Re: [RFC PATCH V2 3/3] PCI/ACPI: hisi: Add ACPI support for HiSilicon SoCs Host Controllers

From: Dongdong Liu
Date: Wed Aug 31 2016 - 23:27:42 EST



å 2016/9/1 6:56, Rafael J. Wysocki åé:
On Wednesday, August 31, 2016 07:48:14 PM Dongdong Liu wrote:
Add specific quirks for PCI config space accessors.This involves:
1. New initialization call hisi_pcie_acpi_init() to get RC config resource
with hardcoded range address and setup ecam mapping.
2. New entry in common quirk array.

Signed-off-by: Dongdong Liu <liudongdong3@xxxxxxxxxx>
Signed-off-by: Gabriele Paoloni <gabriele.paoloni@xxxxxxxxxx>

Well, what exactly is the ACPI support you're adding? Is it the ECAM part only
or is there anything more to it?


Hi Rafael, thanks for replying.

Our host bridge is non ECAM only for the RC bus config space;
for any other bus underneath the root bus we support ECAM access.

In our case we cannot use the standard MCFG object to pass the RC itself config space addresses.
The more discuss information can be found:
https://lkml.org/lkml/2016/2/22/1087
[...]
I have looked into this and in our case we cannot use the
standard MCFG object to pass the RC config space addresses.

The reason is that in our HW we have the config base addresses of the
root complex ports that are less than 0x100000 byte distant one from
the other as we only map the first 0x10000 bytes.
Now the MCFG acpi framework always fix the MCFG resource size to 0x100000
for each bus; therefore if we pass our RC addresses through MCFG we end
up with a resource conflict.
To give you a practical example we are in a situation where we have:

port0: [0x00000000b0080000 - 0x00000000b0080000 + 0x10000]
port1: [0x00000000b0090000 - 0x00000000b0090000 + 0x10000]
port2: [0x00000000b00A0000 - 0x00000000b00A0000 + 0x10000]
port3: [0x00000000b00B0000 - 0x00000000b00B0000 + 0x10000]
So if we pass the base addresses through MCFG the resources
will overlap as MCFG will consider 0x100000 size for each base
address of the root complex (only the RC bus uses that address)
So far I do not see many option other than using _DSD to pass
these RC config base addresses.

Thanks and Regards

Gab
[...]

and

https://patchwork.kernel.org/patch/9178791/
[...]
Furthermore, I suspect we do not even need a way to pass the
non-ECAM compliant config space resources to the OS (ie we can't
change FW anymore anyway in some platforms) so the quirks hooks
are likely to hardcode the required config space addresses for
the respective MCFG match.

......
Thanks !
Lorenzo
[...]

So we hard code with our RC itself config resource.

Thanks
Dongdong

Thanks,
Rafael


.