PCI IRQ routing strangeness with 2.4.22 on A7N266-VM (original nForce)

PCI IRQ routing strangeness with 2.4.22 on A7N266-VM (original nForce)

[John Wong]

[-- Attachment #1: Type: text/plain, Size: 2131 bytes --]

On the nForce original with the onboard GeForce2 video, the BIOS does
assign IRQ 11 to the onboard video, however, Linux does not.  Setting
acpi=off and pci=noirq does not fix the problem.  I have a nForce2 board
without onboard video and ACPI/IO-APIC etc all work fine.

There is a strange entry in the dmesg log:

Pin 2-16 already programmed
    ACPI-1121: *** Error: Method execution failed [\_SB_.SRS_] (Node
cdf363c0),
AE_AML_BUFFER_LIMIT
    ACPI-1121: *** Error: Method execution failed [\_SB_.LNKE._SRS]
(Node cdf36dc0), AE_AML_BUFFER_LIMIT
ACPI: Unable to set IRQ for PCI Interrupt Link [LNKE] (likely buggy ACPI
BIOS).
Aborting ACPI-based IRQ routing. Try pci=noacpi or acpi=off
ERROR: Unable to locate IOAPIC for IRQ -19/n    ACPI-1121: *** Error:
Method execution failed [\_SB_.SRS_] (Node cdf363c0),
AE_AML_BUFFER_LIMIT
    ACPI-1121: *** Error: Method execution failed [\_SB_.LNKE._SRS]
(Node cdf36dc0), AE_AML_BUFFER_LIMIT
ACPI: Unable to set IRQ for PCI Interrupt Link [LNKE] (likely buggy ACPI
BIOS).
Aborting ACPI-based IRQ routing. Try pci=noacpi or acpi=off


I will also attach the dmesg as an attachment.

Here is the interrupts:

           CPU0
  0:     193151    IO-APIC-edge  timer
  1:       9633    IO-APIC-edge  keyboard
  2:          0          XT-PIC  cascade
  8:          4    IO-APIC-edge  rtc
  9:          0   IO-APIC-level  acpi
 12:       3012    IO-APIC-edge  PS/2 Mouse
 14:       8515    IO-APIC-edge  ide0
 18:          0   IO-APIC-level  EMU10K1
 22:   42753540   IO-APIC-level  eth0
NMI:          0
LOC:     193041
ERR:          0
MIS:          0

And the lspci -v:

02:00.0 VGA compatible controller: nVidia Corporation NV15 [GeForce2 -
nForce GPU] (rev b1) (prog-if 00 [VGA])
        Subsystem: nVidia Corporation: Unknown device 0c11
        Flags: bus master, 66Mhz, medium devsel, latency 32, IRQ -19
        Memory at e5000000 (32-bit, non-prefetchable) [size=16M]
        Memory at e8000000 (32-bit, prefetchable) [size=128M]
        Expansion ROM at e7ff0000 [disabled] [size=64K]
        Capabilities: [60] Power Management version 2
        Capabilities: [44] AGP version 2.0


[-- Attachment #2: dmesg --]
[-- Type: text/plain, Size: 12117 bytes --]

