* mpc8541 pci1 ioport allocation address space problem
@ 2008-09-20 11:16 Wang Jian
2008-09-21 1:19 ` Wang Jian
2008-09-22 5:47 ` Benjamin Herrenschmidt
0 siblings, 2 replies; 6+ messages in thread
From: Wang Jian @ 2008-09-20 11:16 UTC (permalink / raw)
To: linuxppc-dev
[-- Attachment #1: Type: text/plain, Size: 2413 bytes --]
Hi,
Here I have a 8541 dev board, with 2 e1000 attached to pci0 and a
homebrewed addon board inserted into pci1. I am trying to make it work
under 2.6.26-rc8 (2.6.26 broken so I am working at rc8).
The ioports allocation reads
$ cat /proc/ioports
00000000-000fffff : /pci@e0008000
00001000-0000103f : 0000:00:0a.0
00001040-0000107f : 0000:00:0b.0
ffefe000-ffffdfff : /pci@e0009000
ffeff000-ffeff00f : 0001:01:0a.0
ffeff010-ffeff01f : 0001:01:0b.0
ffeff020-ffeff02f : 0001:01:0c.0
ffeff030-ffeff03f : 0001:01:0d.0
The port allocation for pci1 looks ridiculous. The addon board doesn't work.
After poking around, I find pci_process_bridge_OF_ranges() in
arch/powerpc/kernel/pci-common.c
--snip--
if (primary)
isa_io_base = (unsigned long)hose->io_base_virt;
--snip--
then fixup_resource(), _IO_BASE = isa_io_base. fix_resource will use
isa_io_base as base address to assign io port address space.
This is reasonable, but on my board, pci1's hose->io_base_virt is
smaller than pci0's. This lead to
[ 0.064214] PCI:0000:00:0a.0 Resource 4
0000000000001000-000000000000103f [20101] fixup...
[ 0.064224] fixup_resource() offset=00000000, io_base_virt=fdeb4000,
_IO_BASE=fdeb4000, io_base_phys=e2000000
[ 0.064232] PCI:0000:00:0a.0 0000000000001000-000000000000103f
[ 0.065129] PCI:0001:01:0a.0 Resource 2
0000000000001000-000000000000100f [20101] fixup...
[ 0.065139] fixup_resource() offset=ffefe000, io_base_virt=fddb2000,
_IO_BASE=fdeb4000, io_base_phys=e3000000
[ 0.065147] PCI:0001:01:0a.0 00000000ffeff000-00000000ffeff00f
offset = fddb2000 - fdeb4000 = ffefe000
So far, I think the workaround is to replace the code above with
--snip--
if (!isa_io_base)
isa_io_base = (unsigned long)hose->io_base_virt;
else if ((unsigned long)hose->io_base_virt < isa_io_base)
isa_io_base = (unsigned long)hose->io_base_virt;
--snip--
Then ioport allocation reads (But I haven't test if it works)
$ cat /proc/ioports
00000000-000fffff : /pci@e0009000
00001000-0000100f : 0001:01:0a.0
00001010-0000101f : 0001:01:0b.0
00001020-0000102f : 0001:01:0c.0
00001030-0000103f : 0001:01:0d.0
00102000-00201fff : /pci@e0008000
00103000-0010303f : 0000:00:0a.0
00103040-0010307f : 0000:00:0b.0
Can someone gives a better generic fix for this?
PS: I attach the kernel log (without workaround) for reference. I added
some printk code.
[-- Attachment #2: ppc_pci.txt --]
[-- Type: text/plain, Size: 6113 bytes --]
[ 0.064140] PCI: Found 0000:00:0a.0 [8086/1076] 000200 00
[ 0.064157] pci_read_bases(): IO: (0000:00:0a.0) l=00001001,start=00001000,flag=00000000
[ 0.064172] pci 0000:00:0a.0: calling 0xc01e9f98
[ 0.064182] PCI:0000:00:0a.0 Resource 0 0000000080000000-000000008001ffff [20204] fixup...
[ 0.064190] PCI:0000:00:0a.0 0000000080000000-000000008001ffff
[ 0.064198] PCI:0000:00:0a.0 Resource 2 0000000080020000-000000008002ffff [20204] fixup...
[ 0.064206] PCI:0000:00:0a.0 0000000080020000-000000008002ffff
[ 0.064214] PCI:0000:00:0a.0 Resource 4 0000000000001000-000000000000103f [20101] fixup...
[ 0.064224] fixup_resource() offset=00000000, io_base_virt=fdeb4000, _IO_BASE=fdeb4000, io_base_phys=e2000000
[ 0.064232] PCI:0000:00:0a.0 0000000000001000-000000000000103f
[ 0.064241] PCI:0000:00:0a.0 Resource 6 0000000000000000-000000000000ffff [27200] is unassigned
[ 0.064268] PCI: Found 0000:00:0b.0 [8086/1076] 000200 00
[ 0.064284] pci_read_bases(): IO: (0000:00:0b.0) l=00001041,start=00001040,flag=00000000
[ 0.064300] pci 0000:00:0b.0: calling 0xc01e9f98
[ 0.064309] PCI:0000:00:0b.0 Resource 0 0000000080040000-000000008005ffff [20204] fixup...
[ 0.064316] PCI:0000:00:0b.0 0000000080040000-000000008005ffff
[ 0.064325] PCI:0000:00:0b.0 Resource 2 0000000080060000-000000008006ffff [20204] fixup...
[ 0.064353] PCI:0000:00:0b.0 0000000080060000-000000008006ffff
[ 0.064362] PCI:0000:00:0b.0 Resource 4 0000000000001040-000000000000107f [20101] fixup...
