUX_KERNEL_APPENDED_DTB=y > -BR2_LINUX_KERNEL_USE_INTREE_DTS=y > -# might need omap4-panda or omap4-panda-a4 instead > -BR2_LINUX_KERNEL_INTREE_DTS_NAME="omap4-panda-es" > - > -# Bootloaders > +BR2_LINUX_KERNEL_ZIMAGE=y > +BR2_LINUX_KERNEL_DTS_SUPPORT=y > +BR2_LINUX_KERNEL_INTREE_DTS_NAME="omap4-panda-es omap4-panda omap4-panda-a4" > +BR2_LINUX_KERNEL_INSTALL_TARGET=y > +BR2_TARGET_ROOTFS_EXT2=y In your report, you said that the failure was caused by the filesystem being ext2, but the bootloader trying to mount it as ext3. And now you're producing an ext2. Is it because the bootloader default kernel arguments have changed? > +# BR2_TARGET_ROOTFS_TAR is not set > BR2_TARGET_UBOOT=y You also need to lock the bootloader version, like was done before. Could you rework those points and submit an updated version? Thanks a lot! Thomas
Hi Thomas, 2015-12-27 9:06 GMT-02:00 Thomas Petazzoni < thomas.petazzoni@free-electrons.com>: > > Dear Sergio Prado, > > On Thu, 24 Dec 2015 15:33:14 -0200, Sergio Prado wrote: > > VFPv3 and NEON was enabled and a readme.txt file was added. > > > > Fixes bug #7580. > > > > Tested on Pandaboard ES. > > > > Signed-off-by: Sergio Prado <sergio.prado@e-labworks.com> > > Thanks a lot for this update. I have a number of comments, see below. > > > --- > > diff --git a/board/pandaboard/readme.txt b/board/pandaboard/readme.txt > > new file mode 100644 > > index 000000000000..7a0e1660c0cb > > --- /dev/null > > +++ b/board/pandaboard/readme.txt > > @@ -0,0 +1,74 @@ > > +Pandaboard > > +==================== > > Nit: excess "=====" signs. OK. I'll remove them. > > > + > > +This file documents the Buildroot support for the Pandaboard, a > > +low-power, low-cost single-board computer development platform based > > +on the Texas Instruments OMAP4 system on a chip (SoC). > > + > > + > > One too many empty new line. OK. > > > +Configuring and building Buildroot > > +---------------------------------- > > + > > +Start from the defconfig: > > + > > + $ make pandaboard_defconfig > > + > > +You can edit build options the usual way: > > + > > + $ make menuconfig > > + > > +When you are happy with the setup, run: > > + > > + $ make > > + > > +The result of the build with the default settings should be these files: > > + > > + output/images > > + ├── MLO > > + ├── omap4-panda-a4.dtb > > + ├── omap4-panda.dtb > > + ├── omap4-panda-es.dtb > > + ├── rootfs.ext2 > > + ├── u-boot.img > > + └── zImage > > + > > + > > Ditto. OK. > > > +Setting up your SD card > > +----------------------- > > + > > +*Important*: pay attention which partition you are modifying so you don't > > +accidentally erase the wrong file system, e.g your host computer or your > > +external storage! > > + > > +In the default setup you need to create two partitions on your SD card: > > +a boot partition and a rootfs partition. > > + > > +The sdcard needs to be formatted in an specific way. You can use the > > +shell script below to format the sdcard (this script was extracted > > +from http://elinux.org/Panda_How_to_MLO_%26_u-boot). > > + > > +#!/bin/sh > > +if [ ! "$1" = "/dev/sda" ] ; then > > I think this check is silly. I know it is in the original script, but > it is still silly. > > > + DRIVE=$1 > > + if [ -b "$DRIVE" ] ; then > > + dd if=/dev/zero of=$DRIVE bs=1024 count=1024 > > + SIZE=`fdisk -l $DRIVE | grep Disk | awk '{print $5}'` > > + echo DISK SIZE - $SIZE bytes > > + CYLINDERS=`echo $SIZE/255/63/512 | bc` > > + echo CYLINDERS - $CYLINDERS > > + { > > + echo ,9,0x0C,* > > + echo ,,,- > > + } | sfdisk -D -H 255 -S 63 -C $CYLINDERS $DRIVE > > + mkfs.vfat -F 32 -n "boot" ${DRIVE}1 > > + mke2fs -j -L "rootfs" ${DRIVE}2 > > + fi > > +fi > > Could you instead replace this with a genimage configuration