linux-pci.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Shiju Jose <shiju.jose@huawei.com>
To: <linux-acpi@vger.kernel.org>, <linux-pci@vger.kernel.org>,
	<linux-kernel@vger.kernel.org>, <rjw@rjwysocki.net>,
	<bp@alien8.de>, <james.morse@arm.com>, <helgaas@kernel.org>,
	<lenb@kernel.org>, <tony.luck@intel.com>,
	<dan.carpenter@oracle.com>, <gregkh@linuxfoundation.org>,
	<zhangliguang@linux.alibaba.com>, <tglx@linutronix.de>
Cc: <linuxarm@huawei.com>, <jonathan.cameron@huawei.com>,
	<tanxiaofei@huawei.com>, <yangyicong@hisilicon.com>,
	Shiju Jose <shiju.jose@huawei.com>
Subject: [RESEND PATCH v7 1/6] ACPI / APEI: Add support to queuing up the non-fatal HW errors and notify
Date: Tue, 21 Apr 2020 14:21:31 +0100	[thread overview]
Message-ID: <20200421132136.1595-2-shiju.jose@huawei.com> (raw)
In-Reply-To: <20200421132136.1595-1-shiju.jose@huawei.com>

Add queuing up the non-fatal HW errors to the work queue and interface to
notify the registered kernel drivers for the error recovery.
The interface enables drivers to register the callback functions to
receive the error notification, including the vendor-specific HW errors
and supports handling the non-fatal HW errors in the process context.

Suggested-by: Borislav Petkov <bp@alien8.de>
Signed-off-by: Shiju Jose <shiju.jose@huawei.com>
---
 drivers/acpi/apei/ghes.c | 127 +++++++++++++++++++++++++++++++++++++++
 include/acpi/ghes.h      |  28 +++++++++
 2 files changed, 155 insertions(+)

diff --git a/drivers/acpi/apei/ghes.c b/drivers/acpi/apei/ghes.c
index 24c9642e8fc7..5c0ab5422311 100644
--- a/drivers/acpi/apei/ghes.c
+++ b/drivers/acpi/apei/ghes.c
@@ -33,6 +33,7 @@
 #include <linux/irq_work.h>
 #include <linux/llist.h>
 #include <linux/genalloc.h>
+#include <linux/kfifo.h>
 #include <linux/pci.h>
 #include <linux/pfn.h>
 #include <linux/aer.h>
@@ -63,6 +64,11 @@
 #define GHES_ESTATUS_CACHES_SIZE	4
 
 #define GHES_ESTATUS_IN_CACHE_MAX_NSEC	10000000000ULL
+
+#define GHES_EVENT_RING_SIZE	256
+#define GHES_GDATA_POOL_MIN_ALLOC_ORDER	3
+#define GHES_GDATA_POOL_MIN_SIZE	65536
+
 /* Prevent too many caches are allocated because of RCU */
 #define GHES_ESTATUS_CACHE_ALLOCED_MAX	(GHES_ESTATUS_CACHES_SIZE * 3 / 2)
 
@@ -122,6 +128,19 @@ static DEFINE_MUTEX(ghes_list_mutex);
  */
 static DEFINE_SPINLOCK(ghes_notify_lock_irq);
 
+struct ghes_event_entry {
+	struct acpi_hest_generic_data *gdata;
+	int error_severity;
+};
+
+static DEFINE_KFIFO(ghes_event_ring, struct ghes_event_entry,
+		    GHES_EVENT_RING_SIZE);
+
+static DEFINE_SPINLOCK(ghes_event_ring_lock);
+
+static struct gen_pool *ghes_gdata_pool;
+static unsigned long ghes_gdata_pool_size_request;
+
 static struct gen_pool *ghes_estatus_pool;
 static unsigned long ghes_estatus_pool_size_request;
 
@@ -188,6 +207,40 @@ int ghes_estatus_pool_init(int num_ghes)
 	return -ENOMEM;
 }
 
