[PATCH 0/6] AMD MCE injection improvs

[PATCH 0/6] AMD MCE injection improvs

[Borislav Petkov]

From: Borislav Petkov <borislav.petkov@amd.com>

So this basically moves the mce-inject module to arch/x86/ras/ and adds
a bunch of features to it.

More importantly, though, it creates all the glue to add the
arch/x86/ras/ directory to Kconfig. The idea is to gradually collect
RAS-related stuff there.

Tony, please take a look and let me know how do you find the general
layout, etc.

Thanks.

Borislav Petkov (6):
  x86, RAS: Add a barebones RAS subtree
  x86, RAS: Move AMD MCE injection code into ras/
  x86, RAS: Convert mce-inject module to debugfs
  x86, RAS: Add function enabling direct writes to MCE MSRs
  x86, RAS: Add attributes needed for HW injection
  x86, RAS: Add an injector function

 arch/x86/Kconfig              |  10 ++
 arch/x86/Makefile             |   3 +
 arch/x86/ras/Kconfig          |  14 +++
 arch/x86/ras/Makefile         |   2 +
 arch/x86/ras/amd/Makefile     |   3 +
 arch/x86/ras/amd/mce-inject.c | 274 ++++++++++++++++++++++++++++++++++++++++++
 drivers/edac/Kconfig          |  10 --
 drivers/edac/mce_amd_inj.c    | 173 --------------------------
 8 files changed, 306 insertions(+), 183 deletions(-)
 create mode 100644 arch/x86/ras/Kconfig
 create mode 100644 arch/x86/ras/Makefile
 create mode 100644 arch/x86/ras/amd/Makefile
 create mode 100644 arch/x86/ras/amd/mce-inject.c
 create mode 100644 arch/x86/ras/ras.c
 delete mode 100644 drivers/edac/mce_amd_inj.c

-- 
1.8.0.rc0.18.gf84667d

[PATCH 1/6] x86, RAS: Add a barebones RAS subtree

[Borislav Petkov]

From: Borislav Petkov <borislav.petkov@amd.com>

This should collect all x86 RAS-related stuff.

Signed-off-by: Borislav Petkov <borislav.petkov@amd.com>
---
 arch/x86/Kconfig      | 8 ++++++++
 arch/x86/Makefile     | 3 +++
 arch/x86/ras/Makefile | 1 +
 3 files changed, 12 insertions(+)
 create mode 100644 arch/x86/ras/Makefile
 create mode 100644 arch/x86/ras/ras.c

diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index e0323d67ee42..e5cae3de52ff 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -687,6 +687,14 @@ config X86_CYCLONE_TIMER
 	def_bool y
 	depends on X86_SUMMIT
 
+config X86_RAS
+	def_bool y
+	prompt "X86 RAS features"
+	---help---
+	A collection of Reliability, Availability and Serviceability software
+	features which enable hardware error logging and reporting. Leave it
+	at 'y' unless you really know what you're doing.
+
 source "arch/x86/Kconfig.cpu"
 
 config HPET_TIMER
diff --git a/arch/x86/Makefile b/arch/x86/Makefile
index 58790bd85c1d..2810e50b670d 100644
--- a/arch/x86/Makefile
+++ b/arch/x86/Makefile
@@ -163,6 +163,9 @@ libs-y  += arch/x86/lib/
 # See arch/x86/Kbuild for content of core part of the kernel
 core-y += arch/x86/
 
+# RAS support
+core-$(CONFIG_X86_RAS)		+= arch/x86/ras/
+
 # drivers-y are linked after core-y
 drivers-$(CONFIG_MATH_EMULATION) += arch/x86/math-emu/
 drivers-$(CONFIG_PCI)            += arch/x86/pci/
diff --git a/arch/x86/ras/Makefile b/arch/x86/ras/Makefile
new file mode 100644
index 000000000000..7a70bb5cd057
--- /dev/null
+++ b/arch/x86/ras/Makefile
@@ -0,0 +1 @@
+obj-y		:= ras.o
diff --git a/arch/x86/ras/ras.c b/arch/x86/ras/ras.c
new file mode 100644
index 000000000000..e69de29bb2d1
-- 
1.8.0.rc0.18.gf84667d

