| Message ID | n5mdi8$5go$1@ger.gmane.org |
|---|---|
| State | Not Applicable |
| Headers | show |
Thanks again, Chris. I was sitting here thinking whether or not I should bug you when you email arrived. 'make linux-menuconfig' fails because: BR2_LINUX_KERNEL_PATCH="$(TOPDIR)/../beaglebone_buildroot_files/Patch_files/linux-4.1.13-ti-r36" Is there anything I need to put in that directory? On Sat, Dec 26, 2015, at 08:00 AM, Chris LaRocque wrote: > Craig Swank wrote: > > > Here is another thing I've found: > > > > https://github.com/beagleboard/linux/issues/49 > > > > In that issue the person had some addresses set incorrectly in uEnv.txt, > > but mine doesn't have any addresses set: > > > > bootpart=0:1 > > bootdir= > > uenvcmd=run loadimage;run loadramdisk;run findfdt;run loadfdt;run ram > > boot > > > > Is there something that needs to be set in uEnv.txt? > > > > > > On Wed, Dec 23, 2015, at 02:52 PM, Thomas Petazzoni wrote: > >> Dear Craig Swank, > >> > >> On Wed, 23 Dec 2015 14:45:36 -0800, Craig Swank wrote: > >> > Is it strange to you that when I do the build the output/build > >> > directory has both of these dirs? > >> > > >> > ➜ buildroot git:(master) ✗ ls output/build/linux-* > >> > output/build/linux-4.1.4-ti-r9 > >> > output/build/linux-headers-3.12.10 > >> > >> No, it is perfectly fine. linux-headers is used when building the > >> toolchain. linux is used to build the kernel itself. > >> > >> Thomas > >> -- > >> Thomas Petazzoni, CTO, Free Electrons > >> Embedded Linux, Kernel and Android engineering > >> http://free-electrons.com > > > > > Hello Craig > > Hopefully the attachments will be accepted. > > The kernel defconfig from beagleboard.org: > > 1> doesn't set DEPRECATED_PARAM_STRUCT [=y] in the kernel .config. > uEnv.txt > (uboot) passes in the console config as "console=ttyO0,115200n8" > > 2> doesn't set EXT2_FS [=y] and the default buildroot .config FS image is > ext2. at least for the beaglebone black and I'm trying to keep things > sane. > > You'll want to do a make source and them make linux-menuconfig and change > the parameters above. > > I haven't created a patch file because I'm lazy. > > Also attached is my patch (hack) for dtc, derived from Robert Nelsons > bb.org-overlays. > > The dtc patch file goes into BR2_GLOBAL_PATCH_DIR , > BR2_GLOBAL_PATCH_DIR"/dtc" > > Look in the buildroot .config for how I set BR2_GLOBAL_PATCH_DIR. you'll > need to point that to your own global patch directory. > > Regards > > Chris > _______________________________________________ > buildroot mailing list > buildroot@busybox.net > http://lists.busybox.net/mailman/listinfo/buildroot > Email had 2 attachments: > + host-dtc-1.4.1.patch > 33k (text/x-patch) > + cleaned_buildroot_v41_12182015_01.config > 61k (text/plain)
My build succeeded, but I'm having trouble with booting now. When I
boot with the uEnv.txt you pasted:
bootfile=zImage
fdtfile=am335x-boneblack.dtb
loadaddr=0x80007fc0
fdtaddr=0x80F80000
loadfdt=fatload mmc 0:1 ${fdtaddr} ${fdtfile}
loaduimage=fatload mmc 0:1 ${loadaddr} ${bootfile}
console=ttyO0,115200n8
mmcroot=/dev/mmcblk0p2
mmcrootfstype=ext2
uenvcmd=mmc rescan; run loaduimage; run loadfdt; run fdtboot
fdtboot=run mmc_args; run mmcargs; bootz ${loadaddr} - ${fdtaddr}
mmc_args=setenv bootargs console=${console} ${optargs} root=${mmcroot}
rw
rootfstype=${mmcrootfstype} root wait
I got an error about mmcargs not being defined. The fdtboot line has
both "run mmc_args" and "run mmcargs". I tried removing the 'run
mmcargs' part so now the uEnv.txt looks like:
bootfile=zImage
fdtfile=am335x-boneblack.dtb
loadaddr=0x80007fc0
fdtaddr=0x80F80000
loadfdt=fatload mmc 0:1 ${fdtaddr} ${fdtfile}
loaduimage=fatload mmc 0:1 ${loadaddr} ${bootfile}
console=ttyO0,115200n8
mmcroot=/dev/mmcblk0p2
mmcrootfstype=ext2
uenvcmd=mmc rescan; run loaduimage; run loadfdt; run fdtboot
fdtboot=run mmc_args; bootz ${loadaddr} - ${fdtaddr}
mmc_args=setenv bootargs console=${console} ${optargs} root=${mmcroot}
rw
rootfstype=${mmcrootfstype} root wait
Next, when I boot with the new uEnv.txt file the error I get is:
U-Boot 2015.10 (Dec 26 2015 - 10:23:47 -0700)
Watchdog enabled
I2C: ready
DRAM: 512 MiB
NAND: 0 MiB
MMC: OMAP SD/MMC: 0, OMAP SD/MMC: 1
*** Error - No Valid Environment Area found
*** Warning - bad CRC, using default environment
Net: <ethaddr> not set. Validating first E-fuse MAC
cpsw, usb_ether
Hit any key to stop autoboot: 0
switch to partitions #0, OK
mmc0 is current device
SD/MMC found on device 0
reading boot.scr
** Unable to read file boot.scr **
reading uEnv.txt
471 bytes read in 3 ms (153.3 KiB/s)
Loaded environment from uEnv.txt
Importing environment from mmc ...
Running uenvcmd ...
reading zImage
7462256 bytes read in 404 ms (17.6 MiB/s)
reading am335x-boneblack.dtb
59295 bytes read in 10 ms (5.7 MiB/s)
Kernel image @ 0x80007fc0 [ 0x000000 - 0x71dd70 ]
data abort
pc : [<9ff62ef8>] lr : [<9ff62f55>]
reloc pc : [<80810ef8>] lr : [<80810f55>]
sp : 9ef31778 ip : 80725d30 fp : 9ef34f74
r10: 00000002 r9 : 9ef31ed8 r8 : 9ffa35ac
r7 : 9ffa35b0 r6 : 00000003 r5 : 9ef34f74 r4 : 9ffa3514
r3 : 00000000 r2 : 9ffa3514 r1 : 9ef34f74 r0 : 00000000
Flags: nZcv IRQs off FIQs on Mode SVC_32
Resetting CPU ...
resetting ...
I really need to learn more about u-boot, but google searches suggest
that it means the loadaddr or fdtaddr is incorrect but I'm unable to
find a solution. Any suggestions?
I appreciate you working with me, Chris. I'm obsessed with getting this
to work and I feel I'm getting close.
On Sat, Dec 26, 2015, at 08:00 AM, Chris LaRocque wrote:
> Craig Swank wrote:
>
> > Here is another thing I've found:
> >
> > https://github.com/beagleboard/linux/issues/49
> >
> > In that issue the person had some addresses set incorrectly in uEnv.txt,
> > but mine doesn't have any addresses set:
> >
> > bootpart=0:1
> > bootdir=
> > uenvcmd=run loadimage;run loadramdisk;run findfdt;run loadfdt;run ram
> > boot
> >
> > Is there something that needs to be set in uEnv.txt?
> >
> >
> > On Wed, Dec 23, 2015, at 02:52 PM, Thomas Petazzoni wrote:
> >> Dear Craig Swank,
> >>
> >> On Wed, 23 Dec 2015 14:45:36 -0800, Craig Swank wrote:
> >> > Is it strange to you that when I do the build the output/build
> >> > directory has both of these dirs?
> >> >
> >> > ➜ buildroot git:(master) ✗ ls output/build/linux-*
> >> > output/build/linux-4.1.4-ti-r9
> >> > output/build/linux-headers-3.12.10
> >>
> >> No, it is perfectly fine. linux-headers is used when building the
> >> toolchain. linux is used to build the kernel itself.
> >>
> >> Thomas
> >> --
> >> Thomas Petazzoni, CTO, Free Electrons
> >> Embedded Linux, Kernel and Android engineering
> >> http://free-electrons.com
> >
> >
> Hello Craig
>
> Hopefully the attachments will be accepted.
>
> The kernel defconfig from beagleboard.org:
>
> 1> doesn't set DEPRECATED_PARAM_STRUCT [=y] in the kernel .config.
> uEnv.txt
> (uboot) passes in the console config as "console=ttyO0,115200n8"
>
> 2> doesn't set EXT2_FS [=y] and the default buildroot .config FS image is
> ext2. at least for the beaglebone black and I'm trying to keep things
> sane.
>
> You'll want to do a make source and them make linux-menuconfig and change
> the parameters above.
>
> I haven't created a patch file because I'm lazy.
>
> Also attached is my patch (hack) for dtc, derived from Robert Nelsons
> bb.org-overlays.
>
> The dtc patch file goes into BR2_GLOBAL_PATCH_DIR ,
> BR2_GLOBAL_PATCH_DIR"/dtc"
>
> Look in the buildroot .config for how I set BR2_GLOBAL_PATCH_DIR. you'll
> need to point that to your own global patch directory.
>
> Regards
>
> Chris
> _______________________________________________
> buildroot mailing list
> buildroot@busybox.net
> http://lists.busybox.net/mailman/listinfo/buildroot
> Email had 2 attachments:
> + host-dtc-1.4.1.patch
> 33k (text/x-patch)
> + cleaned_buildroot_v41_12182015_01.config
> 61k (text/plain)
Hi Craig, On Sat, Dec 26, 2015 at 1:34 PM, Craig Swank <craigswank@fastmail.fm> wrote: > My build succeeded, but I'm having trouble with booting now. When I > boot with the uEnv.txt you pasted: > > bootfile=zImage > fdtfile=am335x-boneblack.dtb > loadaddr=0x80007fc0 > fdtaddr=0x80F80000 > loadfdt=fatload mmc 0:1 ${fdtaddr} ${fdtfile} > loaduimage=fatload mmc 0:1 ${loadaddr} ${bootfile} > console=ttyO0,115200n8 > mmcroot=/dev/mmcblk0p2 > mmcrootfstype=ext2 > uenvcmd=mmc rescan; run loaduimage; run loadfdt; run fdtboot > fdtboot=run mmc_args; run mmcargs; bootz ${loadaddr} - ${fdtaddr} > mmc_args=setenv bootargs console=${console} ${optargs} root=${mmcroot} > rw > rootfstype=${mmcrootfstype} root wait > > I got an error about mmcargs not being defined. The fdtboot line has > both "run mmc_args" and "run mmcargs". I tried removing the 'run > mmcargs' part so now the uEnv.txt looks like: > > bootfile=zImage > fdtfile=am335x-boneblack.dtb > loadaddr=0x80007fc0 > fdtaddr=0x80F80000 > loadfdt=fatload mmc 0:1 ${fdtaddr} ${fdtfile} > loaduimage=fatload mmc 0:1 ${loadaddr} ${bootfile} > console=ttyO0,115200n8 > mmcroot=/dev/mmcblk0p2 > mmcrootfstype=ext2 > uenvcmd=mmc rescan; run loaduimage; run loadfdt; run fdtboot > fdtboot=run mmc_args; bootz ${loadaddr} - ${fdtaddr} > mmc_args=setenv bootargs console=${console} ${optargs} root=${mmcroot} > rw > rootfstype=${mmcrootfstype} root wait > > > Next, when I boot with the new uEnv.txt file the error I get is: > > U-Boot 2015.10 (Dec 26 2015 - 10:23:47 -0700) > > Watchdog enabled > I2C: ready > DRAM: 512 MiB > NAND: 0 MiB > MMC: OMAP SD/MMC: 0, OMAP SD/MMC: 1 > *** Error - No Valid Environment Area found > *** Warning - bad CRC, using default environment > > Net: <ethaddr> not set. Validating first E-fuse MAC > cpsw, usb_ether > Hit any key to stop autoboot: 0 > switch to partitions #0, OK > mmc0 is current device > SD/MMC found on device 0 > reading boot.scr > ** Unable to read file boot.scr ** > reading uEnv.txt > 471 bytes read in 3 ms (153.3 KiB/s) > Loaded environment from uEnv.txt > Importing environment from mmc ... > Running uenvcmd ... > reading zImage > 7462256 bytes read in 404 ms (17.6 MiB/s) > reading am335x-boneblack.dtb > 59295 bytes read in 10 ms (5.7 MiB/s) > Kernel image @ 0x80007fc0 [ 0x000000 - 0x71dd70 ] > data abort > pc : [<9ff62ef8>] lr : [<9ff62f55>] > reloc pc : [<80810ef8>] lr : [<80810f55>] > sp : 9ef31778 ip : 80725d30 fp : 9ef34f74 > r10: 00000002 r9 : 9ef31ed8 r8 : 9ffa35ac > r7 : 9ffa35b0 r6 : 00000003 r5 : 9ef34f74 r4 : 9ffa3514 > r3 : 00000000 r2 : 9ffa3514 r1 : 9ef34f74 r0 : 00000000 > Flags: nZcv IRQs off FIQs on Mode SVC_32 > Resetting CPU ... > > resetting ... > > I really need to learn more about u-boot, but google searches suggest > that it means the loadaddr or fdtaddr is incorrect but I'm unable to > find a solution. Any suggestions? I ran into this issue too. My workaround was to specify a custom uboot version and use the 2015.07 version instead. I didn't have time to debug what had changed, so I'm curious as well what the right solution is. Frank > > I appreciate you working with me, Chris. I'm obsessed with getting this > to work and I feel I'm getting close. > > > > > On Sat, Dec 26, 2015, at 08:00 AM, Chris LaRocque wrote: >> Craig Swank wrote: >> >> > Here is another thing I've found: >> > >> > https://github.com/beagleboard/linux/issues/49 >> > >> > In that issue the person had some addresses set incorrectly in uEnv.txt, >> > but mine doesn't have any addresses set: >> > >> > bootpart=0:1 >> > bootdir= >> > uenvcmd=run loadimage;run loadramdisk;run findfdt;run loadfdt;run ram >> > boot >> > >> > Is there something that needs to be set in uEnv.txt? >> > >> > >> > On Wed, Dec 23, 2015, at 02:52 PM, Thomas Petazzoni wrote: >> >> Dear Craig Swank, >> >> >> >> On Wed, 23 Dec 2015 14:45:36 -0800, Craig Swank wrote: >> >> > Is it strange to you that when I do the build the output/build >> >> > directory has both of these dirs? >> >> > >> >> > ➜ buildroot git:(master) ✗ ls output/build/linux-* >> >> > output/build/linux-4.1.4-ti-r9 >> >> > output/build/linux-headers-3.12.10 >> >> >> >> No, it is perfectly fine. linux-headers is used when building the >> >> toolchain. linux is used to build the kernel itself. >> >> >> >> Thomas >> >> -- >> >> Thomas Petazzoni, CTO, Free Electrons >> >> Embedded Linux, Kernel and Android engineering >> >> http://free-electrons.com >> > >> > >> Hello Craig >> >> Hopefully the attachments will be accepted. >> >> The kernel defconfig from beagleboard.org: >> >> 1> doesn't set DEPRECATED_PARAM_STRUCT [=y] in the kernel .config. >> uEnv.txt >> (uboot) passes in the console config as "console=ttyO0,115200n8" >> >> 2> doesn't set EXT2_FS [=y] and the default buildroot .config FS image is >> ext2. at least for the beaglebone black and I'm trying to keep things >> sane. >> >> You'll want to do a make source and them make linux-menuconfig and change >> the parameters above. >> >> I haven't created a patch file because I'm lazy. >> >> Also attached is my patch (hack) for dtc, derived from Robert Nelsons >> bb.org-overlays. >> >> The dtc patch file goes into BR2_GLOBAL_PATCH_DIR , >> BR2_GLOBAL_PATCH_DIR"/dtc" >> >> Look in the buildroot .config for how I set BR2_GLOBAL_PATCH_DIR. you'll >> need to point that to your own global patch directory. >> >> Regards >> >> Chris >> _______________________________________________ >> buildroot mailing list >> buildroot@busybox.net >> http://lists.busybox.net/mailman/listinfo/buildroot >> Email had 2 attachments: >> + host-dtc-1.4.1.patch >> 33k (text/x-patch) >> + cleaned_buildroot_v41_12182015_01.config >> 61k (text/plain) > > > -- > Craig Swank > craigswank@fastmail.fm > _______________________________________________ > buildroot mailing list > buildroot@busybox.net > http://lists.busybox.net/mailman/listinfo/buildroot
YES! Using Chris's uEnv.txt and u-boot 2015.07 instead of 2015.10 it finally booted up correctly. Many thanks to Thomas, Chris, and, last but not least, Frank, who swooped in with the secret sauce at the very end. I'm not even close to being able to figure out why the 07 version worked and 10 did not, but I'm just too happy to care. Craig On Sat, Dec 26, 2015, at 03:02 PM, Frank Hunleth wrote: > Hi Craig, > > On Sat, Dec 26, 2015 at 1:34 PM, Craig Swank <craigswank@fastmail.fm> > wrote: > > My build succeeded, but I'm having trouble with booting now. When I > > boot with the uEnv.txt you pasted: > > > > bootfile=zImage > > fdtfile=am335x-boneblack.dtb > > loadaddr=0x80007fc0 > > fdtaddr=0x80F80000 > > loadfdt=fatload mmc 0:1 ${fdtaddr} ${fdtfile} > > loaduimage=fatload mmc 0:1 ${loadaddr} ${bootfile} > > console=ttyO0,115200n8 > > mmcroot=/dev/mmcblk0p2 > > mmcrootfstype=ext2 > > uenvcmd=mmc rescan; run loaduimage; run loadfdt; run fdtboot > > fdtboot=run mmc_args; run mmcargs; bootz ${loadaddr} - ${fdtaddr} > > mmc_args=setenv bootargs console=${console} ${optargs} root=${mmcroot} > > rw > > rootfstype=${mmcrootfstype} root wait > > > > I got an error about mmcargs not being defined. The fdtboot line has > > both "run mmc_args" and "run mmcargs". I tried removing the 'run > > mmcargs' part so now the uEnv.txt looks like: > > > > bootfile=zImage > > fdtfile=am335x-boneblack.dtb > > loadaddr=0x80007fc0 > > fdtaddr=0x80F80000 > > loadfdt=fatload mmc 0:1 ${fdtaddr} ${fdtfile} > > loaduimage=fatload mmc 0:1 ${loadaddr} ${bootfile} > > console=ttyO0,115200n8 > > mmcroot=/dev/mmcblk0p2 > > mmcrootfstype=ext2 > > uenvcmd=mmc rescan; run loaduimage; run loadfdt; run fdtboot > > fdtboot=run mmc_args; bootz ${loadaddr} - ${fdtaddr} > > mmc_args=setenv bootargs console=${console} ${optargs} root=${mmcroot} > > rw > > rootfstype=${mmcrootfstype} root wait > > > > > > Next, when I boot with the new uEnv.txt file the error I get is: > > > > U-Boot 2015.10 (Dec 26 2015 - 10:23:47 -0700) > > > > Watchdog enabled > > I2C: ready > > DRAM: 512 MiB > > NAND: 0 MiB > > MMC: OMAP SD/MMC: 0, OMAP SD/MMC: 1 > > *** Error - No Valid Environment Area found > > *** Warning - bad CRC, using default environment > > > > Net: <ethaddr> not set. Validating first E-fuse MAC > > cpsw, usb_ether > > Hit any key to stop autoboot: 0 > > switch to partitions #0, OK > > mmc0 is current device > > SD/MMC found on device 0 > > reading boot.scr > > ** Unable to read file boot.scr ** > > reading uEnv.txt > > 471 bytes read in 3 ms (153.3 KiB/s) > > Loaded environment from uEnv.txt > > Importing environment from mmc ... > > Running uenvcmd ... > > reading zImage > > 7462256 bytes read in 404 ms (17.6 MiB/s) > > reading am335x-boneblack.dtb > > 59295 bytes read in 10 ms (5.7 MiB/s) > > Kernel image @ 0x80007fc0 [ 0x000000 - 0x71dd70 ] > > data abort > > pc : [<9ff62ef8>] lr : [<9ff62f55>] > > reloc pc : [<80810ef8>] lr : [<80810f55>] > > sp : 9ef31778 ip : 80725d30 fp : 9ef34f74 > > r10: 00000002 r9 : 9ef31ed8 r8 : 9ffa35ac > > r7 : 9ffa35b0 r6 : 00000003 r5 : 9ef34f74 r4 : 9ffa3514 > > r3 : 00000000 r2 : 9ffa3514 r1 : 9ef34f74 r0 : 00000000 > > Flags: nZcv IRQs off FIQs on Mode SVC_32 > > Resetting CPU ... > > > > resetting ... > > > > I really need to learn more about u-boot, but google searches suggest > > that it means the loadaddr or fdtaddr is incorrect but I'm unable to > > find a solution. Any suggestions? > > I ran into this issue too. My workaround was to specify a custom uboot > version and use the 2015.07 version instead. I didn't have time to > debug what had changed, so I'm curious as well what the right solution > is. > > Frank > > > > > I appreciate you working with me, Chris. I'm obsessed with getting this > > to work and I feel I'm getting close. > > > > > > > > > > On Sat, Dec 26, 2015, at 08:00 AM, Chris LaRocque wrote: > >> Craig Swank wrote: > >> > >> > Here is another thing I've found: > >> > > >> > https://github.com/beagleboard/linux/issues/49 > >> > > >> > In that issue the person had some addresses set incorrectly in uEnv.txt, > >> > but mine doesn't have any addresses set: > >> > > >> > bootpart=0:1 > >> > bootdir= > >> > uenvcmd=run loadimage;run