[4.4 PATCH] powerpc/eeh: Null check uses of eeh_pe_bus_get

[4.4 PATCH] powerpc/eeh: Null check uses of eeh_pe_bus_get

[Russell Currey]

eeh_pe_bus_get() can return NULL if a PCI bus isn't found for a given PE.
Some callers don't check this, and can cause a null pointer dereference
under certain circumstances.

Fix this by checking NULL everywhere eeh_pe_bus_get() is called.

Fixes: 8a6b1bc70dbb ("powerpc/eeh: EEH core to handle special event")
Cc: stable@vger.kernel.org # v3.11+
Signed-off-by: Russell Currey <ruscur@russell.cc>
Reviewed-by: Andrew Donnellan <andrew.donnellan@au1.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
---
this specific patch for 4.4 only
---
 arch/powerpc/kernel/eeh_driver.c             | 8 ++++++++
 arch/powerpc/platforms/powernv/eeh-powernv.c | 5 +++++
 2 files changed, 13 insertions(+)

diff --git a/arch/powerpc/kernel/eeh_driver.c b/arch/powerpc/kernel/eeh_driver.c
index 247a0dc..c07bfb5 100644
--- a/arch/powerpc/kernel/eeh_driver.c
+++ b/arch/powerpc/kernel/eeh_driver.c
@@ -909,6 +909,14 @@ static void eeh_handle_special_event(void)
 				/* Notify all devices to be down */
 				eeh_pe_state_clear(pe, EEH_PE_PRI_BUS);
 				bus = eeh_pe_bus_get(phb_pe);
+				if (!bus) {
+					pr_err("%s: Cannot find PCI bus for "
+					       "PHB#%d-PE#%x\n",
+					       __func__,
+					       pe->phb->global_number,
+					       pe->addr);
+					break;
+				}
 				eeh_pe_dev_traverse(pe,
 					eeh_report_failure, NULL);
 				pcibios_remove_pci_devices(bus);
diff --git a/arch/powerpc/platforms/powernv/eeh-powernv.c b/arch/powerpc/platforms/powernv/eeh-powernv.c
index 2ba60259..4d822f4 100644
--- a/arch/powerpc/platforms/powernv/eeh-powernv.c
+++ b/arch/powerpc/platforms/powernv/eeh-powernv.c
@@ -956,6 +956,11 @@ static int pnv_eeh_reset(struct eeh_pe *pe, int option)
 		}
 
 		bus = eeh_pe_bus_get(pe);
+		if (!bus) {
+			pr_err("%s: Cannot find PCI bus for PHB#%d-PE#%x\n",
+			       __func__, pe->phb->global_number, pe->addr);
+			return -EIO;
+		}
 		if (pci_is_root_bus(bus) ||
 			pci_is_root_bus(bus->parent))
 			ret = pnv_eeh_root_reset(hose, option);
-- 
2.10.1

Re: [4.4 PATCH] powerpc/eeh: Null check uses of eeh_pe_bus_get

[Greg KH]

On Thu, Oct 27, 2016 at 10:19:45AM +1100, Russell Currey wrote:
> eeh_pe_bus_get() can return NULL if a PCI bus isn't found for a given PE.
> Some callers don't check this, and can cause a null pointer dereference
> under certain circumstances.
> 
> Fix this by checking NULL everywhere eeh_pe_bus_get() is called.
> 
> Fixes: 8a6b1bc70dbb ("powerpc/eeh: EEH core to handle special event")
> Cc: stable@vger.kernel.org # v3.11+
> Signed-off-by: Russell Currey <ruscur@russell.cc>
> Reviewed-by: Andrew Donnellan <andrew.donnellan@au1.ibm.com>
> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
> ---
> this specific patch for 4.4 only

Why?  I can't take a "specific patch only" unless you have a whole lot
of good description about why, and how, and what, and everything else...

Why can't I just take a patch that is already in Linus's tree?  Whenever
we don't do that, it always (yes really), ends up being wrong.

thanks,

greg k-h

Re: [4.4 PATCH] powerpc/eeh: Null check uses of eeh_pe_bus_get

[Russell Currey]

On Thu, 2016-10-27 at 07:12 +0200, Greg KH wrote:
> On Thu, Oct 27, 2016 at 10:19:45AM +1100, Russell Currey wrote:
> > eeh_pe_bus_get() can return NULL if a PCI bus isn't found for a given PE.
> > Some callers don't check this, and can cause a null pointer dereference
> > under certain circumstances.
> > 
> > Fix this by checking NULL everywhere eeh_pe_bus_get() is called.
> > 
> > Fixes: 8a6b1bc70dbb ("powerpc/eeh: EEH core to handle special event")
> > Cc: stable@vger.kernel.org # v3.11+
> > Signed-off-by: Russell Currey <ruscur@russell.cc>
> > Reviewed-by: Andrew Donnellan <andrew.donnellan@au1.ibm.com>
> > Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
> > ---
> > this specific patch for 4.4 only
> 
> Why?  I can't take a "specific patch only" unless you have a whole lot
> of good description about why, and how, and what, and everything else...
> 
> Why can't I just take a patch that is already in Linus's tree?  Whenever
> we don't do that, it always (yes really), ends up being wrong.
> 
> thanks,
> 
> greg k-h

Sorry, that was worded poorly.  This patch is commit
04fec21c06e35b169a83e75a84a015ab4606bf5e upstream.

What I _meant_ is that this patch specifically applies to 4.4.

Re: [4.4 PATCH] powerpc/eeh: Null check uses of eeh_pe_bus_get

[Greg KH]

On Thu, Oct 27, 2016 at 04:22:15PM +1100, Russell Currey wrote:
> On Thu, 2016-10-27 at 07:12 +0200, Greg KH wrote:
> > On Thu, Oct 27, 2016 at 10:19:45AM +1100, Russell Currey wrote:
> > > eeh_pe_bus_get() can return NULL if a PCI bus isn't found for a given PE.
> > > Some callers don't check this, and can cause a null pointer dereference
> > > under certain circumstances.
> > > 
> > > Fix this by checking NULL everywhere eeh_pe_bus_get() is called.
> > > 
> > > Fixes: 8a6b1bc70dbb ("powerpc/eeh: EEH core to handle special event")
> > > Cc: stable@vger.kernel.org # v3.11+
> > > Signed-off-by: Russell Currey <ruscur@russell.cc>
> > > Reviewed-by: Andrew Donnellan <andrew.donnellan@au1.ibm.com>
> > > Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
> > > ---
> > > this specific patch for 4.4 only
> > 
> > Why?��I can't take a "specific patch only" unless you have a whole lot
> > of good description about why, and how, and what, and everything else...
> > 
> > Why can't I just take a patch that is already in Linus's tree?��Whenever
> > we don't do that, it always (yes really), ends up being wrong.
> > 
> > thanks,
> > 
> > greg k-h
> 
> Sorry, that was worded poorly.  This patch is commit
> 04fec21c06e35b169a83e75a84a015ab4606bf5e upstream.
> 
> What I _meant_ is that this patch specifically applies to 4.4.

Ah, that makes a lot more sense, thanks :)

I'll queue this up after this next round of kernels is released in a day
or so.

thanks,

greg k-h