Re: [PATCH 1/6] x86, RAS: Add a barebones RAS subtree

[Tony Luck]

On Mon, Oct 8, 2012 at 10:11 AM, Borislav Petkov <bp@amd64.org> wrote:
> +config X86_RAS
> +       def_bool y
> +       prompt "X86 RAS features"
> +       ---help---
> +       A collection of Reliability, Availability and Serviceability software
> +       features which enable hardware error logging and reporting. Leave it
> +       at 'y' unless you really know what you're doing.
> +

The intent of "X86_RAS" is just to show/hide all the menu
options for the individual features - which will all use
          depends on X86_RAS
right? Having this set to "y" doesn't actually enable any of
the features - they all have their own CONFIG_* variables.

Perhaps we could make that clearer in the help text? And
ditch the "Leave it at 'y' ... ", I don't think it helps anyone.

-Tony

Re: [PATCH 1/6] x86, RAS: Add a barebones RAS subtree

[Borislav Petkov]

On Mon, Oct 08, 2012 at 04:53:46PM -0700, Tony Luck wrote:
> On Mon, Oct 8, 2012 at 10:11 AM, Borislav Petkov <bp@amd64.org> wrote:
> > +config X86_RAS
> > +       def_bool y
> > +       prompt "X86 RAS features"
> > +       ---help---
> > +       A collection of Reliability, Availability and Serviceability software
> > +       features which enable hardware error logging and reporting. Leave it
> > +       at 'y' unless you really know what you're doing.
> > +
> 
> The intent of "X86_RAS" is just to show/hide all the menu
> options for the individual features - which will all use
>           depends on X86_RAS
> right? Having this set to "y" doesn't actually enable any of
> the features - they all have their own CONFIG_* variables.
> 
> Perhaps we could make that clearer in the help text? And
> ditch the "Leave it at 'y' ... ", I don't think it helps anyone.

Good point. I've changed it to a simple bool which opens a submenu when
enabled (diff ontop):

diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 85f666390b5e..d39c2fc20a81 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -688,12 +688,10 @@ config X86_CYCLONE_TIMER
        depends on X86_SUMMIT
 
 config X86_RAS
-       def_bool y
-       prompt "X86 RAS features"
+       bool "X86 RAS features"
        ---help---
        A collection of Reliability, Availability and Serviceability software
-       features which enable hardware error logging and reporting. Leave it
-       at 'y' unless you really know what you're doing.
+       features which enable hardware error logging and reporting.

The def_bool y intent was from earlier but it materialized in a patch
which I sent last week or so which makes CONFIG_X86_MCE enabled by
default so no need to enable those additional RAS stuff by default.

Thanks.

-- 
Regards/Gruss,
Boris.

Advanced Micro Devices GmbH
Einsteinring 24, 85609 Dornach
GM: Alberto Bozzo
Reg: Dornach, Landkreis Muenchen
HRB Nr. 43632 WEEE Registernr: 129 19551

[PATCH 2/6] x86, RAS: Move AMD MCE injection code into ras/

[Borislav Petkov]

From: Borislav Petkov <borislav.petkov@amd.com>

This is the code collecting all AMD MCE injection methods.

No functionality change.

Signed-off-by: Borislav Petkov <borislav.petkov@amd.com>
---
 arch/x86/Kconfig                                            |  2 ++
 arch/x86/ras/Kconfig                                        | 11 +++++++++++
 arch/x86/ras/Makefile                                       |  3 ++-
 arch/x86/ras/amd/Makefile                                   |  3 +++
 drivers/edac/mce_amd_inj.c => arch/x86/ras/amd/mce-inject.c |  2 +-
 drivers/edac/Kconfig                                        | 10 ----------
 6 files changed, 19 insertions(+), 12 deletions(-)
 create mode 100644 arch/x86/ras/Kconfig
 create mode 100644 arch/x86/ras/amd/Makefile
 rename drivers/edac/mce_amd_inj.c => arch/x86/ras/amd/mce-inject.c (99%)

diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index e5cae3de52ff..85f666390b5e 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -695,6 +695,8 @@ config X86_RAS
 	features which enable hardware error logging and reporting. Leave it
 	at 'y' unless you really know what you're doing.
 
