Skip to contentSkip to conversation

BBB doesn't boot

DISCUSSION Gary Thomas 10 participants 49 messages Apr 13, 2014 – Apr 18, 2014

lore
Share & tools
lore
#0 lore raw

I just tried building (core-image-sato) for my BeagleBoneBlack using the latest Poky/Yocto master:

Build Configuration:

BB_VERSION        = "1.23.0"
BUILD_SYS         = "i686-linux"
NATIVELSBSTRING   = "Fedora-13"
TARGET_SYS        = "arm-poky-linux-gnueabi"
MACHINE           = "beaglebone"
DISTRO            = "poky"
DISTRO_VERSION    = "1.6+snapshot-20140411"
TUNE_FEATURES     = "armv7a vfp neon callconvention-hard cortexa8"
TARGET_FPU        = "vfp-neon"

meta
meta-yocto
meta-yocto-bsp = "master:863cc7483f5ee43189537940de8ee5c0964d24cc"

This built the kernel using SRCREV 928d7b2dda

I followed the bring-up instructions from README.hadware and the boot failed to even start the kernel. Here's what I see:

=============================== boot log ========================================= U-Boot 2013.07 (Apr 11 2014 - 15:03:04)

I2C:   ready
DRAM:  512 MiB
WARNING: Caches not enabled
NAND:  0 MiB
MMC:   OMAP SD/MMC: 0, OMAP SD/MMC: 1

*** Warning - readenv() failed, using default environment

musb-hdrc: ConfigData=0xde (UTMI-8, dyn FIFOs, HB-ISO Rx, HB-ISO Tx, SoftConn)
musb-hdrc: MHDRC RTL version 2.0
musb-hdrc: setup fifo_mode 4
musb-hdrc: 28/31 max ep, 16384/16384 memory
USB Peripheral mode controller at 47401000 using PIO, IRQ 0
musb-hdrc: ConfigData=0xde (UTMI-8, dyn FIFOs, HB-ISO Rx, HB-ISO Tx, SoftConn)
musb-hdrc: MHDRC RTL version 2.0
musb-hdrc: setup fifo_mode 4
musb-hdrc: 28/31 max ep, 16384/16384 memory
USB Host mode controller at 47401800 using PIO, IRQ 0
Net: <ethaddr> not set. Validating first E-fuse MAC
Phy not found
PHY reset timed out
cpsw, usb_ether
Hit any key to stop autoboot: 0
mmc0 is current device
SD/MMC found on device 0
reading uEnv.txt
** Unable to read file uEnv.txt **
4981688 bytes read in 613 ms (7.7 MiB/s)
29192 bytes read in 46 ms (619.1 KiB/s)
Booting from mmc ...
## Booting kernel from Legacy Image at 80200000 ...

   Image Name:   Linux-3.14.0-yocto-standard
   Image Type:   ARM Linux Kernel Image (uncompressed)
   Data Size:    4981624 Bytes = 4.8 MiB

Load Address: 80008000
Entry Point: 80008000
Verifying Checksum ... OK
## Flattened Device Tree blob at 80f80000
Booting using the fdt blob at 0x80f80000 Loading Kernel Image ... OK
Using Device Tree in place at 80f80000, end 80f8a207

Starting kernel ...
==================================================================================

Any ideas what I've done wrong?

Thanks

Signature · 4 lines
------------------------------------------------------------
Gary Thomas                 |  Consulting for the
MLB Associates              |    Embedded world
------------------------------------------------------------
#1 lore raw
Quoted 66 lines · Gary Thomas

I just tried building (core-image-sato) for my BeagleBoneBlack using the latest Poky/Yocto master:

Build Configuration:

BB_VERSION        = "1.23.0"
BUILD_SYS         = "i686-linux"
NATIVELSBSTRING   = "Fedora-13"
TARGET_SYS        = "arm-poky-linux-gnueabi"
MACHINE           = "beaglebone"
DISTRO            = "poky"
DISTRO_VERSION    = "1.6+snapshot-20140411"
TUNE_FEATURES     = "armv7a vfp neon callconvention-hard cortexa8"
TARGET_FPU        = "vfp-neon"

meta
meta-yocto
meta-yocto-bsp = "master:863cc7483f5ee43189537940de8ee5c0964d24cc"

This built the kernel using SRCREV 928d7b2dda

I followed the bring-up instructions from README.hadware and the boot failed to even start the kernel. Here's what I see:

=============================== boot log ========================================= U-Boot 2013.07 (Apr 11 2014 - 15:03:04)

I2C:   ready
DRAM:  512 MiB
WARNING: Caches not enabled
NAND:  0 MiB
MMC:   OMAP SD/MMC: 0, OMAP SD/MMC: 1

*** Warning - readenv() failed, using default environment

musb-hdrc: ConfigData=0xde (UTMI-8, dyn FIFOs, HB-ISO Rx, HB-ISO Tx, SoftConn)
musb-hdrc: MHDRC RTL version 2.0
musb-hdrc: setup fifo_mode 4
musb-hdrc: 28/31 max ep, 16384/16384 memory
USB Peripheral mode controller at 47401000 using PIO, IRQ 0
musb-hdrc: ConfigData=0xde (UTMI-8, dyn FIFOs, HB-ISO Rx, HB-ISO Tx, SoftConn)
musb-hdrc: MHDRC RTL version 2.0
musb-hdrc: setup fifo_mode 4
musb-hdrc: 28/31 max ep, 16384/16384 memory
USB Host mode controller at 47401800 using PIO, IRQ 0
Net: <ethaddr> not set. Validating first E-fuse MAC
Phy not found
PHY reset timed out
cpsw, usb_ether
Hit any key to stop autoboot: 0
mmc0 is current device
SD/MMC found on device 0
reading uEnv.txt
** Unable to read file uEnv.txt **
4981688 bytes read in 613 ms (7.7 MiB/s)
29192 bytes read in 46 ms (619.1 KiB/s)
Booting from mmc ...
## Booting kernel from Legacy Image at 80200000 ...

   Image Name:   Linux-3.14.0-yocto-standard
   Image Type:   ARM Linux Kernel Image (uncompressed)
   Data Size:    4981624 Bytes = 4.8 MiB

Load Address: 80008000
Entry Point: 80008000
Verifying Checksum ... OK
## Flattened Device Tree blob at 80f80000
Booting using the fdt blob at 0x80f80000 Loading Kernel Image ... OK
Using Device Tree in place at 80f80000, end 80f8a207

Quoted 4 lines

Starting kernel ...
==================================================================================

Any ideas what I've done wrong?

Hmm, everything looks sane. What revision is your BBB? And did you press USER/BOOT button or erased eMMC partition per instructions?

Signature · 1 line
Denys
#2 lore raw
Quoted 67 lines · Denys Dmytriyenko
Quoted 66 lines · Gary Thomas

I just tried building (core-image-sato) for my BeagleBoneBlack using the latest Poky/Yocto master:

Build Configuration:

BB_VERSION        = "1.23.0"
BUILD_SYS         = "i686-linux"
NATIVELSBSTRING   = "Fedora-13"
TARGET_SYS        = "arm-poky-linux-gnueabi"
MACHINE           = "beaglebone"
DISTRO            = "poky"
DISTRO_VERSION    = "1.6+snapshot-20140411"
TUNE_FEATURES     = "armv7a vfp neon callconvention-hard cortexa8"
TARGET_FPU        = "vfp-neon"

meta
meta-yocto
meta-yocto-bsp = "master:863cc7483f5ee43189537940de8ee5c0964d24cc"

This built the kernel using SRCREV 928d7b2dda

I followed the bring-up instructions from README.hadware and the boot failed to even start the kernel. Here's what I see:

=============================== boot log ========================================= U-Boot 2013.07 (Apr 11 2014 - 15:03:04)

I2C:   ready
DRAM:  512 MiB
WARNING: Caches not enabled
NAND:  0 MiB
MMC:   OMAP SD/MMC: 0, OMAP SD/MMC: 1

*** Warning - readenv() failed, using default environment

musb-hdrc: ConfigData=0xde (UTMI-8, dyn FIFOs, HB-ISO Rx, HB-ISO Tx, SoftConn)
musb-hdrc: MHDRC RTL version 2.0
musb-hdrc: setup fifo_mode 4
musb-hdrc: 28/31 max ep, 16384/16384 memory
USB Peripheral mode controller at 47401000 using PIO, IRQ 0
musb-hdrc: ConfigData=0xde (UTMI-8, dyn FIFOs, HB-ISO Rx, HB-ISO Tx, SoftConn)
musb-hdrc: MHDRC RTL version 2.0
musb-hdrc: setup fifo_mode 4
musb-hdrc: 28/31 max ep, 16384/16384 memory
USB Host mode controller at 47401800 using PIO, IRQ 0
Net: <ethaddr> not set. Validating first E-fuse MAC
Phy not found
PHY reset timed out
cpsw, usb_ether
Hit any key to stop autoboot: 0
mmc0 is current device
SD/MMC found on device 0
reading uEnv.txt
** Unable to read file uEnv.txt **
4981688 bytes read in 613 ms (7.7 MiB/s)
29192 bytes read in 46 ms (619.1 KiB/s)
Booting from mmc ...
## Booting kernel from Legacy Image at 80200000 ...

   Image Name:   Linux-3.14.0-yocto-standard
   Image Type:   ARM Linux Kernel Image (uncompressed)
   Data Size:    4981624 Bytes = 4.8 MiB

Load Address: 80008000
Entry Point: 80008000
Verifying Checksum ... OK
## Flattened Device Tree blob at 80f80000
Booting using the fdt blob at 0x80f80000 Loading Kernel Image ... OK
Using Device Tree in place at 80f80000, end 80f8a207

Quoted 8 lines
Quoted 4 lines

Starting kernel ...
==================================================================================

Any ideas what I've done wrong?

Hmm, everything looks sane. What revision is your BBB? And did you press USER/BOOT button or erased eMMC partition per instructions?

Revision A5A, with an LCD cape

Yes, I erased the eMMC before trying this.

Signature · 4 lines
------------------------------------------------------------
Gary Thomas                 |  Consulting for the
MLB Associates              |    Embedded world
------------------------------------------------------------
#45 lore raw
In reply to Gary Thomas

Hi Gary,

Quoted 66 lines · Gary Thomas

I just tried building (core-image-sato) for my BeagleBoneBlack using the latest Poky/Yocto master:

Build Configuration:

BB_VERSION        = "1.23.0"
BUILD_SYS         = "i686-linux"
NATIVELSBSTRING   = "Fedora-13"
TARGET_SYS        = "arm-poky-linux-gnueabi"
MACHINE           = "beaglebone"
DISTRO            = "poky"
DISTRO_VERSION    = "1.6+snapshot-20140411"
TUNE_FEATURES     = "armv7a vfp neon callconvention-hard cortexa8"
TARGET_FPU        = "vfp-neon"

meta
meta-yocto
meta-yocto-bsp = "master:863cc7483f5ee43189537940de8ee5c0964d24cc"

This built the kernel using SRCREV 928d7b2dda

I followed the bring-up instructions from README.hadware and the boot failed to even start the kernel. Here's what I see:

=============================== boot log ========================================= U-Boot 2013.07 (Apr 11 2014 - 15:03:04)

I2C:   ready
DRAM:  512 MiB
WARNING: Caches not enabled
NAND:  0 MiB
MMC:   OMAP SD/MMC: 0, OMAP SD/MMC: 1

*** Warning - readenv() failed, using default environment

musb-hdrc: ConfigData=0xde (UTMI-8, dyn FIFOs, HB-ISO Rx, HB-ISO Tx, SoftConn)
musb-hdrc: MHDRC RTL version 2.0
musb-hdrc: setup fifo_mode 4
musb-hdrc: 28/31 max ep, 16384/16384 memory
USB Peripheral mode controller at 47401000 using PIO, IRQ 0
musb-hdrc: ConfigData=0xde (UTMI-8, dyn FIFOs, HB-ISO Rx, HB-ISO Tx, SoftConn)
musb-hdrc: MHDRC RTL version 2.0
musb-hdrc: setup fifo_mode 4
musb-hdrc: 28/31 max ep, 16384/16384 memory
USB Host mode controller at 47401800 using PIO, IRQ 0
Net: <ethaddr> not set. Validating first E-fuse MAC
Phy not found
PHY reset timed out
cpsw, usb_ether
Hit any key to stop autoboot: 0
mmc0 is current device
SD/MMC found on device 0
reading uEnv.txt
** Unable to read file uEnv.txt **
4981688 bytes read in 613 ms (7.7 MiB/s)
29192 bytes read in 46 ms (619.1 KiB/s)
Booting from mmc ...
## Booting kernel from Legacy Image at 80200000 ...

   Image Name:   Linux-3.14.0-yocto-standard
   Image Type:   ARM Linux Kernel Image (uncompressed)
   Data Size:    4981624 Bytes = 4.8 MiB

Load Address: 80008000
Entry Point: 80008000
Verifying Checksum ... OK
## Flattened Device Tree blob at 80f80000
Booting using the fdt blob at 0x80f80000 Loading Kernel Image ... OK
Using Device Tree in place at 80f80000, end 80f8a207

Quoted 5 lines

Starting kernel ...
==================================================================================

Any ideas what I've done wrong?

This is strange because last week I booted myself a BBB. This actually looks like the serial doesn't output, perhaps the kernel does boot. I wonder if console is right for the kernel cmdline. On Friday I built and booted from master:db80f796e78746014a0f9497638e5c6fd2953ef0 which is a bit earlier that what you have but basically the same thing (updates for kernel which should fix things, I've tested with SRCREV_meta_pn-linux-yocto = "${AUTOREV}" those before they entered in master because I didn't had HDMI output on my master commit).

I'm gonna rebuild from daisy tip and see what happens.

Cheers,
Stefan

Quoted 7 lines

Thanks

Signature · 4 lines
------------------------------------------------------------
Gary Thomas                 |  Consulting for the
MLB Associates              |    Embedded world
------------------------------------------------------------
#46 lore raw
Quoted 85 lines · StefanX

Hi Gary,

Quoted 66 lines · Gary Thomas

I just tried building (core-image-sato) for my BeagleBoneBlack using the latest Poky/Yocto master:

Build Configuration:

BB_VERSION        = "1.23.0"
BUILD_SYS         = "i686-linux"
NATIVELSBSTRING   = "Fedora-13"
TARGET_SYS        = "arm-poky-linux-gnueabi"
MACHINE           = "beaglebone"
DISTRO            = "poky"
DISTRO_VERSION    = "1.6+snapshot-20140411"
TUNE_FEATURES     = "armv7a vfp neon callconvention-hard cortexa8"
TARGET_FPU        = "vfp-neon"

meta
meta-yocto
meta-yocto-bsp = "master:863cc7483f5ee43189537940de8ee5c0964d24cc"

This built the kernel using SRCREV 928d7b2dda

I followed the bring-up instructions from README.hadware and the boot failed to even start the kernel. Here's what I see:

=============================== boot log ========================================= U-Boot 2013.07 (Apr 11 2014 - 15:03:04)

I2C:   ready
DRAM:  512 MiB
WARNING: Caches not enabled
NAND:  0 MiB
MMC:   OMAP SD/MMC: 0, OMAP SD/MMC: 1

*** Warning - readenv() failed, using default environment

musb-hdrc: ConfigData=0xde (UTMI-8, dyn FIFOs, HB-ISO Rx, HB-ISO Tx, SoftConn)
musb-hdrc: MHDRC RTL version 2.0
musb-hdrc: setup fifo_mode 4
musb-hdrc: 28/31 max ep, 16384/16384 memory
USB Peripheral mode controller at 47401000 using PIO, IRQ 0
musb-hdrc: ConfigData=0xde (UTMI-8, dyn FIFOs, HB-ISO Rx, HB-ISO Tx, SoftConn)
musb-hdrc: MHDRC RTL version 2.0
musb-hdrc: setup fifo_mode 4
musb-hdrc: 28/31 max ep, 16384/16384 memory
USB Host mode controller at 47401800 using PIO, IRQ 0
Net: <ethaddr> not set. Validating first E-fuse MAC
Phy not found
PHY reset timed out
cpsw, usb_ether
Hit any key to stop autoboot: 0
mmc0 is current device
SD/MMC found on device 0
reading uEnv.txt
** Unable to read file uEnv.txt **
4981688 bytes read in 613 ms (7.7 MiB/s)
29192 bytes read in 46 ms (619.1 KiB/s)
Booting from mmc ...
## Booting kernel from Legacy Image at 80200000 ...

   Image Name:   Linux-3.14.0-yocto-standard
   Image Type:   ARM Linux Kernel Image (uncompressed)
   Data Size:    4981624 Bytes = 4.8 MiB

Load Address: 80008000
Entry Point: 80008000
Verifying Checksum ... OK
## Flattened Device Tree blob at 80f80000
Booting using the fdt blob at 0x80f80000 Loading Kernel Image ... OK
Using Device Tree in place at 80f80000, end 80f8a207

Quoted 5 lines

Starting kernel ...
==================================================================================

Any ideas what I've done wrong?

This is strange because last week I booted myself a BBB. This actually looks like the serial doesn't output, perhaps the kernel does boot. I wonder if console is right for the kernel cmdline. On Friday I built and booted from master:db80f796e78746014a0f9497638e5c6fd2953ef0 which is a bit earlier that what you have but basically the same thing (updates for kernel which should fix things, I've tested with SRCREV_meta_pn-linux-yocto = "${AUTOREV}" those before they entered in master because I didn't had HDMI output on my master commit).

Quoted 2 lines

I'm gonna rebuild from daisy tip and see what happens.

I just did a clean build from daisy and it worked fine... Could you try again? And perhaps -ccleansstate for linux-yocto? I'm thinking that somehow it didn't picked up the new console= that changed in the machine definition.

Cheers,
Stefan

Quoted 16 lines

Cheers,
Stefan

Quoted 7 lines

Thanks

Signature · 4 lines
------------------------------------------------------------
Gary Thomas                 |  Consulting for the
MLB Associates              |    Embedded world
------------------------------------------------------------
#47 lore raw
Quoted 89 lines · StefanX
Quoted 85 lines · StefanX

Hi Gary,

Quoted 66 lines · Gary Thomas

I just tried building (core-image-sato) for my BeagleBoneBlack using the latest Poky/Yocto master:

Build Configuration:

BB_VERSION        = "1.23.0"
BUILD_SYS         = "i686-linux"
NATIVELSBSTRING   = "Fedora-13"
TARGET_SYS        = "arm-poky-linux-gnueabi"
MACHINE           = "beaglebone"
DISTRO            = "poky"
DISTRO_VERSION    = "1.6+snapshot-20140411"
TUNE_FEATURES     = "armv7a vfp neon callconvention-hard cortexa8"
TARGET_FPU        = "vfp-neon"

meta
meta-yocto
meta-yocto-bsp = "master:863cc7483f5ee43189537940de8ee5c0964d24cc"

This built the kernel using SRCREV 928d7b2dda

I followed the bring-up instructions from README.hadware and the boot failed to even start the kernel. Here's what I see:

=============================== boot log ========================================= U-Boot 2013.07 (Apr 11 2014 - 15:03:04)

I2C:   ready
DRAM:  512 MiB
WARNING: Caches not enabled
NAND:  0 MiB
MMC:   OMAP SD/MMC: 0, OMAP SD/MMC: 1

*** Warning - readenv() failed, using default environment

musb-hdrc: ConfigData=0xde (UTMI-8, dyn FIFOs, HB-ISO Rx, HB-ISO Tx, SoftConn)
musb-hdrc: MHDRC RTL version 2.0
musb-hdrc: setup fifo_mode 4
musb-hdrc: 28/31 max ep, 16384/16384 memory
USB Peripheral mode controller at 47401000 using PIO, IRQ 0
musb-hdrc: ConfigData=0xde (UTMI-8, dyn FIFOs, HB-ISO Rx, HB-ISO Tx, SoftConn)
musb-hdrc: MHDRC RTL version 2.0
musb-hdrc: setup fifo_mode 4
musb-hdrc: 28/31 max ep, 16384/16384 memory
USB Host mode controller at 47401800 using PIO, IRQ 0
Net: <ethaddr> not set. Validating first E-fuse MAC
Phy not found
PHY reset timed out
cpsw, usb_ether
Hit any key to stop autoboot: 0
mmc0 is current device
SD/MMC found on device 0
reading uEnv.txt
** Unable to read file uEnv.txt **
4981688 bytes read in 613 ms (7.7 MiB/s)
29192 bytes read in 46 ms (619.1 KiB/s)
Booting from mmc ...
## Booting kernel from Legacy Image at 80200000 ...

   Image Name:   Linux-3.14.0-yocto-standard
   Image Type:   ARM Linux Kernel Image (uncompressed)
   Data Size:    4981624 Bytes = 4.8 MiB

Load Address: 80008000
Entry Point: 80008000
Verifying Checksum ... OK
## Flattened Device Tree blob at 80f80000
Booting using the fdt blob at 0x80f80000 Loading Kernel Image ... OK
Using Device Tree in place at 80f80000, end 80f8a207

Quoted 5 lines

Starting kernel ...
==================================================================================

Any ideas what I've done wrong?

This is strange because last week I booted myself a BBB. This actually looks like the serial doesn't output, perhaps the kernel does boot. I wonder if console is right for the kernel cmdline. On Friday I built and booted from master:db80f796e78746014a0f9497638e5c6fd2953ef0 which is a bit earlier that what you have but basically the same thing (updates for kernel which should fix things, I've tested with SRCREV_meta_pn-linux-yocto = "${AUTOREV}" those before they entered in master because I didn't had HDMI output on my master commit).

Quoted 7 lines
Quoted 2 lines

I'm gonna rebuild from daisy tip and see what happens.

I just did a clean build from daisy and it worked fine... Could you try again? And perhaps -ccleansstate for linux-yocto? I'm thinking that somehow it didn't picked up the new console= that changed in the machine definition.

I'll try daisy, but my build/test was clean from the mentioned master rev.

I checked the console setup in U-Boot:
console=ttyO0,115200n8
which I think is correct.

