Smart Q7 Tablet

[jshzh]

One guy on the Chinese forum said he could read pdf files larger than 200M under Wince without any problem. It takes some time to open the file though. But under linux 1**M files turn the screen black right away.

[Suwat]

Hi rogue,
I finally did the "clock=pit" as you suggested a while back. It seemed to work so far. I'll keep testing (fingers crossed). I followed the discussion about multiboot on q7 on other boards. One of the prerequisit for multiboot was to make a file called "menu.lst" under "/boot" directory. I thought it's the same file as you suggested. So, I decided to make a new file name "menu.lst" under "/boot". Then, I added only one line "clock=pit", saved the file and rebooted. The q7's clock seemed to run correctly now.

Thank you again for you suggestion.

[heron61]

Originally Posted by jshzh

One guy on the Chinese forum said he could read pdf files larger than 200M under Wince without any problem. It takes some time to open the file though. But under linux 1**M files turn the screen black right away.

What PDF reader works for WinCE. I've heard that the Adobe Reader won't install.

[del1357]

Guys, what's your Q7 battery life like? Mine seems to last only a few (4-6) hours when using comix/evince now - maybe I'm misremembering, but I seem to recall it lasting substantially longer than that when I first got it.

[rogue_ronin]

Originally Posted by Suwat

Hi rogue,
I finally did the "clock=pit" as you suggested a while back. It seemed to work so far. I'll keep testing (fingers crossed). I followed the discussion about multiboot on q7 on other boards. One of the prerequisit for multiboot was to make a file called "menu.lst" under "/boot" directory. I thought it's the same file as you suggested. So, I decided to make a new file name "menu.lst" under "/boot". Then, I added only one line "clock=pit", saved the file and rebooted. The q7's clock seemed to run correctly now.

Thank you again for you suggestion.

Wow, if that works, you're a genius. I wouldn't have thought to do that, because I would expect a complete proper grub install to be necessary.

Let us know if it keeps working -- I'll definitely do the same, should my Q7 ever be returned.

m a r

[Suwat]

Originally Posted by rogue_ronin

Wow, if that works, you're a genius. I wouldn't have thought to do that, ..................Let us know if it keeps working -- I'll definitely do the same, should my Q7 ever be returned.

m a r

There a very thin line between a genius and an idioit. I'm probably an idiot. It's not working! Last time, it might be my own wishful imagination to see that the clock was running ok. After a long test without interuption from suspension and auto-off mode, the clock's still running too fast. It's a little better if the device is on-off periodically.

I checked system log file to see the effect of "clock=pit" to the boot process. Nothing's at all in the log file, neither about "clock=pit" nor about menu.lst. There's only one line about the realtime clock:

Aug 30 10:20:28 Suwat kernel: s3c2410-rtc s3c2410-rtc: setting system clock to 2009-08-30 03:20:13 UTC (1251602413)

Any input is much appreciated.

[Suwat]

Hi rogue, can you explain this? I shutdowned my q7 after the last post at around 14:40 (2:40pm) local time. The clock on q7 was displaying 15:10 (3:10pm); about half an hour too fast. Since the q7 was pluged into main AC, it was on automatically. After system boot, I checked the clock again. It said 14:43 (2:43pm) which was the correct time. So, I checked the log file. This is the first line in the syslog:

Aug 30 14:42:05 Suwat kernel: Linux version 2.6.24.7 (root@yousheng) (gcc version 4.3.2 (Sourcery G++ Lite 2008q3-72) ) #1291 PREEMPT Tue Jun 30 10:25:48 CST 2009

Look at the entry time. It's the correct time of day. So, I got the correct time when I just booted-up the system. And after a while, the clock would be too fast. Then if I re-boot the system, I will get the correct time again. It's illogical to me. How could q7 know the correct time after its clock ran wild?

[rogue_ronin]

I'm not an expert at Linux, just a long-time user. But I believe that there are two "clocks" in a Linux system, one is hardware and one is software. I may have this mixed up, or over-simplified.

So, my guess is that the software clock is running wild, but the hardware clock is good. That may be what the "clock=pit" is all about, telling the kernel to refer to the hardware clock instead of doing its own timing. When you boot, the kernel gets the time from the hardware clock.

Wish I knew more about how it all works, but it's a massively complex system -- and Smart is not helping by using an outdated Ubuntu release, one that is not even official. They could probably switch to Jaunty pretty easily, relative to us, and at least we would be able to use the Ubuntu resources more easily.

$0.01 this time,

m a r

[Houndx]

Originally Posted by heron61

What PDF reader works for WinCE. I've heard that the Adobe Reader won't install.

The folks on the Chinese forums all seem to be using v1 of Foxit Reader. Maybe 1.1, or something link that, but not the latest 2.x versions.

edit: or it looks like maybe Pocketxpdf

[Suwat]

I spent several hours doing some experimentation on the clock. As it turned out, "clock=pit" had nothing to do with the inaccuracy of the clock. I did some reading and found out that "clock=pit" was i386 systems specific. For q7, it will run correctly after bootstrap or clock-server-sync (probably after clock-setting too) as long as the device is not suspended. After suspend, it will start running faster and faster. I still could not find the causes.

