[PATCH 00/17] ACPI, APEI, Patches for 3.1

[PATCH 00/17] ACPI, APEI, Patches for 3.1

[Huang Ying]

Including some bug fixes,  APEI _OSC support, recoverable error
printk support and memory error recovery support.

[PATCH 01/17] ACPI, APEI, HEST, Detect duplicated hardware error
[PATCH 02/17] ACPI, APEI, ERST, Prevent erst_dbg from loading if
[PATCH 03/17] ACPI, APEI, ERST, Fix erst-dbg long record reading
[PATCH 04/17] ACPI, APEI, GHES, Do not ratelimit fatal error printk
[PATCH 05/17] ACPI, APEI, Add apei_exec_run_optional
[PATCH 06/17] ACPI, APEI, Use apei_exec_run_optional in APEI EINJ
[PATCH 07/17] ACPI, APEI, GHES, Prevent GHES to be built as module
[PATCH 08/17] ACPI, APEI, GHES, Support disable GHES at boot time
[PATCH 09/17] ACPI, APEI, Add APEI bit support in generic _OSC call
[PATCH 10/17] ACPI, APEI, Add WHEA _OSC support
[PATCH 11/17] Add Kconfig option ARCH_HAVE_NMI_SAFE_CMPXCHG
[PATCH 12/17] lib, Add lock-less NULL terminated single list
[PATCH 13/17] lib, Make gen_pool memory allocator lockless
[PATCH 14/17] ACPI, APEI, GHES, printk support for recoverable error
[PATCH 15/17] ACPI, APEI, GHES, Error records content based throttle
[PATCH 16/17] HWPoison: add memory_failure_queue()
[PATCH 17/17] ACPI, APEI, GHES: Add hardware memory error recovery

[PATCH 01/17] ACPI, APEI, HEST, Detect duplicated hardware error source ID

[Huang Ying]

The firmware on some machine will report duplicated hardware error
source ID in HEST.  This is considered a firmware bug.  To provide
better warning message, this patch adds duplicated hardware error
source ID detecting and corresponding printk.

This patch fixes #37412 on kernel bugzilla:
https://bugzilla.kernel.org/show_bug.cgi?id=37412

Reported-by: marconifabio@ubuntu-it.org
Signed-off-by: Huang Ying <ying.huang@intel.com>
Tested-by: Mathias <janedo.spam@gmail.com>
---
 drivers/acpi/apei/hest.c |   12 +++++++++++-
 1 file changed, 11 insertions(+), 1 deletion(-)

--- a/drivers/acpi/apei/hest.c
+++ b/drivers/acpi/apei/hest.c
@@ -139,13 +139,23 @@ static int __init hest_parse_ghes(struct
 {
 	struct platform_device *ghes_dev;
 	struct ghes_arr *ghes_arr = data;
-	int rc;
+	int rc, i;
 
 	if (hest_hdr->type != ACPI_HEST_TYPE_GENERIC_ERROR)
 		return 0;
 
 	if (!((struct acpi_hest_generic *)hest_hdr)->enabled)
 		return 0;
+	for (i = 0; i < ghes_arr->count; i++) {
+		struct acpi_hest_header *hdr;
+		ghes_dev = ghes_arr->ghes_devs[i];
+		hdr = *(struct acpi_hest_header **)ghes_dev->dev.platform_data;
+		if (hdr->source_id == hest_hdr->source_id) {
+			pr_warning(FW_WARN HEST_PFX "Duplicated hardware error source ID: %d.\n",
+				   hdr->source_id);
+			return -EIO;
+		}
+	}
 	ghes_dev = platform_device_alloc("GHES", hest_hdr->source_id);
 	if (!ghes_dev)
 		return -ENOMEM;

Re: [PATCH 01/17] ACPI, APEI, HEST, Detect duplicated hardware error source ID

[Len Brown]

applied

thanks,
Len Brown, Intel Open Source Technology Center

[PATCH 02/17] ACPI, APEI, ERST, Prevent erst_dbg from loading if ERST is disabled

[Huang Ying]

erst_dbg module can not work when ERST is disabled.  So disable module
loading to provide clearer information to user.

Signed-off-by: Huang Ying <ying.huang@intel.com>
---
 drivers/acpi/apei/erst-dbg.c |    4 ++++
 1 file changed, 4 insertions(+)

--- a/drivers/acpi/apei/erst-dbg.c
+++ b/drivers/acpi/apei/erst-dbg.c
@@ -213,6 +213,10 @@ static struct miscdevice erst_dbg_dev =
 
 static __init int erst_dbg_init(void)
 {
+	if (erst_disable) {
+		pr_info(ERST_DBG_PFX "ERST support is disabled.\n");
+		return -ENODEV;
+	}
 	return misc_register(&erst_dbg_dev);
 }
 

Re: [PATCH 02/17] ACPI, APEI, ERST, Prevent erst_dbg from loading if ERST is disabled

[Len Brown]

applied

thanks,
Len Brown, Intel Open Source Technology Center

[PATCH 03/17] ACPI, APEI, ERST, Fix erst-dbg long record reading issue

[Huang Ying]

From: Chen Gong <gong.chen@linux.intel.com>

When we debug ERST table with erst-dbg, if the error record in ERST
table is too long(>4K), it can't be read out.  So this patch increases
the buffer size to 16K to ensure such error records can be read from
ERST table.

Signed-off-by: Chen Gong <gong.chen@linux.intel.com>
Signed-off-by: Huang Ying <ying.huang@intel.com>
---
 drivers/acpi/apei/erst-dbg.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

--- a/drivers/acpi/apei/erst-dbg.c
+++ b/drivers/acpi/apei/erst-dbg.c
@@ -33,7 +33,7 @@
 
 #define ERST_DBG_PFX			"ERST DBG: "
 
-#define ERST_DBG_RECORD_LEN_MAX		4096
+#define ERST_DBG_RECORD_LEN_MAX		0x4000
 
 static void *erst_dbg_buf;
 static unsigned int erst_dbg_buf_len;

Re: [PATCH 03/17] ACPI, APEI, ERST, Fix erst-dbg long record reading issue

[Len Brown]

applied

thanks,
Len Brown, Intel Open Source Technology Center

[PATCH 04/17] ACPI, APEI, GHES, Do not ratelimit fatal error printk before panic

[Huang Ying]

printk is used by GHES to report hardware errors.  Normally, the
printk will be ratelimited to avoid too many hardware error reports in
kernel log.  Because there may be thousands or even millions of
corrected hardware errors during system running.

That is different for fatal hardware error, because system will go
panic as soon as possible, there will be no more than several error
records.  And these error records are valuable for system fault
diagnosis, so they should not be ratelimited.

Signed-off-by: Huang Ying <ying.huang@intel.com>
---
 drivers/acpi/apei/ghes.c |   25 ++++++++++++++-----------
 1 file changed, 14 insertions(+), 11 deletions(-)

--- a/drivers/acpi/apei/ghes.c
+++ b/drivers/acpi/apei/ghes.c
@@ -360,11 +360,8 @@ static void ghes_do_proc(struct ghes *gh
 	}
 }
 
-static void ghes_print_estatus(const char *pfx, struct ghes *ghes)
+static void __ghes_print_estatus(const char *pfx, struct ghes *ghes)
 {
-	/* Not more than 2 messages every 5 seconds */
-	static DEFINE_RATELIMIT_STATE(ratelimit, 5*HZ, 2);
-
 	if (pfx == NULL) {
 		if (ghes_severity(ghes->estatus->error_severity) <=
 		    GHES_SEV_CORRECTED)
@@ -372,12 +369,18 @@ static void ghes_print_estatus(const cha
 		else
 			pfx = KERN_ERR HW_ERR;
 	}
-	if (__ratelimit(&ratelimit)) {
-		printk(
-	"%s""Hardware error from APEI Generic Hardware Error Source: %d\n",
-	pfx, ghes->generic->header.source_id);
-		apei_estatus_print(pfx, ghes->estatus);
-	}
+	printk("%s""Hardware error from APEI Generic Hardware Error Source: %d\n",
+	       pfx, ghes->generic->header.source_id);
+	apei_estatus_print(pfx, ghes->estatus);
+}
+
+static void ghes_print_estatus(const char *pfx, struct ghes *ghes)
+{
+	/* Not more than 2 messages every 5 seconds */
+	static DEFINE_RATELIMIT_STATE(ratelimit, 5*HZ, 2);
+
+	if (__ratelimit(&ratelimit))
+		__ghes_print_estatus(pfx, ghes);
 }
 
 static int ghes_proc(struct ghes *ghes)
@@ -476,7 +479,7 @@ static int ghes_notify_nmi(struct notifi
 
 	if (sev_global >= GHES_SEV_PANIC) {
 		oops_begin();
-		ghes_print_estatus(KERN_EMERG HW_ERR, ghes_global);
+		__ghes_print_estatus(KERN_EMERG HW_ERR, ghes_global);
 		/* reboot to log the error! */
 		if (panic_timeout == 0)
 			panic_timeout = ghes_panic_timeout;

Re: [PATCH 04/17] ACPI, APEI, GHES, Do not ratelimit fatal error printk before panic

[Len Brown]

applied

thanks,
Len Brown, Intel Open Source Technology Center

[PATCH 05/17] ACPI, APEI, Add apei_exec_run_optional

[Huang Ying]

Some actions in APEI ERST and EINJ tables are optional, for example,
ACPI_EINJ_BEGIN_OPERATION action is used to do some preparation for
error injection, and firmware may choose to do nothing here.  While
some other actions are mandatory, for example, firmware must provide
ACPI_EINJ_GET_ERROR_TYPE implementation.

Original implementation treats all actions as optional (that is, can
have no instructions), that may cause issue if firmware does not
provide some mandatory actions.  To fix this, this patch adds
apei_exec_run_optional, which should be used for optional actions.
The original apei_exec_run should be used for mandatory actions.

Cc: Thomas Renninger <trenn@novell.com>
Signed-off-by: Huang Ying <ying.huang@intel.com>
---
 drivers/acpi/apei/apei-base.c     |    9 +++++----
 drivers/acpi/apei/apei-internal.h |   13 ++++++++++++-
 2 files changed, 17 insertions(+), 5 deletions(-)

--- a/drivers/acpi/apei/apei-base.c
+++ b/drivers/acpi/apei/apei-base.c
@@ -157,9 +157,10 @@ EXPORT_SYMBOL_GPL(apei_exec_noop);
  * Interpret the specified action. Go through whole action table,
  * execute all instructions belong to the action.
  */
-int apei_exec_run(struct apei_exec_context *ctx, u8 action)
+int __apei_exec_run(struct apei_exec_context *ctx, u8 action,
+		    bool optional)
 {
-	int rc;
+	int rc = -ENOENT;
 	u32 i, ip;
 	struct acpi_whea_header *entry;
 	apei_exec_ins_func_t run;
@@ -198,9 +199,9 @@ rewind:
 			goto rewind;
 	}
 
-	return 0;
+	return !optional && rc < 0 ? rc : 0;
 }
-EXPORT_SYMBOL_GPL(apei_exec_run);
+EXPORT_SYMBOL_GPL(__apei_exec_run);
 
 typedef int (*apei_exec_entry_func_t)(struct apei_exec_context *ctx,
 				      struct acpi_whea_header *entry,
--- a/drivers/acpi/apei/apei-internal.h
+++ b/drivers/acpi/apei/apei-internal.h
@@ -50,7 +50,18 @@ static inline u64 apei_exec_ctx_get_outp
 	return ctx->value;
 }
 
-int apei_exec_run(struct apei_exec_context *ctx, u8 action);
+int __apei_exec_run(struct apei_exec_context *ctx, u8 action, bool optional);
+
+static inline int apei_exec_run(struct apei_exec_context *ctx, u8 action)
+{
+	return __apei_exec_run(ctx, action, 0);
+}
+
+/* It is optional whether the firmware provides the action */
+static inline int apei_exec_run_optional(struct apei_exec_context *ctx, u8 action)
+{
+	return __apei_exec_run(ctx, action, 1);
+}
 
 /* Common instruction implementation */
 

[PATCH 06/17] ACPI, APEI, Use apei_exec_run_optional in APEI EINJ and ERST

[Huang Ying]

This patch changes APEI EINJ and ERST to use apei_exec_run for
mandatory actions, and apei_exec_run_optional for optional actions.

Cc: Thomas Renninger <trenn@novell.com>
Signed-off-by: Huang Ying <ying.huang@intel.com>
---
 drivers/acpi/apei/einj.c |    4 ++--
 drivers/acpi/apei/erst.c |   12 ++++++------
 2 files changed, 8 insertions(+), 8 deletions(-)

--- a/drivers/acpi/apei/einj.c
+++ b/drivers/acpi/apei/einj.c
@@ -307,7 +307,7 @@ static int __einj_error_inject(u32 type,
 
 	einj_exec_ctx_init(&ctx);
 
-	rc = apei_exec_run(&ctx, ACPI_EINJ_BEGIN_OPERATION);
+	rc = apei_exec_run_optional(&ctx, ACPI_EINJ_BEGIN_OPERATION);
 	if (rc)
 		return rc;
 	apei_exec_ctx_set_input(&ctx, type);
@@ -345,7 +345,7 @@ static int __einj_error_inject(u32 type,
 	rc = __einj_error_trigger(trigger_paddr);
 	if (rc)
 		return rc;
-	rc = apei_exec_run(&ctx, ACPI_EINJ_END_OPERATION);
+	rc = apei_exec_run_optional(&ctx, ACPI_EINJ_END_OPERATION);
 
 	return rc;
 }
--- a/drivers/acpi/apei/erst.c
+++ b/drivers/acpi/apei/erst.c
@@ -642,7 +642,7 @@ static int __erst_write_to_storage(u64 o
 	int rc;
 
 	erst_exec_ctx_init(&ctx);
-	rc = apei_exec_run(&ctx, ACPI_ERST_BEGIN_WRITE);
+	rc = apei_exec_run_optional(&ctx, ACPI_ERST_BEGIN_WRITE);
 	if (rc)
 		return rc;
 	apei_exec_ctx_set_input(&ctx, offset);
@@ -666,7 +666,7 @@ static int __erst_write_to_storage(u64 o
 	if (rc)
 		return rc;
 	val = apei_exec_ctx_get_output(&ctx);
-	rc = apei_exec_run(&ctx, ACPI_ERST_END);
+	rc = apei_exec_run_optional(&ctx, ACPI_ERST_END);
 	if (rc)
 		return rc;
 
@@ -681,7 +681,7 @@ static int __erst_read_from_storage(u64
 	int rc;
 
 	erst_exec_ctx_init(&ctx);
-	rc = apei_exec_run(&ctx, ACPI_ERST_BEGIN_READ);
+	rc = apei_exec_run_optional(&ctx, ACPI_ERST_BEGIN_READ);
 	if (rc)
 		return rc;
 	apei_exec_ctx_set_input(&ctx, offset);
@@ -709,7 +709,7 @@ static int __erst_read_from_storage(u64
 	if (rc)
 		return rc;
 	val = apei_exec_ctx_get_output(&ctx);
-	rc = apei_exec_run(&ctx, ACPI_ERST_END);
+	rc = apei_exec_run_optional(&ctx, ACPI_ERST_END);
 	if (rc)
 		return rc;
 
@@ -724,7 +724,7 @@ static int __erst_clear_from_storage(u64
 	int rc;
 
 	erst_exec_ctx_init(&ctx);
-	rc = apei_exec_run(&ctx, ACPI_ERST_BEGIN_CLEAR);
+	rc = apei_exec_run_optional(&ctx, ACPI_ERST_BEGIN_CLEAR);
 	if (rc)
 		return rc;
 	apei_exec_ctx_set_input(&ctx, record_id);
@@ -748,7 +748,7 @@ static int __erst_clear_from_storage(u64
 	if (rc)
 		return rc;
 	val = apei_exec_ctx_get_output(&ctx);
-	rc = apei_exec_run(&ctx, ACPI_ERST_END);
+	rc = apei_exec_run_optional(&ctx, ACPI_ERST_END);
 	if (rc)
 		return rc;
 

[PATCH 07/17] ACPI, APEI, GHES, Prevent GHES to be built as module

[Huang Ying]

GHES (Generic Hardware Error Source) is used to process hardware error
notification in firmware first mode.  But because firmware first mode
can be turned on but can not be turned off, it is unreasonable to
unload the GHES module with firmware first mode turned on.  To avoid
confusion, this patch makes GHES can be enabled/disabled in
configuration time, but not built as module and unloaded at run time.