loadramdisk;run findfdt;run loadfdt;run ram > >> > boot > >> > > >> > Is there something that needs to be set in uEnv.txt? > >> > > >> > > >> > On Wed, Dec 23, 2015, at 02:52 PM, Thomas Petazzoni wrote: > >> >> Dear Craig Swank, > >> >> > >> >> On Wed, 23 Dec 2015 14:45:36 -0800, Craig Swank wrote: > >> >> > Is it strange to you that when I do the build the output/build > >> >> > directory has both of these dirs? > >> >> > > >> >> > ➜ buildroot git:(master) ✗ ls output/build/linux-* > >> >> > output/build/linux-4.1.4-ti-r9 > >> >> > output/build/linux-headers-3.12.10 > >> >> > >> >> No, it is perfectly fine. linux-headers is used when building the > >> >> toolchain. linux is used to build the kernel itself. > >> >> > >> >> Thomas > >> >> -- > >> >> Thomas Petazzoni, CTO, Free Electrons > >> >> Embedded Linux, Kernel and Android engineering > >> >> http://free-electrons.com > >> > > >> > > >> Hello Craig > >> > >> Hopefully the attachments will be accepted. > >> > >> The kernel defconfig from beagleboard.org: > >> > >> 1> doesn't set DEPRECATED_PARAM_STRUCT [=y] in the kernel .config. > >> uEnv.txt > >> (uboot) passes in the console config as "console=ttyO0,115200n8" > >> > >> 2> doesn't set EXT2_FS [=y] and the default buildroot .config FS image is > >> ext2. at least for the beaglebone black and I'm trying to keep things > >> sane. > >> > >> You'll want to do a make source and them make linux-menuconfig and change > >> the parameters above. > >> > >> I haven't created a patch file because I'm lazy. > >> > >> Also attached is my patch (hack) for dtc, derived from Robert Nelsons > >> bb.org-overlays. > >> > >> The dtc patch file goes into BR2_GLOBAL_PATCH_DIR , > >> BR2_GLOBAL_PATCH_DIR"/dtc" > >> > >> Look in the buildroot .config for how I set BR2_GLOBAL_PATCH_DIR. you'll > >> need to point that to your own global patch directory. > >> > >> Regards > >> > >> Chris > >> _______________________________________________ > >> buildroot mailing list > >> buildroot@busybox.net > >> http://lists.busybox.net/mailman/listinfo/buildroot > >> Email had 2 attachments: > >> + host-dtc-1.4.1.patch > >> 33k (text/x-patch) > >> + cleaned_buildroot_v41_12182015_01.config > >> 61k (text/plain) > > > > > > -- > > Craig Swank > > craigswank@fastmail.fm > > _______________________________________________ > > buildroot mailing list > > buildroot@busybox.net > > http://lists.busybox.net/mailman/listinfo/buildroot
Glad that you're up and running Craig. I use the August 2015 uboot and the only change I make is to disable mmc 0 'cause I want to use the pins for something else. Have fun. Pushing back the boundaries of ignorance. On Dec 26, 2015 18:44, "Craig Swank" <craigswank@fastmail.fm> wrote: > YES! Using Chris's uEnv.txt and u-boot 2015.07 instead of 2015.10 it > finally booted up correctly. Many thanks to Thomas, Chris, and, last > but not least, Frank, who swooped in with the secret sauce at the very > end. > > I'm not even close to being able to figure out why the 07 version worked > and 10 did not, but I'm just too happy to care. > > Craig > > > On Sat, Dec 26, 2015, at 03:02 PM, Frank Hunleth wrote: > > Hi Craig, > > > > On Sat, Dec 26, 2015 at 1:34 PM, Craig Swank <craigswank@fastmail.fm> > > wrote: > > > My build succeeded, but I'm having trouble with booting now. When I > > > boot with the uEnv.txt you pasted: > > > > > > bootfile=zImage > > > fdtfile=am335x-boneblack.dtb > > > loadaddr=0x80007fc0 > > > fdtaddr=0x80F80000 > > > loadfdt=fatload mmc 0:1 ${fdtaddr} ${fdtfile} > > > loaduimage=fatload mmc 0:1 ${loadaddr} ${bootfile} > > > console=ttyO0,115200n8 > > > mmcroot=/dev/mmcblk0p2 > > > mmcrootfstype=ext2 > > > uenvcmd=mmc rescan; run loaduimage; run loadfdt; run fdtboot > > > fdtboot=run mmc_args; run mmcargs; bootz ${loadaddr} - ${fdtaddr} > > > mmc_args=setenv bootargs console=${console} ${optargs} root=${mmcroot} > > > rw > > > rootfstype=${mmcrootfstype} root wait > > > > > > I got an error about mmcargs not being defined. The fdtboot line has > > > both "run mmc_args" and "run mmcargs". I tried removing the 'run > > > mmcargs' part so now the uEnv.txt looks like: > > > > > > bootfile=zImage > > > fdtfile=am335x-boneblack.dtb > > > loadaddr=0x80007fc0 > > > fdtaddr=0x80F80000 > > > loadfdt=fatload mmc 0:1 ${fdtaddr} ${fdtfile} > > > loaduimage=fatload mmc 0:1 ${loadaddr} ${bootfile} > > > console=ttyO0,115200n8 > > > mmcroot=/dev/mmcblk0p2 > > > mmcrootfstype=ext2 > > > uenvcmd=mmc rescan; run loaduimage; run loadfdt; run fdtboot > > > fdtboot=run mmc_args; bootz ${loadaddr} - ${fdtaddr} > > > mmc_args=setenv bootargs console=${console} ${optargs} root=${mmcroot} > > > rw > > > rootfstype=${mmcrootfstype} root wait > > > > > > > > > Next, when I boot with the new uEnv.txt file the error I get is: > > > > > > U-Boot 2015.10 (Dec 26 2015 - 10:23:47 -0700) > > > > > > Watchdog enabled > > > I2C: ready > > > DRAM: 512 MiB > > > NAND: 0 MiB > > > MMC: OMAP SD/MMC: 0, OMAP SD/MMC: 1 > > > *** Error - No Valid Environment Area found > > > *** Warning - bad CRC, using default environment > > > > > > Net: <ethaddr> not set. Validating first E-fuse MAC > > > cpsw, usb_ether > > > Hit any key to stop autoboot: 0 > > > switch to partitions #0, OK > > > mmc0 is current device > > > SD/MMC found on device 0 > > > reading boot.scr > > > ** Unable to read file boot.scr ** > > > reading uEnv.txt > > > 471 bytes read in 3 ms (153.3 KiB/s) > > > Loaded environment from uEnv.txt > > > Importing environment from mmc ... > > > Running uenvcmd ... > > > reading zImage > > > 7462256 bytes read in 404 ms (17.6 MiB/s) > > > reading am335x-boneblack.dtb > > > 59295 bytes read in 10 ms (5.7 MiB/s) > > > Kernel image @ 0x80007fc0 [ 0x000000 - 0x71dd70 ] > > > data abort > > > pc : [<9ff62ef8>] lr : [<9ff62f55>] > > > reloc pc : [<80810ef8>] lr : [<80810f55>] > > > sp : 9ef31778 ip : 80725d30 fp : 9ef34f74 > > > r10: 00000002 r9 : 9ef31ed8 r8 : 9ffa35ac > > > r7 : 9ffa35b0 r6 : 00000003 r5 : 9ef34f74 r4 : 9ffa3514 > > > r3 : 00000000 r2 : 9ffa3514 r1 : 9ef34f74 r0 : 00000000 > > > Flags: nZcv IRQs off FIQs on Mode SVC_32 > > > Resetting CPU ... > > > > > > resetting ... > > > > > > I really need to learn more about u-boot, but google searches suggest > > > that it means the loadaddr or fdtaddr is incorrect but I'm unable to > > > find a solution. Any suggestions? > > > > I ran into this issue too. My workaround was to specify a custom uboot > > version and use the 2015.07 version instead. I didn't have time to > > debug what had changed, so I'm curious as well what the right solution > > is. > > > > Frank > > > > > > > > I appreciate you working with me, Chris. I'm obsessed with getting > this > > > to work and I feel I'm getting close. > > > > > > > > > > > > > > > On Sat, Dec 26, 2015, at 08:00 AM, Chris LaRocque wrote: > > >> Craig Swank wrote: > > >> > > >> > Here is another thing I've found: > > >> > > > >> > https://github.com/beagleboard/linux/issues/49 > > >> > > > >> > In that issue the person had some addresses set incorrectly in > uEnv.txt, > > >> > but mine doesn't have any addresses set: > > >> > > > >> > bootpart=0:1 > > >> > bootdir= > > >> > uenvcmd=run loadimage;run loadramdisk;run findfdt;run loadfdt;run > ram > > >> > boot > > >> > > > >> > Is there something that needs to be set in uEnv.txt? > > >> > > > >> > > > >> > On Wed, Dec 23, 2015, at 02:52 PM, Thomas Petazzoni wrote: > > >> >> Dear Craig Swank, > > >> >> > > >> >> On Wed, 23 Dec 2015 14:45:36 -0800, Craig Swank wrote: > > >> >> > Is it strange to you that when I do the build the output/build > > >> >> > directory has both of these dirs? > > >> >> > > > >> >> > ➜ buildroot git:(master) ✗ ls output/build/linux-* > > >> >> > output/build/linux-4.1.4-ti-r9 > > >> >> > output/build/linux-headers-3.12.10 > > >> >> > > >> >> No, it is perfectly fine. linux-headers is used when building the > > >> >> toolchain. linux is used to build the kernel itself. > > >> >> > > >> >> Thomas > > >> >> -- > > >> >> Thomas Petazzoni, CTO, Free Electrons > > >> >> Embedded Linux, Kernel and Android engineering > > >> >> http://free-electrons.com > > >> > > > >> > > > >> Hello Craig > > >> > > >> Hopefully the attachments will be accepted. > > >> > > >> The kernel defconfig from beagleboard.org: > > >> > > >> 1> doesn't set DEPRECATED_PARAM_STRUCT [=y] in the kernel .config. > > >> uEnv.txt > > >> (uboot) passes in the console config as "console=ttyO0,115200n8" > > >> > > >> 2> doesn't set EXT2_FS [=y] and the default buildroot .config FS > image is > > >> ext2. at least for the beaglebone black and I'm trying to keep things > > >> sane. > > >> > > >> You'll want to do a make source and them make linux-menuconfig and > change > > >> the parameters above. > > >> > > >> I haven't created a patch file because I'm lazy. > > >> > > >> Also attached is my patch (hack) for dtc, derived from Robert Nelsons > > >> bb.org-overlays. > > >> > > >> The dtc patch file goes into BR2_GLOBAL_PATCH_DIR , > > >> BR2_GLOBAL_PATCH_DIR"/dtc" > > >> > > >> Look in the buildroot .config for how I set BR2_GLOBAL_PATCH_DIR. > you'll > > >> need to point that to your own global patch directory. > > >> > > >> Regards > > >> > > >> Chris > > >> _______________________________________________ > > >> buildroot mailing list > > >> buildroot@busybox.net > > >> http://lists.busybox.net/mailman/listinfo/buildroot > > >> Email had 2 attachments: > > >> + host-dtc-1.4.1.patch > > >> 33k (text/x-patch) > > >> + cleaned_buildroot_v41_12182015_01.config > > >> 61k (text/plain) > > > > > > > > > -- > > > Craig Swank > > > craigswank@fastmail.fm > > > _______________________________________________ > > > buildroot mailing list > > > buildroot@busybox.net > > > http://lists.busybox.net/mailman/listinfo/buildroot > > > -- > Craig Swank > craigswank@fastmail.fm >
Hello all, On Sat, 26 Dec 2015 15:44:48 -0800, Craig Swank wrote: > YES! Using Chris's uEnv.txt and u-boot 2015.07 instead of 2015.10 it > finally booted up correctly. Many thanks to Thomas, Chris, and, last > but not least, Frank, who swooped in with the secret sauce at the very > end. > > I'm not even close to being able to figure out why the 07 version worked > and 10 did not, but I'm just too happy to care. Glad to hear that you solved the different problems. Our beaglebone_defconfig is utterly old, and it would be really great to update it to use newer kernel and u-boot versions. Thanks! Thomas
