Hello,

Here is the IOPCCardFamily file that I have worked on today. I complied the source files offered from Apple and also fixed some bugs. You do not have to edit the info.plist to match your machine. If your pccard shows "IOPCCardBridge::start failed", you can give it a try.

How to use?

• Backup your original IOPCCardFamily driver (and IOPCIFamily driver)
• Replace your IOPCCardFamily driver (and IOPCIFamily driver) with the attached one that contains all necessary modifications
• Reboot
• As root, type "dmesg | grep PC"; you should see something being found

Welcome to try it. Don't forget to backup the old files before trying it.

[IMPORTANT]
If you get the kernel panic, PLEASE check if you have replaced the following files to make your cardbus slot working before:

AppleACPIPlatform
AppleAPIC
IOACPIFamily

If you did, you probably need to find a way to restore those files first.
If you still get kernel panic, you can take the picture of it if possible. Thanks.

[KNOWN ISSUES]
It may cause kernel panic with AGPGart installed.....
If you have the error message "IOPCCardBridge::checkBridgeBusIDs invalid sub/cardbus/pci settings of 0x0" before trying the tiger file, it still may not work after trying the tiger file.

Chun-Nan

Kext File: Click to view attachment
Source Code: Click to view attachment

[UPDATE]
1/17/2008: rev1 and source code update
Please replace IOPCCardFamily and IOPCIFamily at the same time.
You DO NOT have to edit info.plist because it will use "ADP2,1" by default.
If possible, let me know the result. Thanks.

Click to view attachment
Click to view attachment
Click to view attachment

1/30/2008: rev2 and source code update
Please replace IOPCCardFamily and IOPCIFamily at the same time.
You DO NOT have to edit info.plist because it will use "ADP2,1" by default.
If you are using other revisions, please update to rev2.
If possible, let me know the result. Thanks.

Click to view attachment
Click to view attachment

2/7/2008: rev3 and source code update
Please replace IOPCCardFamily and IOPCIFamily at the same time.
You DO NOT have to edit info.plist because it will use "ADP2,1" by default.
If you are using other revisions, please update to rev3.
If possible, let me know the result. Thanks.

Click to view attachment
Click to view attachment

5/25/2008: beta3 installation package update
Please choose IOPCCardFamily and IOPCIFamily at the same time.
If possible, let me know the result. Thanks.