Signed-off-by: Huang Ying <ying.huang@intel.com>
Reviewed-by: Andi Kleen <ak@linux.intel.com>
Reviewed-by: Matthew Garrett <mjg@redhat.com>
---
 drivers/acpi/apei/Kconfig |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

--- a/drivers/acpi/apei/Kconfig
+++ b/drivers/acpi/apei/Kconfig
@@ -10,7 +10,7 @@ config ACPI_APEI
 	  error injection.
 
 config ACPI_APEI_GHES
-	tristate "APEI Generic Hardware Error Source"
+	bool "APEI Generic Hardware Error Source"
 	depends on ACPI_APEI && X86
 	select ACPI_HED
 	help

Re: [PATCH 07/17] ACPI, APEI, GHES, Prevent GHES to be built as module

[Len Brown]

applied

thanks,
Len Brown, Intel Open Source Technology Center

[PATCH 08/17] ACPI, APEI, GHES, Support disable GHES at boot time

[Huang Ying]

Some machine may have broken firmware so that GHES and firmware first
mode should be disabled.  This patch adds support to that.

Signed-off-by: Huang Ying <ying.huang@intel.com>
Reviewed-by: Andi Kleen <ak@linux.intel.com>
Reviewed-by: Matthew Garrett <mjg@redhat.com>
---
 drivers/acpi/apei/ghes.c |    8 ++++++++
 drivers/acpi/apei/hest.c |   17 +++++++++--------
 include/acpi/apei.h      |    1 +
 3 files changed, 18 insertions(+), 8 deletions(-)

--- a/drivers/acpi/apei/ghes.c
+++ b/drivers/acpi/apei/ghes.c
@@ -77,6 +77,9 @@ struct ghes {
 	};
 };
 
+int ghes_disable;
+module_param_named(disable, ghes_disable, bool, 0);
+
 static int ghes_panic_timeout	__read_mostly = 30;
 
 /*
@@ -665,6 +668,11 @@ static int __init ghes_init(void)
 		return -EINVAL;
 	}
 
+	if (ghes_disable) {
+		pr_info(GHES_PFX "GHES is not enabled!\n");
+		return -EINVAL;
+	}
+
 	rc = ghes_ioremap_init();
 	if (rc)
 		goto err;
--- a/drivers/acpi/apei/hest.c
+++ b/drivers/acpi/apei/hest.c
@@ -231,16 +231,17 @@ void __init acpi_hest_init(void)
 		goto err;
 	}
 
-	rc = apei_hest_parse(hest_parse_ghes_count, &ghes_count);
-	if (rc)
-		goto err;
-
-	rc = hest_ghes_dev_register(ghes_count);
-	if (!rc) {
-		pr_info(HEST_PFX "Table parsing has been initialized.\n");
-		return;
+	if (!ghes_disable) {
+		rc = apei_hest_parse(hest_parse_ghes_count, &ghes_count);
+		if (rc)
+			goto err;
+		rc = hest_ghes_dev_register(ghes_count);
+		if (rc)
+			goto err;
 	}
 
+	pr_info(HEST_PFX "Table parsing has been initialized.\n");
+	return;
 err:
 	hest_disable = 1;
 }
--- a/include/acpi/apei.h
+++ b/include/acpi/apei.h
@@ -18,6 +18,7 @@
 
 extern int hest_disable;
 extern int erst_disable;
+extern int ghes_disable;
 
 #ifdef CONFIG_ACPI_APEI
 void __init acpi_hest_init(void);

Re: [PATCH 08/17] ACPI, APEI, GHES, Support disable GHES at boot time

[Don Zickus]

On Wed, Jul 13, 2011 at 01:14:19PM +0800, Huang Ying wrote:
> Some machine may have broken firmware so that GHES and firmware first
> mode should be disabled.  This patch adds support to that.

Does this still work if you have disabled ghes module support in the
previous patch?

Cheers,
Don

> 
> Signed-off-by: Huang Ying <ying.huang@intel.com>
> Reviewed-by: Andi Kleen <ak@linux.intel.com>
> Reviewed-by: Matthew Garrett <mjg@redhat.com>
> ---
>  drivers/acpi/apei/ghes.c |    8 ++++++++
>  drivers/acpi/apei/hest.c |   17 +++++++++--------
>  include/acpi/apei.h      |    1 +
>  3 files changed, 18 insertions(+), 8 deletions(-)
> 
> --- a/drivers/acpi/apei/ghes.c
> +++ b/drivers/acpi/apei/ghes.c
> @@ -77,6 +77,9 @@ struct ghes {
>  	};
>  };
>  
> +int ghes_disable;
> +module_param_named(disable, ghes_disable, bool, 0);
> +
>  static int ghes_panic_timeout	__read_mostly = 30;
>  
>  /*
> @@ -665,6 +668,11 @@ static int __init ghes_init(void)
>  		return -EINVAL;
>  	}
>  
> +	if (ghes_disable) {
> +		pr_info(GHES_PFX "GHES is not enabled!\n");
> +		return -EINVAL;
> +	}
> +
>  	rc = ghes_ioremap_init();
>  	if (rc)
>  		goto err;
> --- a/drivers/acpi/apei/hest.c
> +++ b/drivers/acpi/apei/hest.c
> @@ -231,16 +231,17 @@ void __init acpi_hest_init(void)
>  		goto err;
>  	}
>  
> -	rc = apei_hest_parse(hest_parse_ghes_count, &ghes_count);
> -	if (rc)
> -		goto err;
> -
> -	rc = hest_ghes_dev_register(ghes_count);
> -	if (!rc) {
> -		pr_info(HEST_PFX "Table parsing has been initialized.\n");
> -		return;
> +	if (!ghes_disable) {
> +		rc = apei_hest_parse(hest_parse_ghes_count, &ghes_count);
> +		if (rc)
> +			goto err;
> +		rc = hest_ghes_dev_register(ghes_count);
> +		if (rc)
> +			goto err;
>  	}
>  
> +	pr_info(HEST_PFX "Table parsing has been initialized.\n");
> +	return;
>  err:
>  	hest_disable = 1;
>  }
> --- a/include/acpi/apei.h
> +++ b/include/acpi/apei.h
> @@ -18,6 +18,7 @@
>  
>  extern int hest_disable;
>  extern int erst_disable;
> +extern int ghes_disable;
>  
>  #ifdef CONFIG_ACPI_APEI
>  void __init acpi_hest_init(void);
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/

Re: [PATCH 08/17] ACPI, APEI, GHES, Support disable GHES at boot time

[Huang Ying]

On 07/14/2011 10:47 PM, Don Zickus wrote:
> On Wed, Jul 13, 2011 at 01:14:19PM +0800, Huang Ying wrote:
>> Some machine may have broken firmware so that GHES and firmware first
>> mode should be disabled.  This patch adds support to that.
> 
> Does this still work if you have disabled ghes module support in the
> previous patch?

Yes.  Module param can be specified in kernel command line, for example
"ghes.disable=y".

Best Regards,
Huang Ying

Re: [PATCH 08/17] ACPI, APEI, GHES, Support disable GHES at boot time

[Don Zickus]

On Fri, Jul 15, 2011 at 08:31:00AM +0800, Huang Ying wrote:
> On 07/14/2011 10:47 PM, Don Zickus wrote:
> > On Wed, Jul 13, 2011 at 01:14:19PM +0800, Huang Ying wrote:
> >> Some machine may have broken firmware so that GHES and firmware first
> >> mode should be disabled.  This patch adds support to that.
> > 
> > Does this still work if you have disabled ghes module support in the
> > previous patch?
> 
> Yes.  Module param can be specified in kernel command line, for example
> "ghes.disable=y".

Ok.  I just tested that too.  Didn't realize a built-in could still use
module params.

Tested-and-Reviewed-by: Don Zickus <dzickus@redhat.com>

[PATCH 09/17] ACPI, APEI, Add APEI bit support in generic _OSC call

[Huang Ying]

In APEI firmware first mode, hardware error is reported by hardware to
firmware firstly, then firmware reports the error to Linux in a GHES
error record via POLL/SCI/IRQ/NMI etc.

This may result in some issues if OS has no full APEI support.  So
some firmware implementation will work in a back-compatible mode by
default.  Where firmware will only notify OS in old-fashion, without
GHES record.  For example, for a fatal hardware error, only NMI is
signaled, no GHES record.

To gain full APEI power on these machines, APEI bit in generic _OSC
call can be specified to tell firmware that Linux has full APEI
support.  This patch adds the APEI bit support in generic _OSC call.

Signed-off-by: Huang Ying <ying.huang@intel.com>
Reviewed-by: Andi Kleen <ak@linux.intel.com>
Reviewed-by: Matthew Garrett <mjg@redhat.com>
---
 drivers/acpi/bus.c   |   16 ++++++++++++++--
 include/linux/acpi.h |    2 ++
 2 files changed, 16 insertions(+), 2 deletions(-)

--- a/drivers/acpi/bus.c
+++ b/drivers/acpi/bus.c
@@ -39,6 +39,7 @@
 #include <linux/pci.h>
 #include <acpi/acpi_bus.h>
 #include <acpi/acpi_drivers.h>
+#include <acpi/apei.h>
 #include <linux/dmi.h>
 #include <linux/suspend.h>
 
@@ -519,6 +520,7 @@ out_kfree:
 }
 EXPORT_SYMBOL(acpi_run_osc);
 
+bool osc_sb_apei_support_acked;
 static u8 sb_uuid_str[] = "0811B06E-4A27-44F9-8D60-3CBBC22E7B48";
 static void acpi_bus_osc_support(void)
 {
@@ -541,11 +543,21 @@ static void acpi_bus_osc_support(void)
 #if defined(CONFIG_ACPI_PROCESSOR) || defined(CONFIG_ACPI_PROCESSOR_MODULE)
 	capbuf[OSC_SUPPORT_TYPE] |= OSC_SB_PPC_OST_SUPPORT;
 #endif
+
+#ifdef CONFIG_ACPI_APEI_GHES
+	if (!ghes_disable)
+		capbuf[OSC_SUPPORT_TYPE] |= OSC_SB_APEI_SUPPORT;
+#endif
 	if (ACPI_FAILURE(acpi_get_handle(NULL, "\\_SB", &handle)))
 		return;
-	if (ACPI_SUCCESS(acpi_run_osc(handle, &context)))
+	if (ACPI_SUCCESS(acpi_run_osc(handle, &context))) {
+		u32 *capbuf_ret = context.ret.pointer;
+		if (context.ret.length > OSC_SUPPORT_TYPE)
+			osc_sb_apei_support_acked =
+				capbuf_ret[OSC_SUPPORT_TYPE] & OSC_SB_APEI_SUPPORT;
 		kfree(context.ret.pointer);
-	/* do we need to check the returned cap? Sounds no */
+	}
+	/* do we need to check other returned cap? Sounds no */
 }
 
 /* --------------------------------------------------------------------------
--- a/include/linux/acpi.h
+++ b/include/linux/acpi.h
@@ -280,6 +280,8 @@ acpi_status acpi_run_osc(acpi_handle han
 #define OSC_SB_CPUHP_OST_SUPPORT	8
 #define OSC_SB_APEI_SUPPORT		16
 
+extern bool osc_sb_apei_support_acked;
+
 /* PCI defined _OSC bits */
 /* _OSC DW1 Definition (OS Support Fields) */
 #define OSC_EXT_PCI_CONFIG_SUPPORT		1

Re: [PATCH 09/17] ACPI, APEI, Add APEI bit support in generic _OSC call

[Don Zickus]

On Wed, Jul 13, 2011 at 01:14:20PM +0800, Huang Ying wrote:
> In APEI firmware first mode, hardware error is reported by hardware to
> firmware firstly, then firmware reports the error to Linux in a GHES
> error record via POLL/SCI/IRQ/NMI etc.
> 
> This may result in some issues if OS has no full APEI support.  So
> some firmware implementation will work in a back-compatible mode by
> default.  Where firmware will only notify OS in old-fashion, without
> GHES record.  For example, for a fatal hardware error, only NMI is
> signaled, no GHES record.
> 
> To gain full APEI power on these machines, APEI bit in generic _OSC
> call can be specified to tell firmware that Linux has full APEI
> support.  This patch adds the APEI bit support in generic _OSC call.

I have been testing this patch for a couple of weeks with no issues that I
have seen.  Without it (and/or the next one), APEI can't really work.

Tested-and-reviewed-by: Don Zickus <dzickus@redhat.com>

> 
> Signed-off-by: Huang Ying <ying.huang@intel.com>
> Reviewed-by: Andi Kleen <ak@linux.intel.com>
> Reviewed-by: Matthew Garrett <mjg@redhat.com>
> ---
>  drivers/acpi/bus.c   |   16 ++++++++++++++--
>  include/linux/acpi.h |    2 ++
>  2 files changed, 16 insertions(+), 2 deletions(-)
> 
> --- a/drivers/acpi/bus.c
> +++ b/drivers/acpi/bus.c
> @@ -39,6 +39,7 @@
>  #include <linux/pci.h>
>  #include <acpi/acpi_bus.h>
>  #include <acpi/acpi_drivers.h>
> +#include <acpi/apei.h>
>  #include <linux/dmi.h>
>  #include <linux/suspend.h>
>  
> @@ -519,6 +520,7 @@ out_kfree:
>  }
>  EXPORT_SYMBOL(acpi_run_osc);
>  
> +bool osc_sb_apei_support_acked;
>  static u8 sb_uuid_str[] = "0811B06E-4A27-44F9-8D60-3CBBC22E7B48";
>  static void acpi_bus_osc_support(void)
>  {
> @@ -541,11 +543,21 @@ static void acpi_bus_osc_support(void)
>  #if defined(CONFIG_ACPI_PROCESSOR) || defined(CONFIG_ACPI_PROCESSOR_MODULE)
>  	capbuf[OSC_SUPPORT_TYPE] |= OSC_SB_PPC_OST_SUPPORT;
>  #endif
> +
> +#ifdef CONFIG_ACPI_APEI_GHES
> +	if (!ghes_disable)
> +		capbuf[OSC_SUPPORT_TYPE] |= OSC_SB_APEI_SUPPORT;
> +#endif
>  	if (ACPI_FAILURE(acpi_get_handle(NULL, "\\_SB", &handle)))
>  		return;
> -	if (ACPI_SUCCESS(acpi_run_osc(handle, &context)))
> +	if (ACPI_SUCCESS(acpi_run_osc(handle, &context))) {
> +		u32 *capbuf_ret = context.ret.pointer;
> +		if (context.ret.length > OSC_SUPPORT_TYPE)
> +			osc_sb_apei_support_acked =
> +				capbuf_ret[OSC_SUPPORT_TYPE] & OSC_SB_APEI_SUPPORT;
>  		kfree(context.ret.pointer);
> -	/* do we need to check the returned cap? Sounds no */
> +	}
> +	/* do we need to check other returned cap? Sounds no */
>  }
>  
>  /* --------------------------------------------------------------------------
> --- a/include/linux/acpi.h
> +++ b/include/linux/acpi.h
> @@ -280,6 +280,8 @@ acpi_status acpi_run_osc(acpi_handle han
>  #define OSC_SB_CPUHP_OST_SUPPORT	8
>  #define OSC_SB_APEI_SUPPORT		16
>  
> +extern bool osc_sb_apei_support_acked;
> +
>  /* PCI defined _OSC bits */
>  /* _OSC DW1 Definition (OS Support Fields) */
>  #define OSC_EXT_PCI_CONFIG_SUPPORT		1
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/

[PATCH 10/17] ACPI, APEI, Add WHEA _OSC support

[Huang Ying]

APEI firmware first mode must be turned on explicitly on some
machines, otherwise there may be no GHES hardware error record for
hardware error notification.  APEI bit in generic _OSC call can be
used to do that, but on some machine, a special WHEA _OSC call must be
used.  This patch adds the support to that WHEA _OSC call.

Signed-off-by: Huang Ying <ying.huang@intel.com>
Reviewed-by: Andi Kleen <ak@linux.intel.com>
Reviewed-by: Matthew Garrett <mjg@redhat.com>
---
 drivers/acpi/apei/apei-base.c     |   26 ++++++++++++++++++++++++++
 drivers/acpi/apei/apei-internal.h |    2 ++
 drivers/acpi/apei/ghes.c          |   10 ++++++++++
 3 files changed, 38 insertions(+)

--- a/drivers/acpi/apei/apei-base.c
+++ b/drivers/acpi/apei/apei-base.c
@@ -604,3 +604,29 @@ struct dentry *apei_get_debugfs_dir(void
 	return dapei;
 }
 EXPORT_SYMBOL_GPL(apei_get_debugfs_dir);
+
+int apei_osc_setup(void)
+{
+	static u8 whea_uuid_str[] = "ed855e0c-6c90-47bf-a62a-26de0fc5ad5c";
+	acpi_handle handle;
+	u32 capbuf[3];
+	struct acpi_osc_context context = {
+		.uuid_str	= whea_uuid_str,
+		.rev		= 1,
+		.cap.length	= sizeof(capbuf),
+		.cap.pointer	= capbuf,
+	};
+
+	capbuf[OSC_QUERY_TYPE] = OSC_QUERY_ENABLE;
+	capbuf[OSC_SUPPORT_TYPE] = 0;
+	capbuf[OSC_CONTROL_TYPE] = 0;
+
+	if (ACPI_FAILURE(acpi_get_handle(NULL, "\\_SB", &handle))
+	    || ACPI_FAILURE(acpi_run_osc(handle, &context)))
+		return -EIO;
+	else {
+		kfree(context.ret.pointer);
+		return 0;
+	}
+}
+EXPORT_SYMBOL_GPL(apei_osc_setup);
--- a/drivers/acpi/apei/apei-internal.h
+++ b/drivers/acpi/apei/apei-internal.h
@@ -124,4 +124,6 @@ void apei_estatus_print(const char *pfx,
 			const struct acpi_hest_generic_status *estatus);
 int apei_estatus_check_header(const struct acpi_hest_generic_status *estatus);
 int apei_estatus_check(const struct acpi_hest_generic_status *estatus);
+
+int apei_osc_setup(void);
 #endif
--- a/drivers/acpi/apei/ghes.c
+++ b/drivers/acpi/apei/ghes.c
@@ -681,6 +681,16 @@ static int __init ghes_init(void)
 	if (rc)
 		goto err_ioremap_exit;
 
+	rc = apei_osc_setup();
+	if (rc == 0 && osc_sb_apei_support_acked)
+		pr_info(GHES_PFX "APEI firmware first mode is enabled by APEI bit and WHEA _OSC.\n");
+	else if (rc == 0 && !osc_sb_apei_support_acked)
+		pr_info(GHES_PFX "APEI firmware first mode is enabled by WHEA _OSC.\n");
+	else if (rc && osc_sb_apei_support_acked)
+		pr_info(GHES_PFX "APEI firmware first mode is enabled by APEI bit.\n");
+	else
+		pr_info(GHES_PFX "Failed to enable APEI firmware first mode.\n");
+
 	return 0;
 err_ioremap_exit:
 	ghes_ioremap_exit();

Re: [PATCH 10/17] ACPI, APEI, Add WHEA _OSC support

[Don Zickus]

On Wed, Jul 13, 2011 at 01:14:21PM +0800, Huang Ying wrote:
> APEI firmware first mode must be turned on explicitly on some
> machines, otherwise there may be no GHES hardware error record for
> hardware error notification.  APEI bit in generic _OSC call can be
> used to do that, but on some machine, a special WHEA _OSC call must be
> used.  This patch adds the support to that WHEA _OSC call.

Without this patch APEI can't do anything.  I have been using this for a
few weeks to do some internal APEI testing on various machines with no
issues (at least with regards to this patch).

Tested-and-reviewed-by: Don Zickus <dzickus@redhat.com>

> 
> Signed-off-by: Huang Ying <ying.huang@intel.com>
> Reviewed-by: Andi Kleen <ak@linux.intel.com>
> Reviewed-by: Matthew Garrett <mjg@redhat.com>
> ---
>  drivers/acpi/apei/apei-base.c     |   26 ++++++++++++++++++++++++++
>  drivers/acpi/apei/apei-internal.h |    2 ++
>  drivers/acpi/apei/ghes.c          |   10 ++++++++++
>  3 files changed, 38 insertions(+)
> 
> --- a/drivers/acpi/apei/apei-base.c
> +++ b/drivers/acpi/apei/apei-base.c
> @@ -604,3 +604,29 @@ struct dentry *apei_get_debugfs_dir(void
>  	return dapei;
>  }
>  EXPORT_SYMBOL_GPL(apei_get_debugfs_dir);
> +
> +int apei_osc_setup(void)
> +{
> +	static u8 whea_uuid_str[] = "ed855e0c-6c90-47bf-a62a-26de0fc5ad5c";
> +	acpi_handle handle;
> +	u32 capbuf[3];
> +	struct acpi_osc_context context = {
> +		.uuid_str	= whea_uuid_str,
> +		.rev		= 1,
> +		.cap.length	= sizeof(capbuf),
> +		.cap.pointer	= capbuf,
> +	};
> +
> +	capbuf[OSC_QUERY_TYPE] = OSC_QUERY_ENABLE;
> +	capbuf[OSC_SUPPORT_TYPE] = 0;
> +	capbuf[OSC_CONTROL_TYPE] = 0;
> +
> +	if (ACPI_FAILURE(acpi_get_handle(NULL, "\\_SB", &handle))
> +	    || ACPI_FAILURE(acpi_run_osc(handle, &context)))
> +		return -EIO;
> +	else {
> +		kfree(context.ret.pointer);
> +		return 0;
> +	}
> +}
> +EXPORT_SYMBOL_GPL(apei_osc_setup);
> --- a/drivers/acpi/apei/apei-internal.h
> +++ b/drivers/acpi/apei/apei-internal.h
> @@ -124,4 +124,6 @@ void apei_estatus_print(const char *pfx,
>  			const struct acpi_hest_generic_status *estatus);
>  int apei_estatus_check_header(const struct acpi_hest_generic_status *estatus);
>  int apei_estatus_check(const struct acpi_hest_generic_status *estatus);
> +
> +int apei_osc_setup(void);
>  #endif
> --- a/drivers/acpi/apei/ghes.c
> +++ b/drivers/acpi/apei/ghes.c
> @@ -681,6 +681,16 @@ static int __init ghes_init(void)
>  	if (rc)
>  		goto err_ioremap_exit;
>  
> +	rc = apei_osc_setup();
> +	if (rc == 0 && osc_sb_apei_support_acked)
> +		pr_info(GHES_PFX "APEI firmware first mode is enabled by APEI bit and WHEA _OSC.\n");
> +	else if (rc == 0 && !osc_sb_apei_support_acked)
> +		pr_info(GHES_PFX "APEI firmware first mode is enabled by WHEA _OSC.\n");
> +	else if (rc && osc_sb_apei_support_acked)
> +		pr_info(GHES_PFX "APEI firmware first mode is enabled by APEI bit.\n");
> +	else
> +		pr_info(GHES_PFX "Failed to enable APEI firmware first mode.\n");
> +
>  	return 0;
>  err_ioremap_exit:
>  	ghes_ioremap_exit();
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/

[PATCH 11/17] Add Kconfig option ARCH_HAVE_NMI_SAFE_CMPXCHG

[Huang Ying]

cmpxchg() is widely used by lockless code, including NMI-safe lockless
code.  But on some architectures, the cmpxchg() implementation is not
NMI-safe, on these architectures the lockless code may need a
spin_trylock_irqsave() based implementation.

This patch adds a Kconfig option: ARCH_HAVE_NMI_SAFE_CMPXCHG, so that
NMI-safe lockless code can depend on it or provide different
implementation according to it.

On many architectures, cmpxchg is only NMI-safe for several specific
operand sizes. So, ARCH_HAVE_NMI_SAFE_CMPXCHG define in this patch
only guarantees cmpxchg is NMI-safe for sizeof(unsigned long).

Signed-off-by: Huang Ying <ying.huang@intel.com>
Acked-by: Mike Frysinger <vapier@gentoo.org>
Acked-by: Paul Mundt <lethal@linux-sh.org>
Acked-by: Hans-Christian Egtvedt <hans-christian.egtvedt@atmel.com>
Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Acked-by: Chris Metcalf <cmetcalf@tilera.com>
CC: Richard Henderson <rth@twiddle.net>
CC: Mikael Starvik <starvik@axis.com>
CC: David Howells <dhowells@redhat.com>
CC: Yoshinori Sato <ysato@users.sourceforge.jp>
CC: Tony Luck <tony.luck@intel.com>
CC: Hirokazu Takata <takata@linux-m32r.org>
CC: Geert Uytterhoeven <geert@linux-m68k.org>
CC: Michal Simek <monstr@monstr.eu>
CC: Ralf Baechle <ralf@linux-mips.org>
CC: Kyle McMartin <kyle@mcmartin.ca>
CC: Martin Schwidefsky <schwidefsky@de.ibm.com>
CC: Chen Liqin <liqin.chen@sunplusct.com>
CC: "David S. Miller" <davem@davemloft.net>
CC: Ingo Molnar <mingo@redhat.com>
CC: Chris Zankel <chris@zankel.net>
---
 arch/Kconfig         |    3 +++
 arch/alpha/Kconfig   |    1 +
 arch/avr32/Kconfig   |    1 +
 arch/frv/Kconfig     |    1 +
 arch/ia64/Kconfig    |    1 +
 arch/m68k/Kconfig    |    1 +
 arch/parisc/Kconfig  |    1 +
 arch/powerpc/Kconfig |    1 +
 arch/s390/Kconfig    |    1 +
 arch/sh/Kconfig      |    1 +
 arch/sparc/Kconfig   |    1 +
 arch/tile/Kconfig    |    1 +
 arch/x86/Kconfig     |    1 +
 13 files changed, 15 insertions(+)

--- a/arch/Kconfig
+++ b/arch/Kconfig
@@ -178,4 +178,7 @@ config HAVE_ARCH_MUTEX_CPU_RELAX
 config HAVE_RCU_TABLE_FREE
 	bool
 