I'd like to help out, but it might take a bit of hand holding for me since I'm pretty new to buildroot and submitting patches. I suppose the things that would need to be changed would be 1. configs/beaglbone_defconfig 2. the uEnv.txt file needs to be different in output/images 3. Chris attached a patch for the host dtc 4. Two configs for the kernel build need to change I have a diff for configs/beaglebone_defconfig but everything else I'm not sure where to start. If someone has time to coach me through this I'm happy to learn, but if it is easier for someone else to do this I could learn from that too. On Sun, Dec 27, 2015, at 02:08 AM, Thomas Petazzoni wrote: > Hello all, > > On Sat, 26 Dec 2015 15:44:48 -0800, Craig Swank wrote: > > > YES! Using Chris's uEnv.txt and u-boot 2015.07 instead of 2015.10 it > > finally booted up correctly. Many thanks to Thomas, Chris, and, last > > but not least, Frank, who swooped in with the secret sauce at the very > > end. > > > > I'm not even close to being able to figure out why the 07 version worked > > and 10 did not, but I'm just too happy to care. > > Glad to hear that you solved the different problems. > > Our beaglebone_defconfig is utterly old, and it would be really great > to update it to use newer kernel and u-boot versions. > > Thanks! > > Thomas > -- > Thomas Petazzoni, CTO, Free Electrons > Embedded Linux, Kernel and Android engineering > http://free-electrons.com
Oh yeah, and u-boot needs to be version 2015.07 On Sun, Dec 27, 2015, at 05:34 AM, Craig Swank wrote: > I'd like to help out, but it might take a bit of hand holding for me > since I'm pretty new to buildroot and submitting patches. I suppose the > things that would need to be changed would be > > 1. configs/beaglbone_defconfig > 2. the uEnv.txt file needs to be different in output/images > 3. Chris attached a patch for the host dtc > 4. Two configs for the kernel build need to change > > I have a diff for configs/beaglebone_defconfig but everything else I'm > not sure where to start. > > If someone has time to coach me through this I'm happy to learn, but if > it is easier for someone else to do this I could learn from that too. > > > > On Sun, Dec 27, 2015, at 02:08 AM, Thomas Petazzoni wrote: > > Hello all, > > > > On Sat, 26 Dec 2015 15:44:48 -0800, Craig Swank wrote: > > > > > YES! Using Chris's uEnv.txt and u-boot 2015.07 instead of 2015.10 it > > > finally booted up correctly. Many thanks to Thomas, Chris, and, last > > > but not least, Frank, who swooped in with the secret sauce at the very > > > end. > > > > > > I'm not even close to being able to figure out why the 07 version worked > > > and 10 did not, but I'm just too happy to care. > > > > Glad to hear that you solved the different problems. > > > > Our beaglebone_defconfig is utterly old, and it would be really great > > to update it to use newer kernel and u-boot versions. > > > > Thanks! > > > > Thomas > > -- > > Thomas Petazzoni, CTO, Free Electrons > > Embedded Linux, Kernel and Android engineering > > http://free-electrons.com > > > -- > Craig Swank > craigswank@fastmail.fm > _______________________________________________ > buildroot mailing list > buildroot@busybox.net > http://lists.busybox.net/mailman/listinfo/buildroot
Hello I'm running a new beaglebone_defconfig as we speak and may be able to post a patch this afternoon. The DTC patch will take me a couple of days to test. What I posted earlier was a hack that accomplished my needs but it needs to be a little more refined. I'll review the debian install script at github.com/beagleboard/bb.org-overlays to see how to best incorporate both the DTC patch and the device tree overlays. Can I put the DTC patch into the /board/beaglebone directory and have it specified for application in the board defconfig? Suggestions on this aspect would be much appreciated. Regards C Pushing back the boundaries of ignorance. I'd like to help out, but it might take a bit of hand holding for me since I'm pretty new to buildroot and submitting patches. I suppose the things that would need to be changed would be 1. configs/beaglbone_defconfig 2. the uEnv.txt file needs to be different in output/images 3. Chris attached a patch for the host dtc 4. Two configs for the kernel build need to change I have a diff for configs/beaglebone_defconfig but everything else I'm not sure where to start. If someone has time to coach me through this I'm happy to learn, but if it is easier for someone else to do this I could learn from that too. On Sun, Dec 27, 2015, at 02:08 AM, Thomas Petazzoni wrote: > Hello all, > > On Sat, 26 Dec 2015 15:44:48 -0800, Craig Swank wrote: > > > YES! Using Chris's uEnv.txt and u-boot 2015.07 instead of 2015.10 it > > finally booted up correctly. Many thanks to Thomas, Chris, and, last > > but not least, Frank, who swooped in with the secret sauce at the very > > end. > > > > I'm not even close to being able to figure out why the 07 version worked > > and 10 did not, but I'm just too happy to care. > > Glad to hear that you solved the different problems. > > Our beaglebone_defconfig is utterly old, and it would be really great > to update it to use newer kernel and u-boot versions. > > Thanks! > > Thomas > -- > Thomas Petazzoni, CTO, Free Electrons > Embedded Linux, Kernel and Android engineering > http://free-electrons.com -- Craig Swank craigswank@fastmail.fm
Yes, I'll test the DTC. I've already created some overlays but there is a difference between the version 3.X and 4.X capemanager. You need to look at the structure of the new device trees and patch the old cape overlays to make them work. I'm struggling with that as well. You might want to verify that the modules are built. C Pushing back the boundaries of ignorance. On Dec 27, 2015 12:01, "Craig Swank" <craigswank@fastmail.fm> wrote: > I'll be interested in how your tests for the DTC patch go. I'm not having > luck with the bone_capemgr at the moment and am considering filing a bug at > beagleboard/linux. > > Will your tests involve compiling a .dtbo file and doing an echo command > to /sys/devices/platform/bone_capemgr/slots? When I try to load my dtbo > file (for dallas one wire temperature sensor), the expected kernel modules > don't get loaded even though no errors show up in dmesg. > > > > > On Sun, Dec 27, 2015, at 08:28 AM, Chris LaRocque wrote: > > > Hello > > I'm running a new beaglebone_defconfig as we speak and may be able to post > a patch this afternoon. > > The DTC patch will take me a couple of days to test. What I posted earlier > was a hack that accomplished my needs but it needs to be a little more > refined. I'll review the debian install script at > github.com/beagleboard/bb.org-overlays to see how to best incorporate > both the DTC patch and the device tree overlays. Can I put the DTC patch > into the /board/beaglebone directory and have it specified for application > in the board defconfig? Suggestions on this aspect would be much > appreciated. > > Regards > > C > > > Pushing back the boundaries of ignorance. > I'd like to help out, but it might take a bit of hand holding for me > since I'm pretty new to buildroot and submitting patches. I suppose the > things that would need to be changed would be > > 1. configs/beaglbone_defconfig > 2. the uEnv.txt file needs to be different in output/images > 3. Chris attached a patch for the host dtc > 4. Two configs for the kernel build need to change > > I have a diff for configs/beaglebone_defconfig but everything else I'm > not sure where to start. > > If someone has time to coach me through this I'm happy to learn, but if > it is easier for someone else to do this I could learn from that too. > > > > On Sun, Dec 27, 2015, at 02:08 AM, Thomas Petazzoni wrote: > > Hello all, > > > > On Sat, 26 Dec 2015 15:44:48 -0800, Craig Swank wrote: > > > > > YES! Using Chris's uEnv.txt and u-boot 2015.07 instead of 2015.10 it > > > finally booted up correctly. Many thanks to Thomas, Chris, and, last > > > but not least, Frank, who swooped in with the secret sauce at the very > > > end. > > > > > > I'm not even close to being able to figure out why the 07 version > worked > > > and 10 did not, but I'm just too happy to care. > > > > Glad to hear that you solved the different problems. > > > > Our beaglebone_defconfig is utterly old, and it would be really great > > to update it to use newer kernel and u-boot versions. > > > > Thanks! > > > > Thomas > > -- > > Thomas Petazzoni, CTO, Free Electrons > > Embedded Linux, Kernel and Android engineering > > http://free-electrons.com > > > -- > Craig Swank > craigswank@fastmail.fm > > > -- > Craig Swank > craigswank@fastmail.fm > > >
The modules are there. I'm able to modprobe them from the command line.
As far as the difference between the device tree structures, I'm just
trying to learn about this. For instance, my overlay looks like:
/dts-v1/; /plugin/; / { compatible = "ti,beaglebone", "ti,beaglebone-
black"; part-number = "BB-W1"; version = "00A0";
exclusive-use = "P8.11";
fragment@0 { target = <&am33xx_pinmux>;
__overlay__ { bb_w1_pins: pinmux_bb_w1_pins {
pinctrl-single,pins = <0x34 0x37 /* gpmc_ad13.gpio1_13,
OMAP_PIN_INPUT_PULLUP | OMAP_MUX_MODE7 - w1-gpio */ >;
}; }; };
fragment@1 { target = <&ocp>; __overlay__
{ onewire@0 {
status = "okay"; compatible
= "w1-gpio"; pinctrl-names =
"default"; pinctrl-0 =
<&bb_w1_pins>; gpios = <&gpio2
13 0>; }; }; }; };
I see am33xx_pinmux in am335x-boneblack.dts, but I don't see 'ocp'. Is
this what you mean by looking at the structure of the new device tree?
Since I don't see 'ocp' then does that mean that target (target =
<&ocp>;) could be my problem?
On Sun, Dec 27, 2015, at 10:36 AM, Chris LaRocque wrote:
> Yes, I'll test the DTC. I've already created some overlays but there
> is a difference between the version 3.X and 4.X capemanager. You need
> to look at the structure of the new device trees and patch the old
> cape overlays to make them work. I'm struggling with that as well.
> You might want to verify that the modules are built.
> C
> Pushing back the boundaries of ignorance.
> On Dec 27, 2015 12:01, "Craig Swank" <craigswank@fastmail.fm> wrote:
>> __
>> I'll be interested in how your tests for the DTC patch go. I'm not
>> having luck with the bone_capemgr at the moment and am considering
>> filing a bug at beagleboard/linux.
