Re: [PATCH v5 2/2] PCI: HIP: Add handling of HiSilicon HIP PCIe controller errors

[Bjorn Helgaas]

On Fri, Apr 03, 2020 at 01:23:13PM +0300, Dan Carpenter wrote:
> On Wed, Mar 25, 2020 at 12:36:39PM -0500, Bjorn Helgaas wrote:
> > On Wed, Mar 25, 2020 at 01:55:18PM +0000, Shiju Jose wrote:
> > > The HiSilicon HIP PCIe controller is capable of handling errors
> > > on root port and perform port reset separately at each root port.
> > > 
> > > This patch add error handling driver for HIP PCIe controller to log
> > > and report recoverable errors. Perform root port reset and restore
> > > link status after the recovery.
> > > 
> > > Following are some of the PCIe controller's recoverable errors
> > > 1. completion transmission timeout error.
> > > 2. CRS retry counter over the threshold error.
> > > 3. ECC 2 bit errors
> > > 4. AXI bresponse/rresponse errors etc.
> > > 
> > > Also fix the following Smatch warning:
> > > warn: should '((((1))) << (9 + i))' be a 64 bit type?
> > > if (err->val_bits & BIT(HISI_PCIE_LOCAL_VALID_ERR_MISC + i))
> > >      ^^^ This should be BIT_ULL() because it goes up to 9 + 32.
> > > Reported-by: kbuild test robot <lkp@xxxxxxxxx>
> > > Reported-by: Dan Carpenter <dan.carpenter@xxxxxxxxxx>
> > 
> > I'm glad you did this fix, and thanks for acknowledging Dan, but I
> > don't think it's necessary to mention it in the commit log here
> > because it won't really be useful in the future.  It's only relevant
> > when comparing the unmerged versions of this series, e.g., v4 compared
> > to v3.

To elaborate on that a little, I think the commit log should describe
the change specifically made by the patch.  You should be able to
"git diff HEAD^" and match up the commit log with that diff output.
You can't do that with this Smatch paragraph.

> It's the kbuild template which suggests adding the Reported-by tags but
> you're right that it's not really appropriate for patches that haven't
> been merged yet.  I wish there were a correct tag.  I just saw yesterday
> where a maintainer insisted that someone add a Suggested-by tag and I
> don't think that's appropriate either.

Adding tags for every reviewer or bot comment seems like overkill.  I
think the "lore" links are about the right level of attribution for
this sort of thing, e.g., here:

  https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=77d6b9094819ba55353de0ef92957f3f54f2c36c

The Link: tag there gives you the whole v2 thread including review
comments.  And Matthew's cover letter even included a link to the
original v1 posting.  That seems perfect to me.

Bjorn