PCI ROM resource allocation issue with 2.6.17-rc2

PCI ROM resource allocation issue with 2.6.17-rc2

[Matthew Reppert]

I've been running 2.6.12-rc2-mm3 for a long time.  Recently I upgraded
a bunch of OS packages (Debian unstable), so I thought I may as well
upgrade the kernel, too.  I've got a dual-head setup driven by a Radeon
9200 and a Radeon 7000.  When I booted 2.6.17-rc2, X never came up; I
got "RADEON: Cannot read V_BIOS" and "RADEON: VIdeo BIOS not detected
in PCI space!" for the RADEON 7000, and it eventually gets in a loop of
spitting out "RADEON: Idle timed out, resetting engine ... " messages
in Xorg.log.  Doing a diff of working and broken logs uncovered that the
Radeon 7000's PCI ROM resource area had moved from ff8c000 to c6900000.
Once I removed the Radeon 7000 screen from the Xorg config, X came up fine
on the one head.  Adding stupid amounts of printks to the PCI subsystem in
.17-rc2 uncovered that at some point, the ROM area is discovered to be
at ff8c0000, but is later reallocated to c6900000.

I've also got a Promise PDC20268 whose expansion ROM seems to have made a
similar move (from ff8f8000 to c6920000), but the ATA devices attached to
that controller seem to work fine under 2.6.17-rc2.

I have a copy of relevant dmesg and lspci output, as well as a copy of
Xorg.log files, at http://sacredchao.net/~arashi/pci-problem/ .  I'll
try to binary-search for the last version of the kernel that works later
this week (hopefully by Tuesday afternoon), I just haven't had time to
since I've discovered the problem.

Matt

Re: PCI ROM resource allocation issue with 2.6.17-rc2

[Andrew Morton]

Matthew Reppert <arashi@sacredchao.net> wrote:
>
> I've been running 2.6.12-rc2-mm3 for a long time.  Recently I upgraded
> a bunch of OS packages (Debian unstable), so I thought I may as well
> upgrade the kernel, too.  I've got a dual-head setup driven by a Radeon
> 9200 and a Radeon 7000.  When I booted 2.6.17-rc2, X never came up; I
> got "RADEON: Cannot read V_BIOS" and "RADEON: VIdeo BIOS not detected
> in PCI space!" for the RADEON 7000, and it eventually gets in a loop of
> spitting out "RADEON: Idle timed out, resetting engine ... " messages
> in Xorg.log.  Doing a diff of working and broken logs uncovered that the
> Radeon 7000's PCI ROM resource area had moved from ff8c000 to c6900000.
> Once I removed the Radeon 7000 screen from the Xorg config, X came up fine
> on the one head.  Adding stupid amounts of printks to the PCI subsystem in
> .17-rc2 uncovered that at some point, the ROM area is discovered to be
> at ff8c0000, but is later reallocated to c6900000.
> 
> I've also got a Promise PDC20268 whose expansion ROM seems to have made a
> similar move (from ff8f8000 to c6920000), but the ATA devices attached to
> that controller seem to work fine under 2.6.17-rc2.
> 
> I have a copy of relevant dmesg and lspci output, as well as a copy of
> Xorg.log files, at http://sacredchao.net/~arashi/pci-problem/ .  I'll
> try to binary-search for the last version of the kernel that works later
> this week (hopefully by Tuesday afternoon), I just haven't had time to
> since I've discovered the problem.
> 

I'm pretty sure this has come up before, but I cannot for the life of me
remember what the answer was.

You did everything right though - thanks.  Here's
diff -u dmesg-working dmesg-broken:

--- dmesg-working	2006-04-23 22:17:45.000000000 -0700
+++ dmesg-broken	2006-04-23 22:17:23.000000000 -0700
@@ -1,4 +1,4 @@
-Linux version 2.6.12-rc2-mm3 (miz@minerva) (gcc version 4.0.3 (Debian 4.0.3-1)) #4 PREEMPT Sun Apr 23 13:43:42 EDT 2006
+Linux version 2.6.17-rc2 (miz@minerva) (gcc version 4.0.3 (Debian 4.0.3-1)) #1 PREEMPT Sun Apr 23 08:32:41 EDT 2006
 BIOS-provided physical RAM map:
  BIOS-e820: 0000000000000000 - 000000000009fc00 (usable)
  BIOS-e820: 000000000009fc00 - 00000000000a0000 (reserved)
@@ -10,29 +10,26 @@
  BIOS-e820: 00000000fff00000 - 0000000100000000 (reserved)
 511MB LOWMEM available.
 On node 0 totalpages: 131008
-  DMA zone: 4096 pages, LIFO batch:1
-  Normal zone: 126912 pages, LIFO batch:16
-  HighMem zone: 0 pages, LIFO batch:1
+  DMA zone: 4096 pages, LIFO batch:0
+  Normal zone: 126912 pages, LIFO batch:31
 DMI 2.3 present.
-ACPI: RSDP (v000 AMI                                   ) @ 0x000ff980
-ACPI: RSDT (v001 D815EA EA81510A 0x20000628 MSFT 0x00001011) @ 0x1fff0000
-ACPI: FADT (v001 D815EA EA81510A 0x20000628 MSFT 0x00001011) @ 0x1fff1000
-ACPI: DSDT (v001 D815EA EA81510A 0x00000012 MSFT 0x0100000b) @ 0x00000000
-Allocating PCI resources starting at 20000000 (gap: 20000000:dfb80000)
+Allocating PCI resources starting at 30000000 (gap: 20000000:dfb80000)
 Built 1 zonelists
+Kernel command line: root=/dev/hda1 read-only init=/bin/sh
 Local APIC disabled by BIOS -- you can enable it with "lapic"
-mapped APIC to ffffd000 (01404000)
+mapped APIC to ffffd000 (01401000)
+Enabling fast FPU save and restore... done.
+Enabling unmasked SIMD FPU exception support... done.
 Initializing CPU#0
-Kernel command line: root=/dev/hda1 read-only init=/bin/sh
-PID hash table entries: 2048 (order: 11, 32768 bytes)
-Detected 864.158 MHz processor.
+PID hash table entries: 2048 (order: 11, 8192 bytes)
+Detected 863.902 MHz processor.
 Using tsc for high-res timesource
 Console: colour VGA+ 80x25
-Dentry cache hash table entries: 131072 (order: 7, 524288 bytes)
-Inode-cache hash table entries: 65536 (order: 6, 262144 bytes)
-Memory: 513580k/524032k available (3597k kernel code, 9972k reserved, 1035k data, 188k init, 0k highmem)
+Dentry cache hash table entries: 65536 (order: 6, 262144 bytes)
+Inode-cache hash table entries: 32768 (order: 5, 131072 bytes)
+Memory: 515328k/524032k available (2078k kernel code, 8216k reserved, 1249k data, 144k init, 0k highmem)
 Checking if this processor honours the WP bit even in supervisor mode... Ok.
-Calibrating delay loop... 1708.03 BogoMIPS (lpj=854016)
+Calibrating delay using timer specific routine.. 1729.66 BogoMIPS (lpj=8648342)
 Mount-cache hash table entries: 512
 CPU: After generic identify, caps: 0383f9ff 00000000 00000000 00000000 00000000 00000000 00000000
 CPU: After vendor identify, caps: 0383f9ff 00000000 00000000 00000000 00000000 00000000 00000000
@@ -42,123 +39,24 @@
 Intel machine check architecture supported.
 Intel machine check reporting enabled on CPU#0.
 CPU: Intel Pentium III (Coppermine) stepping 06
-Enabling fast FPU save and restore... done.
-Enabling unmasked SIMD FPU exception support... done.
 Checking 'hlt' instruction... OK.
+SMP alternatives: switching to UP code
+Freeing SMP alternatives: 0k freed
 ACPI: setting ELCR to 0200 (from 0e00)
-softlockup thread 0 started up.
 NET: Registered protocol family 16
+ACPI: bus type pci registered
 PCI: PCI BIOS revision 2.10 entry at 0xfda95, last bus=2
-PCI: Using configuration type 1
-mtrr: v2.0 (20020519)
-ACPI: Subsystem revision 20050309
+Setting up standard PCI resources
+ACPI: Subsystem revision 20060127
 ACPI: Interpreter enabled
 ACPI: Using PIC for interrupt routing
 ACPI: PCI Root Bridge [PCI0] (0000:00)
 PCI: Probing PCI hardware (bus 00)
-ACPI: Assume root bridge [\_SB_.PCI0] segment is 0
-PCI: Scanning bus 0000:00
-PCI: Found 0000:00:00.0 [8086/1130] 000600 00
-PCI: Calling quirk c0277c70 for 0000:00:00.0
-PCI: Calling quirk c0278250 for 0000:00:00.0
-PCI: Calling quirk c03ff370 for 0000:00:00.0
-PCI: Calling quirk c03ff5a0 for 0000:00:00.0
-PCI: Calling quirk c03ff780 for 0000:00:00.0
-PCI: Found 0000:00:01.0 [8086/1131] 000604 01
-PCI: Calling quirk c0277c70 for 0000:00:01.0
-PCI: Calling quirk c0278250 for 0000:00:01.0
-PCI: Calling quirk c03ff370 for 0000:00:01.0
-PCI: Calling quirk c03ff5a0 for 0000:00:01.0
-PCI: Calling quirk c03ff780 for 0000:00:01.0
-PCI: Found 0000:00:1e.0 [8086/244e] 000604 01
-PCI: Calling quirk c0277c70 for 0000:00:1e.0
-PCI: Calling quirk c0278250 for 0000:00:1e.0
-PCI: Calling quirk c03ff370 for 0000:00:1e.0
-PCI: Calling quirk c03ff5a0 for 0000:00:1e.0
-PCI: Calling quirk c03ff780 for 0000:00:1e.0
-PCI: Found 0000:00:1f.0 [8086/2440] 000601 00
-PCI: Calling quirk c02776b0 for 0000:00:1f.0
-PCI: Calling quirk c0277c70 for 0000:00:1f.0
-PCI: Calling quirk c059edf0 for 0000:00:1f.0
-PCI: Calling quirk c0278250 for 0000:00:1f.0
-PCI: Calling quirk c03ff370 for 0000:00:1f.0
-PCI: Calling quirk c03ff5a0 for 0000:00:1f.0
-PCI: Calling quirk c03ff780 for 0000:00:1f.0
-PCI: Found 0000:00:1f.1 [8086/244b] 000101 00
-PCI: Calling quirk c0277c70 for 0000:00:1f.1
-PCI: Calling quirk c0278250 for 0000:00:1f.1
-PCI: Calling quirk c03ff370 for 0000:00:1f.1
-PCI: Calling quirk c03ff5a0 for 0000:00:1f.1
-PCI: Calling quirk c03ff780 for 0000:00:1f.1
-PCI: Found 0000:00:1f.2 [8086/2442] 000c03 00
-PCI: Calling quirk c0277c70 for 0000:00:1f.2
-PCI: Calling quirk c0278250 for 0000:00:1f.2
-PCI: Calling quirk c03ff370 for 0000:00:1f.2
-PCI: Calling quirk c03ff5a0 for 0000:00:1f.2
-PCI: Calling quirk c03ff780 for 0000:00:1f.2
-PCI: Found 0000:00:1f.3 [8086/2443] 000c05 00
-PCI: Calling quirk c0277c70 for 0000:00:1f.3
-PCI: Calling quirk c0278250 for 0000:00:1f.3
-PCI: Calling quirk c03ff370 for 0000:00:1f.3
-PCI: Calling quirk c03ff5a0 for 0000:00:1f.3
-PCI: Calling quirk c03ff780 for 0000:00:1f.3
-PCI: Found 0000:00:1f.4 [8086/2444] 000c03 00
-PCI: Calling quirk c0277c70 for 0000:00:1f.4
-PCI: Calling quirk c0278250 for 0000:00:1f.4
-PCI: Calling quirk c03ff370 for 0000:00:1f.4
-PCI: Calling quirk c03ff5a0 for 0000:00:1f.4
-PCI: Calling quirk c03ff780 for 0000:00:1f.4
-PCI: Fixups for bus 0000:00
-PCI: Scanning behind PCI bridge 0000:00:01.0, config 020200, pass 0
-PCI: Scanning bus 0000:02
-PCI: Found 0000:02:00.0 [1002/5960] 000300 00
-PCI: Calling quirk c0277c70 for 0000:02:00.0
-PCI: Calling quirk c0278250 for 0000:02:00.0
-PCI: Calling quirk c03ff370 for 0000:02:00.0
-PCI: Calling quirk c03ff780 for 0000:02:00.0
+PCI quirk: region 0400-047f claimed by ICH4 ACPI/GPIO/TCO
+PCI quirk: region 0500-053f claimed by ICH4 GPIO
 Boot video device is 0000:02:00.0
-PCI: Found 0000:02:00.1 [1002/5940] 000380 00
-PCI: Calling quirk c0277c70 for 0000:02:00.1
-PCI: Calling quirk c0278250 for 0000:02:00.1