>>
>> Will your tests involve compiling a .dtbo file and doing an echo
>> command to /sys/devices/platform/bone_capemgr/slots? When I try to
>> load my dtbo file (for dallas one wire temperature sensor), the
>> expected kernel modules don't get loaded even though no errors show
>> up in dmesg.
>>
>>
>>
>>
>> On Sun, Dec 27, 2015, at 08:28 AM, Chris LaRocque wrote:
>>>
>>>
>>> Hello
>>>
>>> I'm running a new beaglebone_defconfig as we speak and may be able
>>> to post a patch this afternoon.
>>> The DTC patch will take me a couple of days to test. What I posted
>>> earlier was a hack that accomplished my needs but it needs to be a
>>> little more refined. I'll review the debian install script at github.com/beagleboard/bb.org-
>>> overlays to see how to best incorporate both the DTC patch and the
>>> device tree overlays. Can I put the DTC patch into the
>>> /board/beaglebone directory and have it specified for application in
>>> the board defconfig? Suggestions on this aspect would be much
>>> appreciated.
>>> Regards
>>> C
>>>
>>> Pushing back the boundaries of ignorance.
>>> I'd like to help out, but it might take a bit of hand holding for me
>>> since I'm pretty new to buildroot and submitting patches. I suppose
>>> the things that would need to be changed would be
>>>
>>> 1. configs/beaglbone_defconfig
>>> 2. the uEnv.txt file needs to be different in output/images
>>> 3. Chris attached a patch for the host dtc
>>> 4. Two configs for the kernel build need to change
>>>
>>> I have a diff for configs/beaglebone_defconfig but everything else
>>> I'm not sure where to start.
>>>
>>> If someone has time to coach me through this I'm happy to learn,
>>> but if it is easier for someone else to do this I could learn from
>>> that too.
>>>
>>>
>>>
>>> On Sun, Dec 27, 2015, at 02:08 AM, Thomas Petazzoni wrote:
>>> > Hello all,
>>> >
>>> > On Sat, 26 Dec 2015 15:44:48 -0800, Craig Swank wrote:
>>> >
>>> > > YES! Using Chris's uEnv.txt and u-boot 2015.07 instead of
>>> > > 2015.10 it finally booted up correctly. Many thanks to Thomas,
>>> > > Chris, and, last but not least, Frank, who swooped in with the
>>> > > secret sauce at the very end.
>>> > >
>>> > > I'm not even close to being able to figure out why the 07
>>> > > version worked and 10 did not, but I'm just too happy to care.
>>> >
>>> > Glad to hear that you solved the different problems.
>>> >
>>> > Our beaglebone_defconfig is utterly old, and it would be really
>>> > great to update it to use newer kernel and u-boot versions.
>>> >
>>> > Thanks!
>>> >
>>> > Thomas
>>> > --
>>> > Thomas Petazzoni, CTO, Free Electrons Embedded Linux, Kernel and
>>> > Android engineering http://free-electrons.com
>>>
>>>
>>> --
>>> Craig Swank craigswank@fastmail.fm
>>
>> --
>> Craig Swank craigswank@fastmail.fm
>>
>>
--
Craig Swank
craigswank@fastmail.fm
Never mind. I found that the 'ocp' section is in am33xx.dtsi. And looking at the history of that file you can see that gpio in the ocp node used to go from 1-4 and now they are 0-3 (gpio0, gpio1, ...). Fixing that doesn't help with the kernel modules not being loaded, however. On Sun, Dec 27, 2015, at 10:48 AM, Craig Swank wrote: > The modules are there. I'm able to modprobe them from the > command line. > > As far as the difference between the device tree structures, I'm just > trying to learn about this. For instance, my overlay looks like: > > /dts-v1/; /plugin/; / { compatible = "ti,beaglebone", "ti,beaglebone- > black"; part-number = "BB-W1"; version = "00A0"; > > exclusive-use = "P8.11"; > > fragment@0 { target = <&am33xx_pinmux>; > __overlay__ { bb_w1_pins: pinmux_bb_w1_pins { > pinctrl-single,pins = <0x34 0x37 /* gpmc_ad13.gpio1_13, > OMAP_PIN_INPUT_PULLUP | OMAP_MUX_MODE7 - w1-gpio */ >; > }; }; }; > > fragment@1 { target = <&ocp>; > __overlay__ { onewire@0 { > status = "okay"; compatible > = "w1-gpio"; pinctrl-names = > "default"; pinctrl-0 = > <&bb_w1_pins>; gpios = > <&gpio2 13 0>; }; }; }; > }; > > I see am33xx_pinmux in am335x-boneblack.dts, but I don't see 'ocp'. > Is this what you mean by looking at the structure of the new device > tree? Since I don't see 'ocp' then does that mean that target (target > = <&ocp>;) could be my problem? > > > On Sun, Dec 27, 2015, at 10:36 AM, Chris LaRocque wrote: >> Yes, I'll test the DTC. I've already created some overlays but there >> is a difference between the version 3.X and 4.X capemanager. You need >> to look at the structure of the new device trees and patch the old >> cape overlays to make them work. I'm struggling with that as well. >> You might want to verify that the modules are built. >> C >> Pushing back the boundaries of ignorance. >> On Dec 27, 2015 12:01, "Craig Swank" <craigswank@fastmail.fm> wrote: >>> __ >>> I'll be interested in how your tests for the DTC patch go. I'm not >>> having luck with the bone_capemgr at the moment and am considering >>> filing a bug at beagleboard/linux. >>> >>> Will your tests involve compiling a .dtbo file and doing an echo >>> command to /sys/devices/platform/bone_capemgr/slots? When I try to >>> load my dtbo file (for dallas one wire temperature sensor), the >>> expected kernel modules don't get loaded even though no errors show >>> up in dmesg. >>> >>> >>> >>> >>> On Sun, Dec 27, 2015, at 08:28 AM, Chris LaRocque wrote: >>>> >>>> >>>> Hello >>>> >>>> I'm running a new beaglebone_defconfig as we speak and may be able >>>> to post a patch this afternoon. >>>> The DTC patch will take me a couple of days to test. What I posted >>>> earlier was a hack that accomplished my needs but it needs to be a >>>> little more refined. I'll review the debian install script at github.com/beagleboard/bb.org- >>>> overlays to see how to best incorporate both the DTC patch and the >>>> device tree overlays. Can I put the DTC patch into the >>>> /board/beaglebone directory and have it specified for application >>>> in the board defconfig? Suggestions on this aspect would be much >>>> appreciated. >>>> Regards >>>> C >>>> >>>> Pushing back the boundaries of ignorance. >>>> I'd like to help out, but it might take a bit of hand holding for >>>> me since I'm pretty new to buildroot and submitting patches. I >>>> suppose the things that would need to be changed would be >>>> >>>> 1. configs/beaglbone_defconfig >>>> 2. the uEnv.txt file needs to be different in output/images >>>> 3. Chris attached a patch for the host dtc >>>> 4. Two configs for the kernel build need to change >>>> >>>> I have a diff for configs/beaglebone_defconfig but everything else >>>> I'm not sure where to start. >>>> >>>> If someone has time to coach me through this I'm happy to learn, >>>> but if it is easier for someone else to do this I could learn from >>>> that too. >>>> >>>> >>>> >>>> On Sun, Dec 27, 2015, at 02:08 AM, Thomas Petazzoni wrote: >>>> > Hello all, >>>> > >>>> > On Sat, 26 Dec 2015 15:44:48 -0800, Craig Swank wrote: >>>> > >>>> > > YES! Using Chris's uEnv.txt and u-boot 2015.07 instead of >>>> > > 2015.10 it finally booted up correctly. Many thanks to Thomas, >>>> > > Chris, and, last but not least, Frank, who swooped in with the >>>> > > secret sauce at the very end. >>>> > > >>>> > > I'm not even close to being able to figure out why the 07 >>>> > > version worked and 10 did not, but I'm just too happy to care. >>>> > >>>> > Glad to hear that you solved the different problems. >>>> > >>>> > Our beaglebone_defconfig is utterly old, and it would be really >>>> > great to update it to use newer kernel and u-boot versions. >>>> > >>>> > Thanks! >>>> > >>>> > Thomas >>>> > -- >>>> > Thomas Petazzoni, CTO, Free Electrons Embedded Linux, Kernel and >>>> > Android engineering http://free-electrons.com >>>> >>>> >>>> -- >>>> Craig Swank craigswank@fastmail.fm >>> >>> -- >>> Craig Swank craigswank@fastmail.fm >>> >>> > > -- > Craig Swank craigswank@fastmail.fm > > > _________________________________________________ > buildroot mailing list buildroot@busybox.net > http://lists.busybox.net/mailman/listinfo/buildroot -- Craig Swank craigswank@fastmail.fm
--- a/checks.c 2015-12-10 14:50:21.537561920 -0500 +++ b/checks.c 2015-12-01 17:11:59.266242047 -0500 @@ -458,6 +458,8 @@ struct node *node, struct property *prop) { struct marker *m = prop->val.markers; + struct fixup *f, **fp; + struct fixup_entry *fe, **fep; struct node *refnode; cell_t phandle; @@ -466,14 +468,73 @@ refnode = get_node_by_ref(dt, m->ref); if (! refnode) { + if (!dt->is_plugin) { FAIL(c, "Reference to non-existent node or label \"%s\"\n", m->ref); continue; } + /* allocate fixup entry */ + fe = xmalloc(sizeof(*fe)); + + fe->node = node; + fe->prop = prop; + fe->offset = m->offset; + fe->next = NULL; + + /* search for an already existing fixup */ + for_each_fixup(dt, f) + if (strcmp(f->ref, m->ref) == 0) + break; + + /* no fixup found, add new */ + if (f == NULL) { + f = xmalloc(sizeof(*f)); + f->ref = m->ref; + f->entries = NULL; + f->next = NULL; + + /* add it to the tree */ + fp = &dt->fixups; + while (*fp) + fp = &(*fp)->next; + *fp = f; + } + + /* and now append fixup entry */ + fep = &f->entries; + while (*fep) + fep = &(*fep)->next; + *fep = fe; + + /* mark the entry as unresolved */ + *((cell_t *)(prop->val.val + m->offset)) = + cpu_to_fdt32(0xdeadbeef); + continue; + } + + /* if it's a local reference, we need to record it */ + if (symbol_fixup_support) { + + /* allocate a new local fixup entry */ + fe = xmalloc(sizeof(*fe)); + + fe->node = node; + fe->prop = prop; + fe->offset = m->offset; + fe->next = NULL; + + /* append it to the local fixups */ + fep = &dt->local_fixups; + while (*fep) + fep = &(*fep)->next; + *fep = fe; + } + phandle = get_node_phandle(dt, refnode); *((cell_t *)(prop->val.val + m->offset)) = cpu_to_fdt32(phandle); } + } ERROR(phandle_references, NULL, NULL, fixup_phandle_references, NULL, &duplicate_node_names, &explicit_phandles); @@ -652,6 +713,45 @@ } TREE_WARNING(obsolete_chosen_interrupt_controller, NULL); +static void check_auto_label_phandles(struct check *c, struct node *dt, + struct node *node) +{ + struct label *l; + struct symbol *s, **sp; + int has_label; + + if (!symbol_fixup_support) + return; + + has_label = 0; + for_each_label(node->labels, l) { + has_label = 1; + break; + } + + if (!has_label) + return; + + /* force allocation of a phandle for this node */ + (void)get_node_phandle(dt, node); + + /* add the symbol */ + for_each_label(node->labels, l) { + + s = xmalloc(sizeof(*s)); + s->label = l; + s->node = node; + s->next = NULL; + + /* add it to the symbols list */ + sp = &dt->symbols; + while (*sp) + sp = &((*sp)->next); + *sp = s; + } +} +NODE_WARNING(auto_label_phandles, NULL); + static struct check *check_table[] = { &duplicate_node_names, &duplicate_property_names, &node_name_chars, &node_name_format, &property_name_chars, @@ -670,6 +770,8 @@ &avoid_default_addr_size, &obsolete_chosen_interrupt_controller, + &auto_label_phandles, + &always_fail, }; --- a/Documentation/dt-object-internal.txt 1969-12-31 19:00:00.000000000 -0500 +++ b/Documentation/dt-object-internal.txt 2015-12-01 17:11:59.249242025 -0500 @@ -0,0 +1,301 @@ +Device Tree Dynamic Object format internals +------------------------------------------- + +The Device Tree for most platforms is a static representation of +the hardware capabilities. This is insufficient for many platforms +that need to dynamically insert device tree fragments to the +running kernel's live tree. + +This document explains the the device tree object format and the +modifications made to the device tree compiler, which make it possible. + +1. Simplified Problem Definition +-------------------------------- + +Assume we have a platform which boots using following simplified device tree. + +---- foo.dts ----------------------------------------------------------------- + /* FOO platform */ + / { + compatible = "corp,foo"; + + /* shared resources */ + res: res { + }; + + /* On chip peripherals */ + ocp: ocp { + /* peripherals that are always instantiated */ + peripheral1 { ... }; + } + }; +---- foo.dts ----------------------------------------------------------------- + +We have a number of peripherals that after probing (using some undefined method) +should result in different device tree configuration. + +We cannot boot with this static tree because due to the configuration of the +foo platform there exist multiple conficting peripherals DT fragments. + +So for the bar peripheral we would have this: + +---- foo+bar.dts ------------------------------------------------------------- + /* FOO platform + bar peripheral */ + / { + compatible = "corp,foo"; + + /* shared resources */ + res: res { + }; + + /* On chip peripherals */ + ocp: ocp { + /* peripherals that are always instantiated */ + peripheral1 { ... }; + + /* bar peripheral */ + bar { + compatible = "corp,bar"; + ... /* various properties and child nodes */ + } + } + }; +---- foo+bar.dts ------------------------------------------------------------- + +While for the baz peripheral we would have this: + +---- foo+baz.dts ------------------------------------------------------------- + /* FOO platform + baz peripheral */ + / { + compatible = "corp,foo"; + + /* shared resources */ + res: res { + /* baz resources */ + baz_res: res_baz { ... }; + }; + + /* On chip peripherals */ + ocp: ocp { + /* peripherals that are always instantiated */ + peripheral1 { ... }; + + /* baz peripheral */ + baz { + compatible = "corp,baz"; + /* reference to another point in the tree */ + ref-to-res = <&baz_res>; + ... /* various properties and child nodes */ + } + } + }; +---- foo+baz.dts ------------------------------------------------------------- + +We note that the baz case is more complicated, since the baz peripheral needs to +reference another node in the DT tree. + +2. Device Tree Object Format Requirements +----------------------------------------- + +Since the device tree is used for booting a number of very different hardware +platforms it is imperative that we tread very carefully. + +2.a) No changes to the Device Tree binary format. We cannot modify the tree +format at all and all the information we require should be encoded using device +tree itself. We can add nodes that can be safely ignored by both bootloaders and +the kernel. + +2.b) Changes to the DTS source format should be absolutely minimal, and should +only be needed for the DT fragment definitions, and not the base boot DT. + +2.c) An explicit option should be used to instruct DTC to generate the required +information needed for object resolution. Platforms that don't use the +dynamic object format can safely ignore it. + +2.d) Finally, DT syntax changes should be kept to a minimum. It should be +possible to express everything using the existing DT syntax. + +3. Implementation +----------------- + +The basic unit of addressing in Device Tree is the phandle. Turns out it's +relatively simple to extend the way phandles are generated and referenced +so that it's possible to dynamically convert symbolic references (labels) +to phandle values. + +We can roughly divide the operation into two steps. + +3.a) Compilation of the base board DTS file using the '-@' option +generates a valid DT blob with an added __symbols__ node at the root node, +containing a list of all nodes that are marked with a label. + +Using the foo.dts file above the following node will be generated; + +$ dtc -@ -O dtb -o foo.dtb -b 0 foo.dts +$ fdtdump foo.dtb +... +/ { + ... + res { + ... + linux,phandle = <0x00000001>; + phandle = <0x00000001>; + ... + }; + ocp { + ... + linux,phandle = <0x00000002>; + phandle = <0x00000002>; + ... + }; + __symbols__ { + res="/res"; + ocp="/ocp"; + }; +}; + +Notice that all the nodes that had a label have been recorded, and that +phandles have been generated for them. + +This blob can be used to boot the board normally, the __symbols__ node will +be safely ignored both by the bootloader and the kernel (the only loss will +be a few bytes of memory and disk space). + +3.b) The Device Tree fragments must be compiled with the same option but they +must also have a tag (/plugin/) that allows undefined references to labels +that are not present at compilation time to be recorded so that the runtime +loader can fix them. + +So the bar peripheral's DTS format would be of the form: + +/plugin/; /* allow undefined label references and record them */ +/ { + .... /* various properties for loader use; i.e. part id etc. */ + fragment@0 { + target = <&ocp>; + __overlay__ { + /* bar peripheral */ + bar { + compatible = "corp,bar"; + ... /* various properties and child nodes */ + } + }; + }; +}; + +Note that there's a target property that specifies the location where the +contents of the overlay node will be placed, and it references the label +in the foo.dts file. + +$ dtc -@ -O dtb -o bar.dtbo -b 0 bar.dts +$ fdtdump bar.dtbo +... +/ { + ... /* properties */ + fragment@0 { + target = <0xdeadbeef>; + __overlay__ { + bar { + compatible = "corp,bar"; + ... /* various properties and child nodes */ + } + }; + }; + __fixups__ { + ocp = "/fragment@0:target:0"; + }; +}; + +No __symbols__ has been generated (no label in bar.dts). +Note that the target's ocp label is undefined, so the phandle handle +value is filled with the illegal value '0xdeadbeef', while a __fixups__ +node has been generated, which marks the location in the tree where +the label lookup should store the runtime phandle value of the ocp node. + +The format of the __fixups__ node entry is + + <label> = "<local-full-path>:<property-name>:<offset>"; + +<label> Is the label we're referring +<local-full-path> Is the full path of the node the reference is +<property-name> Is the name of the property containing the + reference +<offset> The offset (in bytes) of where the property's + phandle value is located. + +Doing the same with the baz peripheral's DTS format is a little bit more +involved, since baz contains references to local labels which require +local fixups. + +/plugin/; /* allow undefined label references and record them */ +/ { + .... /* various properties for loader use; i.e. part id etc. */ + fragment@0 { + target = <&res>; + __overlay__ { + /* baz resources */ + baz_res: res_baz { ... }; + }; + }; + fragment@1 { + target = <&ocp>; + __overlay__ { + /* baz peripheral */ + baz { + compatible = "corp,baz"; + /* reference to another point in the tree */ + ref-to-res = <&baz_res>; + ... /* various properties and child nodes */ + } + }; + }; +}; + +Note that &bar_res reference. + +$ dtc -@ -O dtb -o baz.dtbo -b 0 baz.dts +$ fdtdump baz.dtbo +... +/ { + ... /* properties */ + fragment@0 { + target = <0xdeadbeef>; + __overlay__ { + res_baz { + .... + linux,phandle = <0x00000001>; + phandle = <0x00000001>; + }; + }; + }; + fragment@1 { + target = <0xdeadbeef>; + __overlay__ { + baz { + compatible = "corp,baz"; + ... /* various properties and child nodes */ + ref-to-res = <0x00000001>; + } + }; + }; + __fixups__ { + res = "/fragment@0:target:0"; + ocp = "/fragment@1:target:0"; + }; + __local_fixups__ { + fragment@1 { + __overlay__ { + baz { + ref-to-res = <0>; + }; + }; + }; + }; +}; + +This is similar to the bar case, but the reference of a local label by the +baz node generates a __local_fixups__ entry that records the place that the +local reference is being made. Since phandles are allocated starting at 1 +the run time loader must apply an offset to each phandle in every dynamic +DT object loaded. The __local_fixups__ node records the place of every +local reference so that the loader can apply the offset. --- a/Documentation/manual.txt 2015-12-10 14:50:21.537561920 -0500 +++ b/Documentation/manual.txt 