Skip to contentSkip to conversation

What's in pci-2.6.git

PATCH In review Jesse Barnes 3 participants 4 messages Jun 16, 2009 PCI

lore
Share & tools
lore
#0 lore raw

I'm getting the queue together for my next pull request. See below for the shortlog.

On a high level there are a few important, new features & highlights:
- improved AER support
Thanks to Yanmin the AER code can now handle multiple errors and chips that don't correctly set the source ID. It does this by scanning the bus when such a condition is detected, looking for the errors. Huang contributed some code to help test this feature as well; driver developers are encouraged to use it.
- PCI resource allocation fixes
Yinghai sent some fixes for resource allocation for transparent brdiges & prefetchable regions. We also try to use _CRS data by default now.
- no more pci_find_slot! (Kudos again to Alex for his work this

    cycle.)

- and the usual collection of misc cleanups & fixes

In looking at the shortlog I just realized the authorship info for the check_enable_amd_mmconf and overflow quirk patches are wrong, so we're still at least one rebase away from having a tree that's ready to pull.

I think the only other patchset left outstanding is Kenji-san's ASPM cleanups; I'm waiting on some more review for those. I think we'll be able to squeeze them in this cycle assuming that happens soon. Oh and there's my Moorestown platform support code. I got some good review on that, so I'll be cleaning it up and sending out another patch this week, with an eye towards merging it this week as well (though I'm not sure if some of the other bits will be ready; I'll talk with the Moorestown folks about that, maybe there's no hurry).

If all goes well I'll be sending Linus a pull request later this week, so please test & report issues as soon as possible!

Thanks,
Jesse

Alex Chiang (16):
PCI Hotplug: cpqphp: stray whitespace cleanups
PCI Hotplug: cpqphp: fix comment style
PCI Hotplug: cpqphp: obey 80 column convention in cpqphp.h PCI Hotplug: cpqphp: remove useless prototypes in cpqphp_core.c PCI Hotplug: cpqphp: eliminate stray braces
PCI Hotplug: cpqphp: refactor cpqhp_probe
PCI Hotplug: cpqphp: clean up cpqphp_ctrl.c
PCI Hotplug: cpqphp: refactor cpqphp_save_slot_config
PCI Hotplug: cpqphp: style cleanups
PCI Hotplug: cpqphp: refactor cpqhp_save_config
PCI Hotplug: cpqphp: clean up accesses to pcibios_get_irq_routing_table() PCI Hotplug: cpqphp: eliminate dead code - PCI_ScanBusNonBridge PCI Hotplug: cpqphp: constify slot_name()
PCI Hotplug: cpqphp: don't use pci_find_slot()
PCI: remove deprecated pci_find_slot() interface
PCI: eliminate redundant pci_stop_dev() call from pci_destroy_dev()

Andrew Patterson (1):
PCI: Add support for turning PCIe ECRC on or off

Greg Kroah-Hartman (1):
PCIE: remove driver_data direct access of struct device

Hidetoshi Seto (2):
PCI MSI: Remove unused/obsolete macros and definitions PCI MSI: Define PCI_MSI_MASK_32/64

Huang Ying (4):
PCI: Add pci_bus_set_ops
PCI: PCIE AER: export aer_irq
PCI: PCIE AER: Document for PCIE AER software error injection PCI AER: software error injection

Ingo Molnar (1):
x86/PCI: add description for check_enable_amd_mmconf boot parameter

Jesse Barnes (1):
PCI: use ACPI _CRS data by default

Kenji Kaneshige (11):
PCI: use pci_is_root_bus() in acpi_pci_get_bridge_handle()
PCI: use pci_is_root_bus() in acpi_find_root_bridge_handle()
PCI: use pci_is_root_bus() in pci_find_upstream_pcie_bridge()
PCI: use pci_is_root_bus() in pci_read_bridge_bases()
PCI: use pci_is_root_bus() in pci_get_interrupt_pin()
PCI: use pci_is_root_bus() in pci_common_swizzle()
PCI: Remove untested Electromechanical Interlock (EMI) support in pciehp. PCI hotplug: fix return value of has_foo() functions
PCI hotplug: create symlink to hotplug driver module
PCI hotplug: remove redundant .owner initializations
PCI: remove invalid comment of msi_mask_irq()

