From: Russ Anderson <rja@efs.americas.sgi.com>
To: linux-ia64@vger.kernel.org
Subject: [patch] Add mca recovery failure messages
Date: Wed, 26 Apr 2006 20:06:10 +0000 [thread overview]
Message-ID: <200604262006.k3QK6AA5287600@efs.americas.sgi.com> (raw)
[patch] Add mca recovery failure messages
When the mca recovery code encounters a condition that makes
the MCA non-recoverable, print the reason it could not recover.
This will make it easier to identify why the recovery code did
not recover.
Signed-off-by: Russ Anderson (rja@sgi.com)
---
arch/ia64/kernel/mca_drv.c | 54 ++++++++++++++++++++++++++++++---------------
1 file changed, 36 insertions(+), 18 deletions(-)
Index: test/arch/ia64/kernel/mca_drv.c
=================================--- test.orig/arch/ia64/kernel/mca_drv.c 2006-04-20 09:01:10.880579074 -0500
+++ test/arch/ia64/kernel/mca_drv.c 2006-04-26 14:43:52.507645704 -0500
@@ -62,6 +62,11 @@ typedef enum {
ISOLATE_NONE
} isolate_status_t;
+typedef enum {
+ MCA_NOT_RECOVERED = 0,
+ MCA_RECOVERED = 1
+} recovery_status_t;
+
/*
* This pool keeps pointers to the section part of SAL error record
*/
@@ -71,6 +76,18 @@ static struct {
int max_idx; /* Maximum index of section pointer list pool */
} slidx_pool;
+int
+fatal_mca(const char *fmt, ...)
+{
+ va_list args;
+
+ va_start(args, fmt);
+ vprintk(fmt, args);
+ va_end(args);
+
+ return MCA_NOT_RECOVERED;
+}
+
/**
* mca_page_isolate - isolate a poisoned page in order not to use it later
* @paddr: poisoned memory location
@@ -424,7 +441,7 @@ recover_from_read_error(slidx_table_t *s
/* Is target address valid? */
if (!pbci->tv)
- return 0;
+ return fatal_mca(KERN_ALERT "MCA: target address not valid\n");
/*
* cpu read or memory-mapped io read
@@ -442,7 +459,7 @@ recover_from_read_error(slidx_table_t *s
/* Is minstate valid? */
if (!peidx_bottom(peidx) || !(peidx_bottom(peidx)->valid.minstate))
- return 0;
+ return fatal_mca(KERN_ALERT "MCA: minstate not valid\n");
psr1 =(struct ia64_psr *)&(peidx_minstate_area(peidx)->pmsa_ipsr);
psr2 =(struct ia64_psr *)&(peidx_minstate_area(peidx)->pmsa_xpsr);
@@ -476,12 +493,13 @@ recover_from_read_error(slidx_table_t *s
psr2->bn = 1;
psr2->i = 0;
- return 1;
+ return MCA_RECOVERED;
}
}
- return 0;
+ return fatal_mca(KERN_ALERT "MCA: kernel context not recovered,"
+ " iip 0x%lx\n", pmsa->pmsa_iip);
}
/**
@@ -567,13 +585,13 @@ recover_from_processor_error(int platfor
* The machine check is corrected.
*/
if (psp->cm = 1)
- return 1;
+ return MCA_RECOVERED;
/*
* The error was not contained. Software must be reset.
*/
if (psp->us || psp->ci = 0)
- return 0;
+ return fatal_mca(KERN_ALERT "MCA: error not contained\n");
/*
* The cache check and bus check bits have four possible states
@@ -584,20 +602,22 @@ recover_from_processor_error(int platfor
* 1 1 Memory error, attempt recovery
*/
if (psp->bc = 0 || pbci = NULL)
- return 0;
+ return fatal_mca(KERN_ALERT "MCA: No bus check\n");
/*
* Sorry, we cannot handle so many.
*/
if (peidx_bus_check_num(peidx) > 1)
- return 0;
+ return fatal_mca(KERN_ALERT "MCA: Too many bus checks\n");
/*
* Well, here is only one bus error.
*/
- if (pbci->ib || pbci->cc)
- return 0;
+ if (pbci->ib)
+ return fatal_mca(KERN_ALERT "MCA: Internal Bus error\n");
+ if (pbci->cc)
+ return fatal_mca(KERN_ALERT "MCA: Cache-cache error\n");
if (pbci->eb && pbci->bsi > 0)
- return 0;
+ return fatal_mca(KERN_ALERT "MCA: External bus check fatal status\n");
/*
* This is a local MCA and estimated as recoverble external bus error.
@@ -609,7 +629,7 @@ recover_from_processor_error(int platfor
/*
* On account of strange SAL error record, we cannot recover.
*/
- return 0;
+ return fatal_mca(KERN_ALERT "MCA: Strange SAL record\n");
}
/**
@@ -638,12 +658,10 @@ mca_try_to_recover(void *rec, struct ia6
/* Now, OS can recover when there is one processor error section */
if (n_proc_err > 1)
- return 0;
- else if (n_proc_err = 0) {
+ return fatal_mca(KERN_ALERT "MCA: Too Many Errors\n");
+ else if (n_proc_err = 0)
/* Weird SAL record ... We need not to recover */
-
- return 1;
- }
+ return fatal_mca(KERN_ALERT "MCA: Weird SAL record\n");
/* Make index of processor error section */
mca_make_peidx((sal_log_processor_info_t*)
@@ -654,7 +672,7 @@ mca_try_to_recover(void *rec, struct ia6
/* Check whether MCA is global or not */
if (is_mca_global(&peidx, &pbci, sos))
- return 0;
+ return fatal_mca(KERN_ALERT "MCA: global MCA\n");
/* Try to recover a processor error */
return recover_from_processor_error(platform_err, &slidx, &peidx,
--
Russ Anderson, OS RAS/Partitioning Project Lead
SGI - Silicon Graphics Inc rja@sgi.com
next reply other threads:[~2006-04-26 20:06 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2006-04-26 20:06 Russ Anderson [this message]
2006-04-26 23:54 ` [patch] Add mca recovery failure messages Keith Owens
2006-04-27 15:07 ` Russ Anderson
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=200604262006.k3QK6AA5287600@efs.americas.sgi.com \
--to=rja@efs.americas.sgi.com \
--cc=linux-ia64@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox