Quantcast

Frustrated findings from running Yocto/Denzil on Gumstix/Overo

classic Classic list List threaded Threaded
12 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Frustrated findings from running Yocto/Denzil on Gumstix/Overo

Arne Lie
Hi again,

Findings from today after running the Yocto/Denzil build from this morning, using:
PREFERRED_PROVIDER_udev = "udev"
in conf/local.conf

System:
Yocto (Built by Poky 7.0.2) 1.2.2 overo ttyO2

Linux:
Linux overo 3.2.0 #1 PREEMPT Thu Jan 17 12:54:54 CET 2013 armv7l GNU/Linux

root@overo:~# systemd-analyze                                                  
Startup finished in 3571ms (kernel) + 33319ms (userspace) = 36891ms

My goal is to make a much faster userspace boot-time, 33 seconds it 20 too much. I see reports on 8 seconds (EarthStorm/Tobi).
And, hopefully, to make the system run on fewer milliamperes, currently, I am somewhere between 240-300 mA...

So, how to I avoid starting not needed services/devices, e.g., bluetooth/rfcomm/hci services? I have spent some time learning about systemd, which I now understand has replaced sysinitv. But I cannot understand which servcies is started, from where... And,  it seems that my buildsystem is a hybrid systemd/sysinitv, and it also reports about LSB: E.g.,
* /etc/inittab is still there, but is it in use?
* /etc/init.d/ is populated with some files, but are they used?
* /etc/rcX.d directories still exist, and they are populated with some symbolic links
* my system uses BusyBox, why?
* opkg seems not to be able to install anything
* many systemd tools I find in the internet literature is not there

The frustrating thing is that the Linux/Gumstix universe is now moving so fast that most of the online literature is not valid anymore. And I cannot seem to find any valid documentation for the build I have. And, I know that the bleeding edge is at Linux 3.6, so...

Hopefully, somebody has some pointers that I do not find via Google..??

Best regards

Arne


> -----Original Message-----
> From: Arne Lie [mailto:[hidden email]]
> Sent: 18. januar 2013 09:31
> To: General mailing list for gumstix users.
> Subject: Re: [Gumstix-users] Building Yocto for Gumstix/Overo (RE:
> Connection Refused to Steve's Git repo)
>
> Hi,
>
> Just did a bitbake jumpnow-console-image with your
> PREFERRED_PROVIDER_udev = "udev" , and it went all the way with building
> xloader, u-boot, uImage, and rootfs. Thanks so much!!! :-)
>
> Best regards
> Arne
>

------------------------------------------------------------------------------
Master HTML5, CSS3, ASP.NET, MVC, AJAX, Knockout.js, Web API and
much more. Get web development skills now with LearnDevNow -
350+ hours of step-by-step video tutorials by Microsoft MVPs and experts.
SALE $99.99 this month only -- learn more at:
http://p.sf.net/sfu/learnmore_122812
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Frustrated findings from running Yocto/Denzil on Gumstix/Overo

Andreas Müller-2
On Fri, Jan 18, 2013 at 4:44 PM, Arne Lie <[hidden email]> wrote:

> Hi again,
>
> Findings from today after running the Yocto/Denzil build from this morning, using:
> PREFERRED_PROVIDER_udev = "udev"
> in conf/local.conf
>
> System:
> Yocto (Built by Poky 7.0.2) 1.2.2 overo ttyO2
>
> Linux:
> Linux overo 3.2.0 #1 PREEMPT Thu Jan 17 12:54:54 CET 2013 armv7l GNU/Linux
>
> root@overo:~# systemd-analyze
> Startup finished in 3571ms (kernel) + 33319ms (userspace) = 36891ms
>
> My goal is to make a much faster userspace boot-time, 33 seconds it 20 too much. I see reports on 8 seconds (EarthStorm/Tobi).
> And, hopefully, to make the system run on fewer milliamperes, currently, I am somewhere between 240-300 mA...
>
> So, how to I avoid starting not needed services/devices, e.g., bluetooth/rfcomm/hci services? I have spent some time learning about systemd, which I now understand has replaced sysinitv. But I cannot understand which servcies is started, from where... And,  it seems that my buildsystem is a hybrid systemd/sysinitv, and it also reports about LSB: E.g.,
> * /etc/inittab is still there, but is it in use?
> * /etc/init.d/ is populated with some files, but are they used?
> * /etc/rcX.d directories still exist, and they are populated with some symbolic links
> * my system uses BusyBox, why?
> * opkg seems not to be able to install anything
> * many systemd tools I find in the internet literature is not there
>
> The frustrating thing is that the Linux/Gumstix universe is now moving so fast that most of the online literature is not valid anymore. And I cannot seem to find any valid documentation for the build I have. And, I know that the bleeding edge is at Linux 3.6, so...
>
> Hopefully, somebody has some pointers that I do not find via Google..??
>
> Best regards
>
> Arne
>
>
>> -----Original Message-----
>> From: Arne Lie [mailto:[hidden email]]
>> Sent: 18. januar 2013 09:31
>> To: General mailing list for gumstix users.
>> Subject: Re: [Gumstix-users] Building Yocto for Gumstix/Overo (RE:
>> Connection Refused to Steve's Git repo)
>>
>> Hi,
>>
>> Just did a bitbake jumpnow-console-image with your
>> PREFERRED_PROVIDER_udev = "udev" , and it went all the way with building
>> xloader, u-boot, uImage, and rootfs. Thanks so much!!! :-)
>>
>> Best regards
>> Arne
>>
>
systemd-analyse blame

gives more details. For creatin a graph there is also an option (I
don't have it here on my win PC.. - systemd-analyse --help) Note: on
first boot it might last longer for postint tasks. Same for further
boots?