+source "arch/x86/ras/Kconfig"
+
 source "arch/x86/Kconfig.cpu"
 
 config HPET_TIMER
diff --git a/arch/x86/ras/Kconfig b/arch/x86/ras/Kconfig
new file mode 100644
index 000000000000..9362f54bfe0f
--- /dev/null
+++ b/arch/x86/ras/Kconfig
@@ -0,0 +1,11 @@
+menu "AMD RAS features"
+	depends on X86_RAS && CPU_SUP_AMD
+
+config X86_AMD_MCE_INJECT
+	tristate "Simple MCE injection interface over /sysfs"
+	default n
+	help
+	  This is a simple interface to inject MCEs over /sysfs and test
+	  the MCE decoding code.
+
+endmenu
diff --git a/arch/x86/ras/Makefile b/arch/x86/ras/Makefile
index 7a70bb5cd057..4d8d37a301c4 100644
--- a/arch/x86/ras/Makefile
+++ b/arch/x86/ras/Makefile
@@ -1 +1,2 @@
-obj-y		:= ras.o
+obj-y				:= ras.o
+obj-$(CONFIG_X86_MCE)		+= amd/
diff --git a/arch/x86/ras/amd/Makefile b/arch/x86/ras/amd/Makefile
new file mode 100644
index 000000000000..74f2a1e676b4
--- /dev/null
+++ b/arch/x86/ras/amd/Makefile
@@ -0,0 +1,3 @@
+amd_mce_inject-y				:= mce-inject.o
+obj-$(CONFIG_X86_AMD_MCE_INJECT)		+= amd_mce_inject.o
+
diff --git a/drivers/edac/mce_amd_inj.c b/arch/x86/ras/amd/mce-inject.c
similarity index 99%
rename from drivers/edac/mce_amd_inj.c
rename to arch/x86/ras/amd/mce-inject.c
index 66b5151c1080..90aa317b09a6 100644
--- a/drivers/edac/mce_amd_inj.c
+++ b/arch/x86/ras/amd/mce-inject.c
@@ -16,7 +16,7 @@
 #include <linux/module.h>
 #include <asm/mce.h>
 
-#include "mce_amd.h"
+#include "../../../drivers/edac/mce_amd.h"
 
 struct edac_mce_attr {
 	struct attribute attr;
diff --git a/drivers/edac/Kconfig b/drivers/edac/Kconfig
index bb82d6be793c..0e6afbd7749c 100644
--- a/drivers/edac/Kconfig
+++ b/drivers/edac/Kconfig
@@ -59,16 +59,6 @@ config EDAC_DECODE_MCE
 	  which occur really early upon boot, before the module infrastructure
 	  has been initialized.
 
-config EDAC_MCE_INJ
-	tristate "Simple MCE injection interface over /sysfs"
-	depends on EDAC_DECODE_MCE
-	default n
-	help
-	  This is a simple interface to inject MCEs over /sysfs and test
-	  the MCE decoding code in EDAC.
-
-	  This is currently AMD-only.
-
 config EDAC_MM_EDAC
 	tristate "Main Memory EDAC (Error Detection And Correction) reporting"
 	help
-- 
1.8.0.rc0.18.gf84667d

[PATCH 3/6] x86, RAS: Convert mce-inject module to debugfs

[Borislav Petkov]

From: Borislav Petkov <borislav.petkov@amd.com>

This is a module which is used for debugging MCE decoding paths so its
userspace interface should go to debugfs, where it belongs conceptually.

While at it, add a warning to the Kconfig text that this interface is
unstable and no userspace scripts should rely all too much on it.

