OK had the 770 for a month, like it enough, but cant get to grips with some of it..

Booting from MMC no problem but how do i install fanoush MMC fast write speed direct from the 770?

I just cant figure it out.

Tried updating the kernel from a ubuntu live cd which resulted in an unbootable tablet unless i removed the MMC.

I would like to try the speed boost but iam struggling with it.

Could i have clear instructions on just flashing the kernel direct from 770 please?

I have a 1gb transend MMC.

Regards, Steph.

Ok so ive tried flashing from Ububtu with no success after every attempt
i get the following.

USB device found found at bus 008, device address 007
Found device SU-18, hardware revision 1802
NOLO version 0.9.14
Version of 'sw-release': SU-18_2006SE_3.2006.49-2_PR_MR0
Sending kernel image (1245 kB)...
100% (1245 of 1245 kB, avg. 883 kB/s)
Flashing kernel... done.
Sending request 0x83 failed: Cannot send after transport endpoint shutdown

After this the 770 will get stuck at the nokia logo unless i remove the mmc card ?

Where did i go wrong?

This means it does not work with your card. What happens when you remove the card, boot system and insert the card later? Could you open osso-xterm insert card, wait few seconds and run 'dmesg' command and post output here or PM it to me? I think I can't help you but I'm still interested in mmc related error messages from kernel with such card.

Perhaps I could make also kernel with MMC bus speed maximum set only to 24MHz not 48. That could help with some cards and still provide at least 2x speedup.

Thanks for the reply, my card is actually a Nokia branded 1gb item that came bundled with Navicore.

I can boot normaly with the mmc removed when i insert it i get about 10secs then either a lockup or a system crash.

I did manage to execute dmesg but system went down before i could get any of it. i did see error "internode " powering down.

Thanks very much for helping but it seems the nokia cards a no go.


Perhaps if iam very quick i may be able to save the log after a few attempts. Will try some more to retrive the log.

dmesg (alot removed to large to post)