[ 0.064371] fixup_resource() offset=00000000, io_base_virt=fdeb4000, _IO_BASE=fdeb4000, io_base_phys=e2000000
[ 0.064379] PCI:0000:00:0b.0 0000000000001040-000000000000107f
[ 0.064388] PCI:0000:00:0b.0 Resource 6 0000000000000000-000000000000ffff [27200] is unassigned
[ 0.064408] PCI: Fixups for bus 0000:00
[ 0.064413] PCI: Fixup bus 0 (PHB)
[ 0.064420] Try to map irq for 0000:00:00.0...
[ 0.064447] Try to map irq for 0000:00:0a.0...
[ 0.064459] -> got one, spec 2 cells (0x00000000 0x00000001...) on /soc8541@e0000000/pic@40000
[ 0.064478] -> mapped to linux irq 16
[ 0.064482] Try to map irq for 0000:00:0b.0...
[ 0.064498] -> got one, spec 2 cells (0x00000001 0x00000001...) on /soc8541@e0000000/pic@40000
[ 0.064510] -> mapped to linux irq 17
[ 0.064517] PCI: Bus scan for 0000:00 returning with max=00
[ 0.064973] PCI: Scanning bus 0001:01
[ 0.064989] PCI: Found 0001:01:00.0 [1057/000c] 000b20 00
[ 0.065011] pci 0001:01:00.0: calling fixup_hide_host_resource_fsl+0x0/0x50
[ 0.065024] pci 0001:01:00.0: calling 0xc01e9f98
[ 0.065055] PCI: Found 0001:01:0a.0 [104c/a106] 000000 00
[ 0.065069] pci_read_bases(): IO: (0001:01:0a.0) l=00001001,start=00001000,flag=00000000
[ 0.065087] pci 0001:01:0a.0: calling 0xc01e9f98
[ 0.065097] PCI:0001:01:0a.0 Resource 0 00000000a0400000-00000000a07fffff [21208] fixup...
[ 0.065105] PCI:0001:01:0a.0 00000000a0400000-00000000a07fffff
[ 0.065113] PCI:0001:01:0a.0 Resource 1 00000000a0800000-00000000a0ffffff [20200] fixup...
[ 0.065121] PCI:0001:01:0a.0 00000000a0800000-00000000a0ffffff
[ 0.065129] PCI:0001:01:0a.0 Resource 2 0000000000001000-000000000000100f [20101] fixup...
[ 0.065139] fixup_resource() offset=ffefe000, io_base_virt=fddb2000, _IO_BASE=fdeb4000, io_base_phys=e3000000
[ 0.065147] PCI:0001:01:0a.0 00000000ffeff000-00000000ffeff00f
[ 0.065169] PCI: Found 0001:01:0b.0 [104c/a106] 000000 00
[ 0.065184] pci_read_bases(): IO: (0001:01:0b.0) l=00001011,start=00001010,flag=00000000
[ 0.065202] pci 0001:01:0b.0: calling 0xc01e9f98
[ 0.065211] PCI:0001:01:0b.0 Resource 0 00000000a1000000-00000000a13fffff [21208] fixup...
[ 0.065219] PCI:0001:01:0b.0 00000000a1000000-00000000a13fffff
[ 0.065227] PCI:0001:01:0b.0 Resource 1 00000000a1800000-00000000a1ffffff [20200] fixup...
[ 0.065235] PCI:0001:01:0b.0 00000000a1800000-00000000a1ffffff
[ 0.065243] PCI:0001:01:0b.0 Resource 2 0000000000001010-000000000000101f [20101] fixup...
[ 0.065253] fixup_resource() offset=ffefe000, io_base_virt=fddb2000, _IO_BASE=fdeb4000, io_base_phys=e3000000
[ 0.065261] PCI:0001:01:0b.0 00000000ffeff010-00000000ffeff01f
[ 0.065284] PCI: Found 0001:01:0c.0 [104c/a106] 000000 00
[ 0.065298] pci_read_bases(): IO: (0001:01:0c.0) l=00001021,start=00001020,flag=00000000
[ 0.065316] pci 0001:01:0c.0: calling 0xc01e9f98
[ 0.065325] PCI:0001:01:0c.0 Resource 0 00000000a2000000-00000000a23fffff [21208] fixup...
[ 0.065333] PCI:0001:01:0c.0 00000000a2000000-00000000a23fffff
[ 0.065341] PCI:0001:01:0c.0 Resource 1 00000000a2800000-00000000a2ffffff [20200] fixup...
[ 0.065360] PCI:0001:01:0c.0 00000000a2800000-00000000a2ffffff
[ 0.065369] PCI:0001:01:0c.0 Resource 2 0000000000001020-000000000000102f [20101] fixup...
[ 0.065378] fixup_resource() offset=ffefe000, io_base_virt=fddb2000, _IO_BASE=fdeb4000, io_base_phys=e3000000
[ 0.065387] PCI:0001:01:0c.0 00000000ffeff020-00000000ffeff02f
[ 0.065413] PCI: Found 0001:01:0d.0 [104c/a106] 000000 00
[ 0.065428] pci_read_bases(): IO: (0001:01:0d.0) l=00001031,start=00001030,flag=00000000
[ 0.065447] pci 0001:01:0d.0: calling 0xc01e9f98
[ 0.065456] PCI:0001:01:0d.0 Resource 0 00000000a3000000-00000000a33fffff [21208] fixup...
[ 0.065464] PCI:0001:01:0d.0 00000000a3000000-00000000a33fffff
[ 0.065472] PCI:0001:01:0d.0 Resource 1 00000000a3800000-00000000a3ffffff [20200] fixup...