Signed-off-by: Borislav Petkov <borislav.petkov@amd.com>
---
 arch/x86/ras/Kconfig          |  11 ++-
 arch/x86/ras/amd/mce-inject.c | 182 ++++++++++++++++++------------------------
 2 files changed, 86 insertions(+), 107 deletions(-)

diff --git a/arch/x86/ras/Kconfig b/arch/x86/ras/Kconfig
index 9362f54bfe0f..4334ab92b3b0 100644
--- a/arch/x86/ras/Kconfig
+++ b/arch/x86/ras/Kconfig
@@ -2,10 +2,13 @@ menu "AMD RAS features"
 	depends on X86_RAS && CPU_SUP_AMD
 
 config X86_AMD_MCE_INJECT
-	tristate "Simple MCE injection interface over /sysfs"
+	tristate "Inject MCEs"
+	depends on DEBUG_FS
 	default n
-	help
-	  This is a simple interface to inject MCEs over /sysfs and test
-	  the MCE decoding code.
+	---help---
+	  This is a simple debugfs interface to inject MCEs and test different
+	  aspects of the MCE handling code.
+
+	  WARNING: Do not even assume that this interface is staying stable!
 
 endmenu
diff --git a/arch/x86/ras/amd/mce-inject.c b/arch/x86/ras/amd/mce-inject.c
index 90aa317b09a6..66969eab5591 100644
--- a/arch/x86/ras/amd/mce-inject.c
+++ b/arch/x86/ras/amd/mce-inject.c
@@ -1,173 +1,149 @@
 /*
- * A simple MCE injection facility for testing the MCE decoding code. This
- * driver should be built as module so that it can be loaded on production
- * kernels for testing purposes.
+ * A simple MCE injection facility for testing different aspects of the RAS
+ * code. This driver should be built as module so that it can be loaded
+ * on production kernels for testing purposes.
  *
  * This file may be distributed under the terms of the GNU General Public
  * License version 2.
  *
- * Copyright (c) 2010:  Borislav Petkov <borislav.petkov@amd.com>
- *			Advanced Micro Devices Inc.
+ * Copyright (c) 2010-12:  Borislav Petkov <borislav.petkov@amd.com>
+ *			   Advanced Micro Devices Inc.
  */
 
 #include <linux/kobject.h>
+#include <linux/debugfs.h>
 #include <linux/device.h>
-#include <linux/edac.h>
 #include <linux/module.h>
 #include <asm/mce.h>
 
 #include "../../../drivers/edac/mce_amd.h"
 
-struct edac_mce_attr {
-	struct attribute attr;
-	ssize_t (*show) (struct kobject *kobj, struct edac_mce_attr *attr, char *buf);
-	ssize_t (*store)(struct kobject *kobj, struct edac_mce_attr *attr,
-			 const char *buf, size_t count);
-};
-
-#define EDAC_MCE_ATTR(_name, _mode, _show, _store)			\
-static struct edac_mce_attr mce_attr_##_name = __ATTR(_name, _mode, _show, _store)
-
-static struct kobject *mce_kobj;
-
 /*
  * Collect all the MCi_XXX settings
  */
 static struct mce i_mce;
+static struct dentry *dfs_inj;
 
-#define MCE_INJECT_STORE(reg)						\
-static ssize_t edac_inject_##reg##_store(struct kobject *kobj,		\
-					 struct edac_mce_attr *attr,	\
-					 const char *data, size_t count)\
+#define MCE_INJECT_SET(reg)						\
+static int inj_##reg##_set(void *data, u64 val)				\
 {									\
-	int ret = 0;							\
-	unsigned long value;						\
-									\
-	ret = strict_strtoul(data, 16, &value);				\
-	if (ret < 0)							\
-		printk(KERN_ERR "Error writing MCE " #reg " field.\n");	\
-									\
-	i_mce.reg = value;						\
+	struct mce *m = (struct mce *)data;				\
 									\
-	return count;							\
+	m->reg = val;							\
+	return 0;							\
 }
 
-MCE_INJECT_STORE(status);
-MCE_INJECT_STORE(misc);
-MCE_INJECT_STORE(addr);
+MCE_INJECT_SET(status);
+MCE_INJECT_SET(misc);
+MCE_INJECT_SET(addr);
 
-#define MCE_INJECT_SHOW(reg)						\
-static ssize_t edac_inject_##reg##_show(struct kobject *kobj,		\
-					struct edac_mce_attr *attr,	\
-					char *buf)			\
+#define MCE_INJECT_GET(reg)						\
+static int inj_##reg##_get(void *data, u64 *val)			\
 {									\
-	return sprintf(buf, "0x%016llx\n", i_mce.reg);			\
+	struct mce *m = (struct mce *)data;				\
+									\
+	*val = m->reg;							\
+	return 0;							\
 }
 
