I think we have a SD slot driver for you.

I believe the Dell Mini 9 uses the same SD slot that your Aspire One uses. We have recently started writing a driver that is partially working at this time. I think it would be beneficial to both camps if we were to work together on this issue. So, as a moderator of MyDellMini.com's forums, I am extending an invitation to come work with us on this driver. At the time of this writing (I say that because we seem to be making progress FAST, 30 hours ago we didn't have a driver at all), we have a driver that can read SD cards (not SDHC). But not write to them. It's also not terribly fast yet. What we need, people who can program, especially someone who has written a hardware driver before, especially someone who has done it in OS X! The driver we have is open source, we're all working together to improve it (well, those of us that can code). What we don't need yet, newb's.. If you can't install a kext and figure out how to load it, you're only going to be in the way, for the time being.
Anyways, the the link to the relevant thread..

 

Why do you have a sticky that says the SD slot does not work?

edit: after doing some googling, I'm guessing you thought you were in the Linux section. This post was made in the OS X section. We already know the slot works in Linux, in fact, we're using the Linux driver (it's built into the kernel in Linux) to help write the OS X driver.

 

Yeah sorry, didn't notice the post was in OSX.

 

we have a sticky saying sd doesnt work, becaus until youc ame along ( ) they didnt
and becaus i got pissed off with people asking over and over how to make them work..
I cant test, but not code, and I'm a mod here so you know lol...

what a great job though...
Rory

 

Yeah, I was wondering about the sticky when "sideways" came in here claiming the slot worked fine.. I was like WTF, why did we spend the past 2 months brainstorming this thing.. heheh.. Then it dawned on me he thought I was talking about Linux.. But that's cool.

 

This is great news! If we figure the card readers out, the only thing left really is sleep! Thanks for all the hard work, I just got my AAO and I'm waiting for my DW 1390 wireless to come in before I take the plunge! :ugeek:

 

I can't code so I can't be of any help I guess, but if I can with doing something else other than coding, I'll be at your service
Maybe you can give us a heads up about how sleep is doing with the wind? I read somewhere that Chameleon is planning a new release with fixes for both hyperthreading and sleep, but I can't get to find any 'official' info about that.

 

Unaclocker, thanks for your post! How exciting! I'm looking forward to having the SD slots fully working!

 

So huh what? does this work for mac? im lost...

 

Nothing to see here, move along.

 

It works - sometimes...

The kext does not seem to be loaded automatically at startup. So there has to be an SD in the slot at startup. Then - mostly - the SD card appears. Loading the kext in Terminal via kextload leads to a "loaded successfully", but afterwards the card is not detected anyway.

I also checked the system using the PCI tools mentioned in the Dell thread. It seems that my JMicron controller sometimes is not recognized. In this case it is listed as invalid vendor id (fffff). If it is recognized properly it will show up as four different PCI devices (for the different memory card systems).

Think this will lead to a solution - let's have a steady look at the project.

I was already warming up my solder iron to insert a new USB reader. Now I will use it for replacing the internal mic with a Logitech USB microphone. That one works with iChat

JL

 

SD Card driver Beta 1 release

Hi. I'm one of the two developers working on the SD Card driver for the Dell Mini 9 which, apparently, has the same host controller as the Aspire One. I have a SourceForge project all set up now at: http://osxsdhci.sourceforge.net From the Downloads link, you can download the latest version of the driver which does reads and writes at around 1.5MB/sec. Not phenomenal, but not too shabby either.

I have a few more improvements to make that will enable 4-bit bus and high speed modes which help transfer rates (I have them working, but not the tests of the card that are needed to make sure the card supports those modes).

It only supports a single card slot at the moment. Unfortunately, I don't have any way of testing a multi-card slot machine as I have a Mini 9, not an Aspire One. I do know that there will have to be some major changes to the driver to support multi-slot, though.

Lastly, I don't monitor this forum too much so you can feel free to post bugs/feature requests/discussions etc to the relevant places on SorceForge. Lastly, if you want to see the code (or contribute to the code ) it's under Subversion control at https://osxsdhci.svn.sourceforge.net/svnroot/osxsdhci/

 

If your PCI vendor id occasionally shows up as 0xffff then you have a hardware problem. The operating system (in this case OSX) matches drivers to devices by vendor ID so there's no way that the driver will load correctly when that happens. In any case, check out the new driver I just posted and see if that works better. On my Mini 9, I can get card insertion/removal events and the card will show up as soon as it's inserted.

 

I just tried the beta1 driver.

Booted with nothing in either slot.

Here's what I get in the console:

Code:

PCI configuration changed (bridge=0 device=4 yenta=0)
SDHCI: BaseClock = 50MHz
SDHCI: SD Clock = 390 KHz
powerSD: 0x14832b2

That's what I get when I insert the card in either slot. And that's all I get - Disk Utility sees nothing.

No messages when I pull the card out. Subsequent inserts give no messages to console.

UPDATE:

If I boot with a card in the *either* slot, it does appear in finder. Read and write work.

Ah - and now for some reason the card is recognized on insert - but insert recognition only works once until the next reboot.

UPDATE2: now it seems to be random on whether or not it will recognize the card on hot insert.

This is excellent progress - thanks!

 

Hmm. Well, this should be "fun" to debug Can you do an lspci -nn and either post the results here or on the discussion forum on SourceForge? I don't have an Aspire One, I have a Dell Mini. I'm frankly (pleasantly ) surprised that you got it to work at all with both slots. I want to see if it shows up as one PCI device or two. If it shows up as two, the driver should work but I'll need to move a lock from a global variable to part of a class to increase performance if you talk to both slots at once.

 

I have a dual boot with Vista and before when I forgot to remove the card it would brake and I needed to reformat it. Now with this beta1 driver i can actually read and write. Progress is fast and it looks like we can have full SD support soon.

 

Ok, this one is strange, and I can't seem to figure out what's going on. On a clean boot with no cards inserted, I started out with no JMicron devices. On first insert, I get three new devices: vendor 197b, subsys 2382, 2383, 2384 - all listed as System peripheral.

Then if I remove and replug the card, I get a 4th device: subsys 2381 (SD Host Controller).

This is all working with a single slot. If I use the other slot, duplicated device entries show up for subsys IDs 2382, 2383 and 2384

Also of note: when I boot with both slots filled (the only reliable way to get the system to see the cards and mount them), the system mounts each card as its own drive. Writes are successful to each individual card as would be expected on a working system. So it looks like the system somehow sees them as separate devices. In this case I have two device entries for each of 2381, 2382, 2383 and 2384.

 

Yeah, so the 2382, 2383, and 2384 are MMC, Memory Stick, and xD respectively. 2381 is the SD Flash reader. It's good that you get two sets - that means that the simplest fix to my driver will fix the driver side. But..... This post: http://guide.ubuntuforums.org/showthread.php?t=974525 as well as yours leads me to believe that the machine doesn't even power up the host controller unless the card is in the slot. That's going to be a PITA because without the PCI device, the OS isn't going to try to load the driver. For those who've tried, do they work reliably under Windows? They apparently don't under Linux. I wonder if there's a BIOS setting for that? I also wonder if Microcenter would notice if I "accidentally" wound up in the BIOS setup...

 

The BIOS setup is very limited. (InsydeH20 3.5)

You have boot options (quick boot, network boot) and boot device priorities. And security password setup... nothing else.

 

I have a dual boot with XP. XP recognizes the card after insertion. Under OSX I experience the same strange behaviour as the_case - you have to insert the card while booting. After that it will recognise any card you put in the slot.