Signature · 4 lines
------------------------------------------------------------
Gary Thomas                 |  Consulting for the
MLB Associates              |    Embedded world
------------------------------------------------------------
#48 lore raw
Quoted 98 lines · Gary Thomas
Quoted 89 lines · StefanX
Quoted 85 lines · StefanX

Hi Gary,

Quoted 66 lines · Gary Thomas

I just tried building (core-image-sato) for my BeagleBoneBlack using the latest Poky/Yocto master:

Build Configuration:

BB_VERSION        = "1.23.0"
BUILD_SYS         = "i686-linux"
NATIVELSBSTRING   = "Fedora-13"
TARGET_SYS        = "arm-poky-linux-gnueabi"
MACHINE           = "beaglebone"
DISTRO            = "poky"
DISTRO_VERSION    = "1.6+snapshot-20140411"
TUNE_FEATURES     = "armv7a vfp neon callconvention-hard cortexa8"
TARGET_FPU        = "vfp-neon"

meta
meta-yocto
meta-yocto-bsp = "master:863cc7483f5ee43189537940de8ee5c0964d24cc"

This built the kernel using SRCREV 928d7b2dda

I followed the bring-up instructions from README.hadware and the boot failed to even start the kernel. Here's what I see:

=============================== boot log ========================================= U-Boot 2013.07 (Apr 11 2014 - 15:03:04)

I2C:   ready
DRAM:  512 MiB
WARNING: Caches not enabled
NAND:  0 MiB
MMC:   OMAP SD/MMC: 0, OMAP SD/MMC: 1

*** Warning - readenv() failed, using default environment

musb-hdrc: ConfigData=0xde (UTMI-8, dyn FIFOs, HB-ISO Rx, HB-ISO Tx, SoftConn)
musb-hdrc: MHDRC RTL version 2.0
musb-hdrc: setup fifo_mode 4
musb-hdrc: 28/31 max ep, 16384/16384 memory
USB Peripheral mode controller at 47401000 using PIO, IRQ 0
musb-hdrc: ConfigData=0xde (UTMI-8, dyn FIFOs, HB-ISO Rx, HB-ISO Tx, SoftConn)
musb-hdrc: MHDRC RTL version 2.0
musb-hdrc: setup fifo_mode 4
musb-hdrc: 28/31 max ep, 16384/16384 memory
USB Host mode controller at 47401800 using PIO, IRQ 0
Net: <ethaddr> not set. Validating first E-fuse MAC
Phy not found
PHY reset timed out
cpsw, usb_ether
Hit any key to stop autoboot: 0
mmc0 is current device
SD/MMC found on device 0
reading uEnv.txt
** Unable to read file uEnv.txt **
4981688 bytes read in 613 ms (7.7 MiB/s)
29192 bytes read in 46 ms (619.1 KiB/s)
Booting from mmc ...
## Booting kernel from Legacy Image at 80200000 ...

   Image Name:   Linux-3.14.0-yocto-standard
   Image Type:   ARM Linux Kernel Image (uncompressed)
   Data Size:    4981624 Bytes = 4.8 MiB

Load Address: 80008000
Entry Point: 80008000
Verifying Checksum ... OK
## Flattened Device Tree blob at 80f80000
Booting using the fdt blob at 0x80f80000 Loading Kernel Image ... OK
Using Device Tree in place at 80f80000, end 80f8a207

Quoted 5 lines

Starting kernel ...
==================================================================================

Any ideas what I've done wrong?

This is strange because last week I booted myself a BBB. This actually looks like the serial doesn't output, perhaps the kernel does boot. I wonder if console is right for the kernel cmdline. On Friday I built and booted from master:db80f796e78746014a0f9497638e5c6fd2953ef0 which is a bit earlier that what you have but basically the same thing (updates for kernel which should fix things, I've tested with SRCREV_meta_pn-linux-yocto = "${AUTOREV}" those before they entered in master because I didn't had HDMI output on my master commit).

Quoted 7 lines
Quoted 2 lines

I'm gonna rebuild from daisy tip and see what happens.

I just did a clean build from daisy and it worked fine... Could you try again? And perhaps -ccleansstate for linux-yocto? I'm thinking that somehow it didn't picked up the new console= that changed in the machine definition.

Quoted 6 lines

I'll try daisy, but my build/test was clean from the mentioned master rev.

I checked the console setup in U-Boot:
console=ttyO0,115200n8
which I think is correct.

Actually, I just compared my master tree with the daisy branch and they are the same, save for the release strings, so there's no need to rebuild.

Still at a loss though.

StefanX, perhaps you could let me try your exact image on my board?

Signature · 4 lines
------------------------------------------------------------
Gary Thomas                 |  Consulting for the
MLB Associates              |    Embedded world
------------------------------------------------------------
#3 lore raw
In reply to Gary Thomas
Quoted 71 lines · Gary Thomas
Quoted 70 lines · Denys Dmytriyenko
Quoted 69 lines · Gary Thomas

I just tried building (core-image-sato) for my BeagleBoneBlack using the latest Poky/Yocto master:

Build Configuration:

BB_VERSION        = "1.23.0"
BUILD_SYS         = "i686-linux"
NATIVELSBSTRING   = "Fedora-13"
TARGET_SYS        = "arm-poky-linux-gnueabi"
MACHINE           = "beaglebone"
DISTRO            = "poky"
DISTRO_VERSION    = "1.6+snapshot-20140411"
TUNE_FEATURES     = "armv7a vfp neon callconvention-hard cortexa8"
TARGET_FPU        = "vfp-neon"

meta
meta-yocto
meta-yocto-bsp = "master:863cc7483f5ee43189537940de8ee5c0964d24cc"

This built the kernel using SRCREV 928d7b2dda

I followed the bring-up instructions from README.hadware and the boot failed to even start the kernel. Here's what I see:

=============================== boot log ========================================= U-Boot 2013.07 (Apr 11 2014 - 15:03:04)

I2C:   ready
DRAM:  512 MiB
WARNING: Caches not enabled
NAND:  0 MiB
MMC:   OMAP SD/MMC: 0, OMAP SD/MMC: 1

*** Warning - readenv() failed, using default environment

musb-hdrc: ConfigData=0xde (UTMI-8, dyn FIFOs, HB-ISO Rx, HB-ISO Tx, SoftConn)
musb-hdrc: MHDRC RTL version 2.0
musb-hdrc: setup fifo_mode 4
musb-hdrc: 28/31 max ep, 16384/16384 memory
USB Peripheral mode controller at 47401000 using PIO, IRQ 0
musb-hdrc: ConfigData=0xde (UTMI-8, dyn FIFOs, HB-ISO Rx, HB-ISO Tx, SoftConn)
musb-hdrc: MHDRC RTL version 2.0
musb-hdrc: setup fifo_mode 4
musb-hdrc: 28/31 max ep, 16384/16384 memory
USB Host mode controller at 47401800 using PIO, IRQ 0
Net: <ethaddr> not set. Validating first E-fuse MAC
Phy not found
PHY reset timed out
cpsw, usb_ether
Hit any key to stop autoboot: 0
mmc0 is current device
SD/MMC found on device 0
reading uEnv.txt
** Unable to read file uEnv.txt **
4981688 bytes read in 613 ms (7.7 MiB/s)
29192 bytes read in 46 ms (619.1 KiB/s)
Booting from mmc ...
## Booting kernel from Legacy Image at 80200000 ...

   Image Name:   Linux-3.14.0-yocto-standard
   Image Type:   ARM Linux Kernel Image (uncompressed)
   Data Size:    4981624 Bytes = 4.8 MiB

Load Address: 80008000
Entry Point: 80008000
Verifying Checksum ... OK
## Flattened Device Tree blob at 80f80000
Booting using the fdt blob at 0x80f80000 Loading Kernel Image ... OK
Using Device Tree in place at 80f80000, end 80f8a207

Starting kernel ...
==================================================================================

Quoted 7 lines
Quoted 5 lines
Quoted 1 line

Any ideas what I've done wrong?

Hmm, everything looks sane. What revision is your BBB? And did you press USER/BOOT button or erased eMMC partition per instructions?

Revision A5A, with an LCD cape

Hmm, I'm wondering if LCD cape conflicts here - there's no cape support in this BSP. Can you try w/o it?

As of the board revision - I was testing on different ones from A1 to A6 and didn't see any major issue...

Signature · 1 line
Denys
#4 lore raw
Quoted 74 lines · Denys Dmytriyenko
Quoted 73 lines · Gary Thomas
Quoted 72 lines · Denys Dmytriyenko
Quoted 70 lines · Gary Thomas

I just tried building (core-image-sato) for my BeagleBoneBlack using the latest Poky/Yocto master:

Build Configuration:

BB_VERSION        = "1.23.0"
BUILD_SYS         = "i686-linux"
NATIVELSBSTRING   = "Fedora-13"
TARGET_SYS        = "arm-poky-linux-gnueabi"
MACHINE           = "beaglebone"
DISTRO            = "poky"
DISTRO_VERSION    = "1.6+snapshot-20140411"
TUNE_FEATURES     = "armv7a vfp neon callconvention-hard cortexa8"
TARGET_FPU        = "vfp-neon"

meta
meta-yocto
meta-yocto-bsp = "master:863cc7483f5ee43189537940de8ee5c0964d24cc"

This built the kernel using SRCREV 928d7b2dda

I followed the bring-up instructions from README.hadware and the boot failed to even start the kernel. Here's what I see:

=============================== boot log ========================================= U-Boot 2013.07 (Apr 11 2014 - 15:03:04)

I2C:   ready
DRAM:  512 MiB
WARNING: Caches not enabled
NAND:  0 MiB
MMC:   OMAP SD/MMC: 0, OMAP SD/MMC: 1

*** Warning - readenv() failed, using default environment

musb-hdrc: ConfigData=0xde (UTMI-8, dyn FIFOs, HB-ISO Rx, HB-ISO Tx, SoftConn)
musb-hdrc: MHDRC RTL version 2.0
musb-hdrc: setup fifo_mode 4
musb-hdrc: 28/31 max ep, 16384/16384 memory
USB Peripheral mode controller at 47401000 using PIO, IRQ 0
musb-hdrc: ConfigData=0xde (UTMI-8, dyn FIFOs, HB-ISO Rx, HB-ISO Tx, SoftConn)
musb-hdrc: MHDRC RTL version 2.0
musb-hdrc: setup fifo_mode 4
musb-hdrc: 28/31 max ep, 16384/16384 memory
USB Host mode controller at 47401800 using PIO, IRQ 0
Net: <ethaddr> not set. Validating first E-fuse MAC
Phy not found
PHY reset timed out
cpsw, usb_ether
Hit any key to stop autoboot: 0
mmc0 is current device
SD/MMC found on device 0
reading uEnv.txt
** Unable to read file uEnv.txt **
4981688 bytes read in 613 ms (7.7 MiB/s)
29192 bytes read in 46 ms (619.1 KiB/s)
Booting from mmc ...
## Booting kernel from Legacy Image at 80200000 ...

   Image Name:   Linux-3.14.0-yocto-standard
   Image Type:   ARM Linux Kernel Image (uncompressed)
   Data Size:    4981624 Bytes = 4.8 MiB

Load Address: 80008000
Entry Point: 80008000
Verifying Checksum ... OK
## Flattened Device Tree blob at 80f80000
Booting using the fdt blob at 0x80f80000 Loading Kernel Image ... OK
Using Device Tree in place at 80f80000, end 80f8a207

Starting kernel ...
==================================================================================

Any ideas what I've done wrong?

Quoted 8 lines
Quoted 5 lines
Quoted 3 lines

Hmm, everything looks sane. What revision is your BBB? And did you press USER/BOOT button or erased eMMC partition per instructions?

Revision A5A, with an LCD cape

Hmm, I'm wondering if LCD cape conflicts here - there's no cape support in this BSP. Can you try w/o it?

Sure I can try it but I don't think that's it. I got the kernel that StefanX built and booted and tried it on my board and it came up. No clue why the kernels are different - ostensibly we both built the same image from the same meta data, but they are slightly different (only in size - I compared the System.map files from both builds and they contain exactly the same bits, just a few changes in memory layout which I can't explain).

I'm trying another build from scratch using a different build host to see if that makes a difference.

Quoted 3 lines

As of the board revision - I was testing on different ones from A1 to A6 and didn't see any major issue...

Signature · 4 lines
------------------------------------------------------------
Gary Thomas                 |  Consulting for the
MLB Associates              |    Embedded world
------------------------------------------------------------
#5 lore raw
Quoted 84 lines · Gary Thomas
Quoted 74 lines · Denys Dmytriyenko
Quoted 73 lines · Gary Thomas
Quoted 72 lines · Denys Dmytriyenko
Quoted 70 lines · Gary Thomas

I just tried building (core-image-sato) for my BeagleBoneBlack using the latest Poky/Yocto master:

Build Configuration:

BB_VERSION        = "1.23.0"
BUILD_SYS         = "i686-linux"
NATIVELSBSTRING   = "Fedora-13"
TARGET_SYS        = "arm-poky-linux-gnueabi"
MACHINE           = "beaglebone"
DISTRO            = "poky"
DISTRO_VERSION    = "1.6+snapshot-20140411"
TUNE_FEATURES     = "armv7a vfp neon callconvention-hard cortexa8"
TARGET_FPU        = "vfp-neon"

meta
meta-yocto
meta-yocto-bsp = "master:863cc7483f5ee43189537940de8ee5c0964d24cc"

This built the kernel using SRCREV 928d7b2dda

I followed the bring-up instructions from README.hadware and the boot failed to even start the kernel. Here's what I see:

=============================== boot log ========================================= U-Boot 2013.07 (Apr 11 2014 - 15:03:04)

I2C:   ready
DRAM:  512 MiB
WARNING: Caches not enabled
NAND:  0 MiB
MMC:   OMAP SD/MMC: 0, OMAP SD/MMC: 1

*** Warning - readenv() failed, using default environment

musb-hdrc: ConfigData=0xde (UTMI-8, dyn FIFOs, HB-ISO Rx, HB-ISO Tx, SoftConn)
musb-hdrc: MHDRC RTL version 2.0
musb-hdrc: setup fifo_mode 4
musb-hdrc: 28/31 max ep, 16384/16384 memory
USB Peripheral mode controller at 47401000 using PIO, IRQ 0
musb-hdrc: ConfigData=0xde (UTMI-8, dyn FIFOs, HB-ISO Rx, HB-ISO Tx, SoftConn)
musb-hdrc: MHDRC RTL version 2.0
musb-hdrc: setup fifo_mode 4
musb-hdrc: 28/31 max ep, 16384/16384 memory
USB Host mode controller at 47401800 using PIO, IRQ 0
Net: <ethaddr> not set. Validating first E-fuse MAC
Phy not found
PHY reset timed out
cpsw, usb_ether
Hit any key to stop autoboot: 0
mmc0 is current device
SD/MMC found on device 0
reading uEnv.txt
** Unable to read file uEnv.txt **
4981688 bytes read in 613 ms (7.7 MiB/s)
29192 bytes read in 46 ms (619.1 KiB/s)
Booting from mmc ...
## Booting kernel from Legacy Image at 80200000 ...

   Image Name:   Linux-3.14.0-yocto-standard
   Image Type:   ARM Linux Kernel Image (uncompressed)
   Data Size:    4981624 Bytes = 4.8 MiB

Load Address: 80008000
Entry Point: 80008000
Verifying Checksum ... OK
## Flattened Device Tree blob at 80f80000
Booting using the fdt blob at 0x80f80000 Loading Kernel Image ... OK
Using Device Tree in place at 80f80000, end 80f8a207

Starting kernel ...
==================================================================================

Any ideas what I've done wrong?

Quoted 8 lines
Quoted 5 lines
Quoted 3 lines

Hmm, everything looks sane. What revision is your BBB? And did you press USER/BOOT button or erased eMMC partition per instructions?

Revision A5A, with an LCD cape

Hmm, I'm wondering if LCD cape conflicts here - there's no cape support in this BSP. Can you try w/o it?

Quoted 6 lines

Sure I can try it but I don't think that's it. I got the kernel that StefanX built and booted and tried it on my board and it came up. No clue why the kernels are different - ostensibly we both built the same image from the same meta data, but they are slightly different (only in size - I compared the System.map files from both builds and they contain exactly the same bits, just a few changes in memory layout which I can't explain).

Yeah, good point. Doesn't look like your cape causes the issue... The only other difference is in the host. Do you have access to another Linux box you can try? FWIW, I'm using 64-bit Gentoo with gcc-4.7.3. Yours is 32-bit Fedora 13, right?

Quoted 2 lines

I'm trying another build from scratch using a different build host to see if that makes a difference.

Do you have any other layers or customizations on top? (the metadata above suggests you build just pure Poky though)

Signature · 1 line
Denys
#6 lore raw
Quoted 92 lines · Denys Dmytriyenko
Quoted 84 lines · Gary Thomas
Quoted 74 lines · Denys Dmytriyenko
Quoted 73 lines · Gary Thomas
Quoted 72 lines · Denys Dmytriyenko
Quoted 70 lines · Gary Thomas

I just tried building (core-image-sato) for my BeagleBoneBlack using the latest Poky/Yocto master:

Build Configuration:

BB_VERSION        = "1.23.0"
BUILD_SYS         = "i686-linux"
NATIVELSBSTRING   = "Fedora-13"
TARGET_SYS        = "arm-poky-linux-gnueabi"
MACHINE           = "beaglebone"
DISTRO            = "poky"
DISTRO_VERSION    = "1.6+snapshot-20140411"
TUNE_FEATURES     = "armv7a vfp neon callconvention-hard cortexa8"
TARGET_FPU        = "vfp-neon"

meta
meta-yocto
meta-yocto-bsp = "master:863cc7483f5ee43189537940de8ee5c0964d24cc"

This built the kernel using SRCREV 928d7b2dda

I followed the bring-up instructions from README.hadware and the boot failed to even start the kernel. Here's what I see:

=============================== boot log ========================================= U-Boot 2013.07 (Apr 11 2014 - 15:03:04)

I2C:   ready
DRAM:  512 MiB
WARNING: Caches not enabled
NAND:  0 MiB
MMC:   OMAP SD/MMC: 0, OMAP SD/MMC: 1

*** Warning - readenv() failed, using default environment

musb-hdrc: ConfigData=0xde (UTMI-8, dyn FIFOs, HB-ISO Rx, HB-ISO Tx, SoftConn)
musb-hdrc: MHDRC RTL version 2.0
musb-hdrc: setup fifo_mode 4
musb-hdrc: 28/31 max ep, 16384/16384 memory
USB Peripheral mode controller at 47401000 using PIO, IRQ 0
musb-hdrc: ConfigData=0xde (UTMI-8, dyn FIFOs, HB-ISO Rx, HB-ISO Tx, SoftConn)
musb-hdrc: MHDRC RTL version 2.0
musb-hdrc: setup fifo_mode 4
musb-hdrc: 28/31 max ep, 16384/16384 memory
USB Host mode controller at 47401800 using PIO, IRQ 0
Net: <ethaddr> not set. Validating first E-fuse MAC
Phy not found
PHY reset timed out
cpsw, usb_ether
Hit any key to stop autoboot: 0
mmc0 is current device
SD/MMC found on device 0
reading uEnv.txt
** Unable to read file uEnv.txt **
4981688 bytes read in 613 ms (7.7 MiB/s)
29192 bytes read in 46 ms (619.1 KiB/s)
Booting from mmc ...
## Booting kernel from Legacy Image at 80200000 ...

   Image Name:   Linux-3.14.0-yocto-standard
   Image Type:   ARM Linux Kernel Image (uncompressed)
   Data Size:    4981624 Bytes = 4.8 MiB

Load Address: 80008000
Entry Point: 80008000
Verifying Checksum ... OK
## Flattened Device Tree blob at 80f80000
Booting using the fdt blob at 0x80f80000 Loading Kernel Image ... OK
Using Device Tree in place at 80f80000, end 80f8a207

Starting kernel ...
==================================================================================

Any ideas what I've done wrong?

Quoted 8 lines
Quoted 5 lines
Quoted 3 lines

Hmm, everything looks sane. What revision is your BBB? And did you press USER/BOOT button or erased eMMC partition per instructions?

Revision A5A, with an LCD cape

Hmm, I'm wondering if LCD cape conflicts here - there's no cape support in this BSP. Can you try w/o it?

Quoted 6 lines

Sure I can try it but I don't think that's it. I got the kernel that StefanX built and booted and tried it on my board and it came up. No clue why the kernels are different - ostensibly we both built the same image from the same meta data, but they are slightly different (only in size - I compared the System.map files from both builds and they contain exactly the same bits, just a few changes in memory layout which I can't explain).

Quoted 4 lines

Yeah, good point. Doesn't look like your cape causes the issue... The only other difference is in the host. Do you have access to another Linux box you can try? FWIW, I'm using 64-bit Gentoo with gcc-4.7.3. Yours is 32-bit Fedora 13, right?

I'll try Fedora 17 and Ubuntu 12.04 (x86_64) and see what happens.

Note: I use the Fedora 13 system *all* the time for my other Poky/Yocto builds.

Quoted 6 lines
Quoted 3 lines

I'm trying another build from scratch using a different build host to see if that makes a difference.

Do you have any other layers or customizations on top? (the metadata above suggests you build just pure Poky though)

I was trying this on a pure Poky/Yocto build.