-MCE_INJECT_SHOW(status);
-MCE_INJECT_SHOW(misc);
-MCE_INJECT_SHOW(addr);
+MCE_INJECT_GET(status);
+MCE_INJECT_GET(misc);
+MCE_INJECT_GET(addr);
 
-EDAC_MCE_ATTR(status, 0644, edac_inject_status_show, edac_inject_status_store);
-EDAC_MCE_ATTR(misc, 0644, edac_inject_misc_show, edac_inject_misc_store);
-EDAC_MCE_ATTR(addr, 0644, edac_inject_addr_show, edac_inject_addr_store);
+DEFINE_SIMPLE_ATTRIBUTE(status_fops, inj_status_get, inj_status_set, "%llx\n");
+DEFINE_SIMPLE_ATTRIBUTE(misc_fops, inj_misc_get, inj_misc_set, "%llx\n");
+DEFINE_SIMPLE_ATTRIBUTE(addr_fops, inj_addr_get, inj_addr_set, "%llx\n");
 
 /*
  * This denotes into which bank we're injecting and triggers
  * the injection, at the same time.
  */
-static ssize_t edac_inject_bank_store(struct kobject *kobj,
-				      struct edac_mce_attr *attr,
-				      const char *data, size_t count)
+static int inj_bank_set(void *data, u64 val)
 {
-	int ret = 0;
-	unsigned long value;
+	struct mce *m = (struct mce *)data;
 
-	ret = strict_strtoul(data, 10, &value);
-	if (ret < 0) {
-		printk(KERN_ERR "Invalid bank value!\n");
-		return -EINVAL;
-	}
-
-	if (value > 5)
-		if (boot_cpu_data.x86 != 0x15 || value > 6) {
-			printk(KERN_ERR "Non-existent MCE bank: %lu\n", value);
+	if (val > 5) {
+		if (boot_cpu_data.x86 != 0x15 || val > 6) {
+			pr_err("Non-existent MCE bank: %llu\n", val);
 			return -EINVAL;
 		}
+	}
 
-	i_mce.bank = value;
+	m->bank = val;
 
-	amd_decode_mce(NULL, 0, &i_mce);
+	amd_decode_mce(NULL, 0, m);
 
-	return count;
+	return 0;
 }
 
-static ssize_t edac_inject_bank_show(struct kobject *kobj,
-				     struct edac_mce_attr *attr, char *buf)
+static int inj_bank_get(void *data, u64 *val)
 {
-	return sprintf(buf, "%d\n", i_mce.bank);
-}
+	struct mce *m = (struct mce *)data;
 
-EDAC_MCE_ATTR(bank, 0644, edac_inject_bank_show, edac_inject_bank_store);
+	*val = m->bank;
+	return 0;
+}
 
-static struct edac_mce_attr *sysfs_attrs[] = { &mce_attr_status, &mce_attr_misc,
-					       &mce_attr_addr, &mce_attr_bank
+DEFINE_SIMPLE_ATTRIBUTE(bank_fops, inj_bank_get, inj_bank_set, "%llu\n");
+
+struct dfs_node {
+	char *name;
+	struct dentry *d;
+	const struct file_operations *fops;
+} dfs_fls[] = {
+	{ .name = "status",	.fops = &status_fops },
+	{ .name = "misc",	.fops = &misc_fops },
+	{ .name = "addr",	.fops = &addr_fops },
+	{ .name = "bank",	.fops = &bank_fops },
 };
 
-static int __init edac_init_mce_inject(void)
+static int __init init_mce_inject(void)
 {
-	struct bus_type *edac_subsys = NULL;
-	int i, err = 0;
+	int i;
 
-	edac_subsys = edac_get_sysfs_subsys();
-	if (!edac_subsys)
+	dfs_inj = debugfs_create_dir("mce-inject", NULL);
+	if (!dfs_inj)
 		return -EINVAL;
 