[ 0.065480] PCI:0001:01:0d.0 00000000a3800000-00000000a3ffffff
[ 0.065488] PCI:0001:01:0d.0 Resource 2 0000000000001030-000000000000103f [20101] fixup...
[ 0.065498] fixup_resource() offset=ffefe000, io_base_virt=fddb2000, _IO_BASE=fdeb4000, io_base_phys=e3000000
[ 0.065506] PCI:0001:01:0d.0 00000000ffeff030-00000000ffeff03f
[ 0.065527] PCI: Fixups for bus 0001:01
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: mpc8541 pci1 ioport allocation address space problem
2008-09-20 11:16 mpc8541 pci1 ioport allocation address space problem Wang Jian
@ 2008-09-21 1:19 ` Wang Jian
2008-09-23 6:55 ` Benjamin Herrenschmidt
2008-09-22 5:47 ` Benjamin Herrenschmidt
1 sibling, 1 reply; 6+ messages in thread
From: Wang Jian @ 2008-09-21 1:19 UTC (permalink / raw)
To: linuxppc-dev
The following patch works fine.
diff --git a/arch/powerpc/kernel/pci-common.c b/arch/powerpc/kernel/pci-common.c
index 063cdd4..4b913e0 100644
--- a/arch/powerpc/kernel/pci-common.c
+++ b/arch/powerpc/kernel/pci-common.c
@@ -582,9 +582,19 @@ void __devinit pci_process_bridge_OF_ranges(struct pci_controller *hose,
hose->io_base_virt = ioremap(cpu_addr, size);
/* Expect trouble if pci_addr is not 0 */
+#if 0
if (primary)
isa_io_base =
(unsigned long)hose->io_base_virt;
+#endif
+#if 1
+ if (!isa_io_base)
+ isa_io_base =
+ (unsigned long)hose->io_base_virt;
+ else if ((unsigned long)hose->io_base_virt < isa_io_base)
+ isa_io_base =
+ (unsigned long)hose->io_base_virt;
+#endif
#endif /* CONFIG_PPC32 */
/* pci_io_size and io_base_phys always represent IO
* space starting at 0 so we factor in pci_addr
--
On Sat, Sep 20, 2008 at 07:16:25PM +0800, Wang Jian wrote:
> Hi,
>
> Here I have a 8541 dev board, with 2 e1000 attached to pci0 and a
> homebrewed addon board inserted into pci1. I am trying to make it work
> under 2.6.26-rc8 (2.6.26 broken so I am working at rc8).
>
> The ioports allocation reads
>
> $ cat /proc/ioports
> 00000000-000fffff : /pci@e0008000
> 00001000-0000103f : 0000:00:0a.0
> 00001040-0000107f : 0000:00:0b.0
> ffefe000-ffffdfff : /pci@e0009000
> ffeff000-ffeff00f : 0001:01:0a.0
> ffeff010-ffeff01f : 0001:01:0b.0
> ffeff020-ffeff02f : 0001:01:0c.0
> ffeff030-ffeff03f : 0001:01:0d.0
>
> The port allocation for pci1 looks ridiculous. The addon board doesn't work.
>
> After poking around, I find pci_process_bridge_OF_ranges() in
> arch/powerpc/kernel/pci-common.c
>
> --snip--
> if (primary)
> isa_io_base = (unsigned long)hose->io_base_virt;
> --snip--
>
> then fixup_resource(), _IO_BASE = isa_io_base. fix_resource will use
> isa_io_base as base address to assign io port address space.
>
> This is reasonable, but on my board, pci1's hose->io_base_virt is
> smaller than pci0's. This lead to
>
> [ 0.064214] PCI:0000:00:0a.0 Resource 4
> 0000000000001000-000000000000103f [20101] fixup...
> [ 0.064224] fixup_resource() offset=00000000, io_base_virt=fdeb4000,
> _IO_BASE=fdeb4000, io_base_phys=e2000000
> [ 0.064232] PCI:0000:00:0a.0 0000000000001000-000000000000103f
>
> [ 0.065129] PCI:0001:01:0a.0 Resource 2
> 0000000000001000-000000000000100f [20101] fixup...
> [ 0.065139] fixup_resource() offset=ffefe000, io_base_virt=fddb2000,
> _IO_BASE=fdeb4000, io_base_phys=e3000000
> [ 0.065147] PCI:0001:01:0a.0 00000000ffeff000-00000000ffeff00f
>
> offset = fddb2000 - fdeb4000 = ffefe000
>
> So far, I think the workaround is to replace the code above with
>
> --snip--
> if (!isa_io_base)
> isa_io_base = (unsigned long)hose->io_base_virt;
> else if ((unsigned long)hose->io_base_virt < isa_io_base)
> isa_io_base = (unsigned long)hose->io_base_virt;
> --snip--
>
> Then ioport allocation reads (But I haven't test if it works)
>
> $ cat /proc/ioports
> 00000000-000fffff : /pci@e0009000
> 00001000-0000100f : 0001:01:0a.0
> 00001010-0000101f : 0001:01:0b.0
> 00001020-0000102f : 0001:01:0c.0
> 00001030-0000103f : 0001:01:0d.0
> 00102000-00201fff : /pci@e0008000
> 00103000-0010303f : 0000:00:0a.0
> 00103040-0010307f : 0000:00:0b.0
>
> Can someone gives a better generic fix for this?
>
> PS: I attach the kernel log (without workaround) for reference. I added
> some printk code.