+int ghes_gdata_pool_init(void)
+{
+	unsigned long addr, len;
+	int rc;
+
+	ghes_gdata_pool = gen_pool_create(GHES_GDATA_POOL_MIN_ALLOC_ORDER, -1);
+	if (!ghes_gdata_pool)
+		return -ENOMEM;
+
+	if (ghes_gdata_pool_size_request < GHES_GDATA_POOL_MIN_SIZE)
+		ghes_gdata_pool_size_request = GHES_GDATA_POOL_MIN_SIZE;
+
+	len = ghes_gdata_pool_size_request;
+	addr = (unsigned long)vmalloc(PAGE_ALIGN(len));
+	if (!addr)
+		goto err_pool_alloc;
+
+	vmalloc_sync_mappings();
+
+	rc = gen_pool_add(ghes_gdata_pool, addr, PAGE_ALIGN(len), -1);
+	if (rc)
+		goto err_pool_add;
+
+	return 0;
+
+err_pool_add:
+	vfree((void *)addr);
+
+err_pool_alloc:
+	gen_pool_destroy(ghes_gdata_pool);
+
+	return -ENOMEM;
+}
+
 static int map_gen_v2(struct ghes *ghes)
 {
 	return apei_map_generic_address(&ghes->generic_v2->read_ack_register);
@@ -247,6 +300,10 @@ static struct ghes *ghes_new(struct acpi_hest_generic *generic)
 		goto err_unmap_status_addr;
 	}
 
+	ghes_gdata_pool_size_request += generic->records_to_preallocate *
+					generic->max_sections_per_record *
+					generic->max_raw_data_length;
+
 	return ghes;
 
 err_unmap_status_addr:
@@ -490,6 +547,49 @@ static void ghes_handle_aer(struct acpi_hest_generic_data *gdata)
 #endif
 }
 
+static BLOCKING_NOTIFIER_HEAD(ghes_event_notify_list);
+
+/**
+ * ghes_register_event_notifier - register an event notifier
+ * for the non-fatal HW errors.
+ * @nb: pointer to the notifier_block structure of the event handler.
+ *
+ * return 0 : SUCCESS, non-zero : FAIL
+ */
+int ghes_register_event_notifier(struct notifier_block *nb)
+{
+	return blocking_notifier_chain_register(&ghes_event_notify_list, nb);
+}
+EXPORT_SYMBOL_GPL(ghes_register_event_notifier);
+
+/**
+ * ghes_unregister_event_notifier - unregister the previously
+ * registered event notifier.
+ * @nb: pointer to the notifier_block structure of the event handler.
+ */
+void ghes_unregister_event_notifier(struct notifier_block *nb)
+{
+	blocking_notifier_chain_unregister(&ghes_event_notify_list, nb);
+}
+EXPORT_SYMBOL_GPL(ghes_unregister_event_notifier);
+
+static void ghes_event_work_func(struct work_struct *work)
+{
+	struct ghes_event_entry entry;
+	u32 len;
+
+	while (kfifo_get(&ghes_event_ring, &entry)) {
+		blocking_notifier_call_chain(&ghes_event_notify_list,
+					     entry.error_severity,
+					     entry.gdata);
+		len = acpi_hest_get_size(entry.gdata) +
+				acpi_hest_get_error_length(entry.gdata);
+		gen_pool_free(ghes_gdata_pool, (unsigned long)entry.gdata, len);
+	}
+}
+
+static DECLARE_WORK(ghes_event_work, ghes_event_work_func);
+
 static void ghes_do_proc(struct ghes *ghes,
 			 const struct acpi_hest_generic_status *estatus)
 {
@@ -498,6 +598,8 @@ static void ghes_do_proc(struct ghes *ghes,
 	guid_t *sec_type;
 	const guid_t *fru_id = &guid_null;
 	char *fru_text = "";
+	struct ghes_event_entry event_entry;
+	u32 len;
 
 	sev = ghes_severity(estatus->error_severity);
 	apei_estatus_for_each_section(estatus, gdata) {
@@ -509,6 +611,25 @@ static void ghes_do_proc(struct ghes *ghes,
 		if (gdata->validation_bits & CPER_SEC_VALID_FRU_TEXT)
 			fru_text = gdata->fru_text;
 
+		len = acpi_hest_get_record_size(gdata);
+		event_entry.gdata = (void *)gen_pool_alloc(ghes_gdata_pool,
+							   len);
+		if (!event_entry.gdata) {
+			pr_warn(GHES_PFX "ghes gdata pool alloc fail\n");
+			break;
+		}
+
+		memcpy(event_entry.gdata, gdata, len);
+		event_entry.error_severity = sev;
+
+		if (kfifo_in_spinlocked(&ghes_event_ring, &event_entry, 1,
+					&ghes_event_ring_lock))
+			schedule_work(&ghes_event_work);
+		else {
+			pr_warn(GHES_PFX "ghes event queue full\n");
+			break;
+		}
+
 		if (guid_equal(sec_type, &CPER_SEC_PLATFORM_MEM)) {
 			struct cper_sec_mem_err *mem_err = acpi_hest_get_payload(gdata);
 
