mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Andi Kleen <andi@firstfloor.org>
To: linux-kernel@vger.kernel.org, hpa@zytor.com, x86@kernel.org
Cc: Andi Kleen <ak@linux.intel.com>
Subject: [PATCH 14/31] x86: MCE: Add table driven machine check grading
Date: Wed, 27 May 2009 01:54:16 +0200	[thread overview]
Message-ID: <10e478c24139e29e7e74529edd694858ec2fb7ea.1243381848.git.ak@linux.intel.com> (raw)
In-Reply-To: <37501061dc5d5581fefcaff92c2606e39cc61913.1243381848.git.ak@linux.intel.com>
In-Reply-To: <c63915cfa51e24394524ddaa66a6f2cb2fcfe66f.1243381848.git.ak@linux.intel.com>

From: Andi Kleen <ak@linux.intel.com>

Impact: Feature

The machine check grading (as in deciding what should be done for a given
register value) has to be done multiple times soon and it's also getting more complicated.
So it makes sense to consolidate it into a single function. To get smaller
and more straight forward and possibly more extensible code I opted towards
a new table driven method. The various rules are put into a table
when is then executed by a very simple interpreter.

The grading engine is in a new file mce-severity.c. I also added a private
include file mce-internal.h, because mce.h is already a bit too cluttered.

This is dead code right now, but will be used in followon patches.

Signed-off-by: Andi Kleen <ak@linux.intel.com>
---
 arch/x86/kernel/cpu/mcheck/Makefile       |    1 +
 arch/x86/kernel/cpu/mcheck/mce-internal.h |   10 +++++
 arch/x86/kernel/cpu/mcheck/mce-severity.c |   61 +++++++++++++++++++++++++++++
 3 files changed, 72 insertions(+), 0 deletions(-)
 create mode 100644 arch/x86/kernel/cpu/mcheck/mce-internal.h
 create mode 100644 arch/x86/kernel/cpu/mcheck/mce-severity.c

diff --git a/arch/x86/kernel/cpu/mcheck/Makefile b/arch/x86/kernel/cpu/mcheck/Makefile
index 60ee182..45004fa 100644
--- a/arch/x86/kernel/cpu/mcheck/Makefile
+++ b/arch/x86/kernel/cpu/mcheck/Makefile
@@ -1,5 +1,6 @@
 obj-y				=  mce.o therm_throt.o
 