>
> [ 0.064140] PCI: Found 0000:00:0a.0 [8086/1076] 000200 00
> [ 0.064157] pci_read_bases(): IO: (0000:00:0a.0) l=00001001,start=00001000,flag=00000000
> [ 0.064172] pci 0000:00:0a.0: calling 0xc01e9f98
> [ 0.064182] PCI:0000:00:0a.0 Resource 0 0000000080000000-000000008001ffff [20204] fixup...
> [ 0.064190] PCI:0000:00:0a.0 0000000080000000-000000008001ffff
> [ 0.064198] PCI:0000:00:0a.0 Resource 2 0000000080020000-000000008002ffff [20204] fixup...
> [ 0.064206] PCI:0000:00:0a.0 0000000080020000-000000008002ffff
> [ 0.064214] PCI:0000:00:0a.0 Resource 4 0000000000001000-000000000000103f [20101] fixup...
> [ 0.064224] fixup_resource() offset=00000000, io_base_virt=fdeb4000, _IO_BASE=fdeb4000, io_base_phys=e2000000
> [ 0.064232] PCI:0000:00:0a.0 0000000000001000-000000000000103f
> [ 0.064241] PCI:0000:00:0a.0 Resource 6 0000000000000000-000000000000ffff [27200] is unassigned
> [ 0.064268] PCI: Found 0000:00:0b.0 [8086/1076] 000200 00
> [ 0.064284] pci_read_bases(): IO: (0000:00:0b.0) l=00001041,start=00001040,flag=00000000
> [ 0.064300] pci 0000:00:0b.0: calling 0xc01e9f98
> [ 0.064309] PCI:0000:00:0b.0 Resource 0 0000000080040000-000000008005ffff [20204] fixup...
> [ 0.064316] PCI:0000:00:0b.0 0000000080040000-000000008005ffff
> [ 0.064325] PCI:0000:00:0b.0 Resource 2 0000000080060000-000000008006ffff [20204] fixup...
> [ 0.064353] PCI:0000:00:0b.0 0000000080060000-000000008006ffff
> [ 0.064362] PCI:0000:00:0b.0 Resource 4 0000000000001040-000000000000107f [20101] fixup...
> [ 0.064371] fixup_resource() offset=00000000, io_base_virt=fdeb4000, _IO_BASE=fdeb4000, io_base_phys=e2000000
> [ 0.064379] PCI:0000:00:0b.0 0000000000001040-000000000000107f
> [ 0.064388] PCI:0000:00:0b.0 Resource 6 0000000000000000-000000000000ffff [27200] is unassigned
> [ 0.064408] PCI: Fixups for bus 0000:00
> [ 0.064413] PCI: Fixup bus 0 (PHB)
> [ 0.064420] Try to map irq for 0000:00:00.0...
> [ 0.064447] Try to map irq for 0000:00:0a.0...
> [ 0.064459] -> got one, spec 2 cells (0x00000000 0x00000001...) on /soc8541@e0000000/pic@40000
> [ 0.064478] -> mapped to linux irq 16
> [ 0.064482] Try to map irq for 0000:00:0b.0...
> [ 0.064498] -> got one, spec 2 cells (0x00000001 0x00000001...) on /soc8541@e0000000/pic@40000
> [ 0.064510] -> mapped to linux irq 17
> [ 0.064517] PCI: Bus scan for 0000:00 returning with max=00
> [ 0.064973] PCI: Scanning bus 0001:01
> [ 0.064989] PCI: Found 0001:01:00.0 [1057/000c] 000b20 00
> [ 0.065011] pci 0001:01:00.0: calling fixup_hide_host_resource_fsl+0x0/0x50
> [ 0.065024] pci 0001:01:00.0: calling 0xc01e9f98
> [ 0.065055] PCI: Found 0001:01:0a.0 [104c/a106] 000000 00
> [ 0.065069] pci_read_bases(): IO: (0001:01:0a.0) l=00001001,start=00001000,flag=00000000
> [ 0.065087] pci 0001:01:0a.0: calling 0xc01e9f98
> [ 0.065097] PCI:0001:01:0a.0 Resource 0 00000000a0400000-00000000a07fffff [21208] fixup...
> [ 0.065105] PCI:0001:01:0a.0 00000000a0400000-00000000a07fffff
> [ 0.065113] PCI:0001:01:0a.0 Resource 1 00000000a0800000-00000000a0ffffff [20200] fixup...
> [ 0.065121] PCI:0001:01:0a.0 00000000a0800000-00000000a0ffffff
> [ 0.065129] PCI:0001:01:0a.0 Resource 2 0000000000001000-000000000000100f [20101] fixup...
> [ 0.065139] fixup_resource() offset=ffefe000, io_base_virt=fddb2000, _IO_BASE=fdeb4000, io_base_phys=e3000000
> [ 0.065147] PCI:0001:01:0a.0 00000000ffeff000-00000000ffeff00f
> [ 0.065169] PCI: Found 0001:01:0b.0 [104c/a106] 000000 00
> [ 0.065184] pci_read_bases(): IO: (0001:01:0b.0) l=00001011,start=00001010,flag=00000000
> [ 0.065202] pci 0001:01:0b.0: calling 0xc01e9f98
> [ 0.065211] PCI:0001:01:0b.0 Resource 0 00000000a1000000-00000000a13fffff [21208] fixup...
> [ 0.065219] PCI:0001:01:0b.0 00000000a1000000-00000000a13fffff
> [ 0.065227] PCI:0001:01:0b.0 Resource 1 00000000a1800000-00000000a1ffffff [20200] fixup...