file ? You > can look at the support for the following boards for examples: > > ./board/via/imx6_vab820/genimage.cfg > ./board/intel/galileo/genimage.cfg > ./board/wandboard/genimage.cfg That was my first attempt, but it didn't work. The ROM code from OMAP processors (and I think also Sitara processors) need the SD card to be formatted with a special geometry in the partition table. That's the reason of the script above. I did not find any option in the genimage tool to configure thinks like sectors and cylinders. So I do not see an easy way to generate a sdcard image in this case. What do you think? > > > diff --git a/configs/pandaboard_defconfig b/configs/pandaboard_defconfig > > index 50c3714d84d9..53870b0e1f8d 100644 > > --- a/configs/pandaboard_defconfig > > +++ b/configs/pandaboard_defconfig > > @@ -1,39 +1,19 @@ > > -# Architecture > > BR2_arm=y > > BR2_cortex_a9=y > > - > > -# system > > -BR2_TARGET_GENERIC_GETTY=y > > +BR2_ARM_ENABLE_NEON=y > > +BR2_ARM_FPU_VFPV3=y > > BR2_TARGET_GENERIC_GETTY_PORT="ttyO2" > > -BR2_ROOTFS_DEVICE_CREATION_DYNAMIC_DEVTMPFS=y > > - > > -# filesystem > > -BR2_TARGET_ROOTFS_EXT2=y > > -# BR2_TARGET_ROOTFS_TAR is not set > > - > > -# Lock to 3.12 headers to avoid breaking with newer kernels > > -BR2_KERNEL_HEADERS_VERSION=y > > -BR2_DEFAULT_KERNEL_VERSION="3.12.2" > > -BR2_PACKAGE_HOST_LINUX_HEADERS_CUSTOM_3_12=y > > You need to keep something like this to lock the kernel headers version > to a version matching the kernel version being used. Right. My bad. > > > - > > -# Kernel > > BR2_LINUX_KERNEL=y > > -BR2_LINUX_KERNEL_CUSTOM_VERSION=y > > -BR2_LINUX_KERNEL_CUSTOM_VERSION_VALUE="3.12.2" > > You also need to keep something like this to lock the kernel version > being used. Right now, you're using the "latest" kernel, which is not > what we want, as we want defconfigs to use a fixed kernel and > bootloader version, which have been tested by the submitter of the > defconfig. OK. > > > BR2_LINUX_KERNEL_DEFCONFIG="omap2plus" > > -BR2_LINUX_KERNEL_UIMAGE_LOADADDR="0x80008000" > > -BR2_LINUX_KERNEL_APPENDED_UIMAGE=y > > -BR2_LINUX_KERNEL_APPENDED_DTB=y > > -BR2_LINUX_KERNEL_USE_INTREE_DTS=y > > -# might need omap4-panda or omap4-panda-a4 instead > > -BR2_LINUX_KERNEL_INTREE_DTS_NAME="omap4-panda-es" > > - > > -# Bootloaders > > +BR2_LINUX_KERNEL_ZIMAGE=y > > +BR2_LINUX_KERNEL_DTS_SUPPORT=y > > +BR2_LINUX_KERNEL_INTREE_DTS_NAME="omap4-panda-es omap4-panda omap4-panda-a4" > > +BR2_LINUX_KERNEL_INSTALL_TARGET=y > > +BR2_TARGET_ROOTFS_EXT2=y > > In your report, you said that the failure was caused by the filesystem > being ext2, but the bootloader trying to mount it as ext3. And now > you're producing an ext2. Is it because the bootloader default kernel > arguments have changed? Yes, the old bootloader's default kernel arguments had a "rootfstype=ext3" option, and the new one doesn't, so the kernel is able to find out and mount any supported filesystem. I kept the ext2 format because most of the boards use it. Is this the right approach, or should I change to ext4? > > > +# BR2_TARGET_ROOTFS_TAR is not set > > BR2_TARGET_UBOOT=y > > You also need to lock the bootloader version, like was done before. Right. > > Could you rework those points and submit an updated version? Sure. I'll just wait for your comments to send an updated version of the patch. > > Thanks a lot! Thanks! > > Thomas > -- > Thomas Petazzoni, CTO, Free Electrons > Embedded Linux, Kernel and Android engineering > http://free-electrons.com
Sergio, On Tue, 29 Dec 2015 10:21:52 -0200, Sergio Prado wrote: > > Could you instead replace this with a genimage configuration file ? You > > can look at the support for the following boards for examples: > > > > ./board/via/imx6_vab820/genimage.cfg > > ./board/intel/galileo/genimage.cfg > > ./board/wandboard/genimage.cfg > > That was my first attempt, but it didn't work. The ROM code from OMAP > processors (and I think also Sitara processors) need the SD card to be > formatted with a special geometry in the partition table. That's the reason > of the script above. I did not find any option in the genimage tool to > configure thinks like sectors and cylinders. So I do not see an easy way to > generate a sdcard image in this case. What do you think? Ah ok. I don't know if genimage can handle such scenarios. I am Cc'ing Vivien Didelot, who did quite a bit of work on genimage in Buildroot. I think I remember he telling me that he had a working genimage configuration for the BeagleBone, which has, AFAIK, the same requirements as the PandaBoard in terms of geometry and all. > Yes, the old bootloader's default kernel arguments had a "rootfstype=ext3" > option, and the new one doesn't, so the kernel is able to find out and > mount any supported filesystem. I kept the ext2 format because most of the > boards use it. Is this the right approach, or should I change to ext4? No, it's alright to keep ext2. Thanks! Thomas
Hello, On Tue, 29 Dec 2015 14:00:21 +0100, Thomas Petazzoni wrote: > > That was my first attempt, but it didn't work. The ROM code from OMAP > > processors (and I think also Sitara processors) need the SD card to be > > formatted with a special geometry in the partition table. That's the reason > > of the script above. I did not find any option in the genimage tool to > > configure thinks like sectors and cylinders. So I do not see an easy way to > > generate a sdcard image in this case. What do you think? > > Ah ok. I don't know if genimage can handle such scenarios. I am Cc'ing > Vivien Didelot, who did quite a bit of work on genimage in Buildroot. I > think I remember he telling me that he had a working genimage > configuration for the BeagleBone, which has, AFAIK, the same > requirements as the PandaBoard in terms of geometry and all. I found Vivien's initial patch for the BeagleBone: http://lists.busybox.net/pipermail/buildroot/2015-March/122406.html Though I don't see where it handles the specific geometry requirements. Maybe those requirements have been relaxed in the AM335x compared to the OMAP4 ? Best regards, Thomas
Hello Thomas, 2015-12-29 11:01 GMT-02:00 Thomas Petazzoni < thomas.petazzoni@free-electrons.com>: > > Hello, > > On Tue, 29 Dec 2015 14:00:21 +0100, Thomas Petazzoni wrote: > > > > That was my first attempt, but it didn't work. The ROM code from OMAP > > > processors (and I think also Sitara processors) need the SD card to be > > > formatted with a special geometry in the partition table. That's the reason > > > of the script above. I did not find any option in the genimage tool to > > > configure thinks like sectors and cylinders. So I do not see an easy way to > > > generate a sdcard image in this case. What do you think? > > > > Ah ok. I don't know if genimage can handle such scenarios. I am Cc'ing > > Vivien Didelot, who did quite a bit of work on genimage in Buildroot. I > > think I remember he telling me that he had a working genimage > > configuration for the BeagleBone, which has, AFAIK, the same > > requirements as the PandaBoard in terms of geometry and all. > > I found Vivien's initial patch for the BeagleBone: > > http://lists.busybox.net/pipermail/buildroot/2015-March/122406.html > > Though I don't see where it handles the specific geometry requirements. > Maybe those requirements have been relaxed in the AM335x compared to > the OMAP4 ? I've just confirmed that the Beaglebone Black has the same geometry requirements. The generated sdcard image works in this case because the processor's ROM code will by default load U-Boot from the internal eMMC, that will read the images (kernel, dtb, rootfs) from the external sdcard. So if you have U-Boot in the internal eMMC, it will work. But if you try to boot U-Boot from external sdcard (holding S2) it won't work. Maybe we can think of trying to add in the genimage tool the feature of configuring the geometry in the generated image. Any thoughts Vivien? In the meantime, can we keep the previous approach of just documenting the process of generating the image in the readme.txt? > > Best regards, Thanks, > > Thomas > -- > Thomas Petazzoni, CTO, Free Electrons > Embedded Linux, Kernel and Android engineering > http://free-electrons.com Sergio Prado Embedded Labworks Office: +55 11 2628-3461 Mobile: +55 11 97123-3420
Dear Sergio Prado, On Tue, 29 Dec 2015 15:14:27 -0200, Sergio Prado wrote: > I've just confirmed that the Beaglebone Black has the same geometry > requirements. > > The generated sdcard image works in this case because the processor's ROM > code will by default load U-Boot from the internal eMMC, that will read the > images (kernel, dtb, rootfs) from the external sdcard. So if you have > U-Boot in the internal eMMC, it will work. > > But if you try to boot U-Boot from external sdcard (holding S2) it won't > work. Ah, not good then. > In the meantime, can we keep the previous approach of just documenting the > process of generating the image in the readme.txt? Yes, definitely. Thanks! Thomas
>>>>> "Thomas" == Thomas Petazzoni <thomas.petazzoni@free-electrons.com> writes: Hi, > Hello, > On Tue, 29 Dec 2015 14:00:21 +0100, Thomas Petazzoni wrote: >> > That was my first attempt, but it didn't work. The ROM code from OMAP >> > processors (and I think also Sitara processors) need the SD card to be >> > formatted with a special geometry in the partition table. That's the reason >> > of the script above. I did not find any option in the genimage tool to >> > configure thinks like sectors and cylinders. So I do not see an easy way to >> > generate a sdcard image in this case. What do you think? >> >> Ah ok. I don't know if genimage can handle such scenarios. I am Cc'ing >> Vivien Didelot, who did quite a bit of work on genimage in Buildroot. I >> think I remember he telling me that he had a working genimage >> configuration for the BeagleBone, which has, AFAIK, the same >> requirements as the PandaBoard in terms of geometry and all. > I found Vivien's initial patch for the BeagleBone: > http://lists.busybox.net/pipermail/buildroot/2015-March/122406.html > Though I don't see where it handles the specific geometry requirements. > Maybe those requirements have been relaxed in the AM335x compared to > the OMAP4 ? Yes, I believe so, I atleast haven't seen the issuess with more recent TI components. The boot problems on omap3 (and 4?) were afaik a combination of boot rom issues/erratas (and therefore depending on chip revisions) and sw limitations (E.G. early xloader versions had various limitations). https://groups.google.com/forum/#!topic/beagleboard/ro5k5r4Cuq4 http://elinux.org/Panda_How_to_MLO_%26_u-boot I do have an early pandaboard somewhere that I can try to dig out and do some tests with.