Click to view attachment
Click to view attachment

==========
{ "Cirrus PD6729", IS_CIRRUS|IS_PCI, ID(CIRRUS, 6729) }
{ "Cirrus PD6730", IS_CIRRUS|IS_PCI, PCI_VENDOR_ID_CIRRUS, -1 }
{ "Cirrus PD6832", IS_CIRRUS|IS_CARDBUS, ID(CIRRUS, 6832) }
{ "ENE 1211", IS_TI|IS_CARDBUS, ID(ENE, 1211) } <= new added
{ "ENE 1225", IS_TI|IS_CARDBUS, ID(ENE, 1225) } <= new added
{ "ENE 1410", IS_TI|IS_CARDBUS, ID(ENE, 1410) } <= new added
{ "ENE 710/2/4", IS_TI|IS_CARDBUS, ID(ENE, 1411) } <= new added
{ "ENE 712/4", IS_TI|IS_CARDBUS, ID(ENE, 1412) } <= new added
{ "ENE 1420", IS_TI|IS_CARDBUS, ID(ENE, 1420) } <= new added
{ "ENE 720/2/4", IS_TI|IS_CARDBUS, ID(ENE, 1421) } <= new added
{ "ENE 722/4", IS_TI|IS_CARDBUS, ID(ENE, 1422) } <= new added
{ "O2Micro OZ6729", IS_O2MICRO|IS_PCI|IS_VG_PWR, ID(O2, 6729) }
{ "O2Micro OZ6730", IS_O2MICRO|IS_PCI|IS_VG_PWR, ID(O2, 6730) }
{ "O2Micro OZ6832/33", IS_O2MICRO|IS_CARDBUS, ID(O2, 6832) }
{ "O2Micro OZ6836/60", IS_O2MICRO|IS_CARDBUS, ID(O2, 6836) }
{ "O2Micro OZ6812", IS_O2MICRO|IS_CARDBUS, ID(O2, 6812) }
{ "O2Micro OZ6922", IS_O2MICRO|IS_CARDBUS, ID(O2, 6922) }
{ "O2Micro OZ6933/711E1", IS_O2MICRO|IS_CARDBUS, ID(O2, 6933) }
{ "O2Micro OZ6912/711E0", IS_O2MICRO|IS_CARDBUS, ID(O2, 6912) }
{ "O2Micro OZ711EC1/M1", IS_O2MICRO|IS_CARDBUS, ID(O2, 7112) }
{ "O2Micro OZ711EC1", IS_O2MICRO|IS_CARDBUS, ID(O2, 7113) }
{ "O2Micro OZ711M1/MC1", IS_O2MICRO|IS_CARDBUS, ID(O2, 7114) }
{ "O2Micro OZ711MP1/MS1", IS_O2MICRO|IS_CARDBUS, ID(O2, 7134) } <= new added
{ "O2Micro OZ711EZ1", IS_O2MICRO|IS_CARDBUS, ID(O2, 7135) } <= new added
{ "O2Micro OZ711SP1", IS_O2MICRO|IS_CARDBUS, ID(O2, 7136) } <= new added
{ "O2Micro OZ711E2", IS_O2MICRO|IS_CARDBUS, ID(O2, 71E2) }
{ "O2Micro OZ711M2", IS_O2MICRO|IS_CARDBUS, ID(O2, 7212) }
{ "O2Micro OZ711M3/MC3", IS_O2MICRO|IS_CARDBUS, ID(O2, 7223) }
{ "O2Micro OZ711MP3/MS3", IS_O2MICRO|IS_CARDBUS, ID(O2, 7233) } <= new added
{ "Ricoh RL5C465", IS_RICOH|IS_CARDBUS, ID(RICOH, RL5C465) }
{ "Ricoh RL5C466", IS_RICOH|IS_CARDBUS, ID(RICOH, RL5C466) }
{ "Ricoh RL5C475", IS_RICOH|IS_CARDBUS, ID(RICOH, RL5C475) }
{ "Ricoh RL5C476", IS_RICOH|IS_CARDBUS, ID(RICOH, RL5C476) }
{ "Ricoh RL5C477", IS_RICOH|IS_CARDBUS, ID(RICOH, RL5C477) } <= new added
{ "Ricoh RL5C478", IS_RICOH|IS_CARDBUS, ID(RICOH, RL5C478) }
{ "SMC 34C90", IS_CARDBUS, ID(SMC, 34C90) }
{ "TI 1130", IS_TI|IS_CARDBUS, ID(TI, 1130) }
{ "TI 1131", IS_TI|IS_CARDBUS, ID(TI, 1131) }
{ "TI 1031", IS_TI|IS_CARDBUS, ID(TI, 1031) }
{ "TI 1210", IS_TI|IS_CARDBUS, ID(TI, 1210) }
{ "TI 1211", IS_TI|IS_CARDBUS, ID(TI, 1211) }
{ "TI 1220", IS_TI|IS_CARDBUS, ID(TI, 1220) }
{ "TI 1221", IS_TI|IS_CARDBUS, ID(TI, 1221) }
{ "TI 1225", IS_TI|IS_CARDBUS, ID(TI, 1225) }
{ "TI 1250A", IS_TI|IS_CARDBUS, ID(TI, 1250A) }
{ "TI 1251A", IS_TI|IS_CARDBUS, ID(TI, 1251A) }
{ "TI 1251B", IS_TI|IS_CARDBUS, ID(TI, 1251B) }
{ "TI 1410", IS_TI|IS_CARDBUS, ID(TI, 1410) }
{ "TI 1420", IS_TI|IS_CARDBUS, ID(TI, 1420) }
{ "TI 1450", IS_TI|IS_CARDBUS, ID(TI, 1450) }
{ "TI 1451", IS_TI|IS_CARDBUS, ID(TI, 1451) }
{ "TI 1510", IS_TI|IS_CARDBUS, ID(TI, 1510) }
{ "TI 1520", IS_TI|IS_CARDBUS, ID(TI, 1520) }
{ "TI 1620", IS_TI|IS_CARDBUS, ID(TI, 1620) }
{ "TI 4410", IS_TI|IS_CARDBUS, ID(TI, 4410) }
{ "TI 4450", IS_TI|IS_CARDBUS, ID(TI, 4450) }
{ "TI 4451", IS_TI|IS_CARDBUS, ID(TI, 4451) }
{ "TI 4510", IS_TI|IS_CARDBUS, ID(TI, 4510) }
{ "TI 4520", IS_TI|IS_CARDBUS, ID(TI, 4520) }
{ "TI 7410", IS_TI|IS_CARDBUS, ID(TI, 7410) }
{ "TI 7510", IS_TI|IS_CARDBUS, ID(TI, 7510) }
{ "TI 7610", IS_TI|IS_CARDBUS, ID(TI, 7610) }
{ "TI 7420", IS_TI|IS_CARDBUS, ID(TI, 7510) } <= new added
{ "TI 7620", IS_TI|IS_CARDBUS, ID(TI, 7610) } <= new added
{ "TI XX12", IS_TI|IS_CARDBUS, ID(TI, XX12) } <= new added
{ "TI XX21", IS_TI|IS_CARDBUS, ID(TI, XX21) } <= new added
{ "TI 6515", IS_TI|IS_CARDBUS, ID(TI, 6515) } <= new added
{ "Toshiba ToPIC95-A", IS_CARDBUS|IS_TOPIC, ID(TOSHIBA, TOPIC95_A) }
{ "Toshiba ToPIC95-B", IS_CARDBUS|IS_TOPIC, ID(TOSHIBA, TOPIC95_B) }
{ "Toshiba ToPIC97", IS_CARDBUS|IS_TOPIC, ID(TOSHIBA, TOPIC97) }
{ "Toshiba ToPIC100", IS_CARDBUS|IS_TOPIC, ID(TOSHIBA, TOPIC100) }
==========
I don't know if the new added devices will work or not. However, you can give it a try and let me know the result. In addition, if your pccard slot works under linux and it is not listed on the above the list (and if you believe it may work under mac), let me know the device id and I will try to add it to the next revision.

