85 replies to this topic

[ludacrisvp]

For what its worth to note, my laptop does still completely boot when it seems to 'hang' at PCI configuration begin. Once booted the actual boot log does show up in the console logs to review.
I recently changed my WLAN card from the dell broadcom 4311 b/g card to a dual radio 2.4/5.0Ghz dell broadcom 4331 wireless N WLAN card.
This card change caused the above PCI configuration begin issue on my laptop.
I can boot with the npci=0x2000 and it will properly display the boot progress completely.
This is true for Lion and Mountain Lion DP1, have not tested DP2

[Gringo Vermelho]

This happens to me as well if I remove npci=0x2000, only tested on 10.6.8 though. I haven't tried it on Lion.

P5Q-E, E8500, GTX460

[larry100]

Hi,

I've got this error installing Lion on a ION based PC and on a HP Compaq Elite 8100 equipped with a NVidia 9400 GT card, so, in my case, it seems to me NVidia related.

Nothing solved 'til I put the following kexts into S/L/E:

http://prasys.info/l...s_by_nawcom.zip

Hope this would help.

L

[Captain America]

What works for me (MSI GX740 235-US) is "pci=off npci=0x3000 GraphicsEnabler=No"

[Hackintosh2000]

ludacrisvp, on 22 March 2012 - 07:19 PM, said:

ludacrisvp, on 22 March 2012 - 07:19 PM, said:

For what its worth to note, my laptop does still completely boot when it seems to 'hang' at PCI configuration begin. Once booted the actual boot log does show up in the console logs to review.
I recently changed my WLAN card from the dell broadcom 4311 b/g card to a dual radio 2.4/5.0Ghz dell broadcom 4331 wireless N WLAN card.
This card change caused the above PCI configuration begin issue on my laptop.
I can boot with the npci=0x2000 and it will properly display the boot progress completely.
This is true for Lion and Mountain Lion DP1, have not tested DP2

I recently changed my WLAN card from the dell broadcom 4311 b/g card to a dual radio 2.4/5.0Ghz dell broadcom 4331 wireless N WLAN card.
This card change caused the above PCI configuration begin issue on my laptop.
I can boot with the npci=0x2000 and it will properly display the boot progress completely.
This is true for Lion and Mountain Lion DP1, have not tested DP2

For what its worth to note, my laptop does still completely boot when it seems to 'hang' at PCI configuration begin. Once booted the actual boot log does show up in the console logs to review.

Gringo Vermelho, on 22 March 2012 - 09:23 PM, said:

This happens to me as well if I remove npci=0x2000, only tested on 10.6.8 though. I haven't tried it on Lion.

P5Q-E, E8500, GTX460

Same here.  I didn't check the log though once the system booted.  I was working on a Dell Studio Slim 540s for a friend and this started happening with 10.6.8.  When I deleted the patched DSDT it would boot and stop outputing at PCI Configuration Begin but the HDD access light keeps flashing and after a little while the desktop appears.  When I add npci=0x3000 verbose mode outputs normally and the computer boots normally.  Unfortunately without that DSDT I'm drowning in kexts.  I need to figure out how to patch my DSDT.

[Gringo Vermelho]

Since installing a USB 2.0 expansion card in the PCI-E x1 slot above the video card, I don't need the npci flag anymore.
 asus_p5q_e.jpg   101.22K   1 downloads
After adding the USB card and replacing my GTX 460 with a GTX 660 I had to re-extract and repatch my DSDT due to changed memory locations (OperationRegion, SystemMemory).

I've also changed this...

	 Device (PCI0)
	 {
		 Name (_HID, EisaId ("PNP0A08")) // _HID: Hardware ID
		 Name (_ADR, Zero) // _ADR: Address
		 Method (^BN00, 0, NotSerialized)
		 {
			 Return (Zero)
		 }

		 Method (_BBN, 0, NotSerialized) // _BBN: BIOS Bus Number
		 {
			 Return (BN00 ())
		 }

		 Name (_UID, Zero) // _UID: Unique ID
		 Method (_PRT, 0, NotSerialized) // _PRT: PCI Routing Table
		 {
			 If (PICM)
			 {
				 Return (AR00)
			 }

			 Return (PR00)
		 }

		 Method (_S3D, 0, NotSerialized) // _S3D: S3 Device State
		 {
			 If (LOr (LEqual (OSFL (), One), LEqual (OSFL (), 0x02)))
			 {
				 Return (0x02)
			 }
			 Else
			 {
				 Return (0x03)
			 }
		 }

		 Name (_CID, EisaId ("PNP0A03")) // _CID: Compatible ID

...to this:

	 Device (PCI0)
	 {
		 Name (_HID, EisaId ("PNP0A08")) // _HID: Hardware ID
		 Name (_CID, EisaId ("PNP0A03")) // _CID: Compatible ID
		 Name (_ADR, Zero) // _ADR: Address
		 Name (_UID, Zero) // _UID: Unique ID
		 Name (_BBN, Zero) // _BBN: BIOS Bus Number
		 Method (_PRT, 0, NotSerialized) // _PRT: PCI Routing Table
		 {
			 If (PICM)
			 {
				 Return (AR00)
			 }

			 Return (PR00)
		 }

		 Method (_S3D, 0, NotSerialized) // _S3D: S3 Device State
		 {
			 If (LOr (LEqual (OSFL (), One), LEqual (OSFL (), 0x02)))
			 {
				 Return (0x02)
			 }
			 Else
			 {
				 Return (0x03)
			 }
		 }

Much cleaner!

System.log excerpt:

Feb 22 00:29:15 localhost kernel[0]: [ PCI configuration begin ]
Feb 22 00:29:15 localhost kernel[0]: console relocated to 0xf89000000
Feb 22 00:29:15 localhost kernel[0]: PCI configuration changed (bridge=1 device=1 cardbus=0)
Feb 22 00:29:15 localhost kernel[0]: [ PCI configuration end, bridges 7 devices 20 ]

I'm pretty sure the fact that it works now is somehow related to the card having its own PCI bridge:

03:00.0 PCI bridge [0604]: Pericom Semiconductor Device [12d8:e111] (rev 02)
04:00.0 USB controller [0c03]: NEC Corporation OHCI USB Controller [1033:0035] (rev 43)
04:00.1 USB controller [0c03]: NEC Corporation OHCI USB Controller [1033:0035] (rev 43)
04:00.2 USB controller [0c03]: NEC Corporation uPD72010x USB 2.0 Controller [1033:00e0] (rev 04)

ASUS P5Q-E, 10.8.2

Links of interest:
http://en.wikipedia....ki/Root_complex
http://www.acpi.info/acpi_faq.htm (do a page search for "root bus")
http://opensource.ap...onfigurator.cpp
http://www.projectos...indpost&p=15259
Also google _BBN root bus