+config ARCH_HAVE_NMI_SAFE_CMPXCHG
+	bool
+
 source "kernel/gcov/Kconfig"
--- a/arch/alpha/Kconfig
+++ b/arch/alpha/Kconfig
@@ -13,6 +13,7 @@ config ALPHA
 	select AUTO_IRQ_AFFINITY if SMP
 	select GENERIC_IRQ_SHOW
 	select ARCH_WANT_OPTIONAL_GPIOLIB
+	select ARCH_HAVE_NMI_SAFE_CMPXCHG
 	help
 	  The Alpha is a 64-bit general-purpose processor designed and
 	  marketed by the Digital Equipment Corporation of blessed memory,
--- a/arch/avr32/Kconfig
+++ b/arch/avr32/Kconfig
@@ -10,6 +10,7 @@ config AVR32
 	select GENERIC_IRQ_PROBE
 	select HARDIRQS_SW_RESEND
 	select GENERIC_IRQ_SHOW
+	select ARCH_HAVE_NMI_SAFE_CMPXCHG
 	help
 	  AVR32 is a high-performance 32-bit RISC microprocessor core,
 	  designed for cost-sensitive embedded applications, with particular
--- a/arch/frv/Kconfig
+++ b/arch/frv/Kconfig
@@ -7,6 +7,7 @@ config FRV
 	select HAVE_PERF_EVENTS
 	select HAVE_GENERIC_HARDIRQS
 	select GENERIC_IRQ_SHOW
+	select ARCH_HAVE_NMI_SAFE_CMPXCHG
 
 config ZONE_DMA
 	bool
--- a/arch/ia64/Kconfig
+++ b/arch/ia64/Kconfig
@@ -27,6 +27,7 @@ config IA64
 	select GENERIC_PENDING_IRQ if SMP
 	select IRQ_PER_CPU
 	select GENERIC_IRQ_SHOW
+	select ARCH_HAVE_NMI_SAFE_CMPXCHG
 	default y
 	help
 	  The Itanium Processor Family is Intel's 64-bit successor to
--- a/arch/m68k/Kconfig
+++ b/arch/m68k/Kconfig
@@ -6,6 +6,7 @@ config M68K
 	select GENERIC_ATOMIC64 if MMU
 	select HAVE_GENERIC_HARDIRQS if !MMU
 	select GENERIC_IRQ_SHOW if !MMU
+	select ARCH_HAVE_NMI_SAFE_CMPXCHG if RMW_INSNS
 
 config RWSEM_GENERIC_SPINLOCK
 	bool
--- a/arch/parisc/Kconfig
+++ b/arch/parisc/Kconfig
@@ -15,6 +15,7 @@ config PARISC
 	select HAVE_GENERIC_HARDIRQS
 	select GENERIC_IRQ_PROBE
 	select IRQ_PER_CPU
+	select ARCH_HAVE_NMI_SAFE_CMPXCHG
 
 	help
 	  The PA-RISC microprocessor is designed by Hewlett-Packard and used
--- a/arch/powerpc/Kconfig
+++ b/arch/powerpc/Kconfig
@@ -134,6 +134,7 @@ config PPC
 	select GENERIC_IRQ_SHOW_LEVEL
 	select HAVE_RCU_TABLE_FREE if SMP
 	select HAVE_SYSCALL_TRACEPOINTS
+	select ARCH_HAVE_NMI_SAFE_CMPXCHG
 
 config EARLY_PRINTK
 	bool
--- a/arch/s390/Kconfig
+++ b/arch/s390/Kconfig
@@ -81,6 +81,7 @@ config S390
 	select INIT_ALL_POSSIBLE
 	select HAVE_IRQ_WORK
 	select HAVE_PERF_EVENTS
+	select ARCH_HAVE_NMI_SAFE_CMPXCHG
 	select HAVE_KERNEL_GZIP
 	select HAVE_KERNEL_BZIP2
 	select HAVE_KERNEL_LZMA
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -11,6 +11,7 @@ config SUPERH
 	select HAVE_DMA_ATTRS
 	select HAVE_IRQ_WORK
 	select HAVE_PERF_EVENTS
+	select ARCH_HAVE_NMI_SAFE_CMPXCHG if (GUSA_RB || CPU_SH4A)
 	select PERF_USE_VMALLOC
 	select HAVE_KERNEL_GZIP
 	select HAVE_KERNEL_BZIP2
--- a/arch/sparc/Kconfig
+++ b/arch/sparc/Kconfig
@@ -53,6 +53,7 @@ config SPARC64
 	select HAVE_PERF_EVENTS
 	select PERF_USE_VMALLOC
 	select IRQ_PREFLOW_FASTEOI
+	select ARCH_HAVE_NMI_SAFE_CMPXCHG
 
 config ARCH_DEFCONFIG
 	string
--- a/arch/tile/Kconfig
+++ b/arch/tile/Kconfig
@@ -12,6 +12,7 @@ config TILE
 	select GENERIC_PENDING_IRQ if SMP
 	select GENERIC_IRQ_SHOW
 	select SYS_HYPERVISOR
+	select ARCH_HAVE_NMI_SAFE_CMPXCHG if !M386
 
 # FIXME: investigate whether we need/want these options.
 #	select HAVE_IOREMAP_PROT
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -70,6 +70,7 @@ config X86
 	select IRQ_FORCED_THREADING
 	select USE_GENERIC_SMP_HELPERS if SMP
 	select HAVE_BPF_JIT if (X86_64 && NET)
+	select ARCH_HAVE_NMI_SAFE_CMPXCHG
 
 config INSTRUCTION_DECODER
 	def_bool (KPROBES || PERF_EVENTS)

[PATCH 12/17] lib, Add lock-less NULL terminated single list

[Huang Ying]

Cmpxchg is used to implement adding new entry to the list, deleting
all entries from the list, deleting first entry of the list and some
other operations.

Because this is a single list, so the tail can not be accessed in O(1).

If there are multiple producers and multiple consumers, llist_add can
be used in producers and llist_del_all can be used in consumers.  They
can work simultaneously without lock.  But llist_del_first can not be
used here.  Because llist_del_first depends on list->first->next does
not changed if list->first is not changed during its operation, but
llist_del_first, llist_add, llist_add (or llist_del_all, llist_add,
llist_add) sequence in another consumer may violate that.

If there are multiple producers and one consumer, llist_add can be
used in producers and llist_del_all or llist_del_first can be used in
the consumer.

This can be summarized as follow:

           |   add    | del_first |  del_all
 add       |    -     |     -     |     -
 del_first |          |     L     |     L
 del_all   |          |           |     -

Where "-" stands for no lock is needed, while "L" stands for lock is
needed.

The list entries deleted via llist_del_all can be traversed with
traversing function such as llist_for_each etc.  But the list entries
can not be traversed safely before deleted from the list.  The order
of deleted entries is from the newest to the oldest added one.  If you
want to traverse from the oldest to the newest, you must reverse the
order by yourself before traversing.

The basic atomic operation of this list is cmpxchg on long.  On
architectures that don't have NMI-safe cmpxchg implementation, the
list can NOT be used in NMI handler.  So code uses the list in NMI
handler should depend on CONFIG_ARCH_HAVE_NMI_SAFE_CMPXCHG.

Signed-off-by: Huang Ying <ying.huang@intel.com>
Reviewed-by: Andi Kleen <ak@linux.intel.com>
Reviewed-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
---
 include/linux/llist.h |  126 ++++++++++++++++++++++++++++++++++++++++++++++++
 lib/Kconfig           |    3 +
 lib/Makefile          |    2 
 lib/llist.c           |  129 ++++++++++++++++++++++++++++++++++++++++++++++++++
 4 files changed, 260 insertions(+)
 create mode 100644 include/linux/llist.h
 create mode 100644 lib/llist.c

--- /dev/null
+++ b/include/linux/llist.h
@@ -0,0 +1,126 @@
+#ifndef LLIST_H
+#define LLIST_H
+/*
+ * Lock-less NULL terminated single linked list
+ *
+ * If there are multiple producers and multiple consumers, llist_add
+ * can be used in producers and llist_del_all can be used in
+ * consumers.  They can work simultaneously without lock.  But
+ * llist_del_first can not be used here.  Because llist_del_first
+ * depends on list->first->next does not changed if list->first is not
+ * changed during its operation, but llist_del_first, llist_add,
+ * llist_add (or llist_del_all, llist_add, llist_add) sequence in
+ * another consumer may violate that.
+ *
+ * If there are multiple producers and one consumer, llist_add can be
+ * used in producers and llist_del_all or llist_del_first can be used
+ * in the consumer.
+ *
+ * This can be summarized as follow:
+ *
+ *           |   add    | del_first |  del_all
+ * add       |    -     |     -     |     -
+ * del_first |          |     L     |     L
+ * del_all   |          |           |     -
+ *
+ * Where "-" stands for no lock is needed, while "L" stands for lock
+ * is needed.
+ *
+ * The list entries deleted via llist_del_all can be traversed with
+ * traversing function such as llist_for_each etc.  But the list
+ * entries can not be traversed safely before deleted from the list.
+ * The order of deleted entries is from the newest to the oldest added
+ * one.  If you want to traverse from the oldest to the newest, you
+ * must reverse the order by yourself before traversing.
+ *
+ * The basic atomic operation of this list is cmpxchg on long.  On
+ * architectures that don't have NMI-safe cmpxchg implementation, the
+ * list can NOT be used in NMI handler.  So code uses the list in NMI
+ * handler should depend on CONFIG_ARCH_HAVE_NMI_SAFE_CMPXCHG.
+ */
+
+struct llist_head {
+	struct llist_node *first;
+};
+
+struct llist_node {
+	struct llist_node *next;
+};
+
+#define LLIST_HEAD_INIT(name)	{ NULL }
+#define LLIST_HEAD(name)	struct llist_head name = LLIST_HEAD_INIT(name)
+
+/**
+ * init_llist_head - initialize lock-less list head
+ * @head:	the head for your lock-less list
+ */
+static inline void init_llist_head(struct llist_head *list)
+{
+	list->first = NULL;
+}
+
+/**
+ * llist_entry - get the struct of this entry
+ * @ptr:	the &struct llist_node pointer.
+ * @type:	the type of the struct this is embedded in.
+ * @member:	the name of the llist_node within the struct.
+ */
+#define llist_entry(ptr, type, member)		\
+	container_of(ptr, type, member)
+
+/**
+ * llist_for_each - iterate over some deleted entries of a lock-less list
+ * @pos:	the &struct llist_node to use as a loop cursor
+ * @node:	the first entry of deleted list entries
+ *
+ * In general, some entries of the lock-less list can be traversed
+ * safely only after being deleted from list, so start with an entry
+ * instead of list head.
+ *
+ * If being used on entries deleted from lock-less list directly, the
+ * traverse order is from the newest to the oldest added entry.  If
+ * you want to traverse from the oldest to the newest, you must
+ * reverse the order by yourself before traversing.
+ */
+#define llist_for_each(pos, node)			\
+	for ((pos) = (node); pos; (pos) = (pos)->next)
+
+/**
+ * llist_for_each_entry - iterate over some deleted entries of lock-less list of given type
+ * @pos:	the type * to use as a loop cursor.
+ * @node:	the fist entry of deleted list entries.
+ * @member:	the name of the llist_node with the struct.
+ *
+ * In general, some entries of the lock-less list can be traversed
+ * safely only after being removed from list, so start with an entry
+ * instead of list head.
+ *
+ * If being used on entries deleted from lock-less list directly, the
+ * traverse order is from the newest to the oldest added entry.  If
+ * you want to traverse from the oldest to the newest, you must
+ * reverse the order by yourself before traversing.
+ */
+#define llist_for_each_entry(pos, node, member)				\
+	for ((pos) = llist_entry((node), typeof(*(pos)), member);	\
+	     &(pos)->member != NULL;					\
+	     (pos) = llist_entry((pos)->member.next, typeof(*(pos)), member))
+
+/**
+ * llist_empty - tests whether a lock-less list is empty
+ * @head:	the list to test
+ *
+ * Not guaranteed to be accurate or up to date.  Just a quick way to
+ * test whether the list is empty without deleting something from the
+ * list.
+ */
+static inline int llist_empty(const struct llist_head *head)
+{
+	return ACCESS_ONCE(head->first) == NULL;
+}
+
+void llist_add(struct llist_node *new, struct llist_head *head);
+void llist_add_batch(struct llist_node *new_first, struct llist_node *new_last,
+		     struct llist_head *head);
+struct llist_node *llist_del_first(struct llist_head *head);
+struct llist_node *llist_del_all(struct llist_head *head);
+#endif /* LLIST_H */
--- a/lib/Kconfig
+++ b/lib/Kconfig
@@ -262,4 +262,7 @@ config AVERAGE
 
 	  If unsure, say N.
 
+config LLIST
+	bool
+
 endmenu
--- a/lib/Makefile
+++ b/lib/Makefile
@@ -112,6 +112,8 @@ obj-$(CONFIG_AVERAGE) += average.o
 
 obj-$(CONFIG_CPU_RMAP) += cpu_rmap.o
 
+obj-$(CONFIG_LLIST) += llist.o
+
 hostprogs-y	:= gen_crc32table
 clean-files	:= crc32table.h
 
--- /dev/null
+++ b/lib/llist.c
@@ -0,0 +1,129 @@
+/*
+ * Lock-less NULL terminated single linked list
+ *
+ * The basic atomic operation of this list is cmpxchg on long.  On
+ * architectures that don't have NMI-safe cmpxchg implementation, the
+ * list can NOT be used in NMI handler.  So code uses the list in NMI
+ * handler should depend on CONFIG_ARCH_HAVE_NMI_SAFE_CMPXCHG.
+ *
+ * Copyright 2010,2011 Intel Corp.
+ *   Author: Huang Ying <ying.huang@intel.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License version
+ * 2 as published by the Free Software Foundation;
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ */
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/interrupt.h>
+#include <linux/llist.h>
+
+#include <asm/system.h>
+
+/**
+ * llist_add - add a new entry
+ * @new:	new entry to be added
+ * @head:	the head for your lock-less list
+ */
+void llist_add(struct llist_node *new, struct llist_head *head)
+{
+	struct llist_node *entry, *old_entry;
+
+#ifndef CONFIG_ARCH_HAVE_NMI_SAFE_CMPXCHG
+	BUG_ON(in_nmi());
+#endif
+
+	entry = head->first;
+	do {
+		old_entry = entry;
+		new->next = entry;
+		cpu_relax();
+	} while ((entry = cmpxchg(&head->first, old_entry, new)) != old_entry);
+}
+EXPORT_SYMBOL_GPL(llist_add);
+
+/**
+ * llist_add_batch - add several linked entries in batch
+ * @new_first:	first entry in batch to be added
+ * @new_last:	last entry in batch to be added
+ * @head:	the head for your lock-less list
+ */
+void llist_add_batch(struct llist_node *new_first, struct llist_node *new_last,
+		     struct llist_head *head)
+{
+	struct llist_node *entry, *old_entry;
+
+#ifndef CONFIG_ARCH_HAVE_NMI_SAFE_CMPXCHG
+	BUG_ON(in_nmi());
+#endif
+
+	entry = head->first;
+	do {
+		old_entry = entry;
+		new_last->next = entry;
+		cpu_relax();
+	} while ((entry = cmpxchg(&head->first, old_entry, new_first)) != old_entry);
+}
+EXPORT_SYMBOL_GPL(llist_add_batch);
+
+/**
+ * llist_del_first - delete the first entry of lock-less list
+ * @head:	the head for your lock-less list
+ *
+ * If list is empty, return NULL, otherwise, return the first entry
+ * deleted, this is the newest added one.
+ *
+ * Only one llist_del_first user can be used simultaneously with
+ * multiple llist_add users without lock.  Because otherwise
+ * llist_del_first, llist_add, llist_add (or llist_del_all, llist_add,
+ * llist_add) sequence in another user may change @head->first->next,
+ * but keep @head->first.  If multiple consumers are needed, please
+ * use llist_del_all or use lock between consumers.
+ */
+struct llist_node *llist_del_first(struct llist_head *head)
+{
+	struct llist_node *entry, *old_entry, *next;
+
+#ifndef CONFIG_ARCH_HAVE_NMI_SAFE_CMPXCHG
+	BUG_ON(in_nmi());
+#endif
+
+	entry = head->first;
+	do {
+		if (entry == NULL)
+			return NULL;
+		old_entry = entry;
+		next = entry->next;
+		cpu_relax();
+	} while ((entry = cmpxchg(&head->first, old_entry, next)) != old_entry);
+
+	return entry;
+}
+EXPORT_SYMBOL_GPL(llist_del_first);
+
+/**
+ * llist_del_all - delete all entries from lock-less list
+ * @head:	the head of lock-less list to delete all entries
+ *
+ * If list is empty, return NULL, otherwise, delete all entries and
+ * return the pointer to the first entry.  The order of entries
+ * deleted is from the newest to the oldest added one.
+ */
+struct llist_node *llist_del_all(struct llist_head *head)
+{
+#ifndef CONFIG_ARCH_HAVE_NMI_SAFE_CMPXCHG
+	BUG_ON(in_nmi());
+#endif
+
+	return xchg(&head->first, NULL);
+}
+EXPORT_SYMBOL_GPL(llist_del_all);

[PATCH 13/17] lib, Make gen_pool memory allocator lockless

[Huang Ying]

This version of the gen_pool memory allocator supports lockless
operation.