> [ 0.065235] PCI:0001:01:0b.0 00000000a1800000-00000000a1ffffff
> [ 0.065243] PCI:0001:01:0b.0 Resource 2 0000000000001010-000000000000101f [20101] fixup...
> [ 0.065253] fixup_resource() offset=ffefe000, io_base_virt=fddb2000, _IO_BASE=fdeb4000, io_base_phys=e3000000
> [ 0.065261] PCI:0001:01:0b.0 00000000ffeff010-00000000ffeff01f
> [ 0.065284] PCI: Found 0001:01:0c.0 [104c/a106] 000000 00
> [ 0.065298] pci_read_bases(): IO: (0001:01:0c.0) l=00001021,start=00001020,flag=00000000
> [ 0.065316] pci 0001:01:0c.0: calling 0xc01e9f98
> [ 0.065325] PCI:0001:01:0c.0 Resource 0 00000000a2000000-00000000a23fffff [21208] fixup...
> [ 0.065333] PCI:0001:01:0c.0 00000000a2000000-00000000a23fffff
> [ 0.065341] PCI:0001:01:0c.0 Resource 1 00000000a2800000-00000000a2ffffff [20200] fixup...
> [ 0.065360] PCI:0001:01:0c.0 00000000a2800000-00000000a2ffffff
> [ 0.065369] PCI:0001:01:0c.0 Resource 2 0000000000001020-000000000000102f [20101] fixup...
> [ 0.065378] fixup_resource() offset=ffefe000, io_base_virt=fddb2000, _IO_BASE=fdeb4000, io_base_phys=e3000000
> [ 0.065387] PCI:0001:01:0c.0 00000000ffeff020-00000000ffeff02f
> [ 0.065413] PCI: Found 0001:01:0d.0 [104c/a106] 000000 00
> [ 0.065428] pci_read_bases(): IO: (0001:01:0d.0) l=00001031,start=00001030,flag=00000000
> [ 0.065447] pci 0001:01:0d.0: calling 0xc01e9f98
> [ 0.065456] PCI:0001:01:0d.0 Resource 0 00000000a3000000-00000000a33fffff [21208] fixup...
> [ 0.065464] PCI:0001:01:0d.0 00000000a3000000-00000000a33fffff
> [ 0.065472] PCI:0001:01:0d.0 Resource 1 00000000a3800000-00000000a3ffffff [20200] fixup...
> [ 0.065480] PCI:0001:01:0d.0 00000000a3800000-00000000a3ffffff
> [ 0.065488] PCI:0001:01:0d.0 Resource 2 0000000000001030-000000000000103f [20101] fixup...
> [ 0.065498] fixup_resource() offset=ffefe000, io_base_virt=fddb2000, _IO_BASE=fdeb4000, io_base_phys=e3000000
> [ 0.065506] PCI:0001:01:0d.0 00000000ffeff030-00000000ffeff03f
> [ 0.065527] PCI: Fixups for bus 0001:01
> _______________________________________________
> Linuxppc-dev mailing list
> Linuxppc-dev@ozlabs.org
> https://ozlabs.org/mailman/listinfo/linuxppc-dev
^ permalink raw reply related [flat|nested] 6+ messages in thread
* Re: mpc8541 pci1 ioport allocation address space problem
2008-09-20 11:16 mpc8541 pci1 ioport allocation address space problem Wang Jian
2008-09-21 1:19 ` Wang Jian
@ 2008-09-22 5:47 ` Benjamin Herrenschmidt
2008-09-23 14:24 ` Wang Jian
1 sibling, 1 reply; 6+ messages in thread
From: Benjamin Herrenschmidt @ 2008-09-22 5:47 UTC (permalink / raw)
To: Wang Jian; +Cc: linuxppc-dev
On Sat, 2008-09-20 at 19:16 +0800, Wang Jian wrote:
> Hi,
>
> Here I have a 8541 dev board, with 2 e1000 attached to pci0 and a
> homebrewed addon board inserted into pci1. I am trying to make it work
> under 2.6.26-rc8 (2.6.26 broken so I am working at rc8).
>
> The ioports allocation reads
>
> $ cat /proc/ioports
> 00000000-000fffff : /pci@e0008000
> 00001000-0000103f : 0000:00:0a.0
> 00001040-0000107f : 0000:00:0b.0
> ffefe000-ffffdfff : /pci@e0009000
> ffeff000-ffeff00f : 0001:01:0a.0
> ffeff010-ffeff01f : 0001:01:0b.0
> ffeff020-ffeff02f : 0001:01:0c.0
> ffeff030-ffeff03f : 0001:01:0d.0
>
> The port allocation for pci1 looks ridiculous. The addon board doesn't work.
>
> After poking around, I find pci_process_bridge_OF_ranges() in
> arch/powerpc/kernel/pci-common.c
.../...
This is expected and should work. Depending on the relative
physical addresses of IO space and the order in which the bridges are
discovered, the bridge IO ports will look at either positive or negative
values. This should be fine, as port numbers are supposed to be 32 bits
and in-kernel arithmetic should do the right thing... I suppose unless a
driver stores those in a 64 bits integer and doesn't sign extend.
I would like to change that whole thing to something more similar to
64 bits where I reserve a portion of the address space for IO ports,
though address space on 32 bits platforms is scarce, but nothing I have
time to toy with right now.
Cheers,
Ben.
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: mpc8541 pci1 ioport allocation address space problem
2008-09-21 1:19 ` Wang Jian
@ 2008-09-23 6:55 ` Benjamin Herrenschmidt
0 siblings, 0 replies; 6+ messages in thread
From: Benjamin Herrenschmidt @ 2008-09-23 6:55 UTC (permalink / raw)
To: Wang Jian; +Cc: linuxppc-dev
On Sun, 2008-09-21 at 09:19 +0800, Wang Jian wrote:
> The following patch works fine.
