On Tuesday, September 6, 2016 7:49:49 PM CEST Tomasz Nowicki wrote:
On 05.09.2016 04:25, Bjorn Helgaas wrote:
On Mon, Aug 08, 2016 at 03:05:39PM +0200, Tomasz Nowicki wrote:
Some platforms may not be fully compliant with generic set of PCI config accessors. For these cases we implement the way to overwrite accessors set. Algorithm traverses available quirk list (static array), matches against <oem_id, oem_table_id, rev, domain, bus number range> and returns pci_config_window structure with fancy PCI config ops. oem_id, oem_table_id and rev come from MCFG table standard header.
It is possible to define custom init call which is responsible for setting up PCI configuration access accordingly to quirk requirements. If custom init call is not defined, use standard pci_acpi_setup_ecam_mapping().
pci_generic_ecam_ops will be used for platforms free from quirks.
Signed-off-by: Tomasz Nowicki tn@semihalf.com Signed-off-by: Dongdong Liu liudongdong3@huawei.com Signed-off-by: Christopher Covington cov@codeaurora.org
drivers/pci/host/Makefile | 1 + drivers/pci/host/mcfg-quirks.c | 86 ++++++++++++++++++++++++++++++++++++++++++ drivers/pci/host/mcfg-quirks.h | 20 ++++++++++ include/linux/pci-acpi.h | 2 + 4 files changed, 109 insertions(+) create mode 100644 drivers/pci/host/mcfg-quirks.c create mode 100644 drivers/pci/host/mcfg-quirks.h
If the object is to work around defects in the ACPI MCFG table, I think I'd put the quirks closer to drivers/acpi/pci_mcfg.c, where we parse that table. What if we actually put them directly *in* drivers/acpi/pci_mcfg.c?
Then we need to export quirk init calls or ecam ops from drivers/pci/host/* directory.
Currently we keep mcfg quirk handling code and related drivers in one place drivers/pci/host/
I think that's the wrong place to have it, it just leads to people trying to share code with the host drivers in that directory, which hasn't really worked out so far, it just adds complexity.
Arnd