Linux version 2.4.22 (root@johnw) (gcc version 3.3.2 20030812 (Debian prerelease)) #1 Thu Aug 28 17:28:52 PDT 2003
BIOS-provided physical RAM map:
 BIOS-e820: 0000000000000000 - 0000000000080000 (usable)
 BIOS-e820: 000000000009fc00 - 00000000000a0000 (reserved)
 BIOS-e820: 00000000000f0000 - 0000000000100000 (reserved)
 BIOS-e820: 0000000000100000 - 000000000dfec000 (usable)
 BIOS-e820: 000000000dfec000 - 000000000dfef000 (ACPI data)
 BIOS-e820: 000000000dfef000 - 000000000dfff000 (reserved)
 BIOS-e820: 000000000dfff000 - 000000000e000000 (ACPI NVS)
 BIOS-e820: 00000000fec00000 - 00000000fec01000 (reserved)
 BIOS-e820: 00000000fee00000 - 00000000fee01000 (reserved)
 BIOS-e820: 00000000ffff0000 - 0000000100000000 (reserved)
223MB LOWMEM available.
ACPI: have wakeup address 0xc0001000
On node 0 totalpages: 57324
zone(0): 4096 pages.
zone(1): 53228 pages.
zone(2): 0 pages.
ACPI: RSDP (v000 ASUS                                      ) @ 0x000f8090
ACPI: RSDT (v001 ASUS   A7N266VM 0x42302e31 MSFT 0x31313031) @ 0x0dfec000
ACPI: FADT (v001 ASUS   A7N266VM 0x42302e31 MSFT 0x31313031) @ 0x0dfec100
ACPI: BOOT (v001 ASUS   A7N266VM 0x42302e31 MSFT 0x31313031) @ 0x0dfec040
ACPI: MADT (v001 ASUS   A7N266VM 0x42302e31 MSFT 0x31313031) @ 0x0dfec080
ACPI: DSDT (v001   ASUS A7N266VM 0x00001000 MSFT 0x0100000b) @ 0x00000000
ACPI: Local APIC address 0xfee00000
ACPI: LAPIC (acpi_id[0x00] lapic_id[0x00] enabled)
Processor #0 Pentium(tm) Pro APIC version 16
ACPI: LAPIC_NMI (acpi_id[0x00] polarity[0x1] trigger[0x1] lint[0x1])
ACPI: IOAPIC (id[0x02] address[0xfec00000] global_irq_base[0x0])
IOAPIC[0]: Assigned apic_id 2
IOAPIC[0]: apic_id 2, version 17, address 0xfec00000, IRQ 0-23
ACPI: INT_SRC_OVR (bus[0] irq[0x0] global_irq[0x2] polarity[0x0] trigger[0x1])
ACPI: INT_SRC_OVR (bus[0] irq[0x9] global_irq[0x9] polarity[0x1] trigger[0x3])
Using ACPI (MADT) for SMP configuration information
Kernel command line: root=/dev/hda1 ro
Initializing CPU#0
Detected 1662.406 MHz processor.
Console: colour VGA+ 80x25
Calibrating delay loop... 3316.12 BogoMIPS
Memory: 224268k/229296k available (1213k kernel code, 4516k reserved, 311k data, 276k init, 0k highmem)
Dentry cache hash table entries: 32768 (order: 6, 262144 bytes)
Inode cache hash table entries: 16384 (order: 5, 131072 bytes)
Mount cache hash table entries: 512 (order: 0, 4096 bytes)
Buffer cache hash table entries: 16384 (order: 4, 65536 bytes)
Page-cache hash table entries: 65536 (order: 6, 262144 bytes)
CPU: L1 I Cache: 64K (64 bytes/line), D cache 64K (64 bytes/line)
CPU: L2 Cache: 256K (64 bytes/line)
Intel machine check architecture supported.
Intel machine check reporting enabled on CPU#0.
CPU:     After generic, caps: 0383fbff c1c3fbff 00000000 00000000
CPU:             Common caps: 0383fbff c1c3fbff 00000000 00000000
CPU: AMD Athlon(TM) XP 2000+ stepping 01
Enabling fast FPU save and restore... done.
Enabling unmasked SIMD FPU exception support... done.
Checking 'hlt' instruction... OK.
POSIX conformance testing by UNIFIX
enabled ExtINT on CPU#0
ESR value before enabling vector: 00000000
ESR value after enabling vector: 00000000
ENABLING IO-APIC IRQs
init IO_APIC IRQs
 IO-APIC (apicid-pin) 2-0, 2-16, 2-17, 2-18, 2-19, 2-20, 2-21, 2-22, 2-23 not connected.
..TIMER: vector=0x31 pin1=2 pin2=0
..MP-BIOS bug: 8254 timer not connected to IO-APIC
...trying to set up timer (IRQ0) through the 8259A ... 
..... (found pin 0) ...works.
number of MP IRQ sources: 16.
number of IO-APIC #2 registers: 24.
testing the IO APIC.......................

