Re: [PATCH v6 3/6] iommu/tegra-smmu: Rename struct tegra_smmu_swgroup *group to *swgrp

From: Nicolin Chen <nicoleotsuka@gmail.com>
To: Thierry Reding <thierry.reding@gmail.com>
Cc: joro@8bytes.org, will@kernel.org, vdumpa@nvidia.com,
	jonathanh@nvidia.com, linux-tegra@vger.kernel.org,
	iommu@lists.linux-foundation.org, linux-kernel@vger.kernel.org,
	digetx@gmail.com
Subject: Re: [PATCH v6 3/6] iommu/tegra-smmu: Rename struct tegra_smmu_swgroup *group to *swgrp
Date: Thu, 7 Oct 2021 13:29:45 -0700	[thread overview]
Message-ID: <20211007202945.GB20821@Asurada-Nvidia> (raw)
In-Reply-To: <YV8m+0qP2ibQBvNs@orome.fritz.box>

On Thu, Oct 07, 2021 at 06:57:31PM +0200, Thierry Reding wrote:
> On Mon, Sep 13, 2021 at 06:38:55PM -0700, Nicolin Chen wrote:
> > There are both tegra_smmu_swgroup and tegra_smmu_group structs
> > using "group" for their pointer instances. This gets confusing
> > to read the driver sometimes.
> > 
> > So this patch renames "group" of struct tegra_smmu_swgroup to
> > "swgrp" as a cleanup. Also renames its "find" function.
> > 
> > Note that we already have "swgroup" being used for an unsigned
> > int type variable that is inside struct tegra_smmu_swgroup, so
> > it's not able to use "swgroup" but only something like "swgrp".
> > 
> > Signed-off-by: Nicolin Chen <nicoleotsuka@gmail.com>
> > ---
> >  drivers/iommu/tegra-smmu.c | 34 +++++++++++++++++-----------------
> >  1 file changed, 17 insertions(+), 17 deletions(-)
> > 
> > diff --git a/drivers/iommu/tegra-smmu.c b/drivers/iommu/tegra-smmu.c
> > index a32ed347e25d..0f3883045ffa 100644
> > --- a/drivers/iommu/tegra-smmu.c
> > +++ b/drivers/iommu/tegra-smmu.c
> > @@ -334,35 +334,35 @@ static void tegra_smmu_domain_free(struct iommu_domain *domain)
> >  }
> >  
> >  static const struct tegra_smmu_swgroup *
> > -tegra_smmu_find_swgroup(struct tegra_smmu *smmu, unsigned int swgroup)
> > +tegra_smmu_find_swgrp(struct tegra_smmu *smmu, unsigned int swgroup)
> 
> This makes things inconsistent now. The tegra_smmu_find_swgroup() name
> indicates that we're looking for some "swgroup" entity within an "smmu"
> object. The entity that we're looking for is a struct tegra_smmu_swgroup
> so I think it makes sense to use that full name in the function name.

This is more like an indirect change to keep consistency between
function name and pointer name.

> >  {
> > -	const struct tegra_smmu_swgroup *group = NULL;
> > +	const struct tegra_smmu_swgroup *swgrp = NULL;
> 
> I don't think the existing naming is confusing. The variable name
> "group" is consistently used for tegra_smmu_swgroup structures and there
> are no cases where we would confuse them with struct tegra_smmu_group
> instances.

If we don't rename it, then PATCH-4 adds to struct tegra_smmu_group
a "struct tegra_smmu_swgroup *group", which results in a confusing
group->group...

> However, I don't feel strongly about it, so I'm fine with changing the
> variable names to "swgrp" if you think that makes things less confusing.

Yea, I'd like to keep this change. I will respin it in next version
after fixing other comments.