On 04.12.2015 17:22, Gabriele Paoloni wrote:
Hi Lorenzo, Arnd (thanks to you both for looking at this)
-----Original Message----- From: Arnd Bergmann [mailto:arnd@arndb.de] Sent: 04 December 2015 13:57 To: Lorenzo Pieralisi Cc: linux-arm-kernel@lists.infradead.org; Gabriele Paoloni; linux- acpi@vger.kernel.org; linux-pci@vger.kernel.org; catalin.marinas@arm.com; linaro-acpi@lists.linaro.org; Liviu.Dudau@arm.com; linux-kernel@vger.kernel.org; will.deacon@arm.com; Wangyijing; Wangzhou (B); hanjun.guo@linaro.org; liudongdong (C); tn@semihalf.com; bhelgaas@google.com; tglx@linutronix.de; xuwei (O); Liguozhu (Kenneth); jiang.liu@linux.intel.com Subject: Re: [RFC PATCH 1/2] PCI/ACPI: Add ACPI support for non ECAM Host Bridge Controllers
On Friday 04 December 2015 12:04:04 Lorenzo Pieralisi wrote:
On Thu, Dec 03, 2015 at 09:58:14PM +0100, Arnd Bergmann wrote:
pci-host-generic.c is just for standard PCI implementations, and it has zero code that would be shared with ACPI: Most of the implementation deals with parsing DT properties, and all that code is entirely differnet for ACPI and already exists in drivers/acpi. The one thing that could be shared is the ECAM config space access, but ACPI already needs something else here because it requires access to the config space at early boot time, way before we even
load that driver, see raw_pci_read/raw_pci_write.
Yes, I agree, basically ACPI has already a concept of "host generic" layer, there is not much point in "merging" it with the pci-host-generic.c driver. One thing is for certain: nothing in this and Tomasz patchsets is arm64 specific, and should not live in arch/arm64.
Ok so now I guess Tomasz is aware of this and probably he is reworking his patchset to move his code into "drivers/acpi/pci_*", Tomasz can you confirm this?
Yes, working on it now and sorry for late response.
Tomasz