IO APIC #2......
.... register #00: 02000000
.......    : physical APIC id: 02
.......    : Delivery Type: 0
.......    : LTS          : 0
.... register #01: 00170011
.......     : max redirection entries: 0017
.......     : PRQ implemented: 0
.......     : IO APIC version: 0011
.... register #02: 00000000
.......     : arbitration: 00
.... IRQ redirection table:
 NR Log Phy Mask Trig IRR Pol Stat Dest Deli Vect:   
 00 001 01  0    0    0   0   0    1    1    31
 01 001 01  0    0    0   0   0    1    1    39
 02 000 00  1    0    0   0   0    0    0    00
 03 001 01  0    0    0   0   0    1    1    41
 04 001 01  0    0    0   0   0    1    1    49
 05 001 01  0    0    0   0   0    1    1    51
 06 001 01  0    0    0   0   0    1    1    59
 07 001 01  0    0    0   0   0    1    1    61
 08 001 01  0    0    0   0   0    1    1    69
 09 001 01  1    1    0   0   0    1    1    71
 0a 001 01  0    0    0   0   0    1    1    79
 0b 001 01  0    0    0   0   0    1    1    81
 0c 001 01  0    0    0   0   0    1    1    89
 0d 001 01  0    0    0   0   0    1    1    91
 0e 001 01  0    0    0   0   0    1    1    99
 0f 001 01  0    0    0   0   0    1    1    A1
 10 000 00  1    0    0   0   0    0    0    00
 11 000 00  1    0    0   0   0    0    0    00
 12 000 00  1    0    0   0   0    0    0    00
 13 000 00  1    0    0   0   0    0    0    00
 14 000 00  1    0    0   0   0    0    0    00
 15 000 00  1    0    0   0   0    0    0    00
 16 000 00  1    0    0   0   0    0    0    00
 17 000 00  1    0    0   0   0    0    0    00
IRQ to pin mappings:
IRQ0 -> 0:0
IRQ1 -> 0:1
IRQ3 -> 0:3
IRQ4 -> 0:4
IRQ5 -> 0:5
IRQ6 -> 0:6
IRQ7 -> 0:7
IRQ8 -> 0:8
IRQ9 -> 0:9
IRQ10 -> 0:10
IRQ11 -> 0:11
IRQ12 -> 0:12
IRQ13 -> 0:13
IRQ14 -> 0:14
IRQ15 -> 0:15
.................................... done.
Using local APIC timer interrupts.
calibrating APIC timer ...
..... CPU clock speed is 1662.4018 MHz.
..... host bus clock speed is 265.9843 MHz.
cpu: 0, clocks: 2659843, slice: 1329921
CPU0<T0:2659840,T1:1329904,D:15,S:1329921,C:2659843>
mtrr: v1.40 (20010327) Richard Gooch (rgooch-r1x6VkxMR+00zabcByZE4g@public.gmane.org)
mtrr: detected mtrr type: Intel
ACPI: Subsystem revision 20030813
PCI: PCI BIOS revision 2.10 entry at 0xf1b40, last bus=2
PCI: Using configuration type 1
ACPI: Interpreter enabled
ACPI: Using IOAPIC for interrupt routing
ACPI: System [ACPI] (supports S0 S1 S4 S5)
ACPI: PCI Interrupt Link [LNKA] (IRQs 16 18, enabled at IRQ 5)
ACPI: PCI Interrupt Link [LNKB] (IRQs 16 18, disabled)
ACPI: PCI Interrupt Link [LNKC] (IRQs 16 18, enabled at IRQ 5)
ACPI: PCI Interrupt Link [LNKD] (IRQs 16 18, disabled)
ACPI: PCI Interrupt Link [LNKE] (IRQs 19, enabled at IRQ 11)
ACPI: PCI Interrupt Link [LNKF] (IRQs 20 21 22, enabled at IRQ 5)
ACPI: PCI Interrupt Link [LNKU] (IRQs 20 21 22, enabled at IRQ 10)
ACPI: PCI Interrupt Link [LNKI] (IRQs 20 21 22, enabled at IRQ 10)
ACPI: PCI Interrupt Link [LNKJ] (IRQs 20 21 22, disabled)
ACPI: PCI Interrupt Link [LNKK] (IRQs 20 21 22, disabled)
ACPI: PCI Interrupt Link [LNKM] (IRQs 20 21 22, disabled)
ACPI: PCI Root Bridge [PCI0] (00:00)
PCI: Probing PCI hardware (bus 00)
ACPI: PCI Interrupt Routing Table [\_SB_.PCI0._PRT]
ACPI: PCI Interrupt Routing Table [\_SB_.PCI0.PCI1._PRT]
ACPI: PCI Interrupt Routing Table [\_SB_.PCI0.PCI2._PRT]
PCI: Probing PCI hardware
ACPI: PCI Interrupt Link [LNKF] enabled at IRQ 22
IOAPIC[0]: Set PCI routing entry (2-22 -> 0xa9 -> IRQ 22 Mode:1 Active:0)
00:00:01[A] -> 2-22 -> IRQ 22
Pin 2-22 already programmed
ACPI: PCI Interrupt Link [LNKU] enabled at IRQ 21
IOAPIC[0]: Set PCI routing entry (2-21 -> 0xb1 -> IRQ 21 Mode:1 Active:0)
00:00:02[A] -> 2-21 -> IRQ 21
Pin 2-21 already programmed
ACPI: PCI Interrupt Link [LNKI] enabled at IRQ 22
Pin 2-22 already programmed
ACPI: PCI Interrupt Link [LNKJ] enabled at IRQ 20
IOAPIC[0]: Set PCI routing entry (2-20 -> 0xb9 -> IRQ 20 Mode:1 Active:0)
00:00:05[A] -> 2-20 -> IRQ 20
ACPI: PCI Interrupt Link [LNKK] enabled at IRQ 20
Pin 2-20 already programmed
ACPI: PCI Interrupt Link [LNKM] enabled at IRQ 21
Pin 2-21 already programmed
ACPI: PCI Interrupt Link [LNKA] enabled at IRQ 18
IOAPIC[0]: Set PCI routing entry (2-18 -> 0xc1 -> IRQ 18 Mode:1 Active:0)
00:01:06[A] -> 2-18 -> IRQ 18
ACPI: PCI Interrupt Link [LNKB] enabled at IRQ 16
IOAPIC[0]: Set PCI routing entry (2-16 -> 0xc9 -> IRQ 16 Mode:1 Active:0)
00:01:06[B] -> 2-16 -> IRQ 16
ACPI: PCI Interrupt Link [LNKC] enabled at IRQ 18
Pin 2-18 already programmed
ACPI: PCI Interrupt Link [LNKD] enabled at IRQ 16
Pin 2-16 already programmed
Pin 2-16 already programmed
Pin 2-18 already programmed
Pin 2-16 already programmed
Pin 2-18 already programmed
Pin 2-18 already programmed
Pin 2-16 already programmed
Pin 2-18 already programmed
Pin 2-16 already programmed
Pin 2-16 already programmed
Pin 2-18 already programmed
Pin 2-16 already programmed
Pin 2-18 already programmed
Pin 2-18 already programmed
Pin 2-16 already programmed
Pin 2-18 already programmed
Pin 2-16 already programmed
Pin 2-16 already programmed
Pin 2-18 already programmed
Pin 2-16 already programmed
Pin 2-18 already programmed
Pin 2-18 already programmed
Pin 2-16 already programmed
Pin 2-18 already programmed
Pin 2-16 already programmed
    ACPI-1121: *** Error: Method execution failed [\_SB_.SRS_] (Node cdf363c0), AE_AML_BUFFER_LIMIT
    ACPI-1121: *** Error: Method execution failed [\_SB_.LNKE._SRS] (Node cdf36dc0), AE_AML_BUFFER_LIMIT
ACPI: Unable to set IRQ for PCI Interrupt Link [LNKE] (likely buggy ACPI BIOS). Aborting ACPI-based IRQ routing. Try pci=noacpi or acpi=off
ERROR: Unable to locate IOAPIC for IRQ -19/n    ACPI-1121: *** Error: Method execution failed [\_SB_.SRS_] (Node cdf363c0), AE_AML_BUFFER_LIMIT
    ACPI-1121: *** Error: Method execution failed [\_SB_.LNKE._SRS] (Node cdf36dc0), AE_AML_BUFFER_LIMIT
ACPI: Unable to set IRQ for PCI Interrupt Link [LNKE] (likely buggy ACPI BIOS). Aborting ACPI-based IRQ routing. Try pci=noacpi or acpi=off
PCI: Using ACPI for IRQ routing
PCI: if you experience problems, try using option 'pci=noacpi' or even 'acpi=off'
Linux NET4.0 for Linux 2.4
Based upon Swansea University Computer Society NET3.039
Initializing RT netlink socket
Starting kswapd
Journalled Block Device driver loaded
pty: 256 Unix98 ptys configured
Serial driver version 5.05c (2001-07-08) with MANY_PORTS SHARE_IRQ SERIAL_PCI enabled
ttyS00 at 0x03f8 (irq = 4) is a 16550A
ttyS01 at 0x02f8 (irq = 3) is a 16550A
Floppy drive(s): fd0 is 1.44M
FDC 0 is a post-1991 82077
Uniform Multi-Platform E-IDE driver Revision: 7.00beta4-2.4
ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx
NFORCE: IDE controller at PCI slot 00:09.0
NFORCE: chipset revision 195
NFORCE: not 100% native mode: will probe irqs later
AMD_IDE: Bios didn't set cable bits corectly. Enabling workaround.
ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx
AMD_IDE: nVidia Corporation nForce IDE (rev c3) UDMA100 controller on pci00:09.0
    ide0: BM-DMA at 0xb800-0xb807, BIOS settings: hda:DMA, hdb:DMA
    ide1: BM-DMA at 0xb808-0xb80f, BIOS settings: hdc:pio, hdd:pio
hda: ST38410A, ATA DISK drive
hdb: WDC WD400JB-00ENA0, ATA DISK drive
blk: queue c02efc20, I/O limit 4095Mb (mask 0xffffffff)
blk: queue c02efd5c, I/O limit 4095Mb (mask 0xffffffff)
ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
hda: attached ide-disk driver.
hda: host protected area => 1
hda: 16841664 sectors (8623 MB) w/512KiB Cache, CHS=1048/255/63, UDMA(66)
hdb: attached ide-disk driver.
hdb: host protected area => 1
hdb: 78165360 sectors (40021 MB) w/8192KiB Cache, CHS=4865/255/63, UDMA(100)
Partition check:
 hda: hda1 hda2
 hdb: unknown partition table
NET4: Linux TCP/IP 1.0 for NET4.0
IP Protocols: ICMP, UDP, TCP, IGMP
IP: routing cache hash table of 2048 buckets, 16Kbytes
TCP: Hash tables configured (established 16384 bind 32768)
NET4: Unix domain sockets 1.0/SMP for Linux NET4.0.
kjournald starting.  Commit interval 5 seconds
EXT3-fs: mounted filesystem with ordered data mode.
VFS: Mounted root (ext3 filesystem) readonly.
Freeing unused kernel memory: 276k freed
Adding Swap: 120476k swap-space (priority -1)
EXT3 FS 2.4-0.9.19, 19 August 2002 on ide0(3,1), internal journal
Real Time Clock Driver v1.10e
PCI: Setting latency timer of device 00:04.0 to 64
Creative EMU10K1 PCI Audio Driver, version 0.20, 17:39:48 Aug 28 2003
emu10k1: EMU10K1 rev 7 model 0x8064 found, IO at 0xc800-0xc81f, IRQ 18
ac97_codec: AC97 Audio codec, id: EMC40 (Unknown)
emu10k1: SBLive! 5.1 card detected
ACPI: Processor [CPU] (supports C1)
ACPI: Power Button (FF) [PWRF]