This makes it safe to use in NMI handlers and other special
unblockable contexts that could otherwise deadlock on locks.  This is
implemented by using atomic operations and retries on any conflicts.
The disadvantage is that there may be livelocks in extreme cases.  For
better scalability, one gen_pool allocator can be used for each CPU.

The lockless operation only works if there is enough memory available.
If new memory is added to the pool a lock has to be still taken.  So
any user relying on locklessness has to ensure that sufficient memory
is preallocated.

The basic atomic operation of this allocator is cmpxchg on long.  On
architectures that don't have NMI-safe cmpxchg implementation, the
allocator can NOT be used in NMI handler.  So code uses the allocator
in NMI handler should depend on CONFIG_ARCH_HAVE_NMI_SAFE_CMPXCHG.

Signed-off-by: Huang Ying <ying.huang@intel.com>
Reviewed-by: Andi Kleen <ak@linux.intel.com>
Reviewed-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
---
 include/linux/bitmap.h   |    1 
 include/linux/genalloc.h |   34 ++++-
 lib/bitmap.c             |    2 
 lib/genalloc.c           |  300 ++++++++++++++++++++++++++++++++++++++---------
 4 files changed, 272 insertions(+), 65 deletions(-)

--- a/include/linux/bitmap.h
+++ b/include/linux/bitmap.h
@@ -145,6 +145,7 @@ extern void bitmap_release_region(unsign
 extern int bitmap_allocate_region(unsigned long *bitmap, int pos, int order);
 extern void bitmap_copy_le(void *dst, const unsigned long *src, int nbits);
 
+#define BITMAP_FIRST_WORD_MASK(start) (~0UL << ((start) % BITS_PER_LONG))
 #define BITMAP_LAST_WORD_MASK(nbits)					\
 (									\
 	((nbits) % BITS_PER_LONG) ?					\
--- a/include/linux/genalloc.h
+++ b/include/linux/genalloc.h
@@ -1,8 +1,26 @@
 /*
- * Basic general purpose allocator for managing special purpose memory
- * not managed by the regular kmalloc/kfree interface.
- * Uses for this includes on-device special memory, uncached memory
- * etc.
+ * Basic general purpose allocator for managing special purpose
+ * memory, for example, memory that is not managed by the regular
+ * kmalloc/kfree interface.  Uses for this includes on-device special
+ * memory, uncached memory etc.
+ *
+ * It is safe to use the allocator in NMI handlers and other special
+ * unblockable contexts that could otherwise deadlock on locks.  This
+ * is implemented by using atomic operations and retries on any
+ * conflicts.  The disadvantage is that there may be livelocks in
+ * extreme cases.  For better scalability, one allocator can be used
+ * for each CPU.
+ *
+ * The lockless operation only works if there is enough memory
+ * available.  If new memory is added to the pool a lock has to be
+ * still taken.  So any user relying on locklessness has to ensure
+ * that sufficient memory is preallocated.
+ *
+ * The basic atomic operation of this allocator is cmpxchg on long.
+ * On architectures that don't have NMI-safe cmpxchg implementation,
+ * the allocator can NOT be used in NMI handler.  So code uses the
+ * allocator in NMI handler should depend on
+ * CONFIG_ARCH_HAVE_NMI_SAFE_CMPXCHG.
  *
  * This source code is licensed under the GNU General Public License,
  * Version 2.  See the file COPYING for more details.
@@ -15,7 +33,7 @@
  *  General purpose special memory pool descriptor.
  */
 struct gen_pool {
-	rwlock_t lock;
+	spinlock_t lock;
 	struct list_head chunks;	/* list of chunks in this pool */
 	int min_alloc_order;		/* minimum allocation order */
 };
@@ -24,8 +42,8 @@ struct gen_pool {
  *  General purpose special memory pool chunk descriptor.
  */
 struct gen_pool_chunk {
-	spinlock_t lock;
 	struct list_head next_chunk;	/* next chunk in pool */
+	atomic_t avail;
 	phys_addr_t phys_addr;		/* physical starting address of memory chunk */
 	unsigned long start_addr;	/* starting address of memory chunk */
 	unsigned long end_addr;		/* ending address of memory chunk */
@@ -56,4 +74,8 @@ static inline int gen_pool_add(struct ge
 extern void gen_pool_destroy(struct gen_pool *);
 extern unsigned long gen_pool_alloc(struct gen_pool *, size_t);
 extern void gen_pool_free(struct gen_pool *, unsigned long, size_t);
+extern void gen_pool_for_each_chunk(struct gen_pool *,
+	void (*)(struct gen_pool *, struct gen_pool_chunk *, void *), void *);
+extern size_t gen_pool_avail(struct gen_pool *);
+extern size_t gen_pool_size(struct gen_pool *);
 #endif /* __GENALLOC_H__ */
--- a/lib/bitmap.c
+++ b/lib/bitmap.c
@@ -271,8 +271,6 @@ int __bitmap_weight(const unsigned long
 }
 EXPORT_SYMBOL(__bitmap_weight);
 
-#define BITMAP_FIRST_WORD_MASK(start) (~0UL << ((start) % BITS_PER_LONG))
-
 void bitmap_set(unsigned long *map, int start, int nr)
 {
 	unsigned long *p = map + BIT_WORD(start);
--- a/lib/genalloc.c
+++ b/lib/genalloc.c
@@ -1,8 +1,26 @@
 /*
- * Basic general purpose allocator for managing special purpose memory
- * not managed by the regular kmalloc/kfree interface.
- * Uses for this includes on-device special memory, uncached memory
- * etc.
+ * Basic general purpose allocator for managing special purpose
+ * memory, for example, memory that is not managed by the regular
+ * kmalloc/kfree interface.  Uses for this includes on-device special
+ * memory, uncached memory etc.
+ *
+ * It is safe to use the allocator in NMI handlers and other special
+ * unblockable contexts that could otherwise deadlock on locks.  This
+ * is implemented by using atomic operations and retries on any
+ * conflicts.  The disadvantage is that there may be livelocks in
+ * extreme cases.  For better scalability, one allocator can be used
+ * for each CPU.
+ *
+ * The lockless operation only works if there is enough memory
+ * available.  If new memory is added to the pool a lock has to be
+ * still taken.  So any user relying on locklessness has to ensure
+ * that sufficient memory is preallocated.
+ *
+ * The basic atomic operation of this allocator is cmpxchg on long.
+ * On architectures that don't have NMI-safe cmpxchg implementation,
+ * the allocator can NOT be used in NMI handler.  So code uses the
+ * allocator in NMI handler should depend on
+ * CONFIG_ARCH_HAVE_NMI_SAFE_CMPXCHG.
  *
  * Copyright 2005 (C) Jes Sorensen <jes@trained-monkey.org>
  *
@@ -13,8 +31,109 @@
 #include <linux/slab.h>
 #include <linux/module.h>
 #include <linux/bitmap.h>
+#include <linux/rculist.h>
+#include <linux/interrupt.h>
 #include <linux/genalloc.h>
 
+static int set_bits_ll(unsigned long *addr, unsigned long mask_to_set)
+{
+	unsigned long val, nval;
+
+	nval = *addr;
+	do {
+		val = nval;
+		if (val & mask_to_set)
+			return -EBUSY;
+		cpu_relax();
+	} while ((nval = cmpxchg(addr, val, val | mask_to_set)) != val);
+
+	return 0;
+}
+
+static int clear_bits_ll(unsigned long *addr, unsigned long mask_to_clear)
+{
+	unsigned long val, nval;
+
+	nval = *addr;
+	do {
+		val = nval;
+		if ((val & mask_to_clear) != mask_to_clear)
+			return -EBUSY;
+		cpu_relax();
+	} while ((nval = cmpxchg(addr, val, val & ~mask_to_clear)) != val);
+
+	return 0;
+}
+
+/*
+ * bitmap_set_ll - set the specified number of bits at the specified position
+ * @map: pointer to a bitmap
+ * @start: a bit position in @map
+ * @nr: number of bits to set
+ *
+ * Set @nr bits start from @start in @map lock-lessly. Several users
+ * can set/clear the same bitmap simultaneously without lock. If two
+ * users set the same bit, one user will return remain bits, otherwise
+ * return 0.
+ */
+static int bitmap_set_ll(unsigned long *map, int start, int nr)
+{
+	unsigned long *p = map + BIT_WORD(start);
+	const int size = start + nr;
+	int bits_to_set = BITS_PER_LONG - (start % BITS_PER_LONG);
+	unsigned long mask_to_set = BITMAP_FIRST_WORD_MASK(start);
+
+	while (nr - bits_to_set >= 0) {
+		if (set_bits_ll(p, mask_to_set))
+			return nr;
+		nr -= bits_to_set;
+		bits_to_set = BITS_PER_LONG;
+		mask_to_set = ~0UL;
+		p++;
+	}
+	if (nr) {
+		mask_to_set &= BITMAP_LAST_WORD_MASK(size);
+		if (set_bits_ll(p, mask_to_set))
+			return nr;
+	}
+
+	return 0;
+}
+
+/*
+ * bitmap_clear_ll - clear the specified number of bits at the specified position
+ * @map: pointer to a bitmap
+ * @start: a bit position in @map
+ * @nr: number of bits to set
+ *
+ * Clear @nr bits start from @start in @map lock-lessly. Several users
+ * can set/clear the same bitmap simultaneously without lock. If two
+ * users clear the same bit, one user will return remain bits,
+ * otherwise return 0.
+ */
+static int bitmap_clear_ll(unsigned long *map, int start, int nr)
+{
+	unsigned long *p = map + BIT_WORD(start);
+	const int size = start + nr;
+	int bits_to_clear = BITS_PER_LONG - (start % BITS_PER_LONG);
+	unsigned long mask_to_clear = BITMAP_FIRST_WORD_MASK(start);
+
+	while (nr - bits_to_clear >= 0) {
+		if (clear_bits_ll(p, mask_to_clear))
+			return nr;
+		nr -= bits_to_clear;
+		bits_to_clear = BITS_PER_LONG;
+		mask_to_clear = ~0UL;
+		p++;
+	}
+	if (nr) {
+		mask_to_clear &= BITMAP_LAST_WORD_MASK(size);
+		if (clear_bits_ll(p, mask_to_clear))
+			return nr;
+	}
+
+	return 0;
+}
 
 /**
  * gen_pool_create - create a new special memory pool
@@ -30,7 +149,7 @@ struct gen_pool *gen_pool_create(int min
 
 	pool = kmalloc_node(sizeof(struct gen_pool), GFP_KERNEL, nid);
 	if (pool != NULL) {
-		rwlock_init(&pool->lock);
+		spin_lock_init(&pool->lock);
 		INIT_LIST_HEAD(&pool->chunks);
 		pool->min_alloc_order = min_alloc_order;
 	}
@@ -63,14 +182,14 @@ int gen_pool_add_virt(struct gen_pool *p
 	if (unlikely(chunk == NULL))
 		return -ENOMEM;
 
-	spin_lock_init(&chunk->lock);
 	chunk->phys_addr = phys;
 	chunk->start_addr = virt;
 	chunk->end_addr = virt + size;
+	atomic_set(&chunk->avail, size);
 
-	write_lock(&pool->lock);
-	list_add(&chunk->next_chunk, &pool->chunks);
-	write_unlock(&pool->lock);
+	spin_lock(&pool->lock);
+	list_add_rcu(&chunk->next_chunk, &pool->chunks);
+	spin_unlock(&pool->lock);
 
 	return 0;
 }
@@ -85,19 +204,19 @@ EXPORT_SYMBOL(gen_pool_add_virt);
  */
 phys_addr_t gen_pool_virt_to_phys(struct gen_pool *pool, unsigned long addr)
 {
-	struct list_head *_chunk;
 	struct gen_pool_chunk *chunk;
+	phys_addr_t paddr = -1;
 
-	read_lock(&pool->lock);
-	list_for_each(_chunk, &pool->chunks) {
-		chunk = list_entry(_chunk, struct gen_pool_chunk, next_chunk);
-
-		if (addr >= chunk->start_addr && addr < chunk->end_addr)
-			return chunk->phys_addr + addr - chunk->start_addr;
+	rcu_read_lock();
+	list_for_each_entry_rcu(chunk, &pool->chunks, next_chunk) {
+		if (addr >= chunk->start_addr && addr < chunk->end_addr) {
+			paddr = chunk->phys_addr + (addr - chunk->start_addr);
+			break;
+		}
 	}
-	read_unlock(&pool->lock);
+	rcu_read_unlock();
 
-	return -1;
+	return paddr;
 }
 EXPORT_SYMBOL(gen_pool_virt_to_phys);
 
@@ -115,7 +234,6 @@ void gen_pool_destroy(struct gen_pool *p
 	int order = pool->min_alloc_order;
 	int bit, end_bit;
 
