linux-kernel - Re: [PATCH RFC 54/77] ntb: Ensure number of MSIs on SNB is enough for the link interrupt

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:	Mon, 7 Oct 2013 20:38:45 +0200
From:	Alexander Gordeev <agordeev@...hat.com>
To:	Jon Mason <jon.mason@...el.com>
Cc:	linux-kernel@...r.kernel.org, Bjorn Helgaas <bhelgaas@...gle.com>,
	Ralf Baechle <ralf@...ux-mips.org>,
	Michael Ellerman <michael@...erman.id.au>,
	Benjamin Herrenschmidt <benh@...nel.crashing.org>,
	Martin Schwidefsky <schwidefsky@...ibm.com>,
	Ingo Molnar <mingo@...hat.com>, Tejun Heo <tj@...nel.org>,
	Dan Williams <dan.j.williams@...el.com>,
	Andy King <acking@...are.com>,
	Matt Porter <mporter@...nel.crashing.org>,
	stable@...r.kernel.org, linux-pci@...r.kernel.org,
	linux-mips@...ux-mips.org, linuxppc-dev@...ts.ozlabs.org,
	linux390@...ibm.com, linux-s390@...r.kernel.org, x86@...nel.org,
	linux-ide@...r.kernel.org, iss_storagedev@...com,
	linux-nvme@...ts.infradead.org, linux-rdma@...r.kernel.org,
	netdev@...r.kernel.org, e1000-devel@...ts.sourceforge.net,
	linux-driver@...gic.com,
	Solarflare linux maintainers <linux-net-drivers@...arflare.com>,
	"VMware, Inc." <pv-drivers@...are.com>, linux-scsi@...r.kernel.org
Subject: Re: [PATCH RFC 54/77] ntb: Ensure number of MSIs on SNB is enough
 for the link interrupt

On Mon, Oct 07, 2013 at 09:50:57AM -0700, Jon Mason wrote:
> On Sat, Oct 05, 2013 at 11:43:04PM +0200, Alexander Gordeev wrote:
> > On Wed, Oct 02, 2013 at 05:48:05PM -0700, Jon Mason wrote:
> > > On Wed, Oct 02, 2013 at 12:49:10PM +0200, Alexander Gordeev wrote:
> > > > Signed-off-by: Alexander Gordeev <agordeev@...hat.com>
> > > > ---
> > > >  drivers/ntb/ntb_hw.c |    2 +-
> > > >  1 files changed, 1 insertions(+), 1 deletions(-)
> > > > 
> > > > diff --git a/drivers/ntb/ntb_hw.c b/drivers/ntb/ntb_hw.c
> > > > index de2062c..eccd5e5 100644
> > > > --- a/drivers/ntb/ntb_hw.c
> > > > +++ b/drivers/ntb/ntb_hw.c
> > > > @@ -1066,7 +1066,7 @@ static int ntb_setup_msix(struct ntb_device *ndev)
> > > >  		/* On SNB, the link interrupt is always tied to 4th vector.  If
> > > >  		 * we can't get all 4, then we can't use MSI-X.
> > > >  		 */
> > > > -		if (ndev->hw_type != BWD_HW) {
> > > > +		if ((rc < SNB_MSIX_CNT) && (ndev->hw_type != BWD_HW)) {
> > > 
> > > Nack, this check is unnecessary.
> > 
> > If SNB can do more than SNB_MSIX_CNT MSI-Xs then this check is needed
> > to enable less than maximum MSI-Xs in case the maximum was not allocated.
> > Otherwise SNB will fallback to single MSI instead of multiple MSI-Xs.
> 
> Per the comment in the code snippet above, "If we can't get all 4,
> then we can't use MSI-X".  There is already a check to see if more
> than 4 were acquired.  So it's not possible to hit this.  Even if it
> was, don't use SNB_MSIX_CNT here (limits.msix_cnt is the preferred
> variable).  Also, the "()" are unnecessary.

The changelog is definitely bogus. I meant here an improvement to the
existing scheme, not a conversion to the new one:

	msix_entries = msix_table_size(val);

Getting i.e. 16 vectors here.

	if (msix_entries > ndev->limits.msix_cnt) {
		rc = -EINVAL;
		goto err;
	}

Upper limit check i.e. succeeds.

	[...]

	rc = pci_enable_msix(pdev, ndev->msix_entries, msix_entries);

pci_enable_msix() does not success and returns i.e. 8 here, should retry.

	if (rc < 0)
		goto err1;
	if (rc > 0) {
		/* On SNB, the link interrupt is always tied to 4th vector.  If
		 * we can't get all 4, then we can't use MSI-X.
		 */
		if (ndev->hw_type != BWD_HW) {

On SNB bail out here, although could have continue with 8 vectors.
Can only use SNB_MSIX_CNT here, since limits.msix_cnt is the upper limit.

			rc = -EIO;
			goto err1;
		}

		[...]
	}

-- 
Regards,
Alexander Gordeev
agordeev@...hat.com
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/