Re: PCI IRQ routing strangeness with 2.4.22 on A7N266-VM (original nForce)

[Andrew de Quincey]

> Pin 2-16 already programmed
>     ACPI-1121: *** Error: Method execution failed [\_SB_.SRS_] (Node
> cdf363c0),
> AE_AML_BUFFER_LIMIT
>     ACPI-1121: *** Error: Method execution failed [\_SB_.LNKE._SRS]
> (Node cdf36dc0), AE_AML_BUFFER_LIMIT

Hi, first of all, my apologies to everyone for the long hiatus in IRQ routing 
patches from me... I've been really busy with other projects.

Secondly, I know what this is. I have a patch for it already. I will send it 
once I've ported it to the latest kernels (and tracked down which machine I 
was developing it on :)

Also, I should be able to fix the issues people were having with KT400s... VIA 
are being quite helpful.



-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf

[PATCH] Drop back to extended IRQ resource type

[Andrew de Quincey]

[-- Attachment #1: Type: text/plain, Size: 843 bytes --]

On Friday 29 Aug 2003 5:50 am, John Wong wrote:
> On the nForce original with the onboard GeForce2 video, the BIOS does
> assign IRQ 11 to the onboard video, however, Linux does not.  Setting
> acpi=off and pci=noirq does not fix the problem.  I have a nForce2 board
> without onboard video and ACPI/IO-APIC etc all work fine.

Hi, the attached patch should resolve this issue. Originally, if an IRQ is 
<=15, the code always used a standard IRQ descriptor when calling the _SRS 
method. 

However, certain BIOSes always expect an extended IRQ descriptor, and fail 
when passed a standard one. This patch drops back to an extended descriptor 
if a set with a standard one fails... you'll still see the BUFFER LIMIT 
message, but you should now see it saying "Retrying with extended IRQ 
descriptor".. and hopefully succeeding the second time.