Signature · 4 lines
------------------------------------------------------------
Gary Thomas                 |  Consulting for the
MLB Associates              |    Embedded world
------------------------------------------------------------
#7 lore raw
Quoted 93 lines · Gary Thomas
Quoted 92 lines · Denys Dmytriyenko
Quoted 90 lines · Gary Thomas
Quoted 74 lines · Denys Dmytriyenko
Quoted 73 lines · Gary Thomas
Quoted 72 lines · Denys Dmytriyenko
Quoted 70 lines · Gary Thomas

I just tried building (core-image-sato) for my BeagleBoneBlack using the latest Poky/Yocto master:

Build Configuration:

BB_VERSION        = "1.23.0"
BUILD_SYS         = "i686-linux"
NATIVELSBSTRING   = "Fedora-13"
TARGET_SYS        = "arm-poky-linux-gnueabi"
MACHINE           = "beaglebone"
DISTRO            = "poky"
DISTRO_VERSION    = "1.6+snapshot-20140411"
TUNE_FEATURES     = "armv7a vfp neon callconvention-hard cortexa8"
TARGET_FPU        = "vfp-neon"

meta
meta-yocto
meta-yocto-bsp = "master:863cc7483f5ee43189537940de8ee5c0964d24cc"

This built the kernel using SRCREV 928d7b2dda

I followed the bring-up instructions from README.hadware and the boot failed to even start the kernel. Here's what I see:

=============================== boot log ========================================= U-Boot 2013.07 (Apr 11 2014 - 15:03:04)

I2C:   ready
DRAM:  512 MiB
WARNING: Caches not enabled
NAND:  0 MiB
MMC:   OMAP SD/MMC: 0, OMAP SD/MMC: 1

*** Warning - readenv() failed, using default environment

musb-hdrc: ConfigData=0xde (UTMI-8, dyn FIFOs, HB-ISO Rx, HB-ISO Tx, SoftConn)
musb-hdrc: MHDRC RTL version 2.0
musb-hdrc: setup fifo_mode 4
musb-hdrc: 28/31 max ep, 16384/16384 memory
USB Peripheral mode controller at 47401000 using PIO, IRQ 0
musb-hdrc: ConfigData=0xde (UTMI-8, dyn FIFOs, HB-ISO Rx, HB-ISO Tx, SoftConn)
musb-hdrc: MHDRC RTL version 2.0
musb-hdrc: setup fifo_mode 4
musb-hdrc: 28/31 max ep, 16384/16384 memory
USB Host mode controller at 47401800 using PIO, IRQ 0
Net: <ethaddr> not set. Validating first E-fuse MAC
Phy not found
PHY reset timed out
cpsw, usb_ether
Hit any key to stop autoboot: 0
mmc0 is current device
SD/MMC found on device 0
reading uEnv.txt
** Unable to read file uEnv.txt **
4981688 bytes read in 613 ms (7.7 MiB/s)
29192 bytes read in 46 ms (619.1 KiB/s)
Booting from mmc ...
## Booting kernel from Legacy Image at 80200000 ...

   Image Name:   Linux-3.14.0-yocto-standard
   Image Type:   ARM Linux Kernel Image (uncompressed)
   Data Size:    4981624 Bytes = 4.8 MiB

Load Address: 80008000
Entry Point: 80008000
Verifying Checksum ... OK
## Flattened Device Tree blob at 80f80000
Booting using the fdt blob at 0x80f80000 Loading Kernel Image ... OK
Using Device Tree in place at 80f80000, end 80f8a207

Starting kernel ...
==================================================================================

Any ideas what I've done wrong?

Quoted 8 lines
Quoted 5 lines
Quoted 3 lines

Hmm, everything looks sane. What revision is your BBB? And did you press USER/BOOT button or erased eMMC partition per instructions?

Revision A5A, with an LCD cape

Hmm, I'm wondering if LCD cape conflicts here - there's no cape support in this BSP. Can you try w/o it?

Sure I can try it but I don't think that's it. I got the kernel that StefanX built and booted and tried it on my board and it came up. No clue why the kernels are different - ostensibly we both built the same image from the same meta data, but they are slightly different (only in size - I compared the System.map files from both builds and they contain exactly the same bits, just a few changes in memory layout which I can't explain).

Quoted 8 lines
Quoted 4 lines

Yeah, good point. Doesn't look like your cape causes the issue... The only other difference is in the host. Do you have access to another Linux box you can try? FWIW, I'm using 64-bit Gentoo with gcc-4.7.3. Yours is 32-bit Fedora 13, right?

I'll try Fedora 17 and Ubuntu 12.04 (x86_64) and see what happens.

Note: I use the Fedora 13 system *all* the time for my other Poky/Yocto builds.

Yeah, it usually shouldn't matter, as OE is very good at isolating host differences. But at this point we need to eliminate every variable...

Quoted 14 lines
Quoted 5 lines
Quoted 2 lines

I'm trying another build from scratch using a different build host to see if that makes a difference.

Do you have any other layers or customizations on top? (the metadata above suggests you build just pure Poky though)

I was trying this on a pure Poky/Yocto build.

Signature · 4 lines
------------------------------------------------------------
Gary Thomas                 |  Consulting for the
MLB Associates              |    Embedded world
------------------------------------------------------------
#8 lore raw
Quoted 99 lines · Denys Dmytriyenko
Quoted 98 lines · Gary Thomas
Quoted 96 lines · Denys Dmytriyenko
Quoted 84 lines · Gary Thomas
Quoted 74 lines · Denys Dmytriyenko
Quoted 73 lines · Gary Thomas
Quoted 72 lines · Denys Dmytriyenko
Quoted 70 lines · Gary Thomas

I just tried building (core-image-sato) for my BeagleBoneBlack using the latest Poky/Yocto master:

Build Configuration:

BB_VERSION        = "1.23.0"
BUILD_SYS         = "i686-linux"
NATIVELSBSTRING   = "Fedora-13"
TARGET_SYS        = "arm-poky-linux-gnueabi"
MACHINE           = "beaglebone"
DISTRO            = "poky"
DISTRO_VERSION    = "1.6+snapshot-20140411"
TUNE_FEATURES     = "armv7a vfp neon callconvention-hard cortexa8"
TARGET_FPU        = "vfp-neon"

meta
meta-yocto
meta-yocto-bsp = "master:863cc7483f5ee43189537940de8ee5c0964d24cc"

This built the kernel using SRCREV 928d7b2dda

I followed the bring-up instructions from README.hadware and the boot failed to even start the kernel. Here's what I see:

=============================== boot log ========================================= U-Boot 2013.07 (Apr 11 2014 - 15:03:04)

I2C:   ready
DRAM:  512 MiB
WARNING: Caches not enabled
NAND:  0 MiB
MMC:   OMAP SD/MMC: 0, OMAP SD/MMC: 1

*** Warning - readenv() failed, using default environment

musb-hdrc: ConfigData=0xde (UTMI-8, dyn FIFOs, HB-ISO Rx, HB-ISO Tx, SoftConn)
musb-hdrc: MHDRC RTL version 2.0
musb-hdrc: setup fifo_mode 4
musb-hdrc: 28/31 max ep, 16384/16384 memory
USB Peripheral mode controller at 47401000 using PIO, IRQ 0
musb-hdrc: ConfigData=0xde (UTMI-8, dyn FIFOs, HB-ISO Rx, HB-ISO Tx, SoftConn)
musb-hdrc: MHDRC RTL version 2.0
musb-hdrc: setup fifo_mode 4
musb-hdrc: 28/31 max ep, 16384/16384 memory
USB Host mode controller at 47401800 using PIO, IRQ 0
Net: <ethaddr> not set. Validating first E-fuse MAC
Phy not found
PHY reset timed out
cpsw, usb_ether
Hit any key to stop autoboot: 0
mmc0 is current device
SD/MMC found on device 0
reading uEnv.txt
** Unable to read file uEnv.txt **
4981688 bytes read in 613 ms (7.7 MiB/s)
29192 bytes read in 46 ms (619.1 KiB/s)
Booting from mmc ...
## Booting kernel from Legacy Image at 80200000 ...

   Image Name:   Linux-3.14.0-yocto-standard
   Image Type:   ARM Linux Kernel Image (uncompressed)
   Data Size:    4981624 Bytes = 4.8 MiB

Load Address: 80008000
Entry Point: 80008000
Verifying Checksum ... OK
## Flattened Device Tree blob at 80f80000
Booting using the fdt blob at 0x80f80000 Loading Kernel Image ... OK
Using Device Tree in place at 80f80000, end 80f8a207

Starting kernel ...
==================================================================================

Any ideas what I've done wrong?

Quoted 8 lines
Quoted 5 lines
Quoted 3 lines

Hmm, everything looks sane. What revision is your BBB? And did you press USER/BOOT button or erased eMMC partition per instructions?

Revision A5A, with an LCD cape

Hmm, I'm wondering if LCD cape conflicts here - there's no cape support in this BSP. Can you try w/o it?

Quoted 6 lines

Sure I can try it but I don't think that's it. I got the kernel that StefanX built and booted and tried it on my board and it came up. No clue why the kernels are different - ostensibly we both built the same image from the same meta data, but they are slightly different (only in size - I compared the System.map files from both builds and they contain exactly the same bits, just a few changes in memory layout which I can't explain).

Yeah, good point. Doesn't look like your cape causes the issue... The only other difference is in the host. Do you have access to another Linux box you can try? FWIW, I'm using 64-bit Gentoo with gcc-4.7.3. Yours is 32-bit Fedora 13, right?

Quoted 10 lines
Quoted 3 lines

I'll try Fedora 17 and Ubuntu 12.04 (x86_64) and see what happens.

Note: I use the Fedora 13 system *all* the time for my other Poky/Yocto builds.

Yeah, it usually shouldn't matter, as OE is very good at isolating host differences. But at this point we need to eliminate every variable...

Quoted 2 lines
Quoted 2 lines
Quoted 2 lines

I'm trying another build from scratch using a different build host to see if that makes a difference.

Very interesting results! These are the results from the build hosts I have:
Fedora 13 (i686) - fails
Fedora 17 (i686) - fails
Ubuntu 12.04 (x86_64) - boots

Note that I routinely build for other targets (which does imply other, mostly older, kernels) using all of these machines with no differences based on the build host.

I don't know what's unique about an x86_64 host, but it does seem to work.

I was trying this to see how the stock Yocto support for the BBB competes with building using meta-beaglebone which I've been using successfully. Based on these results, I'll be sticking with the meta-beaglebone approach for now (not just for the booting issue, but support for my LCD cape and other things that aren't there in the Yocto kernel)

Quoted 4 lines
Quoted 4 lines
Quoted 2 lines

Do you have any other layers or customizations on top? (the metadata above suggests you build just pure Poky though)

I was trying this on a pure Poky/Yocto build.

Signature · 4 lines
------------------------------------------------------------
Gary Thomas                 |  Consulting for the
MLB Associates              |    Embedded world
------------------------------------------------------------
#9 lore raw
Quoted 4 lines · Gary Thomas

Very interesting results! These are the results from the build hosts I have:
Fedora 13 (i686) - fails
Fedora 17 (i686) - fails
Ubuntu 12.04 (x86_64) - boots

Interesting indeed. I have no idea what's so special about Fedora host - this is the first time I hear about issues with it. I may try experimenting with different VMs once I have more time...

Quoted 3 lines

Note that I routinely build for other targets (which does imply other, mostly older, kernels) using all of these machines with no differences based on the build host.

Same here - building for different targets with different kernels from meta-ti and on 32 and 64 bit hosts and never had an issue like that...

Quoted 7 lines

I don't know what's unique about an x86_64 host, but it does seem to work.

I was trying this to see how the stock Yocto support for the BBB competes with building using meta-beaglebone which I've been using successfully. Based on these results, I'll be sticking with the meta-beaglebone approach for now (not just for the booting issue, but support for my LCD cape and other things that aren't there in the Yocto kernel)

This is not a "Yocto" kernel, this is a vanilla mainline 3.14 kernel (as a successor to 3.12 work done in meta-ti). Indeed, cape support is missing, as it is still not merged upstream, going through numerous iterations... The point of meta-beagleboard was to work on cape support and such, with the goal to upstream it, while meta-ti kept on working on SoC and peripheral support. Due to different circumstances, meta-beagleboard still uses 3.8 kernel. Once it is accepted upstream, it will be part of the mainline kernel and one of the later versions of the Yocto Project reference BSP.

Signature · 1 line
Denys
#10 lore raw
Quoted 9 lines · Denys Dmytriyenko
Quoted 4 lines · Gary Thomas

Very interesting results! These are the results from the build hosts I have:
Fedora 13 (i686) - fails
Fedora 17 (i686) - fails
Ubuntu 12.04 (x86_64) - boots

Interesting indeed. I have no idea what's so special about Fedora host - this is the first time I hear about issues with it. I may try experimenting with different VMs once I have more time...

I've been having a look at this. The biggest differences I can find between working and non working builds is the path length to the build directory for the kernel. This is from comparing vmlinux files from working and non working builds.

Works:
/home/paul/poky/build/tmp/work/beaglebone-poky-linux-gnueabi

Doesn't Work:
/media/data1/build1/poky/build/tmp/work/beaglebone-poky-linux-gnueabi

I also have been wondering if the version strings may be making a difference.

http://dan.rpsys.net/uImage-rp2 is a uImage from a broken build where I truncated the path length to a "working" build path length and patched in the same version strings:

const char linux_banner[] =

       "Linux version 3.14.0-yocto-standard (paul@ubuntu-build01) (gcc

version 4.8.2 (GCC) ) #1 PREEMPT Mon Apr 14 16:00:52 BST 2014\n";

const char linux_proc_banner[] = "%s version %s (paul@ubuntu-build01) (gcc version 4.8.2 (GCC) ) %s\n";

to init/version.c.

I don't have hardware and would be interested to know if the kernel linked to above works or not. If it doesn't, it rules out these path and string lengths, if it does work, it points to a problem there.

Cheers,

Richard

#11 lore raw
Quoted 33 lines · Richard Purdie
Quoted 9 lines · Denys Dmytriyenko
Quoted 4 lines · Gary Thomas

Very interesting results! These are the results from the build hosts I have:
Fedora 13 (i686) - fails
Fedora 17 (i686) - fails
Ubuntu 12.04 (x86_64) - boots

Interesting indeed. I have no idea what's so special about Fedora host - this is the first time I hear about issues with it. I may try experimenting with different VMs once I have more time...

I've been having a look at this. The biggest differences I can find between working and non working builds is the path length to the build directory for the kernel. This is from comparing vmlinux files from working and non working builds.

Works:
/home/paul/poky/build/tmp/work/beaglebone-poky-linux-gnueabi

Doesn't Work:
/media/data1/build1/poky/build/tmp/work/beaglebone-poky-linux-gnueabi

I also have been wondering if the version strings may be making a difference.

http://dan.rpsys.net/uImage-rp2 is a uImage from a broken build where I truncated the path length to a "working" build path length and patched in the same version strings:

const char linux_banner[] =

       "Linux version 3.14.0-yocto-standard (paul@ubuntu-build01) (gcc

version 4.8.2 (GCC) ) #1 PREEMPT Mon Apr 14 16:00:52 BST 2014\n";

Quoted 8 lines

const char linux_proc_banner[] = "%s version %s (paul@ubuntu-build01) (gcc version 4.8.2 (GCC) ) %s\n";

to init/version.c.

I don't have hardware and would be interested to know if the kernel linked to above works or not. If it doesn't, it rules out these path and string lengths, if it does work, it points to a problem there.

Richard,

Good catch! It boots:

U-Boot SPL 2013.07 (Apr 10 2014 - 04:47:32)
musb-hdrc: ConfigData=0xde (UTMI-8, dyn FIFOs, HB-ISO Rx, HB-ISO Tx, SoftConn)
musb-hdrc: MHDRC RTL version 2.0
musb-hdrc: setup fifo_mode 4
musb-hdrc: 28/31 max ep, 16384/16384 memory
USB Peripheral mode controller at 47401000 using PIO, IRQ 0
musb-hdrc: ConfigData=0xde (UTMI-8, dyn FIFOs, HB-ISO Rx, HB-ISO Tx, SoftConn)
musb-hdrc: MHDRC RTL version 2.0
musb-hdrc: setup fifo_mode 4
musb-hdrc: 28/31 max ep, 16384/16384 memory
USB Host mode controller at 47401800 using PIO, IRQ 0
OMAP SD/MMC: 0
reading args
spl: error reading image args, err - -1
reading u-boot.img
reading u-boot.img

U-Boot 2013.07 (Apr 10 2014 - 04:47:32)

I2C:   ready
DRAM:  512 MiB
WARNING: Caches not enabled
NAND:  0 MiB
MMC:   OMAP SD/MMC: 0, OMAP SD/MMC: 1

*** Warning - readenv() failed, using default environment

musb-hdrc: ConfigData=0xde (UTMI-8, dyn FIFOs, HB-ISO Rx, HB-ISO Tx, SoftConn)
musb-hdrc: MHDRC RTL version 2.0
musb-hdrc: setup fifo_mode 4
musb-hdrc: 28/31 max ep, 16384/16384 memory
USB Peripheral mode controller at 47401000 using PIO, IRQ 0
musb-hdrc: ConfigData=0xde (UTMI-8, dyn FIFOs, HB-ISO Rx, HB-ISO Tx, SoftConn)
musb-hdrc: MHDRC RTL version 2.0
musb-hdrc: setup fifo_mode 4
musb-hdrc: 28/31 max ep, 16384/16384 memory
USB Host mode controller at 47401800 using PIO, IRQ 0
Net: <ethaddr> not set. Validating first E-fuse MAC
cpsw, usb_ether
Hit any key to stop autoboot: 0
mmc0 is current device
SD/MMC found on device 0
reading uEnv.txt
** Unable to read file uEnv.txt **
4985376 bytes read in 853 ms (5.6 MiB/s)
29192 bytes read in 28 ms (1017.6 KiB/s)
Booting from mmc ...
## Booting kernel from Legacy Image at 80200000 ...

   Image Name:   Linux-3.14.0-yocto-standard
   Image Type:   ARM Linux Kernel Image (uncompressed)
   Data Size:    4985312 Bytes = 4.8 MiB

Load Address: 80008000
Entry Point: 80008000
Verifying Checksum ... OK
## Flattened Device Tree blob at 80f80000
Booting using the fdt blob at 0x80f80000 Loading Kernel Image ... OK
Using Device Tree in place at 80f80000, end 80f8a207

Starting kernel ...

Booting Linux on physical CPU 0x0
Initializing cgroup subsys cpuset
Initializing cgroup subsys cpu
Initializing cgroup subsys cpuacct
Linux version 3.14.0-yocto-standard (paul@ubuntu-build01) (gcc version 4.8.2 (GCC) ) #1 PREEMPT Mon Apr 14 16:00:52 BST 2014
CPU: ARMv7 Processor [413fc082] revision 2 (ARMv7), cr=10c53c7d
CPU: PIPT / VIPT nonaliasing data cache, VIPT aliasing instruction cache
Machine model: TI AM335x BeagleBone
cma: CMA: reserved 16 MiB at 9e800000
Memory policy: Data cache writeback
CPU: All CPU(s) started in SVC mode.
AM335X ES2.0 (sgx neon )
Built 1 zonelists in Zone order, mobility grouping on. Total pages: 129792
Kernel command line: console=ttyO0,115200n8 root=/dev/mmcblk0p2 ro rootfstype=ext4 rootwait
PID hash table entries: 2048 (order: 1, 8192 bytes)
Dentry cache hash table entries: 65536 (order: 6, 262144 bytes)
Inode-cache hash table entries: 32768 (order: 5, 131072 bytes)
allocated 1048576 bytes of page_cgroup
please try 'cgroup_disable=memory' option if you don't want memory cgroups
Memory: 489452K/523264K available (7491K kernel code, 520K rwdata, 2448K rodata, 488K init, 757K bss, 33812K reserved, 0K highmem)

[snip]

Signature · 1 line
Denys
#12 lore raw
Quoted 39 lines · Denys Dmytriyenko
Quoted 38 lines · Richard Purdie
Quoted 9 lines · Denys Dmytriyenko
Quoted 4 lines · Gary Thomas

Very interesting results! These are the results from the build hosts I have:
Fedora 13 (i686) - fails
Fedora 17 (i686) - fails
Ubuntu 12.04 (x86_64) - boots

Interesting indeed. I have no idea what's so special about Fedora host - this is the first time I hear about issues with it. I may try experimenting with different VMs once I have more time...

I've been having a look at this. The biggest differences I can find between working and non working builds is the path length to the build directory for the kernel. This is from comparing vmlinux files from working and non working builds.

Works:
/home/paul/poky/build/tmp/work/beaglebone-poky-linux-gnueabi

Doesn't Work:
/media/data1/build1/poky/build/tmp/work/beaglebone-poky-linux-gnueabi

I also have been wondering if the version strings may be making a difference.

http://dan.rpsys.net/uImage-rp2 is a uImage from a broken build where I truncated the path length to a "working" build path length and patched in the same version strings:

const char linux_banner[] =

       "Linux version 3.14.0-yocto-standard (paul@ubuntu-build01) (gcc

version 4.8.2 (GCC) ) #1 PREEMPT Mon Apr 14 16:00:52 BST 2014\n";

const char linux_proc_banner[] = "%s version %s (paul@ubuntu-build01) (gcc version 4.8.2 (GCC) ) %s\n";

to init/version.c.

Quoted 7 lines
Quoted 3 lines

I don't have hardware and would be interested to know if the kernel linked to above works or not. If it doesn't, it rules out these path and string lengths, if it does work, it points to a problem there.

Richard,

Good catch! It boots:

Thanks Denys, this helps narrow down the issue. I've shared
http://dan.rpsys.net/uImage-rp3 which is the same as the last one but with my changes to version.c reverted. The one should tell us if its the paths or the strings.

I'm guessing the path problem is more likely but anything is possible. This is starting to look like some kind of compiler or linker issue. If it is that, it would help to have more data points about what works and what doesn't. With that in mind could people who have good or bad builds please share the paths they built the kernels in so we can see if we can spot some kind of pattern.

Cheers,

Richard

#13 lore raw
Quoted 44 lines · Richard Purdie
Quoted 43 lines · Denys Dmytriyenko
Quoted 41 lines · Richard Purdie
Quoted 9 lines · Denys Dmytriyenko
Quoted 4 lines · Gary Thomas

Very interesting results! These are the results from the build hosts I have:
Fedora 13 (i686) - fails
Fedora 17 (i686) - fails
Ubuntu 12.04 (x86_64) - boots

Interesting indeed. I have no idea what's so special about Fedora host - this is the first time I hear about issues with it. I may try experimenting with different VMs once I have more time...

I've been having a look at this. The biggest differences I can find between working and non working builds is the path length to the build directory for the kernel. This is from comparing vmlinux files from working and non working builds.

