github-mirror/linux
1
0
Fork 0
mirror of https://github.com/torvalds/linux.git synced 2026-11-07 06:17:02 +01:00
Code Issues Actions Packages Projects Releases Wiki Activity Graph

EDAC/intel-bff: Report frequent filter overflows

Browse Source 
If an instance of a bitfix filter contains some transient errors built
up over time, then resetting the filter will free up slots in the filter
to store persistent errors.

Save a timestamp when "yellow" status is seen and clear the filter.

Log at KERN_WARNING level if the overflow occurred quickly after a
previous overflow on the same bitfix filter instance. Use KERN_NOTICE
for first, or long delayed, overflow.

Co-developed-by: Qiuxu Zhuo <qiuxu.zhuo@intel.com>
Signed-off-by: Qiuxu Zhuo <qiuxu.zhuo@intel.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
Link: https://patch.msgid.link/20260828153002.10290-8-tony.luck@intel.com
This commit is contained in:
Tony Luck 2026-08-28 08:30:02 -07:00
parent 4835abed9f
commit 30713abb1e
1 changed files with 70 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

72
drivers/edac/intel-bff.c
View File

@ -25,13 +25,17 @@
#include <linux/cpuhplock.h>
#include <linux/device-id/x86_cpu.h>
#include <linux/errno.h>
#include <linux/gfp_types.h>
#include <linux/init.h>
#include <linux/jiffies.h>
#include <linux/limits.h>
#include <linux/module.h>
#include <linux/notifier.h>
#include <linux/printk.h>
#include <linux/slab.h>
#include <linux/topology.h>
#include <linux/types.h>
#include <linux/xarray.h>
#include <asm/cpu_device_id.h>
#include <asm/cpufeatures.h>
@ -40,6 +44,16 @@
#include <asm/msr.h>
#include <asm/msr-index.h>
/*
* A 10-minute observation period helps distinguish between:
*
* - A long-term accumulation of transient corrected errors
* (filter stays clear after reset).
*
* - Permanent defects (filter overflows again quickly).
*/
#define BFF_OVERFLOW_INTERVAL secs_to_jiffies(10 * 60)
/* Intel bitfix filter control register defines */
#define MSR_MC0_BFF_CTL 0x000006c0
#define MSR_MCx_BFF_CTL(x) (MSR_MC0_BFF_CTL + (x))
@ -77,6 +91,8 @@ MODULE_DEVICE_TABLE(x86cpu, bff_cpu_ids);
static const enum bff_bank_type *bff_bank_types;
static DEFINE_XARRAY(bff_bank_xa);
/* Diamond Rapids maps APICID[2] to the IMH instance within a socket. */
#define APICID_IMH_NUM GENMASK(2, 2)
#define IMH_NUM(apicid) FIELD_GET(APICID_IMH_NUM, apicid)
@ -149,6 +165,41 @@ static unsigned long bff_get_id(struct mce *mce)
return id;
}
/*
* Save current timestamp for bff_id. Return true if it is within
* BFF_OVERFLOW_INTERVAL of previous timestamp for this bff_id.
*/
static bool bff_overflow_is_frequent(unsigned long bff_id)
{
unsigned long now = jiffies, interval_end;
unsigned long *ts;
if (bff_id == ULONG_MAX)
return false;
ts = xa_load(&bff_bank_xa, bff_id);
if (!ts) {
ts = kzalloc_obj(*ts);
if (!ts) {
pr_warn("Failed to allocate timestamp for bitfix filter 0x%lx\n", bff_id);
return false;
}
if (xa_is_err(xa_store(&bff_bank_xa, bff_id, ts, GFP_KERNEL))) {
kfree(ts);
pr_warn("Failed to record timestamp for bitfix filter 0x%lx\n", bff_id);
return false;
}
*ts = now;
return false;
}
interval_end = *ts + BFF_OVERFLOW_INTERVAL;
*ts = now;
return time_before(now, interval_end);
}
static void bff_reset_and_report(struct mce *mce)
{
/* Reset bitfix filter using the CPU that logged the yellow status */
@ -156,8 +207,18 @@ static void bff_reset_and_report(struct mce *mce)
pr_warn("Failed to reset bitfix filter for CPU %d Bank %d\n",
mce->extcpu, mce->bank);
/* Placeholder use of bff_get_id() */
pr_debug("unique_id = 0x%lx\n", bff_get_id(mce));
/*
* Use the unique id for the bitfix filter instance that overflowed and
* check if this is a repeat within the BFF_OVERFLOW_INTERVAL. If it
* is, then report at WARN severity as the user may want to take action.
*/
if (bff_overflow_is_frequent(bff_get_id(mce))) {
pr_warn_ratelimited(HW_ERR "Socket %d CPU %d Bank %d bitfix filter overflowed frequently\n",
mce->socketid, mce->extcpu, mce->bank);
} else {
pr_notice_ratelimited(HW_ERR "Socket %d CPU %d Bank %d bitfix filter overflowed\n",
mce->socketid, mce->extcpu, mce->bank);
}
}
static int bff_mce_notify(struct notifier_block *nb, unsigned long val, void *data)
@ -216,7 +277,14 @@ static int __init bff_init(void)
static void __exit bff_exit(void)
{
unsigned long bff_id;
unsigned long *ts;
mce_unregister_decode_chain(&bff_notifier);
xa_for_each(&bff_bank_xa, bff_id, ts)
kfree(ts);
xa_destroy(&bff_bank_xa);
}
module_init(bff_init);