netdev - Re: [PATCH V3 04/10] PCI/TPH: Add pci=nostmode to force No ST Mode

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: <20240723224408.GA779931@bhelgaas>
Date: Tue, 23 Jul 2024 17:44:08 -0500
From: Bjorn Helgaas <helgaas@...nel.org>
To: Wei Huang <wei.huang2@....com>
Cc: linux-pci@...r.kernel.org, linux-kernel@...r.kernel.org,
	linux-doc@...r.kernel.org, netdev@...r.kernel.org,
	Jonathan.Cameron@...wei.com, corbet@....net, davem@...emloft.net,
	edumazet@...gle.com, kuba@...nel.org, pabeni@...hat.com,
	alex.williamson@...hat.com, gospo@...adcom.com,
	michael.chan@...adcom.com, ajit.khaparde@...adcom.com,
	somnath.kotur@...adcom.com, andrew.gospodarek@...adcom.com,
	manoj.panicker2@....com, Eric.VanTassell@....com,
	vadim.fedorenko@...ux.dev, horms@...nel.org, bagasdotme@...il.com,
	bhelgaas@...gle.com
Subject: Re: [PATCH V3 04/10] PCI/TPH: Add pci=nostmode to force No ST Mode

On Wed, Jul 17, 2024 at 03:55:05PM -0500, Wei Huang wrote:
> When "No ST mode" is enabled, endpoint devices can generate TPH headers
> but with all steering tags treated as zero. A steering tag of zero is
> interpreted as "using the default policy" by the root complex. This is
> essential to quantify the benefit of steering tags for some given
> workloads.

Capitalize technical terms defined by spec.

> +++ b/Documentation/admin-guide/kernel-parameters.txt
> @@ -4656,6 +4656,7 @@
>  		norid		[S390] ignore the RID field and force use of
>  				one PCI domain per PCI function
>  		notph		[PCIE] Do not use PCIe TPH
> +		nostmode	[PCIE] Force TPH to use No ST Mode

Needs a little more context here about what this means.  Users won't
know where to even look for "No ST Mode" unless they have a copy of
the spec.

> +++ b/drivers/pci/pci-driver.c
> @@ -324,8 +324,13 @@ static long local_pci_probe(void *_ddi)
>  	pci_dev->driver = pci_drv;
>  	rc = pci_drv->probe(pci_dev, ddi->id);
>  	if (!rc) {
> -		if (pci_tph_disabled())
> +		if (pci_tph_disabled()) {
>  			pcie_tph_disable(pci_dev);
> +			return rc;
> +		}
> +
> +		if (pci_tph_nostmode())
> +			pcie_tph_set_nostmode(pci_dev);

Same comment here; can we do this outside the probe() path somehow?

>  		return rc;
>  	}
> diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
> index 4cbfd5b53be8..8745ce1c4a9a 100644
> --- a/drivers/pci/pci.c
> +++ b/drivers/pci/pci.c
> @@ -160,6 +160,9 @@ static bool pcie_ats_disabled;
>  /* If set, the PCIe TPH capability will not be used. */
>  static bool pcie_tph_disabled;
>  
> +/* If TPH is enabled, "No ST Mode" will be enforced. */
> +static bool pcie_tph_nostmode;
> +
>  /* If set, the PCI config space of each device is printed during boot. */
>  bool pci_early_dump;
>  
> @@ -175,6 +178,12 @@ bool pci_tph_disabled(void)
>  }
>  EXPORT_SYMBOL_GPL(pci_tph_disabled);
>  
> +bool pci_tph_nostmode(void)
> +{
> +	return pcie_tph_nostmode;
> +}
> +EXPORT_SYMBOL_GPL(pci_tph_nostmode);

s/pci/pcie/

Unexport unless it's useful for drivers.

Bjorn