linux-kernel - [PATCH 3/3] ACPI: APEI: GHES: Improve ghes_notify

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-Id: <20251203130253.73888-4-xueshuai@linux.alibaba.com>
Date: Wed,  3 Dec 2025 21:02:53 +0800
From: Shuai Xue <xueshuai@...ux.alibaba.com>
To: tony.luck@...el.com,
	guohanjun@...wei.com,
	mchehab@...nel.org,
	yazen.ghannam@....com
Cc: dave.jiang@...el.com,
	Smita.KoralahalliChannabasappa@....com,
	leitao@...ian.org,
	pengdonglin@...omi.com,
	xueshuai@...ux.alibaba.com,
	baolin.wang@...ux.alibaba.com,
	benjamin.cheatham@....com,
	bp@...en8.de,
	dan.j.williams@...el.com,
	james.morse@....com,
	lenb@...nel.org,
	linux-acpi@...r.kernel.org,
	linux-kernel@...r.kernel.org,
	rafael@...nel.org,
	zhuo.song@...ux.alibaba.com
Subject: [PATCH 3/3] ACPI: APEI: GHES: Improve ghes_notify_sea() status check

Performance testing on ARMv8 systems shows significant overhead in error
status handling in SEA error handling.

- ghes_peek_estatus(): 8,138.3 ns (21,160 cycles).
- ghes_clear_estatus(): 2,038.3 ns (5,300 cycles).

Apply the same optimization used in ghes_notify_nmi() to
ghes_notify_sea() by checking for active errors before processing,

Signed-off-by: Shuai Xue <xueshuai@...ux.alibaba.com>
---
 drivers/acpi/apei/ghes.c | 20 +++++++++++++++++---
 1 file changed, 17 insertions(+), 3 deletions(-)

diff --git a/drivers/acpi/apei/ghes.c b/drivers/acpi/apei/ghes.c
index 2c7f3ca6ce50..0be46d63db53 100644
--- a/drivers/acpi/apei/ghes.c
+++ b/drivers/acpi/apei/ghes.c
@@ -1455,6 +1455,9 @@ int ghes_notify_sea(void)
 	static DEFINE_RAW_SPINLOCK(ghes_notify_lock_sea);
 	int rv;
 
+	if (!ghes_has_active_errors(&ghes_sea))
+		return -ENOENT;
+
 	raw_spin_lock(&ghes_notify_lock_sea);
 	rv = ghes_in_nmi_spool_from_list(&ghes_sea, FIX_APEI_GHES_SEA);
 	raw_spin_unlock(&ghes_notify_lock_sea);
@@ -1462,11 +1465,19 @@ int ghes_notify_sea(void)
 	return rv;
 }
 
-static void ghes_sea_add(struct ghes *ghes)
+static int ghes_sea_add(struct ghes *ghes)
 {
+	int rc;
+
+	rc = ghes_map_error_status(ghes);
+	if (rc)
+		return rc;
+
 	mutex_lock(&ghes_list_mutex);
 	list_add_rcu(&ghes->list, &ghes_sea);
 	mutex_unlock(&ghes_list_mutex);
+
+	return 0;
 }
 
 static void ghes_sea_remove(struct ghes *ghes)
@@ -1474,10 +1485,11 @@ static void ghes_sea_remove(struct ghes *ghes)
 	mutex_lock(&ghes_list_mutex);
 	list_del_rcu(&ghes->list);
 	mutex_unlock(&ghes_list_mutex);
+	ghes_unmap_error_status(ghes);
 	synchronize_rcu();
 }
 #else /* CONFIG_ACPI_APEI_SEA */
-static inline void ghes_sea_add(struct ghes *ghes) { }
+static inline int ghes_sea_add(struct ghes *ghes) { return -EINVAL; }
 static inline void ghes_sea_remove(struct ghes *ghes) { }
 #endif /* CONFIG_ACPI_APEI_SEA */
 
@@ -1710,7 +1722,9 @@ static int ghes_probe(struct platform_device *ghes_dev)
 		break;
 
 	case ACPI_HEST_NOTIFY_SEA:
-		ghes_sea_add(ghes);
+		rc = ghes_sea_add(ghes);
+		if (rc)
+			goto err;
 		break;
 	case ACPI_HEST_NOTIFY_NMI:
 		rc = ghes_nmi_add(ghes);
-- 
2.39.3