Fix fallout from PCI: PCIE ASPM support
PATCH 1/1 Fix fallout from PCI: PCIE ASPM support
In reply to a message not in the archiveQuoted 1 line · Linux Kernel Mailing List
PCI: PCIE ASPM support
This one broke almost all defconfigs on powerpc, since they enable CONFIG_EXPERIMENTAL and CONFIG_PCI, and this option defaults to y.
Make it default to 'n' like most other options, and make it depend on ACPI. Take out the unneccessary default 'n' from the debug option as well.
drivers/pci/pcie/Kconfig +1 -3
@@ -32,8 +32,7 @@ source "drivers/pci/pcie/aer/Kconfig" # config PCIEASPM bool "PCI Express ASPM support(Experimental)" - depends on PCI && EXPERIMENTAL - default y + depends on PCI && ACPI && EXPERIMENTAL help This enables PCI Express ASPM (Active State Power Management) and Clock Power Management. ASPM supports state L0/L0s/L1.
@@ -42,7 +41,6 @@ config PCIEASPM config PCIEASPM_DEBUG bool "Debug PCI Express ASPM" depends on PCIEASPM - default n help This enables PCI Express ASPM debug support. It will add per-device interface to control ASPM.
Quoted 10 lines · Olof Johansson
Quoted 1 line · Linux Kernel Mailing List
PCI: PCIE ASPM support
This one broke almost all defconfigs on powerpc, since they enable
CONFIG_EXPERIMENTALandCONFIG_PCI, and this option defaults to y.Make it default to 'n' like most other options, and make it depend on ACPI. Take out the unneccessary default 'n' from the debug option as well.
Is this actually ACPI-dependent?
-hpa
Quoted 5 lines · H. Peter Anvin
Quoted 4 lines · Olof Johansson
Quoted 1 line · Linux Kernel Mailing List
PCI: PCIE ASPM support
Quoted 8 lines
Quoted 6 lines
This one broke almost all defconfigs on powerpc, since they enable
CONFIG_EXPERIMENTALandCONFIG_PCI, and this option defaults to y.Make it default to 'n' like most other options, and make it depend on ACPI. Take out the unneccessary default 'n' from the debug option as well.
Is this actually ACPI-dependent?
It shouldn't be, but it seems like it's currently entangled a bit with the acpi include files. OSC_ACTIVE_STATE_PWR_SUPPORT comes from linux/pci-acpi.h.
I can't include acpi/acpi_bus.h though, since that ends up pulling in asm/ files that don't exist on powerpc.
So if the probing is fixed, no, it's not ACPI-dependent. Seems like something that's just as well to mask behind it until someone's actually used and tested it on a non-ACPI platform though, no?
I would experiment with it but I believe our firmware turns off empty links before the kernel is booted, we don't support hotplug at this time.
-Olof
Quoted 4 lines · Olof Johansson
Quoted 1 line · Linux Kernel Mailing List
PCI: PCIE ASPM support
Quoted 8 lines
This one broke almost all defconfigs on powerpc, since they enable
CONFIG_EXPERIMENTALandCONFIG_PCI, and this option defaults to y.Make it default to 'n' like most other options, and make it depend on ACPI. Take out the unneccessary default 'n' from the debug option as well.
Signed-off-by Olof Johansson <[email protected]>
I'm just going to revert this, it's caused too many problems, sorry.
greg k-h