[-- Attachment #2: linux-2.4.22-irqrestype.patch --]
[-- Type: text/x-diff, Size: 2312 bytes --]


--- linux-2.4.22.orig/drivers/acpi/pci_link.c	2003-08-25 12:44:41.000000000 +0100
+++ linux-2.4.22/drivers/acpi/pci_link.c	2003-08-30 18:54:35.000000000 +0100
@@ -290,7 +290,8 @@
 	struct acpi_buffer	buffer = {sizeof(resource)+1, &resource};
 	int			i = 0;
 	int			valid = 0;
-
+	int			resource_type = 0;
+   
 	ACPI_FUNCTION_TRACE("acpi_pci_link_set");
 
 	if (!link || !irq)
@@ -313,12 +314,23 @@
 		}
 	}
 
+	/* If IRQ<=15, first try with a "normal" IRQ descriptor. If that fails, try with
+	 * an extended one */
+	if (irq <= 15) {
+		resource_type = ACPI_RSTYPE_IRQ;
+	} else {
+		resource_type = ACPI_RSTYPE_EXT_IRQ;
+	}
+
+retry_programming:
+   
 	memset(&resource, 0, sizeof(resource));
 
 	/* NOTE: PCI interrupts are always level / active_low / shared. But not all
 	   interrupts > 15 are PCI interrupts. Rely on the ACPI IRQ definition for 
 	   parameters */
-	if (irq <= 15) {
+	switch(resource_type) {
+	case ACPI_RSTYPE_IRQ:
 		resource.res.id = ACPI_RSTYPE_IRQ;
 		resource.res.length = sizeof(struct acpi_resource);
 		resource.res.data.irq.edge_level = link->irq.edge_level;
@@ -326,8 +338,9 @@
 		resource.res.data.irq.shared_exclusive = ACPI_SHARED;
 		resource.res.data.irq.number_of_interrupts = 1;
 		resource.res.data.irq.interrupts[0] = irq;
-	}
-	else {
+		break;
+	   
+	case ACPI_RSTYPE_EXT_IRQ:
 		resource.res.id = ACPI_RSTYPE_EXT_IRQ;
 		resource.res.length = sizeof(struct acpi_resource);
 		resource.res.data.extended_irq.producer_consumer = ACPI_CONSUMER;
@@ -337,11 +350,22 @@
 		resource.res.data.extended_irq.number_of_interrupts = 1;
 		resource.res.data.extended_irq.interrupts[0] = irq;
 		/* ignore resource_source, it's optional */
+		break;
 	}
 	resource.end.id = ACPI_RSTYPE_END_TAG;
 
 	/* Attempt to set the resource */
 	status = acpi_set_current_resources(link->handle, &buffer);
+   
+
+	/* if we failed and IRQ <= 15, try again with an extended descriptor */
+	if (ACPI_FAILURE(status) && (resource_type == ACPI_RSTYPE_IRQ)) {
+                resource_type = ACPI_RSTYPE_EXT_IRQ;
+                printk(PREFIX "Retrying with extended IRQ descriptor\n");
+                goto retry_programming;
+	}
+  
+	/* check for total failure */
 	if (ACPI_FAILURE(status)) {
 		ACPI_DEBUG_PRINT((ACPI_DB_ERROR, "Error evaluating _SRS\n"));
 		return_VALUE(-ENODEV);

[-- Attachment #3: linux-2.6.0-test4-irqrestype.patch --]
[-- Type: text/x-diff, Size: 2999 bytes --]


--- linux-2.6.0-test4.orig/drivers/acpi/pci_link.c	2003-08-23 00:52:08.000000000 +0100
+++ linux-2.6.0-test4/drivers/acpi/pci_link.c	2003-08-30 18:54:39.000000000 +0100
@@ -294,6 +294,7 @@
 	struct acpi_buffer	buffer = {sizeof(resource)+1, &resource};
 	int			i = 0;
 	int			valid = 0;
+	int			resource_type = 0;
 
 	ACPI_FUNCTION_TRACE("acpi_pci_link_set");
 
@@ -317,20 +318,32 @@
 		}
 	}
 
+	/* If IRQ<=15, first try with a "normal" IRQ descriptor. If that fails, try with
+	 * an extended one */
+	if (irq <= 15) {
+		resource_type = ACPI_RSTYPE_IRQ;
+	} else {
+		resource_type = ACPI_RSTYPE_EXT_IRQ;
+	}
+
+retry_programming:
+   
 	memset(&resource, 0, sizeof(resource));
 
 	/* NOTE: PCI interrupts are always level / active_low / shared. But not all
 	   interrupts > 15 are PCI interrupts. Rely on the ACPI IRQ definition for 
 	   parameters */
-	if (irq <= 15) {
+	switch(resource_type) {
+	case ACPI_RSTYPE_IRQ:
 		resource.res.id = ACPI_RSTYPE_IRQ;
 		resource.res.length = sizeof(struct acpi_resource);
 		resource.res.data.irq.edge_level = link->irq.edge_level;
 		resource.res.data.irq.active_high_low = link->irq.active_high_low;
 		resource.res.data.irq.number_of_interrupts = 1;
 		resource.res.data.irq.interrupts[0] = irq;
-	}
-	else {
+		break;
+	   
+	case ACPI_RSTYPE_EXT_IRQ:
 		resource.res.id = ACPI_RSTYPE_EXT_IRQ;
 		resource.res.length = sizeof(struct acpi_resource);
 		resource.res.data.extended_irq.producer_consumer = ACPI_CONSUMER;
@@ -339,11 +352,21 @@
 		resource.res.data.extended_irq.number_of_interrupts = 1;
 		resource.res.data.extended_irq.interrupts[0] = irq;
 		/* ignore resource_source, it's optional */
+		break;
 	}
 	resource.end.id = ACPI_RSTYPE_END_TAG;
 
