linux-kernel - Re: [V3] PCI: fu740: Drop to 2.5GT/s to fix initial device probing on some boards

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:   Thu, 14 Apr 2022 01:10:33 +0100 (BST)
From:   "Maciej W. Rozycki" <macro@...am.me.uk>
To:     Ben Dooks <ben.dooks@...ethink.co.uk>
cc:     Alexandre Ghiti <alex@...ti.fr>, linux-pci@...r.kernel.org,
        linux-kernel@...r.kernel.org, linux-riscv@...ts.infradead.org,
        Bjorn Helgaas <bhelgaas@...gle.com>,
        Palmer Dabbelt <palmer@...belt.com>,
        Rob Herring <robh@...nel.org>,
        Lorenzo Pieralisi <lorenzo.pieralisi@....com>,
        Greentime Hu <greentime.hu@...ive.com>,
        Paul Walmsley <paul.walmsley@...ive.com>,
        David Abdurachmanov <david.abdurachmanov@...ive.com>,
        Neill Whillans <neill.whillans@...ethink.co.uk>
Subject: Re: [V3] PCI: fu740: Drop to 2.5GT/s to fix initial device probing
 on some boards

On Wed, 23 Mar 2022, Ben Dooks wrote:

> > FWIW, I have tested this and it solved my issue with nvme not being probed,
> > so:
> > 
> > Tested-by: Alexandre Ghiti <alexandre.ghiti@...onical.com>
> 
> Ok, great. Our test rig seems to be still working with this.

 I ran simple verification of your change by interrupting U-Boot after a 
power-up and issuing:

=> setenv boot_pci_enum true

at the command prompt before booting from the uSD card and curiously 
enough the root port comes up with the Link Capabilities Register 
reporting the lack of Link Bandwidth Notification Capability in this 
scenario, while it reports its presence if booted undisturbed, i.e.:

		LnkCap: Port #0, Speed 8GT/s, Width x8, ASPM L0s L1, Exit Latency L0s <4us, L1 <4us
			ClockPM- Surprise- LLActRep+ BwNot- ASPMOptComp+

vs:

		LnkCap: Port #0, Speed 8GT/s, Width x8, ASPM L0s L1, Exit Latency L0s <4us, L1 <4us
			ClockPM- Surprise- LLActRep+ BwNot+ ASPMOptComp+

It is fully reproducible.

 Any idea what might be causing it?  I can't see it being explicitly set 
or cleared anywhere, be it in U-Boot or Linux, so it must be done by the 
device itself depending on something.  And the lack of this capability 
seems to me like non-compliance for a multiple-lane, multiple-speed 
device.

  Maciej