Works:
/home/paul/poky/build/tmp/work/beaglebone-poky-linux-gnueabi

Doesn't Work:
/media/data1/build1/poky/build/tmp/work/beaglebone-poky-linux-gnueabi

I also have been wondering if the version strings may be making a difference.

http://dan.rpsys.net/uImage-rp2 is a uImage from a broken build where I truncated the path length to a "working" build path length and patched in the same version strings:

const char linux_banner[] =

       "Linux version 3.14.0-yocto-standard (paul@ubuntu-build01) (gcc

version 4.8.2 (GCC) ) #1 PREEMPT Mon Apr 14 16:00:52 BST 2014\n";

const char linux_proc_banner[] = "%s version %s (paul@ubuntu-build01) (gcc version 4.8.2 (GCC) ) %s\n";

to init/version.c.

I don't have hardware and would be interested to know if the kernel linked to above works or not. If it doesn't, it rules out these path and string lengths, if it does work, it points to a problem there.

Quoted 8 lines
Quoted 3 lines

Richard,

Good catch! It boots:

Thanks Denys, this helps narrow down the issue. I've shared
http://dan.rpsys.net/uImage-rp3 which is the same as the last one but with my changes to version.c reverted. The one should tell us if its the paths or the strings.

This one also boots for me:

Linux version 3.14.0-yocto-standard (richard@ted) (gcc version 4.8.2 (GCC) ) #2 PREEMPT Tue Apr 15 05:40:19 IST 2014

Quoted 11 lines

I'm guessing the path problem is more likely but anything is possible. This is starting to look like some kind of compiler or linker issue. If it is that, it would help to have more data points about what works and what doesn't. With that in mind could people who have good or bad builds please share the paths they built the kernels in so we can see if we can spot some kind of pattern.

Cheers,

Richard

#14 lore raw
Quoted 49 lines · Denys Dmytriyenko
Quoted 48 lines · Richard Purdie
Quoted 46 lines · Denys Dmytriyenko
Quoted 33 lines · Richard Purdie
Quoted 9 lines · Denys Dmytriyenko
Quoted 4 lines · Gary Thomas

Very interesting results! These are the results from the build hosts I have:
Fedora 13 (i686) - fails
Fedora 17 (i686) - fails
Ubuntu 12.04 (x86_64) - boots

Interesting indeed. I have no idea what's so special about Fedora host - this is the first time I hear about issues with it. I may try experimenting with different VMs once I have more time...

I've been having a look at this. The biggest differences I can find between working and non working builds is the path length to the build directory for the kernel. This is from comparing vmlinux files from working and non working builds.

Works:
/home/paul/poky/build/tmp/work/beaglebone-poky-linux-gnueabi

Doesn't Work:
/media/data1/build1/poky/build/tmp/work/beaglebone-poky-linux-gnueabi

I also have been wondering if the version strings may be making a difference.

http://dan.rpsys.net/uImage-rp2 is a uImage from a broken build where I truncated the path length to a "working" build path length and patched in the same version strings:

const char linux_banner[] =

       "Linux version 3.14.0-yocto-standard (paul@ubuntu-build01) (gcc

version 4.8.2 (GCC) ) #1 PREEMPT Mon Apr 14 16:00:52 BST 2014\n";

Quoted 8 lines

const char linux_proc_banner[] = "%s version %s (paul@ubuntu-build01) (gcc version 4.8.2 (GCC) ) %s\n";

to init/version.c.

I don't have hardware and would be interested to know if the kernel linked to above works or not. If it doesn't, it rules out these path and string lengths, if it does work, it points to a problem there.

Richard,

Good catch! It boots:

Quoted 16 lines
Quoted 4 lines

Thanks Denys, this helps narrow down the issue. I've shared
http://dan.rpsys.net/uImage-rp3 which is the same as the last one but with my changes to version.c reverted. The one should tell us if its the paths or the strings.

This one also boots for me:

Linux version 3.14.0-yocto-standard (richard@ted) (gcc version 4.8.2 (GCC) ) #2 PREEMPT Tue Apr 15 05:40:19 IST 2014

Quoted 6 lines

I'm guessing the path problem is more likely but anything is possible. This is starting to look like some kind of compiler or linker issue. If it is that, it would help to have more data points about what works and what doesn't. With that in mind could people who have good or bad builds please share the paths they built the kernels in so we can see if we can spot some kind of pattern.

BTW, my path is /OE/RAM/poky/tmp/work/beaglebone-poky-linux-gnueabi and it works.

Signature · 1 line
Denys
#15 lore raw
Quoted 67 lines · Denys Dmytriyenko
Quoted 49 lines · Denys Dmytriyenko
Quoted 48 lines · Richard Purdie
Quoted 46 lines · Denys Dmytriyenko
Quoted 33 lines · Richard Purdie
Quoted 9 lines · Denys Dmytriyenko
Quoted 4 lines · Gary Thomas

Very interesting results! These are the results from the build hosts I have:
Fedora 13 (i686) - fails
Fedora 17 (i686) - fails
Ubuntu 12.04 (x86_64) - boots

Interesting indeed. I have no idea what's so special about Fedora host - this is the first time I hear about issues with it. I may try experimenting with different VMs once I have more time...

I've been having a look at this. The biggest differences I can find between working and non working builds is the path length to the build directory for the kernel. This is from comparing vmlinux files from working and non working builds.

Works:
/home/paul/poky/build/tmp/work/beaglebone-poky-linux-gnueabi

Doesn't Work:
/media/data1/build1/poky/build/tmp/work/beaglebone-poky-linux-gnueabi

I also have been wondering if the version strings may be making a difference.

http://dan.rpsys.net/uImage-rp2 is a uImage from a broken build where I truncated the path length to a "working" build path length and patched in the same version strings:

const char linux_banner[] =

       "Linux version 3.14.0-yocto-standard (paul@ubuntu-build01) (gcc

version 4.8.2 (GCC) ) #1 PREEMPT Mon Apr 14 16:00:52 BST 2014\n";

Quoted 8 lines

const char linux_proc_banner[] = "%s version %s (paul@ubuntu-build01) (gcc version 4.8.2 (GCC) ) %s\n";

to init/version.c.

I don't have hardware and would be interested to know if the kernel linked to above works or not. If it doesn't, it rules out these path and string lengths, if it does work, it points to a problem there.

Richard,

Good catch! It boots:

Quoted 16 lines
Quoted 4 lines

Thanks Denys, this helps narrow down the issue. I've shared
http://dan.rpsys.net/uImage-rp3 which is the same as the last one but with my changes to version.c reverted. The one should tell us if its the paths or the strings.

This one also boots for me:

Linux version 3.14.0-yocto-standard (richard@ted) (gcc version 4.8.2 (GCC) ) #2 PREEMPT Tue Apr 15 05:40:19 IST 2014

Quoted 6 lines

I'm guessing the path problem is more likely but anything is possible. This is starting to look like some kind of compiler or linker issue. If it is that, it would help to have more data points about what works and what doesn't. With that in mind could people who have good or bad builds please share the paths they built the kernels in so we can see if we can spot some kind of pattern.

Quoted 3 lines

BTW, my path is /OE/RAM/poky/tmp/work/beaglebone-poky-linux-gnueabi and it works.

I can confirm:
build dir in /home/stefans/b1 works,
but /home/stefans/yocto/poky/build doesn't.

Cheers,
Stefan

Quoted 2 lines
Signature · 1 line
Denys
#16 lore raw
Quoted 70 lines · StefanX
Quoted 67 lines · Denys Dmytriyenko
Quoted 65 lines · Denys Dmytriyenko
Quoted 44 lines · Richard Purdie
Quoted 43 lines · Denys Dmytriyenko
Quoted 41 lines · Richard Purdie
Quoted 9 lines · Denys Dmytriyenko
Quoted 4 lines · Gary Thomas

Very interesting results! These are the results from the build hosts I have:
Fedora 13 (i686) - fails
Fedora 17 (i686) - fails
Ubuntu 12.04 (x86_64) - boots

Interesting indeed. I have no idea what's so special about Fedora host - this is the first time I hear about issues with it. I may try experimenting with different VMs once I have more time...

I've been having a look at this. The biggest differences I can find between working and non working builds is the path length to the build directory for the kernel. This is from comparing vmlinux files from working and non working builds.

Works:
/home/paul/poky/build/tmp/work/beaglebone-poky-linux-gnueabi

Doesn't Work:
/media/data1/build1/poky/build/tmp/work/beaglebone-poky-linux-gnueabi

I also have been wondering if the version strings may be making a difference.

http://dan.rpsys.net/uImage-rp2 is a uImage from a broken build where I truncated the path length to a "working" build path length and patched in the same version strings:

const char linux_banner[] =

       "Linux version 3.14.0-yocto-standard (paul@ubuntu-build01) (gcc

version 4.8.2 (GCC) ) #1 PREEMPT Mon Apr 14 16:00:52 BST 2014\n";

const char linux_proc_banner[] = "%s version %s (paul@ubuntu-build01) (gcc version 4.8.2 (GCC) ) %s\n";

to init/version.c.

I don't have hardware and would be interested to know if the kernel linked to above works or not. If it doesn't, it rules out these path and string lengths, if it does work, it points to a problem there.

Quoted 8 lines
Quoted 3 lines

Richard,

Good catch! It boots:

Thanks Denys, this helps narrow down the issue. I've shared
http://dan.rpsys.net/uImage-rp3 which is the same as the last one but with my changes to version.c reverted. The one should tell us if its the paths or the strings.

This one also boots for me:

Linux version 3.14.0-yocto-standard (richard@ted) (gcc version 4.8.2 (GCC) ) #2 PREEMPT Tue Apr 15 05:40:19 IST 2014

Quoted 6 lines

I'm guessing the path problem is more likely but anything is possible. This is starting to look like some kind of compiler or linker issue. If it is that, it would help to have more data points about what works and what doesn't. With that in mind could people who have good or bad builds please share the paths they built the kernels in so we can see if we can spot some kind of pattern.

Quoted 8 lines
Quoted 3 lines

BTW, my path is /OE/RAM/poky/tmp/work/beaglebone-poky-linux-gnueabi and it works.

I can confirm:
build dir in /home/stefans/b1 works,
but /home/stefans/yocto/poky/build doesn't.

But this works
[stefans@firebird bu]$ pwd
/home/stefans/yocto/poky/bu
[stefans@firebird bu]$ echo pwd| wc -c
28

But 29 doesn't.
So, what now?

Cheers,
Stefan

Quoted 6 lines

Cheers,
Stefan

Quoted 2 lines
Signature · 1 line
Denys
#44 lore raw
In reply to Stanacar, StefanX
Quoted 71 lines · StefanX
Quoted 67 lines · Denys Dmytriyenko
Quoted 65 lines · Denys Dmytriyenko
Quoted 44 lines · Richard Purdie
Quoted 43 lines · Denys Dmytriyenko
Quoted 41 lines · Richard Purdie
Quoted 9 lines · Denys Dmytriyenko
Quoted 4 lines · Gary Thomas

Very interesting results! These are the results from the build hosts I have:
Fedora 13 (i686) - fails
Fedora 17 (i686) - fails
Ubuntu 12.04 (x86_64) - boots

Interesting indeed. I have no idea what's so special about Fedora host - this is the first time I hear about issues with it. I may try experimenting with different VMs once I have more time...

I've been having a look at this. The biggest differences I can find between working and non working builds is the path length to the build directory for the kernel. This is from comparing vmlinux files from working and non working builds.

Works:
/home/paul/poky/build/tmp/work/beaglebone-poky-linux-gnueabi

Doesn't Work:
/media/data1/build1/poky/build/tmp/work/beaglebone-poky-linux-gnueabi

I also have been wondering if the version strings may be making a difference.

http://dan.rpsys.net/uImage-rp2 is a uImage from a broken build where I truncated the path length to a "working" build path length and patched in the same version strings:

const char linux_banner[] =

       "Linux version 3.14.0-yocto-standard (paul@ubuntu-build01) (gcc

version 4.8.2 (GCC) ) #1 PREEMPT Mon Apr 14 16:00:52 BST 2014\n";

const char linux_proc_banner[] = "%s version %s (paul@ubuntu-build01) (gcc version 4.8.2 (GCC) ) %s\n";

to init/version.c.

I don't have hardware and would be interested to know if the kernel linked to above works or not. If it doesn't, it rules out these path and string lengths, if it does work, it points to a problem there.

Quoted 8 lines
Quoted 3 lines

Richard,

Good catch! It boots:

Thanks Denys, this helps narrow down the issue. I've shared
http://dan.rpsys.net/uImage-rp3 which is the same as the last one but with my changes to version.c reverted. The one should tell us if its the paths or the strings.

This one also boots for me:

Linux version 3.14.0-yocto-standard (richard@ted) (gcc version 4.8.2 (GCC) ) #2 PREEMPT Tue Apr 15 05:40:19 IST 2014

Quoted 6 lines

I'm guessing the path problem is more likely but anything is possible. This is starting to look like some kind of compiler or linker issue. If it is that, it would help to have more data points about what works and what doesn't. With that in mind could people who have good or bad builds please share the paths they built the kernels in so we can see if we can spot some kind of pattern.

Quoted 7 lines
Quoted 3 lines

BTW, my path is /OE/RAM/poky/tmp/work/beaglebone-poky-linux-gnueabi and it works.

I can confirm:
build dir in /home/stefans/b1 works,
but /home/stefans/yocto/poky/build doesn't.

My paths:
/local/bbb_yocto_2014-04-14 - works
/home/local/bbb_yocto_2014-04-14 - fails
/raid/work/tmp/bbb_yocto_2014-04-14 - fails
/tmp/bbb_yocto_2014-04-15 - works
/tmp/work/bbb_yocto_2014-04-15 - fails
/tmp/w2/bbb_yocto_2014-04-15 - fails
/tmp/w/bbb_yocto_2014-04-15 - works
Note: the last 5 were all built on the same host

Signature · 4 lines
------------------------------------------------------------
Gary Thomas                 |  Consulting for the
MLB Associates              |    Embedded world
------------------------------------------------------------
#17 lore raw
In reply to Stanacar, StefanX
Quoted 90 lines · StefanX
Quoted 70 lines · StefanX
Quoted 67 lines · Denys Dmytriyenko
Quoted 65 lines · Denys Dmytriyenko
Quoted 44 lines · Richard Purdie
Quoted 43 lines · Denys Dmytriyenko
Quoted 41 lines · Richard Purdie
Quoted 9 lines · Denys Dmytriyenko
Quoted 4 lines · Gary Thomas

Very interesting results! These are the results from the build hosts I have:
Fedora 13 (i686) - fails
Fedora 17 (i686) - fails
Ubuntu 12.04 (x86_64) - boots

Interesting indeed. I have no idea what's so special about Fedora host - this is the first time I hear about issues with it. I may try experimenting with different VMs once I have more time...

I've been having a look at this. The biggest differences I can find between working and non working builds is the path length to the build directory for the kernel. This is from comparing vmlinux files from working and non working builds.

Works:
/home/paul/poky/build/tmp/work/beaglebone-poky-linux-gnueabi

Doesn't Work:
/media/data1/build1/poky/build/tmp/work/beaglebone-poky-linux-gnueabi

I also have been wondering if the version strings may be making a difference.

http://dan.rpsys.net/uImage-rp2 is a uImage from a broken build where I truncated the path length to a "working" build path length and patched in the same version strings:

const char linux_banner[] =

       "Linux version 3.14.0-yocto-standard (paul@ubuntu-build01) (gcc

version 4.8.2 (GCC) ) #1 PREEMPT Mon Apr 14 16:00:52 BST 2014\n";

const char linux_proc_banner[] = "%s version %s (paul@ubuntu-build01) (gcc version 4.8.2 (GCC) ) %s\n";

to init/version.c.

I don't have hardware and would be interested to know if the kernel linked to above works or not. If it doesn't, it rules out these path and string lengths, if it does work, it points to a problem there.

Quoted 8 lines
Quoted 3 lines

Richard,

Good catch! It boots:

Thanks Denys, this helps narrow down the issue. I've shared
http://dan.rpsys.net/uImage-rp3 which is the same as the last one but with my changes to version.c reverted. The one should tell us if its the paths or the strings.

This one also boots for me:

Linux version 3.14.0-yocto-standard (richard@ted) (gcc version 4.8.2 (GCC) ) #2 PREEMPT Tue Apr 15 05:40:19 IST 2014

Quoted 6 lines

I'm guessing the path problem is more likely but anything is possible. This is starting to look like some kind of compiler or linker issue. If it is that, it would help to have more data points about what works and what doesn't. With that in mind could people who have good or bad builds please share the paths they built the kernels in so we can see if we can spot some kind of pattern.

Quoted 8 lines
Quoted 3 lines

BTW, my path is /OE/RAM/poky/tmp/work/beaglebone-poky-linux-gnueabi and it works.

I can confirm:
build dir in /home/stefans/b1 works,
but /home/stefans/yocto/poky/build doesn't.

But this works
[stefans@firebird bu]$ pwd
/home/stefans/yocto/poky/bu
[stefans@firebird bu]$ echo pwd| wc -c
28

Quoted 2 lines

But 29 doesn't.
So, what now?

Hard to know, it might take debugging via JTAG to find out what's happening since when it fails, it's so dead that nothing makes it to the [dmesg] log buffer :-(

Sadly, I can't help much with this as I don't have a JTAG debug setup for the BBB.

Signature · 4 lines
------------------------------------------------------------
Gary Thomas                 |  Consulting for the
MLB Associates              |    Embedded world
------------------------------------------------------------
#18 lore raw
In reply to Stanacar, StefanX
Quoted 9 lines · StefanX
Quoted 8 lines · StefanX
Quoted 7 lines · Denys Dmytriyenko
Quoted 6 lines · Denys Dmytriyenko
Quoted 5 lines · Richard Purdie
Quoted 4 lines · Denys Dmytriyenko
Quoted 3 lines · Richard Purdie
Quoted 2 lines · Denys Dmytriyenko
Quoted 1 line · Gary Thomas

Very interesting results! These are the results from the

build hosts I have:

Quoted 95 lines
Quoted 80 lines
Quoted 80 lines
Quoted 79 lines
Quoted 56 lines
Quoted 55 lines
Quoted 46 lines
Quoted 8 lines
Quoted 3 lines

Fedora 13 (i686) - fails
Fedora 17 (i686) - fails
Ubuntu 12.04 (x86_64) - boots

Interesting indeed. I have no idea what's so special about Fedora host - this is the first time I hear about issues with it. I may try experimenting with different VMs once I have more time...

I've been having a look at this. The biggest differences I can find
between working and non working builds is the path length to the build
directory for the kernel. This is from comparing vmlinux files from
working and non working builds.

Works:
/home/paul/poky/build/tmp/work/beaglebone-poky-linux-gnueabi

Doesn't Work:
/media/data1/build1/poky/build/tmp/work/beaglebone-poky-linux-gn ueabi

I also have been wondering if the version strings may be making a
difference.

http://dan.rpsys.net/uImage-rp2 is a uImage from a broken build where I
truncated the path length to a "working" build path length and patched
in the same version strings:

const char linux_banner[] =

"Linux version 3.14.0-yocto-standard
(paul@ubuntu-build01) (gcc

version 4.8.2 (GCC) ) #1 PREEMPT Mon Apr 14 16:00:52 BST 2014\n";

const char linux_proc_banner[] = "%s version %s
(paul@ubuntu-build01)
(gcc version 4.8.2 (GCC) ) %s\n";

Quoted 7 lines

to init/version.c.

I don't have hardware and would be interested to know if the kernel
linked to above works or not. If it doesn't, it rules out these path and
string lengths, if it does work, it points to a problem there.

Richard,

Quoted 7 lines
Quoted 1 line

Good catch! It boots:

Thanks Denys, this helps narrow down the issue. I've shared
http://dan.rpsys.net/uImage-rp3 which is the same as the last one but
with my changes to version.c reverted. The one should tell us if its the
paths or the strings.

This one also boots for me:

Linux version 3.14.0-yocto-standard (richard@ted) (gcc version 4.8.2 (GCC) ) #2 PREEMPT Tue Apr 15 05:40:19 IST 2014> > >

Quoted 11 lines

I'm guessing the path problem is more likely but anything is possible.
This is starting to look like some kind of compiler or linker issue. If
it is that, it would help to have more data points about what works and
what doesn't. With that in mind could people who have good or bad builds
please share the paths they built the kernels in so we can see if we can
spot some kind of pattern.

Quoted 7 lines
Quoted 2 lines

BTW, my path is /OE/RAM/poky/tmp/work/beaglebone-poky-linux-gnueabi and it works.

I can confirm:
build dir in /home/stefans/b1 works,

but /home/stefans/yocto/poky/build doesn't.

But this works
[stefans@firebird bu]$ pwd
/home/stefans/yocto/poky/bu
[stefans@firebird bu]$ echo pwd| wc -c
28

Quoted 2 lines

But 29 doesn't.
So, what now?

Some other things I tried with a "long" TMPDIR path (note that it's the TMPDIR path that makes the difference - in my tests I've been using /home/paul/poky/build2/much/longer/path/to/tmp). None of this helped:

* kernel built with gcc 4.7.2 and binutils 2.23.2
* u-boot built with gcc 4.7.2 and binutils 2.23.2
* u-boot from http://downloads.angstrom-distribution.org/demo/beaglebone/
* earlyprintk and CONFIG_DEBUG_LL - no additional output printed

I think we're now at the point where we'd benefit from someone with better knowledge debugging the issue.