 	/* Attempt to set the resource */
 	status = acpi_set_current_resources(link->handle, &buffer);
+
+	/* if we failed and IRQ <= 15, try again with an extended descriptor */
+	if (ACPI_FAILURE(status) && (resource_type == ACPI_RSTYPE_IRQ)) {
+                resource_type = ACPI_RSTYPE_EXT_IRQ;
+                printk(PREFIX "Retrying with extended IRQ descriptor\n");
+                goto retry_programming;
+	}
+  
+	/* check for total failure */
 	if (ACPI_FAILURE(status)) {
 		ACPI_DEBUG_PRINT((ACPI_DB_ERROR, "Error evaluating _SRS\n"));
 		return_VALUE(-ENODEV);
@@ -458,14 +481,14 @@
 		irq = link->irq.possible[0];
 	}
 
-		/* 
-		 * Select the best IRQ.  This is done in reverse to promote 
-		 * the use of IRQs 9, 10, 11, and >15.
-		 */
-		for (i=(link->irq.possible_count-1); i>0; i--) {
-			if (acpi_irq_penalty[irq] > acpi_irq_penalty[link->irq.possible[i]])
-				irq = link->irq.possible[i];
-		}
+	/* 
+	 * Select the best IRQ.  This is done in reverse to promote 
+	 * the use of IRQs 9, 10, 11, and >15.
+	 */
+	for (i=(link->irq.possible_count-1); i>0; i--) {
+		if (acpi_irq_penalty[irq] > acpi_irq_penalty[link->irq.possible[i]])
+			irq = link->irq.possible[i];
+	}
 
 	/* Attempt to enable the link device at this IRQ. */
 	if (acpi_pci_link_set(link, irq)) {

Re: [PATCH] Drop back to extended IRQ resource type

[John Wong]

The patch does appear to give the card an IRQ, but it does not appear to
be valid.  Doesn't the IRQ only go up to 24 on a single APIC processor?
The IRQ of 27 does not get registered in /proc/interrupts either.

           CPU0
  0:     401852          XT-PIC  timer
  1:       2680    IO-APIC-edge  i8042
  2:          0          XT-PIC  cascade
  8:          4    IO-APIC-edge  rtc
  9:          0   IO-APIC-level  acpi
 12:       3251    IO-APIC-edge  i8042
 14:       1096    IO-APIC-edge  ide0
 15:         11    IO-APIC-edge  ide1
 18:          0   IO-APIC-level  EMU10K1
 22:   17386129   IO-APIC-level  eth0
NMI:          0
LOC:     401701
ERR:          0
MIS:          0


With the patch applied (using 2.6.0-test4):

02:00.0 VGA compatible controller: nVidia Corporation NV15 [GeForce2 -
nForce GPU] (rev b1) (prog-if 00 [VGA])
        Subsystem: nVidia Corporation: Unknown device 0c11
        Flags: bus master, 66Mhz, medium devsel, latency 32, IRQ 27
        Memory at e5000000 (32-bit, non-prefetchable) [size=16M]
        Memory at e8000000 (32-bit, prefetchable) [size=128M]
        Expansion ROM at e7ff0000 [disabled] [size=64K]
        Capabilities: [60] Power Management version 2
        Capabilities: [44] AGP version 2.0

Without patch (2.4.22, also exhibited on 2.6.0-test4)
02:00.0 VGA compatible controller: nVidia Corporation NV15 [GeForce2 -
nForce GPU] (rev b1) (prog-if 00 [VGA])
        Subsystem: nVidia Corporation: Unknown device 0c11
        Flags: bus master, 66Mhz, medium devsel, latency 32, IRQ -19
        Memory at e5000000 (32-bit, non-prefetchable) [size=16M]
        Memory at e8000000 (32-bit, prefetchable) [size=128M]
        Expansion ROM at e7ff0000 [disabled] [size=64K]
        Capabilities: [60] Power Management version 2
        Capabilities: [44] AGP version 2.0


On Sat, Aug 30, 2003 at 07:56:41PM +0100, Andrew de Quincey wrote:
> On Friday 29 Aug 2003 5:50 am, John Wong wrote:
> > On the nForce original with the onboard GeForce2 video, the BIOS does
> > assign IRQ 11 to the onboard video, however, Linux does not.  Setting
> > acpi=off and pci=noirq does not fix the problem.  I have a nForce2 board
> > without onboard video and ACPI/IO-APIC etc all work fine.
> 
> Hi, the attached patch should resolve this issue. Originally, if an IRQ is 
> <=15, the code always used a standard IRQ descriptor when calling the _SRS 
> method. 
> 
> However, certain BIOSes always expect an extended IRQ descriptor, and fail 
> when passed a standard one. This patch drops back to an extended descriptor 
> if a set with a standard one fails... you'll still see the BUFFER LIMIT 
> message, but you should now see it saying "Retrying with extended IRQ 
> descriptor".. and hopefully succeeding the second time.

