linux-kernel - Re: FYI: i.MX8MP ISP (RKISP1) MI registers corruption: resolved

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: <m3zfc4sp0o.fsf@t19.piap.pl>
Date: Tue, 12 Aug 2025 14:28:07 +0200
From: Krzysztof Hałasa <khalasa@...p.pl>
To: Laurent Pinchart <laurent.pinchart@...asonboard.com>
Cc: Stefan Klug <stefan.klug@...asonboard.com>,  Dafna Hirschfeld
 <dafna@...tmail.com>,  Heiko Stuebner <heiko@...ech.de>,  Paul Elder
 <paul.elder@...asonboard.com>,  Jacopo Mondi
 <jacopo.mondi@...asonboard.com>,  Ondrej Jirman <megi@....cz>,
  linux-media@...r.kernel.org,  linux-rockchip@...ts.infradead.org,
  linux-arm-kernel@...ts.infradead.org,  linux-kernel@...r.kernel.org
Subject: Re: FYI: i.MX8MP ISP (RKISP1) MI registers corruption: resolved

Hi Laurent,

>> - anyway, lowering the frequencies of ISP and MEDIA root clocks fixes
>>   the ISP2 MI corruption. I'm currently investigating PMIC settings
>>   (both my Compulab and SolidRun modules use PCA9450C PMICs), so perhaps
>>   I'll be able to use the higher 500 MHz clocks. It doesn't matter much,
>>   though.

Well, apparently my Compulab UCM module provides overdrive power. At
least the PMIC driver says so (in /sys pseudofiles). I guess SolidRun
won't be different.

1. At least two different designs are affected (Compulab and SolidRun),
2. the PCA9450C is THE i.MX8MP PMIC, most probably connected to the CPU
   exactly as in the datasheet (at least WRT VDD_SOC power lines),
3. NXP people experienced the problem on their (which ones?) boards, too

Well... They specify (IMX8MPIEC = the datasheet):

"Two instances of 4-lane MIPI CSI interface and HDR ISP
• For single Camera, MIPI CSI 1 can support up to 400/500 MHz pixel
  clock in the Nominal/Overdrive mode.
• For single Camera, MIPI CSI 2 can support up to 277 MHz pixel clock.
• For dual Camera, both MIPI CSI can support up to 266 MHz pixel clock.
• 2x ISP supporting 375 Mpixel/s aggregate performance and up to
  3-exposure HDR processing.
    •When one camera is used, support up to 12MP@...ps or 4kp45
    •When two cameras are used, each supports up to 1080p80"

So, while ISP clock is not exactly pixel clock, perhaps they mean that
the ISP clock is to be limited as well.

ISP2 at 400 MHz (single camera) works for me (with a 1080p60 12-bit
IMX462 sensor), but I haven't tested in extreme conditions etc.

> I think it would make sense to lower the default clock frequencies, and
> provide an overlay to enable overdrive mode.
>
> It's also interesting that the issue only affected the second ISP, as
> the first one should also be limited to 400 MHz in normal mode.

Right. Maybe it's not the overdrive mode after all. I will do more tests
soon.
-- 
Krzysztof "Chris" Hałasa

Sieć Badawcza Łukasiewicz
Przemysłowy Instytut Automatyki i Pomiarów PIAP
Al. Jerozolimskie 202, 02-486 Warszawa