To enable/disable

systemctl disable foo.service

should help. See man systemctl.

The available service files should be found at /lib/systemd/system.
See man systemd.unit.

Hope it helps a bit

Andreas

------------------------------------------------------------------------------
Master HTML5, CSS3, ASP.NET, MVC, AJAX, Knockout.js, Web API and
much more. Get web development skills now with LearnDevNow -
350+ hours of step-by-step video tutorials by Microsoft MVPs and experts.
SALE $99.99 this month only -- learn more at:
http://p.sf.net/sfu/learnmore_122812
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Frustrated findings from running Yocto/Denzil on Gumstix/Overo

Scott Ellis
In reply to this post by Arne Lie
I've found a lot of boot time speed depends on the SD card.

Here are measurements from an AirStorm running the jumpnow-console-image
built yesterday. All times are the third boot from a complete power off.

Each card was partitioned and formatted prior to copying the binaries.

A SanDisk Ultra 8GB Class 6

root@overo:~# systemd-analyze
Startup finished in 3246ms (kernel) + 3824ms (userspace) = 7071ms

root@overo:~# systemd-analyze blame
  1203ms networking.service
   954ms populate-volatile.service
   896ms systemd-modules-load.service
   700ms udev-trigger.service
   510ms checkroot.service
   422ms mountall.service
   420ms hostname.service
   405ms sysfs.service
   404ms banner.service
   378ms systemd-sysctl.service
   372ms modutils.service
   371ms alignment.service
   364ms systemd-remount-api-vfs.service
   356ms tmp.mount
   342ms sys-kernel-debug.mount
   321ms remount-rootfs.service
   224ms devpts.service
   218ms systemd-logind.service
   170ms systemd-tmpfiles-setup.service
   126ms dbus.service
   125ms sshdgenkeys.service
   114ms stop-bootlogd.service
   111ms rmnologin.service
    76ms udev.service
    41ms sys-fs-fuse-connections.mount
    40ms systemd-user-sessions.service
    16ms sshd.socket


A PNY 8GB Class 6

root@overo:~# systemd-analyze
Startup finished in 3266ms (kernel) + 20816ms (userspace) = 24083ms

root@overo:~# systemd-analyze blame
 18262ms checkroot.service
  9799ms remount-rootfs.service
  9118ms populate-volatile.service
  9071ms networking.service
  5344ms mountall.service
  5257ms devpts.service
   677ms udev-trigger.service
   658ms systemd-tmpfiles-setup.service
   479ms systemd-modules-load.service
   427ms sshd.socket
   405ms hostname.service
   405ms sysfs.service
   404ms systemd-remount-api-vfs.service
   403ms tmp.mount
   397ms alignment.service
   396ms banner.service
   388ms modutils.service
   350ms sys-kernel-debug.mount
   350ms systemd-sysctl.service
   209ms dbus.service
   174ms sshdgenkeys.service
   148ms sys-fs-fuse-connections.mount
    88ms stop-bootlogd.service
    86ms rmnologin.service
    82ms udev.service
    48ms systemd-user-sessions.service


An HC 4GB Class 4 card

root@overo:~# systemd-analyze
Startup finished in 3208ms (kernel) + 8618ms (userspace) = 11826ms

root@overo:~# systemd-analyze blame
  6765ms checkroot.service
  3709ms networking.service
  3669ms populate-volatile.service
  3391ms remount-rootfs.service
  1006ms systemd-tmpfiles-setup.service
   646ms udev-trigger.service
   529ms systemd-modules-load.service
   507ms mountall.service
   506ms alignment.service
   502ms systemd-remount-api-vfs.service
   500ms hostname.service
   499ms sysfs.service
   497ms banner.service
   494ms tmp.mount
   443ms sys-kernel-debug.mount
   443ms devpts.service
   442ms systemd-sysctl.service
   194ms sshdgenkeys.service
   174ms modutils.service
   125ms dbus.service
   100ms sys-fs-fuse-connections.mount
    89ms stop-bootlogd.service
    77ms udev.service
    77ms rmnologin.service
    58ms systemd-user-sessions.service
    12ms sshd.socket

