linux-kernel - Re: [PATCH v2] PCI: Fix up L1SS capability for Intel Apollolake PCIe bridge

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, 29 Dec 2022 13:07:25 -0600
From:   Bjorn Helgaas <helgaas@...nel.org>
To:     "Lee, Ron" <ron.lee@...el.com>
Cc:     "bhelgaas@...gle.com" <bhelgaas@...gle.com>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        "linux-pci@...r.kernel.org" <linux-pci@...r.kernel.org>,
        "lmajczak@...gle.com" <lmajczak@...gle.com>,
        "Jain, Rajat" <rajatja@...gle.com>,
        Ron Lee <ron.lee.intel@...il.com>
Subject: Re: [PATCH v2] PCI: Fix up L1SS capability for Intel Apollolake PCIe
 bridge

On Tue, Dec 20, 2022 at 09:11:31AM +0000, Lee, Ron wrote:
> > On Fri, Dec 16, 2022 at 04:29:39PM +0000, Lee, Ron wrote:
> > > > On Thu, Dec 15, 2022 at 05:13:57PM +0800, Ron Lee wrote:
> > Even if you haven't seen a battery life issue, I suspect you might be 
> > able to measure a power consumption difference if you looked for it 
> > and likely could see issues with manual ASPM enable/disable using 
> > sysfs.  That might be a legitimate reason for this quirk, and if
> > it is, we should mention it here.
>
> We can arrange the power measurement, but I doubt this quirk has
> correlation to power consumption.  My point is that the ASPM
> behavior is not changed with or without this quirk.

Makes sense.

> ...
> I agree that, I ever try to recover the link by traversing list, but
> it didn't work and the capability list have no method to do reverse
> traversal.  One approach may save the whole capability list before
> suspend, and check each capability link then restore the missing one
> after resume.  Do you think it's practical ? It is appreciated if
> you could recommend a practical solution for this issue.

The issue being "lspci doesn't show L1SS after suspend/resume"?

Is the point of this basically to fix lspci output after
suspend/resume?  Or is there something else this fixes?

It sounds like ASPM and L1SS works correctly after suspend/resume
even without this patch?

Bjorn