Mats Erik Andersson (1):
PCI: expose SMBus on Asus notebook A6L

Michael S. Tsirkin (1):
PCI MSI: let drivers retry when not enough vectors

Michal Miroslaw (1):
PCI quirk: HP hides SMBus controller in Compaq nx9500 laptops

Rafael J. Wysocki (2):
PCI PM: Follow PCI_PM_CTRL_NO_SOFT_RESET during transitions from D3 PCI PM: Fix handling of devices without PM support by pci_target_state()

Shaohua Li (1):
PCI: disable ASPM on VIA root-port-under-bridge configurations

Yinghai Lu (2):
PCI/x86: don't assume prefetchable ranges are 64bit
PCI: improve resource allocation under transparent bridges

Yu Zhao (4):
PCI: fix SR-IOV function dependency link problem
PCI: cleanup Function Level Reset
PCI: support PM D0hot->D3 transition reset
PCI: support Secondary Bus Reset

Zhang, Yanmin (3):
PCI AER: support Multiple Error Received and no error source id PCI AER: support invalid error source IDs
PCI AER: multiple error support

[email protected] (2):
PCI: ibmphp_core.c: fix warning due to missing module_exit() PCI quirk: unhide 'Overflow' device on i828{6,7}5P/PE chipsets

Signature · 1 line
Jesse Barnes, Intel Open Source Technology Center
#1 lore raw

PATCH 1/1 IDE: consistently use type bool for wake enable variable

IDE: consistently use type bool for wake enable parameter

Other functions use type bool, so use that for pci_enable_wake as well.

---

Hi Jesse,

I've had this trivial patch sitting in my local branch for some time. Please consider including.

Cheers,
FJP

drivers/pci/pci.c +1 -1
index 1a91bf9..602052e 100644
@@ -1203,7 +1203,7 @@ void pci_pme_active(struct pci_dev *dev, bool enable)
  * Error code depending on the platform is returned if both the platform and
  * the native mechanism fail to enable the generation of wake-up events
  */
-int pci_enable_wake(struct pci_dev *dev, pci_power_t state, int enable)
+int pci_enable_wake(struct pci_dev *dev, pci_power_t state, bool enable)
 {
 	int error = 0;
 	bool pme_done = false;
include/linux/pci.h +1 -1
index 37f3430..6f6bc19 100644
@@ -724,7 +724,7 @@ int pci_set_power_state(struct pci_dev *dev, pci_power_t state);
 pci_power_t pci_choose_state(struct pci_dev *dev, pm_message_t state);
 bool pci_pme_capable(struct pci_dev *dev, pci_power_t state);
 void pci_pme_active(struct pci_dev *dev, bool enable);
-int pci_enable_wake(struct pci_dev *dev, pci_power_t state, int enable);
+int pci_enable_wake(struct pci_dev *dev, pci_power_t state, bool enable);
 int pci_wake_from_d3(struct pci_dev *dev, bool enable);
 pci_power_t pci_target_state(struct pci_dev *dev);
 int pci_prepare_to_sleep(struct pci_dev *dev);
#2 lore raw

Re: [PATCH] IDE: consistently use type bool for wake enable variable

Quoted 5 lines · On Wed, 17 Jun 2009 00:16:15 +0200 Frans Pop <[email protected]> wrote:

IDE: consistently use type bool for wake enable parameter

Other functions use type bool, so use that for pci_enable_wake as well.

Quoted 7 lines

---

Hi Jesse,

I've had this trivial patch sitting in my local branch for some time. Please consider including.

Yep, looks reasonable. Rafael I assume there's no objection here?

Thanks,

Signature · 1 line
Jesse Barnes, Intel Open Source Technology Center
#3 lore raw

Re: [PATCH] IDE: consistently use type bool for wake enable variable

Quoted 11 lines · Jesse Barnes
Quoted 8 lines · On Wed, 17 Jun 2009 00:16:15 +0200 Frans Pop <[email protected]> wrote:

IDE: consistently use type bool for wake enable parameter

Other functions use type bool, so use that for pci_enable_wake as well.

---

Quoted 6 lines
Quoted 4 lines

Hi Jesse,

I've had this trivial patch sitting in my local branch for some time. Please consider including.

Yep, looks reasonable. Rafael I assume there's no objection here?

Sure, please feel free add my ACK to the patch.

Best,
Rafael