Another SanDisk Ultra 8GB

root@overo:~# systemd-analyze
Startup finished in 3232ms (kernel) + 3777ms (userspace) = 7010ms

root@overo:~# systemd-analyze blame
  1145ms networking.service
   966ms populate-volatile.service
   787ms systemd-modules-load.service
   763ms udev-trigger.service
   431ms sysfs.service
   431ms banner.service
   428ms alignment.service
   426ms hostname.service
   425ms sys-kernel-debug.mount
   421ms mountall.service
   420ms systemd-remount-api-vfs.service
   418ms tmp.mount
   350ms systemd-sysctl.service
   344ms remount-rootfs.service
   319ms systemd-tmpfiles-setup.service
   301ms checkroot.service
   296ms modutils.service
   247ms devpts.service
   154ms sshdgenkeys.service
   140ms dbus.service
   123ms stop-bootlogd.service
    95ms rmnologin.service
    68ms udev.service
    60ms sys-fs-fuse-connections.mount
    56ms systemd-user-sessions.service
    13ms sshd.socket


A PNY 2GB card

root@overo:~# systemd-analyze

Startup finished in 3645ms (kernel) + 18198ms (userspace) = 21844ms
root@overo:~# systemd-analyze blame
 16158ms checkroot.service
  8252ms remount-rootfs.service
  8040ms networking.service
  4011ms populate-volatile.service
  3939ms mountall.service
  2918ms devpts.service
   750ms systemd-modules-load.service
   593ms udev-trigger.service
   469ms banner.service
   468ms sysfs.service
   467ms alignment.service
   432ms hostname.service
   372ms systemd-logind.service
   366ms systemd-sysctl.service
   359ms sys-kernel-debug.mount
   349ms systemd-remount-api-vfs.service
   313ms tmp.mount
   310ms sshd.socket
   181ms dbus.service
   177ms sshdgenkeys.service
   152ms stop-bootlogd.service
   140ms rmnologin.service
   128ms systemd-tmpfiles-setup.service
    94ms modutils.service
    81ms udev.service
    43ms systemd-user-sessions.service
    22ms sys-fs-fuse-connections.mount

Definitely bluetooth, wifi/dhcp and the Caspa kernel driver are time hogs and
I'd remove any that you don't need.
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Frustrated findings from running Yocto/Denzil on Gumstix/Overo

paul@idruna.com
Hi Scott,

Thought I`d give your instructions a go as I was having a Bluetooth
problem, and was hoping upgrading the whole OS and kernel might help.
It made the BT problem worse, so I thought I`d try the 3.6 kernel, do
you have a known good defconfig for this by any chance?  I`m using the
Earth/EarthStorm boards.  Using the omap3-multi defconfig from Steve's
site hangs with the same error as others report on the duovero:

Saw some messages with a similar error, just in case this helps
Waiting for root device /dev/mmcblk0p2...
U-Boot 2012.10-00007-g2d44b29 (Jan 18 2013 - 17:29:44)

OMAP3530-GP ES3.1, CPU-OPP2, L3-165MHz, Max CPU Clock 720 mHz
Gumstix Overo board + LPDDR/NAND
I2C:   ready
DRAM:  512 MiB
NAND:  512 MiB
MMC:   OMAP SD/MMC: 0
*** Warning - bad CRC, using default environment

In:    serial
Out:   serial
Err:   serial
Board revision: 1
Direct connection on mmc2
No EEPROM on expansion board
Die ID #0c780004000000000403990c0b022018
Net:   No ethernet found.
Hit any key to stop autoboot:  0
reading boot.scr

** Unable to read "boot.scr" from mmc 0:1 **
reading uImage

3478736 bytes read
Booting from mmc ...
## Booting kernel from Legacy Image at 82000000 ...
    Image Name:   Linux-3.6.0
    Image Type:   ARM Linux Kernel Image (uncompressed)
    Data Size:    3478672 Bytes = 3.3 MiB
    Load Address: 80008000
    Entry Point:  80008000
    Verifying Checksum ... OK
    Loading Kernel Image ... OK
OK

Starting kernel ...