@@ -1346,6 +1467,12 @@ static int __init ghes_init(void)
 	else
 		pr_info(GHES_PFX "Failed to enable APEI firmware first mode.\n");
 
+	rc = ghes_gdata_pool_init();
+	if (rc) {
+		pr_warn(GHES_PFX "ghes_gdata_pool_init failed\n");
+		goto err;
+	}
+
 	return 0;
 err:
 	return rc;
diff --git a/include/acpi/ghes.h b/include/acpi/ghes.h
index e3f1cddb4ac8..a3dd82069069 100644
--- a/include/acpi/ghes.h
+++ b/include/acpi/ghes.h
@@ -50,6 +50,34 @@ enum {
 	GHES_SEV_PANIC = 0x3,
 };
 
+
+#ifdef CONFIG_ACPI_APEI_GHES
+/**
+ * ghes_register_event_notifier - register an event notifier
+ * for the non-fatal HW errors.
+ * @nb: pointer to the notifier_block structure of the event notifier.
+ *
+ * Return : 0 - SUCCESS, non-zero - FAIL.
+ */
+int ghes_register_event_notifier(struct notifier_block *nb);
+
+/**
+ * ghes_unregister_event_notifier - unregister the previously
+ * registered event notifier.
+ * @nb: pointer to the notifier_block structure of the event notifier.
+ */
+void ghes_unregister_event_notifier(struct notifier_block *nb);
+#else
+static inline int ghes_register_event_notifier(struct notifier_block *nb)
+{
+	return -ENODEV;
+}
+
+static inline void ghes_unregister_event_notifier(struct notifier_block *nb)
+{
+}
+#endif
+
 int ghes_estatus_pool_init(int num_ghes);
 
 /* From drivers/edac/ghes_edac.c */
-- 
2.17.1



  reply	other threads:[~2020-04-21 13:23 UTC|newest]

