cardbus: Add a fixup hook and fix powerpc
PATCH 1/1 cardbus: Add a fixup hook and fix powerpc
The cardbus code creates PCI devices without ever going through the necessary fixup bits and pieces that normal PCI devices go through.
There's in fact a commented out call to pcibios_fixup_bus() in there, it's commented because ... it doesn't work.
I could make pcibios_fixup_bus() do the right thing on powerpc easily but I felt it cleaner instead to provide a specific hook pci_fixup_cardbus for which a weak empty implementation is provided by the PCI core.
This fixes cardbus on powerbooks and probably all other PowerPC platforms which was broken completely for ever on some platforms and since 2.6.31 on others such as PowerBooks when we made the DMA ops mandatory (since those are setup by the fixups).
---
Note: In the long run we might want to streamline the hooks for fixing up new devices vs. new busses and make some stuff common between cardbus and PCI hotplug which is actually a mess.
This is a quick fix that makes it work again in the meantime, and that I would like to see in 2.6.31 and 2.6.32 stable at least.
Another option if you prefer is I can make my pcibios_fixup_bus() do the right thing in all cases I believe (by testing bus->is_added to skip stuff that must not be done twice) and we can remove the comment and stick the call in #ifdef CONFIG_PPC but I felt that solution was a tad cleaner.
Olof, once that's in you should be able to remove the hack you have in the PA-Semi code to work around this.
Cheers,
Ben.
drivers/pcmcia/cardbus.c +1 -1
@@ -214,7 +214,7 @@ int __ref cb_alloc(struct pcmcia_socketunsigned int max, pass; s->functions =pci_scan_slot (bus,PCI_DEVFN (0 ,0 )); -// pcibios_fixup_bus(bus); +pci_fixup_cardbus (bus); max = bus->secondary;for (pass =0 ; pass <2 ; pass++)
include/linux/pci.h +3 -0
@@ -564,6 +564,9 @@ void pcibios_align_resource(void *, stru resource_size_t);void pcibios_update_irq (struct pci_dev *,int irq); +/* Weak but can be overriden by arch */ +void pci_fixup_cardbus (struct pci_bus *); +/* Generic PCI functions used internally */ extern struct pci_bus *pci_find_bus (int domain,int busnr);
arch/powerpc/kernel/pci-common.c +13 -0
@@ -1107,6 +1107,12 @@ void __devinit pcibios_setup_bus_deviceslist_for_each_entry (dev, &bus->devices, bus_list) {struct dev_archdata *sd = &dev->dev.archdata; +/* Cardbus can call us to add new devices to a bus, so ignore +* those who are already fully discovered +*/ +if (dev->is_added) +continue ; +/* Setup OF node pointer in archdata */ sd->of_node =pci_device_to_OF_node (dev);
@@ -1147,6 +1153,13 @@ void __devinit pcibios_fixup_bus(struct }EXPORT_SYMBOL (pcibios_fixup_bus); +void __devinit pci_fixup_cardbus (struct pci_bus *bus) +{ +/* Now fixup devices on that bus */ +pcibios_setup_bus_devices (bus); +} + +static int skip_isa_ioresource_align (struct pci_dev *dev) { if ((ppc_pci_flags & PPC_PCI_CAN_SKIP_ISA_ALIGN) &&
drivers/pci/pci.c +5 -0
@@ -2723,6 +2723,11 @@ int __attribute__ ((weak)) pci_ext_cfg_areturn 1 ; } +void __weak pci_fixup_cardbus (struct pci_bus *bus) +{ +} +EXPORT_SYMBOL (pci_fixup_cardbus); +static int __init pci_setup (char *str) {while (str) {
Hey Benjamin,
Quoted 1 line · Benjamin Herrenschmidt
Signed-off-by Benjamin Herrenschmidt <[email protected]>
Quoted 3 lines
Note: In the long run we might want to streamline the hooks for fixing up new devices vs. new busses and make some stuff common between cardbus and PCI hotplug which is actually a mess.
... though I'd prefer such a generic approach.
Best,
Dominik
Quoted 7 lines · Dominik Brodowski
Quoted 3 lines
Note: In the long run we might want to streamline the hooks for fixing up new devices vs. new busses and make some stuff common between cardbus and PCI hotplug which is actually a mess.
... though I'd prefer such a generic approach.
Thanks. Yes that would be better. I don't have the bandwidth right now to look at that though, maybe early next year.
Cheers,
Ben.
Jesse Barnes
AppliedQuoted 11 lines · On Wed, 09 Dec 2009 17:52:13 +1100 Benjamin Herrenschmidt…
The cardbus code creates PCI devices without ever going through the necessary fixup bits and pieces that normal PCI devices go through.
