Originally Posted by jeffcavicchi

Stuff like this makes me feel as though I should have buckled down and studied engineering in college rather than Comparative Literature...

Nice work, Gary - great to have such a knowledgeable developer working on our device!

Did you really have to quote the whole thing? Now I can't even edit the post to get rid of my ravings.

Originally Posted by garyd9

Did you really have to quote the whole thing? Now I can't even edit the post to get rid of my ravings.

Honestly I almost did the same thing last night since I'm going to school also for stuff like this and felt the same way

As to seeing left-shifts of zeros - It's a common thing in hardware so that you ha ve source code which explicitly says "I am setting this flag to 0". (The idea being that sometimes it might be nonzero, and you just have to change the 0 to something else, rather than add in the whole line.)

e.g. it's easier to have a bunch of flag-setting stuff, and pop in 0s where it's 0, than to not have the flag-setting stuff at all. (it makes hardware people go AAAARRRGH NOW I HAVE TO CALCULATE WHAT'S GOING INTO THAT REGISTER!!!!!)

As to the SMB136 taking over - not surprised, if you recall the first post of my I777 kernel thread, it's the same deal in the GS2, except we have a MAX8922. For whatever reason, the MAX8997 has charger circuitry, it is in the source, but it is never used!

Makes sense to disable all USB when on high-current charging, considering that unless you override the detection modes, anything that triggers high-current charging will not be able to support USB data transfer.

Originally Posted by garyd9

Did you really have to quote the whole thing? Now I can't even edit the post to get rid of my ravings.

Fixed it for ya!

All I left were the non-technical ravings...

You know.. its sad when you start having bad dreams about kernel source.

When my wife woke me up this morning (the first time - at 6:30am), I apparently told her I was dreaming that the charge on the kids was depleted, and I couldn't get them to accept a new charge. My guess was that I pushed too much current into them and it may have damaged their cells. Best of all, I apparently asked her if she could trickle charge them to see if that would get things working again.

(I don't remember any of this.)

I need to figure out how to deplete the charge on my kids - might mean I could get more than 5 hours of sleep a night for once!]

As far as I can tell, children under the age of 8 have a nearly unlimited battery life!

I have a call in to Summit Microelectronics (the people who make the smb136.) Perhaps I can get them to send me a full datasheet on the part (along with programming information.) It never hurts to ask. They might even send me a free sample.

Originally Posted by chip.hughes

Does this kernel have ext2 support?

Please don't post the same question in multiple subforums. From the "development" point of view, I'd suggest examining the output from the following command (run on your tablet): cat /proc/filesystems

Originally Posted by garyd9

I have a call in to Summit Microelectronics (the people who make the smb136.) Perhaps I can get them to send me a full datasheet on the part (along with programming information.) It never hurts to ask. They might even send me a free sample.

That would be useful - also try looking for other kernels that support the SMB136. That's how I found more info on the MAX8998's charge control registers.

Maxim is usually good about datasheets, but the 8997/8998 are exceptions. We're lucky in that the datasheet for the 8922 in the I9100/I777 is available. (even if it provided us information that was disappointing - at least you know not to waste time on trying to change the current of a chip that only supports three current levels!)

Originally Posted by Entropy512

That would be useful - also try looking for other kernels that support the SMB136. That's how I found more info on the MAX8998's charge control registers.

Yeah - did that last night. The original galaxy tab uses the SMB136 as well, and it appears that anyone messing with it didn't get very far.

Worst case scenerio, I can easily set up a parameter in the sysfs for forcing HC mode - but that will disable USB data.