linux-kernel - Re: [PATCH v3 1/2] x86/ioremap: fix the pfn calculation mistake in __ioremap_check

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]

Date:   Fri, 2 Jul 2021 07:49:01 -0700
From:   Dave Hansen <dave.hansen@...el.com>
To:     Yaohui Wang <yaohuiwang@...ux.alibaba.com>, tglx@...utronix.de
Cc:     luto@...nel.org, peterz@...radead.org, mingo@...hat.com,
        bp@...en8.de, x86@...nel.org, linux-kernel@...r.kernel.org,
        luoben@...ux.alibaba.com, Tom Lendacky <thomas.lendacky@....com>,
        Brijesh Singh <brijesh.singh@....com>
Subject: Re: [PATCH v3 1/2] x86/ioremap: fix the pfn calculation mistake in
 __ioremap_check_ram()

On 7/2/21 3:05 AM, Yaohui Wang wrote:
> On 2021/7/1 22:41, Dave Hansen wrote:
>> Was this issue found by inspection, or is it causing an actual problem
>> in practice?
> 
> This issus truly caused terrible perforamnce downgrade in the practice.
> When developing an out of tree module in our testing environment,
> invoking ioremap() on normal RAM causes apparent CLI lag. The Unixbench
> score also decreases a lot (5x slowdown in the worst case).
> 
> Debugging such performance issue is extremely difficult, especially when
> the code of the faulty module itself is already very complex. I tested
> the system in many aspects before finally located this problem.

Do you know why this check:

> 	if ((res->flags & IORESOURCE_SYSTEM_RAM) != IORESOURCE_SYSTEM_RAM)
> 		return false;

did not catch your out-of-tree driver's errant ioremap()?