linux-kernel - Re: [PATCH v6 6/6] iommu/tegra-smmu: Add pagetable mappings to debugfs

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <20211007204131.GC20821@Asurada-Nvidia>
Date:   Thu, 7 Oct 2021 13:41:31 -0700
From:   Nicolin Chen <nicoleotsuka@...il.com>
To:     Thierry Reding <thierry.reding@...il.com>
Cc:     joro@...tes.org, will@...nel.org, vdumpa@...dia.com,
        jonathanh@...dia.com, linux-tegra@...r.kernel.org,
        iommu@...ts.linux-foundation.org, linux-kernel@...r.kernel.org,
        digetx@...il.com
Subject: Re: [PATCH v6 6/6] iommu/tegra-smmu: Add pagetable mappings to
 debugfs

On Thu, Oct 07, 2021 at 07:13:25PM +0200, Thierry Reding wrote:
> > @@ -496,6 +506,8 @@ static void tegra_smmu_as_unprepare(struct tegra_smmu *smmu,
> >  	mutex_unlock(&smmu->lock);
> >  }
> >  
> > +static const struct file_operations tegra_smmu_debugfs_mappings_fops;
> 
> Could the implementation be moved up here to avoid the forward
> declaration?

I thought that keeping all debugfs fops together would be preferable.
But yes, I will move it if you prefer no-additional forward declare.

> > +	seq_printf(s, "\nSWGROUP: %s\n", swgrp->name);
> > +	seq_printf(s, "as->id: %d\nas->attr: %c|%c|%s\nas->pd_dma: %pad\n", as->id,
> > +		   as->attr & SMMU_PD_READABLE ? 'R' : '-',
> > +		   as->attr & SMMU_PD_WRITABLE ? 'W' : '-',
> > +		   as->attr & SMMU_PD_NONSECURE ? "NS" : "S",
> > +		   &as->pd_dma);
> > +	seq_puts(s, "{\n");
> 
> Maybe this can be more compact by putting the name, ID, attributes and
> base address onto a single line? Maybe also use "'-' : 'S'" for the
> non-secure attribute to keep in line with what you've done for readable
> and writable attributes.

Okay. Will change that.

> Then again, this is going to be very verbose output anyway, so maybe it
> isn't worth it.

Are you saying the whole debugfs thing or just attributes? Yet, for
either case, I don't think so, as mappings info would help for sure
from our past experience while the attributes are just one line...