Re: [PATCH v4 7/7] cxl/memdev: Register for and process CPER events
From: Smita Koralahalli
Date: Mon Dec 18 2023 - 13:18:11 EST
On 12/15/2023 3:26 PM, Ira Weiny wrote:
If the firmware has configured CXL event support to be firmware first
the OS can process those events through CPER records. The CXL layer has
unique DPA to HPA knowledge and standard event trace parsing in place.
CPER records contain Bus, Device, Function information which can be used
to identify the PCI device which is sending the event.
Change pci driver registration to include registration for a CXL CPER
notifier to process the events through the trace subsystem.
Define and use scoped based management to simplify the handling of the
pci device object.
Cc: Bjorn Helgaas <bhelgaas@xxxxxxxxxx>
Signed-off-by: Ira Weiny <ira.weiny@xxxxxxxxx>
---
[snip]
+ switch (event_type) {
+ case CXL_CPER_EVENT_GEN_MEDIA:
+ trace_cxl_general_media(cxlmd, type, &gen_media_event_uuid,
+ &event->gen_media);
+ break;
+ case CXL_CPER_EVENT_DRAM:
+ trace_cxl_dram(cxlmd, type, &dram_event_uuid, &event->dram);
+ break;
+ case CXL_CPER_EVENT_MEM_MODULE:
+ trace_cxl_memory_module(cxlmd, type, &mem_mod_event_uuid,
+ &event->mem_module);
+ break;
+ }
+}
Is default case needed here?
+EXPORT_SYMBOL_NS_GPL(cxl_event_trace_record, CXL);
+
+static void __cxl_event_trace_record(const struct cxl_memdev *cxlmd,
+ enum cxl_event_log_type type,
+ struct cxl_event_record_raw *record)
{
union cxl_event *evt = &record->event;
uuid_t *id = &record->id;
@@ -965,8 +986,8 @@ static void cxl_mem_get_records_log(struct cxl_memdev_state *mds,
break;
for (i = 0; i < nr_rec; i++)
- cxl_event_trace_record(cxlmd, type,
- &payload->records[i]);
+ __cxl_event_trace_record(cxlmd, type,
+ &payload->records[i]);
if (payload->flags & CXL_GET_EVENT_FLAG_OVERFLOW)
trace_cxl_overflow(cxlmd, type, payload);
diff --git a/drivers/cxl/cxlmem.h b/drivers/cxl/cxlmem.h
index e5d770e26e02..e7e9508fecac 100644
--- a/drivers/cxl/cxlmem.h
+++ b/drivers/cxl/cxlmem.h
@@ -802,6 +802,10 @@ void set_exclusive_cxl_commands(struct cxl_memdev_state *mds,
void clear_exclusive_cxl_commands(struct cxl_memdev_state *mds,
unsigned long *cmds);
void cxl_mem_get_event_records(struct cxl_memdev_state *mds, u32 status);
+void cxl_event_trace_record(const struct cxl_memdev *cxlmd,
+ enum cxl_event_log_type type,
+ enum cxl_event_type event_type,
+ union cxl_event *event);
int cxl_set_timestamp(struct cxl_memdev_state *mds);
int cxl_poison_state_init(struct cxl_memdev_state *mds);
int cxl_mem_get_poison(struct cxl_memdev *cxlmd, u64 offset, u64 len,
diff --git a/drivers/cxl/pci.c b/drivers/cxl/pci.c
index 0155fb66b580..638275569d63 100644
--- a/drivers/cxl/pci.c
+++ b/drivers/cxl/pci.c
@@ -1,5 +1,6 @@
// SPDX-License-Identifier: GPL-2.0-only
/* Copyright(c) 2020 Intel Corporation. All rights reserved. */
+#include <asm-generic/unaligned.h>
#include <linux/io-64-nonatomic-lo-hi.h>
#include <linux/moduleparam.h>
#include <linux/module.h>
@@ -969,6 +970,58 @@ static struct pci_driver cxl_pci_driver = {
},
};
+#define CXL_EVENT_HDR_FLAGS_REC_SEVERITY GENMASK(1, 0)
+static void cxl_cper_event_call(enum cxl_event_type ev_type,
+ struct cxl_cper_event_rec *rec)
+{
+ struct cper_cxl_event_devid *device_id = &rec->hdr.device_id;
+ struct pci_dev *pdev __free(pci_dev_put) = NULL;
+ struct cxl_dev_state *cxlds = NULL;
+ enum cxl_event_log_type log_type;
+ unsigned int devfn;
+ u32 hdr_flags;
+
+ devfn = PCI_DEVFN(device_id->device_num, device_id->func_num);
+ pdev = pci_get_domain_bus_and_slot(device_id->segment_num,
+ device_id->bus_num, devfn);
+ if (!pdev)
+ return;
+
+ guard(device)(&pdev->dev);
+ if (pdev->driver == &cxl_pci_driver)
+ cxlds = pci_get_drvdata(pdev);
+ if (!cxlds)
+ return;
+
+ /* Fabricate a log type */
+ hdr_flags = get_unaligned_le24(rec->event.generic.hdr.flags);
+ log_type = FIELD_GET(CXL_EVENT_HDR_FLAGS_REC_SEVERITY, hdr_flags);
+
+ cxl_event_trace_record(cxlds->cxlmd, log_type, ev_type, &rec->event);
Currently, when I run this, I see two trace events printed. One from
here, and another as a non_standard_event from ghes. I think both should
be unified?
I remember Dan pointing out to me this when I sent decoding for protocol
errors and its still pending on me for protocol errors.
Thanks,
Smita
+}
+
+static int __init cxl_pci_driver_init(void)
+{
+ int rc;
+
+ rc = pci_register_driver(&cxl_pci_driver);
+ if (rc)
+ return rc;
+
+ rc = cxl_cper_register_notifier(cxl_cper_event_call);
+ if (rc)
+ pci_unregister_driver(&cxl_pci_driver);
+
+ return rc;
+}
+
+static void __exit cxl_pci_driver_exit(void)
+{
+ cxl_cper_unregister_notifier(cxl_cper_event_call);
+ pci_unregister_driver(&cxl_pci_driver);
+}
+
+module_init(cxl_pci_driver_init);
+module_exit(cxl_pci_driver_exit);
MODULE_LICENSE("GPL v2");
-module_pci_driver(cxl_pci_driver);
MODULE_IMPORT_NS(CXL);
diff --git a/include/linux/pci.h b/include/linux/pci.h
index 60ca768bc867..290d0a2651b2 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -1170,6 +1170,7 @@ int pci_get_interrupt_pin(struct pci_dev *dev, struct pci_dev **bridge);
u8 pci_common_swizzle(struct pci_dev *dev, u8 *pinp);
struct pci_dev *pci_dev_get(struct pci_dev *dev);
void pci_dev_put(struct pci_dev *dev);
+DEFINE_FREE(pci_dev_put, struct pci_dev *, if (_T) pci_dev_put(_T))
void pci_remove_bus(struct pci_bus *b);
void pci_stop_and_remove_bus_device(struct pci_dev *dev);
void pci_stop_and_remove_bus_device_locked(struct pci_dev *dev);
@@ -1871,6 +1872,7 @@ void pci_cfg_access_unlock(struct pci_dev *dev);
void pci_dev_lock(struct pci_dev *dev);
int pci_dev_trylock(struct pci_dev *dev);
void pci_dev_unlock(struct pci_dev *dev);
+DEFINE_GUARD(pci_dev, struct pci_dev *, pci_dev_lock(_T), pci_dev_unlock(_T))
/*
* PCI domain support. Sometimes called PCI segment (eg by ACPI),