-PCI: Calling quirk c03ff370 for 0000:02:00.1
-PCI: Calling quirk c03ff780 for 0000:02:00.1
-PCI: Fixups for bus 0000:02
-PCI: Bus scan for 0000:02 returning with max=02
-PCI: Scanning behind PCI bridge 0000:00:1e.0, config 010100, pass 0
-PCI: Scanning bus 0000:01
-PCI: Found 0000:01:0a.0 [105a/4d68] 000180 00
-PCI: Calling quirk c0277c70 for 0000:01:0a.0
-PCI: Calling quirk c0278250 for 0000:01:0a.0
-PCI: Calling quirk c03ff370 for 0000:01:0a.0
-PCI: Calling quirk c03ff780 for 0000:01:0a.0
-PCI: Found 0000:01:0b.0 [1073/0012] 000401 00
-PCI: Calling quirk c0277c70 for 0000:01:0b.0
-PCI: Calling quirk c0278250 for 0000:01:0b.0
-PCI: Calling quirk c03ff370 for 0000:01:0b.0
-PCI: Calling quirk c03ff780 for 0000:01:0b.0
-PCI: Found 0000:01:0c.0 [1002/5159] 000300 00
-PCI: Calling quirk c0277c70 for 0000:01:0c.0
-PCI: Calling quirk c0278250 for 0000:01:0c.0
-PCI: Calling quirk c03ff370 for 0000:01:0c.0
-PCI: Calling quirk c03ff780 for 0000:01:0c.0
-PCI: Found 0000:01:0d.0 [10ec/8139] 000200 00
-PCI: Calling quirk c0277c70 for 0000:01:0d.0
-PCI: Calling quirk c0278250 for 0000:01:0d.0
-PCI: Calling quirk c03ff370 for 0000:01:0d.0
-PCI: Calling quirk c03ff780 for 0000:01:0d.0
-PCI: Fixups for bus 0000:01
-PCI: Bus scan for 0000:01 returning with max=01
-PCI: Scanning behind PCI bridge 0000:00:01.0, config 020200, pass 1
-PCI: Scanning behind PCI bridge 0000:00:1e.0, config 010100, pass 1
-PCI: Bus scan for 0000:00 returning with max=02
+PCI: Transparent bridge - 0000:00:1e.0
 ACPI: PCI Interrupt Routing Table [\_SB_.PCI0._PRT]
-ACPI: Can't get handler for 0000:00:01.0
-ACPI: Can't get handler for 0000:02:00.0
-ACPI: Can't get handler for 0000:02:00.1
-ACPI: Can't get handler for 0000:01:0a.0
-ACPI: Can't get handler for 0000:01:0b.0
-ACPI: Can't get handler for 0000:01:0c.0
-ACPI: Can't get handler for 0000:01:0d.0
 ACPI: PCI Interrupt Routing Table [\_SB_.PCI0.PCI1._PRT]
 ACPI: Power Resource [URP1] (off)
 ACPI: Power Resource [URP2] (off)
@@ -177,69 +75,42 @@
 usbcore: registered new driver hub
 PCI: Using ACPI for IRQ routing
 PCI: If a device doesn't work, try "pci=routeirq".  If it helps, post a report
-audit: initializing netlink socket (disabled)
-audit(1145849409.148:0): initialized
-inotify device minor=63
+PCI: Bridge: 0000:00:01.0
+  IO window: d000-dfff
+  MEM window: ff900000-ff9fffff
+  PREFETCH window: d6a00000-f6afffff
+PCI: Bridge: 0000:00:1e.0
+  IO window: c000-cfff
+  MEM window: ff800000-ff8fffff
+  PREFETCH window: c6900000-d69fffff
+PCI: Setting latency timer of device 0000:00:1e.0 to 64
 Initializing Cryptographic API
-PCI: Calling quirk c0277b30 for 0000:00:00.0
-PCI: Calling quirk c02782a0 for 0000:00:00.0
-PCI: Calling quirk c0277b30 for 0000:00:01.0
-PCI: Calling quirk c02782a0 for 0000:00:01.0
-PCI: Calling quirk c0277b30 for 0000:00:1e.0
-PCI: Calling quirk c02782a0 for 0000:00:1e.0
-PCI: Calling quirk c0277b30 for 0000:00:1f.0
-PCI: Calling quirk c02782a0 for 0000:00:1f.0
-PCI: Calling quirk c0277b30 for 0000:00:1f.1
-PCI: Calling quirk c02782a0 for 0000:00:1f.1
-PCI: Calling quirk c0277b30 for 0000:00:1f.2
-PCI: Calling quirk c02782a0 for 0000:00:1f.2
-PCI: Calling quirk c0277b30 for 0000:00:1f.3
-PCI: Calling quirk c02782a0 for 0000:00:1f.3
-PCI: Calling quirk c0277b30 for 0000:00:1f.4
-PCI: Calling quirk c02782a0 for 0000:00:1f.4
-PCI: Calling quirk c0277b30 for 0000:02:00.0
-PCI: Calling quirk c0277b30 for 0000:02:00.1
-PCI: Calling quirk c0277b30 for 0000:01:0a.0
-PCI: Calling quirk c0277b30 for 0000:01:0b.0
-PCI: Calling quirk c0277b30 for 0000:01:0c.0
-PCI: Calling quirk c0277b30 for 0000:01:0d.0
+io scheduler noop registered
+io scheduler cfq registered (default)
 ACPI: Power Button (FF) [PWRF]
 ACPI: Sleep Button (CM) [SBTN]
-ACPI: CPU0 (power states: C1[C1])
-Real Time Clock Driver v1.12
+Real Time Clock Driver v1.12ac
+Non-volatile memory driver v1.2
 hw_random hardware driver 1.0.0 loaded
 Linux agpgart interface v0.101 (c) Dave Jones
 agpgart: Detected an Intel i815 Chipset.
 agpgart: AGP aperture is 64M @ 0xf8000000
-[drm] Initialized drm 1.0.0 20040925
+[drm] Initialized drm 1.0.1 20051102
 PCI: Enabling device 0000:01:0c.0 (0080 -> 0083)
 ACPI: PCI Interrupt Link [LNKB] enabled at IRQ 11
 PCI: setting IRQ 11 as level-triggered
 ACPI: PCI Interrupt 0000:01:0c.0[A] -> Link [LNKB] -> GSI 11 (level, low) -> IRQ 11
-[drm] Initialized radeon 1.16.0 20050311 on minor 0: ATI Technologies Inc Radeon RV100 QY [Radeon 7000/VE]
+[drm] Initialized radeon 1.24.0 20060225 on minor 0
 ACPI: PCI Interrupt Link [LNKA] enabled at IRQ 11
 ACPI: PCI Interrupt 0000:02:00.0[A] -> Link [LNKA] -> GSI 11 (level, low) -> IRQ 11
-[drm] Initialized radeon 1.16.0 20050311 on minor 1: ATI Technologies Inc RV280 [Radeon 9200 PRO]
-ACPI: No ACPI bus support for i8042
-serio: i8042 AUX port at 0x60,0x64 irq 12
-serio: i8042 KBD port at 0x60,0x64 irq 1
-Serial: 8250/16550 driver $Revision: 1.90 $ 8 ports, IRQ sharing disabled
-ACPI: No ACPI bus support for serial8250
-ttyS0 at I/O 0x3f8 (irq = 4) is a 16550A
-ttyS1 at I/O 0x2f8 (irq = 3) is a 16550A
-io scheduler noop registered
-io scheduler anticipatory registered
-io scheduler deadline registered
-io scheduler cfq registered
-floppy: ignoring I/O port region 0x3f4-0x3f5
-floppy: ignoring I/O port region 0x3f7-0x3f7
-ACPI: Floppy Controller [FDC0] at I/O 0x3f0-0x3f1, 0x3f2-0x3f3 irq 6 dma channel 2
-ACPI: [FDC0] doesn't declare FD_DCR; also claiming 0x3f7
-Floppy drive(s): fd0 is 1.44M
-ACPI: No ACPI bus support for serio0
-ACPI: No ACPI bus support for serio1
-FDC 0 is a post-1991 82077
-ACPI: No ACPI bus support for floppy.0
+[drm] Initialized radeon 1.24.0 20060225 on minor 1
+Serial: 8250/16550 driver $Revision: 1.90 $ 4 ports, IRQ sharing disabled
+serial8250: ttyS0 at I/O 0x3f8 (irq = 4) is a 16550A
+serial8250: ttyS1 at I/O 0x2f8 (irq = 3) is a 16550A
+parport0: PC-style at 0x378 [PCSPP,TRISTATE,EPP]
+loop: loaded (max 8 devices)
+e100: Intel(R) PRO/100 Network Driver, 3.5.10-k2-NAPI
+e100: Copyright(c) 1999-2005 Intel Corporation
 8139too Fast Ethernet driver 0.9.27
 ACPI: PCI Interrupt Link [LNKF] enabled at IRQ 11
 ACPI: PCI Interrupt 0000:01:0d.0[A] -> Link [LNKF] -> GSI 11 (level, low) -> IRQ 11
@@ -256,20 +127,16 @@
 hda: WDC WD300BB-00CCB0, ATA DISK drive
 hdb: ST380011A, ATA DISK drive
 ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
-ACPI: No ACPI bus support for 0.0
-ACPI: No ACPI bus support for 0.1
 Probing IDE interface ide1...
 hdc: _NEC DVD_RW ND-3520A, ATAPI CD/DVD-ROM drive
 hdd: TOSHIBA DVD-ROM SD-M1612, ATAPI CD/DVD-ROM drive
 ide1 at 0x170-0x177,0x376 on irq 15
-ACPI: No ACPI bus support for 1.0
-ACPI: No ACPI bus support for 1.1
 PDC20268: IDE controller at PCI slot 0000:01:0a.0
 ACPI: PCI Interrupt Link [LNKG] enabled at IRQ 9
 PCI: setting IRQ 9 as level-triggered
 ACPI: PCI Interrupt 0000:01:0a.0[A] -> Link [LNKG] -> GSI 9 (level, low) -> IRQ 9
 PDC20268: chipset revision 2
-PDC20268: ROM enabled at 0xff8f8000
+PDC20268: ROM enabled at 0xc6920000
 PDC20268: 100% native mode on irq 9
     ide2: BM-DMA at 0xcf90-0xcf97, BIOS settings: hde:pio, hdf:pio
     ide3: BM-DMA at 0xcf98-0xcf9f, BIOS settings: hdg:pio, hdh:pio
@@ -278,16 +145,11 @@
 hdg: Maxtor 4D060H3, ATA DISK drive
 hdh: Maxtor 6Y060L0, ATA DISK drive
 ide3 at 0xcfa8-0xcfaf,0xcfe2 on irq 9
-ACPI: No ACPI bus support for 3.0
-ACPI: No ACPI bus support for 3.1
-Probing IDE interface ide2...
-Probing IDE interface ide4...
-Probing IDE interface ide5...
 hda: max request size: 128KiB
 hda: 58633344 sectors (30020 MB) w/2048KiB Cache, CHS=58168/16/63, UDMA(100)
 hda: cache flushes not supported
  hda: hda1 hda2
-hdb: max request size: 1024KiB
+hdb: max request size: 512KiB
 hdb: 156301488 sectors (80026 MB) w/2048KiB Cache, CHS=16383/255/63, UDMA(100)
 hdb: cache flushes supported
  hdb: hdb1 hdb2 hdb3 hdb4
@@ -302,72 +164,82 @@
 hdc: ATAPI 48X DVD-ROM DVD-R CD-R/RW drive, 2048kB Cache, UDMA(33)
 Uniform CD-ROM driver Revision: 3.20
 hdd: ATAPI 48X DVD-ROM drive, 512kB Cache, UDMA(33)
-USB Universal Host Controller Interface driver v2.2
+USB Universal Host Controller Interface driver v3.0
 ACPI: PCI Interrupt Link [LNKD] enabled at IRQ 11
 ACPI: PCI Interrupt 0000:00:1f.2[D] -> Link [LNKD] -> GSI 11 (level, low) -> IRQ 11
 PCI: Setting latency timer of device 0000:00:1f.2 to 64
-uhci_hcd 0000:00:1f.2: Intel Corporation 82801BA/BAM USB (Hub #1)
+uhci_hcd 0000:00:1f.2: UHCI Host Controller
 uhci_hcd 0000:00:1f.2: new USB bus registered, assigned bus number 1
 uhci_hcd 0000:00:1f.2: irq 11, io base 0x0000ef40
-ACPI: No ACPI bus support for usb1
+usb usb1: configuration #1 chosen from 1 choice
 hub 1-0:1.0: USB hub found
 hub 1-0:1.0: 2 ports detected
-ACPI: No ACPI bus support for 1-0:1.0
 ACPI: PCI Interrupt Link [LNKH] enabled at IRQ 10
 PCI: setting IRQ 10 as level-triggered
 ACPI: PCI Interrupt 0000:00:1f.4[C] -> Link [LNKH] -> GSI 10 (level, low) -> IRQ 10
 PCI: Setting latency timer of device 0000:00:1f.4 to 64
-uhci_hcd 0000:00:1f.4: Intel Corporation 82801BA/BAM USB (Hub #2)
+uhci_hcd 0000:00:1f.4: UHCI Host Controller
 uhci_hcd 0000:00:1f.4: new USB bus registered, assigned bus number 2
 uhci_hcd 0000:00:1f.4: irq 10, io base 0x0000ef80
-ACPI: No ACPI bus support for usb2
+usb usb2: configuration #1 chosen from 1 choice
 hub 2-0:1.0: USB hub found
 hub 2-0:1.0: 2 ports detected
-ACPI: No ACPI bus support for 2-0:1.0
+Initializing USB Mass Storage driver...
 usb 1-1: new low speed USB device using uhci_hcd and address 2
-ACPI: No ACPI bus support for 1-1
-ACPI: No ACPI bus support for 1-1:1.0
-ACPI: No ACPI bus support for 1-1:1.1
+usb 1-1: configuration #1 chosen from 1 choice
 usb 1-2: new full speed USB device using uhci_hcd and address 3
-ACPI: No ACPI bus support for 1-2
+usb 1-2: configuration #1 chosen from 1 choice
 hub 1-2:1.0: USB hub found
 hub 1-2:1.0: 4 ports detected
-ACPI: No ACPI bus support for 1-2:1.0
-usbcore: registered new driver usblp
-drivers/usb/class/usblp.c: v0.13: USB Printer Device Class driver
-Initializing USB Mass Storage driver...
+usb 1-2.1: new low speed USB device using uhci_hcd and address 4
+usb 1-2.1: configuration #1 chosen from 1 choice
+usb 1-2.3: new low speed USB device using uhci_hcd and address 5
+usb 1-2.3: configuration #1 chosen from 1 choice
 usbcore: registered new driver usb-storage
 USB Mass Storage support registered.
 usbcore: registered new driver hiddev
-usb 1-2.1: new low speed USB device using uhci_hcd and address 4
-ACPI: No ACPI bus support for 1-2.1
-ACPI: No ACPI bus support for 1-2.1:1.0
-usb 1-2.3: new low speed USB device using uhci_hcd and address 5
-ACPI: No ACPI bus support for 1-2.3
-ACPI: No ACPI bus support for 1-2.3:1.0
+input: Logitech USB Receiver as /class/input/input0
 input: USB HID v1.10 Keyboard [Logitech USB Receiver] on usb-0000:00:1f.2-1
+input: Logitech USB Receiver as /class/input/input1
 input: USB HID v1.10 Mouse [Logitech USB Receiver] on usb-0000:00:1f.2-1
+input: Logitech USB-PS/2 Optical Mouse as /class/input/input2
 input: USB HID v1.10 Mouse [Logitech USB-PS/2 Optical Mouse] on usb-0000:00:1f.2-2.1
-input,hiddev96: USB HID v1.00 Mouse [WACOM CTE-440-U V4.0-3] on usb-0000:00:1f.2-2.3
 usbcore: registered new driver usbhid
-drivers/usb/input/hid-core.c: v2.01:USB HID core driver
+drivers/usb/input/hid-core.c: v2.6:USB HID core driver
+serio: i8042 AUX port at 0x60,0x64 irq 12
+serio: i8042 KBD port at 0x60,0x64 irq 1
 mice: PS/2 mouse device common for all mice
-Advanced Linux Sound Architecture Driver Version 1.0.9rc2  (Thu Mar 24 10:33:39 2005 UTC).
+Advanced Linux Sound Architecture Driver Version 1.0.11rc4 (Wed Mar 22 10:27:24 2006 UTC).
 ALSA device list:
   No soundcards found.
-oprofile: using timer interrupt.
 NET: Registered protocol family 2
-IP: routing cache hash table of 1024 buckets, 32Kbytes
-TCP established hash table entries: 32768 (order: 6, 262144 bytes)
-TCP bind hash table entries: 32768 (order: 7, 917504 bytes)
-TCP: Hash tables configured (established 32768 bind 32768)
+IP route cache hash table entries: 4096 (order: 2, 16384 bytes)
+TCP established hash table entries: 16384 (order: 4, 65536 bytes)
+TCP bind hash table entries: 8192 (order: 3, 32768 bytes)
+TCP: Hash tables configured (established 16384 bind 8192)
+TCP reno registered
+TCP bic registered
+Initializing IPsec netlink socket
 NET: Registered protocol family 1
+NET: Registered protocol family 10
+IPv6 over IPv4 tunneling driver
 NET: Registered protocol family 17
+NET: Registered protocol family 15
+Using IPI Shortcut mode
+kjournald starting.  Commit interval 5 seconds
 EXT3-fs: mounted filesystem with ordered data mode.
 VFS: Mounted root (ext3 filesystem) readonly.
-Freeing unused kernel memory: 188k freed
-kjournald starting.  Commit interval 5 seconds
-input: AT Translated Set 2 keyboard on isa0060/serio0
+Freeing unused kernel memory: 144k freed
+input: AT Translated Set 2 keyboard as /class/input/input3
+EXT3 FS on hda1, internal journal
+Adding 1060280k swap on /dev/hda2.  Priority:-1 extents:1 across:1060280k
+EXT3 FS on hda1, internal journal
 kjournald starting.  Commit interval 5 seconds
 EXT3 FS on hdb2, internal journal
 EXT3-fs: mounted filesystem with ordered data mode.
+ACPI: PCI Interrupt 0000:01:0b.0[A] -> Link [LNKH] -> GSI 10 (level, low) -> IRQ 10
+input: Wacom Graphire4 4x5 as /class/input/input4
+usbcore: registered new driver wacom
+drivers/usb/input/wacom.c: v1.45:USB Wacom Graphire and Wacom Intuos tablet driver
+eth0: link up, 100Mbps, full-duplex, lpa 0x45E1
+eth0: no IPv6 routers present

