diff mbox

[3/3] powerpc/eeh: Escalate error on non-existing PE

Message ID 1389507226-32002-3-git-send-email-shangw@linux.vnet.ibm.com (mailing list archive)
State Superseded
Delegated to: Benjamin Herrenschmidt
Headers show

Commit Message

Gavin Shan Jan. 12, 2014, 6:13 a.m. UTC 
Sometimes, especially in sinario of loading another kernel with kdump,
we got EEH error on non-existing PE. That means the PEEV / PEST in
the corresponding PHB would be messy and we can't handle that case.
The patch escalates the error to fenced PHB so that the PHB could be
rested in order to revoer the errors on non-existing PEs.

Reported-by: Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
Signed-off-by: Gavin Shan <shangw@linux.vnet.ibm.com>
Tested-by: Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
---
 arch/powerpc/platforms/powernv/eeh-ioda.c |   24 +++++++++++++++++++-----
 1 file changed, 19 insertions(+), 5 deletions(-)
diff mbox

Patch

diff --git a/arch/powerpc/platforms/powernv/eeh-ioda.c b/arch/powerpc/platforms/powernv/eeh-ioda.c
index 6971146..0bf6b2b 100644
--- a/arch/powerpc/platforms/powernv/eeh-ioda.c
+++ b/arch/powerpc/platforms/powernv/eeh-ioda.c
@@ -980,12 +980,26 @@  static int ioda_eeh_next_error(struct eeh_pe **pe)
 
 			break;
 		case OPAL_EEH_PE_ERROR:
-			if (ioda_eeh_get_pe(hose, frozen_pe_no, pe))
-				break;
+			/*
+			 * If we can't find the corresponding PE, the
+			 * PEEV / PEST would be messy. So we force an
+			 * fenced PHB so that it can be recovered.
+			 */
+			if (ioda_eeh_get_pe(hose, frozen_pe_no, pe)) {
+				if (!ioda_eeh_get_phb_pe(hose, pe)) {
+					ret = EEH_NEXT_ERR_FENCED_PHB;
+					pr_err("EEH: Escalated fenced PHB#%x "
+					       "detected\n",
+						hose->global_number);
+				} else {
+					ret = EEH_NEXT_ERR_NONE;
+				}
+			} else {
+				pr_err("EEH: Frozen PE#%x on PHB#%x detected\n",
+					(*pe)->addr, (*pe)->phb->global_number);
+				ret = EEH_NEXT_ERR_FROZEN_PE;
+			}
 
-			pr_err("EEH: Frozen PE#%x on PHB#%x detected\n",
-				(*pe)->addr, (*pe)->phb->global_number);
-			ret = EEH_NEXT_ERR_FROZEN_PE;
 			break;
 		default:
 			pr_warn("%s: Unexpected error type %d\n",