[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Qemu-devel] [PATCH v3 3/4] arm: Add PCIe host bridge in virt machin
From: |
Peter Maydell |
Subject: |
Re: [Qemu-devel] [PATCH v3 3/4] arm: Add PCIe host bridge in virt machine |
Date: |
Fri, 30 Jan 2015 17:21:11 +0000 |
On 29 January 2015 at 15:06, Alexander Graf <address@hidden> wrote:
> Now that we have a working "generic" PCIe host bridge driver, we can plug
> it into ARM's virt machine to always have PCIe available to normal ARM VMs.
> - /* 0x10000000 .. 0x40000000 reserved for PCI */
> + /* PCIe region layout: [ MMIO | PIO | ECAM ] */
> + [VIRT_PCIE] = { 0x10000000, 0x30000000 },
> + /*
> + * Align all the regions to their respective positions. Eventually
> + * we want to have:
> + *
> + * < MMIO space >
> + * < IO Port space >
> + * < ECAM space >
> + *
> + * all naturally aligned after each other. The MMIO region always starts
> + * at 0 and shrinks depending on the size of the other 2 regions.
> + */
This is still very opaque to me. Can't we just have a comment that
tells the reader how big the MMIO, IO and ECAM windows are without
requiring them to cross reference it with the previous array and
do a bunch of mental arithmetic?
For preference, this comment should live in the a15memmap[] array
where we document all the rest of the memory map.
thanks
-- PMM