And will break platforms that explicitely want IO port 0 to be on the
primary bus, though I'm not sure there's many of these left...
Ben.
> diff --git a/arch/powerpc/kernel/pci-common.c b/arch/powerpc/kernel/pci-common.c
> index 063cdd4..4b913e0 100644
> --- a/arch/powerpc/kernel/pci-common.c
> +++ b/arch/powerpc/kernel/pci-common.c
> @@ -582,9 +582,19 @@ void __devinit pci_process_bridge_OF_ranges(struct pci_controller *hose,
> hose->io_base_virt = ioremap(cpu_addr, size);
>
> /* Expect trouble if pci_addr is not 0 */
> +#if 0
> if (primary)
> isa_io_base =
> (unsigned long)hose->io_base_virt;
> +#endif
> +#if 1
> + if (!isa_io_base)
> + isa_io_base =
> + (unsigned long)hose->io_base_virt;
> + else if ((unsigned long)hose->io_base_virt < isa_io_base)
> + isa_io_base =
> + (unsigned long)hose->io_base_virt;
> +#endif
> #endif /* CONFIG_PPC32 */
> /* pci_io_size and io_base_phys always represent IO
> * space starting at 0 so we factor in pci_addr
> --
>
> On Sat, Sep 20, 2008 at 07:16:25PM +0800, Wang Jian wrote:
> > Hi,
> >
> > Here I have a 8541 dev board, with 2 e1000 attached to pci0 and a
> > homebrewed addon board inserted into pci1. I am trying to make it work
> > under 2.6.26-rc8 (2.6.26 broken so I am working at rc8).
> >
> > The ioports allocation reads
> >
> > $ cat /proc/ioports
> > 00000000-000fffff : /pci@e0008000
> > 00001000-0000103f : 0000:00:0a.0
> > 00001040-0000107f : 0000:00:0b.0
> > ffefe000-ffffdfff : /pci@e0009000
> > ffeff000-ffeff00f : 0001:01:0a.0
> > ffeff010-ffeff01f : 0001:01:0b.0
> > ffeff020-ffeff02f : 0001:01:0c.0
> > ffeff030-ffeff03f : 0001:01:0d.0
> >
> > The port allocation for pci1 looks ridiculous. The addon board doesn't work.
> >
> > After poking around, I find pci_process_bridge_OF_ranges() in
> > arch/powerpc/kernel/pci-common.c
> >
> > --snip--
> > if (primary)
> > isa_io_base = (unsigned long)hose->io_base_virt;
> > --snip--
> >
> > then fixup_resource(), _IO_BASE = isa_io_base. fix_resource will use
> > isa_io_base as base address to assign io port address space.
> >
> > This is reasonable, but on my board, pci1's hose->io_base_virt is
> > smaller than pci0's. This lead to
> >
> > [ 0.064214] PCI:0000:00:0a.0 Resource 4
> > 0000000000001000-000000000000103f [20101] fixup...
> > [ 0.064224] fixup_resource() offset=00000000, io_base_virt=fdeb4000,
> > _IO_BASE=fdeb4000, io_base_phys=e2000000
> > [ 0.064232] PCI:0000:00:0a.0 0000000000001000-000000000000103f
> >
> > [ 0.065129] PCI:0001:01:0a.0 Resource 2
> > 0000000000001000-000000000000100f [20101] fixup...
> > [ 0.065139] fixup_resource() offset=ffefe000, io_base_virt=fddb2000,
> > _IO_BASE=fdeb4000, io_base_phys=e3000000
> > [ 0.065147] PCI:0001:01:0a.0 00000000ffeff000-00000000ffeff00f
> >
> > offset = fddb2000 - fdeb4000 = ffefe000
> >
> > So far, I think the workaround is to replace the code above with
> >
> > --snip--
> > if (!isa_io_base)
> > isa_io_base = (unsigned long)hose->io_base_virt;
> > else if ((unsigned long)hose->io_base_virt < isa_io_base)
> > isa_io_base = (unsigned long)hose->io_base_virt;
> > --snip--
> >
> > Then ioport allocation reads (But I haven't test if it works)
> >
> > $ cat /proc/ioports
> > 00000000-000fffff : /pci@e0009000
> > 00001000-0000100f : 0001:01:0a.0
> > 00001010-0000101f : 0001:01:0b.0
> > 00001020-0000102f : 0001:01:0c.0
> > 00001030-0000103f : 0001:01:0d.0
> > 00102000-00201fff : /pci@e0008000
> > 00103000-0010303f : 0000:00:0a.0
> > 00103040-0010307f : 0000:00:0b.0
> >
> > Can someone gives a better generic fix for this?
> >
> > PS: I attach the kernel log (without workaround) for reference. I added
> > some printk code.
> >
>
> > [ 0.064140] PCI: Found 0000:00:0a.0 [8086/1076] 000200 00
> > [ 0.064157] pci_read_bases(): IO: (0000:00:0a.0) l=00001001,start=00001000,flag=00000000
> > [ 0.064172] pci 0000:00:0a.0: calling 0xc01e9f98
> > [ 0.064182] PCI:0000:00:0a.0 Resource 0 0000000080000000-000000008001ffff [20204] fixup...
> > [ 0.064190] PCI:0000:00:0a.0 0000000080000000-000000008001ffff
> > [ 0.064198] PCI:0000:00:0a.0 Resource 2 0000000080020000-000000008002ffff [20204] fixup...