> 
> --- linux-2.4.22.orig/drivers/acpi/pci_link.c	2003-08-25 12:44:41.000000000 +0100
> +++ linux-2.4.22/drivers/acpi/pci_link.c	2003-08-30 18:54:35.000000000 +0100
> @@ -290,7 +290,8 @@
>  	struct acpi_buffer	buffer = {sizeof(resource)+1, &resource};
>  	int			i = 0;
>  	int			valid = 0;
> -
> +	int			resource_type = 0;
> +   
>  	ACPI_FUNCTION_TRACE("acpi_pci_link_set");
>  
>  	if (!link || !irq)
> @@ -313,12 +314,23 @@
>  		}
>  	}
>  
> +	/* If IRQ<=15, first try with a "normal" IRQ descriptor. If that fails, try with
> +	 * an extended one */
> +	if (irq <= 15) {
> +		resource_type = ACPI_RSTYPE_IRQ;
> +	} else {
> +		resource_type = ACPI_RSTYPE_EXT_IRQ;
> +	}
> +
> +retry_programming:
> +   
>  	memset(&resource, 0, sizeof(resource));
>  
>  	/* NOTE: PCI interrupts are always level / active_low / shared. But not all
>  	   interrupts > 15 are PCI interrupts. Rely on the ACPI IRQ definition for 
>  	   parameters */
> -	if (irq <= 15) {
> +	switch(resource_type) {
> +	case ACPI_RSTYPE_IRQ:
>  		resource.res.id = ACPI_RSTYPE_IRQ;
>  		resource.res.length = sizeof(struct acpi_resource);
>  		resource.res.data.irq.edge_level = link->irq.edge_level;
> @@ -326,8 +338,9 @@
>  		resource.res.data.irq.shared_exclusive = ACPI_SHARED;
>  		resource.res.data.irq.number_of_interrupts = 1;
>  		resource.res.data.irq.interrupts[0] = irq;
> -	}
> -	else {
> +		break;
> +	   
> +	case ACPI_RSTYPE_EXT_IRQ:
>  		resource.res.id = ACPI_RSTYPE_EXT_IRQ;
>  		resource.res.length = sizeof(struct acpi_resource);
>  		resource.res.data.extended_irq.producer_consumer = ACPI_CONSUMER;
> @@ -337,11 +350,22 @@
>  		resource.res.data.extended_irq.number_of_interrupts = 1;
>  		resource.res.data.extended_irq.interrupts[0] = irq;
>  		/* ignore resource_source, it's optional */
> +		break;
>  	}
>  	resource.end.id = ACPI_RSTYPE_END_TAG;
>  
>  	/* Attempt to set the resource */
>  	status = acpi_set_current_resources(link->handle, &buffer);
> +   
> +
> +	/* if we failed and IRQ <= 15, try again with an extended descriptor */
> +	if (ACPI_FAILURE(status) && (resource_type == ACPI_RSTYPE_IRQ)) {
> +                resource_type = ACPI_RSTYPE_EXT_IRQ;
> +                printk(PREFIX "Retrying with extended IRQ descriptor\n");
> +                goto retry_programming;
> +	}
> +  
> +	/* check for total failure */
>  	if (ACPI_FAILURE(status)) {
>  		ACPI_DEBUG_PRINT((ACPI_DB_ERROR, "Error evaluating _SRS\n"));
>  		return_VALUE(-ENODEV);

> 
> --- linux-2.6.0-test4.orig/drivers/acpi/pci_link.c	2003-08-23 00:52:08.000000000 +0100
> +++ linux-2.6.0-test4/drivers/acpi/pci_link.c	2003-08-30 18:54:39.000000000 +0100
> @@ -294,6 +294,7 @@
>  	struct acpi_buffer	buffer = {sizeof(resource)+1, &resource};
>  	int			i = 0;
>  	int			valid = 0;
> +	int			resource_type = 0;
>  
>  	ACPI_FUNCTION_TRACE("acpi_pci_link_set");
>  
> @@ -317,20 +318,32 @@
>  		}
>  	}
>  
> +	/* If IRQ<=15, first try with a "normal" IRQ descriptor. If that fails, try with
> +	 * an extended one */
> +	if (irq <= 15) {
> +		resource_type = ACPI_RSTYPE_IRQ;
> +	} else {
> +		resource_type = ACPI_RSTYPE_EXT_IRQ;
> +	}
> +
> +retry_programming:
> +   
>  	memset(&resource, 0, sizeof(resource));
>  
>  	/* NOTE: PCI interrupts are always level / active_low / shared. But not all
>  	   interrupts > 15 are PCI interrupts. Rely on the ACPI IRQ definition for 
>  	   parameters */
> -	if (irq <= 15) {
> +	switch(resource_type) {
> +	case ACPI_RSTYPE_IRQ:
>  		resource.res.id = ACPI_RSTYPE_IRQ;
>  		resource.res.length = sizeof(struct acpi_resource);
>  		resource.res.data.irq.edge_level = link->irq.edge_level;
>  		resource.res.data.irq.active_high_low = link->irq.active_high_low;
>  		resource.res.data.irq.number_of_interrupts = 1;
>  		resource.res.data.irq.interrupts[0] = irq;
> -	}
> -	else {
> +		break;
> +	   
> +	case ACPI_RSTYPE_EXT_IRQ:
>  		resource.res.id = ACPI_RSTYPE_EXT_IRQ;
>  		resource.res.length = sizeof(struct acpi_resource);
>  		resource.res.data.extended_irq.producer_consumer = ACPI_CONSUMER;
> @@ -339,11 +352,21 @@
>  		resource.res.data.extended_irq.number_of_interrupts = 1;
>  		resource.res.data.extended_irq.interrupts[0] = irq;
>  		/* ignore resource_source, it's optional */
> +		break;
>  	}
>  	resource.end.id = ACPI_RSTYPE_END_TAG;
>  
>  	/* Attempt to set the resource */
>  	status = acpi_set_current_resources(link->handle, &buffer);
> +
> +	/* if we failed and IRQ <= 15, try again with an extended descriptor */
> +	if (ACPI_FAILURE(status) && (resource_type == ACPI_RSTYPE_IRQ)) {
> +                resource_type = ACPI_RSTYPE_EXT_IRQ;
> +                printk(PREFIX "Retrying with extended IRQ descriptor\n");
> +                goto retry_programming;
> +	}
> +  
> +	/* check for total failure */
>  	if (ACPI_FAILURE(status)) {
>  		ACPI_DEBUG_PRINT((ACPI_DB_ERROR, "Error evaluating _SRS\n"));
>  		return_VALUE(-ENODEV);
> @@ -458,14 +481,14 @@
>  		irq = link->irq.possible[0];
>  	}
>  
> -		/* 
> -		 * Select the best IRQ.  This is done in reverse to promote 
> -		 * the use of IRQs 9, 10, 11, and >15.
> -		 */
> -		for (i=(link->irq.possible_count-1); i>0; i--) {
> -			if (acpi_irq_penalty[irq] > acpi_irq_penalty[link->irq.possible[i]])
> -				irq = link->irq.possible[i];
> -		}
> +	/* 
> +	 * Select the best IRQ.  This is done in reverse to promote 
> +	 * the use of IRQs 9, 10, 11, and >15.
> +	 */
> +	for (i=(link->irq.possible_count-1); i>0; i--) {
> +		if (acpi_irq_penalty[irq] > acpi_irq_penalty[link->irq.possible[i]])
> +			irq = link->irq.possible[i];
> +	}
>  
>  	/* Attempt to enable the link device at this IRQ. */
>  	if (acpi_pci_link_set(link, irq)) {