Thread overview: 59+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <Shiju Jose>
2020-01-15 11:01 ` [RFC PATCH 0/2] ACPI: APEI: Add support to notify the vendor specific HW errors Shiju Jose
2020-01-15 11:01   ` [RFC PATCH 1/2] " Shiju Jose
2020-01-15 11:01   ` [RFC PATCH 2/2] PCI:hip08:Add driver to handle HiSilicon hip08 PCIe controller's errors Shiju Jose
2020-01-15 14:13     ` Bjorn Helgaas
2020-01-17  9:40       ` Shiju Jose
2020-01-24 12:39 ` [PATCH v2 0/2] ACPI: APEI: Add support to notify the vendor specific HW errors Shiju Jose
2020-01-24 12:39   ` [PATCH v2 1/2] " Shiju Jose
2020-01-24 12:39   ` [PATCH v2 2/2] PCI: hip: Add handling of HiSilicon hip PCIe controller's errors Shiju Jose
2020-01-24 14:30     ` Bjorn Helgaas
2020-01-26 18:12     ` kbuild test robot
2020-01-26 18:12     ` [RFC PATCH] PCI: hip: hisi_pcie_sec_type can be static kbuild test robot
2020-02-03 16:51 ` [PATCH v3 0/2] ACPI: APEI: Add support to notify the vendor specific HW errors Shiju Jose
2020-02-03 16:51   ` [PATCH v3 1/2] " Shiju Jose
2020-02-03 16:51   ` [PATCH v3 2/2] PCI: HIP: Add handling of HiSilicon HIP PCIe controller's errors Shiju Jose
2020-02-04 14:31     ` Dan Carpenter
2020-02-07 10:31 ` [PATCH v4 0/2] ACPI: APEI: Add support to notify the vendor specific HW errors Shiju Jose
2020-02-07 10:31   ` [PATCH v4 1/2] " Shiju Jose
2020-03-11 17:29     ` James Morse
2020-03-12 12:10       ` Shiju Jose
2020-03-13 15:17         ` James Morse
2020-03-13 17:08           ` Shiju Jose
2020-02-07 10:31   ` [PATCH v4 2/2] PCI: HIP: Add handling of HiSilicon HIP PCIe controller errors Shiju Jose
2020-03-09  9:23   ` [PATCH v4 0/2] ACPI: APEI: Add support to notify the vendor specific HW errors Shiju Jose
2020-03-11 17:27     ` James Morse
2020-03-25 16:42 ` [PATCH v6 0/2] ACPI / " Shiju Jose
2020-03-25 16:42   ` [PATCH v6 1/2] " Shiju Jose
2020-03-27 18:22     ` Borislav Petkov
2020-03-30 10:14       ` Shiju Jose
2020-03-30 10:33         ` Borislav Petkov
2020-03-30 11:55           ` Shiju Jose
2020-03-30 13:42             ` Borislav Petkov
2020-03-30 15:44               ` Shiju Jose
2020-03-31  9:09                 ` Borislav Petkov
2020-04-08  9:20                   ` Shiju Jose
2020-04-08 10:03       ` James Morse
2020-04-21 13:18         ` Shiju Jose
2020-05-11 11:20         ` Shiju Jose
2020-03-25 16:42   ` [PATCH v6 2/2] PCI: hip: Add handling of HiSilicon HIP PCIe controller errors Shiju Jose
2020-03-27 15:07   ` [PATCH v6 0/2] ACPI / APEI: Add support to notify the vendor specific HW errors Bjorn Helgaas
2020-04-07 12:00 ` [v7 PATCH 0/6] ACPI / APEI: Add support to notify non-fatal " Shiju Jose
2020-04-07 12:00   ` [v7 PATCH 1/6] ACPI / APEI: Add support to queuing up the non-fatal HW errors and notify Shiju Jose
2020-04-08 19:41     ` kbuild test robot
2020-04-08 19:41     ` [RFC PATCH] ACPI / APEI: ghes_gdata_pool_init() can be static kbuild test robot
2020-04-07 12:00   ` [v7 PATCH 2/6] ACPI / APEI: Add callback for memory errors to the GHES notifier Shiju Jose
2020-04-07 12:00   ` [v7 PATCH 3/6] ACPI / APEI: Add callback for AER " Shiju Jose
2020-04-07 12:00   ` [v7 PATCH 4/6] ACPI / APEI: Add callback for ARM HW errors " Shiju Jose
2020-04-07 12:00   ` [v7 PATCH 5/6] ACPI / APEI: Add callback for non-standard " Shiju Jose
2020-04-07 12:00   ` [v7 PATCH 6/6] PCI: hip: Add handling of HiSilicon HIP PCIe controller errors Shiju Jose
2020-04-21 13:21 ` [RESEND PATCH v7 0/6] ACPI / APEI: Add support to notify non-fatal HW errors Shiju Jose
2020-04-21 13:21   ` Shiju Jose [this message]
2020-04-21 14:12     ` [RESEND PATCH v7 1/6] ACPI / APEI: Add support to queuing up the non-fatal HW errors and notify Dan Carpenter
2020-04-21 13:21   ` [RESEND PATCH v7 2/6] ACPI / APEI: Add callback for memory errors to the GHES notifier Shiju Jose
2020-04-21 13:21   ` [RESEND PATCH v7 3/6] ACPI / APEI: Add callback for AER " Shiju Jose
2020-04-21 13:21   ` [RESEND PATCH v7 4/6] ACPI / APEI: Add callback for ARM HW errors " Shiju Jose
2020-04-21 14:14     ` Dan Carpenter
2020-04-21 15:18       ` Shiju Jose
2020-04-21 13:21   ` [RESEND PATCH v7 5/6] ACPI / APEI: Add callback for non-standard " Shiju Jose
2020-04-21 13:21   ` [RESEND PATCH v7 6/6] PCI: hip: Add handling of HiSilicon HIP PCIe controller errors Shiju Jose
2020-04-21 14:20     ` Dan Carpenter

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20200421132136.1595-2-shiju.jose@huawei.com \
    --to=shiju.jose@huawei.com \
    --cc=bp@alien8.de \
    --cc=dan.carpenter@oracle.com \
    --cc=gregkh@linuxfoundation.org \
    --cc=helgaas@kernel.org \
    --cc=james.morse@arm.com \
    --cc=jonathan.cameron@huawei.com \
    --cc=lenb@kernel.org \
    --cc=linux-acpi@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-pci@vger.kernel.org \
    --cc=linuxarm@huawei.com \
    --cc=rjw@rjwysocki.net \
    --cc=tanxiaofei@huawei.com \
    --cc=tglx@linutronix.de \
    --cc=tony.luck@intel.com \
    --cc=yangyicong@hisilicon.com \
    --cc=zhangliguang@linux.alibaba.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).