There's in fact a commented out call to
pcibios_fixup_bus()in there, it's commented because ... it doesn't work.I could make
pcibios_fixup_bus()do the right thing on powerpc easily but I felt it cleaner instead to provide a specific hookpci_fixup_cardbusfor which a weak empty implementation is provided by the PCI core.
Quoted 6 lines
This fixes cardbus on powerbooks and probably all other PowerPC platforms which was broken completely for ever on some platforms and since 2.6.31 on others such as PowerBooks when we made the DMA ops mandatory (since those are setup by the fixups).
Signed-off-by Benjamin Herrenschmidt <[email protected]>
Applied, thanks. Had to fix up a conflict since someone had "helpfully" changed the comment from a // to a /* .. */ in cardbus.c.
Signature · 1 line
Jesse Barnes, Intel Open Source Technology Center
Quoted 26 lines · On Wed, 09 Dec 2009 17:52:13 +1100 Benjamin Herrenschmidt…
The cardbus code creates PCI devices without ever going through the necessary fixup bits and pieces that normal PCI devices go through.
There's in fact a commented out call to
pcibios_fixup_bus()in there, it's commented because ... it doesn't work.I could make
pcibios_fixup_bus()do the right thing on powerpc easily but I felt it cleaner instead to provide a specific hookpci_fixup_cardbusfor which a weak empty implementation is provided by the PCI core.This fixes cardbus on powerbooks and probably all other PowerPC platforms which was broken completely for ever on some platforms and since 2.6.31 on others such as PowerBooks when we made the DMA ops mandatory (since those are setup by the fixups).
Signed-off-by Benjamin Herrenschmidt <[email protected]>---
Note: In the long run we might want to streamline the hooks for fixing up new devices vs. new busses and make some stuff common between cardbus and PCI hotplug which is actually a mess.
This is a quick fix that makes it work again in the meantime, and that I would like to see in 2.6.31 and 2.6.32 stable at least.
Quoted 9 lines
Another option if you prefer is I can make my
pcibios_fixup_bus()do the right thing in all cases I believe (by testingbus->is_addedto skip stuff that must not be done twice) and we can remove the comment and stick the call in #ifdefCONFIG_PPCbut I felt that solution was a tad cleaner.Olof, once that's in you should be able to remove the hack you have in the PA-Semi code to work around this.
Oops, looks like this fails for the modular case? I get an unresolved symbol error when building this with my default config...
Care to resend with the fix against my for-linus branch?
Thanks,
Signature · 1 line
Jesse Barnes, Intel Open Source Technology Center
Quoted 8 lines · Jesse Barnes
Quoted 3 lines
Olof, once that's in you should be able to remove the hack you have in the PA-Semi code to work around this.
Oops, looks like this fails for the modular case? I get an unresolved symbol error when building this with my default config...
Care to resend with the fix against my for-linus branch?
Care to send a log ? I fail to see where the problem is since the patch has:
drivers/pci/pci.c 2009-12-09 17:34:16.000000000 +1100 renamed from drivers/pci/pci.c 2009-12-09 17:33:24.000000000 +1100 +4 -0
@@ -2723,6 +2723,11 @@ int __attribute__ ((weak)) pci_ext_cfg_a return 1; } +void __weak pci_fixup_cardbus(struct pci_bus *bus) +{ +} +EXPORT_SYMBOL(pci_fixup_cardbus);
Or is that broken in some way ?
I'm starting to regret trying to use weak stuff :-) I may turn it into a good old:
#ifndef pcibios_fixup_cardbus
static inline void pcibios_fixup_cardbus(struct pci_bus *bus) { } #define pcibios_fixup_cardbus
#endif
In a header and have ppc define it.
(And call it pcibios_ instead of pci_ which matches better what other arch fixups are called :-)
Cheers,
Ben.
Quoted 11 lines · On Wed, 09 Dec 2009 17:52:13 +1100 Benjamin Herrenschmidt…
The cardbus code creates PCI devices without ever going through the necessary fixup bits and pieces that normal PCI devices go through.
There's in fact a commented out call to
pcibios_fixup_bus()in there, it's commented because ... it doesn't work.I could make
pcibios_fixup_bus()do the right thing on powerpc easily but I felt it cleaner instead to provide a specific hookpci_fixup_cardbusfor which a weak empty implementation is provided by the PCI core.
Quoted 6 lines
This fixes cardbus on powerbooks and probably all other PowerPC platforms which was broken completely for ever on some platforms and since 2.6.31 on others such as PowerBooks when we made the DMA ops mandatory (since those are setup by the fixups).
Signed-off-by Benjamin Herrenschmidt <[email protected]>
Ah the link failure was my fault. I had fixed up the conflict incorrectly. It's applied now.
Signature · 1 line
Jesse Barnes, Intel Open Source Technology Center