Re: [PATCH 02/11] PCI: Try to allocate mem64 above 4G at first

[joeyli]

æ äï2012-07-04 æ 10:56 +0800ïlee joey æåï
> 
> 
> ---------- Forwarded message ----------
> From: Bjorn Helgaas <bhelgaas@xxxxxxxxxx>
> Date: 2012/6/7
> Subject: Re: [PATCH 02/11] PCI: Try to allocate mem64 above 4G at
> first
> To: Steven Newbury <steve@xxxxxxxxxxxxxxx>
> Cc: Yinghai Lu <yinghai@xxxxxxxxxx>, "H. Peter Anvin" <hpa@xxxxxxxxx>,
> David Miller <davem@xxxxxxxxxxxxx>, Tony Luck <tony.luck@xxxxxxxxx>,
> Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx>, Andrew Morton
> <akpm@xxxxxxxxxxxxxxxxxxxx>, linux-pci@xxxxxxxxxxxxxxx,
> linux-kernel@xxxxxxxxxxxxxxx
> 
> 
> On Wed, Jun 6, 2012 at 2:44 AM, Steven Newbury <steve@xxxxxxxxxxxxxxx>
> wrote:
> > On Tue,   5 Jun 2012, 06:04:57 BST, Yinghai Lu <yinghai@xxxxxxxxxx>
> wrote:
> >> > Linux has a long history of allocating bottom-up.  Windows has a
> long
> >> > history of allocating top-down.  You're proposing a third
> alternative,
> >> > allocating bottom-up starting at 4GB for 64-bit BARs.  If we
> change
> >> > this area, I would prefer something that follows Windows because
> I
> >> > think it will be closer to what's been tested by Windows.  Do you
> >> > think your alternative is better?
> >>
> >> hope we can figure out how windows is making it work.
> >>
> >> Steve, Can you check if Windows is working with your test case ?
> >>
> >> If it works, we may try do the same thing from Linux, so you will
> not
> >> need to append "pci=nocrs pci=alloc_high"...
> >>
> > Unfortunately I don't have a 64 bit version of Windows to test
> with.  Vista(32 bit) fails to even boot when docked, hot-plugging
> fails to allocate resources, but at least doesn't crash.
> >
> > From what I've read about the (64 bit) Windows allocation stragegy
> it's closer to Yinghai's method than the Linux default, preferring 64
> bit resources (>4G) when possible.  I'll try to find the specification
> document again.
> 
> 
> Here's the host bridge info from the BIOS (from
> https://bugzilla.kernel.org/show_bug.cgi?id=10461 attachment
> https://bugzilla.kernel.org/attachment.cgi?id=72869):
> 
> ACPI: PCI Root Bridge [PCI0] (domain 0000 [bus 00-ff])
> pci_root PNP0A03:00: host bridge window [io  0x0000-0x0cf7]
> pci_root PNP0A03:00: host bridge window [io  0x0d00-0xffff]
> pci_root PNP0A03:00: host bridge window [mem 0x000a0000-0x000bffff]
> pci_root PNP0A03:00: host bridge window [mem 0x000d0000-0x000dffff]
> pci_root PNP0A03:00: host bridge window [mem 0xe0000000-0xf7ffffff]
> pci_root PNP0A03:00: host bridge window [mem 0xfc000000-0xfebfffff]
> pci_root PNP0A03:00: host bridge window [mem 0xfec10000-0xfecfffff]
> pci_root PNP0A03:00: host bridge window [mem 0xfed1c000-0xfed1ffff]
> pci_root PNP0A03:00: host bridge window [mem 0xfed90000-0xfed9ffff]
> pci_root PNP0A03:00: host bridge window [mem 0xfed40000-0xfed44fff]
> pci_root PNP0A03:00: host bridge window [mem 0xfeda7000-0xfedfffff]
> pci_root PNP0A03:00: host bridge window [mem 0xfee10000-0xff9fffff]
> pci_root PNP0A03:00: host bridge window [mem 0xffc00000-0xffdfffff]
> 
> There's no aperture above 4GB.  So I don't think any version of
> Windows will ever assign a BAR above 4GB.
> --
> To unsubscribe from this list: send the line "unsubscribe
> linux-kernel" in
> the body of a message to majordomo@xxxxxxxxxxxxxxx
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/
> 
> 

Hope have any help...

Here have a document from MSDN talk about the pci allocate strategy on
Windows server 2003, XP and vista:
	http://msdn.microsoft.com/en-us/library/windows/hardware/gg462986.aspx


Per page 4, looks Microsoft have different strategy on different Windows
version

On XP and server 2003: First, they ignored BIOS's boot configuration and
allocate below 4G. If fail, then try to allocate above 4GB.

On Vista: it always respects the boot configuration of devices above 4
GB.

But, this document didn't cover the behavior on Windows 7, not sure it's
the same with Vista.


Thanks

Joey Lee



--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/