-	mce_kobj = kobject_create_and_add("mce", &edac_subsys->dev_root->kobj);
-	if (!mce_kobj) {
-		printk(KERN_ERR "Error creating a mce kset.\n");
-		err = -ENOMEM;
-		goto err_mce_kobj;
-	}
+	for (i = 0; i < ARRAY_SIZE(dfs_fls); i++) {
+		dfs_fls[i].d = debugfs_create_file(dfs_fls[i].name,
+						    S_IRUSR | S_IWUSR,
+						    dfs_inj,
+						    &i_mce,
+						    dfs_fls[i].fops);
 
-	for (i = 0; i < ARRAY_SIZE(sysfs_attrs); i++) {
-		err = sysfs_create_file(mce_kobj, &sysfs_attrs[i]->attr);
-		if (err) {
-			printk(KERN_ERR "Error creating %s in sysfs.\n",
-					sysfs_attrs[i]->attr.name);
-			goto err_sysfs_create;
-		}
+		if (!dfs_fls[i].d)
+			goto err_dfs_add;
 	}
+
 	return 0;
 
-err_sysfs_create:
+err_dfs_add:
 	while (--i >= 0)
-		sysfs_remove_file(mce_kobj, &sysfs_attrs[i]->attr);
+		debugfs_remove(dfs_fls[i].d);
 
-	kobject_del(mce_kobj);
+	debugfs_remove(dfs_inj);
+	dfs_inj = NULL;
 
-err_mce_kobj:
-	edac_put_sysfs_subsys();
-
-	return err;
+	return -ENOMEM;
 }
 
-static void __exit edac_exit_mce_inject(void)
+static void __exit exit_mce_inject(void)
 {
 	int i;
 
-	for (i = 0; i < ARRAY_SIZE(sysfs_attrs); i++)
-		sysfs_remove_file(mce_kobj, &sysfs_attrs[i]->attr);
+	for (i = 0; i < ARRAY_SIZE(dfs_fls); i++)
+		debugfs_remove(dfs_fls[i].d);
 
-	kobject_del(mce_kobj);
+	memset(&dfs_fls, 0, sizeof(dfs_fls));
 
-	edac_put_sysfs_subsys();
+	debugfs_remove(dfs_inj);
+	dfs_inj = NULL;
 }
-
-module_init(edac_init_mce_inject);
-module_exit(edac_exit_mce_inject);
+module_init(init_mce_inject);
+module_exit(exit_mce_inject);
 
 MODULE_LICENSE("GPL");
 MODULE_AUTHOR("Borislav Petkov <borislav.petkov@amd.com>");
 MODULE_AUTHOR("AMD Inc.");
-MODULE_DESCRIPTION("MCE injection facility for testing MCE decoding");
+MODULE_DESCRIPTION("MCE injection facility for RAS testing");
-- 
1.8.0.rc0.18.gf84667d

[PATCH 4/6] x86, RAS: Add function enabling direct writes to MCE MSRs

[Borislav Petkov]

From: Borislav Petkov <borislav.petkov@amd.com>

Normally, writing to MCE MSRs causes a #GP. Add a function to enable
direct accesses to those MSRs.

Signed-off-by: Borislav Petkov <borislav.petkov@amd.com>
---
 arch/x86/ras/amd/mce-inject.c | 24 ++++++++++++++++++++++++
 1 file changed, 24 insertions(+)

diff --git a/arch/x86/ras/amd/mce-inject.c b/arch/x86/ras/amd/mce-inject.c
index 66969eab5591..c0c811579ce9 100644
--- a/arch/x86/ras/amd/mce-inject.c
+++ b/arch/x86/ras/amd/mce-inject.c
@@ -55,6 +55,30 @@ DEFINE_SIMPLE_ATTRIBUTE(misc_fops, inj_misc_get, inj_misc_set, "%llx\n");
 DEFINE_SIMPLE_ATTRIBUTE(addr_fops, inj_addr_get, inj_addr_set, "%llx\n");
 