Cheers,
Paul

Signature · 2 lines
Paul Eggleton
Intel Open Source Technology Centre
#19 lore raw
Quoted 122 lines

-----Original Message-----
From: [email protected] [mailto:yocto- [email protected]] On Behalf Of Paul Eggleton
Sent: Tuesday, April 15, 2014 5:50 PM
To: [email protected]
Subject: Re: [yocto] BBB doesn't boot

Quoted 8 lines · StefanX
Quoted 7 lines · StefanX
Quoted 6 lines · Denys Dmytriyenko
Quoted 5 lines · Denys Dmytriyenko
Quoted 4 lines · Richard Purdie
Quoted 3 lines · Denys Dmytriyenko
Quoted 2 lines · Richard Purdie
Quoted 1 line · Denys Dmytriyenko

On Mon, Apr 14, 2014 at 02:11:05PM -0600, Gary Thomas

Quoted 2 lines · wrote:
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines

Very interesting results! These are the results from the

build hosts I have:

Quoted 9 lines
Quoted 9 lines
Quoted 9 lines
Quoted 9 lines
Quoted 9 lines
Quoted 9 lines
Quoted 9 lines
Quoted 8 lines
Quoted 3 lines

Fedora 13 (i686) - fails
Fedora 17 (i686) - fails
Ubuntu 12.04 (x86_64) - boots

Interesting indeed. I have no idea what's so special about Fedora host - this is the first time I hear about issues with it. I may try experimenting with different VMs once I have more time...

Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines

I've been having a look at this. The biggest differences I can find between working and non working builds is the path length to the build directory for the kernel. This is from comparing vmlinux files from working and non working builds.

Works:
/home/paul/poky/build/tmp/work/beaglebone-poky-linux-

gnuea

Quoted 55 lines
Quoted 55 lines
Quoted 55 lines
Quoted 54 lines
Quoted 33 lines
Quoted 33 lines
Quoted 32 lines

bi

Doesn't Work:
/media/data1/build1/poky/build/tmp/work/beaglebone-poky-li nux-gn
ueabi

I also have been wondering if the version strings may be making a difference.

http://dan.rpsys.net/uImage-rp2 is a uImage from a broken build where I truncated the path length to a "working" build path length and patched in the same version strings:

const char linux_banner[] =

"Linux version 3.14.0-yocto-standard
(paul@ubuntu-build01) (gcc

version 4.8.2 (GCC) ) #1 PREEMPT Mon Apr 14 16:00:52 BST 2014\n";

const char linux_proc_banner[] = "%s version %s (paul@ubuntu-build01)
(gcc version 4.8.2 (GCC) ) %s\n";

to init/version.c.

I don't have hardware and would be interested to know if the kernel linked to above works or not. If it doesn't, it rules out these path and string lengths, if it does work, it points to a problem there.

Quoted 7 lines
Quoted 3 lines

Richard,

Good catch! It boots:

Thanks Denys, this helps narrow down the issue. I've shared
http://dan.rpsys.net/uImage-rp3 which is the same as the last one but with my changes to version.c reverted. The one should tell us if its the paths or the strings.

This one also boots for me:

Linux version 3.14.0-yocto-standard (richard@ted) (gcc version 4.8.2
(GCC) ) #2 PREEMPT Tue Apr 15 05:40:19 IST 2014> > >

Quoted 8 lines

I'm guessing the path problem is more likely but anything is possible.
This is starting to look like some kind of compiler or linker issue. If
it is that, it would help to have more data points about what works and what doesn't. With that in mind could people who have good or bad builds please share the paths they built the kernels in so we can see if we can spot some kind of pattern.

Quoted 2 lines
Quoted 2 lines
Quoted 2 lines

BTW, my path is /OE/RAM/poky/tmp/work/beaglebone-poky-linux-gnueabi and it

works.

Quoted 13 lines
Quoted 5 lines

I can confirm:
build dir in /home/stefans/b1 works,

but /home/stefans/yocto/poky/build doesn't.

But this works
[stefans@firebird bu]$ pwd
/home/stefans/yocto/poky/bu
[stefans@firebird bu]$ echo pwd| wc -c
28

Quoted 2 lines

But 29 doesn't.
So, what now?

Some other things I tried with a "long" TMPDIR path (note that it's the TMPDIR path that makes the difference - in my tests I've been using /home/paul/poky/build2/much/longer/path/to/tmp). None of this helped:

* kernel built with gcc 4.7.2 and binutils 2.23.2
* u-boot built with gcc 4.7.2 and binutils 2.23.2
* u-boot from http://downloads.angstrom-
distribution.org/demo/beaglebone/
* earlyprintk and CONFIG_DEBUG_LL - no additional output printed

Quoted 6 lines

I think we're now at the point where we'd benefit from someone with better knowledge debugging the issue.

Cheers,
Paul

I don't know if this is helpful but this TMPDIR path size issue reminded me of a bug I verified some time ago:
https://bugzilla.yoctoproject.org/show_bug.cgi?id=3989

Regards,
Corneliu

#20 lore raw
Quoted 125 lines · CorneliuX
Quoted 125 lines

-----Original Message-----
From: [email protected] [mailto:yocto- [email protected]] On Behalf Of Paul Eggleton
Sent: Tuesday, April 15, 2014 5:50 PM
To: [email protected]
Subject: Re: [yocto] BBB doesn't boot

Quoted 8 lines · StefanX
Quoted 7 lines · StefanX
Quoted 6 lines · Denys Dmytriyenko
Quoted 5 lines · Denys Dmytriyenko
Quoted 4 lines · Richard Purdie
Quoted 3 lines · Denys Dmytriyenko
Quoted 2 lines · Richard Purdie
Quoted 1 line · Denys Dmytriyenko

On Mon, Apr 14, 2014 at 02:11:05PM -0600, Gary Thomas

Quoted 2 lines · wrote:
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines

Very interesting results! These are the results from the

build hosts I have:

Quoted 9 lines
Quoted 9 lines
Quoted 9 lines
Quoted 9 lines
Quoted 9 lines
Quoted 9 lines
Quoted 9 lines
Quoted 8 lines
Quoted 3 lines

Fedora 13 (i686) - fails
Fedora 17 (i686) - fails
Ubuntu 12.04 (x86_64) - boots

Interesting indeed. I have no idea what's so special about Fedora host - this is the first time I hear about issues with it. I may try experimenting with different VMs once I have more time...

Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines

I've been having a look at this. The biggest differences I can find between working and non working builds is the path length to the build directory for the kernel. This is from comparing vmlinux files from working and non working builds.

Works:
/home/paul/poky/build/tmp/work/beaglebone-poky-linux-

gnuea

Quoted 55 lines
Quoted 55 lines
Quoted 55 lines
Quoted 54 lines
Quoted 33 lines
Quoted 33 lines
Quoted 32 lines

bi

Doesn't Work:
/media/data1/build1/poky/build/tmp/work/beaglebone-poky-li nux-gn
ueabi

I also have been wondering if the version strings may be making a difference.

http://dan.rpsys.net/uImage-rp2 is a uImage from a broken build where I truncated the path length to a "working" build path length and patched in the same version strings:

const char linux_banner[] =

"Linux version 3.14.0-yocto-standard
(paul@ubuntu-build01) (gcc

version 4.8.2 (GCC) ) #1 PREEMPT Mon Apr 14 16:00:52 BST 2014\n";

const char linux_proc_banner[] = "%s version %s (paul@ubuntu-build01)
(gcc version 4.8.2 (GCC) ) %s\n";

to init/version.c.

I don't have hardware and would be interested to know if the kernel linked to above works or not. If it doesn't, it rules out these path and string lengths, if it does work, it points to a problem there.

Quoted 7 lines
Quoted 3 lines

Richard,

Good catch! It boots:

Thanks Denys, this helps narrow down the issue. I've shared
http://dan.rpsys.net/uImage-rp3 which is the same as the last one but with my changes to version.c reverted. The one should tell us if its the paths or the strings.

This one also boots for me:

Linux version 3.14.0-yocto-standard (richard@ted) (gcc version 4.8.2
(GCC) ) #2 PREEMPT Tue Apr 15 05:40:19 IST 2014> > >

Quoted 8 lines

I'm guessing the path problem is more likely but anything is possible.
This is starting to look like some kind of compiler or linker issue. If
it is that, it would help to have more data points about what works and what doesn't. With that in mind could people who have good or bad builds please share the paths they built the kernels in so we can see if we can spot some kind of pattern.

Quoted 2 lines
Quoted 2 lines
Quoted 2 lines

BTW, my path is /OE/RAM/poky/tmp/work/beaglebone-poky-linux-gnueabi and it

works.

Quoted 13 lines
Quoted 5 lines

I can confirm:
build dir in /home/stefans/b1 works,

but /home/stefans/yocto/poky/build doesn't.

But this works
[stefans@firebird bu]$ pwd
/home/stefans/yocto/poky/bu
[stefans@firebird bu]$ echo pwd| wc -c
28

Quoted 2 lines

But 29 doesn't.
So, what now?

Some other things I tried with a "long" TMPDIR path (note that it's the TMPDIR path that makes the difference - in my tests I've been using /home/paul/poky/build2/much/longer/path/to/tmp). None of this helped:

* kernel built with gcc 4.7.2 and binutils 2.23.2
* u-boot built with gcc 4.7.2 and binutils 2.23.2
* u-boot from http://downloads.angstrom-
distribution.org/demo/beaglebone/
* earlyprintk and CONFIG_DEBUG_LL - no additional output printed

I think we're now at the point where we'd benefit from someone with better knowledge debugging the issue.

Quoted 6 lines
Quoted 3 lines

Cheers,
Paul

I don't know if this is helpful but this TMPDIR path size issue reminded me of a bug I verified some time ago:
https://bugzilla.yoctoproject.org/show_bug.cgi?id=3989

Probably not related since the only component which seems to be having issues is the kernel (you can use a kernel which boots with any file system, no matter the build path length)

Signature · 4 lines
------------------------------------------------------------
Gary Thomas                 |  Consulting for the
MLB Associates              |    Embedded world
------------------------------------------------------------
#21 lore raw
In reply to Paul Eggleton
Quoted 113 lines · Paul Eggleton
Quoted 9 lines · StefanX
Quoted 8 lines · StefanX
Quoted 7 lines · Denys Dmytriyenko
Quoted 6 lines · Denys Dmytriyenko
Quoted 5 lines · Richard Purdie
Quoted 4 lines · Denys Dmytriyenko
Quoted 3 lines · Richard Purdie
Quoted 2 lines · Denys Dmytriyenko
Quoted 1 line · Gary Thomas

Very interesting results! These are the results from the

build hosts I have:

Quoted 95 lines
Quoted 80 lines
Quoted 80 lines
Quoted 79 lines
Quoted 56 lines
Quoted 55 lines
Quoted 46 lines
Quoted 8 lines
Quoted 3 lines

Fedora 13 (i686) - fails
Fedora 17 (i686) - fails
Ubuntu 12.04 (x86_64) - boots

Interesting indeed. I have no idea what's so special about Fedora host - this is the first time I hear about issues with it. I may try experimenting with different VMs once I have more time...

I've been having a look at this. The biggest differences I can find
between working and non working builds is the path length to the build
directory for the kernel. This is from comparing vmlinux files from
working and non working builds.

Works:
/home/paul/poky/build/tmp/work/beaglebone-poky-linux-gnueabi

Doesn't Work:
/media/data1/build1/poky/build/tmp/work/beaglebone-poky-linux-gn ueabi

I also have been wondering if the version strings may be making a
difference.

http://dan.rpsys.net/uImage-rp2 is a uImage from a broken build where I
truncated the path length to a "working" build path length and patched
in the same version strings:

const char linux_banner[] =

"Linux version 3.14.0-yocto-standard
(paul@ubuntu-build01) (gcc

version 4.8.2 (GCC) ) #1 PREEMPT Mon Apr 14 16:00:52 BST 2014\n";

const char linux_proc_banner[] = "%s version %s
(paul@ubuntu-build01)
(gcc version 4.8.2 (GCC) ) %s\n";

Quoted 7 lines

to init/version.c.

I don't have hardware and would be interested to know if the kernel
linked to above works or not. If it doesn't, it rules out these path and
string lengths, if it does work, it points to a problem there.

Richard,

Quoted 7 lines
Quoted 1 line

Good catch! It boots:

Thanks Denys, this helps narrow down the issue. I've shared
http://dan.rpsys.net/uImage-rp3 which is the same as the last one but
with my changes to version.c reverted. The one should tell us if its the
paths or the strings.

This one also boots for me:

Linux version 3.14.0-yocto-standard (richard@ted) (gcc version 4.8.2 (GCC) ) #2 PREEMPT Tue Apr 15 05:40:19 IST 2014> > >

Quoted 11 lines

I'm guessing the path problem is more likely but anything is possible.
This is starting to look like some kind of compiler or linker issue. If
it is that, it would help to have more data points about what works and
what doesn't. With that in mind could people who have good or bad builds
please share the paths they built the kernels in so we can see if we can
spot some kind of pattern.

Quoted 7 lines
Quoted 2 lines

BTW, my path is /OE/RAM/poky/tmp/work/beaglebone-poky-linux-gnueabi and it works.

I can confirm:
build dir in /home/stefans/b1 works,

but /home/stefans/yocto/poky/build doesn't.

But this works
[stefans@firebird bu]$ pwd
/home/stefans/yocto/poky/bu
[stefans@firebird bu]$ echo pwd| wc -c
28

Quoted 2 lines

But 29 doesn't.
So, what now?

Some other things I tried with a "long" TMPDIR path (note that it's the TMPDIR path that makes the difference - in my tests I've been using /home/paul/poky/build2/much/longer/path/to/tmp). None of this helped:

Quoted 7 lines

* kernel built with gcc 4.7.2 and binutils 2.23.2
* u-boot built with gcc 4.7.2 and binutils 2.23.2
* u-boot from http://downloads.angstrom-distribution.org/demo/beaglebone/
* earlyprintk and CONFIG_DEBUG_LL - no additional output printed

I think we're now at the point where we'd benefit from someone with better knowledge debugging the issue.

Ok, should we expand the search area? Since this is supposed to be vanilla 3.14 kernel, can we try other platforms and see if they are similarly affected? I'll try pinging our kernel guys for any ideas...

Signature · 1 line
Denys
#22 lore raw
Quoted 8 lines · On Tuesday 15 April 2014 12:26:39 Denys Dmytriyenko wrote:
Quoted 7 lines · Paul Eggleton
Quoted 6 lines · StefanX
Quoted 5 lines · StefanX
Quoted 4 lines · Denys Dmytriyenko
Quoted 3 lines · Denys Dmytriyenko
Quoted 2 lines · Richard Purdie
Quoted 1 line · Denys Dmytriyenko

On Tue, Apr 15, 2014 at 01:20:03AM +0100, Richard Purdie

Quoted 2 lines · wrote:
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines · Denys Dmytriyenko
Quoted 1 line

On Mon, Apr 14, 2014 at 02:11:05PM -0600, Gary Thomas

Quoted 129 lines · wrote:
Quoted 129 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines

Very interesting results! These are the results from the

build hosts I have:

Quoted 107 lines
Quoted 98 lines
Quoted 98 lines
Quoted 97 lines
Quoted 67 lines
Quoted 64 lines
Quoted 56 lines
Quoted 9 lines
Quoted 3 lines

Fedora 13 (i686) - fails
Fedora 17 (i686) - fails
Ubuntu 12.04 (x86_64) - boots

Interesting indeed. I have no idea what's so special about Fedora host - this is the first time I hear about issues with
it. I may try experimenting with different VMs once I have more time...

I've been having a look at this. The biggest differences I can
find
between working and non working builds is the path length to the
build
directory for the kernel. This is from comparing vmlinux files
from
working and non working builds.

Works:
/home/paul/poky/build/tmp/work/beaglebone-poky-linux-gnueabi

Doesn't Work:
/media/data1/build1/poky/build/tmp/work/beaglebone-poky-linu x-gn
ueabi

I also have been wondering if the version strings may be making
a
difference.

http://dan.rpsys.net/uImage-rp2 is a uImage from a broken build
where I
truncated the path length to a "working" build path length and
patched
in the same version strings:

const char linux_banner[] =

"Linux version 3.14.0-yocto-standard
(paul@ubuntu-build01) (gcc

version 4.8.2 (GCC) ) #1 PREEMPT Mon Apr 14 16:00:52 BST 2014\n";

const char linux_proc_banner[] = "%s version %s
(paul@ubuntu-build01)
(gcc version 4.8.2 (GCC) ) %s\n";

to init/version.c.

Quoted 7 lines

I don't have hardware and would be interested to know if the kernel
linked to above works or not. If it doesn't, it rules out these
path and
string lengths, if it does work, it points to a problem there.

Quoted 3 lines

Richard,

Good catch! It boots:

Quoted 8 lines

Thanks Denys, this helps narrow down the issue. I've shared
http://dan.rpsys.net/uImage-rp3 which is the same as the last one
but
with my changes to version.c reverted. The one should tell us if its
the
paths or the strings.

This one also boots for me:

Linux version 3.14.0-yocto-standard (richard@ted) (gcc version 4.8.2
(GCC) ) #2 PREEMPT Tue Apr 15 05:40:19 IST 2014> > >

Quoted 15 lines

I'm guessing the path problem is more likely but anything is possible.
This is starting to look like some kind of compiler or linker issue.
If
it is that, it would help to have more data points about what works
and
what doesn't. With that in mind could people who have good or bad
builds
please share the paths they built the kernels in so we can see if we
can
spot some kind of pattern.

Quoted 8 lines
Quoted 3 lines

BTW, my path is /OE/RAM/poky/tmp/work/beaglebone-poky-linux-gnueabi and
it works.

I can confirm:
build dir in /home/stefans/b1 works,

but /home/stefans/yocto/poky/build doesn't.

Quoted 8 lines

But this works
[stefans@firebird bu]$ pwd
/home/stefans/yocto/poky/bu
[stefans@firebird bu]$ echo pwd| wc -c
28

But 29 doesn't.
So, what now?

Some other things I tried with a "long" TMPDIR path (note that it's the TMPDIR path that makes the difference - in my tests I've been using /home/paul/poky/build2/much/longer/path/to/tmp). None of this helped:

* kernel built with gcc 4.7.2 and binutils 2.23.2
* u-boot built with gcc 4.7.2 and binutils 2.23.2
* u-boot from http://downloads.angstrom-distribution.org/demo/beaglebone/
* earlyprintk and CONFIG_DEBUG_LL - no additional output printed

Quoted 6 lines
Quoted 2 lines

I think we're now at the point where we'd benefit from someone with better knowledge debugging the issue.

Ok, should we expand the search area? Since this is supposed to be vanilla 3.14 kernel, can we try other platforms and see if they are similarly affected? I'll try pinging our kernel guys for any ideas...

As far as I know it has only been observed with beaglebone (both white and black, if it makes a difference). FWIW, qemuarm images from the autobuilder boot just fine, and apparently the same is true of edgerouter (different architecture but also uses u-boot).

Cheers,
Paul

Signature · 2 lines
Paul Eggleton
Intel Open Source Technology Centre
#23 lore raw
Quoted 145 lines · Paul Eggleton
Quoted 8 lines · On Tuesday 15 April 2014 12:26:39 Denys Dmytriyenko wrote:
Quoted 7 lines · Paul Eggleton
Quoted 6 lines · StefanX
Quoted 5 lines · StefanX
Quoted 4 lines · Denys Dmytriyenko
Quoted 3 lines · Denys Dmytriyenko
Quoted 2 lines · Richard Purdie
Quoted 1 line · Denys Dmytriyenko

On Tue, Apr 15, 2014 at 01:20:03AM +0100, Richard Purdie

Quoted 2 lines · wrote:
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines · Denys Dmytriyenko
Quoted 1 line

On Mon, Apr 14, 2014 at 02:11:05PM -0600, Gary Thomas

Quoted 132 lines · wrote:
Quoted 131 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines

Very interesting results! These are the results from the

build hosts I have:

Quoted 107 lines
Quoted 98 lines
Quoted 98 lines
Quoted 97 lines
Quoted 67 lines
Quoted 64 lines
Quoted 56 lines
Quoted 9 lines
Quoted 3 lines

Fedora 13 (i686) - fails
Fedora 17 (i686) - fails
Ubuntu 12.04 (x86_64) - boots

Interesting indeed. I have no idea what's so special about Fedora host - this is the first time I hear about issues with
it. I may try experimenting with different VMs once I have more time...

I've been having a look at this. The biggest differences I can
find
between working and non working builds is the path length to the
build
directory for the kernel. This is from comparing vmlinux files
from
working and non working builds.

Works:
/home/paul/poky/build/tmp/work/beaglebone-poky-linux-gnueabi

Doesn't Work:
/media/data1/build1/poky/build/tmp/work/beaglebone-poky-linu x-gn
ueabi

I also have been wondering if the version strings may be making
a
difference.

http://dan.rpsys.net/uImage-rp2 is a uImage from a broken build
where I
truncated the path length to a "working" build path length and
patched
in the same version strings:

const char linux_banner[] =

"Linux version 3.14.0-yocto-standard
(paul@ubuntu-build01) (gcc

version 4.8.2 (GCC) ) #1 PREEMPT Mon Apr 14 16:00:52 BST 2014\n";

const char linux_proc_banner[] = "%s version %s
(paul@ubuntu-build01)
(gcc version 4.8.2 (GCC) ) %s\n";

to init/version.c.

Quoted 7 lines

I don't have hardware and would be interested to know if the kernel
linked to above works or not. If it doesn't, it rules out these
path and
string lengths, if it does work, it points to a problem there.

Quoted 3 lines

Richard,

Good catch! It boots:

Quoted 8 lines

Thanks Denys, this helps narrow down the issue. I've shared
http://dan.rpsys.net/uImage-rp3 which is the same as the last one
but
with my changes to version.c reverted. The one should tell us if its
the
paths or the strings.

This one also boots for me:

Linux version 3.14.0-yocto-standard (richard@ted) (gcc version 4.8.2
(GCC) ) #2 PREEMPT Tue Apr 15 05:40:19 IST 2014> > >

Quoted 15 lines

I'm guessing the path problem is more likely but anything is possible.
This is starting to look like some kind of compiler or linker issue.
If
it is that, it would help to have more data points about what works
and
what doesn't. With that in mind could people who have good or bad
builds
please share the paths they built the kernels in so we can see if we
can
spot some kind of pattern.

Quoted 8 lines
Quoted 3 lines

BTW, my path is /OE/RAM/poky/tmp/work/beaglebone-poky-linux-gnueabi and
it works.

I can confirm:
build dir in /home/stefans/b1 works,

but /home/stefans/yocto/poky/build doesn't.