Great. I will try it later at my tiger, JaS 10.4.8. Thanks

Hi Chun-Nan

i get iopccardfamily error on load, with tiger vanilla 10.4.11 (of course with copy and permissions repair )

do it needs a patched iopccardfamily also?

regards.

Hi,

I will check it this week. I was working the one for Leopard. Thanks.

Chun-Nan

sorry bad typo
i meant iopcifamily and not iopccardfamily.
leo kext you've made has iopccardfamily and iopcifamily kext
does also tiger needs it (i mean the iopcifamily)?
i can confirm leo kext doens't work on tiger

Hi guys,

I have updated rev1 of IOPCIFamily and IOPCCardFamily.

Chun-Nan

Hi Chun-Nan,

kernel panic with rev1 on tiger 10.4.11 (Kalyway 10.4.10 updated via software update).

Regards,
Mariusz.

Hi Chun-Nan

my computer has ENE CB1410 cardbus and running 10.4.11, try to use various IOPCC & IOPCI with following results :
- got same kernel panic as Mariusz on yr rev1 of IOPCI & IOPCC
- got this on yr rev1 of IOPCC and gruntster IOPCI


- got "IOPCCardBridge::checkBridgeBusIDs invalid sub/cardbus/pci settings of 0x0" on manually hexedit stock IOPCC and stock IOPCI
- got working cardbus on manually hexedit stock IOPCC and gruntster IOPCI with the following message


Regards,

Barbie

Hi Chun-Nan!
I tried your kexts for Tiger to help your development.
Very dangerous!

That's all.
I know the problem in MacOSX version. I have 10.4.6 but you write the driver for 10.4.8 and up.
Day by day I am trying to explain that the problem with PCMCI not with bus number, not with addresses, not with DeviceID.
Look once more.
Default IOPCCardFamily.kext
IOPCIFamily v2.0-2.2 - (yes, I can use its)

IOPCIFamily v1.6-1.7

In this case I can insert Adaptec SlimSCSI 1480 and use it. It works!
Now I use my compilation of Apple's opensource IOPCIFamily-32. It works fine without any tunings.