2015-12-01 17:11:59.249242025 -0500 @@ -119,6 +119,14 @@ Make space for <number> reserve map entries Relevant for dtb and asm output only. + -@ + Generates a __symbols__ node at the root node of the resulting blob + for any node labels used, and for any local references using phandles + it also generates a __local_fixups__ node that tracks them. + + When using the /plugin/ tag all unresolved label references to + be tracked in the __fixups__ node, making dynamic resolution possible. + -S <bytes> Ensure the blob at least <bytes> long, adding additional space if needed. @@ -153,6 +161,8 @@ devicetree: '/' nodedef + plugindecl: '/' 'plugin' '/' ';' + nodedef: '{' list_of_property list_of_subnode '}' ';' property: label PROPNAME '=' propdata ';' --- a/dtc.c 2015-12-10 14:50:21.537561920 -0500 +++ b/dtc.c 2015-12-01 17:11:59.264242045 -0500 @@ -31,6 +31,7 @@ int minsize; /* Minimum blob size */ int padsize; /* Additional padding to blob */ int phandle_format = PHANDLE_BOTH; /* Use linux,phandle or phandle properties */ +int symbol_fixup_support = 0; static void fill_fullpaths(struct node *tree, const char *prefix) { @@ -53,7 +54,7 @@ #define FDT_VERSION(version) _FDT_VERSION(version) #define _FDT_VERSION(version) #version static const char usage_synopsis[] = "dtc [options] <input file>"; -static const char usage_short_opts[] = "qI:O:o:V:d:R:S:p:fb:i:H:sW:E:hv"; +static const char usage_short_opts[] = "qI:O:o:V:d:R:S:p:fb:i:H:sW:E:@hv"; static struct option const usage_long_opts[] = { {"quiet", no_argument, NULL, 'q'}, {"in-format", a_argument, NULL, 'I'}, @@ -71,6 +72,7 @@ {"phandle", a_argument, NULL, 'H'}, {"warning", a_argument, NULL, 'W'}, {"error", a_argument, NULL, 'E'}, + {"symbols", no_argument, NULL, '@'}, {"help", no_argument, NULL, 'h'}, {"version", no_argument, NULL, 'v'}, {NULL, no_argument, NULL, 0x0}, @@ -101,6 +103,7 @@ "\t\tboth - Both \"linux,phandle\" and \"phandle\" properties", "\n\tEnable/disable warnings (prefix with \"no-\")", "\n\tEnable/disable errors (prefix with \"no-\")", + "\n\tEnable symbols/fixup support", "\n\tPrint this help and exit", "\n\tPrint version and exit", NULL, @@ -233,7 +236,9 @@ case 'E': parse_checks_option(false, true, optarg); break; - + case '@': + symbol_fixup_support = 1; + break; case 'h': usage(NULL); default: --- a/dtc.h 2015-12-10 14:50:21.538561922 -0500 +++ b/dtc.h 2015-12-01 17:11:59.250242026 -0500 @@ -54,6 +54,7 @@ extern int minsize; /* Minimum blob size */ extern int padsize; /* Additional padding to blob */ extern int phandle_format; /* Use linux,phandle or phandle properties */ +extern int symbol_fixup_support;/* enable symbols & fixup support */ #define PHANDLE_LEGACY 0x1 #define PHANDLE_EPAPR 0x2 @@ -132,6 +133,26 @@ struct label *next; }; +struct fixup_entry { + int offset; + struct node *node; + struct property *prop; + struct fixup_entry *next; + bool local_fixup_generated; +}; + +struct fixup { + char *ref; + struct fixup_entry *entries; + struct fixup *next; +}; + +struct symbol { + struct label *label; + struct node *node; + struct symbol *next; +}; + struct property { bool deleted; char *name; @@ -158,6 +179,13 @@ int addr_cells, size_cells; struct label *labels; + + struct symbol *symbols; + struct fixup_entry *local_fixups; + bool emit_local_fixup_node; + + bool is_plugin; + struct fixup *fixups; }; #define for_each_label_withdel(l0, l) \ @@ -181,6 +209,18 @@ for_each_child_withdel(n, c) \ if (!(c)->deleted) +#define for_each_fixup(n, f) \ + for ((f) = (n)->fixups; (f); (f) = (f)->next) + +#define for_each_fixup_entry(f, fe) \ + for ((fe) = (f)->entries; (fe); (fe) = (fe)->next) + +#define for_each_symbol(n, s) \ + for ((s) = (n)->symbols; (s); (s) = (s)->next) + +#define for_each_local_fixup_entry(n, fe) \ + for ((fe) = (n)->local_fixups; (fe); (fe) = (fe)->next) + void add_label(struct label **labels, char *label); void delete_labels(struct label **labels); --- a/dtc-lexer.l 2015-12-10 14:50:21.537561920 -0500 +++ b/dtc-lexer.l 2015-12-01 17:11:59.249242025 -0500 @@ -113,6 +113,11 @@ return DT_V1; } +<*>"/plugin/" { + DPRINT("Keyword: /plugin/\n"); + return DT_PLUGIN; + } + <*>"/memreserve/" { DPRINT("Keyword: /memreserve/\n"); BEGIN_DEFAULT(); --- a/dtc-parser.y 2015-12-10 14:50:21.537561920 -0500 +++ b/dtc-parser.y 2015-12-01 17:11:59.249242025 -0500 @@ -19,6 +19,7 @@ */ %{ #include <stdio.h> +#include <inttypes.h> #include "dtc.h" #include "srcpos.h" @@ -52,9 +53,11 @@ struct node *nodelist; struct reserve_info *re; uint64_t integer; + bool is_plugin; } %token DT_V1 +%token DT_PLUGIN %token DT_MEMRESERVE %token DT_LSHIFT DT_RSHIFT DT_LE DT_GE DT_EQ DT_NE DT_AND DT_OR %token DT_BITS @@ -71,6 +74,7 @@ %type <data> propdata %type <data> propdataprefix +%type <is_plugin> plugindecl %type <re> memreserve %type <re> memreserves %type <array> arrayprefix @@ -101,10 +105,22 @@ %% sourcefile: - DT_V1 ';' memreserves devicetree + DT_V1 ';' plugindecl memreserves devicetree { - the_boot_info = build_boot_info($3, $4, - guess_boot_cpuid($4)); + $5->is_plugin = $3; + the_boot_info = build_boot_info($4, $5, + guess_boot_cpuid($5)); + } + ; + +plugindecl: + /* empty */ + { + $$ = false; + } + | DT_PLUGIN ';' + { + $$ = true; } ; --- a/fdtdump.c 2015-12-10 14:50:21.538561922 -0500 +++ b/fdtdump.c 2015-12-01 17:11:59.250242026 -0500 @@ -8,6 +8,14 @@ #include <stdlib.h> #include <string.h> #include <ctype.h> +#include <sys/types.h> +#include <sys/stat.h> +#include <alloca.h> +#include <dirent.h> +#include <limits.h> +#include <unistd.h> +#include <fcntl.h> +#include <errno.h> #include <libfdt.h> #include <libfdt_env.h> @@ -143,6 +151,95 @@ } } +static void dump_live_internal(const char *path, bool debug, int depth) +{ + int maxsz = strlen(path) + 1 + PATH_MAX; + char *new_path = alloca(maxsz + 1); + struct stat sb; + struct dirent *de; + char *buf, *p; + int buf_alloc, shift, chunk, left, fd, ret; + DIR *d; + + shift = 4; + buf_alloc = 4 * 1024; /* 4K (maximum chunk) */ + buf = alloca(buf_alloc + sizeof(uint32_t)); + buf[buf_alloc] = '\0'; /* always terminate (just in case) */ + + d = opendir(path); + if (d == NULL) + die("Could not open %s directory\n", path); + + /* first dump the properties (files) */ + while ((de = readdir(d)) != NULL) { + /* properties are files */ + if (de->d_type != DT_REG) + continue; + snprintf(new_path, maxsz, "%s/%s", path, de->d_name); + new_path[maxsz] = '\0'; + printf("%*s%s", depth * shift, "", de->d_name); + + if (stat(new_path, &sb) != 0) + die("could not open: %s\n", new_path); + + fd = open(new_path, O_RDONLY); + if (fd == -1) + die("Could not open: %s\n", new_path); + + chunk = sb.st_size > buf_alloc ? buf_alloc : sb.st_size; + p = buf; + left = chunk; + while (left > 0) { + do { + ret = read(fd, p, left); + } while (ret == -1 && (errno == EAGAIN || errno == EINTR)); + if (ret == -1) + die("Read failed on: %s\n", new_path); + left -= ret; + p += ret; + } + close(fd); + + if (chunk < sb.st_size) + printf(" (trunc)"); + utilfdt_print_data(buf, chunk); + printf(";\n"); + } + + /* now recurse to the directories */ + rewinddir(d); + while ((de = readdir(d)) != NULL) { + /* properties are files */ + if (de->d_type != DT_DIR) + continue; + /* skip current and parent directories */ + if (strcmp(de->d_name, ".") == 0 || + strcmp(de->d_name, "..") == 0) + continue; + snprintf(new_path, maxsz, "%s/%s", path, de->d_name); + new_path[maxsz] = '\0'; + printf("%*s%s {\n", depth * shift, "", de->d_name); + dump_live_internal(new_path, debug, depth + 1); + printf("%*s};\n", depth * shift, ""); + } +} + +static void dump_live(const char *path, bool debug) +{ + char *fixed_path = alloca(strlen(path) + 1); + char *p; + + /* strip trailing / */ + strcpy(fixed_path, path); + p = fixed_path + strlen(fixed_path) - 1; + while (*p == '/' && p > fixed_path) + *p-- = '\0'; + printf("/* dump of live tree at %s */\n", fixed_path); + printf("/ {\n"); + dump_live_internal(fixed_path, debug, 1); + printf("};\n"); +} + /* Usage related data. */ static const char usage_synopsis[] = "fdtdump [options] <file>"; static const char usage_short_opts[] = "ds" USAGE_COMMON_SHORT_OPTS; @@ -165,6 +262,7 @@ bool debug = false; bool scan = false; off_t len; + struct stat sb; while ((opt = util_getopt_long()) != EOF) { switch (opt) { @@ -182,6 +280,15 @@ usage("missing input filename"); file = argv[optind]; + if (stat(file, &sb) != 0) + die("could not open: %s\n", file); + + /* dump live tree if it's a directory */ + if (S_ISDIR(sb.st_mode)) { + dump_live(file, debug); + return 0; + } + buf = utilfdt_read_len(file, &len); if (!buf) die("could not read: %s\n", file); --- a/flattree.c 2015-12-10 14:50:21.538561922 -0500 +++ b/flattree.c 2015-12-01 17:11:59.250242026 -0500 @@ -255,6 +255,204 @@ return i; } +static void emit_local_fixups(struct node *tree, struct emitter *emit, + void *etarget, struct data *strbuf, struct version_info *vi, + struct node *node) +{ + struct fixup_entry *fe, *fen; + struct node *child; + int nameoff, count; + cell_t *buf; + struct data d; + + if (node->emit_local_fixup_node) { + + /* emit the external fixups (do not emit /) */ + if (node != tree) { + emit->beginnode(etarget, NULL); + emit->string(etarget, node->name, 0); + emit->align(etarget, sizeof(cell_t)); + } + + for_each_local_fixup_entry(tree, fe) { + if (fe->node != node || fe->local_fixup_generated) + continue; + + /* count the number of fixup entries */ + count = 0; + for_each_local_fixup_entry(tree, fen) { + if (fen->prop != fe->prop) + continue; + fen->local_fixup_generated = true; + count++; + } + + /* allocate buffer */ + buf = xmalloc(count * sizeof(cell_t)); + + /* collect all the offsets in buffer */ + count = 0; + for_each_local_fixup_entry(tree, fen) { + if (fen->prop != fe->prop) + continue; + fen->local_fixup_generated = true; + buf[count++] = cpu_to_fdt32(fen->offset); + } + d = empty_data; + d.len = count * sizeof(cell_t); + d.val = (char *)buf; + + nameoff = stringtable_insert(strbuf, fe->prop->name); + emit->property(etarget, fe->prop->labels); + emit->cell(etarget, count * sizeof(cell_t)); + emit->cell(etarget, nameoff); + + if ((vi->flags & FTF_VARALIGN) && + (count * sizeof(cell_t)) >= 8) + emit->align(etarget, 8); + + emit->data(etarget, d); + emit->align(etarget, sizeof(cell_t)); + + free(buf); + } + } + + for_each_child(node, child) + emit_local_fixups(tree, emit, etarget, strbuf, vi, child); + + if (node->emit_local_fixup_node && node != tree) + emit->endnode(etarget, tree->labels); +} + +static void emit_symbols_node(struct node *tree, struct emitter *emit, + void *etarget, struct data *strbuf, + struct version_info *vi) +{ + struct symbol *sym; + int nameoff, vallen; + + /* do nothing if no symbols */ + if (!tree->symbols) + return; + + emit->beginnode(etarget, NULL); + emit->string(etarget, "__symbols__", 0); + emit->align(etarget, sizeof(cell_t)); + + for_each_symbol(tree, sym) { + + vallen = strlen(sym->node->fullpath); + + nameoff = stringtable_insert(strbuf, sym->label->label); + + emit->property(etarget, NULL); + emit->cell(etarget, vallen + 1); + emit->cell(etarget, nameoff); + + if ((vi->flags & FTF_VARALIGN) && vallen >= 8) + emit->align(etarget, 8); + + emit->string(etarget, sym->node->fullpath, + strlen(sym->node->fullpath)); + emit->align(etarget, sizeof(cell_t)); + } + + emit->endnode(etarget, NULL); +} + +static void emit_local_fixups_node(struct node *tree, struct emitter *emit, + void *etarget, struct data *strbuf, + struct version_info *vi) +{ + struct fixup_entry *fe; + struct node *node; + + /* do nothing if no local fixups */ + if (!tree->local_fixups) + return; + + /* mark all nodes that need a local fixup generated (and parents) */ + for_each_local_fixup_entry(tree, fe) { + node = fe->node; + while (node != NULL && !node->emit_local_fixup_node) { + node->emit_local_fixup_node = true; + node = node->parent; + } + } + + /* emit the local fixups node now */ + emit->beginnode(etarget, NULL); + emit->string(etarget, "__local_fixups__", 0); + emit->align(etarget, sizeof(cell_t)); + + emit_local_fixups(tree, emit, etarget, strbuf, vi, tree); + + emit->endnode(etarget, tree->labels); +} + +static void emit_fixups_node(struct node *tree, struct emitter *emit, + void *etarget, struct data *strbuf, + struct version_info *vi) +{ + struct fixup *f; + struct fixup_entry *fe; + char *name, *s; + const char *fullpath; + int namesz, nameoff, vallen; + + /* do nothing if no fixups */ + if (!tree->fixups) + return; + + /* emit the external fixups */ + emit->beginnode(etarget, NULL); + emit->string(etarget, "__fixups__", 0); + emit->align(etarget, sizeof(cell_t)); + + for_each_fixup(tree, f) { + + namesz = 0; + for_each_fixup_entry(f, fe) { + fullpath = fe->node->fullpath; + if (fullpath[0] == '\0') + fullpath = "/"; + namesz += strlen(fullpath) + 1; + namesz += strlen(fe->prop->name) + 1; + namesz += 32; /* space for :<number> + '\0' */ + } + + name = xmalloc(namesz); + + s = name; + for_each_fixup_entry(f, fe) { + fullpath = fe->node->fullpath; + if (fullpath[0] == '\0') + fullpath = "/"; + snprintf(s, name + namesz - s, "%s:%s:%d", fullpath, + fe->prop->name, fe->offset); + s += strlen(s) + 1; + } + + nameoff = stringtable_insert(strbuf, f->ref); + vallen = s - name - 1; + + emit->property(etarget, NULL); + emit->cell(etarget, vallen + 1); + emit->cell(etarget, nameoff); + + if ((vi->flags & FTF_VARALIGN) && vallen >= 8) + emit->align(etarget, 8); + + emit->string(etarget, name, vallen); + emit->align(etarget, sizeof(cell_t)); + + free(name); + } + + emit->endnode(etarget, tree->labels); +} + static void flatten_tree(struct node *tree, struct emitter *emit, void *etarget, struct data *strbuf, struct version_info *vi) @@ -310,6 +508,10 @@ flatten_tree(child, emit, etarget, strbuf, vi); } + emit_symbols_node(tree, emit, etarget, strbuf, vi); + emit_local_fixups_node(tree, emit, etarget, strbuf, vi); + emit_fixups_node(tree, emit, etarget, strbuf, vi); + emit->endnode(etarget, tree->labels); } --- a/libfdt/fdt_ro.c 2014-11-11 22:29:16.000000000 -0500 +++ b/libfdt/fdt_ro.c 2015-12-01 17:11:42.213219672 -0500 @@ -154,9 +154,9 @@ return fdt_subnode_offset_namelen(fdt, parentoffset, name, strlen(name)); } -int fdt_path_offset(const void *fdt, const char *path) +int fdt_path_offset_namelen(const void *fdt, const char *path, int namelen) { - const char *end = path + strlen(path); + const char *end = path + namelen; const char *p = path; int offset = 0; @@ -164,7 +164,7 @@ /* see if we have an alias */ if (*path != '/') { - const char *q = strchr(path, '/'); + const char *q = memchr(path, '/', end - p); if (!q) q = end; @@ -177,14 +177,15 @@ p = q; } - while (*p) { + while (p < end) { const char *q; - while (*p == '/') + while (*p == '/') { p++; - if (! *p) - return offset; - q = strchr(p, '/'); + if (p == end) + return offset; + } + q = memchr(p, '/', end - p); if (! q) q = end; @@ -198,6 +199,11 @@ return offset; } +int fdt_path_offset(const void *fdt, const char *path) +{ + return fdt_path_offset_namelen(fdt, path, strlen(path)); +} + const char *fdt_get_name(const void *fdt, int nodeoffset, int *len) { const struct fdt_node_header *nh = _fdt_offset_ptr(fdt, nodeoffset); --- a/libfdt/libfdt.h 2014-11-11 22:29:16.000000000 -0500 +++ b/libfdt/libfdt.h 2015-12-01 17:11:42.213219672 -0500 @@ -323,6 +323,17 @@ int fdt_subnode_offset(const void *fdt, int parentoffset, const char *name); /** + * fdt_path_offset_namelen - find a tree node by its full path + * @fdt: pointer to the device tree blob + * @path: full path of the node to locate + * @namelen: number of characters of path to consider + * + * Identical to fdt_path_offset(), but only consider the first namelen + * characters of path as the path name. + */ +int fdt_path_offset_namelen(const void *fdt, const char *path, int namelen); + +/** * fdt_path_offset - find a tree node by its full path * @fdt: pointer to the device tree blob * @path: full path of the node to locate --- a/libfdt/version.lds 2014-11-11 22:29:16.000000000 -0500 +++ b/libfdt/version.lds 2015-12-01 17:11:42.213219672 -0500 @@ -8,6 +8,7 @@ fdt_get_mem_rsv; fdt_subnode_offset_namelen; fdt_subnode_offset; + fdt_path_offset_namelen; fdt_path_offset; fdt_get_name; fdt_get_property_namelen; @@ -54,6 +55,8 @@ fdt_get_property_by_offset; fdt_getprop_by_offset; fdt_next_property_offset; + fdt_first_subnode; + fdt_next_subnode; local: *; --- a/tests/run_tests.sh 2014-11-11 22:29:16.000000000 -0500 +++ b/tests/run_tests.sh 2015-12-01 17:11:42.213219672 -0500 @@ -42,20 +42,20 @@ shorten_echo () { limit=32 - echo -n "$1" + printf "$1" shift for x; do if [ ${#x} -le $limit ]; then - echo -n " $x" + printf " $x" else short=$(echo "$x" | head -c$limit) - echo -n " \"$short\"...<${#x} bytes>" + printf " \"$short\"...<${#x} bytes>" fi done } run_test () { - echo -n "$@: " + printf "$*: " if [ -n "$VALGRIND" -a -f $1.supp ]; then VGSUPP="--suppressions=$1.supp" fi @@ -63,7 +63,7 @@ } run_sh_test () { - echo -n "$@: " + printf "$*: " base_run_test sh "$@" } @@ -106,12 +106,12 @@ run_wrap_error_test () { shorten_echo "$@" - echo -n " {!= 0}: " + printf " {!= 0}: " base_run_test wrap_error "$@" } run_dtc_test () { - echo -n "dtc $@: " + printf "dtc $*: " base_run_test wrap_test $VALGRIND $DTC "$@" } @@ -126,7 +126,7 @@ run_fdtget_test () { expect="$1" shift - echo -n "fdtget-runtest.sh "$expect" $@: " + printf "fdtget-runtest.sh %s $*: " "$(echo $expect)" base_run_test sh fdtget-runtest.sh "$expect" "$@" } @@ -134,14 +134,14 @@ expect="$1" shift shorten_echo fdtput-runtest.sh "$expect" "$@" - echo -n ": " + printf ": " base_run_test sh fdtput-runtest.sh "$expect" "$@" } run_fdtdump_test() { file="$1" shorten_echo fdtdump-runtest.sh "$file" - echo -n ": " + printf ": " base_run_test sh fdtdump-runtest.sh "$file" } @@ -279,6 +279,8 @@ run_dtc_test -I dts -O dtb -o embedded_nul.test.dtb embedded_nul.dts run_dtc_test -I dts -O dtb -o embedded_nul_equiv.test.dtb embedded_nul_equiv.dts run_test dtbs_equal_ordered embedded_nul.test.dtb embedded_nul_equiv.test.dtb + + run_dtc_test -I dts -O dtb bad-size-cells.dts } dtc_tests () { @@ -399,6 +401,8 @@ tree1_tests dtc_tree1_merge.test.dtb test_tree1.dtb run_dtc_test -I dts -O dtb -o dtc_tree1_merge_labelled.test.dtb test_tree1_merge_labelled.dts tree1_tests dtc_tree1_merge_labelled.test.dtb test_tree1.dtb + run_dtc_test -I dts -O dtb -o dtc_tree1_label_noderef.test.dtb test_tree1_label_noderef.dts + run_test dtbs_equal_unordered dtc_tree1_label_noderef.test.dtb test_tree1.dtb run_dtc_test -I dts -O dtb -o multilabel_merge.test.dtb multilabel_merge.dts run_test references multilabel.test.dtb run_test dtbs_equal_ordered multilabel.test.dtb multilabel_merge.test.dtb @@ -610,6 +614,28 @@ run_wrap_test $DTPUT $dtb -cp /chosen run_wrap_test $DTPUT $dtb -cp /chosen/son + # Start again with a fresh dtb + run_dtc_test -O dtb -p $(stat -c %s $text) -o $dtb $dts + + # Node delete + run_wrap_test $DTPUT $dtb -c /chosen/node1 /chosen/node2 /chosen/node3 + run_fdtget_test "node3\nnode2\nnode1" $dtb -l /chosen + run_wrap_test $DTPUT $dtb -r /chosen/node1 /chosen/node2 + run_fdtget_test "node3" $dtb -l /chosen + + # Delete the non-existent node + run_wrap_error_test $DTPUT $dtb -r /non-existent/node + + # Property delete + run_fdtput_test "eva" $dtb /chosen/ name "" -ts "eva" + run_fdtput_test "016" $dtb /chosen/ age "" -ts "016" + run_fdtget_test "age\nname\nbootargs\nlinux,platform" $dtb -p /chosen + run_wrap_test $DTPUT $dtb -d /chosen/ name age + run_fdtget_test "bootargs\nlinux,platform" $dtb -p /chosen + + # Delete the non-existent property + run_wrap_error_test $DTPUT $dtb -d /chosen non-existent-prop + # TODO: Add tests for verbose mode? }