> > [ 0.064206] PCI:0000:00:0a.0 0000000080020000-000000008002ffff
> > [ 0.064214] PCI:0000:00:0a.0 Resource 4 0000000000001000-000000000000103f [20101] fixup...
> > [ 0.064224] fixup_resource() offset=00000000, io_base_virt=fdeb4000, _IO_BASE=fdeb4000, io_base_phys=e2000000
> > [ 0.064232] PCI:0000:00:0a.0 0000000000001000-000000000000103f
> > [ 0.064241] PCI:0000:00:0a.0 Resource 6 0000000000000000-000000000000ffff [27200] is unassigned
> > [ 0.064268] PCI: Found 0000:00:0b.0 [8086/1076] 000200 00
> > [ 0.064284] pci_read_bases(): IO: (0000:00:0b.0) l=00001041,start=00001040,flag=00000000
> > [ 0.064300] pci 0000:00:0b.0: calling 0xc01e9f98
> > [ 0.064309] PCI:0000:00:0b.0 Resource 0 0000000080040000-000000008005ffff [20204] fixup...
> > [ 0.064316] PCI:0000:00:0b.0 0000000080040000-000000008005ffff
> > [ 0.064325] PCI:0000:00:0b.0 Resource 2 0000000080060000-000000008006ffff [20204] fixup...
> > [ 0.064353] PCI:0000:00:0b.0 0000000080060000-000000008006ffff
> > [ 0.064362] PCI:0000:00:0b.0 Resource 4 0000000000001040-000000000000107f [20101] fixup...
> > [ 0.064371] fixup_resource() offset=00000000, io_base_virt=fdeb4000, _IO_BASE=fdeb4000, io_base_phys=e2000000
> > [ 0.064379] PCI:0000:00:0b.0 0000000000001040-000000000000107f
> > [ 0.064388] PCI:0000:00:0b.0 Resource 6 0000000000000000-000000000000ffff [27200] is unassigned
> > [ 0.064408] PCI: Fixups for bus 0000:00
> > [ 0.064413] PCI: Fixup bus 0 (PHB)
> > [ 0.064420] Try to map irq for 0000:00:00.0...
> > [ 0.064447] Try to map irq for 0000:00:0a.0...
> > [ 0.064459] -> got one, spec 2 cells (0x00000000 0x00000001...) on /soc8541@e0000000/pic@40000
> > [ 0.064478] -> mapped to linux irq 16
> > [ 0.064482] Try to map irq for 0000:00:0b.0...
> > [ 0.064498] -> got one, spec 2 cells (0x00000001 0x00000001...) on /soc8541@e0000000/pic@40000
> > [ 0.064510] -> mapped to linux irq 17
> > [ 0.064517] PCI: Bus scan for 0000:00 returning with max=00
> > [ 0.064973] PCI: Scanning bus 0001:01
> > [ 0.064989] PCI: Found 0001:01:00.0 [1057/000c] 000b20 00
> > [ 0.065011] pci 0001:01:00.0: calling fixup_hide_host_resource_fsl+0x0/0x50
> > [ 0.065024] pci 0001:01:00.0: calling 0xc01e9f98
> > [ 0.065055] PCI: Found 0001:01:0a.0 [104c/a106] 000000 00
> > [ 0.065069] pci_read_bases(): IO: (0001:01:0a.0) l=00001001,start=00001000,flag=00000000
> > [ 0.065087] pci 0001:01:0a.0: calling 0xc01e9f98
> > [ 0.065097] PCI:0001:01:0a.0 Resource 0 00000000a0400000-00000000a07fffff [21208] fixup...
> > [ 0.065105] PCI:0001:01:0a.0 00000000a0400000-00000000a07fffff
> > [ 0.065113] PCI:0001:01:0a.0 Resource 1 00000000a0800000-00000000a0ffffff [20200] fixup...
> > [ 0.065121] PCI:0001:01:0a.0 00000000a0800000-00000000a0ffffff
> > [ 0.065129] PCI:0001:01:0a.0 Resource 2 0000000000001000-000000000000100f [20101] fixup...
> > [ 0.065139] fixup_resource() offset=ffefe000, io_base_virt=fddb2000, _IO_BASE=fdeb4000, io_base_phys=e3000000
> > [ 0.065147] PCI:0001:01:0a.0 00000000ffeff000-00000000ffeff00f
> > [ 0.065169] PCI: Found 0001:01:0b.0 [104c/a106] 000000 00
> > [ 0.065184] pci_read_bases(): IO: (0001:01:0b.0) l=00001011,start=00001010,flag=00000000
> > [ 0.065202] pci 0001:01:0b.0: calling 0xc01e9f98
> > [ 0.065211] PCI:0001:01:0b.0 Resource 0 00000000a1000000-00000000a13fffff [21208] fixup...
> > [ 0.065219] PCI:0001:01:0b.0 00000000a1000000-00000000a13fffff
> > [ 0.065227] PCI:0001:01:0b.0 Resource 1 00000000a1800000-00000000a1ffffff [20200] fixup...
> > [ 0.065235] PCI:0001:01:0b.0 00000000a1800000-00000000a1ffffff
> > [ 0.065243] PCI:0001:01:0b.0 Resource 2 0000000000001010-000000000000101f [20101] fixup...
> > [ 0.065253] fixup_resource() offset=ffefe000, io_base_virt=fddb2000, _IO_BASE=fdeb4000, io_base_phys=e3000000
> > [ 0.065261] PCI:0001:01:0b.0 00000000ffeff010-00000000ffeff01f
> > [ 0.065284] PCI: Found 0001:01:0c.0 [104c/a106] 000000 00
> > [ 0.065298] pci_read_bases(): IO: (0001:01:0c.0) l=00001021,start=00001020,flag=00000000
> > [ 0.065316] pci 0001:01:0c.0: calling 0xc01e9f98
> > [ 0.065325] PCI:0001:01:0c.0 Resource 0 00000000a2000000-00000000a23fffff [21208] fixup...