Uncompressing Linux... done, booting the kernel.
[    0.000000] Booting Linux on physical CPU 0
[    0.000000] Initializing cgroup subsys cpuset
[    0.000000] Initializing cgroup subsys cpu
[    0.000000] Linux version 3.6.0 (paul@ubuntu) (gcc version 4.6.4
20120303 (prerelease) (GCC) ) #1 PREEMPT Sun Jan 20 09:29:05 PST 2013
[    0.000000] CPU: ARMv7 Processor [411fc083] revision 3 (ARMv7),
cr=10c53c7d
[    0.000000] CPU: PIPT / VIPT nonaliasing data cache, VIPT nonaliasing
instruction cache
[    0.000000] Machine: Gumstix Overo
[    0.000000] Reserving 12582912 bytes SDRAM for VRAM
[    0.000000] Memory policy: ECC disabled, Data cache writeback
[    0.000000] OMAP3430/3530 ES3.1 (l2cache iva sgx neon isp 720mhz )
[    0.000000] Clocking rate (Crystal/Core/MPU): 26.0/332/600 MHz
[    0.000000] Built 1 zonelists in Zone order, mobility grouping on.
Total pages: 126720
[    0.000000] Kernel command line: console=ttyO2,115200n8 mpurate=500
vram=12M omapfb.mode=dvi:1024x768MR-16@60 omapdss.def_disp=dvi
root=/dev/mmcblk0p2 rw rootfstype=ext3 rootwait
[    0.000000] PID hash table entries: 2048 (order: 1, 8192 bytes)
[    0.000000] Dentry cache hash table entries: 65536 (order: 6, 262144
bytes)
[    0.000000] Inode-cache hash table entries: 32768 (order: 5, 131072
bytes)
[    0.000000] Memory: 499MB = 499MB total
[    0.000000] Memory: 498704k/498704k available, 25584k reserved, 0K
highmem
[    0.000000] Virtual kernel memory layout:
[    0.000000]     vector  : 0xffff0000 - 0xffff1000   (   4 kB)
[    0.000000]     fixmap  : 0xfff00000 - 0xfffe0000   ( 896 kB)
[    0.000000]     vmalloc : 0xe0800000 - 0xff000000   ( 488 MB)
[    0.000000]     lowmem  : 0xc0000000 - 0xe0000000   ( 512 MB)
[    0.000000]     pkmap   : 0xbfe00000 - 0xc0000000   (   2 MB)
[    0.000000]     modules : 0xbf000000 - 0xbfe00000   (  14 MB)
[    0.000000]       .text : 0xc0008000 - 0xc065ed18   (6492 kB)
[    0.000000]       .init : 0xc065f000 - 0xc0695e20   ( 220 kB)
[    0.000000]       .data : 0xc0696000 - 0xc0702238   ( 433 kB)
[    0.000000]        .bss : 0xc070225c - 0xc078c3e8   ( 553 kB)
[    0.000000] NR_IRQS:474
[    0.000000] IRQ: Found an INTC at 0xfa200000 (revision 4.0) with 96
interrupts
[    0.000000] Total of 96 interrupts on 1 active controller
[    0.000000] OMAP clockevent source: GPTIMER1 at 32768 Hz
[    0.000000] sched_clock: 32 bits at 32kHz, resolution 30517ns, wraps
every 131071999ms
[    0.000000] OMAP clocksource: 32k_counter at 32768 Hz
[    0.000000] Console: colour dummy device 80x30
[    0.000213] Calibrating delay loop... 399.14 BogoMIPS (lpj=1560576)
[    0.054718] pid_max: default: 32768 minimum: 301
[    0.054840] Security Framework initialized
[    0.054931] Mount-cache hash table entries: 512
[    0.055786] Initializing cgroup subsys debug
[    0.055816] Initializing cgroup subsys cpuacct
[    0.055816] Initializing cgroup subsys devices
[    0.055816] Initializing cgroup subsys freezer
[    0.055847] Initializing cgroup subsys perf_event
[    0.055969] CPU: Testing write buffer coherency: ok
[    0.056335] Setting up static identity map for 0x80464f10 - 0x80464f68
[    0.061157] devtmpfs: initialized
[    0.066436] dummy:
[    0.066833] NET: Registered protocol family 16
[    0.068206] DMA: preallocated 256 KiB pool for atomic coherent
allocations
[    0.068420] GPMC revision 5.0
[    0.068450] gpmc: irq-20 could not claim: err -22
[    0.070587] OMAP GPIO hardware version 2.5
[    0.073486] omap_mux_init: Add partition: #1: core, flags: 0
[    0.085083] Reprogramming SDRC clock to 332000000 Hz
[    0.085113] dpll3_m2_clk rate change failed: -22
[    0.085113] Found NAND on CS0
[    0.085144] Registering NAND on CS0
[    0.086517]  usbhs_omap: alias fck already exists
[    0.087493] platform iva.0: Enabled OPP corresponding to 720MHz
[    0.094329] hw-breakpoint: debug architecture 0x4 unsupported.
[    0.097991]  omap-mcbsp.2: alias fck already exists
[    0.098297]  omap-mcbsp.3: alias fck already exists
[    0.099517] Switched to new clocking rate (Crystal/Core/MPU):