Re: PCI ROM resource allocation issue with 2.6.17-rc2

[Matthew Reppert]

A bit more information.

On Mon, 2006-04-24 at 00:02 -0400, Matthew Reppert wrote:
> I've been running 2.6.12-rc2-mm3 for a long time.  Recently I upgraded
> a bunch of OS packages (Debian unstable), so I thought I may as well
> upgrade the kernel, too.  I've got a dual-head setup driven by a Radeon
> 9200 and a Radeon 7000.  When I booted 2.6.17-rc2, X never came up; I
> got "RADEON: Cannot read V_BIOS" and "RADEON: VIdeo BIOS not detected
> in PCI space!" for the RADEON 7000, and it eventually gets in a loop of
> spitting out "RADEON: Idle timed out, resetting engine ... " messages
> in Xorg.log.  Doing a diff of working and broken logs uncovered that the
> Radeon 7000's PCI ROM resource area had moved from ff8c000 to c6900000.
> Once I removed the Radeon 7000 screen from the Xorg config, X came up fine
> on the one head.  Adding stupid amounts of printks to the PCI subsystem in
> .17-rc2 uncovered that at some point, the ROM area is discovered to be
> at ff8c0000, but is later reallocated to c6900000.
> 
> I've also got a Promise PDC20268 whose expansion ROM seems to have made a
> similar move (from ff8f8000 to c6920000), but the ATA devices attached to
> that controller seem to work fine under 2.6.17-rc2.