What is happen? I compare sources and found a great difference in Interrupt resolver.
Moreover sound and graphis work with delays if IOPCIFamily 2.x but perfectly if IOPCIFamily 1.x.
Sound sorces explicitly uses DMA. So as ATIRadeon9700 driver, so as in IOPCCardFamily

But in new AppleACPIPlatform there are changes about DMA.
The same problem I observe from other users reports with more recent Tiger and Leo.
The new interrupt policy might works with some users at some motherboard with some BIOS.
But I dunno it is common solution.

Hi guys,I have uploaded the rev1_fix. Hope the kernel panic will disappear. Give it a try and let me know the result. Thanks.Chun-NanHi Slice,As far as I know, the reason why the roginal IOPCIFamily doesn't work with IOPCCardFamily is because the arrangment of BUS Numbering & Memory/IO resource that IOPCIFamily offers to IOPCCardFamily is somehow not accurate. Apple has changed the way they wrote IOPCIFamily after version of 79.1. (You can compare the source code, you will know the difference) Honestly, I haven't found the reason why the arrangement is not accurate. Right now what I do is try to correct the wrong arrangement.I am using Tiger 10.4.11 and the reason you got kernel panic is because I used the wrong kernel to compile (Sorry about that). I have recompiled the file with Tiger Kernel and have tested it on my machine. It works.Chun-NanHi barbie,I have uploaded rev1_fix. You can give it a try and let me know the result.Chun-Nan

Hi Chun-Nan,

your rev1 fix works like a charm for me.
Here is dmesg:



p.s.
last mesages are related to mine PCMCIA USB adpter (EHCI and UHCI) problem ... like in Leo

Great work Chun-Nan!!

Regards,
Mariusz.

YUJUUUULLL!!! ... Thank you Chun-Nan, rev1_fix works for my Acer 5634WLMi, again many thanks Chun-Nam you are the MEN.

Greetings MacOff

PD I have instaled Uphuck 10.4.9 v1.3

dmesg | grep PC
LPC Bridge Vendor ID 0x8086, Device ID 0x27b9
PCI configuration changed (BUS free=1 needed=1 added=0)
PCI configuration changed (bridge=0 device=5 yenta=1)
IOPCCard info: Intel PCIC probe: ENE 712/4 rev 10
IOPCCard info: cs: cb_alloc(bus 7): vendor 0x1106, device 0x3044
IOPCCard info: cs: cb_alloc(bus 7): found 1 functions
IOPCCard info: cs: cb_config(bus 7)
IOPCCard info: fn 0 bar 1: mem 0xa0001000-0xa00017ff
IOPCCard info: fn 0 bar 2: io 0x100-0x17f
IOPCCard info: fn 0 bar 3: mem 0xa0000000-0xa00000ff
IOPCCard info: irq 16
Got boot device = IOService:/AppleACPIPlatformExpert/PCI0@0/AppleACPIPCI/SATA@1F,2/
AppleIntelPIIXATARoot/PRID@0/AppleIntelICHxSATA/ATADeviceNub@0/IOATABlockStorageDriver/
IOATABlockStorageDevice/IOBlockStorageDriver/Hitachi HTS541212H9AT00 Media/IOFDiskPartitionScheme/
Untitled 1@1

Cheer Chun-Nan!
You made that I dream! I use only IOPCIFamily.kext that you fix.

OK!
But more essencial for me that now I can set OpenGL memory more then f..g 3Mb and its works!
My Radeon 9000IGP (devID=5835) never works before!

I still have a problems with sound. May be you look at my ioreg to find remaining mistake?
I admire you

One more problem. My AGPGart works with IOPCIFamily 2.0-2.1, but not works with IOPCIFamily 1.6-1.7 and your rev1. Help!

Hi Slice,

Congratulations that your pccard is working now. If you find any bugs after using rev1_fix, please let me know.

I am sorry to say that I wish I could help you with the sound. I don't think Apple has released all the audio source code, especially the HDA ,AC97, etc. If I had the source code, I might look at it. I know Taruga has been working on it. Maybe you can check this link to see if you can get any help.
http://forum.insanelymac.com/index.php?showtopic=32859

Chun-Nan

@Chun-Nan

Would really appreciate if you could make a similar change to the Leopard version.

good work chun nan.
i've installed new release on tiger 10.4.11 vanilla.
works like a charm but i have same firewire issues of leopard10.5.1
i hope you'll find a way to solve my little prob.

regards

Hi,

If you're looking for support in leopard, please go to this thread:-
CardBus/PCMCIA PC Cards for Leopard

Good luck.