Quoted 8 lines

But this works
[stefans@firebird bu]$ pwd
/home/stefans/yocto/poky/bu
[stefans@firebird bu]$ echo pwd| wc -c
28

But 29 doesn't.
So, what now?

Some other things I tried with a "long" TMPDIR path (note that it's the TMPDIR path that makes the difference - in my tests I've been using /home/paul/poky/build2/much/longer/path/to/tmp). None of this helped:

* kernel built with gcc 4.7.2 and binutils 2.23.2
* u-boot built with gcc 4.7.2 and binutils 2.23.2
* u-boot from http://downloads.angstrom-distribution.org/demo/beaglebone/
* earlyprintk and CONFIG_DEBUG_LL - no additional output printed

I think we're now at the point where we'd benefit from someone with better knowledge debugging the issue.

Quoted 8 lines
Quoted 3 lines

Ok, should we expand the search area? Since this is supposed to be vanilla 3.14 kernel, can we try other platforms and see if they are similarly affected? I'll try pinging our kernel guys for any ideas...

As far as I know it has only been observed with beaglebone (both white and black, if it makes a difference). FWIW, qemuarm images from the autobuilder boot just fine, and apparently the same is true of edgerouter (different architecture but also uses u-boot).

But do those other platforms use uImage or zImage?

Anyway, just for testing, I switched to zImage and it seems to boot fine for me regarldless of the path length...

I just sent 2 patches to test this out - unfortunately booting zImage isntead of uImage requires fiddling with U-boot variables a bit, when using 2013.07. So I also provided a quick update patch to get U-boot 2014.01 that has zImage enabled by default.

Those patches are not yet for merging and they are directly against Poky - could anyone having issues with booting BBB try them? Place new u-boot.img into first boot FAT partition and new zImage kernel image to /boot dir of the rootfs partition. Please let me know if it works for your long paths. Thanks.

Signature · 1 line
Denys
#43 lore raw
In reply to Denys Dmytriyenko
Quoted 119 lines · Denys Dmytriyenko
Quoted 118 lines · Paul Eggleton
Quoted 9 lines · StefanX
Quoted 8 lines · StefanX
Quoted 7 lines · Denys Dmytriyenko
Quoted 6 lines · Denys Dmytriyenko
Quoted 5 lines · Richard Purdie
Quoted 4 lines · Denys Dmytriyenko
Quoted 3 lines · Richard Purdie
Quoted 2 lines · Denys Dmytriyenko
Quoted 1 line · Gary Thomas

Very interesting results! These are the results from the

build hosts I have:

Quoted 95 lines
Quoted 80 lines
Quoted 80 lines
Quoted 79 lines
Quoted 56 lines
Quoted 55 lines
Quoted 46 lines
Quoted 8 lines
Quoted 3 lines

Fedora 13 (i686) - fails
Fedora 17 (i686) - fails
Ubuntu 12.04 (x86_64) - boots

Interesting indeed. I have no idea what's so special about Fedora host - this is the first time I hear about issues with it. I may try experimenting with different VMs once I have more time...

I've been having a look at this. The biggest differences I can find
between working and non working builds is the path length to the build
directory for the kernel. This is from comparing vmlinux files from
working and non working builds.

Works:
/home/paul/poky/build/tmp/work/beaglebone-poky-linux-gnueabi

Doesn't Work:
/media/data1/build1/poky/build/tmp/work/beaglebone-poky-linux-gn ueabi

I also have been wondering if the version strings may be making a
difference.

http://dan.rpsys.net/uImage-rp2 is a uImage from a broken build where I
truncated the path length to a "working" build path length and patched
in the same version strings:

const char linux_banner[] =

"Linux version 3.14.0-yocto-standard
(paul@ubuntu-build01) (gcc

version 4.8.2 (GCC) ) #1 PREEMPT Mon Apr 14 16:00:52 BST 2014\n";

const char linux_proc_banner[] = "%s version %s
(paul@ubuntu-build01)
(gcc version 4.8.2 (GCC) ) %s\n";

Quoted 7 lines

to init/version.c.

I don't have hardware and would be interested to know if the kernel
linked to above works or not. If it doesn't, it rules out these path and
string lengths, if it does work, it points to a problem there.

Richard,

Quoted 7 lines
Quoted 1 line

Good catch! It boots:

Thanks Denys, this helps narrow down the issue. I've shared
http://dan.rpsys.net/uImage-rp3 which is the same as the last one but
with my changes to version.c reverted. The one should tell us if its the
paths or the strings.

This one also boots for me:

Linux version 3.14.0-yocto-standard (richard@ted) (gcc version 4.8.2 (GCC) ) #2 PREEMPT Tue Apr 15 05:40:19 IST 2014> > >

Quoted 11 lines

I'm guessing the path problem is more likely but anything is possible.
This is starting to look like some kind of compiler or linker issue. If
it is that, it would help to have more data points about what works and
what doesn't. With that in mind could people who have good or bad builds
please share the paths they built the kernels in so we can see if we can
spot some kind of pattern.

Quoted 7 lines
Quoted 2 lines

BTW, my path is /OE/RAM/poky/tmp/work/beaglebone-poky-linux-gnueabi and it works.

I can confirm:
build dir in /home/stefans/b1 works,

but /home/stefans/yocto/poky/build doesn't.

But this works
[stefans@firebird bu]$ pwd
/home/stefans/yocto/poky/bu
[stefans@firebird bu]$ echo pwd| wc -c
28

Quoted 2 lines

But 29 doesn't.
So, what now?

Some other things I tried with a "long" TMPDIR path (note that it's the TMPDIR path that makes the difference - in my tests I've been using /home/paul/poky/build2/much/longer/path/to/tmp). None of this helped:

* kernel built with gcc 4.7.2 and binutils 2.23.2
* u-boot built with gcc 4.7.2 and binutils 2.23.2
* u-boot from http://downloads.angstrom-distribution.org/demo/beaglebone/
* earlyprintk and CONFIG_DEBUG_LL - no additional output printed

Quoted 7 lines
Quoted 2 lines

I think we're now at the point where we'd benefit from someone with better knowledge debugging the issue.

Ok, should we expand the search area? Since this is supposed to be vanilla 3.14 kernel, can we try other platforms and see if they are similarly affected? I'll try pinging our kernel guys for any ideas...

FWIW, I filed https://bugzilla.yoctoproject.org/show_bug.cgi?id=6165
that contains all the info I had.
A link to two of my kernel is there, one working one not.

Cheers,
Stefan

#24 lore raw
In reply to Denys Dmytriyenko
Quoted 165 lines · Denys Dmytriyenko
Quoted 145 lines · Paul Eggleton
Quoted 8 lines · On Tuesday 15 April 2014 12:26:39 Denys Dmytriyenko wrote:
Quoted 7 lines · Paul Eggleton
Quoted 6 lines · StefanX
Quoted 5 lines · StefanX
Quoted 4 lines · Denys Dmytriyenko
Quoted 3 lines · Denys Dmytriyenko
Quoted 2 lines · Richard Purdie
Quoted 1 line · Denys Dmytriyenko

On Tue, Apr 15, 2014 at 01:20:03AM +0100, Richard Purdie

Quoted 2 lines · wrote:
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines · Denys Dmytriyenko
Quoted 1 line

On Mon, Apr 14, 2014 at 02:11:05PM -0600, Gary Thomas

Quoted 132 lines · wrote:
Quoted 131 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines

Very interesting results! These are the results from
the

build hosts I have:

Quoted 107 lines
Quoted 98 lines
Quoted 98 lines
Quoted 97 lines
Quoted 67 lines
Quoted 64 lines
Quoted 56 lines
Quoted 9 lines
Quoted 3 lines

Fedora 13 (i686) - fails
Fedora 17 (i686) - fails
Ubuntu 12.04 (x86_64) - boots

Interesting indeed. I have no idea what's so special about Fedora host - this is the first time I hear about issues with
it. I may try experimenting with different VMs once I have more time...

I've been having a look at this. The biggest differences I can
find
between working and non working builds is the path length to the
build
directory for the kernel. This is from comparing vmlinux files
from
working and non working builds.

Works:
/home/paul/poky/build/tmp/work/beaglebone-poky-linux-gnueabi

Doesn't Work:
/media/data1/build1/poky/build/tmp/work/beaglebone-poky-linu x-gn
ueabi

I also have been wondering if the version strings may be making
a
difference.

http://dan.rpsys.net/uImage-rp2 is a uImage from a broken build
where I
truncated the path length to a "working" build path length and
patched
in the same version strings:

const char linux_banner[] =

"Linux version 3.14.0-yocto-standard
(paul@ubuntu-build01) (gcc

version 4.8.2 (GCC) ) #1 PREEMPT Mon Apr 14 16:00:52 BST 2014\n";

const char linux_proc_banner[] = "%s version %s
(paul@ubuntu-build01)
(gcc version 4.8.2 (GCC) ) %s\n";

to init/version.c.

Quoted 7 lines

I don't have hardware and would be interested to know if the kernel
linked to above works or not. If it doesn't, it rules out these
path and
string lengths, if it does work, it points to a problem there.

Quoted 3 lines

Richard,

Good catch! It boots:

Quoted 8 lines

Thanks Denys, this helps narrow down the issue. I've shared
http://dan.rpsys.net/uImage-rp3 which is the same as the last one
but
with my changes to version.c reverted. The one should tell us if its
the
paths or the strings.

This one also boots for me:

Linux version 3.14.0-yocto-standard (richard@ted) (gcc version 4.8.2
(GCC) ) #2 PREEMPT Tue Apr 15 05:40:19 IST 2014> > >

Quoted 15 lines

I'm guessing the path problem is more likely but anything is possible.
This is starting to look like some kind of compiler or linker issue.
If
it is that, it would help to have more data points about what works
and
what doesn't. With that in mind could people who have good or bad
builds
please share the paths they built the kernels in so we can see if we
can
spot some kind of pattern.

Quoted 8 lines
Quoted 3 lines

BTW, my path is /OE/RAM/poky/tmp/work/beaglebone-poky-linux-gnueabi and
it works.

I can confirm:
build dir in /home/stefans/b1 works,

but /home/stefans/yocto/poky/build doesn't.

Quoted 8 lines

But this works
[stefans@firebird bu]$ pwd
/home/stefans/yocto/poky/bu
[stefans@firebird bu]$ echo pwd| wc -c
28

But 29 doesn't.
So, what now?

Some other things I tried with a "long" TMPDIR path (note that it's the TMPDIR path that makes the difference - in my tests I've been using /home/paul/poky/build2/much/longer/path/to/tmp). None of this helped:

* kernel built with gcc 4.7.2 and binutils 2.23.2
* u-boot built with gcc 4.7.2 and binutils 2.23.2
* u-boot from http://downloads.angstrom-distribution.org/demo/beaglebone/
* earlyprintk and CONFIG_DEBUG_LL - no additional output printed

I think we're now at the point where we'd benefit from someone with better knowledge debugging the issue.

Quoted 8 lines
Quoted 3 lines

Ok, should we expand the search area? Since this is supposed to be vanilla 3.14 kernel, can we try other platforms and see if they are similarly affected? I'll try pinging our kernel guys for any ideas...

As far as I know it has only been observed with beaglebone (both white and black, if it makes a difference). FWIW, qemuarm images from the autobuilder boot just fine, and apparently the same is true of edgerouter (different architecture but also uses u-boot).

But do those other platforms use uImage or zImage?

Anyway, just for testing, I switched to zImage and it seems to boot fine for me regarldless of the path length...

I just sent 2 patches to test this out - unfortunately booting zImage isntead of uImage requires fiddling with U-boot variables a bit, when using 2013.07. So I also provided a quick update patch to get U-boot 2014.01 that has zImage enabled by default.

Quoted 4 lines

Those patches are not yet for merging and they are directly against Poky - could anyone having issues with booting BBB try them? Place new u-boot.img into first boot FAT partition and new zImage kernel image to /boot dir of the rootfs partition. Please let me know if it works for your long paths. Thanks.

Ah, never mind, it still doesn't work with zImage - I got fooled by sstate :(

Signature · 1 line
Denys
#25 lore raw
Quoted 16 lines · Denys Dmytriyenko
Quoted 16 lines
Quoted 16 lines
Quoted 15 lines
Quoted 11 lines

Some other things I tried with a "long" TMPDIR path (note that it's the TMPDIR path that makes the difference - in my tests I've been using /home/paul/poky/build2/much/longer/path/to/tmp). None of this helped:

* kernel built with gcc 4.7.2 and binutils 2.23.2
* u-boot built with gcc 4.7.2 and binutils 2.23.2
* u-boot from http://downloads.angstrom-distribution.org/demo/beaglebone/
* earlyprintk and CONFIG_DEBUG_LL - no additional output printed

I think we're now at the point where we'd benefit from someone with better knowledge debugging the issue.

Ok, should we expand the search area? Since this is supposed to be vanilla 3.14 kernel, can we try other platforms and see if they are similarly affected? I'll try pinging our kernel guys for any ideas...

Quoted 6 lines
Quoted 6 lines
Quoted 4 lines

As far as I know it has only been observed with beaglebone (both white and black, if it makes a difference). FWIW, qemuarm images from the autobuilder boot just fine, and apparently the same is true of edgerouter (different architecture but also uses u-boot).

But do those other platforms use uImage or zImage?

I don't yet know what is going on, but building in the same directory with sources (B = S) makes it work regarless of the path length:

/OE/RAM/poky-111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111/22222222222222222222222222222222222222222222222222222222222222222222/3333333333333333333333333333333333333333333333333333/tmp/work/beaglebone-poky-linux-gnueabi/linux-yocto/3.14+gitAUTOINC+928d7b2dda_0143c6ebb4-r0/linux

So, I just commented out setting kernel-specific B in linux-yocto.inc and any kernel now boots with long path:

#B = "${WORKDIR}/linux-${PACKAGE_ARCH}-${LINUX_KERNEL_TYPE}-build"

I'm copying Richard and Bruce directly to see if they may have a quick insight and/or accept it as a workaround for the release. I'll keep digging further, but if anyone cares to verify the above workaround works for them, I would appreciate. Thanks!

Signature · 1 line
Denys
#26 lore raw
Quoted 32 lines · Denys Dmytriyenko
Quoted 16 lines · Denys Dmytriyenko
Quoted 16 lines
Quoted 16 lines
Quoted 15 lines
Quoted 11 lines

Some other things I tried with a "long" TMPDIR path (note that it's the TMPDIR path that makes the difference - in my tests I've been using /home/paul/poky/build2/much/longer/path/to/tmp). None of this helped:

* kernel built with gcc 4.7.2 and binutils 2.23.2
* u-boot built with gcc 4.7.2 and binutils 2.23.2
* u-boot from http://downloads.angstrom-distribution.org/demo/beaglebone/
* earlyprintk and CONFIG_DEBUG_LL - no additional output printed

I think we're now at the point where we'd benefit from someone with better knowledge debugging the issue.

Ok, should we expand the search area? Since this is supposed to be vanilla 3.14 kernel, can we try other platforms and see if they are similarly affected? I'll try pinging our kernel guys for any ideas...

Quoted 6 lines
Quoted 6 lines
Quoted 4 lines

As far as I know it has only been observed with beaglebone (both white and black, if it makes a difference). FWIW, qemuarm images from the autobuilder boot just fine, and apparently the same is true of edgerouter (different architecture but also uses u-boot).

But do those other platforms use uImage or zImage?

I don't yet know what is going on, but building in the same directory with sources (B = S) makes it work regarless of the path length:

/OE/RAM/poky-111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111/22222222222222222222222222222222222222222222222222222222222222222222/3333333333333333333333333333333333333333333333333333/tmp/work/beaglebone-poky-linux-gnueabi/linux-yocto/3.14+gitAUTOINC+928d7b2dda_0143c6ebb4-r0/linux

So, I just commented out setting kernel-specific B in linux-yocto.inc and any kernel now boots with long path:

Quoted 7 lines

#B = "${WORKDIR}/linux-${PACKAGE_ARCH}-${LINUX_KERNEL_TYPE}-build"

I'm copying Richard and Bruce directly to see if they may have a quick insight and/or accept it as a workaround for the release. I'll keep digging further, but if anyone cares to verify the above workaround works for them, I would appreciate. Thanks!

Verified - I rebuilt the kernel in a working tree with a longer path (one in fact that had failed before) and it boots fine.

Wonder what ${B} != ${S} is doing wacky...?

Signature · 4 lines
------------------------------------------------------------
Gary Thomas                 |  Consulting for the
MLB Associates              |    Embedded world
------------------------------------------------------------
#34 lore raw
In reply to Denys Dmytriyenko
Quoted 29 lines · Denys Dmytriyenko
Quoted 16 lines · Denys Dmytriyenko
Quoted 16 lines
Quoted 16 lines
Quoted 15 lines
Quoted 11 lines

Some other things I tried with a "long" TMPDIR path (note that it's the TMPDIR path that makes the difference - in my tests I've been using /home/paul/poky/build2/much/longer/path/to/tmp). None of this helped:

* kernel built with gcc 4.7.2 and binutils 2.23.2
* u-boot built with gcc 4.7.2 and binutils 2.23.2
* u-boot from http://downloads.angstrom-distribution.org/demo/beaglebone/
* earlyprintk and CONFIG_DEBUG_LL - no additional output printed

I think we're now at the point where we'd benefit from someone with better knowledge debugging the issue.

Ok, should we expand the search area? Since this is supposed to be vanilla 3.14 kernel, can we try other platforms and see if they are similarly affected? I'll try pinging our kernel guys for any ideas...

Quoted 6 lines
Quoted 6 lines
Quoted 4 lines

As far as I know it has only been observed with beaglebone (both white and black, if it makes a difference). FWIW, qemuarm images from the autobuilder boot just fine, and apparently the same is true of edgerouter (different architecture but also uses u-boot).

But do those other platforms use uImage or zImage?

I don't yet know what is going on, but building in the same directory with sources (B = S) makes it work regarless of the path length:

/OE/RAM/poky-111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111/22222222222222222222222222222222222222222222222222222222222222222222/3333333333333333333333333333333333333333333333333333/tmp/work/beaglebone-poky-linux-gnueabi/linux-yocto/3.14+gitAUTOINC+928d7b2dda_0143c6ebb4-r0/linux

Quoted 7 lines

So, I just commented out setting kernel-specific B in linux-yocto.inc and any kernel now boots with long path:

#B = "${WORKDIR}/linux-${PACKAGE_ARCH}-${LINUX_KERNEL_TYPE}-build"

I'm copying Richard and Bruce directly to see if they may have a quick insight and/or accept it as a workaround for the release. I'll keep digging further,

I've never seen this before, and when I'm back in the office Wed/Thursday I can have a closer look.

As for the BBB using this in the release, I really don't want to have one board that breaks the build and source separation, since it has always been in place to keep things clean (and I know of a few random scripts, etc, that expect it) .. but more importantly, I want to root cause this, since it is a lurking problem.

Bruce

Quoted 3 lines

but if anyone cares to verify the above workaround works for them, I would appreciate. Thanks!

#35 lore raw
In reply to Denys Dmytriyenko
Quoted 8 lines · Denys Dmytriyenko

I don't yet know what is going on, but building in the same directory with sources (B = S) makes it work regarless of the path length:

/OE/RAM/poky-111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111/22222222222222222222222222222222222222222222222222222222222222222222/3333333333333333333333333333333333333333333333333333/tmp/work/beaglebone-poky-linux-gnueabi/linux-yocto/3.14+gitAUTOINC+928d7b2dda_0143c6ebb4-r0/linux

So, I just commented out setting kernel-specific B in linux-yocto.inc and any kernel now boots with long path:

Quoted 6 lines

#B = "${WORKDIR}/linux-${PACKAGE_ARCH}-${LINUX_KERNEL_TYPE}-build"

I'm copying Richard and Bruce directly to see if they may have a quick insight and/or accept it as a workaround for the release. I'll keep digging further, but if anyone cares to verify the above workaround works for them, I would appreciate. Thanks!

I'm travelling and don't have hardware so I'm limited in what I can look at right now. I suspect this workaround "works" because it makes the "beaglebone-standard-build" extra characters on the path. I have a feeling your -1111111 test above may have reused sstate or something like that and given misleading results. I'd be interested in the vmlinux file from the build above to see if the poky-111111 pathnames are in there (they get stripped out when you create the zImage).

Cheers,

Richard

#36 lore raw
Quoted 16 lines · Richard Purdie
Quoted 8 lines · Denys Dmytriyenko

I don't yet know what is going on, but building in the same directory with sources (B = S) makes it work regarless of the path length:

/OE/RAM/poky-111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111/22222222222222222222222222222222222222222222222222222222222222222222/3333333333333333333333333333333333333333333333333333/tmp/work/beaglebone-poky-linux-gnueabi/linux-yocto/3.14+gitAUTOINC+928d7b2dda_0143c6ebb4-r0/linux

So, I just commented out setting kernel-specific B in linux-yocto.inc and any kernel now boots with long path:

Quoted 6 lines

#B = "${WORKDIR}/linux-${PACKAGE_ARCH}-${LINUX_KERNEL_TYPE}-build"

I'm copying Richard and Bruce directly to see if they may have a quick insight and/or accept it as a workaround for the release. I'll keep digging further, but if anyone cares to verify the above workaround works for them, I would appreciate. Thanks!

Quoted 7 lines

I'm travelling and don't have hardware so I'm limited in what I can look at right now. I suspect this workaround "works" because it makes the "beaglebone-standard-build" extra characters on the path. I have a feeling your -1111111 test above may have reused sstate or something like that and given misleading results. I'd be interested in the vmlinux file from the build above to see if the poky-111111 pathnames are in there (they get stripped out when you create the zImage).

Nope, I was fooled by sstate once, so this time I tested with cleansstate and compared timestamps of the kernel when it boots - it is definitely the new one. And to make sure 'beaglebone-standard-build' extra suffix doesn't affect it, I increased the path length even further - that extra level with 333 is there to over-compensate, as it was failing before even with just 111 and 222 levels only... Looks like Gary was also able to verify it.