Also, on 2.6.17-rc2, if I do a hexdump of the PCI config space for the
RADEON 7000 via sysfs once Linux boots, it still says the ROM is located
at ff8c0000, even though I get this message during boot:

PCI: pbus will assign resource 0000:01:0c.0
PCI: assigning resource #6 for 0000:01:0c.0 (start 0)
  got res [c6900000:c691ffff] bus [c6900000:c691ffff] flags 7200 for BAR 6 of
0000:01:0c.0

The first two lines of that bit from dmesg are from extra printks I put
in the for loop at the end of pbus_assign_resources_sorted and at the
top of pci_assign_resource.

The Promise controller's PCI config space says it's at c6920000.


> I have a copy of relevant dmesg and lspci output, as well as a copy of
> Xorg.log files, at http://sacredchao.net/~arashi/pci-problem/ .  I'll
> try to binary-search for the last version of the kernel that works later
> this week (hopefully by Tuesday afternoon), I just haven't had time to
> since I've discovered the problem.
> 
> Matt

Re: PCI ROM resource allocation issue with 2.6.17-rc2

[Dave Airlie]

>> I've been running 2.6.12-rc2-mm3 for a long time.  Recently I upgraded
>> a bunch of OS packages (Debian unstable), so I thought I may as well
>> upgrade the kernel, too.  I've got a dual-head setup driven by a Radeon
>> 9200 and a Radeon 7000.  When I booted 2.6.17-rc2, X never came up; I
>> got "RADEON: Cannot read V_BIOS" and "RADEON: VIdeo BIOS not detected
>> in PCI space!" for the RADEON 7000, and it eventually gets in a loop of
>> spitting out "RADEON: Idle timed out, resetting engine ... " messages
>> in Xorg.log.  Doing a diff of working and broken logs uncovered that the
>> Radeon 7000's PCI ROM resource area had moved from ff8c000 to c6900000.
>> Once I removed the Radeon 7000 screen from the Xorg config, X came up fine
>> on the one head.  Adding stupid amounts of printks to the PCI subsystem in
>> .17-rc2 uncovered that at some point, the ROM area is discovered to be
>> at ff8c0000, but is later reallocated to c6900000.
>>

welcome to X have a nice day :-)

X is horribly horribly broken in this area, if you remove the 
pci_enable_device from drivers/char/drm/drm_stub.c

and restart everything does it still happen?

The problem is that X uses the fact that the pci bars are disabled to 
decide whether to POST the card using the BIOS, however the card isn't 
actually posted but pci_enable_device enables the BARs...

however not doing pci_enable_device causes interrupts to not work on the 
cards in a lot of circumstances..

Dave.

-- 
David Airlie, Software Engineer
http://www.skynet.ie/~airlied / airlied at skynet.ie
Linux kernel - DRI, VAX / pam_smb / ILUG

Re: PCI ROM resource allocation issue with 2.6.17-rc2

[Linus Torvalds]

On Sun, 23 Apr 2006, Andrew Morton wrote:
>
> Matthew Reppert <arashi@sacredchao.net> wrote:
> >
> > I've been running 2.6.12-rc2-mm3 for a long time.  Recently I upgraded
> > a bunch of OS packages (Debian unstable), so I thought I may as well
> > upgrade the kernel, too.  I've got a dual-head setup driven by a Radeon
> > 9200 and a Radeon 7000.  When I booted 2.6.17-rc2, X never came up; I
> > got "RADEON: Cannot read V_BIOS" and "RADEON: VIdeo BIOS not detected
> > in PCI space!" for the RADEON 7000, and it eventually gets in a loop of
> > spitting out "RADEON: Idle timed out, resetting engine ... " messages
> > in Xorg.log.  Doing a diff of working and broken logs uncovered that the
> > Radeon 7000's PCI ROM resource area had moved from ff8c000 to c6900000.
> > Once I removed the Radeon 7000 screen from the Xorg config, X came up fine
> > on the one head.  Adding stupid amounts of printks to the PCI subsystem in
> > .17-rc2 uncovered that at some point, the ROM area is discovered to be
> > at ff8c0000, but is later reallocated to c6900000.

The relocation sounds strange (and isn't visible in your dmesg that I can 
see). But it _should_ be perfectly fine: it's moving the ROM from the 
non-prefetchable region of the 00:1e.0 bridge into the prefetchable one. 

I don't see quite why it would do it (yes, ROM's are prefetchable, but the 
old location was _valid_, and I don't see why we didn't re-use it), but it 
_really_ shouldn't matter. 

I would expect some silly interaction with X, as usual. It would be nice 
to see the whole dmesg, especially with the debugging messages - I suspect 
the remapping of the ROM happens only when X starts up, and that your 
dmesg is from just the kernel boot from before that?

But it might be that I just missed it (or that we don't have good debug 
output for that case).

> > I've also got a Promise PDC20268 whose expansion ROM seems to have made a
> > similar move (from ff8f8000 to c6920000), but the ATA devices attached to
> > that controller seem to work fine under 2.6.17-rc2.

Exactly the same deal. It's moved from non-prefetchable to prefetchable.

That said, the PDC202xx driver doesn't even _use_ the ROM resource, so I 
don't see why it bothers to enable it.

You can try this stupid patch to see if it moves the ROM resource back 
into the non-prefetchable region. Maybe it matters even if it shouldn't.

		Linus
---
diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c
index a10ed9d..7af4610 100644
--- a/drivers/pci/probe.c
+++ b/drivers/pci/probe.c
@@ -217,7 +217,7 @@ #endif
 			sz = pci_size(l, sz, (u32)PCI_ROM_ADDRESS_MASK);
 			if (sz) {
 				res->flags = (l & IORESOURCE_ROM_ENABLE) |
-				  IORESOURCE_MEM | IORESOURCE_PREFETCH |
+				  IORESOURCE_MEM | /* IORESOURCE_PREFETCH | */
 				  IORESOURCE_READONLY | IORESOURCE_CACHEABLE;
 				res->start = l & PCI_ROM_ADDRESS_MASK;
 				res->end = res->start + (unsigned long) sz;

Re: PCI ROM resource allocation issue with 2.6.17-rc2

[Linus Torvalds]

On Mon, 24 Apr 2006, Dave Airlie wrote:
> 
> however not doing pci_enable_device causes interrupts to not work on the cards
> in a lot of circumstances..

