Skip to contentSkip to conversation

Hugetlb: Test to detect bug with freeing gigantic hugetlb pages

PATCH Superseded Tarun Sahu 2 participants 3 messages Apr 13, 2023 – Apr 19, 2023

v2 of 3 Superseded Latest is v3 →

lore
Share & tools
lore

Tarun Sahu

author
#0 lore raw

PATCH v2 1/1 Hugetlb: Test to detect bug with freeing gigantic hugetlb pages

Before kernel version 5.10-rc7, there was a bug that resulted in a "Bad Page State" error when freeing gigantic hugepages. This happened because the struct page entry compound_nr, which overlapped with page->mapping in the first tail page, was not cleared, causing the error. To ensure that this issue does not reoccur as struct page keeps changes and some fields are managed by folio, this test checks that freeing gigantic hugepages does not produce the above-mentioned error.

---
 .../kernel/mem/hugetlb/hugemmap/hugemmap32.c  | 85 +++++++++++++++++++
 testcases/kernel/mem/hugetlb/lib/hugetlb.h    |  6 ++
 2 files changed, 91 insertions(+)
 create mode 100644 testcases/kernel/mem/hugetlb/hugemmap/hugemmap32.c
testcases/kernel/mem/hugetlb/hugemmap/hugemmap32.c new file +85 -0
new file mode 100644
index 000000000..6b90d2546
@@ -0,0 +1,85 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Copyright (C) 2023, IBM Corporation.
+ * Author: Tarun Sahu
+ */
+
+/*\
+ * [Description]
+ *
+ * Before kernel version 5.10-rc7, there was a bug that resulted in a "Bad Page
+ * State" error when freeing gigantic hugepages. This happened because the
+ * struct page entry compound_nr, which overlapped with page->mapping in the
+ * first tail page, was not cleared, causing the error. To ensure that this
+ * issue does not reoccur as struct page keeps changing and some fields are
+ * managed by folio, this test checks that freeing gigantic hugepages does not
+ * produce the above-mentioned error.
+ */
+
+#define _GNU_SOURCE
+#include <dirent.h>
+
+#include <stdio.h>
+
+#include "hugetlb.h"
+
+#define PATH_GIGANTIC_HUGEPAGE "/sys/kernel/mm/hugepages"
+#define GIGANTIC_MIN_ORDER 10
+
+static int org_g_hpages;
+static char g_hpage_path[4096];
+
+static void run_test(void)
+{
+	if (FILE_PRINTF(g_hpage_path, "%d", 1))
+		tst_brk(TCONF, "Can't update the gigantic hugepages.");
+	SAFE_FILE_PRINTF(g_hpage_path, "%d", 0);
+
+	if (tst_taint_check())
+		tst_res(TFAIL, "Freeing Gigantic pages resulted in Bad Page State bug.");
+	else
+		tst_res(TPASS, "Successfully freed the gigantic hugepages");
+}
+
+static void setup(void)
+{
+	DIR *dir;
+	struct dirent *ent;
+	unsigned long hpage_size;
+
+	dir = SAFE_OPENDIR(PATH_GIGANTIC_HUGEPAGE);
+	while ((ent = SAFE_READDIR(dir))) {
+		if (strstr(ent->d_name, "hugepages-") != NULL) {
+			if ((sscanf(ent->d_name, "hugepages-%lukB", &hpage_size) == 1) &&
+				is_hugetlb_gigantic(hpage_size * 1024)) {
+				sprintf(g_hpage_path, "%s/%s/%s", PATH_GIGANTIC_HUGEPAGE,
+						ent->d_name, "nr_hugepages");
+				break;
+			}
+		}
+	}
+	if (!g_hpage_path[0])
+		tst_brk(TCONF, "Gigantic hugepages not supported");
+
+	SAFE_CLOSEDIR(dir);
+	SAFE_FILE_LINES_SCANF(g_hpage_path, "%d", &org_g_hpages);
+}
+
+static void cleanup(void)
+{
+	if (g_hpage_path[0])
+		SAFE_FILE_PRINTF(g_hpage_path, "%d", org_g_hpages);
+}
+
+static struct tst_test test = {
+	.tags = (struct tst_tag[]) {
+	    {"linux-git", "ba9c1201beaa"},
+	    {"linux-git", "7118fc2906e9"},
+	    {}
+	},
+	.needs_root = 1,
+	.setup = setup,
+	.cleanup = cleanup,
+	.test_all = run_test,
+	.taint_check = TST_TAINT_B,
+};
testcases/kernel/mem/hugetlb/lib/hugetlb.h +6 -0
index 241dab708..34fe08c24 100644
@@ -39,6 +39,12 @@
 # endif
 #endif
 
+/* Check if hugetlb page is gigantic */
+static inline int is_hugetlb_gigantic(unsigned long hpage_size)
+{
+	return (hpage_size / getpagesize()) >> 11;
+}
+
 /*
  * to get the lower nine permission bits
  * from shmid_ds.ipc_perm.mode
Signature · 5 lines
2.31.1


-- 
Mailing list info: https://lists.linux.it/listinfo/ltp
#1 lore raw

Hi!