-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf

Re: [PATCH] Drop back to extended IRQ resource type

[Andrew de Quincey]

On Tuesday 02 Sep 2003 4:48 pm, John Wong wrote:
> The patch does appear to give the card an IRQ, but it does not appear to
> be valid.  Doesn't the IRQ only go up to 24 on a single APIC processor?
> The IRQ of 27 does not get registered in /proc/interrupts either.

No, theres no limit like that with APIC.. really depends on the hardware 
configuration in the machine (number and type of IO-APIC(s) used).

IRQs do not appear in /proc/interrupts unless a device driver claims it... I 
have the same thing with my Geforce4 card... I'm using the drivers that come 
with X.

Does the card work OK?



-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf

Re: [PATCH] Drop back to extended IRQ resource type

[Andrew de Quincey]

On Tuesday 02 Sep 2003 4:48 pm, John Wong wrote:
> The patch does appear to give the card an IRQ, but it does not appear to
> be valid.  Doesn't the IRQ only go up to 24 on a single APIC processor?
> The IRQ of 27 does not get registered in /proc/interrupts either.

Check the debug printout of the IO-APIC in your dmesg to see if IRQ 27 is 
valid... it should have lines saying its mapping IRQ 27 to  particular input 
on an IO-APIC.



-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf

Re: [PATCH] Drop back to extended IRQ resource type

[John Wong]

My card does work, but I'm am using the XFree86 nv driver since nVidia's
binary driver doesn't yet work with kernel 2.6.  I will try the 2.4.22
patch and the nVidia binary driver afterwards.

My dmesg does show a mapping to IRQ 27.

Pin 2-16 already programmed
    ACPI-1121: *** Error: Method execution failed [\_SB_.SRS_] (Node
cefdbca0), 
AE_AML_BUFFER_LIMIT
    ACPI-1121: *** Error: Method execution failed [\_SB_.LNKE._SRS]
(Node cef437
a0), AE_AML_BUFFER_LIMIT
ACPI: Retrying with extended IRQ descriptor
ACPI: PCI Interrupt Link [LNKE] enabled at IRQ 11
IOAPIC[0]: Set PCI routing entry (2-11 -> 0xd1 -> IRQ 27 Mode:1
Active:0)
00:02:00[A] -> 2-11 -> IRQ 27


On Tue, Sep 02, 2003 at 05:53:36PM +0100, Andrew de Quincey wrote:
> On Tuesday 02 Sep 2003 4:48 pm, John Wong wrote:
> > The patch does appear to give the card an IRQ, but it does not appear to
> > be valid.  Doesn't the IRQ only go up to 24 on a single APIC processor?
> > The IRQ of 27 does not get registered in /proc/interrupts either.
> 
> Check the debug printout of the IO-APIC in your dmesg to see if IRQ 27 is 
> valid... it should have lines saying its mapping IRQ 27 to  particular input 
> on an IO-APIC.
> 
> 


-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf

Re: [PATCH] Drop back to extended IRQ resource type

[Andrew de Quincey]

On Tuesday 02 Sep 2003 5:23 pm, John Wong wrote:
> My card does work, but I'm am using the XFree86 nv driver since nVidia's
> binary driver doesn't yet work with kernel 2.6.  I will try the 2.4.22
> patch and the nVidia binary driver afterwards.

Great thanks... There IS a third party patch for v2.5 (maybe usable in v2.6) 
for the nvidia driver somewhere... I've lost the URL unfortunately.

I'm glad to see the dropback thing works on other boards.. well it looks like 
it works anyway. :)



-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf