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: <ZwkvMIqC2DjLZJrg@PC2K9PVX.TheFacebook.com>
Date: Fri, 11 Oct 2024 09:59:12 -0400
From: Gregory Price <gourry@...rry.net>
To: Bjorn Helgaas <helgaas@...nel.org>
Cc: linux-pci@...r.kernel.org, linux-cxl@...r.kernel.org,
	linux-kernel@...r.kernel.org, lukas@...ner.de,
	dan.j.williams@...el.com, bhelgaas@...gle.com, dave@...olabs.net,
	dave.jiang@...el.com, vishal.l.verma@...el.com,
	Jonathan.Cameron@...wei.com
Subject: Re: [PATCH] PCI/DOE: Poll DOE Busy bit for up to 1 second in
 pci_doe_send_req

On Thu, Oct 10, 2024 at 05:16:28PM -0500, Bjorn Helgaas wrote:
> On Fri, Oct 04, 2024 at 12:28:28PM -0400, Gregory Price wrote:
> > During initial device probe, the PCI DOE busy bit for some CXL
> > devices may be left set for a longer period than expected by the
> > current driver logic. Despite local comments stating DOE Busy is
> > unlikely to be detected, it appears commonly specifically during
> > boot when CXL devices are being probed.
> > 
> > This was observed on a single socket AMD platform with 2 CXL memory
> > expanders attached to the single socket. It was not the case that
> > concurrent accesses were being made, as validated by monitoring
> > mailbox commands on the device side.
> > 
> > This behavior has been observed with multiple CXL memory expanders
> > from different vendors - so it appears unrelated to the model.
> > 
> > In all observed tests, only a small period of the retry window is
> > actually used - typically only a handful of loop iterations.
> > 
> > Polling on the PCI DOE Busy Bit for (at max) one PCI DOE timeout
> > interval (1 second), resolves this issues cleanly.
> > 
> > Per PCIe r6.2 sec 6.30.3, the DOE Busy Bit being cleared does not
> > raise an interrupt, so polling is the best option in this scenario.
> > 
> > Subsqeuent code in doe_statemachine_work and abort paths also wait
> > for up to 1 PCI DOE timeout interval, so this order of (potential)
> > additional delay is presumed acceptable.
> 
> I provisionally applied this to pci/doe for v6.13 with Lukas and
> Jonathan's reviewed-by.  
> 
> Can we include a sample of any dmesg logging or other errors users
> would see because of this problem?  I'll update the commit log with
> any of this information to help users connect an issue with this fix.
>

The only indication in dmesg you will see is a line like

[   24.542625] endpoint6: DOE failed -EBUSY

produced by cxl_cdat_get_length or cxl_cdat_read_table


Do you want an updated patch with the nits fixed?
 
> > Suggested-by: Lukas Wunner <lukas@...ner.de>
> > Signed-off-by: Gregory Price <gourry@...rry.net>
> > ---
> >  drivers/pci/doe.c | 14 +++++++++++++-
> >  1 file changed, 13 insertions(+), 1 deletion(-)
> > 
> > diff --git a/drivers/pci/doe.c b/drivers/pci/doe.c
> > index 652d63df9d22..27ba5d281384 100644
> > --- a/drivers/pci/doe.c
> > +++ b/drivers/pci/doe.c
> > @@ -149,14 +149,26 @@ static int pci_doe_send_req(struct pci_doe_mb *doe_mb,
> >  	size_t length, remainder;
> >  	u32 val;
> >  	int i;
> > +	unsigned long timeout_jiffies;
> >  
> >  	/*
> >  	 * Check the DOE busy bit is not set. If it is set, this could indicate
> >  	 * someone other than Linux (e.g. firmware) is using the mailbox. Note
> >  	 * it is expected that firmware and OS will negotiate access rights via
> >  	 * an, as yet to be defined, method.
> > +	 *
> > +	 * Wait up to one PCI_DOE_TIMEOUT period to allow the prior command to
> > +	 * finish. Otherwise, simply error out as unable to field the request.
> > +	 *
> > +	 * PCIe r6.2 sec 6.30.3 states no interrupt is raised when the DOE Busy
> > +	 * bit is cleared, so polling here is our best option for the moment.
> >  	 */
> > -	pci_read_config_dword(pdev, offset + PCI_DOE_STATUS, &val);
> > +	timeout_jiffies = jiffies + PCI_DOE_TIMEOUT;
> > +	do {
> > +		pci_read_config_dword(pdev, offset + PCI_DOE_STATUS, &val);
> > +	} while (FIELD_GET(PCI_DOE_STATUS_BUSY, val) &&
> > +		 !time_after(jiffies, timeout_jiffies));
> > +
> >  	if (FIELD_GET(PCI_DOE_STATUS_BUSY, val))
> >  		return -EBUSY;
> >  
> > -- 
> > 2.43.0
> > 

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