But, you are right about vmlinux - it doesn't have those paths in there any more! I've seen them there when building with B != S, but they are gone when building with B = S. Wondering why. You can check it for yourself:

http://arago-project.org/files/short-term/misc/vmlinux-3.14.0-yocto-standard

Signature · 1 line
Denys
#37 lore raw
Quoted 32 lines · Denys Dmytriyenko
Quoted 16 lines · Richard Purdie
Quoted 8 lines · Denys Dmytriyenko

I don't yet know what is going on, but building in the same directory with sources (B = S) makes it work regarless of the path length:

/OE/RAM/poky-111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111/22222222222222222222222222222222222222222222222222222222222222222222/3333333333333333333333333333333333333333333333333333/tmp/work/beaglebone-poky-linux-gnueabi/linux-yocto/3.14+gitAUTOINC+928d7b2dda_0143c6ebb4-r0/linux

So, I just commented out setting kernel-specific B in linux-yocto.inc and any kernel now boots with long path:

Quoted 6 lines

#B = "${WORKDIR}/linux-${PACKAGE_ARCH}-${LINUX_KERNEL_TYPE}-build"

I'm copying Richard and Bruce directly to see if they may have a quick insight and/or accept it as a workaround for the release. I'll keep digging further, but if anyone cares to verify the above workaround works for them, I would appreciate. Thanks!

Quoted 7 lines

I'm travelling and don't have hardware so I'm limited in what I can look at right now. I suspect this workaround "works" because it makes the "beaglebone-standard-build" extra characters on the path. I have a feeling your -1111111 test above may have reused sstate or something like that and given misleading results. I'd be interested in the vmlinux file from the build above to see if the poky-111111 pathnames are in there (they get stripped out when you create the zImage).

Nope, I was fooled by sstate once, so this time I tested with cleansstate and compared timestamps of the kernel when it boots - it is definitely the new one. And to make sure 'beaglebone-standard-build' extra suffix doesn't affect it, I increased the path length even further - that extra level with 333 is there to over-compensate, as it was failing before even with just 111 and 222 levels only... Looks like Gary was also able to verify it.

Quoted 5 lines

But, you are right about vmlinux - it doesn't have those paths in there any more! I've seen them there when building with B != S, but they are gone when building with B = S. Wondering why. You can check it for yourself:

http://arago-project.org/files/short-term/misc/vmlinux-3.14.0-yocto-standard

And, as a follow up, all those long paths in vmlinux when built with B != S point to sources. Here are few examples:

B != S strings:

/OE/RAM/poky/tmp/work/beaglebone-poky-linux-gnueabi/linux-yocto/3.14+gitAUTOINC+928d7b2dda_0143c6ebb4-r0/linux/init/main.c earlycon
4Malformed early option '%s'
3Starting init: %s exists but couldn't execute it (error %d)
...
6Trying to unpack rootfs image as initramfs... /OE/RAM/poky/tmp/work/beaglebone-poky-linux-gnueabi/linux-yocto/3.14+gitAUTOINC+928d7b2dda_0143c6ebb4-r0/linux/init/initramfs.c 6rootfs image is not initramfs (%s); looks like an initrd
TRAILER!!!
...
%lu.%02lu BogoMIPS (lpj=%lu) /OE/RAM/poky/tmp/work/beaglebone-poky-linux-gnueabi/linux-yocto/3.14+gitAUTOINC+928d7b2dda_0143c6ebb4-r0/linux/arch/arm/vfp/vfpmodule.c 6VFP support v0.3:

B = S same strings:

init/main.c
earlycon
4Malformed early option '%s'
3Starting init: %s exists but couldn't execute it (error %d)
...
6Trying to unpack rootfs image as initramfs... init/initramfs.c
6rootfs image is not initramfs (%s); looks like an initrd TRAILER!!!
...
%lu.%02lu BogoMIPS (lpj=%lu) arch/arm/vfp/vfpmodule.c
6VFP support v0.3:

So, that explains why B = S works regardless of the location, as it doesn't embed full path to source files...

Signature · 1 line
Denys
#38 lore raw
Quoted 76 lines · Denys Dmytriyenko
Quoted 32 lines · Denys Dmytriyenko
Quoted 16 lines · Richard Purdie
Quoted 8 lines · Denys Dmytriyenko

I don't yet know what is going on, but building in the same directory with sources (B = S) makes it work regarless of the path length:

/OE/RAM/poky-111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111/22222222222222222222222222222222222222222222222222222222222222222222/3333333333333333333333333333333333333333333333333333/tmp/work/beaglebone-poky-linux-gnueabi/linux-yocto/3.14+gitAUTOINC+928d7b2dda_0143c6ebb4-r0/linux

So, I just commented out setting kernel-specific B in linux-yocto.inc and any kernel now boots with long path:

Quoted 6 lines

#B = "${WORKDIR}/linux-${PACKAGE_ARCH}-${LINUX_KERNEL_TYPE}-build"

I'm copying Richard and Bruce directly to see if they may have a quick insight and/or accept it as a workaround for the release. I'll keep digging further, but if anyone cares to verify the above workaround works for them, I would appreciate. Thanks!

Quoted 7 lines

I'm travelling and don't have hardware so I'm limited in what I can look at right now. I suspect this workaround "works" because it makes the "beaglebone-standard-build" extra characters on the path. I have a feeling your -1111111 test above may have reused sstate or something like that and given misleading results. I'd be interested in the vmlinux file from the build above to see if the poky-111111 pathnames are in there (they get stripped out when you create the zImage).

Nope, I was fooled by sstate once, so this time I tested with cleansstate and compared timestamps of the kernel when it boots - it is definitely the new one. And to make sure 'beaglebone-standard-build' extra suffix doesn't affect it, I increased the path length even further - that extra level with 333 is there to over-compensate, as it was failing before even with just 111 and 222 levels only... Looks like Gary was also able to verify it.

Quoted 5 lines

But, you are right about vmlinux - it doesn't have those paths in there any more! I've seen them there when building with B != S, but they are gone when building with B = S. Wondering why. You can check it for yourself:

http://arago-project.org/files/short-term/misc/vmlinux-3.14.0-yocto-standard

And, as a follow up, all those long paths in vmlinux when built with B != S point to sources. Here are few examples:

B != S strings:

/OE/RAM/poky/tmp/work/beaglebone-poky-linux-gnueabi/linux-yocto/3.14+gitAUTOINC+928d7b2dda_0143c6ebb4-r0/linux/init/main.c earlycon
4Malformed early option '%s'
3Starting init: %s exists but couldn't execute it (error %d)
...
6Trying to unpack rootfs image as initramfs... /OE/RAM/poky/tmp/work/beaglebone-poky-linux-gnueabi/linux-yocto/3.14+gitAUTOINC+928d7b2dda_0143c6ebb4-r0/linux/init/initramfs.c 6rootfs image is not initramfs (%s); looks like an initrd
TRAILER!!!
...
%lu.%02lu BogoMIPS (lpj=%lu) /OE/RAM/poky/tmp/work/beaglebone-poky-linux-gnueabi/linux-yocto/3.14+gitAUTOINC+928d7b2dda_0143c6ebb4-r0/linux/arch/arm/vfp/vfpmodule.c 6VFP support v0.3:

B = S same strings:

init/main.c
earlycon
4Malformed early option '%s'
3Starting init: %s exists but couldn't execute it (error %d)
...
6Trying to unpack rootfs image as initramfs... init/initramfs.c
6rootfs image is not initramfs (%s); looks like an initrd TRAILER!!!
...
%lu.%02lu BogoMIPS (lpj=%lu) arch/arm/vfp/vfpmodule.c
6VFP support v0.3:

Quoted 3 lines

So, that explains why B = S works regardless of the location, as it doesn't embed full path to source files...

seemingly there could be rodata segment overflow issue. Can you do the kernel build with say linker from dora build
and see if it still fails.

Quoted 7 lines

--
Denys
--
_______________________________________________
yocto mailing list
[email protected]
https://lists.yoctoproject.org/listinfo/yocto

#42 lore raw
In reply to Denys Dmytriyenko
Quoted 76 lines · Denys Dmytriyenko
Quoted 32 lines · Denys Dmytriyenko
Quoted 16 lines · Richard Purdie
Quoted 8 lines · Denys Dmytriyenko

I don't yet know what is going on, but building in the same directory with sources (B = S) makes it work regarless of the path length:

/OE/RAM/poky-111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111/22222222222222222222222222222222222222222222222222222222222222222222/3333333333333333333333333333333333333333333333333333/tmp/work/beaglebone-poky-linux-gnueabi/linux-yocto/3.14+gitAUTOINC+928d7b2dda_0143c6ebb4-r0/linux

So, I just commented out setting kernel-specific B in linux-yocto.inc and any kernel now boots with long path:

Quoted 6 lines

#B = "${WORKDIR}/linux-${PACKAGE_ARCH}-${LINUX_KERNEL_TYPE}-build"

I'm copying Richard and Bruce directly to see if they may have a quick insight and/or accept it as a workaround for the release. I'll keep digging further, but if anyone cares to verify the above workaround works for them, I would appreciate. Thanks!

Quoted 7 lines

I'm travelling and don't have hardware so I'm limited in what I can look at right now. I suspect this workaround "works" because it makes the "beaglebone-standard-build" extra characters on the path. I have a feeling your -1111111 test above may have reused sstate or something like that and given misleading results. I'd be interested in the vmlinux file from the build above to see if the poky-111111 pathnames are in there (they get stripped out when you create the zImage).

Nope, I was fooled by sstate once, so this time I tested with cleansstate and compared timestamps of the kernel when it boots - it is definitely the new one. And to make sure 'beaglebone-standard-build' extra suffix doesn't affect it, I increased the path length even further - that extra level with 333 is there to over-compensate, as it was failing before even with just 111 and 222 levels only... Looks like Gary was also able to verify it.

Quoted 5 lines

But, you are right about vmlinux - it doesn't have those paths in there any more! I've seen them there when building with B != S, but they are gone when building with B = S. Wondering why. You can check it for yourself:

http://arago-project.org/files/short-term/misc/vmlinux-3.14.0-yocto-standard

And, as a follow up, all those long paths in vmlinux when built with B != S point to sources. Here are few examples:

B != S strings:

/OE/RAM/poky/tmp/work/beaglebone-poky-linux-gnueabi/linux-yocto/3.14+gitAUTOINC+928d7b2dda_0143c6ebb4-r0/linux/init/main.c earlycon
4Malformed early option '%s'
3Starting init: %s exists but couldn't execute it (error %d)
...
6Trying to unpack rootfs image as initramfs... /OE/RAM/poky/tmp/work/beaglebone-poky-linux-gnueabi/linux-yocto/3.14+gitAUTOINC+928d7b2dda_0143c6ebb4-r0/linux/init/initramfs.c 6rootfs image is not initramfs (%s); looks like an initrd
TRAILER!!!
...
%lu.%02lu BogoMIPS (lpj=%lu) /OE/RAM/poky/tmp/work/beaglebone-poky-linux-gnueabi/linux-yocto/3.14+gitAUTOINC+928d7b2dda_0143c6ebb4-r0/linux/arch/arm/vfp/vfpmodule.c 6VFP support v0.3:

B = S same strings:

init/main.c
earlycon
4Malformed early option '%s'
3Starting init: %s exists but couldn't execute it (error %d)
...
6Trying to unpack rootfs image as initramfs... init/initramfs.c
6rootfs image is not initramfs (%s); looks like an initrd TRAILER!!!
...
%lu.%02lu BogoMIPS (lpj=%lu) arch/arm/vfp/vfpmodule.c
6VFP support v0.3:

Quoted 3 lines

So, that explains why B = S works regardless of the location, as it doesn't embed full path to source files...

But qemux86-64 or genericx86-64 or qemuarm kernel has the same strings, and it boots just fine:

$ strings
yocto-autobuilder/yocto-slave/nightly-x86-64/build/build/tmp/work/qemux86_64-poky-linux/linux-yocto/3.14+gitAUTOINC+fc8c30398d_0143c6ebb4-r0/linux-qemux86_64-standard-build/vmlinux | grep -A 2 -B 2 -E "init/main.c"
4Malformed early option '%s'
3Starting init: %s exists but couldn't execute it (error %d)
/home/stefans/yocto-autobuilder/yocto-slave/nightly-x86-64/build/build/tmp/work/qemux86_64-poky-linux/linux-yocto/3.14+gitAUTOINC+fc8c30398d_0143c6ebb4-r0/linux/init/main.c
4Parameter %s is obsolete, ignored
Interrupts were enabled *very* early, fixing it

qemuarm:

$ strings
yocto-autobuilder/yocto-slave/nightly-arm/build/build/tmp/work/qemuarm-poky-linux-gnueabi/linux-yocto/3.14+gitAUTOINC+928d7b2dda_c966987f88-r0/linux-qemuarm-standard-build/vmlinux | grep -A 2 -B 2 -E "init/main.c"
tab_
__ks
/home/stefans/yocto-autobuilder/yocto-slave/nightly-arm/build/build/tmp/work/qemuarm-poky-linux-gnueabi/linux-yocto/3.14+gitAUTOINC+928d7b2dda_c966987f88-r0/linux/init/main.c
earlycon
4Malformed early option '%s'

Cheers,
Stefan

Quoted 2 lines
Signature · 1 line
Denys
#39 lore raw
In reply to Khem Raj
Quoted 77 lines · Khem Raj
Quoted 76 lines · Denys Dmytriyenko
Quoted 32 lines · Denys Dmytriyenko
Quoted 16 lines · Richard Purdie
Quoted 8 lines · Denys Dmytriyenko

I don't yet know what is going on, but building in the same directory with sources (B = S) makes it work regarless of the path length:

/OE/RAM/poky-111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111/22222222222222222222222222222222222222222222222222222222222222222222/3333333333333333333333333333333333333333333333333333/tmp/work/beaglebone-poky-linux-gnueabi/linux-yocto/3.14+gitAUTOINC+928d7b2dda_0143c6ebb4-r0/linux

So, I just commented out setting kernel-specific B in linux-yocto.inc and any kernel now boots with long path:

Quoted 6 lines

#B = "${WORKDIR}/linux-${PACKAGE_ARCH}-${LINUX_KERNEL_TYPE}-build"

I'm copying Richard and Bruce directly to see if they may have a quick insight and/or accept it as a workaround for the release. I'll keep digging further, but if anyone cares to verify the above workaround works for them, I would appreciate. Thanks!

Quoted 7 lines

I'm travelling and don't have hardware so I'm limited in what I can look at right now. I suspect this workaround "works" because it makes the "beaglebone-standard-build" extra characters on the path. I have a feeling your -1111111 test above may have reused sstate or something like that and given misleading results. I'd be interested in the vmlinux file from the build above to see if the poky-111111 pathnames are in there (they get stripped out when you create the zImage).

Nope, I was fooled by sstate once, so this time I tested with cleansstate and compared timestamps of the kernel when it boots - it is definitely the new one. And to make sure 'beaglebone-standard-build' extra suffix doesn't affect it, I increased the path length even further - that extra level with 333 is there to over-compensate, as it was failing before even with just 111 and 222 levels only... Looks like Gary was also able to verify it.

Quoted 5 lines

But, you are right about vmlinux - it doesn't have those paths in there any more! I've seen them there when building with B != S, but they are gone when building with B = S. Wondering why. You can check it for yourself:

http://arago-project.org/files/short-term/misc/vmlinux-3.14.0-yocto-standard

And, as a follow up, all those long paths in vmlinux when built with B != S point to sources. Here are few examples:

B != S strings:

/OE/RAM/poky/tmp/work/beaglebone-poky-linux-gnueabi/linux-yocto/3.14+gitAUTOINC+928d7b2dda_0143c6ebb4-r0/linux/init/main.c earlycon
4Malformed early option '%s'
3Starting init: %s exists but couldn't execute it (error %d)
...
6Trying to unpack rootfs image as initramfs... /OE/RAM/poky/tmp/work/beaglebone-poky-linux-gnueabi/linux-yocto/3.14+gitAUTOINC+928d7b2dda_0143c6ebb4-r0/linux/init/initramfs.c 6rootfs image is not initramfs (%s); looks like an initrd
TRAILER!!!
...
%lu.%02lu BogoMIPS (lpj=%lu) /OE/RAM/poky/tmp/work/beaglebone-poky-linux-gnueabi/linux-yocto/3.14+gitAUTOINC+928d7b2dda_0143c6ebb4-r0/linux/arch/arm/vfp/vfpmodule.c 6VFP support v0.3:

B = S same strings:

init/main.c
earlycon
4Malformed early option '%s'
3Starting init: %s exists but couldn't execute it (error %d)
...
6Trying to unpack rootfs image as initramfs... init/initramfs.c
6rootfs image is not initramfs (%s); looks like an initrd TRAILER!!!
...
%lu.%02lu BogoMIPS (lpj=%lu) arch/arm/vfp/vfpmodule.c
6VFP support v0.3:

Quoted 8 lines
Quoted 3 lines

So, that explains why B = S works regardless of the location, as it doesn't embed full path to source files...

seemingly there could be rodata segment overflow issue. Can you do the kernel build with say linker from dora build
and see if it still fails.

It would be interesting to see if the rodata segment (or segment it contributes to) between working and non-working versions were crossing some magic power of 2 value.

Are there two images published from the same build machines with just the path difference? I would like to see one that was just under the wire and one just a bit too long. Denys, you have that?

#40 lore raw

https://bugzilla.yoctoproject.org/show_bug.cgi?id=6165

Link to the two kernels at the end of the first comment.

Quoted 137 lines · William Mills
Quoted 111 lines · Khem Raj
Quoted 108 lines · Denys Dmytriyenko
Quoted 53 lines · Denys Dmytriyenko
Quoted 28 lines · Richard Purdie
Quoted 17 lines · Denys Dmytriyenko

I don't yet know what is going on, but building in the same directory with
sources (B = S) makes it work regarless of the path length:

/OE/RAM/poky-111111111111111111111111111111 111111111111111111111111111111111111111111111111111111111111 111111111111111/222222222222222222222222222222 22222222222222222222222222222222222222/333333333333333333333333333333 3333333333333333333333/tmp/work/beaglebone-poky-linux- gnueabi/linux-yocto/3.14+gitAUTOINC+928d7b2dda_0143c6ebb4-r0/linux

So, I just commented out setting kernel-specific B in linux-yocto.inc and any
kernel now boots with long path:

#B = "${WORKDIR}/linux-${PACKAGE_ARCH}-${LINUX_KERNEL_TYPE}-build"

Quoted 8 lines

I'm copying Richard and Bruce directly to see if they may have a quick insight
and/or accept it as a workaround for the release. I'll keep digging further,
but if anyone cares to verify the above workaround works for them, I would
appreciate. Thanks!

Quoted 10 lines

I'm travelling and don't have hardware so I'm limited in what I can look
at right now. I suspect this workaround "works" because it makes the "beaglebone-standard-build" extra characters on the path. I have a feeling your -1111111 test above may have reused sstate or something like that and given misleading results. I'd be interested in the vmlinux
file from the build above to see if the poky-111111 pathnames are in there (they get stripped out when you create the zImage).

Nope, I was fooled by sstate once, so this time I tested with cleansstate and
compared timestamps of the kernel when it boots - it is definitely the new
one. And to make sure 'beaglebone-standard-build' extra suffix doesn't affect
it, I increased the path length even further - that extra level with 333 is
there to over-compensate, as it was failing before even with just 111 and 222
levels only... Looks like Gary was also able to verify it.

Quoted 9 lines

But, you are right about vmlinux - it doesn't have those paths in there any
more! I've seen them there when building with B != S, but they are gone when
building with B = S. Wondering why. You can check it for yourself:

http://arago-project.org/files/short-term/misc/vmlinux-
3.14.0-yocto-standard

And, as a follow up, all those long paths in vmlinux when built with B != S
point to sources. Here are few examples:

B != S strings:

/OE/RAM/poky/tmp/work/beaglebone-poky-linux-gnueabi/ linux-yocto/3.14+gitAUTOINC+928d7b2dda_0143c6ebb4-r0/linux/init/main.c earlycon
4Malformed early option '%s'
3Starting init: %s exists but couldn't execute it (error %d)
...
6Trying to unpack rootfs image as initramfs... /OE/RAM/poky/tmp/work/beaglebone-poky-linux-gnueabi/ linux-yocto/3.14+gitAUTOINC+928d7b2dda_0143c6ebb4-r0/ linux/init/initramfs.c
6rootfs image is not initramfs (%s); looks like an initrd TRAILER!!!
...
%lu.%02lu BogoMIPS (lpj=%lu) /OE/RAM/poky/tmp/work/beaglebone-poky-linux-gnueabi/ linux-yocto/3.14+gitAUTOINC+928d7b2dda_0143c6ebb4-r0/ linux/arch/arm/vfp/vfpmodule.c
6VFP support v0.3:

B = S same strings:

init/main.c
earlycon
4Malformed early option '%s'
3Starting init: %s exists but couldn't execute it (error %d)
...
6Trying to unpack rootfs image as initramfs... init/initramfs.c
6rootfs image is not initramfs (%s); looks like an initrd TRAILER!!!
...
%lu.%02lu BogoMIPS (lpj=%lu) arch/arm/vfp/vfpmodule.c
6VFP support v0.3:

Quoted 10 lines
Quoted 5 lines

So, that explains why B = S works regardless of the location, as it doesn't
embed full path to source files...

seemingly there could be rodata segment overflow issue. Can you do the kernel build with say linker from dora build
and see if it still fails.

It would be interesting to see if the rodata segment (or segment it contributes to) between working and non-working versions were crossing some magic power of 2 value.

Are there two images published from the same build machines with just the path difference? I would like to see one that was just under the wire and one just a bit too long. Denys, you have that? --
_______________________________________________
yocto mailing list
[email protected]
https://lists.yoctoproject.org/listinfo/yocto

#27 lore raw
In reply to Gary Thomas
Quoted 41 lines · Gary Thomas
Quoted 32 lines · Denys Dmytriyenko
Quoted 16 lines · Denys Dmytriyenko
Quoted 16 lines
Quoted 16 lines
Quoted 15 lines
Quoted 11 lines

Some other things I tried with a "long" TMPDIR path (note that it's the TMPDIR path that makes the difference - in my tests I've been using /home/paul/poky/build2/much/longer/path/to/tmp). None of this helped:

* kernel built with gcc 4.7.2 and binutils 2.23.2
* u-boot built with gcc 4.7.2 and binutils 2.23.2
* u-boot from http://downloads.angstrom-distribution.org/demo/beaglebone/
* earlyprintk and CONFIG_DEBUG_LL - no additional output printed

I think we're now at the point where we'd benefit from someone with better knowledge debugging the issue.