Well, you could use "pci_enable_device_bars(0)" instead.

That will set up interrupt routing _without_ enabling any BAR's, however, 
that's probably crazy too, since that means that if an interrupt happens, 
you're really really screwed and can't do anything about it. So that's 
probably even more broken than what you do now.

How about delaying the "pci_enable_device()" until when you actually need 
it, ie do it in drm_irq_install() instead?

Of course, I wonder how you could POST the device without having the BAR's 
enabled, and I sure as hell wouldn't want the POST sequence to decide 
where to put them, since it has no clue what is allocated..

Sounds like X is being really bogus again.

			Linus

Re: PCI ROM resource allocation issue with 2.6.17-rc2

[Arjan van de Ven]

On Mon, 2006-04-24 at 10:07 -0700, Linus Torvalds wrote:
> 
> On Mon, 24 Apr 2006, Dave Airlie wrote:
> > 
> > however not doing pci_enable_device causes interrupts to not work on the cards
> > in a lot of circumstances..
> 
> Well, you could use "pci_enable_device_bars(0)" instead.
> 
> That will set up interrupt routing _without_ enabling any BAR's, however, 
> that's probably crazy too, since that means that if an interrupt happens, 
> you're really really screwed and can't do anything about it. So that's 
> probably even more broken than what you do now.
> 
> How about delaying the "pci_enable_device()" until when you actually need 
> it, ie do it in drm_irq_install() instead?
> 
> Of course, I wonder how you could POST the device without having the BAR's 
> enabled, 

you haven't spent enough time reading the X pci code then ;)
(or rather, you've done the same thing but hey who's counting)

X does all that *itself* based on what X thinks is best.

Yes that's silly and X should be taken out and shot for that.
What's worse, this is the kind of thing that is really hard to work
around in a away that isn't going to make having a fixed X work as
well... you can't not enable it for old X and enable it for not-insane X
at the same time ;)

Re: PCI ROM resource allocation issue with 2.6.17-rc2

[Linus Torvalds]

On Mon, 24 Apr 2006, Arjan van de Ven wrote:
> 
> you haven't spent enough time reading the X pci code then ;)
> (or rather, you've done the same thing but hey who's counting)
> 
> X does all that *itself* based on what X thinks is best.

Yeah, I knew that used to be true, but I was hoping the new interfaces 
would have made it obsolete. Especially as the DRM layer _does_ now enable 
the device on demand.

Maybe just add a DRM command to do it, so that old X versions (who don't 
know about it) will just do it by hand, and then new X versions can do

	if (drm_ioctl(fd, DRM_SETUP_THE_DAMN_RESOURCES) < 0) {
		/*
		 * I don't know what errno the drm-ioctl actually
		 * returns for unrecognized commands, so this is
		 * just an example
		 */
		if (errno == ENOTTY) {
			old kernel: do it by hand
		}
	}

which allows us to go forward in a sane way, and finally leave the broken 
X PCI-configuration-by-hand crap behind.

Please?

			Linus

Re: PCI ROM resource allocation issue with 2.6.17-rc2

[Jon Smirl]

On 4/24/06, Matthew Reppert <arashi@sacredchao.net> wrote:
> > I've also got a Promise PDC20268 whose expansion ROM seems to have made a
> > similar move (from ff8f8000 to c6920000), but the ATA devices attached to
> > that controller seem to work fine under 2.6.17-rc2.
>
> Also, on 2.6.17-rc2, if I do a hexdump of the PCI config space for the
> RADEON 7000 via sysfs once Linux boots, it still says the ROM is located
> at ff8c0000, even though I get this message during boot:
>
> PCI: pbus will assign resource 0000:01:0c.0
> PCI: assigning resource #6 for 0000:01:0c.0 (start 0)
>   got res [c6900000:c691ffff] bus [c6900000:c691ffff] flags 7200 for BAR 6 of
> 0000:01:0c.0

To make the sysfs rom attribute work, "echo 1 >rom". Then use 'hexdump
-C rom | more' to see the ROM contents. If you get the video ROM
contents when you do this, then the ROM is where the kernel thinks it
is. If you get FFFF or some other ROM then something like X has moved
the ROM without the kernel's knowledge.  Obviously, moving ROMs
without telling the kernel is a good way to mess up your system.

If your system locks up after "echo 1 >rom" on a disk controller, then
you have one of the few disk controllers that didn't bother to
implement full address decoding for the ROM. "echo 1 >rom". should
always work for video ROMs.

Read http://people.freedesktop.org/~jonsmirl/graphics.html if you want
to know more about the evils of X and it's use of the PCI bus.

--
Jon Smirl
jonsmirl@gmail.com

Re: PCI ROM resource allocation issue with 2.6.17-rc2

[Dave Airlie]

>
> Maybe just add a DRM command to do it, so that old X versions (who don't
> know about it) will just do it by hand, and then new X versions can do
>
> 	if (drm_ioctl(fd, DRM_SETUP_THE_DAMN_RESOURCES) < 0) {
> 		/*
> 		 * I don't know what errno the drm-ioctl actually
> 		 * returns for unrecognized commands, so this is
> 		 * just an example
> 		 */
> 		if (errno == ENOTTY) {
> 			old kernel: do it by hand
> 		}
> 	}
>
> which allows us to go forward in a sane way, and finally leave the broken
> X PCI-configuration-by-hand crap behind.

It doesn't help of course, the fb drivers also pci_enable the devices, 
really X needs a kicking square, I'm trying to figure out some sort of fix 
here, but X does't some really stupid things with PCI resources...

We really need a userspace way to pci_enable_device that X can call (via 
sysfs) so for cards that don't have a DRM or fb loaded we still get 
something..

Dave.

-- 
David Airlie, Software Engineer
http://www.skynet.ie/~airlied / airlied at skynet.ie
Linux kernel - DRI, VAX / pam_smb / ILUG

Re: PCI ROM resource allocation issue with 2.6.17-rc2

[Jon Smirl]

On 4/25/06, Dave Airlie <airlied@linux.ie> wrote:
>
> >
> > Maybe just add a DRM command to do it, so that old X versions (who don't
> > know about it) will just do it by hand, and then new X versions can do
> >
> >       if (drm_ioctl(fd, DRM_SETUP_THE_DAMN_RESOURCES) < 0) {
> >               /*
> >                * I don't know what errno the drm-ioctl actually
> >                * returns for unrecognized commands, so this is
> >                * just an example
> >                */
> >               if (errno == ENOTTY) {
> >                       old kernel: do it by hand
> >               }
> >       }
> >
> > which allows us to go forward in a sane way, and finally leave the broken
> > X PCI-configuration-by-hand crap behind.
>
> It doesn't help of course, the fb drivers also pci_enable the devices,
> really X needs a kicking square, I'm trying to figure out some sort of fix
> here, but X does't some really stupid things with PCI resources...
>
> We really need a userspace way to pci_enable_device that X can call (via
> sysfs) so for cards that don't have a DRM or fb loaded we still get
> something..

