From 0c00d359f27c8c88bf7a8af9c5f5fee8c6d9fb15 Mon Sep 17 00:00:00 2001
From: Danny Auble <da@schedmd.com>
Date: Wed, 18 Dec 2013 11:48:04 -0800
Subject: [PATCH] BGQ - make sure if multiple steps fail in a single block at
the same time that spans multiple midplanes the cnodes are correctly
accounted for that are in error.
---
src/plugins/select/bluegene/select_bluegene.c | 22 +++++++++++++++----
1 file changed, 18 insertions(+), 4 deletions(-)
diff --git a/src/plugins/select/bluegene/select_bluegene.c b/src/plugins/select/bluegene/select_bluegene.c
index a4a4b84a470..69bbb0adba6 100644
--- a/src/plugins/select/bluegene/select_bluegene.c
+++ b/src/plugins/select/bluegene/select_bluegene.c
@@ -2899,18 +2899,32 @@ extern int select_p_fail_cnode(struct step_record *step_ptr)
itr2 = list_iterator_create(bg_record->ba_mp_list);
while ((found_ba_mp = (ba_mp_t *)list_next(itr2))) {
- if (!found_ba_mp->used
- || !bit_test(step_ptr->step_node_bitmap,
- found_ba_mp->index))
+ if (!found_ba_mp->used)
continue;
+ if (!bit_test(step_ptr->step_node_bitmap,
+ found_ba_mp->index)) {
+ /* Make sure we get the count of this midplane
+ even if it isn't in this particular step.
+ */
+ bg_record->cnode_err_cnt += bit_set_count(
+ found_ba_mp->cnode_err_bitmap);
+ continue;
+ }
+
/* perhaps this block isn't involved in this
error */
if (jobinfo->units_avail
&& found_ba_mp->cnode_usable_bitmap
&& bit_overlap(found_ba_mp->cnode_usable_bitmap,
- ba_mp->cnode_err_bitmap))
+ ba_mp->cnode_err_bitmap)) {
+ /* Make sure we get the count of this midplane
+ even if it isn't in this particular step.
+ */
+ bg_record->cnode_err_cnt += bit_set_count(
+ found_ba_mp->cnode_err_bitmap);
continue;
+ }
if (!found_ba_mp->cnode_err_bitmap)
found_ba_mp->cnode_err_bitmap =
--
GitLab