linux-kernel - Re: [PATCH v1 1/2] iommu/tegra-smmu: Defer attachment of display clients

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

Message-ID: <7c1ebbf2-5d27-fa2a-3cae-bedab4c8e30a@gmail.com>
Date:   Fri, 9 Apr 2021 15:55:41 +0300
From:   Dmitry Osipenko <digetx@...il.com>
To:     Thierry Reding <thierry.reding@...il.com>
Cc:     Joerg Roedel <joro@...tes.org>,
        Jonathan Hunter <jonathanh@...dia.com>,
        Krishna Reddy <vdumpa@...dia.com>,
        Nicolin Chen <nicoleotsuka@...il.com>,
        Will Deacon <will@...nel.org>,
        iommu@...ts.linux-foundation.org, linux-tegra@...r.kernel.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH v1 1/2] iommu/tegra-smmu: Defer attachment of display
 clients

08.04.2021 17:07, Dmitry Osipenko пишет:
>> Whatever happened to the idea of creating identity mappings based on the
>> obscure tegra_fb_mem (or whatever it was called) command-line option? Is
>> that command-line not universally passed to the kernel from bootloaders
>> that initialize display?
> This is still a good idea! The command-line isn't universally passed
> just because it's up to a user to override the cmdline and then we get a
> hang (a very slow boot in reality) on T30 since display client takes out
> the whole memory bus with the constant SMMU faults. For example I don't
> have that cmdline option in my current setups.
> 

Thinking a bit more about the identity.. have you consulted with the
memory h/w people about whether it's always safe to enable ASID in a
middle of DMA request?