The above event that the clock reversed back to the correct time occured only that once. I tried several ways to duplicate it without success. It's probably a fluke. Maybe it's like you said, there's a hardware rtc-clock and the software clock applet that displays time. Maybe there's intervals that the software clock goes back and update the rtc-clock. Hence, the reversal of hardware time because I did the reboot too early.

I will keep investigate anyway. Wish me luck.

*** Correction to the above bold sentence. 25 hours later, here's the correct info:

For q7, it will run correctly after bootstrap for as long as the device is not suspended. Details in later post.

[celtica96]

You two are doing some great work.

The following package may help you learn more detail about the inner workings of the Q7.

It's called "hardinfo" (not the name I would have chosen).

In any case, it's pretty handy. For example, hardinfo provides data on the OS, kernel modules, memory, benchmarks, etc.

One section that may have helped you is called "Boots" but nothing is there.

Hardinfo is discussed here: http://wiki.hardinfo.org/HomePage.

You should be able to use "sudo apt-get install hardinfo" to install from the repository.handhelds.org site. It will appear under Preferences as "System Profiler..." when you pull down the Panel Menu.

As always, install at your own risk....

***Warning***
Your system will require a reset should you check the Battery section!

[rogue_ronin]

Originally Posted by Suwat

I spent several hours doing some experimentation on the clock. As it turned out, "clock=pit" had nothing to do with the inaccuracy of the clock. I did some reading and found out that "clock=pit" was i386 systems specific. For q7, it will run correctly after bootstrap or clock-server-sync (probably after clock-setting too) as long as the device is not suspended. After suspend, it will start running faster and faster. I still could not find the causes.

The above event that the clock reversed back to the correct time occured only that once. I tried several ways to duplicate it without success. It's probably a fluke. Maybe it's like you said, there's a hardware rtc-clock and the software clock applet that displays time. Maybe there's intervals that the software clock goes back and update the rtc-clock. Hence, the reversal of hardware time because I did the reboot too early.

I will keep investigate anyway. Wish me luck.

Luck!

I am certain that the Linux kernel has its own methods of keeping time and synchronization -- at least 3 or 4. That "clock=pit" is 386-specific also does not surprise me.

The issue is ARM, and the kernel options when it was compiled. When my Q7 comes back (if it ever does) I will try to mimic the current desktop on Jaunty by installing Mer, then adding LXDE and removing Hildon. This should give me a clock to compare to the original. If the newer kernel has proper compilation, I expect that the clock has been taken care of. Jaunty is the first Ubuntu release that has official support for ARM, and I expect that at least minimal bug-checking has been done.

m a r

[rogue_ronin]

Originally Posted by celtica96

You two are doing some great work.

The following package may help you learn more detail about the inner workings of the Q7.

It's called "hardinfo" (not the name I would have chosen).

In any case, it's pretty handy. For example, hardinfo provides data on the OS, kernel modules, memory, benchmarks, etc.

One section that may have helped you is called "Boots" but nothing is there.

Hardinfo is discussed here: http://wiki.hardinfo.org/HomePage.

You should be able to use "sudo apt-get install hardinfo" to install from the repository.handhelds.org site. It will appear under Preferences as "System Profiler..." when you pull down the Panel Menu.

As always, install at your own risk....

***Warning***
Your system will require a reset should you check the Battery section!

Hardinfo is pretty rockin'. I install it on all my machines.

I am thinking that bootup is probably just taking us directly into openbox from X, skipping a desktop manager. The kernel start may be a giant binary blob as the hardware doesn't change between devices, right?

Can't check, no Q7 on this island -- went back to China three days ago before I knew it was here.

Thanks for the kind words -- but I'm just a user, and a reasonable guesser.

m a r

[Suwat]

@celtica96, Thanks for the kind words.

@rogue and interested people,
I found some interesting links you guys might like to see:

1. Linux Q5/7 Realtime Clock Driver source code in C

2. SmartQ5(&7?) Boot Process from Mer people--only general description

As for me, I understood only a bit of them.

[Suwat]

Anyone knows for sure whether smartq7 comes with USB 2.0 hardware? When I bought mine I believed it came with 2 USB v2.0. Now, I'm not so sure. I checked syslog and there was no mention of EHCI driver during boot. There was only OHCI driver which was only USB 1.0. If the hardware supports USB2.0, is there a way to inject the correct driver into boot process?

Here's the spec on eletroworld web site:

Connectivity: 2.0 USB HOST x 2 (USB x 1 & miniUSB x1) supported to connect the high-speed EDGE, HSDPA network adapter or mobile phone

Did it mean USB version 2.0 or just 2 USB HOST ports? I'm not a native english speaker and might have misinterpreted the above.