+obj-$(CONFIG_X86_NEW_MCE)	+= mce-severity.o
 obj-$(CONFIG_X86_OLD_MCE)	+= k7.o p4.o p6.o
 obj-$(CONFIG_X86_ANCIENT_MCE)	+= winchip.o p5.o
 obj-$(CONFIG_X86_MCE_P4THERMAL)	+= mce_intel.o
diff --git a/arch/x86/kernel/cpu/mcheck/mce-internal.h b/arch/x86/kernel/cpu/mcheck/mce-internal.h
new file mode 100644
index 0000000..f126b4a
--- /dev/null
+++ b/arch/x86/kernel/cpu/mcheck/mce-internal.h
@@ -0,0 +1,10 @@
+#include <asm/mce.h>
+
+enum severity_level {
+	MCE_NO_SEVERITY,
+	MCE_SOME_SEVERITY,
+	MCE_UC_SEVERITY,
+	MCE_PANIC_SEVERITY,
+};
+
+int mce_severity(struct mce *a, int tolerant, char **msg);
diff --git a/arch/x86/kernel/cpu/mcheck/mce-severity.c b/arch/x86/kernel/cpu/mcheck/mce-severity.c
new file mode 100644
index 0000000..c189e89
--- /dev/null
+++ b/arch/x86/kernel/cpu/mcheck/mce-severity.c
@@ -0,0 +1,61 @@
+/*
+ * MCE grading rules.
+ * Copyright 2008, 2009 Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; version 2
+ * of the License.
+ *
+ * Author: Andi Kleen
+ */
+#include <linux/kernel.h>
+#include <asm/mce.h>
+
+#include "mce-internal.h"
+
+/*
+ * Grade an mce by severity. In general the most severe ones are processed
+ * first. Since there are quite a lot of combinations test the bits in a
+ * table-driven way. The rules are simply processed in order, first
+ * match wins.
+ */
+
+static struct severity {
+	u64 mask;
+	u64 result;
+	unsigned char sev;
+	unsigned char mcgmask;
+	unsigned char mcgres;
+	char *msg;
+} severities[] = {
+#define SEV(s) .sev = MCE_ ## s ## _SEVERITY
+#define BITCLR(x, s, m, r...) { .mask = x, .result = 0, SEV(s), .msg = m, ## r }
+#define BITSET(x, s, m, r...) { .mask = x, .result = x, SEV(s), .msg = m, ## r }
+#define MCGMASK(x, res, s, m, r...) \
+	{ .mcgmask = x, .mcgres = res, SEV(s), .msg = m, ## r }
+	BITCLR(MCI_STATUS_VAL, NO, "Invalid"),
+	BITCLR(MCI_STATUS_EN, NO, "Not enabled"),
+	BITSET(MCI_STATUS_PCC, PANIC, "Processor context corrupt"),
+	MCGMASK(MCG_STATUS_RIPV, 0, PANIC, "No restart IP"),
+	BITSET(MCI_STATUS_UC|MCI_STATUS_OVER, PANIC, "Overflowed uncorrected"),
+	BITSET(MCI_STATUS_UC, UC, "Uncorrected"),
+	BITSET(0, SOME, "No match")	/* always matches. keep at end */
+};
+
+int mce_severity(struct mce *a, int tolerant, char **msg)
+{
+	struct severity *s;
+	for (s = severities;; s++) {
+		if ((a->status & s->mask) != s->result)
+			continue;
+		if ((a->mcgstatus & s->mcgmask) != s->mcgres)
+			continue;
+		if (s->sev > MCE_NO_SEVERITY && (a->status & MCI_STATUS_UC) &&
+			tolerant < 1)
+			return MCE_PANIC_SEVERITY;
+		if (msg)
+			*msg = s->msg;
+		return s->sev;
+	}
+}
-- 
1.6.0.2


  reply	other threads:[~2009-05-26 23:58 UTC|newest]