26.0/332/500 MHz
[    0.099761] OMAP DMA hardware revision 4.0
[    0.111267] bio: create slab <bio-0> at 0
[    0.112152] fixed-dummy:
[    0.112579] vads7846: 3300 mV
[    0.113555] SCSI subsystem initialized
[    0.114013] usbcore: registered new interface driver usbfs
[    0.114166] usbcore: registered new interface driver hub
[    0.114501] usbcore: registered new device driver usb
[    0.115509] omap_i2c omap_i2c.1: bus 1 rev1.3.12 at 2600 kHz
[    0.118957] twl 1-0048: PIH (irq 7) chaining IRQs 320..328
[    0.119110] twl 1-0048: power (irq 325) chaining IRQs 328..335
[    0.119873] twl4030_gpio twl4030_gpio: gpio (irq 320) chaining IRQs
336..353
[    0.121429] VUSB1V5: 1500 mV normal standby
[    0.122161] VUSB1V8: 1800 mV normal standby
[    0.122802] VUSB3V1: 3100 mV normal standby
[    0.127349] musb-omap2430 musb-omap2430: musb core is not yet ready
[    0.127380] twl4030_usb twl4030_usb: Initialized TWL4030 USB module
[    0.128204] vdd_mpu_iva: 600 <--> 1450 mV normal
[    0.128784] vdd_core: 600 <--> 1450 mV normal
[    0.129486] VMMC1: 1850 <--> 3150 mV at 3150 mV normal standby
[    0.130218] VDAC: 1800 mV normal standby
[    0.130920] VDVI: 1800 mV normal standby
[    0.131591] omap_i2c omap_i2c.3: bus 3 rev1.3.12 at 400 kHz
[    0.132659] omap-iommu omap-iommu.0: isp registered
[    0.132965] Advanced Linux Sound Architecture Driver Version 1.0.25.
[    0.134155] Switching to clocksource 32k_counter
[    0.173767] usbhs_omap usbhs_omap: xclk60mhsp2_ck set parentfailed
error:-22
[    0.174682] NET: Registered protocol family 2
[    0.175567] TCP established hash table entries: 16384 (order: 5,
131072 bytes)
[    0.176025] TCP bind hash table entries: 16384 (order: 4, 65536 bytes)
[    0.176269] TCP: Hash tables configured (established 16384 bind 16384)
[    0.176391] TCP: reno registered
[    0.176422] UDP hash table entries: 256 (order: 0, 4096 bytes)
[    0.176452] UDP-Lite hash table entries: 256 (order: 0, 4096 bytes)
[    0.176818] NET: Registered protocol family 1
[    0.177398] RPC: Registered named UNIX socket transport module.
[    0.177429] RPC: Registered udp transport module.
[    0.177429] RPC: Registered tcp transport module.
[    0.177459] RPC: Registered tcp NFSv4.1 backchannel transport module.
[    0.178466] platform iva.0: Enabled OPP corresponding to 720MHz
[    0.181457] audit: initializing netlink socket (disabled)
[    0.181549] type=2000 audit(0.179:1): initialized
[    0.183166] VFS: Disk quotas dquot_6.5.2
[    0.183258] Dquot-cache hash table entries: 1024 (order 0, 4096 bytes)
[    0.183929] NFS: Registering the id_resolver key type
[    0.184020] Key type id_resolver registered
[    0.184051] Key type id_legacy registered
[    0.184112] jffs2: version 2.2. (NAND) (SUMMARY)  Â© 2001-2006 Red
Hat, Inc.
[    0.184753] fuse init (API version 7.20)
[    0.185394] msgmni has been set to 974
[    0.188110] Block layer SCSI generic (bsg) driver version 0.4 loaded
(major 253)
[    0.188140] io scheduler noop registered
[    0.188140] io scheduler deadline registered (default)
[    0.189392] OMAP DSS rev 2.0
[    0.193481] omap_uart.0: ttyO0 at MMIO 0x4806a000 (irq = 72) is a
OMAP UART0
[    0.194152] omap_uart.1: ttyO1 at MMIO 0x4806c000 (irq = 73) is a
OMAP UART1
[    0.194793] omap_uart.2: ttyO2 at MMIO 0x49020000 (irq = 74) is a
OMAP UART2
[    0.879364] console [ttyO2] enabled
[    0.895233] brd: module loaded
[    0.904632] loop: module loaded
[    0.909942] twl4030_madc twl4030_madc: clk disabled, enabling
[    0.918518] NAND device: Manufacturer ID: 0x2c, Chip ID: 0xbc (Micron
NAND 512MiB 1,8V 16-bit), page size: 2048, OOB size: 64
[    0.931671] Creating 5 MTD partitions on "omap2-nand.0":
[    0.937286] 0x000000000000-0x000000080000 : "xloader"
[    0.944854] 0x000000080000-0x000000240000 : "uboot"
[    0.953247] 0x000000240000-0x000000280000 : "uboot environment"
[    0.961242] 0x000000280000-0x000000680000 : "linux"
[    0.971893] 0x000000680000-0x000020000000 : "rootfs"
[    1.480957] spi spi1.0: no RX DMA engine channel for McSPI
[    1.486816] omap2_mcspi omap2_mcspi.1: can't setup spi1.0, status -11
[    1.493621] omap2_mcspi omap2_mcspi.1: can't create new device for
ads7846
[    1.500885] spi spi1.1: no RX DMA engine channel for McSPI
[    1.506683] omap2_mcspi omap2_mcspi.1: can't setup spi1.1, status -11
[    1.513488] omap2_mcspi omap2_mcspi.1: can't create new device for
lgphilips_lb035q02_panel-spi
[    1.524169] smsc911x: Driver version 2008-10-21
[    1.529541] smsc911x: Driver version 2008-10-21
[    1.534912] usbcore: registered new interface driver asix
[    1.540740] usbcore: registered new interface driver cdc_ether
[    1.547027] usbcore: registered new interface driver cdc_ncm
[    1.553405] ehci_hcd: USB 2.0 'Enhanced' Host Controller (EHCI) Driver
[    1.560699] ehci-omap ehci-omap.0: OMAP-EHCI Host Controller
[    1.567108] ehci-omap ehci-omap.0: new USB bus registered, assigned
bus number 1
[    2.610961] ehci-omap ehci-omap.0: irq 77, io mem 0x48064800
[    2.626556] ehci-omap ehci-omap.0: USB 2.0 started, EHCI 1.00
[    2.632873] usb usb1: New USB device found, idVendor=1d6b, idProduct=0002
[    2.640045] usb usb1: New USB device strings: Mfr=3, Product=2,
SerialNumber=1
[    2.647674] usb usb1: Product: OMAP-EHCI Host Controller
[    2.653289] usb usb1: Manufacturer: Linux 3.6.0 ehci_hcd
[    2.658905] usb usb1: SerialNumber: ehci-omap.0
[    2.664825] hub 1-0:1.0: USB hub found
[    2.668853] hub 1-0:1.0: 3 ports detected
[    2.673736] Initializing USB Mass Storage driver...
[    2.679046] usbcore: registered new interface driver usb-storage
[    2.685394] USB Mass Storage support registered.
[    2.690277] musb-hdrc: version 6.0, ?dma?, otg (peripheral+host)
[    2.698730] musb-hdrc musb-hdrc: USB OTG mode controller at fa0ab000
using DMA, IRQ 92
[    2.707855] mousedev: PS/2 mouse device common for all mice
[    2.714538] twl_rtc twl_rtc: Power up reset detected.
[    2.720001] twl_rtc twl_rtc: Enabling TWL-RTC
[    2.727447] twl_rtc twl_rtc: rtc core: registered twl_rtc as rtc0
[    2.734832] i2c /dev entries driver
[    2.740722] cpuidle: using governor ladder
[    2.745086] cpuidle: using governor menu
[    2.749877] omap_hsmmc omap_hsmmc.0: Failed to get debounce clk
[    2.756164] omap_hsmmc omap_hsmmc.0: unable to obtain RX DMA engine
channel 62
[    2.764251] omap_hsmmc omap_hsmmc.1: Failed to get debounce clk
[    2.770751] omap_hsmmc omap_hsmmc.1: unable to obtain RX DMA engine
channel 48
[    2.782379] usbcore: registered new interface driver usbhid
[    2.788299] usbhid: USB HID core driver
[    2.793029] usbcore: registered new interface driver snd-usb-audio
[    2.801239] overo SoC init
[    2.804321] soc-audio soc-audio: ASoC machine overo should use
snd_soc_register_card()
[    2.897094] twl4030-codec twl4030-codec: ASoC: Failed to create
Capture debugfs file
[    2.910278] soc-audio soc-audio:  twl4030-hifi <-> omap-mcbsp.2
mapping ok
[    2.923278] oprofile: hardware counters not available
[    2.928649] oprofile: using timer interrupt.
[    2.933807] TCP: cubic registered
[    2.937316] NET: Registered protocol family 17
[    2.942077] NET: Registered protocol family 15
[    2.946990] Key type dns_resolver registered
[    2.951721] VFP support v0.3: implementor 41 architecture 3 part 30
variant c rev 1
[    2.965393] ThumbEE CPU extension supported.
[    2.979522] clock: disabling unused clocks to save power
[    2.986572] registered taskstats version 1
[    2.992126] omapfb omapfb: no driver for display: lcd35
[    2.997802] usb 1-2: new high-speed USB device number 2 using ehci-omap
[    3.004821] omapfb omapfb: cannot parse default modes
[    3.034637] Console: switching to colour frame buffer device 90x35
[    3.113708] VDVI: incomplete constraints, leaving on
[    3.119140] VMMC1: incomplete constraints, leaving on
[    3.126190] twl_rtc twl_rtc: setting system clock to 2000-01-01
00:00:10 UTC (946684810)
[    3.135070] ALSA device list:
[    3.138214]   #0: overo
[    3.141632] Waiting for root device /dev/mmcblk0p2...


