[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <5a173967-098a-9e40-1373-5502506488f4@cogentembedded.com>
Date: Sat, 18 Jul 2020 11:44:25 +0300
From: Sergei Shtylyov <sergei.shtylyov@...entembedded.com>
To: Tiezhu Yang <yangtiezhu@...ngson.cn>,
Lorenzo Pieralisi <lorenzo.pieralisi@....com>,
Rob Herring <robh@...nel.org>,
Bjorn Helgaas <bhelgaas@...gle.com>,
Thomas Bogendoerfer <tsbogend@...ha.franken.de>,
Huacai Chen <chenhc@...ote.com>,
Jiaxun Yang <jiaxun.yang@...goat.com>
Cc: linux-mips@...r.kernel.org, linux-pci@...r.kernel.org,
linux-kernel@...r.kernel.org, Xuefeng Li <lixuefeng@...ngson.cn>
Subject: Re: [PATCH v4] PCI: loongson: Use DECLARE_PCI_FIXUP_EARLY for
bridge_class_quirk()
On 18.07.2020 11:37, Tiezhu Yang wrote:
> According to the datasheet of Loongson LS7A bridge chip, the old version
> of Loongson LS7A PCIE port has a wrong value about PCI class which is
> 0x060000, the correct value should be 0x060400, this bug can be fixed by
> "dev->class = PCI_CLASS_BRIDGE_PCI << 8;" at the software level. As far
> as I know, the latest version of LS7A has already fixed the value at the
> hardware level.
>
> In order to maintain downward compatibility, use DECLARE_PCI_FIXUP_EARLY
> instead of DECLARE_PCI_FIXUP_HEADER for bridge_class_quirk() to fix it as
> early as possible.
>
> Otherwise, in the function pci_setup_device(), the related code about
> "dev->class" such as "class = dev->class >> 8;" and "dev->transparent
> = ((dev->class & 0xff) == 1);" maybe get wrong value due to without
"Due to" not needed there.
> EARLY fixup.
>
> Fixes: 1f58cca5cf2b ("PCI: Add Loongson PCI Controller support")
> Signed-off-by: Tiezhu Yang <yangtiezhu@...ngson.cn>
[...]
MBR, Sergei
Powered by blists - more mailing lists