diff --git a/board/pandaboard/readme.txt b/board/pandaboard/readme.txt new file mode 100644 index 000000000000..7a0e1660c0cb --- /dev/null +++ b/board/pandaboard/readme.txt @@ -0,0 +1,74 @@ +Pandaboard +==================== + +This file documents the Buildroot support for the Pandaboard, a +low-power, low-cost single-board computer development platform based +on the Texas Instruments OMAP4 system on a chip (SoC). + + +Configuring and building Buildroot +---------------------------------- + +Start from the defconfig: + + $ make pandaboard_defconfig + +You can edit build options the usual way: + + $ make menuconfig + +When you are happy with the setup, run: + + $ make + +The result of the build with the default settings should be these files: + + output/images + ├── MLO + ├── omap4-panda-a4.dtb + ├── omap4-panda.dtb + ├── omap4-panda-es.dtb + ├── rootfs.ext2 + ├── u-boot.img + └── zImage + + +Setting up your SD card +----------------------- + +*Important*: pay attention which partition you are modifying so you don't +accidentally erase the wrong file system, e.g your host computer or your +external storage! + +In the default setup you need to create two partitions on your SD card: +a boot partition and a rootfs partition. + +The sdcard needs to be formatted in an specific way. You can use the +shell script below to format the sdcard (this script was extracted +from http://elinux.org/Panda_How_to_MLO_%26_u-boot). + +#!/bin/sh +if [ ! "$1" = "/dev/sda" ] ; then + DRIVE=$1 + if [ -b "$DRIVE" ] ; then + dd if=/dev/zero of=$DRIVE bs=1024 count=1024 + SIZE=`fdisk -l $DRIVE | grep Disk | awk '{print $5}'` + echo DISK SIZE - $SIZE bytes + CYLINDERS=`echo $SIZE/255/63/512 | bc` + echo CYLINDERS - $CYLINDERS + { + echo ,9,0x0C,* + echo ,,,- + } | sfdisk -D -H 255 -S 63 -C $CYLINDERS $DRIVE + mkfs.vfat -F 32 -n "boot" ${DRIVE}1 + mke2fs -j -L "rootfs" ${DRIVE}2 + fi +fi + +The next step is to mount the sdcard's first partition and copy MLO +and u-boot.img to it. + +The last step is to copy the rootfs image to the sdcard's second +partition using 'dd': + + $ sudo dd if=output/images/rootfs.ext2 of=/dev/sdX2 bs=1M conv=fsync diff --git a/configs/pandaboard_defconfig b/configs/pandaboard_defconfig index 50c3714d84d9..53870b0e1f8d 100644 --- a/configs/pandaboard_defconfig +++ b/configs/pandaboard_defconfig @@ -1,39 +1,19 @@ -# Architecture BR2_arm=y BR2_cortex_a9=y - -# system -BR2_TARGET_GENERIC_GETTY=y +BR2_ARM_ENABLE_NEON=y +BR2_ARM_FPU_VFPV3=y BR2_TARGET_GENERIC_GETTY_PORT="ttyO2" -BR2_ROOTFS_DEVICE_CREATION_DYNAMIC_DEVTMPFS=y - -# filesystem -BR2_TARGET_ROOTFS_EXT2=y -# BR2_TARGET_ROOTFS_TAR is not set - -# Lock to 3.12 headers to avoid breaking with newer kernels -BR2_KERNEL_HEADERS_VERSION=y -BR2_DEFAULT_KERNEL_VERSION="3.12.2" -BR2_PACKAGE_HOST_LINUX_HEADERS_CUSTOM_3_12=y - -# Kernel BR2_LINUX_KERNEL=y -BR2_LINUX_KERNEL_CUSTOM_VERSION=y -BR2_LINUX_KERNEL_CUSTOM_VERSION_VALUE="3.12.2" BR2_LINUX_KERNEL_DEFCONFIG="omap2plus" -BR2_LINUX_KERNEL_UIMAGE_LOADADDR="0x80008000" -BR2_LINUX_KERNEL_APPENDED_UIMAGE=y -BR2_LINUX_KERNEL_APPENDED_DTB=y -BR2_LINUX_KERNEL_USE_INTREE_DTS=y -# might need omap4-panda or omap4-panda-a4 instead -BR2_LINUX_KERNEL_INTREE_DTS_NAME="omap4-panda-es" - -# Bootloaders +BR2_LINUX_KERNEL_ZIMAGE=y +BR2_LINUX_KERNEL_DTS_SUPPORT=y +BR2_LINUX_KERNEL_INTREE_DTS_NAME="omap4-panda-es omap4-panda omap4-panda-a4" +BR2_LINUX_KERNEL_INSTALL_TARGET=y +BR2_TARGET_ROOTFS_EXT2=y +# BR2_TARGET_ROOTFS_TAR is not set BR2_TARGET_UBOOT=y +BR2_TARGET_UBOOT_BUILD_SYSTEM_KCONFIG=y +BR2_TARGET_UBOOT_BOARD_DEFCONFIG="omap4_panda" BR2_TARGET_UBOOT_FORMAT_IMG=y -BR2_TARGET_UBOOT_BOARDNAME="omap4_panda" BR2_TARGET_UBOOT_SPL=y BR2_TARGET_UBOOT_SPL_NAME="MLO" -BR2_TARGET_UBOOT_LATEST_VERSION=n -BR2_TARGET_UBOOT_CUSTOM_VERSION=y -BR2_TARGET_UBOOT_CUSTOM_VERSION_VALUE="2012.07"
VFPv3 and NEON was enabled and a readme.txt file was added. Fixes bug #7580. Tested on Pandaboard ES. Signed-off-by: Sergio Prado <sergio.prado@e-labworks.com> --- board/pandaboard/readme.txt | 74 ++++++++++++++++++++++++++++++++++++++++++++ configs/pandaboard_defconfig | 40 ++++++------------------ 2 files changed, 84 insertions(+), 30 deletions(-) create mode 100644 board/pandaboard/readme.txt