summaryrefslogtreecommitdiffstats
path: root/drivers/acpi
diff options
context:
space:
mode:
authorHuang Ying <ying.huang@intel.com>2011-07-13 13:14:25 +0800
committerLen Brown <len.brown@intel.com>2011-08-03 11:15:57 -0400
commit67eb2e99076708cc790019a6a08ca3e0ae130a3a (patch)
treedc3863496a4b6c4e30450f1b94d3e1c87b858e7a /drivers/acpi
parent7f184275aa306046fe7edcbef3229754f0d97402 (diff)
downloadlinux-67eb2e99076708cc790019a6a08ca3e0ae130a3a.tar.bz2
ACPI, APEI, GHES, printk support for recoverable error via NMI
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> Signed-off-by: Len Brown <len.brown@intel.com>
Diffstat (limited to 'drivers/acpi')
-rw-r--r--drivers/acpi/apei/Kconfig2
-rw-r--r--drivers/acpi/apei/ghes.c209
2 files changed, 193 insertions, 18 deletions
diff --git a/drivers/acpi/apei/Kconfig b/drivers/acpi/apei/Kconfig
index 3f45dde17aec..35596eaaca17 100644
--- 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
diff --git a/drivers/acpi/apei/ghes.c b/drivers/acpi/apei/ghes.c
index b1390a61cde1..d1a40218e17e 100644
--- 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_area;
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 __iomem *vaddr_ptr)
__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 ghes *ghes)
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 *ghes)
}
}
-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 notifier_block *this,
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 notifier_block *this,
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 notifier_block *this,
}
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_notifier_nmi = {
.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 platform_device *ghes_dev)
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 platform_device *ghes_dev)
{
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 platform_device *ghes_dev)
* 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();
}