Ok, should we expand the search area? Since this is supposed to be vanilla 3.14 kernel, can we try other platforms and see if they are similarly affected? I'll try pinging our kernel guys for any ideas...

Quoted 6 lines
Quoted 6 lines
Quoted 4 lines

As far as I know it has only been observed with beaglebone (both white and black, if it makes a difference). FWIW, qemuarm images from the autobuilder boot just fine, and apparently the same is true of edgerouter (different architecture but also uses u-boot).

But do those other platforms use uImage or zImage?

I don't yet know what is going on, but building in the same directory with sources (B = S) makes it work regarless of the path length:

/OE/RAM/poky-111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111/22222222222222222222222222222222222222222222222222222222222222222222/3333333333333333333333333333333333333333333333333333/tmp/work/beaglebone-poky-linux-gnueabi/linux-yocto/3.14+gitAUTOINC+928d7b2dda_0143c6ebb4-r0/linux

So, I just commented out setting kernel-specific B in linux-yocto.inc and any kernel now boots with long path:

Quoted 7 lines

#B = "${WORKDIR}/linux-${PACKAGE_ARCH}-${LINUX_KERNEL_TYPE}-build"

I'm copying Richard and Bruce directly to see if they may have a quick insight and/or accept it as a workaround for the release. I'll keep digging further, but if anyone cares to verify the above workaround works for them, I would appreciate. Thanks!

Quoted 4 lines

Verified - I rebuilt the kernel in a working tree with a longer path (one in fact that had failed before) and it boots fine.

Wonder what ${B} != ${S} is doing wacky...?

Gary, et al,

I've just submitted a patch to oe-core and yocto MLs that fixes this issue - could you please test it in your setup and confirm? Thanks!

Signature · 1 line
Denys
#28 lore raw
Quoted 45 lines · Denys Dmytriyenko
Quoted 44 lines · Gary Thomas
Quoted 32 lines · Denys Dmytriyenko
Quoted 16 lines · Denys Dmytriyenko
Quoted 16 lines
Quoted 16 lines
Quoted 15 lines
Quoted 11 lines

Some other things I tried with a "long" TMPDIR path (note that it's the TMPDIR path that makes the difference - in my tests I've been using /home/paul/poky/build2/much/longer/path/to/tmp). None of this helped:

* kernel built with gcc 4.7.2 and binutils 2.23.2
* u-boot built with gcc 4.7.2 and binutils 2.23.2
* u-boot from http://downloads.angstrom-distribution.org/demo/beaglebone/
* earlyprintk and CONFIG_DEBUG_LL - no additional output printed

I think we're now at the point where we'd benefit from someone with better knowledge debugging the issue.

Ok, should we expand the search area? Since this is supposed to be vanilla 3.14 kernel, can we try other platforms and see if they are similarly affected? I'll try pinging our kernel guys for any ideas...

Quoted 6 lines
Quoted 6 lines
Quoted 4 lines

As far as I know it has only been observed with beaglebone (both white and black, if it makes a difference). FWIW, qemuarm images from the autobuilder boot just fine, and apparently the same is true of edgerouter (different architecture but also uses u-boot).

But do those other platforms use uImage or zImage?

I don't yet know what is going on, but building in the same directory with sources (B = S) makes it work regarless of the path length:

/OE/RAM/poky-111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111/22222222222222222222222222222222222222222222222222222222222222222222/3333333333333333333333333333333333333333333333333333/tmp/work/beaglebone-poky-linux-gnueabi/linux-yocto/3.14+gitAUTOINC+928d7b2dda_0143c6ebb4-r0/linux

So, I just commented out setting kernel-specific B in linux-yocto.inc and any kernel now boots with long path:

Quoted 7 lines

#B = "${WORKDIR}/linux-${PACKAGE_ARCH}-${LINUX_KERNEL_TYPE}-build"

I'm copying Richard and Bruce directly to see if they may have a quick insight and/or accept it as a workaround for the release. I'll keep digging further, but if anyone cares to verify the above workaround works for them, I would appreciate. Thanks!

Verified - I rebuilt the kernel in a working tree with a longer path (one in fact that had failed before) and it boots fine.

Quoted 7 lines
Quoted 1 line

Wonder what ${B} != ${S} is doing wacky...?

Gary, et al,

I've just submitted a patch to oe-core and yocto MLs that fixes this issue - could you please test it in your setup and confirm? Thanks!

I updated Stefan's bug w/ more explanation.
I verified that Stefan's uImage-bad failed for me and then added the following to uEnv.txt:
fdtaddr=0x88000000

uImage-bad (and uImage-good) worked w/ the above change to uEnv.txt. Denys' patch fixes all the defaults in u-boot so that no uEnv.txt change is needed.

#32 lore raw
In reply to Denys Dmytriyenko
Quoted 45 lines · Denys Dmytriyenko
Quoted 44 lines · Gary Thomas
Quoted 32 lines · Denys Dmytriyenko
Quoted 16 lines · Denys Dmytriyenko
Quoted 16 lines
Quoted 16 lines
Quoted 15 lines
Quoted 11 lines

Some other things I tried with a "long" TMPDIR path (note that it's the TMPDIR path that makes the difference - in my tests I've been using /home/paul/poky/build2/much/longer/path/to/tmp). None of this helped:

* kernel built with gcc 4.7.2 and binutils 2.23.2
* u-boot built with gcc 4.7.2 and binutils 2.23.2
* u-boot from http://downloads.angstrom-distribution.org/demo/beaglebone/
* earlyprintk and CONFIG_DEBUG_LL - no additional output printed

I think we're now at the point where we'd benefit from someone with better knowledge debugging the issue.

Ok, should we expand the search area? Since this is supposed to be vanilla 3.14 kernel, can we try other platforms and see if they are similarly affected? I'll try pinging our kernel guys for any ideas...

Quoted 6 lines
Quoted 6 lines
Quoted 4 lines

As far as I know it has only been observed with beaglebone (both white and black, if it makes a difference). FWIW, qemuarm images from the autobuilder boot just fine, and apparently the same is true of edgerouter (different architecture but also uses u-boot).

But do those other platforms use uImage or zImage?

I don't yet know what is going on, but building in the same directory with sources (B = S) makes it work regarless of the path length:

/OE/RAM/poky-111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111/22222222222222222222222222222222222222222222222222222222222222222222/3333333333333333333333333333333333333333333333333333/tmp/work/beaglebone-poky-linux-gnueabi/linux-yocto/3.14+gitAUTOINC+928d7b2dda_0143c6ebb4-r0/linux

So, I just commented out setting kernel-specific B in linux-yocto.inc and any kernel now boots with long path:

Quoted 7 lines

#B = "${WORKDIR}/linux-${PACKAGE_ARCH}-${LINUX_KERNEL_TYPE}-build"

I'm copying Richard and Bruce directly to see if they may have a quick insight and/or accept it as a workaround for the release. I'll keep digging further, but if anyone cares to verify the above workaround works for them, I would appreciate. Thanks!

Verified - I rebuilt the kernel in a working tree with a longer path (one in fact that had failed before) and it boots fine.

Quoted 7 lines
Quoted 1 line

Wonder what ${B} != ${S} is doing wacky...?

Gary, et al,

I've just submitted a patch to oe-core and yocto MLs that fixes this issue - could you please test it in your setup and confirm? Thanks!

Yes, verified it works great! Thanks for figuring this one out.

Signature · 4 lines
------------------------------------------------------------
Gary Thomas                 |  Consulting for the
MLB Associates              |    Embedded world
------------------------------------------------------------
#33 lore raw
Quoted 48 lines · Gary Thomas
Quoted 47 lines · Denys Dmytriyenko
Quoted 45 lines · Gary Thomas
Quoted 32 lines · Denys Dmytriyenko
Quoted 16 lines · Denys Dmytriyenko
Quoted 16 lines
Quoted 16 lines
Quoted 15 lines
Quoted 11 lines

Some other things I tried with a "long" TMPDIR path (note that it's the TMPDIR path that makes the difference - in my tests I've been using /home/paul/poky/build2/much/longer/path/to/tmp). None of this helped:

* kernel built with gcc 4.7.2 and binutils 2.23.2
* u-boot built with gcc 4.7.2 and binutils 2.23.2
* u-boot from http://downloads.angstrom-distribution.org/demo/beaglebone/
* earlyprintk and CONFIG_DEBUG_LL - no additional output printed

I think we're now at the point where we'd benefit from someone with better knowledge debugging the issue.

Ok, should we expand the search area? Since this is supposed to be vanilla 3.14 kernel, can we try other platforms and see if they are similarly affected? I'll try pinging our kernel guys for any ideas...

Quoted 6 lines
Quoted 6 lines
Quoted 4 lines

As far as I know it has only been observed with beaglebone (both white and black, if it makes a difference). FWIW, qemuarm images from the autobuilder boot just fine, and apparently the same is true of edgerouter (different architecture but also uses u-boot).

But do those other platforms use uImage or zImage?

I don't yet know what is going on, but building in the same directory with sources (B = S) makes it work regarless of the path length:

/OE/RAM/poky-111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111/22222222222222222222222222222222222222222222222222222222222222222222/3333333333333333333333333333333333333333333333333333/tmp/work/beaglebone-poky-linux-gnueabi/linux-yocto/3.14+gitAUTOINC+928d7b2dda_0143c6ebb4-r0/linux

So, I just commented out setting kernel-specific B in linux-yocto.inc and any kernel now boots with long path:

Quoted 7 lines

#B = "${WORKDIR}/linux-${PACKAGE_ARCH}-${LINUX_KERNEL_TYPE}-build"

I'm copying Richard and Bruce directly to see if they may have a quick insight and/or accept it as a workaround for the release. I'll keep digging further, but if anyone cares to verify the above workaround works for them, I would appreciate. Thanks!

Verified - I rebuilt the kernel in a working tree with a longer path (one in fact that had failed before) and it boots fine.

Wonder what ${B} != ${S} is doing wacky...?

Quoted 7 lines
Quoted 5 lines

Gary, et al,

I've just submitted a patch to oe-core and yocto MLs that fixes this issue - could you please test it in your setup and confirm? Thanks!

Yes, verified it works great! Thanks for figuring this one out.

And thank you for stumbling upon it in the first place :) and your continued help validating it until completion! Thanks Stefan, Bill and everyone else involed. Now let's hope Richard takes it into the release shortly... :)

Signature · 1 line
Denys
#29 lore raw
In reply to William Mills
Quoted 79 lines · William Mills
Quoted 68 lines · Denys Dmytriyenko
Quoted 66 lines · Gary Thomas
Quoted 52 lines · Denys Dmytriyenko
Quoted 34 lines · Denys Dmytriyenko
Quoted 34 lines
Quoted 32 lines
Quoted 18 lines
Quoted 13 lines

Some other things I tried with a "long" TMPDIR path (note that it's the
TMPDIR path that makes the difference - in my tests I've been using /home/paul/poky/build2/much/longer/path/to/tmp). None of this helped:

* kernel built with gcc 4.7.2 and binutils 2.23.2
* u-boot built with gcc 4.7.2 and binutils 2.23.2
* u-boot from
http://downloads.angstrom-distribution.org/demo/beaglebone/
* earlyprintk and CONFIG_DEBUG_LL - no additional output printed

Quoted 3 lines

I think we're now at the point where we'd benefit from someone with better
knowledge debugging the issue.

Quoted 5 lines

Ok, should we expand the search area? Since this is supposed to be vanilla
3.14 kernel, can we try other platforms and see if they are similarly
affected? I'll try pinging our kernel guys for any ideas...

As far as I know it has only been observed with beaglebone (both white and
black, if it makes a difference). FWIW, qemuarm images from the autobuilder
boot just fine, and apparently the same is true of edgerouter (different
architecture but also uses u-boot).

Quoted 1 line
Quoted 1 line

But do those other platforms use uImage or zImage?

I don't yet know what is going on, but building in the same directory with
sources (B = S) makes it work regarless of the path length:

/OE/RAM/poky-111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111/22222222222222222222222222222222222222222222222222222222222222222222/3333333333333333333333333333333333333333333333333333/tmp/work/beaglebone-poky-linux-gnueabi/linux-yocto/3.14+gitAUTOINC+928d7b2dda_0143c6ebb4-r0/linux

So, I just commented out setting kernel-specific B in linux-yocto.inc and any
kernel now boots with long path:

#B = "${WORKDIR}/linux-${PACKAGE_ARCH}-${LINUX_KERNEL_TYPE}-build"

Quoted 8 lines

I'm copying Richard and Bruce directly to see if they may have a quick insight
and/or accept it as a workaround for the release. I'll keep digging further,
but if anyone cares to verify the above workaround works for them, I would
appreciate. Thanks!

Verified - I rebuilt the kernel in a working tree with a longer path (one in fact that had failed before) and it boots fine.

Quoted 9 lines
Quoted 1 line

Wonder what ${B} != ${S} is doing wacky...?

Gary, et al,

I've just submitted a patch to oe-core and yocto MLs that fixes this issue -
could you please test it in your setup and confirm? Thanks!

Quoted 4 lines

I updated Stefan's bug w/ more explanation.
I verified that Stefan's uImage-bad failed for me and then added the following to uEnv.txt:
fdtaddr=0x88000000

hmmm so it seems kernel size grew enough to overwrite the area where uboot would put divice tree ?
has happened to me on ppc with kernel+initramfs where initramfs kept growing

Quoted 10 lines

uImage-bad (and uImage-good) worked w/ the above change to uEnv.txt. Denys' patch fixes all the defaults in u-boot so that no uEnv.txt change is needed.

--
_______________________________________________
yocto mailing list
[email protected]
https://lists.yoctoproject.org/listinfo/yocto

#30 lore raw
Quoted 80 lines · Khem Raj
Quoted 79 lines · William Mills
Quoted 77 lines · Denys Dmytriyenko
Quoted 63 lines · Gary Thomas
Quoted 52 lines · Denys Dmytriyenko
Quoted 34 lines · Denys Dmytriyenko
Quoted 34 lines
Quoted 32 lines
Quoted 18 lines
Quoted 13 lines

Some other things I tried with a "long" TMPDIR path (note that it's the
TMPDIR path that makes the difference - in my tests I've been using /home/paul/poky/build2/much/longer/path/to/tmp). None of this helped:

* kernel built with gcc 4.7.2 and binutils 2.23.2
* u-boot built with gcc 4.7.2 and binutils 2.23.2
* u-boot from
http://downloads.angstrom-distribution.org/demo/beaglebone/
* earlyprintk and CONFIG_DEBUG_LL - no additional output printed

Quoted 3 lines

I think we're now at the point where we'd benefit from someone with better
knowledge debugging the issue.

Quoted 5 lines

Ok, should we expand the search area? Since this is supposed to be vanilla
3.14 kernel, can we try other platforms and see if they are similarly
affected? I'll try pinging our kernel guys for any ideas...

As far as I know it has only been observed with beaglebone (both white and
black, if it makes a difference). FWIW, qemuarm images from the autobuilder
boot just fine, and apparently the same is true of edgerouter (different
architecture but also uses u-boot).

Quoted 1 line
Quoted 1 line

But do those other platforms use uImage or zImage?

I don't yet know what is going on, but building in the same directory with
sources (B = S) makes it work regarless of the path length:

/OE/RAM/poky-111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111/22222222222222222222222222222222222222222222222222222222222222222222/3333333333333333333333333333333333333333333333333333/tmp/work/beaglebone-poky-linux-gnueabi/linux-yocto/3.14+gitAUTOINC+928d7b2dda_0143c6ebb4-r0/linux

So, I just commented out setting kernel-specific B in linux-yocto.inc and any
kernel now boots with long path:

#B = "${WORKDIR}/linux-${PACKAGE_ARCH}-${LINUX_KERNEL_TYPE}-build"

Quoted 8 lines

I'm copying Richard and Bruce directly to see if they may have a quick insight
and/or accept it as a workaround for the release. I'll keep digging further,
but if anyone cares to verify the above workaround works for them, I would
appreciate. Thanks!

Quoted 4 lines

Verified - I rebuilt the kernel in a working tree with a longer path (one in fact that had failed before) and it boots fine.

Wonder what ${B} != ${S} is doing wacky...?

Gary, et al,

I've just submitted a patch to oe-core and yocto MLs that fixes this issue -
could you please test it in your setup and confirm? Thanks!

Quoted 7 lines
Quoted 4 lines

I updated Stefan's bug w/ more explanation.
I verified that Stefan's uImage-bad failed for me and then added the following to uEnv.txt:
fdtaddr=0x88000000

hmmm so it seems kernel size grew enough to overwrite the area where uboot would put divice tree ?

Exactly! The overall kernel size was very close to the limit of previous space allocation in U-boot and enabling another feature or growing the build path a bit would overflow into and corrupt the device tree.

Quoted 1 line

has happened to me on ppc with kernel+initramfs where initramfs kept growing

I was told no other platforms were affected, including qemuarm! :) Yeah, I know, it can happen to anyone, even though by default it works fine...

Quoted 10 lines
Quoted 9 lines

uImage-bad (and uImage-good) worked w/ the above change to uEnv.txt. Denys' patch fixes all the defaults in u-boot so that no uEnv.txt change is needed.

--
_______________________________________________
yocto mailing list
[email protected]
https://lists.yoctoproject.org/listinfo/yocto

#31 lore raw
Quoted 94 lines · Denys Dmytriyenko
Quoted 80 lines · Khem Raj
Quoted 79 lines · William Mills
Quoted 77 lines · Denys Dmytriyenko
Quoted 63 lines · Gary Thomas
Quoted 52 lines · Denys Dmytriyenko
Quoted 34 lines · Denys Dmytriyenko
Quoted 34 lines
Quoted 32 lines
Quoted 18 lines
Quoted 13 lines

Some other things I tried with a "long" TMPDIR path (note that it's the
TMPDIR path that makes the difference - in my tests I've been using /home/paul/poky/build2/much/longer/path/to/tmp). None of this helped:

* kernel built with gcc 4.7.2 and binutils 2.23.2
* u-boot built with gcc 4.7.2 and binutils 2.23.2
* u-boot from
http://downloads.angstrom-distribution.org/demo/beaglebone/
* earlyprintk and CONFIG_DEBUG_LL - no additional output printed

Quoted 3 lines

I think we're now at the point where we'd benefit from someone with better
knowledge debugging the issue.

Quoted 5 lines

Ok, should we expand the search area? Since this is supposed to be vanilla
3.14 kernel, can we try other platforms and see if they are similarly
affected? I'll try pinging our kernel guys for any ideas...

As far as I know it has only been observed with beaglebone (both white and
black, if it makes a difference). FWIW, qemuarm images from the autobuilder
boot just fine, and apparently the same is true of edgerouter (different
architecture but also uses u-boot).

Quoted 1 line
Quoted 1 line

But do those other platforms use uImage or zImage?

I don't yet know what is going on, but building in the same directory with
sources (B = S) makes it work regarless of the path length:

/OE/RAM/poky-111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111/22222222222222222222222222222222222222222222222222222222222222222222/3333333333333333333333333333333333333333333333333333/tmp/work/beaglebone-poky-linux-gnueabi/linux-yocto/3.14+gitAUTOINC+928d7b2dda_0143c6ebb4-r0/linux

So, I just commented out setting kernel-specific B in linux-yocto.inc and any
kernel now boots with long path:

#B = "${WORKDIR}/linux-${PACKAGE_ARCH}-${LINUX_KERNEL_TYPE}-build"

Quoted 8 lines

I'm copying Richard and Bruce directly to see if they may have a quick insight
and/or accept it as a workaround for the release. I'll keep digging further,
but if anyone cares to verify the above workaround works for them, I would
appreciate. Thanks!

Quoted 4 lines

Verified - I rebuilt the kernel in a working tree with a longer path (one in fact that had failed before) and it boots fine.

Wonder what ${B} != ${S} is doing wacky...?

Gary, et al,

I've just submitted a patch to oe-core and yocto MLs that fixes this issue -
could you please test it in your setup and confirm? Thanks!

Quoted 7 lines
Quoted 4 lines

I updated Stefan's bug w/ more explanation.
I verified that Stefan's uImage-bad failed for me and then added the following to uEnv.txt:
fdtaddr=0x88000000

hmmm so it seems kernel size grew enough to overwrite the area where uboot would put divice tree ?

Exactly! The overall kernel size was very close to the limit of previous space m, allocation in U-boot and enabling another feature or growing the build path a bit would overflow into and corrupt the device tree.

Quoted 5 lines
Quoted 1 line

has happened to me on ppc with kernel+initramfs where initramfs kept growing

I was told no other platforms were affected, including qemuarm! :) Yeah, I know, it can happen to anyone, even though by default it works fine...

I did not look at qemuarm but I assume it was using different values for for loadaddr & fdtaddr. (qemuarm is using fdt now right?)

Khem, yes this would have effected initrd also if the kernel was just a touch bigger. The patch Denys pushed moves the intrd load address as well. We should be good now.

If initramfs gets trashed you get a message when the kernel tries to decompresses it that it is invalid. If fdt gets trashed the kernel does not know how to send messages to uart so you get nothing. I am not sure but when I was debugging with JTAG, it looked like the kernel had actually started and was doing _something_. It just had no hardware to talk to.

If a kernel boots in a forest and has no way to talk, does it have any bogomips?

Quoted 11 lines
Quoted 11 lines
Quoted 10 lines

uImage-bad (and uImage-good) worked w/ the above change to uEnv.txt. Denys' patch fixes all the defaults in u-boot so that no uEnv.txt change is needed.

--
_______________________________________________
yocto mailing list
[email protected]
https://lists.yoctoproject.org/listinfo/yocto