 /*
+ * Caller needs to be make sure this cpu doesn't disappear
+ * from under us, i.e.: get_cpu/put_cpu.
+ */
+static int toggle_hw_mce_inject(unsigned int cpu, bool enable)
+{
+	u32 l, h;
+	int err;
+
+	err = rdmsr_on_cpu(cpu, MSR_K7_HWCR, &l, &h);
+	if (err) {
+		pr_err("%s: error reading HWCR\n", __func__);
+		return err;
+	}
+
+	enable ? (l |= BIT(18)) : (l &= ~BIT(18));
+
+	err = wrmsr_on_cpu(cpu, MSR_K7_HWCR, l, h);
+	if (err)
+		pr_err("%s: error writing HWCR\n", __func__);
+
+	return err;
+}
+
+/*
  * This denotes into which bank we're injecting and triggers
  * the injection, at the same time.
  */
-- 
1.8.0.rc0.18.gf84667d

[PATCH 5/6] x86, RAS: Add attributes needed for HW injection

[Borislav Petkov]

From: Borislav Petkov <borislav.petkov@amd.com>

hw_inject denotes whether we want to do a hardware or a software
injection and, in the case of hardware injection, we want to do that on
a particular cpu, thus the 'cpu' attribute.

Signed-off-by: Borislav Petkov <borislav.petkov@amd.com>
---
 arch/x86/ras/amd/mce-inject.c | 55 +++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 55 insertions(+)

diff --git a/arch/x86/ras/amd/mce-inject.c b/arch/x86/ras/amd/mce-inject.c
index c0c811579ce9..b55d69fbd6e3 100644
--- a/arch/x86/ras/amd/mce-inject.c
+++ b/arch/x86/ras/amd/mce-inject.c
@@ -79,6 +79,59 @@ static int toggle_hw_mce_inject(unsigned int cpu, bool enable)
 }
 
 /*
+ * HW or SW injection
+ */
+static int hw_inj_get(void *data, u64 *val)
+{
+	struct mce *m = (struct mce *)data;
+
+	*val = !!(m->inject_flags & MCJ_EXCEPTION);
+
+	return 0;
+}
+
+static int hw_inj_set(void *data, u64 val)
+{
+	struct mce *m = (struct mce *)data;
+
+	switch (val) {
+	case 0:
+		m->inject_flags &= (u8)~MCJ_EXCEPTION;
+		break;
+
+	case 1:
+		m->inject_flags |= MCJ_EXCEPTION;
+		break;
+
+	default:
+		pr_err("%s: Only 0 or 1 allowed!\n", __func__);
+		return -EINVAL;
+	}
+	return 0;
+}
+
+DEFINE_SIMPLE_ATTRIBUTE(hw_inj_fops, hw_inj_get, hw_inj_set, "%llu\n");
+
+/*
+ * On which CPU to inject?
+ */
+MCE_INJECT_GET(extcpu);
+
+static int inj_extcpu_set(void *data, u64 val)
+{
+	struct mce *m = (struct mce *)data;
+
+	if (val >= nr_cpu_ids || !cpu_online(val)) {
+		pr_err("%s: Invalid CPU: %llu\n", __func__, val);
+		return -EINVAL;
+	}
+	m->extcpu = val;
+	return 0;
+}
+
+DEFINE_SIMPLE_ATTRIBUTE(extcpu_fops, inj_extcpu_get, inj_extcpu_set, "%llu\n");
+
+/*
  * This denotes into which bank we're injecting and triggers
  * the injection, at the same time.
  */
@@ -119,6 +172,8 @@ struct dfs_node {
 	{ .name = "misc",	.fops = &misc_fops },
 	{ .name = "addr",	.fops = &addr_fops },
 	{ .name = "bank",	.fops = &bank_fops },
+	{ .name = "hw_inject",	.fops = &hw_inj_fops },
+	{ .name = "cpu",	.fops = &extcpu_fops },
 };
 
 static int __init init_mce_inject(void)
-- 
1.8.0.rc0.18.gf84667d

[PATCH 6/6] x86, RAS: Add an injector function

[Borislav Petkov]

From: Borislav Petkov <borislav.petkov@amd.com>

Selectively inject either a real MCE or a sw-only version which
exercises the decoding code only. The hardware-injected MCE triggers a
machine check exception (#MC) so that the MCE handler can be bothered to
do something too.