Hi,

I have installed this REV6 in my Hp Notebook nx7300 loaded with kalyway 10.5.1 and after installation the PCARD driver and on rebooting OS Crashed. Power botton image came in to transparent shade and asking to press and hold the switch ...like.

I copied back the backed up old pc card kexts and tried re booting and same result only ... its permenantly crashed.

In kalyway I had another incident too, Installed the Intel wireless Driver for 3945 and on reboot same way it crashed.
Infact I am afried to install anything on Kaly way..!

I had tried many things in other versions of leopard - Hotfix, 9a5.....(I forgot the patch number) and there was no this kind of issue yet. I will try this pc card rev 6 in hotfix leopard version and let me see what will happen.

If anybody can give a suggestion on this matter would be apreciated. I am still not decided which version of leopard to be finalised. Once it loaded then I have to run my day to day functionality

Regards
Gopan

Hi Chun-Nan!
I have the sources of AC97 sound. But I think another developers working on it. I want to say you that it works good with IOPCIFamily 1.6 compiled from sources-32. As you are deep in the project you can find what is change. Why it becomes buggy with your IOPCIFamily? QuickTime Player doesn't play music.
And more.
I didn't understand yet why AGPGart doesn't work with your IOPCIFamily. It calls createNub many times and ATA becomes non-working

I'll try to rewrite the AGPGart but what about other users that uses other AGP driver?

Hi Chun-Nan!
I made an analysis and found mistakes.
The main is that IOPCIFamily 2.2r1 make ("IOPCIConfigured" = Yes) for PCI bus as whole while Apple driver make it for individual devices! That is why AGPGart try to configure FireWire, USB, ATA and so on that it can't.
I made ioregs in single boot mode with three different IOPCIFamily.kext v1.6, v2.0 by Apple and your v2.2r1.
The results:
PCMCI
v1.6 & v2.2r1 good
"Bridge Memory Ranges" = (0xffffffffc0202900,0xffffffffc0202fff,0xffffffffc0206000,0xffffffffc02fffff,0xfffffffffff00000,0xfffff)
v2.0 - bad
"Bridge Memory Ranges" = (0xffffffffc0202900,0xffffffffc0202fff,0xffffffffc0205000,0xffffffffc02fffff,0xfffffffffff00000,0xfffff)

Sound
v1.6 good
| "assigned-addresses" = <10a5000200000000004400c00000000000010000>
v2.2r1 bad
| "assigned-addresses" = <10a5008200000000004400c00000000000010000>
BTW v2.0 & 2.2r1 have "IODeviceMemory" - it is good while not used

ATI Radeon
V1.6 - no delays, OGLMemory=0
| | | "IODeviceMemory" = (({"address"=0xffffffffd0000000,"length"=0x10000000}),
{"offset"=0x9000,"parent"=({"address"=0x0,"length"=0x10000}),"length"=0x100},
({"address"=0xffffffffc0100000,"length"=0x10000}))
| | | "assigned-addresses" = <1028014200000000000000d0000000000000001014280101000000000090000000000000000
100001828010200000000000010c00000000000000100>
v2.0 - delays. OGLMemory=0
| | | "IODeviceMemory" = (({"address"=0xffffffffd0000000,"length"=0x10000000}),({"address"=0xffffffffc0100000,"length"=0x10000}))
| | | "assigned-addresses" = <102801c200000000000000d0000000000000001014280181000000000090000000000000000
100001828018200000000000010c00000000000000100>
v2.2r1 no delays, good OGLMemory, but no DMA translations
| | | "IODeviceMemory" = (({"address"=0xffffffffd0000000,"length"=0x10000000}),({"address"=0xffffffffc0100000,"length"=0x10000}))
| | | "assigned-addresses" = <102801c200000000000000d0000000000000001014280181000000000090000000000000000
100001828018200000000000010c00000000000000100>
I think best values in v1.6

AGP host controller
v1.6
| | | "IODeviceMemory" = (({"address"=0xffffffffc4000000,"length"=0x4000000}),({"address"=0xffffffffc0000000,"length"=0x1000}))
| | | "reg" = <00000000>
| | | "assigned-addresses" = <1000004200000000000000c400000000000000041400000200000000000000c00000000000100
00>
v2.0 - good, but IODeviceMemory will be corrected later by AGPGart
| | "IOPCIConfigured" = Yes
| | "acpi-reg" = <0000000000000000000000000000000000000000>
| | "reg" = <0000000000000000000000000000000000000000>
v2.2r1
| | "reg" = <0000000000000000000000000000000000000000>
| | "IOPCIResourced" = Yes

I can't compile your sources because with XCode 2.4.1 I have no i386/cpu_number.h
I hope you correct all mistakes.

