[PATCH 1/2] powerpc/eeh: Enable PCI_COMMAND_MASTER for PCI bridges

Subsystems: linux for powerpc (32-bit and 64-bit), pci enhanced error handling (eeh) for powerpc, the rest

DORMANTno replies

2 messages, 1 author, 2013-11-12 · open the first message on its own page

[PATCH 1/2] powerpc/eeh: Enable PCI_COMMAND_MASTER for PCI bridges

From: Gavin Shan <hidden>
Date: 2013-11-12 06:49:34

On PHB3, we will fail to fetch IODA tables without PCI_COMMAND_MASTER
on PCI bridges. According to one experiment I had, the MSIx interrupts
didn't raise from the adapter without the bit applied to all upstream
PCI bridges including root port of the adapter. The patch forces to
have that bit enabled accordingly.

Signed-off-by: Gavin Shan <redacted>
---
 arch/powerpc/kernel/eeh.c |    9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/arch/powerpc/kernel/eeh.c b/arch/powerpc/kernel/eeh.c
index 1fb331d..180af13 100644
--- a/arch/powerpc/kernel/eeh.c
+++ b/arch/powerpc/kernel/eeh.c

@@ -687,6 +687,15 @@ void eeh_save_bars(struct eeh_dev *edev)
 
 	for (i = 0; i < 16; i++)
 		eeh_ops->read_config(dn, i * 4, 4, &edev->config_space[i]);
+
+	/*
+	 * For PCI bridges including root port, we need enable bus
+	 * master explicitly. Otherwise, it can't fetch IODA table
+	 * entries correctly. So we cache the bit in advance so that
+	 * we can restore it after reset, either PHB range or PE range.
+	 */
+	if (edev->mode & EEH_DEV_BRIDGE)
+		edev->config_space[1] |= PCI_COMMAND_MASTER;
 }
 
 /**

-- 
1.7.9.5

[PATCH 2/2] powerpc/eeh: More accurate log

From: Gavin Shan <hidden>
Date: 2013-11-12 06:49:34

We possibly has fenced PHB except frozen PE. So the output log
doesn't cover all cases and the patch fixes it.

Signed-off-by: Gavin Shan <redacted>
---
 arch/powerpc/kernel/eeh_event.c |    9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/arch/powerpc/kernel/eeh_event.c b/arch/powerpc/kernel/eeh_event.c
index d27c5af..72d748b 100644
--- a/arch/powerpc/kernel/eeh_event.c
+++ b/arch/powerpc/kernel/eeh_event.c

@@ -74,8 +74,13 @@ static int eeh_event_handler(void * dummy)
 		pe = event->pe;
 		if (pe) {
 			eeh_pe_state_mark(pe, EEH_PE_RECOVERING);
-			pr_info("EEH: Detected PCI bus error on PHB#%d-PE#%x\n",
-				 pe->phb->global_number, pe->addr);
+			if (pe->type & EEH_PE_PHB)
+				pr_info("EEH: Detected error on PHB#%d\n",
+					 pe->phb->global_number);
+			else
+				pr_info("EEH: Detected PCI bus error on "
+					"PHB#%d-PE#%x\n",
+					pe->phb->global_number, pe->addr);
 			eeh_handle_event(pe);
 			eeh_pe_state_clear(pe, EEH_PE_RECOVERING);
 		} else {

-- 
1.7.9.5

`h`	back out one level
`j`	next message in thread
`k`	previous message in thread
`l`	drill in
`Esc`	close help / fold thread tree
`?`	toggle this help