ACPI: APEI: Fix missing ERST record id

Read a record is cleared by others, but the deleted record cache entry is
still created by erst_get_record_id_next. When next enumerate the records,
get the cached deleted record, then erst_read() return -ENOENT and try to
get next record, loop back to first ID will return 0 in function
__erst_record_id_cache_add_one and then set record_id as
APEI_ERST_INVALID_RECORD_ID, finished this time read operation.
It will result in read the records just in the cache hereafter.

This patch cleared the deleted record cache, fix the issue that
"./erst-inject -p" shows record counts not equal to "./erst-inject -n".

A reproducer of the problem(retry many times):

[root@localhost erst-inject]# ./erst-inject -c 0xaaaaa00011
[root@localhost erst-inject]# ./erst-inject -p
rc: 273
rcd sig: CPER
rcd id: 0xaaaaa00012
rc: 273
rcd sig: CPER
rcd id: 0xaaaaa00013
rc: 273
rcd sig: CPER
rcd id: 0xaaaaa00014
[root@localhost erst-inject]# ./erst-inject -i 0xaaaaa000006
[root@localhost erst-inject]# ./erst-inject -i 0xaaaaa000007
[root@localhost erst-inject]# ./erst-inject -i 0xaaaaa000008
[root@localhost erst-inject]# ./erst-inject -p
rc: 273
rcd sig: CPER
rcd id: 0xaaaaa00012
rc: 273
rcd sig: CPER
rcd id: 0xaaaaa00013
rc: 273
rcd sig: CPER
rcd id: 0xaaaaa00014
[root@localhost erst-inject]# ./erst-inject -n
total error record count: 6
Signed-off-by: Liu Xinpeng <liuxp11@chinatelecom.cn>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>

arch/x86/kernel/cpu/mce/apei.c

@@ -177,16 +177,14 @@ ssize_t apei_read_mce(struct mce *m, u64 *record_id)
 	/* no more record */
 	if (*record_id == APEI_ERST_INVALID_RECORD_ID)
 		goto out;
-	rc = erst_read(*record_id, &rcd.hdr, sizeof(rcd));
+	rc = erst_read_record(*record_id, &rcd.hdr, sizeof(rcd), sizeof(rcd),
+			&CPER_CREATOR_MCE);
 	/* someone else has cleared the record, try next one */
 	if (rc == -ENOENT)
 		goto retry;
 	else if (rc < 0)
 		goto out;
-	/* try to skip other type records in storage */
-	else if (rc != sizeof(rcd) ||
-		 !guid_equal(&rcd.hdr.creator_id, &CPER_CREATOR_MCE))
-		goto retry;
+
 	memcpy(m, &rcd.mce, sizeof(*m));
 	rc = sizeof(*m);
 out:

drivers/acpi/apei/erst-dbg.c

@@ -111,7 +111,8 @@ static ssize_t erst_dbg_read(struct file *filp, char __user *ubuf,
 		goto out;
 	}
 retry:
-	rc = len = erst_read(id, erst_dbg_buf, erst_dbg_buf_len);
+	rc = len = erst_read_record(id, erst_dbg_buf, erst_dbg_buf_len,
+			erst_dbg_buf_len, NULL);
 	/* The record may be cleared by others, try read next record */
 	if (rc == -ENOENT)
 		goto retry_next;

drivers/acpi/apei/erst.c

@@ -856,6 +856,74 @@ ssize_t erst_read(u64 record_id, struct cper_record_header *record,
 }
 EXPORT_SYMBOL_GPL(erst_read);
 
+static void erst_clear_cache(u64 record_id)
+{
+	int i;
+	u64 *entries;
+
+	mutex_lock(&erst_record_id_cache.lock);
+
+	entries = erst_record_id_cache.entries;
+	for (i = 0; i < erst_record_id_cache.len; i++) {
+		if (entries[i] == record_id)
+			entries[i] = APEI_ERST_INVALID_RECORD_ID;
+	}
+	__erst_record_id_cache_compact();
+
+	mutex_unlock(&erst_record_id_cache.lock);
+}
+
+ssize_t erst_read_record(u64 record_id, struct cper_record_header *record,
+		size_t buflen, size_t recordlen, const guid_t *creatorid)
+{
+	ssize_t len;
+
+	/*
+	 * if creatorid is NULL, read any record for erst-dbg module
+	 */
+	if (creatorid == NULL) {
+		len = erst_read(record_id, record, buflen);
+		if (len == -ENOENT)
+			erst_clear_cache(record_id);
+
+		return len;
+	}
+
+	len = erst_read(record_id, record, buflen);
+	/*
+	 * if erst_read return value is -ENOENT skip to next record_id,
+	 * and clear the record_id cache.
+	 */
+	if (len == -ENOENT) {
+		erst_clear_cache(record_id);
+		goto out;
+	}
+
+	if (len < 0)
+		goto out;
+
+	/*
+	 * if erst_read return value is less than record head length,
+	 * consider it as -EIO, and clear the record_id cache.
+	 */
+	if (len < recordlen) {
+		len = -EIO;
+		erst_clear_cache(record_id);
+		goto out;
+	}
+
+	/*
+	 * if creatorid is not wanted, consider it as not found,
+	 * for skipping to next record_id.
+	 */
+	if (!guid_equal(&record->creator_id, creatorid))
+		len = -ENOENT;
+
+out:
+	return len;
+}
+EXPORT_SYMBOL_GPL(erst_read_record);
+
 int erst_clear(u64 record_id)
 {
 	int rc, i;
@@ -996,16 +1064,13 @@ static ssize_t erst_reader(struct pstore_record *record)
 		goto out;
 	}
 
-	len = erst_read(record_id, &rcd->hdr, rcd_len);
+	len = erst_read_record(record_id, &rcd->hdr, rcd_len, sizeof(*rcd),
+			&CPER_CREATOR_PSTORE);
 	/* The record may be cleared by others, try read next record */
 	if (len == -ENOENT)
 		goto skip;
-	else if (len < 0 || len < sizeof(*rcd)) {
-		rc = -EIO;
+	else if (len < 0)
 		goto out;
-	}
-	if (!guid_equal(&rcd->hdr.creator_id, &CPER_CREATOR_PSTORE))
-		goto skip;
 
 	record->buf = kmalloc(len, GFP_KERNEL);
 	if (record->buf == NULL) {

include/acpi/apei.h

@@ -46,6 +46,8 @@ int erst_get_record_id_next(int *pos, u64 *record_id);
 void erst_get_record_id_end(void);
 ssize_t erst_read(u64 record_id, struct cper_record_header *record,
 		  size_t buflen);
+ssize_t erst_read_record(u64 record_id, struct cper_record_header *record,
+		size_t buflen, size_t recordlen, const guid_t *creatorid);
 int erst_clear(u64 record_id);
 
 int arch_apei_enable_cmcff(struct acpi_hest_header *hest_hdr, void *data);