> > [ 0.065333] PCI:0001:01:0c.0 00000000a2000000-00000000a23fffff
> > [ 0.065341] PCI:0001:01:0c.0 Resource 1 00000000a2800000-00000000a2ffffff [20200] fixup...
> > [ 0.065360] PCI:0001:01:0c.0 00000000a2800000-00000000a2ffffff
> > [ 0.065369] PCI:0001:01:0c.0 Resource 2 0000000000001020-000000000000102f [20101] fixup...
> > [ 0.065378] fixup_resource() offset=ffefe000, io_base_virt=fddb2000, _IO_BASE=fdeb4000, io_base_phys=e3000000
> > [ 0.065387] PCI:0001:01:0c.0 00000000ffeff020-00000000ffeff02f
> > [ 0.065413] PCI: Found 0001:01:0d.0 [104c/a106] 000000 00
> > [ 0.065428] pci_read_bases(): IO: (0001:01:0d.0) l=00001031,start=00001030,flag=00000000
> > [ 0.065447] pci 0001:01:0d.0: calling 0xc01e9f98
> > [ 0.065456] PCI:0001:01:0d.0 Resource 0 00000000a3000000-00000000a33fffff [21208] fixup...
> > [ 0.065464] PCI:0001:01:0d.0 00000000a3000000-00000000a33fffff
> > [ 0.065472] PCI:0001:01:0d.0 Resource 1 00000000a3800000-00000000a3ffffff [20200] fixup...
> > [ 0.065480] PCI:0001:01:0d.0 00000000a3800000-00000000a3ffffff
> > [ 0.065488] PCI:0001:01:0d.0 Resource 2 0000000000001030-000000000000103f [20101] fixup...
> > [ 0.065498] fixup_resource() offset=ffefe000, io_base_virt=fddb2000, _IO_BASE=fdeb4000, io_base_phys=e3000000
> > [ 0.065506] PCI:0001:01:0d.0 00000000ffeff030-00000000ffeff03f
> > [ 0.065527] PCI: Fixups for bus 0001:01
>
> > _______________________________________________
> > Linuxppc-dev mailing list
> > Linuxppc-dev@ozlabs.org
> > https://ozlabs.org/mailman/listinfo/linuxppc-dev
> _______________________________________________
> Linuxppc-dev mailing list
> Linuxppc-dev@ozlabs.org
> https://ozlabs.org/mailman/listinfo/linuxppc-dev
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: mpc8541 pci1 ioport allocation address space problem
2008-09-22 5:47 ` Benjamin Herrenschmidt
@ 2008-09-23 14:24 ` Wang Jian
2008-09-23 21:16 ` Benjamin Herrenschmidt
0 siblings, 1 reply; 6+ messages in thread
From: Wang Jian @ 2008-09-23 14:24 UTC (permalink / raw)
To: benh; +Cc: linuxppc-dev
Confirmed.
I had this patch and 2 fixes on dts applied simultaneously, and the board
works. I thought this patch was necessary.
I reverted this patch and tried again, the board still works. Thanks for
your insight.
Anyway, the io port read from /proc/ioports is misleading, so I wish it
can be changed a little.
Regards
Benjamin Herrenschmidt wrote:
> This is expected and should work. Depending on the relative
> physical addresses of IO space and the order in which the bridges are
> discovered, the bridge IO ports will look at either positive or negative
> values. This should be fine, as port numbers are supposed to be 32 bits
> and in-kernel arithmetic should do the right thing... I suppose unless a
> driver stores those in a 64 bits integer and doesn't sign extend.
>
> I would like to change that whole thing to something more similar to
> 64 bits where I reserve a portion of the address space for IO ports,
> though address space on 32 bits platforms is scarce, but nothing I have
> time to toy with right now.
>
> Cheers,
> Ben.
>
> _______________________________________________
> Linuxppc-dev mailing list
> Linuxppc-dev@ozlabs.org
> https://ozlabs.org/mailman/listinfo/linuxppc-dev
>
>
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: mpc8541 pci1 ioport allocation address space problem
2008-09-23 14:24 ` Wang Jian
@ 2008-09-23 21:16 ` Benjamin Herrenschmidt
0 siblings, 0 replies; 6+ messages in thread
From: Benjamin Herrenschmidt @ 2008-09-23 21:16 UTC (permalink / raw)
To: Wang Jian; +Cc: linuxppc-dev
On Tue, 2008-09-23 at 22:24 +0800, Wang Jian wrote:
>
> I had this patch and 2 fixes on dts applied simultaneously, and the
> board
> works. I thought this patch was necessary.
>
> I reverted this patch and tried again, the board still works. Thanks
> for
> your insight.
>
> Anyway, the io port read from /proc/ioports is misleading, so I wish
> it
> can be changed a little.
>
Ordering things differently in the DTS so that the lower address bus
gets probed first might help.
Cheers
Ben.
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2008-09-23 21:16 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2008-09-20 11:16 mpc8541 pci1 ioport allocation address space problem Wang Jian
2008-09-21 1:19 ` Wang Jian
2008-09-23 6:55 ` Benjamin Herrenschmidt
2008-09-22 5:47 ` Benjamin Herrenschmidt
2008-09-23 14:24 ` Wang Jian
2008-09-23 21:16 ` Benjamin Herrenschmidt
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).