Signed-off-by: Borislav Petkov <borislav.petkov@amd.com>
---
 arch/x86/ras/amd/mce-inject.c | 50 +++++++++++++++++++++++++++++++++++++++++--
 1 file changed, 48 insertions(+), 2 deletions(-)

diff --git a/arch/x86/ras/amd/mce-inject.c b/arch/x86/ras/amd/mce-inject.c
index b55d69fbd6e3..24a1c870cd18 100644
--- a/arch/x86/ras/amd/mce-inject.c
+++ b/arch/x86/ras/amd/mce-inject.c
@@ -131,6 +131,53 @@ static int inj_extcpu_set(void *data, u64 val)
 
 DEFINE_SIMPLE_ATTRIBUTE(extcpu_fops, inj_extcpu_get, inj_extcpu_set, "%llu\n");
 
+static void trigger_mce(void *info)
+{
+	asm volatile("int $18");
+}
+
+static void do_inject(void)
+{
+	u64 mcg_status = 0;
+	unsigned int cpu = i_mce.extcpu;
+	int this_cpu;
+	u8 b = i_mce.bank;
+
+	if (!(i_mce.inject_flags & MCJ_EXCEPTION)) {
+		amd_decode_mce(NULL, 0, &i_mce);
+		return;
+	}
+
+	/* prep MCE global settings for the injection */
+	mcg_status = MCG_STATUS_MCIP | MCG_STATUS_EIPV;
+
+	if (!(i_mce.status & MCI_STATUS_PCC))
+		mcg_status |= MCG_STATUS_RIPV;
+
+	this_cpu = get_cpu();
+
+	toggle_hw_mce_inject(cpu, true);
+
+	wrmsr_on_cpu(cpu, MSR_IA32_MCG_STATUS,
+		     (u32)mcg_status, (u32)(mcg_status >> 32));
+
+	wrmsr_on_cpu(cpu, MSR_IA32_MCx_STATUS(b),
+		     (u32)i_mce.status, (u32)(i_mce.status >> 32));
+
+	wrmsr_on_cpu(cpu, MSR_IA32_MCx_ADDR(b),
+		     (u32)i_mce.addr, (u32)(i_mce.addr >> 32));
+
+	wrmsr_on_cpu(cpu, MSR_IA32_MCx_MISC(b),
+		     (u32)i_mce.misc, (u32)(i_mce.misc >> 32));
+
+	toggle_hw_mce_inject(cpu, false);
+
+	smp_call_function_single(cpu, trigger_mce, NULL, 0);
+
+	put_cpu();
+
+}
+
 /*
  * This denotes into which bank we're injecting and triggers
  * the injection, at the same time.
@@ -147,8 +194,7 @@ static int inj_bank_set(void *data, u64 val)
 	}
 
 	m->bank = val;
-
-	amd_decode_mce(NULL, 0, m);
+	do_inject();
 
 	return 0;
 }
-- 
1.8.0.rc0.18.gf84667d

Re: [PATCH 0/6] AMD MCE injection improvs

[Tony Luck]

On Mon, Oct 8, 2012 at 10:11 AM, Borislav Petkov <bp@amd64.org> wrote:
>  create mode 100644 arch/x86/ras/ras.c

Overall it looks good - but I'm a bit puzzled by this ras.c file that gets
created as an empty file in part1, and is still empty at the end of the
series.

What is going to go into it?

-Tony

Re: [PATCH 0/6] AMD MCE injection improvs

[Borislav Petkov]

On Mon, Oct 08, 2012 at 04:58:34PM -0700, Tony Luck wrote:
> On Mon, Oct 8, 2012 at 10:11 AM, Borislav Petkov <bp@amd64.org> wrote:
> >  create mode 100644 arch/x86/ras/ras.c
> 
> Overall it looks good - but I'm a bit puzzled by this ras.c file that gets
> created as an empty file in part1, and is still empty at the end of the
> series.
> 
> What is going to go into it?

Yeah, maybe it should be called core.c and all the shared functionality
would go there. I put it there initially for the first patch but I
realize now, we won't need it just yet - I'll drop it.

Thanks.

-- 
Regards/Gruss,
Boris.

Advanced Micro Devices GmbH
Einsteinring 24, 85609 Dornach
GM: Alberto Bozzo
Reg: Dornach, Landkreis Muenchen
HRB Nr. 43632 WEEE Registernr: 129 19551