linux-kernel - Re: [PATCH] iommu/arm-smmu-v3: Fix L1 stream table index calculation for AmpereOne

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: <0e84f3c0-09d6-4485-ac76-ca296d1ee07e@os.amperecomputing.com>
Date: Tue, 1 Oct 2024 12:38:56 -0700
From: Yang Shi <yang@...amperecomputing.com>
To: Jason Gunthorpe <jgg@...pe.ca>
Cc: Nicolin Chen <nicolinc@...dia.com>, will@...nel.org,
 robin.murphy@....com, linux-arm-kernel@...ts.infradead.org,
 linux-kernel@...r.kernel.org
Subject: Re: [PATCH] iommu/arm-smmu-v3: Fix L1 stream table index calculation
 for AmpereOne



On 10/1/24 12:18 PM, Jason Gunthorpe wrote:
> On Tue, Oct 01, 2024 at 12:09:03PM -0700, Yang Shi wrote:
>>> Also, there are other places doing "1 << smmu->sid_bits", e.g.
>>> arm_smmu_init_strtab_linear().
>> The disassembly shows it uses "sbfiz x21, x20, 6, 32" instead of lsl. 1UL
>> should be used if we want to take extra caution and don't prefer rely on
>> compiler.
> Still, we should be fixing them all if sid_bits == 32, all those
> shifts should be throwing a UBSAN error. It would be crazy to have a

OK, will cover this is v2.

> 32 bit linear table but I guess it is allowed?

I'm not smmu expert, but if sid_bits is 32, it looks like the linear 
table will consume 256GB IIUC? That is crazy.

>
> Jason