[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <aPr6dBDUUohRUzYg@wunner.de>
Date: Fri, 24 Oct 2025 06:03:00 +0200
From: Lukas Wunner <lukas@...ner.de>
To: Shuai Xue <xueshuai@...ux.alibaba.com>
Cc: Bjorn Helgaas <bhelgaas@...gle.com>, linux-pci@...r.kernel.org,
linux-kernel@...r.kernel.org, linuxppc-dev@...ts.ozlabs.org,
kbusch@...nel.org, sathyanarayanan.kuppuswamy@...ux.intel.com,
mahesh@...ux.ibm.com, oohall@...il.com, Jonathan.Cameron@...wei.com,
terry.bowman@....com, tianruidong@...ux.alibaba.com
Subject: Re: [PATCH v6 4/5] PCI/ERR: Use pcie_aer_is_native() to check for
native AER control
On Fri, Oct 24, 2025 at 11:38:10AM +0800, Shuai Xue wrote:
> The remaining question is whether it would make more sense to rename
> pcie_clear_device_status() to pci_clear_device_error_status() and refine
> its behavior by adding a mask specifically for bits 0 to 3. Here's an
> example of the proposed change:
I don't see much value in renaming the function.
However clearing only bits 0-3 makes sense. PCIe r5.0 defined bit 6
as Emergency Power Reduction Detected with type RW1C in 2019. The
last time we touched pcie_clear_device_status() was in 2018 with
ec752f5d54d7 and we've been clearing all bits since forever,
not foreseeing that new ones with type RW1C might be added later.
I suggest defining a new macro in include/uapi/linux/pci_regs.h
instead of using 0xf, say PCI_EXP_DEVSTA_ERR. Then you don't
need the code comment because the code is self-explanatory.
Thanks,
Lukas
Powered by blists - more mailing lists