[&space;&space;&space;20.735839]&space;omapdsp:&space;mapping&space;in&space;ARM&s pace;MMU,&space;v=0xe0400000,&space;p=0x12e00000,& space;sz=0x100000
[&space;&space;&space;20.736053]&space;omapdsp:&space;mapping&space;in&space;ARM&s pace;MMU,&space;v=0xe0500000,&space;p=0x12930000,& space;sz=0x10000
[&space;&space;&space;20.736145]&space;omapdsp:&space;mapping&space;in&space;ARM&s pace;MMU,&space;v=0xe0510000,&space;p=0x12920000,& space;sz=0x10000
[&space;&space;&space;20.736236]&space;omapdsp:&space;mapping&space;in&space;ARM&s pace;MMU,&space;v=0xe0520000,&space;p=0x12910000,& space;sz=0x10000
[&space;&space;&space;20.736297]&space;omapdsp:&space;mapping&space;in&space;ARM&s pace;MMU,&space;v=0xe0530000,&space;p=0x12900000,& space;sz=0x10000
[&space;&space;&space;20.736389]&space;omapdsp:&space;mapping&space;in&space;ARM&s pace;MMU,&space;v=0xe0540000,&space;p=0x12df0000,& space;sz=0x10000
[&space;&space;&space;20.736480]&space;omapdsp:&space;mapping&space;in&space;ARM&s pace;MMU,&space;v=0xe0550000,&space;p=0x12de0000,& space;sz=0x10000
[&space;&space;&space;20.736572]&space;omapdsp:&space;mapping&space;in&space;ARM&s pace;MMU,&space;v=0xe0560000,&space;p=0x12940000,& space;sz=0x10000
[&space;&space;&space;20.736663]&space;omapdsp:&space;mapping&space;in&space;ARM&s pace;MMU,&space;v=0xe0570000,&space;p=0x12950000,& space;sz=0x10000
[&space;&space;&space;21.153594]&space;omapdsp:&space;IPBUF&space;configuration
[&space;&space;&space;21.153625]&space;&space;&space;&space;&space;&space;&space;& space;&space;&space;&space;&space;512&space;words& space;*&space;16&space;lines&space;at&space;0xe020 0000.
[&space;&space;&space;21.153869]&space;omapdsp:&space;found&space;4&space;task( s)
[&space;&space;&space;21.154144]&space;omapdsp:&space;task&space;0:&space;name&spa ce;pcm0
[&space;&space;&space;21.154724]&space;omapdsp:&space;taskdev&space;pcm0&space;ena bled.
[&space;&space;&space;21.163391]&space;omapdsp:&space;task&space;1:&space;name&spa ce;pcm1
[&space;&space;&space;21.163940]&space;omapdsp:&space;taskdev&space;pcm1&space;ena bled.
[&space;&space;&space;21.172912]&space;omapdsp:&space;task&space;2:&space;name&spa ce;avsync
[&space;&space;&space;21.173431]&space;omapdsp:&space;taskdev&space;avsync&space;e nabled.
[&space;&space;&space;21.182434]&space;omapdsp:&space;task&space;3:&space;name&spa ce;audiopp
[&space;&space;&space;21.183044]&space;omapdsp:&space;taskdev&space;audiopp&space; enabled.
[&space;&space;&space;21.723022]&space;hci_cmd_task:&space;hci0&space;command&spac e;tx&space;timeout
[&space;&space;&space;21.723144]&space;brf6150:&space;Frame&space;for&space;non-running&space;device
[&space;&space;&space;22.298889]&space;aic23&space;powering&space;up
[&space;&space;&space;22.303619]&space;aic23_init_power()&space;done
[&space;&space;&space;22.695648]&space;omapdsp:&space;mmap&space;info:&space;vmadr &space;=&space;40016000,&space;padr&space;=&space; 12e30000,&space;len&space;=&space;2000
[&space;&space;&space;25.075836]&space;aic23&space;powering&space;down
[&space;&space;&space;33.664184]&space;aic23&space;powering&space;up
[&space;&space;&space;33.668579]&space;aic23_init_power()&space;done
[&space;&space;&space;40.223480]&space;aic23&space;powering&space;down
[&space;&space;&space;45.957397]&space;aic23&space;powering&space;up
[&space;&space;&space;45.962310]&space;aic23_init_power()&space;done
[&space;&space;&space;45.973815]&space;omapdsp:&space;mmap&space;info:&space;vmadr &space;=&space;40016000,&space;padr&space;=&space; 12e10000,&space;len&space;=&space;2000
[&space;&space;&space;48.426635]&space;aic23&space;powering&space;down
[&space;&space;&space;63.204498]&space;JFFS2&space;notice:&space;(327)&space;check _node_data:&space;wrong&space;data&space;CRC&space ;in&space;data&space;node&space;at&space;0x039a280 0:&space;read&space;0xbfdf3ca,&space;calculated&sp ace;0xf3b9dd99.
[&space;&space;&space;67.995361]&space;JFFS2&space;notice:&space;(327)&space;check _node_data:&space;wrong&space;data&space;CRC&space ;in&space;data&space;node&space;at&space;0x04d2280 0:&space;read&space;0x8179469b,&space;calculated&s pace;0xdaa450c4.
[&space;&space;&space;68.447296]&space;JFFS2&space;notice:&space;(327)&space;check _node_data:&space;wrong&space;data&space;CRC&space ;in&space;data&space;node&space;at&space;0x04d2d00 0:&space;read&space;0x1c8d7b31,&space;calculated&s pace;0x152ff38d.
[&space;&space;&space;95.238433]&space;hwa742:&space;setting&space;update&space;mo de&space;to&space;disabled
[&space;&space;&space;95.238494]&space;omapfb_notifier_cb():&space;event&space;=&s pace;DISABLED
[&space;&space;103.703765]&space;hwa742:&space;setting&space;update&space;mo de&space;to&space;manual
[&space;&space;103.703826]&space;omapfb_notifier_cb():&space;event&space;=&s pace;READY
[&space;&space;103.783966]&space;aic23&space;powering&space;up
[&space;&space;103.788513]&space;aic23_init_power()&space;done
[&space;&space;109.224029]&space;aic23&space;powering&space;down
[&space;&space;110.495422]&space;aic23&space;powering&space;up
[&space;&space;110.500061]&space;aic23_init_power()&space;done
[&space;&space;113.587341]&space;aic23&space;powering&space;down
[&space;&space;115.330780]&space;aic23&space;powering&space;up
[&space;&space;115.335021]&space;aic23_init_power()&space;done
[&space;&space;125.700134]&space;aic23&space;powering&space;down
[&space;&space;125.925628]&space;aic23&space;powering&space;up
[&space;&space;125.929870]&space;aic23_init_power()&space;done
[&space;&space;129.176788]&space;aic23&space;powering&space;down
[&space;&space;129.728210]&space;aic23&space;powering&space;up
[&space;&space;129.732482]&space;aic23_init_power()&space;done
[&space;&space;132.099884]&space;aic23&space;powering&space;down
[&space;&space;132.540191]&space;aic23&space;powering&space;up
[&space;&space;132.545318]&space;aic23_init_power()&space;done
[&space;&space;145.912750]&space;aic23&space;powering&space;down
[&space;&space;149.824523]&space;aic23&space;powering&space;up
[&space;&space;149.828796]&space;aic23_init_power()&space;done
[&space;&space;152.196807]&space;aic23&space;powering&space;down
[&space;&space;163.683227]&space;aic23&space;powering&space;up
[&space;&space;163.687500]&space;aic23_init_power()&space;done
[&space;&space;168.262817]&space;aic23&space;powering&space;down
[&space;&space;169.784027]&space;aic23&space;powering&space;up
[&space;&space;169.788238]&space;aic23_init_power()&space;done
[&space;&space;172.153289]&space;aic23&space;powering&space;down
[&space;&space;172.499664]&space;aic23&space;powering&space;up
[&space;&space;172.503875]&space;aic23_init_power()&space;done
[&space;&space;175.974060]&space;aic23&space;powering&space;down
[&space;&space;179.785400]&space;aic23&space;powering&space;up
[&space;&space;179.789642]&space;aic23_init_power()&space;done
[&space;&space;192.060424]&space;aic23&space;powering&space;down
[&space;&space;200.916412]&space;mmci-omap&space;mmci-omap.2:&space;cover&space;is&space;now&space;open
[&space;&space;209.350189]&space;aic23&space;powering&space;up
[&space;&space;209.354400]&space;aic23_init_power()&space;done
[&space;&space;216.559570]&space;aic23&space;powering&space;down
[&space;&space;221.326599]&space;aic23&space;powering&space;up
[&space;&space;221.330810]&space;aic23_init_power()&space;done
[&space;&space;225.591003]&space;aic23&space;powering&space;down
[&space;&space;230.226745]&space;aic23&space;powering&space;up
[&space;&space;230.230926]&space;aic23_init_power()&space;done
[&space;&space;232.590820]&space;aic23&space;powering&space;down
[&space;&space;238.678924]&space;aic23&space;powering&space;up
[&space;&space;238.683135]&space;aic23_init_power()&space;done
/home/user&space;#