------------------------------------------------------------------------------
Master Visual Studio, SharePoint, SQL, ASP.NET, C# 2012, HTML5, CSS,
MVC, Windows 8 Apps, JavaScript and much more. Keep your skills current
with LearnDevNow - 3,200 step-by-step video tutorials by Microsoft
MVPs and experts. ON SALE this month only -- learn more at:
http://p.sf.net/sfu/learnmore_123012
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Frustrated findings from running Yocto/Denzil on Gumstix/Overo

Scott Ellis
Hi Paul,

I don't have any 3.6 configs for the Overo.

The McBSP driver changes in 3.6 broke all my drivers that depend on it. Until
I get that working again, I've been sticking with 3.2.

What's the bluetooth problem?

Those jumpnow images don't have any BT userland support so unless you modified
the recipes and added some packages...

I've been adding the following when I need bluetooth

bluez4
bluez-hcidump

and the following for dev images

bluez4-dev
bluez-hcidump-dev


I've been using images built from the layer below on a system that talks over
RFCOMM between a Gumstix and a couple of Arduinos and it's been working well.

https://github.com/Pansenti/meta-pansenti

There's a custom bt-attach recipe in there that adds a startup script I've
been using. It requires a manually configured /etc/bluetooth/rfcomm.conf.
You'll probably want to change or remove that.


The "waiting for root device" error on the Duovero was a hardware issue with
our COM.

I haven't seen that error on an Overo when the card was properly prepared.

Scott
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Frustrated findings from running Yocto/Denzil on Gumstix/Overo

qmay123
Thanks for posting those results!

I'd be curious to see the results compared to flashing to NAND, a class 10 card, and an industrial grade card. (not a request by any means, just a comment).
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Frustrated findings from running Yocto/Denzil on Gumstix/Overo

Scott Ellis
That wasn't a dedicated test. I just grabbed from the cards sitting on my desk
at the time.

It was only to illustrate the point that the SD card matters for boot times.

Once the system is running it's not as important since it usually all memory.
Unless of course your app relies on a lot of SD card activity.

At least that's what I've found.
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Frustrated findings from running Yocto/Denzil on Gumstix/Overo

paul@idruna.com
In reply to this post by Scott Ellis
Hi Scott,

> The McBSP driver changes in 3.6 broke all my drivers that depend on it.
> Until I get that working again, I've been sticking with 3.2.

A new version of Linux broke something? :o

> What's the bluetooth problem?

I`m using rfcomm sockets to connect to a bluetooth barcode scanner,
passcode 0000.  I`m using rfcomm because I gave up trying to get it
working in HID mode.  On my 3.2 kernel with bluez 4.89, sometimes it
works fine, but a lot of the time the connection is continually refused,
even if I use the rfcomm command to connect.  My image is based on open
embedded from 3 years ago, so I thought maybe my userland stuff was too
out of date.  I added bluez 4.101 to your recipe, and the rfcomm command
fails with an operation in progress message, or something similar.  It
seems to work fine on my Ubuntu 12 VM running kernel 3.2.026 and bluez
4.98 (same Asus BT dongle).  Maybe I should try to get bluez 4.98
working on my image.

> The "waiting for root device" error on the Duovero was a hardware issue with
> our COM.
>
> I haven't seen that error on an Overo when the card was properly prepared.

I`m guessing my defconfig needs something tweaking for the Overo, or I
downloaded the wrong defconfig from Steve Sakoman's git repository.

Thanks,
--
Paul Nolan, CEO Idruna Software Inc.

------------------------------------------------------------------------------
Master Visual Studio, SharePoint, SQL, ASP.NET, C# 2012, HTML5, CSS,
MVC, Windows 8 Apps, JavaScript and much more. Keep your skills current
with LearnDevNow - 3,200 step-by-step video tutorials by Microsoft
MVPs and experts. SALE $99.99 this month only -- learn more at:
http://p.sf.net/sfu/learnmore_122412
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Frustrated findings from running Yocto/Denzil on Gumstix/Overo

Arne Lie
In reply to this post by Arne Lie
Andreas,

> -----Original Message-----
> From: Andreas Müller [mailto:[hidden email]]
> Sent: 18. januar 2013 19:12
> To: General mailing list for gumstix users.
> Subject: Re: [Gumstix-users] Frustrated findings from running Yocto/Denzil
> on Gumstix/Overo
>
> To enable/disable
>
> systemctl disable foo.service
>
> should help. See man systemctl.
>
> The available service files should be found at /lib/systemd/system.
> See man systemd.unit.
>
> Hope it helps a bit
>
> Andreas
>
Andreas, thanks for the information! I finally concluded that my jumpnow-console-image from Monday did not create a healthy system: systemctl disable foo.service did not alter my bootup for any service. Also, the build was missing a lot of applications, like cpufreq-set, and a lot of linux binaries used busybox variants that are far from as powerful, e.g., ps command. Instead I downloaded Sakoman ready binaries from http://feeds.sakoman.com/feeds/yocto-1.3/images/ which is also newer (Linux 3.6, I guess it is from "danny" branch, not "denzil"). Now I got a healthy system where I can fine tune by disabling services with systemctl. Using a Sandisc SD card, I can now boot in about 12 seconds, but I find power consumption still a bit high, though (about 230 mA).