-
 	list_for_each_safe(_chunk, _next_chunk, &pool->chunks) {
 		chunk = list_entry(_chunk, struct gen_pool_chunk, next_chunk);
 		list_del(&chunk->next_chunk);
@@ -137,44 +255,50 @@ EXPORT_SYMBOL(gen_pool_destroy);
  * @size: number of bytes to allocate from the pool
  *
  * Allocate the requested number of bytes from the specified pool.
- * Uses a first-fit algorithm.
+ * Uses a first-fit algorithm. Can not be used in NMI handler on
+ * architectures without NMI-safe cmpxchg implementation.
  */
 unsigned long gen_pool_alloc(struct gen_pool *pool, size_t size)
 {
-	struct list_head *_chunk;
 	struct gen_pool_chunk *chunk;
-	unsigned long addr, flags;
+	unsigned long addr = 0;
 	int order = pool->min_alloc_order;
-	int nbits, start_bit, end_bit;
+	int nbits, start_bit = 0, end_bit, remain;
+
+#ifndef CONFIG_ARCH_HAVE_NMI_SAFE_CMPXCHG
+	BUG_ON(in_nmi());
+#endif
 
 	if (size == 0)
 		return 0;
 
 	nbits = (size + (1UL << order) - 1) >> order;
-
-	read_lock(&pool->lock);
-	list_for_each(_chunk, &pool->chunks) {
-		chunk = list_entry(_chunk, struct gen_pool_chunk, next_chunk);
+	rcu_read_lock();
+	list_for_each_entry_rcu(chunk, &pool->chunks, next_chunk) {
+		if (size > atomic_read(&chunk->avail))
+			continue;
 
 		end_bit = (chunk->end_addr - chunk->start_addr) >> order;
-
-		spin_lock_irqsave(&chunk->lock, flags);
-		start_bit = bitmap_find_next_zero_area(chunk->bits, end_bit, 0,
-						nbits, 0);
-		if (start_bit >= end_bit) {
-			spin_unlock_irqrestore(&chunk->lock, flags);
+retry:
+		start_bit = bitmap_find_next_zero_area(chunk->bits, end_bit,
+						       start_bit, nbits, 0);
+		if (start_bit >= end_bit)
 			continue;
+		remain = bitmap_set_ll(chunk->bits, start_bit, nbits);
+		if (remain) {
+			remain = bitmap_clear_ll(chunk->bits, start_bit,
+						 nbits - remain);
+			BUG_ON(remain);
+			goto retry;
 		}
 
 		addr = chunk->start_addr + ((unsigned long)start_bit << order);
-
-		bitmap_set(chunk->bits, start_bit, nbits);
-		spin_unlock_irqrestore(&chunk->lock, flags);
-		read_unlock(&pool->lock);
-		return addr;
+		size = nbits << order;
+		atomic_sub(size, &chunk->avail);
+		break;
 	}
-	read_unlock(&pool->lock);
-	return 0;
+	rcu_read_unlock();
+	return addr;
 }
 EXPORT_SYMBOL(gen_pool_alloc);
 
@@ -184,33 +308,95 @@ EXPORT_SYMBOL(gen_pool_alloc);
  * @addr: starting address of memory to free back to pool
  * @size: size in bytes of memory to free
  *
- * Free previously allocated special memory back to the specified pool.
+ * Free previously allocated special memory back to the specified
+ * pool.  Can not be used in NMI handler on architectures without
+ * NMI-safe cmpxchg implementation.
  */
 void gen_pool_free(struct gen_pool *pool, unsigned long addr, size_t size)
 {
-	struct list_head *_chunk;
 	struct gen_pool_chunk *chunk;
-	unsigned long flags;
 	int order = pool->min_alloc_order;
-	int bit, nbits;
+	int start_bit, nbits, remain;
 
-	nbits = (size + (1UL << order) - 1) >> order;
-
-	read_lock(&pool->lock);
-	list_for_each(_chunk, &pool->chunks) {
-		chunk = list_entry(_chunk, struct gen_pool_chunk, next_chunk);
+#ifndef CONFIG_ARCH_HAVE_NMI_SAFE_CMPXCHG
+	BUG_ON(in_nmi());
+#endif
 
+	nbits = (size + (1UL << order) - 1) >> order;
+	rcu_read_lock();
+	list_for_each_entry_rcu(chunk, &pool->chunks, next_chunk) {
 		if (addr >= chunk->start_addr && addr < chunk->end_addr) {
 			BUG_ON(addr + size > chunk->end_addr);
-			spin_lock_irqsave(&chunk->lock, flags);
-			bit = (addr - chunk->start_addr) >> order;
-			while (nbits--)
-				__clear_bit(bit++, chunk->bits);
-			spin_unlock_irqrestore(&chunk->lock, flags);
-			break;
+			start_bit = (addr - chunk->start_addr) >> order;
+			remain = bitmap_clear_ll(chunk->bits, start_bit, nbits);
+			BUG_ON(remain);
+			size = nbits << order;
+			atomic_add(size, &chunk->avail);
+			rcu_read_unlock();
+			return;
 		}
 	}
-	BUG_ON(nbits > 0);
-	read_unlock(&pool->lock);
+	rcu_read_unlock();
+	BUG();
 }
 EXPORT_SYMBOL(gen_pool_free);
+
+/**
+ * gen_pool_for_each_chunk - call func for every chunk of generic memory pool
+ * @pool:	the generic memory pool
+ * @func:	func to call
+ * @data:	additional data used by @func
+ *
+ * Call @func for every chunk of generic memory pool.  The @func is
+ * called with rcu_read_lock held.
+ */
+void gen_pool_for_each_chunk(struct gen_pool *pool,
+	void (*func)(struct gen_pool *pool, struct gen_pool_chunk *chunk, void *data),
+	void *data)
+{
+	struct gen_pool_chunk *chunk;
+
+	rcu_read_lock();
+	list_for_each_entry_rcu(chunk, &(pool)->chunks, next_chunk)
+		func(pool, chunk, data);
+	rcu_read_unlock();
+}
+EXPORT_SYMBOL(gen_pool_for_each_chunk);
+
+/**
+ * gen_pool_avail - get available free space of the pool
+ * @pool: pool to get available free space
+ *
+ * Return available free space of the specified pool.
+ */
+size_t gen_pool_avail(struct gen_pool *pool)
+{
+	struct gen_pool_chunk *chunk;
+	size_t avail = 0;
+
+	rcu_read_lock();
+	list_for_each_entry_rcu(chunk, &pool->chunks, next_chunk)
+		avail += atomic_read(&chunk->avail);
+	rcu_read_unlock();
+	return avail;
+}
+EXPORT_SYMBOL_GPL(gen_pool_avail);
+
+/**
+ * gen_pool_size - get size in bytes of memory managed by the pool
+ * @pool: pool to get size
+ *
+ * Return size in bytes of memory managed by the pool.
+ */
+size_t gen_pool_size(struct gen_pool *pool)
+{
+	struct gen_pool_chunk *chunk;
+	size_t size = 0;
+
+	rcu_read_lock();
+	list_for_each_entry_rcu(chunk, &pool->chunks, next_chunk)
+		size += chunk->end_addr - chunk->start_addr;
+	rcu_read_unlock();
+	return size;
+}
+EXPORT_SYMBOL_GPL(gen_pool_size);

[PATCH 14/17] ACPI, APEI, GHES, printk support for recoverable error via NMI

[Huang Ying]

Some APEI GHES recoverable errors are reported via NMI, but printk is
not safe in NMI context.

To solve the issue, a lock-less memory allocator is used to allocate
memory in NMI handler, save the error record into the allocated
memory, put the error record into a lock-less list.  On the other
hand, an irq_work is used to delay the operation from NMI context to
IRQ context.  The irq_work IRQ handler will remove nodes from
lock-less list, printk the error record and do some further processing
include recovery operation, then free the memory.

Signed-off-by: Huang Ying <ying.huang@intel.com>
---
 drivers/acpi/apei/Kconfig |    2 
 drivers/acpi/apei/ghes.c  |  209 ++++++++++++++++++++++++++++++++++++++++++----
 2 files changed, 193 insertions(+), 18 deletions(-)

--- a/drivers/acpi/apei/Kconfig
+++ b/drivers/acpi/apei/Kconfig
@@ -13,6 +13,8 @@ config ACPI_APEI_GHES
 	bool "APEI Generic Hardware Error Source"
 	depends on ACPI_APEI && X86
 	select ACPI_HED
+	select LLIST
+	select GENERIC_ALLOCATOR
 	help
 	  Generic Hardware Error Source provides a way to report
 	  platform hardware errors (such as that from chipset). It
--- a/drivers/acpi/apei/ghes.c
+++ b/drivers/acpi/apei/ghes.c
@@ -12,7 +12,7 @@
  * For more information about Generic Hardware Error Source, please
  * refer to ACPI Specification version 4.0, section 17.3.2.6
  *
- * Copyright 2010 Intel Corp.
+ * Copyright 2010,2011 Intel Corp.
  *   Author: Huang Ying <ying.huang@intel.com>
  *
  * This program is free software; you can redistribute it and/or
@@ -42,6 +42,9 @@
 #include <linux/mutex.h>
 #include <linux/ratelimit.h>
 #include <linux/vmalloc.h>
+#include <linux/irq_work.h>
+#include <linux/llist.h>
+#include <linux/genalloc.h>
 #include <acpi/apei.h>
 #include <acpi/atomicio.h>
 #include <acpi/hed.h>
@@ -53,6 +56,15 @@
 #define GHES_PFX	"GHES: "
 
 #define GHES_ESTATUS_MAX_SIZE		65536
+#define GHES_ESOURCE_PREALLOC_MAX_SIZE	65536
+
+#define GHES_ESTATUS_POOL_MIN_ALLOC_ORDER 3
+
+#define GHES_ESTATUS_NODE_LEN(estatus_len)			\
+	(sizeof(struct ghes_estatus_node) + (estatus_len))
+#define GHES_ESTATUS_FROM_NODE(estatus_node)				\
+	((struct acpi_hest_generic_status *)				\
+	 ((struct ghes_estatus_node *)(estatus_node) + 1))
 
 /*
  * One struct ghes is created for each generic hardware error source.
@@ -77,6 +89,11 @@ struct ghes {
 	};
 };
 
+struct ghes_estatus_node {
+	struct llist_node llnode;
+	struct acpi_hest_generic *generic;
+};
+
 int ghes_disable;
 module_param_named(disable, ghes_disable, bool, 0);
 
@@ -124,6 +141,19 @@ static struct vm_struct *ghes_ioremap_ar
 static DEFINE_RAW_SPINLOCK(ghes_ioremap_lock_nmi);
 static DEFINE_SPINLOCK(ghes_ioremap_lock_irq);
 
+/*
+ * printk is not safe in NMI context.  So in NMI handler, we allocate
+ * required memory from lock-less memory allocator
+ * (ghes_estatus_pool), save estatus into it, put them into lock-less
+ * list (ghes_estatus_llist), then delay printk into IRQ context via
+ * irq_work (ghes_proc_irq_work).  ghes_estatus_size_request record
+ * required pool size by all NMI error source.
+ */
+static struct gen_pool *ghes_estatus_pool;
+static unsigned long ghes_estatus_pool_size_request;
+static struct llist_head ghes_estatus_llist;
+static struct irq_work ghes_proc_irq_work;
+
 static int ghes_ioremap_init(void)
 {
 	ghes_ioremap_area = __get_vm_area(PAGE_SIZE * GHES_IOREMAP_PAGES,
@@ -183,6 +213,55 @@ static void ghes_iounmap_irq(void __iome
 	__flush_tlb_one(vaddr);
 }
 
+static int ghes_estatus_pool_init(void)
+{
+	ghes_estatus_pool = gen_pool_create(GHES_ESTATUS_POOL_MIN_ALLOC_ORDER, -1);
+	if (!ghes_estatus_pool)
+		return -ENOMEM;
+	return 0;
+}
+
+static void ghes_estatus_pool_free_chunk_page(struct gen_pool *pool,
+					      struct gen_pool_chunk *chunk,
+					      void *data)
+{
+	free_page(chunk->start_addr);
+}
+
+static void ghes_estatus_pool_exit(void)
+{
+	gen_pool_for_each_chunk(ghes_estatus_pool,
+				ghes_estatus_pool_free_chunk_page, NULL);
+	gen_pool_destroy(ghes_estatus_pool);
+}
+
+static int ghes_estatus_pool_expand(unsigned long len)
+{
+	unsigned long i, pages, size, addr;
+	int ret;
+
+	ghes_estatus_pool_size_request += PAGE_ALIGN(len);
+	size = gen_pool_size(ghes_estatus_pool);
+	if (size >= ghes_estatus_pool_size_request)
+		return 0;
+	pages = (ghes_estatus_pool_size_request - size) / PAGE_SIZE;
+	for (i = 0; i < pages; i++) {
+		addr = __get_free_page(GFP_KERNEL);
+		if (!addr)
+			return -ENOMEM;
+		ret = gen_pool_add(ghes_estatus_pool, addr, PAGE_SIZE, -1);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
+static void ghes_estatus_pool_shrink(unsigned long len)
+{
+	ghes_estatus_pool_size_request -= PAGE_ALIGN(len);
+}
+
 static struct ghes *ghes_new(struct acpi_hest_generic *generic)
 {
 	struct ghes *ghes;
@@ -344,13 +423,13 @@ static void ghes_clear_estatus(struct gh
 	ghes->flags &= ~GHES_TO_CLEAR;
 }
 
-static void ghes_do_proc(struct ghes *ghes)
+static void ghes_do_proc(const struct acpi_hest_generic_status *estatus)
 {
 	int sev, processed = 0;
 	struct acpi_hest_generic_data *gdata;
 
-	sev = ghes_severity(ghes->estatus->error_severity);
-	apei_estatus_for_each_section(ghes->estatus, gdata) {
+	sev = ghes_severity(estatus->error_severity);
+	apei_estatus_for_each_section(estatus, gdata) {
 #ifdef CONFIG_X86_MCE
 		if (!uuid_le_cmp(*(uuid_le *)gdata->section_type,
 				 CPER_SEC_PLATFORM_MEM)) {
@@ -363,27 +442,37 @@ static void ghes_do_proc(struct ghes *gh
 	}
 }
 
-static void __ghes_print_estatus(const char *pfx, struct ghes *ghes)
+static void __ghes_print_estatus(const char *pfx,
+				 const struct acpi_hest_generic *generic,
+				 const struct acpi_hest_generic_status *estatus)
 {
 	if (pfx == NULL) {
-		if (ghes_severity(ghes->estatus->error_severity) <=
+		if (ghes_severity(estatus->error_severity) <=
 		    GHES_SEV_CORRECTED)
 			pfx = KERN_WARNING HW_ERR;
 		else
 			pfx = KERN_ERR HW_ERR;
 	}
 	printk("%s""Hardware error from APEI Generic Hardware Error Source: %d\n",
-	       pfx, ghes->generic->header.source_id);
-	apei_estatus_print(pfx, ghes->estatus);
+	       pfx, generic->header.source_id);
+	apei_estatus_print(pfx, estatus);
 }
 