Thanks for your effort. Sadly there is nothing there related to MMC except that you opened card door near the end of the log. Also the html quoting makes it quite hard to read :-) Maybe it could be better done when logged in from PC to tablet over ssh as root and running 'cat /proc/kmsg'. This keeps printing the log as it grows (until device hangs or crashes).

But don't bother if it sounds complex. It is possible there won't be anything interesting before the device crashes anyway.

As for the speed hack - it really doesn't work with every card. After all there must be some real reason why Nokia bothered to limit the MMC bus frequency to such low value (12Mhz). Perhaps the card must be more tolerant to signal noise or bad timing produced by 770 when running at higher speeds. My 2GB Kingston MMCmobile is rock solid at 48MHz. I am using it for more than year at such speed with whole system booted from it and swap file enabled and so far had no read/write error or crash that was provably caused by the card. But you are not first who reports such problems so there is not much you can do except trying different card.

Is it possible to add some kind 'white' list to MMC driver and select speed depending on memory card type/vendor? I also used two Kingston MMCMobile cards, 1GB and 2GB with high speed patch and everything is fine except that I also used them to boot system (that means it is never inserted/removed at runtime, so this use case is not covered). So probably Kingston is the first candidate to get added into 'white' list. Of course 'white' list is only needed if the goal is to have a kernel which works reliable with all types of memory cards and gets the best performance for those cards which are known to work fine at high speed with Nokia 770.

Well, it would be hard to maintain such list. In some older firmwares I had kernel with maximum being selectable via file in /sys/devices/platform/... which is IMO better way to solve it but lost the code and so far was lazy to write it again. But there was problem with it that it needed reinsertion of the card, which is hard to do when you boot from it :-) Current linux MMC code determines speed at card insertion and it is not prepared to change it later so it would need more effort than just changing one variable in MMC slot structure. But still it should be possible, I'll try to do it when doing something with mmc code again. I have some work in progress there (mainly SDHC compatibility for hacked SD cards) that waits for me to actually open the device and hack such card in and verify how it works. Part of the changes is upgrade to newer MMC code used in 2.6.18 that could maybe solve some issues with hanging whole kernel while card is detected (i.e. the issue OP has) and better error recovery in multiblock writes. Unfortunately there is lot of interesting stuff to do and so little of spare time.

I would very much appreciate this (especially one compatible with the new HE version).

I bought a N770 for my son to use with his online school and the 770 is just a bit too slow to be usable. His tablet has the original 64MB card used for swap which should be ok at 24Mhz.

(My 770 with 2GB card loves the 52mhz kernel )

-Josh