You could make a null DRM driver that is loaded for every card that
doesn't have a real one. Give it aliases to make it match the X driver
names.

The right answer here is to start working towards a solution where the
OS is actually in control of hardware resources instead of a user app.
There can only be one entity in charge of PCI space or we will all go
insane. If we continue to say that old X binaries have to work we will
still have these same problems in 2060.


>
> Dave.
>
> --
> David Airlie, Software Engineer
> http://www.skynet.ie/~airlied / airlied at skynet.ie
> Linux kernel - DRI, VAX / pam_smb / ILUG
>
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/
>


--
Jon Smirl
jonsmirl@gmail.com

Re: PCI ROM resource allocation issue with 2.6.17-rc2

[Dave Airlie]

> I don't see quite why it would do it (yes, ROM's are prefetchable, but the
> old location was _valid_, and I don't see why we didn't re-use it), but it
> _really_ shouldn't matter.
>
> I would expect some silly interaction with X, as usual. It would be nice
> to see the whole dmesg, especially with the debugging messages - I suspect
> the remapping of the ROM happens only when X starts up, and that your
> dmesg is from just the kernel boot from before that?
>
> But it might be that I just missed it (or that we don't have good debug
> output for that case).

Just on my own system I'm seeing something of an issue while debugging
the dual-head issue,

my secondary head is being assigned non-prefetchable resources outside
the bridge,
PCI: Transparent bridge - 0000:00:1e.0

PCI: Bridge: 0000:00:1e.0
  IO window: b000-bfff
  MEM window: ff900000-ff9fffff
  PREFETCH window: e8000000-efffffff
is the bridge,

02:02.0 VGA compatible controller: ATI Technologies Inc Radeon RV100
QY [Radeon 7000/VE] (prog-if 00 [VGA])
        Subsystem: C.P. Technology Co. Ltd: Unknown device 2049
        Flags: stepping, medium devsel, IRQ 255
        Memory at e8000000 (32-bit, prefetchable) [disabled] [size=128M]
        I/O ports at b000 [disabled] [size=256]
        Memory at ffff0000 (32-bit, non-prefetchable) [disabled] [size=64K]
        Expansion ROM at ff900000 [disabled] [size=128K]
        Capabilities: [50] Power Management version 2

is the device,

when I modprobe radeon which does pci_enable_device, the bars are enabled...

However when X starts it tries to reassign the memory at 0xffff0000
down into the bridge memory... at 0xfff90000,  should the kernel do
this? or does it actually matter if the memory is behind the bridge as
its transparent... maybe I can at least patch X to check for
transparent bridges...

Dave.

Debian Kernel with Squash and UnionFS installed

[Joshua Perrymon]

Hey Guys,

I'm looking to build a live linux distro using the live linux scripts..
I'm not sure if I want to use slax because I haven't used it enough..

I prefer Debian.. But I'm having trouble with the linux live scripts
because SquashFS and UnionFS isn't working correctly..

Anyone know of a new kernel for Debian Linux with Squash and UnionFS
built in?

JP

Joshua Perrymon, C.E.H.
Sr. Security Consultant

Re: PCI ROM resource allocation issue with 2.6.17-rc2

[Linus Torvalds]

On Wed, 26 Apr 2006, Dave Airlie wrote:
>
> my secondary head is being assigned non-prefetchable resources outside
> the bridge,
> PCI: Transparent bridge - 0000:00:1e.0

Note that the "transparent" really means that it forwards all IO resources 
even outside the windows, and the windows are really just for show.

The kernel even used to totally ignore them, now it uses them as a hint 
and will _preferably_ put stuff inside the window for such bridges, if the 
windows have been set up (not all systems will even set up the windows at 
all).

> PCI: Bridge: 0000:00:1e.0
>   IO window: b000-bfff
>   MEM window: ff900000-ff9fffff
>   PREFETCH window: e8000000-efffffff
> is the bridge,
> 
> 02:02.0 VGA compatible controller: ATI Technologies Inc Radeon RV100
> QY [Radeon 7000/VE] (prog-if 00 [VGA])
>         Subsystem: C.P. Technology Co. Ltd: Unknown device 2049
>         Flags: stepping, medium devsel, IRQ 255
>         Memory at e8000000 (32-bit, prefetchable) [disabled] [size=128M]
>         I/O ports at b000 [disabled] [size=256]
>         Memory at ffff0000 (32-bit, non-prefetchable) [disabled] [size=64K]
>         Expansion ROM at ff900000 [disabled] [size=128K]
>         Capabilities: [50] Power Management version 2
> 
> is the device,
> 
> when I modprobe radeon which does pci_enable_device, the bars are enabled...
> 
> However when X starts it tries to reassign the memory at 0xffff0000
> down into the bridge memory... at 0xfff90000,  should the kernel do
> this? or does it actually matter if the memory is behind the bridge as
> its transparent... maybe I can at least patch X to check for
> transparent bridges...

It really shouldn't even matter where it ends up being enabled. Trying to 
move it into the bridge window is as good as anything else, since it was 
disabled to begin with (which means that you can't necessarily trust the 
location that it was disabled _at_ - the ffff0000 value could even be just 
what the firmware left it at after doing PCI BAR sizing, although I 
suspect that it's a perfectly valid address).

		Linus

Re: PCI ROM resource allocation issue with 2.6.17-rc2

[Dave Airlie]

>
> Note that the "transparent" really means that it forwards all IO resources
> even outside the windows, and the windows are really just for show.
>
> The kernel even used to totally ignore them, now it uses them as a hint
> and will _preferably_ put stuff inside the window for such bridges, if the
> windows have been set up (not all systems will even set up the windows at
> all).

Well X has support for it, but something like the quirk in the i386
fixups is needed to make it detect the Intel PCI bridge as a
transparent one..
> >
>
> It really shouldn't even matter where it ends up being enabled. Trying to
> move it into the bridge window is as good as anything else, since it was
> disabled to begin with (which means that you can't necessarily trust the
> location that it was disabled _at_ - the ffff0000 value could even be just
> what the firmware left it at after doing PCI BAR sizing, although I
> suspect that it's a perfectly valid address).

Well it does, as modprobing the DRM enables the device, and we store
the values in the DRM, X then goes and moves it... and the drm gets
annoyed later,

I still haven't tracked down the problem with the BIOS but I'm on its
trail now that I've fixed the PCI resource allocation..

Dave.