-static void ghes_print_estatus(const char *pfx, struct ghes *ghes)
+static void ghes_print_estatus(const char *pfx,
+			       const struct acpi_hest_generic *generic,
+			       const struct acpi_hest_generic_status *estatus)
 {
 	/* Not more than 2 messages every 5 seconds */
-	static DEFINE_RATELIMIT_STATE(ratelimit, 5*HZ, 2);
+	static DEFINE_RATELIMIT_STATE(ratelimit_corrected, 5*HZ, 2);
+	static DEFINE_RATELIMIT_STATE(ratelimit_uncorrected, 5*HZ, 2);
+	struct ratelimit_state *ratelimit;
 
-	if (__ratelimit(&ratelimit))
-		__ghes_print_estatus(pfx, ghes);
+	if (ghes_severity(estatus->error_severity) <= GHES_SEV_CORRECTED)
+		ratelimit = &ratelimit_corrected;
+	else
+		ratelimit = &ratelimit_uncorrected;
+	if (__ratelimit(ratelimit))
+		__ghes_print_estatus(pfx, generic, estatus);
 }
 
 static int ghes_proc(struct ghes *ghes)
@@ -393,8 +482,8 @@ static int ghes_proc(struct ghes *ghes)
 	rc = ghes_read_estatus(ghes, 0);
 	if (rc)
 		goto out;
-	ghes_print_estatus(NULL, ghes);
-	ghes_do_proc(ghes);
+	ghes_print_estatus(NULL, ghes->generic, ghes->estatus);
+	ghes_do_proc(ghes->estatus);
 
 out:
 	ghes_clear_estatus(ghes);
@@ -453,6 +542,40 @@ static int ghes_notify_sci(struct notifi
 	return ret;
 }
 
+static void ghes_proc_in_irq(struct irq_work *irq_work)
+{
+	struct llist_node *llnode, *next, *tail = NULL;
+	struct ghes_estatus_node *estatus_node;
+	struct acpi_hest_generic_status *estatus;
+	u32 len, node_len;
+
+	/*
+	 * Because the time order of estatus in list is reversed,
+	 * revert it back to proper order.
+	 */
+	llnode = llist_del_all(&ghes_estatus_llist);
+	while (llnode) {
+		next = llnode->next;
+		llnode->next = tail;
+		tail = llnode;
+		llnode = next;
+	}
+	llnode = tail;
+	while (llnode) {
+		next = llnode->next;
+		estatus_node = llist_entry(llnode, struct ghes_estatus_node,
+					   llnode);
+		estatus = GHES_ESTATUS_FROM_NODE(estatus_node);
+		len = apei_estatus_len(estatus);
+		node_len = GHES_ESTATUS_NODE_LEN(len);
+		ghes_do_proc(estatus);
+		ghes_print_estatus(NULL, estatus_node->generic, estatus);
+		gen_pool_free(ghes_estatus_pool, (unsigned long)estatus_node,
+			      node_len);
+		llnode = next;
+	}
+}
+
 static int ghes_notify_nmi(struct notifier_block *this,
 				  unsigned long cmd, void *data)
 {
@@ -482,7 +605,8 @@ static int ghes_notify_nmi(struct notifi
 
 	if (sev_global >= GHES_SEV_PANIC) {
 		oops_begin();
-		__ghes_print_estatus(KERN_EMERG HW_ERR, ghes_global);
+		__ghes_print_estatus(KERN_EMERG HW_ERR, ghes_global->generic,
+				     ghes_global->estatus);
 		/* reboot to log the error! */
 		if (panic_timeout == 0)
 			panic_timeout = ghes_panic_timeout;
@@ -490,12 +614,31 @@ static int ghes_notify_nmi(struct notifi
 	}
 
 	list_for_each_entry_rcu(ghes, &ghes_nmi, list) {
+#ifdef CONFIG_ARCH_HAVE_NMI_SAFE_CMPXCHG
+		u32 len, node_len;
+		struct ghes_estatus_node *estatus_node;
+		struct acpi_hest_generic_status *estatus;
+#endif
 		if (!(ghes->flags & GHES_TO_CLEAR))
 			continue;
-		/* Do not print estatus because printk is not NMI safe */
-		ghes_do_proc(ghes);
+#ifdef CONFIG_ARCH_HAVE_NMI_SAFE_CMPXCHG
+		/* Save estatus for further processing in IRQ context */
+		len = apei_estatus_len(ghes->estatus);
+		node_len = GHES_ESTATUS_NODE_LEN(len);
+		estatus_node = (void *)gen_pool_alloc(ghes_estatus_pool,
+						      node_len);
+		if (estatus_node) {
+			estatus_node->generic = ghes->generic;
+			estatus = GHES_ESTATUS_FROM_NODE(estatus_node);
+			memcpy(estatus, ghes->estatus, len);
+			llist_add(&estatus_node->llnode, &ghes_estatus_llist);
+		}
+#endif
 		ghes_clear_estatus(ghes);
 	}
+#ifdef CONFIG_ARCH_HAVE_NMI_SAFE_CMPXCHG
+	irq_work_queue(&ghes_proc_irq_work);
+#endif
 
 out:
 	raw_spin_unlock(&ghes_nmi_lock);
@@ -510,10 +653,26 @@ static struct notifier_block ghes_notifi
 	.notifier_call = ghes_notify_nmi,
 };
 
+static unsigned long ghes_esource_prealloc_size(
+	const struct acpi_hest_generic *generic)
+{
+	unsigned long block_length, prealloc_records, prealloc_size;
+
+	block_length = min_t(unsigned long, generic->error_block_length,
+			     GHES_ESTATUS_MAX_SIZE);
+	prealloc_records = max_t(unsigned long,
+				 generic->records_to_preallocate, 1);
+	prealloc_size = min_t(unsigned long, block_length * prealloc_records,
+			      GHES_ESOURCE_PREALLOC_MAX_SIZE);
+
+	return prealloc_size;
+}
+
 static int __devinit ghes_probe(struct platform_device *ghes_dev)
 {
 	struct acpi_hest_generic *generic;
 	struct ghes *ghes = NULL;
+	unsigned long len;
 	int rc = -EINVAL;
 
 	generic = *(struct acpi_hest_generic **)ghes_dev->dev.platform_data;
@@ -579,6 +738,8 @@ static int __devinit ghes_probe(struct p
 		mutex_unlock(&ghes_list_mutex);
 		break;
 	case ACPI_HEST_NOTIFY_NMI:
+		len = ghes_esource_prealloc_size(generic);
+		ghes_estatus_pool_expand(len);
 		mutex_lock(&ghes_list_mutex);
 		if (list_empty(&ghes_nmi))
 			register_die_notifier(&ghes_notifier_nmi);
@@ -603,6 +764,7 @@ static int __devexit ghes_remove(struct
 {
 	struct ghes *ghes;
 	struct acpi_hest_generic *generic;
+	unsigned long len;
 
 	ghes = platform_get_drvdata(ghes_dev);
 	generic = ghes->generic;
@@ -633,6 +795,8 @@ static int __devexit ghes_remove(struct
 		 * freed after NMI handler finishes.
 		 */
 		synchronize_rcu();
+		len = ghes_esource_prealloc_size(generic);
+		ghes_estatus_pool_shrink(len);
 		break;
 	default:
 		BUG();
@@ -673,14 +837,20 @@ static int __init ghes_init(void)
 		return -EINVAL;
 	}
 
+	init_irq_work(&ghes_proc_irq_work, ghes_proc_in_irq);
+
 	rc = ghes_ioremap_init();
 	if (rc)
 		goto err;
 
-	rc = platform_driver_register(&ghes_platform_driver);
+	rc = ghes_estatus_pool_init();
 	if (rc)
 		goto err_ioremap_exit;
 
+	rc = platform_driver_register(&ghes_platform_driver);
+	if (rc)
+		goto err_pool_exit;
+
 	rc = apei_osc_setup();
 	if (rc == 0 && osc_sb_apei_support_acked)
 		pr_info(GHES_PFX "APEI firmware first mode is enabled by APEI bit and WHEA _OSC.\n");
@@ -692,6 +862,8 @@ static int __init ghes_init(void)
 		pr_info(GHES_PFX "Failed to enable APEI firmware first mode.\n");
 
 	return 0;
+err_pool_exit:
+	ghes_estatus_pool_exit();
 err_ioremap_exit:
 	ghes_ioremap_exit();
 err:
@@ -701,6 +873,7 @@ err:
 static void __exit ghes_exit(void)
 {
 	platform_driver_unregister(&ghes_platform_driver);
+	ghes_estatus_pool_exit();
 	ghes_ioremap_exit();
 }
 

[PATCH 15/17] ACPI, APEI, GHES, Error records content based throttle

[Huang Ying]

printk is used by GHES to report hardware errors.  Ratelimit is
enforced on the printk to avoid too many hardware error reports in
kernel log.  Because there may be thousands or even millions of
corrected hardware errors during system running.

Currently, a simple scheme is used.  That is, the total number of
hardware error reporting is ratelimited.  This may cause some issues
in practice.

For example, there are two kinds of hardware errors occurred in
system.  One is corrected memory error, because the fault memory
address is accessed frequently, there may be hundreds error report
per-second.  The other is corrected PCIe AER error, it will be
reported once per-second.  Because they share one ratelimit control
structure, it is highly possible that only memory error is reported.

To avoid the above issue, an error record content based throttle
algorithm is implemented in the patch.  Where after the first
successful reporting, all error records that are same are throttled for
some time, to let other kinds of error records have the opportunity to
be reported.

In above example, the memory errors will be throttled for some time,
after being printked.  Then the PCIe AER error will be printked
successfully.

Signed-off-by: Huang Ying <ying.huang@intel.com>
---
 drivers/acpi/apei/ghes.c |  184 +++++++++++++++++++++++++++++++++++++++++++++--
 1 file changed, 177 insertions(+), 7 deletions(-)

--- a/drivers/acpi/apei/ghes.c
+++ b/drivers/acpi/apei/ghes.c
@@ -60,6 +60,21 @@
 
 #define GHES_ESTATUS_POOL_MIN_ALLOC_ORDER 3
 
+/* This is just an estimation for memory pool allocation */
+#define GHES_ESTATUS_CACHE_AVG_SIZE	512
+
+#define GHES_ESTATUS_CACHES_SIZE	4
+
+#define GHES_ESTATUS_IN_CACHE_MAX_NSEC	(10 * NSEC_PER_SEC)
+/* Prevent too many caches are allocated because of RCU */
+#define GHES_ESTATUS_CACHE_ALLOCED_MAX	(GHES_ESTATUS_CACHES_SIZE * 3 / 2)
+
+#define GHES_ESTATUS_CACHE_LEN(estatus_len)			\
+	(sizeof(struct ghes_estatus_cache) + (estatus_len))
+#define GHES_ESTATUS_FROM_CACHE(estatus_cache)			\
+	((struct acpi_hest_generic_status *)			\
+	 ((struct ghes_estatus_cache *)(estatus_cache) + 1))
+
 #define GHES_ESTATUS_NODE_LEN(estatus_len)			\
 	(sizeof(struct ghes_estatus_node) + (estatus_len))
 #define GHES_ESTATUS_FROM_NODE(estatus_node)				\
@@ -94,6 +109,14 @@ struct ghes_estatus_node {
 	struct acpi_hest_generic *generic;
 };
 
+struct ghes_estatus_cache {
+	u32 estatus_len;
+	atomic_t count;
+	struct acpi_hest_generic *generic;
+	unsigned long long time_in;
+	struct rcu_head rcu;
+};
+
 int ghes_disable;
 module_param_named(disable, ghes_disable, bool, 0);
 
@@ -154,6 +177,9 @@ static unsigned long ghes_estatus_pool_s
 static struct llist_head ghes_estatus_llist;
 static struct irq_work ghes_proc_irq_work;
 
+struct ghes_estatus_cache *ghes_estatus_caches[GHES_ESTATUS_CACHES_SIZE];
+static atomic_t ghes_estatus_cache_alloced;
+
 static int ghes_ioremap_init(void)
 {
 	ghes_ioremap_area = __get_vm_area(PAGE_SIZE * GHES_IOREMAP_PAGES,
@@ -458,9 +484,9 @@ static void __ghes_print_estatus(const c
 	apei_estatus_print(pfx, estatus);
 }
 