Here is all 3 my ioregs for additional informations if you need

:)Chun-Nan Nice work man made my PCMCIA "Alive" weeeeeeee


CherokeeNMac:~ ***$ sudo dmesg | grep PCPCI configuration changed (BUS free=0 needed=1 added=1)PCI configuration changed (bridge=0 device=3 yenta=1)IOPCCard info: Intel PCIC probe: TI XX12 rev 00FireWire (OHCI) TI ID 803a PCI now active, GUID 5300385641411000; max speed s400.Got boot device = IOService:/AppleACPIPlatformExpert/PCI0@0/AppleACPIPCI/SATA@1F,2/AppleIntelPIIXATARoot/PRID@0/AppleIntelICHxSATA/ATADeviceNub@0/IOATABlockStorageDriver/IOATABlockStorageDevice/IOBlockStorageDriver/FUJITSU MHV2080BH Media/IOFDiskPartitionScheme/Untitled 2@2
CherokeeNMac:~ ***$

Just so people know, Chun-Nan is not available at the moment - he'll be back in a few days.

Don't despair!

Hi Chun-Nan,
I've try yr rev1-fix with result as follow
-first attemp causing computer freeze with messages asf:


-second attemp : seems that yr IOPCI causing a kernel panic with messages similar to above/about AppleUSBUHCI (sorry can't give you dispaly pic as i have no digital camera around)

Anyhow, yr IOPCC success in detecting my PC Card as ENE 1410 when combined with gruntster IOPCI with message asf


regards,

Barbie

Chun-nan resolve only the problem with PCMCI bus but his IOPCIFamily.kext give rise to many problem with another devices. May be we resolve its all together?
If you have two version of IOPCIFamily that not crash your system.
For example located in folders /Old and /New at root of your hdd. Make the follow:
boot with -v -s
/sbin/fsck -fy
/sbim/mount -uw /
ioreg -l -x -w 2048 >ioregYourName-1.txt
rm -r -v /S*/L*/E*/IOPCIF*
cp -r -v /New/IOPCIFamily.kext /S*/L*/E*/
reboot
boot with -v -s
/sbin/fsck -fy
/sbim/mount -uw /
ioreg -l -x -w 2048 >ioregYourName-2.txt
rm -r -v /S*/L*/E*/IOPCIF*
cp -r -v /Old/IOPCIFamily.kext /S*/L*/E*/
reboot
Now compare these txt files and found differencies for your problem devices
Publish differences only!

Hi I am new at this, could someone tell me how I replace kext files? I tried the chmod -R 755 and chown -R root: wheel in the terminal and crashed my system. Can you delete the kext files in the extensions folder and drag the new ones in? I have the TI PCIxx12 cardbus here would love to get it working.

thx for any help

bestman50

Hi Barbie,

According the message above from gruntster, I guess he used part of "IOPCIFamily-34" source code. However, I don't have gruntster's source code IOPCIFamily so I cannot do any comparsions.
I noticed that you have AGP driver installed. Just curious if you have "AGPGart" installed at this moment?

Chun-Nan



Hi Slice,

Right now I am out of town so I cannot do anything till I am back on Saturday.I don't know too much about "AGPGart" because I don't have it installed on my machine.
You can try to use the original IOPCIFamily (2.4) for Leopard and IOPCIFamily (2.2) to see you still have the same problems when you use the rev6 for Leopard and the rev1 for Tiger.
If you still have the same problems with the original IOPCIFamily files (2.4 or 2.2), you probably need to check your AGPGart to find out the reason.
If you don't have the same problems with the orginal IOPCIFamily files (2.4 or 2.2), then there must be some problems with rev6 or rev1.
Please try the orginal IOPCIFamily files (2.4 for Leopard or 2.2 for Tiger) and let me know the result. Thanks.