/Arne

------------------------------------------------------------------------------
Master Visual Studio, SharePoint, SQL, ASP.NET, C# 2012, HTML5, CSS,
MVC, Windows 8 Apps, JavaScript and much more. Keep your skills current
with LearnDevNow - 3,200 step-by-step video tutorials by Microsoft
MVPs and experts. ON SALE this month only -- learn more at:
http://p.sf.net/sfu/learnnow-d2d
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Frustrated findings from running Yocto/Denzil on Gumstix/Overo

Arne Lie
In reply to this post by Scott Ellis
Scott,

> -----Original Message-----
> From: jumpnowdev [mailto:[hidden email]]
> Sent: 19. januar 2013 13:11
> To: [hidden email]
> Subject: Re: [Gumstix-users] Frustrated findings from running Yocto/Denzil
> on Gumstix/Overo
>
> I've found a lot of boot time speed depends on the SD card.
>
> Here are measurements from an AirStorm running the jumpnow-console-
> image built yesterday. All times are the third boot from a complete power
> off.
>
> Each card was partitioned and formatted prior to copying the binaries.
>
> A SanDisk Ultra 8GB Class 6
>
> root@overo:~# systemd-analyze
> Startup finished in 3246ms (kernel) + 3824ms (userspace) = 7071ms
>
[snip]
Thanks a lot, these findings were in fact very helpful. I have one Kingston 4GB Class 4 and one Sandisk 8GB Class 2 card to play with, and the *Sandisk* is at least twice as fast...! Now I have about 3.5s + 7.5s bootime on the latter, using Sakoman's Yocto 1.3 console build with systemd.

Regards Arne

------------------------------------------------------------------------------
Master Visual Studio, SharePoint, SQL, ASP.NET, C# 2012, HTML5, CSS,
MVC, Windows 8 Apps, JavaScript and much more. Keep your skills current
with LearnDevNow - 3,200 step-by-step video tutorials by Microsoft
MVPs and experts. ON SALE this month only -- learn more at:
http://p.sf.net/sfu/learnnow-d2d
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Frustrated findings from running Yocto/Denzil on Gumstix/Overo

Trung Hoang
In reply to this post by Arne Lie
What is the Denzil and danny branch anyhow?

Which is is most stable?


On 25/01/2013, at 7:56 PM, Arne Lie <[hidden email]> wrote:

> Andreas,
>
>> -----Original Message-----
>> From: Andreas Müller [mailto:[hidden email]]
>> Sent: 18. januar 2013 19:12
>> To: General mailing list for gumstix users.
>> Subject: Re: [Gumstix-users] Frustrated findings from running Yocto/Denzil
>> on Gumstix/Overo
>>
>> To enable/disable
>>
>> systemctl disable foo.service
>>
>> should help. See man systemctl.
>>
>> The available service files should be found at /lib/systemd/system.
>> See man systemd.unit.
>>
>> Hope it helps a bit
>>
>> Andreas
>>
> Andreas, thanks for the information! I finally concluded that my jumpnow-console-image from Monday did not create a healthy system: systemctl disable foo.service did not alter my bootup for any service. Also, the build was missing a lot of applications, like cpufreq-set, and a lot of linux binaries used busybox variants that are far from as powerful, e.g., ps command. Instead I downloaded Sakoman ready binaries from http://feeds.sakoman.com/feeds/yocto-1.3/images/ which is also newer (Linux 3.6, I guess it is from "danny" branch, not "denzil"). Now I got a healthy system where I can fine tune by disabling services with systemctl. Using a Sandisc SD card, I can now boot in about 12 seconds, but I find power consumption still a bit high, though (about 230 mA).
>
> /Arne
>
> ------------------------------------------------------------------------------
> Master Visual Studio, SharePoint, SQL, ASP.NET, C# 2012, HTML5, CSS,
> MVC, Windows 8 Apps, JavaScript and much more. Keep your skills current
> with LearnDevNow - 3,200 step-by-step video tutorials by Microsoft
> MVPs and experts. ON SALE this month only -- learn more at:
> http://p.sf.net/sfu/learnnow-d2d
> _______________________________________________
> gumstix-users mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/gumstix-users


------------------------------------------------------------------------------
Master Visual Studio, SharePoint, SQL, ASP.NET, C# 2012, HTML5, CSS,
MVC, Windows 8 Apps, JavaScript and much more. Keep your skills current
with LearnDevNow - 3,200 step-by-step video tutorials by Microsoft
MVPs and experts. ON SALE this month only -- learn more at:
http://p.sf.net/sfu/learnnow-d2d
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Frustrated findings from running Yocto/Denzil on Gumstix/Overo

Scott Ellis
[danny] and [denzil] are two branches of the Yocto project.

[danny] is 1.3
https://www.yoctoproject.org/download/yocto-project-13-poky-80

[denzil] is 1.2.x
https://www.yoctoproject.org/download/yocto-project-121-release-notes-poky-701

[danny] was released a few months ago.

All the releases are claimed to be stable.

From my experience, mostly with [denzil], the projects are still subject
to random breakage, upstream repo moves usually.

The last time I tried [danny] was still broken with the TMPDIR placed on
a separate partition with a short absolute path (gcc-cross wouldn't build).

That was a few weeks ago and may be fixed. It's probably not an issue if
you don't try something dumb like me and attempt to specify another
TMPDIR ;-)




Loading...