Thread overview: 49+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-05-26 23:54 x86 MCE improvements series for 2.6.31 v2 Andi Kleen
2009-05-26 23:54 ` [PATCH 01/31] x86: MCE: Synchronize core after machine check handling Andi Kleen
2009-05-26 23:54   ` [PATCH 02/31] x86: MCE: Improve mce_get_rip v3 Andi Kleen
2009-05-26 23:54     ` [PATCH 03/31] x86: MCE: Fix EIPV behaviour with !PCC Andi Kleen
2009-05-26 23:54       ` [PATCH 04/31] x86: MCE: Use extended sysattrs for the check_interval attribute Andi Kleen
2009-05-26 23:54         ` [PATCH 05/31] x86: MCE: Add machine check exception count in /proc/interrupts Andi Kleen
2009-05-26 23:54           ` [PATCH 06/31] x86: Fix panic with interrupts off (needed for MCE) Andi Kleen
2009-05-26 23:54             ` [PATCH 07/31] x86: MCE: Log corrected errors when panicing Andi Kleen
2009-05-26 23:54               ` [PATCH 08/31] x86: MCE: Remove unused mce_events variable Andi Kleen
2009-05-26 23:54                 ` [PATCH 09/31] x86: MCE: Remove mce_init unused argument Andi Kleen
2009-05-26 23:54                   ` [PATCH 10/31] x86: MCE: Rename and align out2 label Andi Kleen
2009-05-26 23:54                     ` [PATCH 11/31] x86: MCE: Implement bootstrapping for machine check wakeups Andi Kleen
2009-05-26 23:54                       ` [PATCH 12/31] x86: MCE: Remove TSC print heuristic Andi Kleen
2009-05-26 23:54                         ` [PATCH 13/31] x86: MCE: Drop BKL in mce_open Andi Kleen
2009-05-26 23:54                           ` Andi Kleen [this message]
2009-05-26 23:54                             ` [PATCH 15/31] x86: MCE: Check early in exception handler if panic is needed Andi Kleen
2009-05-26 23:54                               ` [PATCH 16/31] x86: MCE: Implement panic synchronization Andi Kleen
2009-05-26 23:54                                 ` [PATCH 17/31] x86: MCE: Switch x86 machine check handler to Monarch election. v2 Andi Kleen
2009-05-26 23:54                                   ` [PATCH 18/31] x86: MCE: Store record length into memory struct mce anchor Andi Kleen
2009-05-26 23:54                                     ` [PATCH 19/31] x86: MCE: Default to panic timeout for machine checks v2 Andi Kleen
2009-05-26 23:54                                       ` [PATCH 20/31] x86: MCE: Improve documentation Andi Kleen
2009-05-26 23:54                                         ` [PATCH 21/31] x86: MCE: Support more than 256 CPUs in struct mce Andi Kleen
2009-05-26 23:54                                           ` [PATCH 22/31] x86: MCE: Extend struct mce user interface with more information Andi Kleen
2009-05-26 23:54                                             ` [PATCH 23/31] x86: MCE: Add MCE poll count to /proc/interrupts Andi Kleen
2009-05-26 23:54                                               ` [PATCH 24/31] x86: MCE: Don't print backtrace on machine checks with DEBUG_BUGVERBOSE Andi Kleen
2009-05-26 23:54                                                 ` [PATCH 25/31] x86: MCE: Implement new status bits v2 Andi Kleen
2009-05-26 23:54                                                   ` [PATCH 26/31] x86: MCE: Export MCE severities coverage via debugfs Andi Kleen
2009-05-26 23:54                                                     ` [PATCH 27/31] x86: MCE: Print header/footer only once for multiple MCEs Andi Kleen
2009-05-26 23:54                                                       ` [PATCH 28/31] x86: MCE: Make non Monarch panic message "Fatal machine check" too v2 Andi Kleen
2009-05-26 23:54                                                         ` [PATCH 29/31] x86: MCE: Rename mce_notify_user to mce_notify_irq Andi Kleen
2009-05-26 23:54                                                           ` [PATCH 30/31] x86: MCE: Define MCE_VECTOR Andi Kleen
2009-05-26 23:54                                                             ` [PATCH 31/31] x86: MCE: Support action-optional machine checks v2 Andi Kleen
2009-05-27  4:31                                                       ` [PATCH 27/31] x86: MCE: Print header/footer only once for multiple MCEs Hidetoshi Seto
2009-05-27  7:10                                                         ` Andi Kleen
2009-05-27  4:31                                       ` [PATCH 19/31] x86: MCE: Default to panic timeout for machine checks v2 Hidetoshi Seto
2009-05-27  7:24                                         ` Andi Kleen
2009-05-27  4:31                                       ` [PATCH] x86: MCE: Fix for mce_panic_timeout Hidetoshi Seto
2009-05-27 10:07                                         ` Andi Kleen
2009-05-28  0:52                                           ` Hidetoshi Seto
2009-05-28  8:15                                             ` Andi Kleen
2009-05-27  4:30             ` [PATCH 06/31] x86: Fix panic with interrupts off (needed for MCE) Hidetoshi Seto
2009-05-27  7:05               ` Andi Kleen
2009-05-27  4:30       ` [PATCH 03/31] x86: MCE: Fix EIPV behaviour with !PCC Hidetoshi Seto
2009-05-27  7:38         ` Andi Kleen
2009-05-27  7:38           ` Huang Ying
2009-05-27  8:53             ` Andi Kleen
2009-05-27  4:29     ` [PATCH 02/31] x86: MCE: Improve mce_get_rip v3 Hidetoshi Seto
2009-05-27  7:23       ` Andi Kleen
2009-05-27  4:29     ` [PATCH] x86: MCE: Fix for getting IP/CS at MCE Hidetoshi Seto

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=10e478c24139e29e7e74529edd694858ec2fb7ea.1243381848.git.ak@linux.intel.com \
    --to=andi@firstfloor.org \
    --cc=ak@linux.intel.com \
    --cc=hpa@zytor.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=x86@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

all inboxes | Powered by JetHome®