Chun-Nan

Hi Chun-Nan
I propose you to discuss my correction to your sources. It's a pity I work only in Tiger, so correction for Leo at your will.
My correction begin with

and ended with

I switch off "cpu_number" cause I have no the header. You must switch it on again.

My sincerely request is to include AGP host controller support (it presents in my codes).
New version is 2.2.2!

Hi Chun-Nan,

Yes I have "AGPGart" installed, it is Joblo10's AGPGart.kext version universal beta 2. Is there anything that i can do with this agpgart to help you resolve tigers's IOPCI issue ?

Regards,

Barbie

I need to rewrite AGPGart to be more compatible with PCMCI and so with new IOPCIFamily. Wait for news at my thread
http://forum.insanelymac.com/index.php?showtopic=71308

Hi Chun-Nan,
Last night I tried your rev1 kext for tiger with JaS 10.4.8. Replace both IOPCIFamily & IOPCCardFamily kext file with your rev1 kext file. I tested a couple of hours, so far my PCMCIA & other hardware working flawlessly as expected. Good job Chun-Nan. Thank you.

P/S: Back to leopard + EFI today because I got fix mouse pointer tearing issue on some application such as firefox & dreamweaver cs3.

Hi Chun-Nan!

I work on AGPGart and IOPCIFamily compatibility.
My corrections have no influence.
1. With default IOPCIConfigurator i.e. wtihout your corrections I have no PCMCI but AGPGart loaded successfully.
2. With your corrections IOPCIFamily create own nub and it is configured so AGPGart can't load. But PCMCI works.

AGP bridge has IOPCIClass=0x060400 while PCMCI 0x060401. We might take it into account.
Or we can check IOPCIAGPCapability to not close this nub.
I can't do anything inside AGPGart because it loaded after IOPCIFamily.

This is my ioreg with AGPGart in first variant.

if you do

sudo cp -R (place where your new kext is) /System/Library/Extentions

you will not need to do any chown/chmod actions

Hi Slice,

Which version of original IOPCIFamily work with your AGPGart? 1.6, 2.2 or 2.4?

Chun-Nan

2.0 & 2.2 that is those who don't support PCMCI.
2.4 is for Leo that I have no. Other user in Leo have default 2.4 and AGPGart loaded.

I think it is better to write subclass CardBridge as separate kext and don't correct IOPCIFamily.
BTW with my corrections I have finest sound.

The alternative would be to fix lots more stuff in iopcifamily - it may well be apple intentionally cripple it.

Why Apple destroy CardBus support in v2.0 being in 1.7? I think because of problems such as we encounter...

With Chun-Nan correction I have good PCMCI IODeviceMemory, while without =0.
It is very interesting for my Radeon problem (very similar bridge) so I shall dig.

Hi Slice,

I agree with you that somehow the IODeviceMemory is not arranged correctly. I am still working on it and trying to find the possible reason.
I checked your source code and I found the following codes:



Should I need to add it to make the AGPGart work? (I don't have AGP device right now) Thanks.

Chun-Nan

Hi Chun-Nan!
Thank you for continue working. I hope we will find a solution.
The mentioned code do nothing to the main problem that I didn't completely understand.
Look again.
You made corrections ONLY to IOPCIConfigurator.cpp. So I have 2 revisions with and without your corrections.
1. WITH your corrections AGPGart don't loaded
2. WITHOUT your correction AGPGart loaded
And nothing more.

May be you need to know something about PCI bus? I can trace all that you want.
Radeon PCI

Intel PCI (host controller)

Useful information:
AGP host controller has header type 0, class=060000
AGP bridge has header type 1, class=060400
Check your codes!

More traces

And PCI structure

Hi Slice,

Is it possible for you to post your FULL source code package of AGPGart or let me know where I can download it? Thanks.

Chun-Nan

No problem!My sources based on joblo AGPGart22 is on www.x86dev.org.The incompatibility comes from IOPCIFamily that close PCI2PCIBridge so AGPGart can't attach to the AGPBridge.

My latest sources with more tunings and output is here.Free for any purpose.

BTW I still don't understand your idea about bus numbering. Can you explain?

Hi guys,

I have updated rev2 of IOPCIFamily and IOPCCardFamily.
http://forum.insanelymac.com/index.php?showtopic=81048

In this revision, I have fixed some "device/bridge command" bugs and it also included the latest IOPCCardFamily-47. Hopefully, it will work better. Please let me know the result if possible, especially Firewire device. Thanks

Chun-Nan

Hi Chun-Nan,

here is dmesg from tiger + rev2:


SystemProfiler says unknow device when connected my DVD->FW adapter (same like in leo).

Regards,
Mariusz.

Hi Chun-Nan!
Thank you for new codes and for cpu_number. It's a pity in my system it is not compilable so I stay with no-rendevouz variant.
I apply your changes to my project and have some questions:

I have no the definition as well as ConfigurationReset

The result always = 0 despite on all previous calculations. May be you want to write
(0 | gIOPCISkip)?


The compiler give me warning about different types. I don't want to ignore any warning because of run-time error.

[quote name='Slice' date='Jan 30 2008, 02:21 PM' post='600903']

The result always = 0 despite on all previous calculations. May be you want to write
(0 | gIOPCISkip)?

no, unless he mean "temporary disabled clause" there must be some variable or complex mask instead of simple "0",
because "0 | variable" means nothing but just "variable" itself, which makes "0" useless also.

2 Chun-Nan
May be you mean

With this changes I have good results

But without AGPGart.
With this version it loaded and trying to attach to any devices. It is probably my mistake but with default IOPCIFamily all is OK!

Hi Slice,

Thank you for the source code file. I will check the source code when I have more time. (School starts so maybe on the weekend)

About bus numbering, here is the sample link. I am sure you can search more on the internet.

http://www.science.unitn.it/~fiorella/guid...100000000000000

About "kIOPCIResetKey", it should be defined in IOPCIPrivate.h.

About "getFreeCount()", it relates to the problem of "IOPCCardBridge::checkBridgeBusIDs invalid sub/cardbus/pci settings of 0x0". Because it doesn't have free bus/range resources, it cause "0x0".
You can check this message on your dmesg "PCI configuration changed (BUS free=0 needed=1 added=1)".

About "gIOPCISkip", that is correct because I don't want it to run the following command but I also don't want to mess up the original source code in case other place needs it.
Thus, I put "int gIOPCISkip = 0;" in the beginning.
Then I put the following codes:
===============================
fPCIBridgeIndex = 0;
fPCIBridgeTailIndex = 0;
do
{
bridge = fPCIBridgeList[fPCIBridgeIndex++];

FOREACH_CHILD(bridge, child)
{
if (child->headerType == kPCIHeaderType2)
{
++fPCIBridgeTailIndex;
// pciBridgeCheckConfiguration(bridge);
gIOPCISkip++;
pciBridgeAllocateResource(bridge);
pciBridgeDistributeResource(bridge);
gIOPCISkip--;
}
}
}
while (fPCIBridgeIndex <= fPCIBridgeTailIndex);
================================
Then I put "if (0 & gIOPCISkip)". Therefore, if gIOPCISkip != 0, it will skip the following command while if gIOPCISkip = 0, it won't.

Chun-Nan

Hey guys,

I've just tried rev1fix and rev2 on an old Dell C640.
I'm using AGPGart 2.2 - the gfx card is an ATI Radeon Mobility 7500. (I have QE on it! Yay!).


Anyway, neither version crashes for me, WITH agpgart 2.2 starting normally.


I'm attaching a set of 4 ioreg dumps I took while testing the various permutations. They're attached in the order I made them. I used ioreg -lx -w0


BTW Slice,
Ever compiled a linux kernel? Compiler warnings are generally ignored - only if there are actual problems do people even consider them - most, especially about different types, can be ignored regardless. It's basically lazy code, relying on the compiler to make it work correctly. GCC is very good at this.



EqUaTe


Click to view attachment
Click to view attachment
Click to view attachment
Click to view attachment

Hi slice,

What is the difference between agpgart 2.2 and your agpgart revsion?

Chun-Nan

Hi Chun-Nan,

Your rev 2 of IOPCI still causing error in my comp. It is about USBF AppleUSBUHCI: UIMInitialize - ioMap is NULL unable to initialize UIM something... just like error generated by the rev 1 IOPCI...

regards

Barbie