-static void ghes_print_estatus(const char *pfx,
-			       const struct acpi_hest_generic *generic,
-			       const struct acpi_hest_generic_status *estatus)
+static int ghes_print_estatus(const char *pfx,
+			      const struct acpi_hest_generic *generic,
+			      const struct acpi_hest_generic_status *estatus)
 {
 	/* Not more than 2 messages every 5 seconds */
 	static DEFINE_RATELIMIT_STATE(ratelimit_corrected, 5*HZ, 2);
@@ -471,8 +497,137 @@ static void ghes_print_estatus(const cha
 		ratelimit = &ratelimit_corrected;
 	else
 		ratelimit = &ratelimit_uncorrected;
-	if (__ratelimit(ratelimit))
+	if (__ratelimit(ratelimit)) {
 		__ghes_print_estatus(pfx, generic, estatus);
+		return 1;
+	}
+	return 0;
+}
+
+/*
+ * GHES error status reporting throttle, to report more kinds of
+ * errors, instead of just most frequently occurred errors.
+ */
+static int ghes_estatus_cached(struct acpi_hest_generic_status *estatus)
+{
+	u32 len;
+	int i, cached = 0;
+	unsigned long long now;
+	struct ghes_estatus_cache *cache;
+	struct acpi_hest_generic_status *cache_estatus;
+
+	len = apei_estatus_len(estatus);
+	rcu_read_lock();
+	for (i = 0; i < GHES_ESTATUS_CACHES_SIZE; i++) {
+		cache = rcu_dereference(ghes_estatus_caches[i]);
+		if (cache == NULL)
+			continue;
+		if (len != cache->estatus_len)
+			continue;
+		cache_estatus = GHES_ESTATUS_FROM_CACHE(cache);
+		if (memcmp(estatus, cache_estatus, len))
+			continue;
+		atomic_inc(&cache->count);
+		now = sched_clock();
+		if (now - cache->time_in < GHES_ESTATUS_IN_CACHE_MAX_NSEC)
+			cached = 1;
+		break;
+	}
+	rcu_read_unlock();
+	return cached;
+}
+
+static struct ghes_estatus_cache *ghes_estatus_cache_alloc(
+	struct acpi_hest_generic *generic,
+	struct acpi_hest_generic_status *estatus)
+{
+	int alloced;
+	u32 len, cache_len;
+	struct ghes_estatus_cache *cache;
+	struct acpi_hest_generic_status *cache_estatus;
+
+	alloced = atomic_add_return(1, &ghes_estatus_cache_alloced);
+	if (alloced > GHES_ESTATUS_CACHE_ALLOCED_MAX) {
+		atomic_dec(&ghes_estatus_cache_alloced);
+		return NULL;
+	}
+	len = apei_estatus_len(estatus);
+	cache_len = GHES_ESTATUS_CACHE_LEN(len);
+	cache = (void *)gen_pool_alloc(ghes_estatus_pool, cache_len);
+	if (!cache) {
+		atomic_dec(&ghes_estatus_cache_alloced);
+		return NULL;
+	}
+	cache_estatus = GHES_ESTATUS_FROM_CACHE(cache);
+	memcpy(cache_estatus, estatus, len);
+	cache->estatus_len = len;
+	atomic_set(&cache->count, 0);
+	cache->generic = generic;
+	cache->time_in = sched_clock();
+	return cache;
+}
+
+static void ghes_estatus_cache_free(struct ghes_estatus_cache *cache)
+{
+	u32 len;
+
+	len = apei_estatus_len(GHES_ESTATUS_FROM_CACHE(cache));
+	len = GHES_ESTATUS_CACHE_LEN(len);
+	gen_pool_free(ghes_estatus_pool, (unsigned long)cache, len);
+	atomic_dec(&ghes_estatus_cache_alloced);
+}
+
+static void ghes_estatus_cache_rcu_free(struct rcu_head *head)
+{
+	struct ghes_estatus_cache *cache;
+
+	cache = container_of(head, struct ghes_estatus_cache, rcu);
+	ghes_estatus_cache_free(cache);
+}
+
+static void ghes_estatus_cache_add(
+	struct acpi_hest_generic *generic,
+	struct acpi_hest_generic_status *estatus)
+{
+	int i, slot = -1, count;
+	unsigned long long now, duration, period, max_period = 0;
+	struct ghes_estatus_cache *cache, *slot_cache = NULL, *new_cache;
+
+	new_cache = ghes_estatus_cache_alloc(generic, estatus);
+	if (new_cache == NULL)
+		return;
+	rcu_read_lock();
+	now = sched_clock();
+	for (i = 0; i < GHES_ESTATUS_CACHES_SIZE; i++) {
+		cache = rcu_dereference(ghes_estatus_caches[i]);
+		if (cache == NULL) {
+			slot = i;
+			slot_cache = NULL;
+			break;
+		}
+		duration = now - cache->time_in;
+		if (duration >= GHES_ESTATUS_IN_CACHE_MAX_NSEC) {
+			slot = i;
+			slot_cache = cache;
+			break;
+		}
+		count = atomic_read(&cache->count);
+		period = duration / (count + 1);
+		if (period > max_period) {
+			max_period = period;
+			slot = i;
+			slot_cache = cache;
+		}
+	}
+	/* new_cache must be put into array after its contents are written */
+	smp_wmb();
+	if (slot != -1 && cmpxchg(ghes_estatus_caches + slot,
+				  slot_cache, new_cache) == slot_cache) {
+		if (slot_cache)
+			call_rcu(&slot_cache->rcu, ghes_estatus_cache_rcu_free);
+	} else
+		ghes_estatus_cache_free(new_cache);
+	rcu_read_unlock();
 }
 
 static int ghes_proc(struct ghes *ghes)
@@ -482,9 +637,11 @@ static int ghes_proc(struct ghes *ghes)
 	rc = ghes_read_estatus(ghes, 0);
 	if (rc)
 		goto out;
-	ghes_print_estatus(NULL, ghes->generic, ghes->estatus);
+	if (!ghes_estatus_cached(ghes->estatus)) {
+		if (ghes_print_estatus(NULL, ghes->generic, ghes->estatus))
+			ghes_estatus_cache_add(ghes->generic, ghes->estatus);
+	}
 	ghes_do_proc(ghes->estatus);
-
 out:
 	ghes_clear_estatus(ghes);
 	return 0;
@@ -546,6 +703,7 @@ static void ghes_proc_in_irq(struct irq_
 {
 	struct llist_node *llnode, *next, *tail = NULL;
 	struct ghes_estatus_node *estatus_node;
+	struct acpi_hest_generic *generic;
 	struct acpi_hest_generic_status *estatus;
 	u32 len, node_len;
 
@@ -569,7 +727,11 @@ static void ghes_proc_in_irq(struct irq_
 		len = apei_estatus_len(estatus);
 		node_len = GHES_ESTATUS_NODE_LEN(len);
 		ghes_do_proc(estatus);
-		ghes_print_estatus(NULL, estatus_node->generic, estatus);
+		if (!ghes_estatus_cached(estatus)) {
+			generic = estatus_node->generic;
+			if (ghes_print_estatus(NULL, generic, estatus))
+				ghes_estatus_cache_add(generic, estatus);
+		}
 		gen_pool_free(ghes_estatus_pool, (unsigned long)estatus_node,
 			      node_len);
 		llnode = next;
@@ -622,6 +784,8 @@ static int ghes_notify_nmi(struct notifi
 		if (!(ghes->flags & GHES_TO_CLEAR))
 			continue;
 #ifdef CONFIG_ARCH_HAVE_NMI_SAFE_CMPXCHG
+		if (ghes_estatus_cached(ghes->estatus))
+			goto next;
 		/* Save estatus for further processing in IRQ context */
 		len = apei_estatus_len(ghes->estatus);
 		node_len = GHES_ESTATUS_NODE_LEN(len);
@@ -633,6 +797,7 @@ static int ghes_notify_nmi(struct notifi
 			memcpy(estatus, ghes->estatus, len);
 			llist_add(&estatus_node->llnode, &ghes_estatus_llist);
 		}
+next:
 #endif
 		ghes_clear_estatus(ghes);
 	}
@@ -847,6 +1012,11 @@ static int __init ghes_init(void)
 	if (rc)
 		goto err_ioremap_exit;
 
+	rc = ghes_estatus_pool_expand(GHES_ESTATUS_CACHE_AVG_SIZE *
+				      GHES_ESTATUS_CACHE_ALLOCED_MAX);
+	if (rc)
+		goto err_pool_exit;
+
 	rc = platform_driver_register(&ghes_platform_driver);
 	if (rc)
 		goto err_pool_exit;

[PATCH 16/17] HWPoison: add memory_failure_queue()

[Huang Ying]

memory_failure() is the entry point for HWPoison memory error
recovery.  It must be called in process context.  But commonly
hardware memory errors are notified via MCE or NMI, so some delayed
execution mechanism must be used.  In MCE handler, a work queue + ring
buffer mechanism is used.

In addition to MCE, now APEI (ACPI Platform Error Interface) GHES
(Generic Hardware Error Source) can be used to report memory errors
too.  To add support to APEI GHES memory recovery, a mechanism similar
to that of MCE is implemented.  memory_failure_queue() is the new
entry point that can be called in IRQ context.  The next step is to
make MCE handler uses this interface too.

Signed-off-by: Huang Ying <ying.huang@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Wu Fengguang <fengguang.wu@intel.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
---
 include/linux/mm.h  |    1 
 mm/memory-failure.c |   92 ++++++++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 93 insertions(+)

--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -1633,6 +1633,7 @@ enum mf_flags {
 };
 extern void memory_failure(unsigned long pfn, int trapno);
 extern int __memory_failure(unsigned long pfn, int trapno, int flags);
+extern void memory_failure_queue(unsigned long pfn, int trapno, int flags);
 extern int unpoison_memory(unsigned long pfn);
 extern int sysctl_memory_failure_early_kill;
 extern int sysctl_memory_failure_recovery;
--- a/mm/memory-failure.c
+++ b/mm/memory-failure.c
@@ -53,6 +53,7 @@
 #include <linux/hugetlb.h>
 #include <linux/memory_hotplug.h>
 #include <linux/mm_inline.h>
+#include <linux/kfifo.h>
 #include "internal.h"
 
 int sysctl_memory_failure_early_kill __read_mostly = 0;
@@ -1178,6 +1179,97 @@ void memory_failure(unsigned long pfn, i
 	__memory_failure(pfn, trapno, 0);
 }
 
+#define MEMORY_FAILURE_FIFO_ORDER	4
+#define MEMORY_FAILURE_FIFO_SIZE	(1 << MEMORY_FAILURE_FIFO_ORDER)
+
+struct memory_failure_entry {
+	unsigned long pfn;
+	int trapno;
+	int flags;
+};
+
+struct memory_failure_cpu {
+	DECLARE_KFIFO(fifo, struct memory_failure_entry,
+		      MEMORY_FAILURE_FIFO_SIZE);
+	spinlock_t lock;
+	struct work_struct work;
+};
+
+static DEFINE_PER_CPU(struct memory_failure_cpu, memory_failure_cpu);
+
+/**
+ * memory_failure_queue - Schedule handling memory failure of a page.
+ * @pfn: Page Number of the corrupted page
+ * @trapno: Trap number reported in the signal to user space.
+ * @flags: Flags for memory failure handling
+ *
+ * This function is called by the low level hardware error handler
+ * when it detects hardware memory corruption of a page. It schedules
+ * the recovering of error page, including dropping pages, killing
+ * processes etc.
+ *
+ * The function is primarily of use for corruptions that
+ * happen outside the current execution context (e.g. when
+ * detected by a background scrubber)
+ *
+ * Can run in IRQ context.
+ */
+void memory_failure_queue(unsigned long pfn, int trapno, int flags)
+{
+	struct memory_failure_cpu *mf_cpu;
+	unsigned long proc_flags;
+	struct memory_failure_entry entry = {
+		.pfn =		pfn,
+		.trapno =	trapno,
+		.flags =	flags,
+	};
+
+	mf_cpu = &get_cpu_var(memory_failure_cpu);
+	spin_lock_irqsave(&mf_cpu->lock, proc_flags);
+	if (kfifo_put(&mf_cpu->fifo, &entry))
+		schedule_work_on(smp_processor_id(), &mf_cpu->work);
+	else
+		pr_err("Memory failure: buffer overflow when queuing memory failure at 0x%#lx\n",
+		       pfn);
+	spin_unlock_irqrestore(&mf_cpu->lock, proc_flags);
+	put_cpu_var(memory_failure_cpu);
+}
+EXPORT_SYMBOL_GPL(memory_failure_queue);
+
+static void memory_failure_work_func(struct work_struct *work)
+{
+	struct memory_failure_cpu *mf_cpu;
+	struct memory_failure_entry entry = { 0, };
+	unsigned long proc_flags;
+	int gotten;
+
+	mf_cpu = &__get_cpu_var(memory_failure_cpu);
+	for (;;) {
+		spin_lock_irqsave(&mf_cpu->lock, proc_flags);
+		gotten = kfifo_get(&mf_cpu->fifo, &entry);
+		spin_unlock_irqrestore(&mf_cpu->lock, proc_flags);
+		if (!gotten)
+			break;
+		__memory_failure(entry.pfn, entry.trapno, entry.flags);
+	}
+}
+
+static int __init memory_failure_init(void)
+{
+	struct memory_failure_cpu *mf_cpu;
+	int cpu;
+
+	for_each_possible_cpu(cpu) {
+		mf_cpu = &per_cpu(memory_failure_cpu, cpu);
+		spin_lock_init(&mf_cpu->lock);
+		INIT_KFIFO(mf_cpu->fifo);
+		INIT_WORK(&mf_cpu->work, memory_failure_work_func);
+	}
+
+	return 0;
+}
+core_initcall(memory_failure_init);
+
 /**
  * unpoison_memory - Unpoison a previously poisoned page
  * @pfn: Page number of the to be unpoisoned page

[PATCH 17/17] ACPI, APEI, GHES: Add hardware memory error recovery support

[Huang Ying]

memory_failure_queue() is called when recoverable memory errors are
notified by firmware to do the recovery work.

Signed-off-by: Huang Ying <ying.huang@intel.com>
---
 drivers/acpi/apei/Kconfig |    7 +++++++
 drivers/acpi/apei/ghes.c  |   24 +++++++++++++++++-------
 2 files changed, 24 insertions(+), 7 deletions(-)

--- a/drivers/acpi/apei/Kconfig
+++ b/drivers/acpi/apei/Kconfig
@@ -32,6 +32,13 @@ config ACPI_APEI_PCIEAER
 	  PCIe AER errors may be reported via APEI firmware first mode.
 	  Turn on this option to enable the corresponding support.
 
+config ACPI_APEI_MEMORY_FAILURE
+	bool "APEI memory error recovering support"
+	depends on ACPI_APEI && MEMORY_FAILURE
+	help
+	  Memory errors may be reported via APEI firmware first mode.
+	  Turn on this option to enable the memory recovering support.
+
 config ACPI_APEI_EINJ
 	tristate "APEI Error INJection (EINJ)"
 	depends on ACPI_APEI && DEBUG_FS
--- a/drivers/acpi/apei/ghes.c
+++ b/drivers/acpi/apei/ghes.c
@@ -466,20 +466,30 @@ static void ghes_clear_estatus(struct gh
 
 static void ghes_do_proc(const struct acpi_hest_generic_status *estatus)
 {
-	int sev, processed = 0;
+	int sev, sec_sev;
 	struct acpi_hest_generic_data *gdata;
 
 	sev = ghes_severity(estatus->error_severity);
 	apei_estatus_for_each_section(estatus, gdata) {
-#ifdef CONFIG_X86_MCE
+		sec_sev = ghes_severity(gdata->error_severity);
 		if (!uuid_le_cmp(*(uuid_le *)gdata->section_type,
 				 CPER_SEC_PLATFORM_MEM)) {
-			apei_mce_report_mem_error(
-				sev == GHES_SEV_CORRECTED,
-				(struct cper_sec_mem_err *)(gdata+1));
-			processed = 1;
-		}
+			struct cper_sec_mem_err *mem_err;
+			mem_err = (struct cper_sec_mem_err *)(gdata+1);
+#ifdef CONFIG_X86_MCE
+			apei_mce_report_mem_error(sev == GHES_SEV_CORRECTED,
+						  mem_err);
 #endif
+#ifdef CONFIG_ACPI_APEI_MEMORY_FAILURE
+			if (sev == GHES_SEV_RECOVERABLE &&
+			    sec_sev == GHES_SEV_RECOVERABLE &&
+			    mem_err->validation_bits & CPER_MEM_VALID_PHYSICAL_ADDRESS) {
+				unsigned long pfn;
+				pfn = mem_err->physical_addr >> PAGE_SHIFT;
+				memory_failure_queue(pfn, 0, 0);
+			}
+#endif
+		}
 	}
 }