Quoted 22 lines
testcases/kernel/mem/hugetlb/hugemmap/hugemmap32.c new file +22 -0
+#define _GNU_SOURCE
+#include <dirent.h>
+
+#include <stdio.h>
+
+#include "hugetlb.h"
+
+#define PATH_GIGANTIC_HUGEPAGE "/sys/kernel/mm/hugepages"
+#define GIGANTIC_MIN_ORDER 10
+
+static int org_g_hpages;
+static char g_hpage_path[4096];
+
+static void run_test(void)
+{
+	if (FILE_PRINTF(g_hpage_path, "%d", 1))
+		tst_brk(TCONF, "Can't update the gigantic hugepages.");
+	SAFE_FILE_PRINTF(g_hpage_path, "%d", 0);
+
+	if (tst_taint_check())
+		tst_res(TFAIL, "Freeing Gigantic pages resulted in Bad Page State bug.");
+	else

testcases/kernel/mem/hugetlb/hugemmap/hugemmap32.c line 50

Quoted 10 lines
testcases/kernel/mem/hugetlb/hugemmap/hugemmap32.c new file +10 -0
+		tst_res(TPASS, "Successfully freed the gigantic hugepages");
+}
+
+static void setup(void)
+{
+	DIR *dir;
+	struct dirent *ent;
+	unsigned long hpage_size;
+
+	dir = SAFE_OPENDIR(PATH_GIGANTIC_HUGEPAGE);

^
This is very minor, but there is nothing gigantic about the path, that's just sysfs hugepates directory, so I suppose that it should be just PATH_HUGEPAGE

testcases/kernel/mem/hugetlb/hugemmap/hugemmap32.c line 52

Quoted 2 lines
testcases/kernel/mem/hugetlb/hugemmap/hugemmap32.c new file +2 -0
+	while ((ent = SAFE_READDIR(dir))) {
+		if (strstr(ent->d_name, "hugepages-") != NULL) {

Isn't the strstr() here reduntant?

I as far as I can tell if the line in sscanf() will not match the call will simply return 0.

Quoted 15 lines
testcases/kernel/mem/hugetlb/hugemmap/hugemmap32.c new file +15 -0
+			if ((sscanf(ent->d_name, "hugepages-%lukB", &hpage_size) == 1) &&
+				is_hugetlb_gigantic(hpage_size * 1024)) {
+				sprintf(g_hpage_path, "%s/%s/%s", PATH_GIGANTIC_HUGEPAGE,
+						ent->d_name, "nr_hugepages");
+				break;
+			}
+		}
+	}
+	if (!g_hpage_path[0])
+		tst_brk(TCONF, "Gigantic hugepages not supported");
+
+	SAFE_CLOSEDIR(dir);
+	SAFE_FILE_LINES_SCANF(g_hpage_path, "%d", &org_g_hpages);
+}
+

testcases/kernel/mem/hugetlb/hugemmap/hugemmap32.c line 77

Quoted 10 lines
testcases/kernel/mem/hugetlb/hugemmap/hugemmap32.c new file +10 -0
+static void cleanup(void)
+{
+	if (g_hpage_path[0])
+		SAFE_FILE_PRINTF(g_hpage_path, "%d", org_g_hpages);
+}
+
+static struct tst_test test = {
+	.tags = (struct tst_tag[]) {
+	    {"linux-git", "ba9c1201beaa"},
+	    {"linux-git", "7118fc2906e9"},

^
This has appears to be wrong. Shouldn't the last digit be 2 instead of 9?

Quoted 28 lines
testcases/kernel/mem/hugetlb/hugemmap/hugemmap32.c new file +8 -0
+	    {}
+	},
+	.needs_root = 1,
+	.setup = setup,
+	.cleanup = cleanup,
+	.test_all = run_test,
+	.taint_check = TST_TAINT_B,
+};
testcases/kernel/mem/hugetlb/lib/hugetlb.h +6 -0
index 241dab708..34fe08c24 100644
@@ -39,6 +39,12 @@
 # endif
 #endif
 
+/* Check if hugetlb page is gigantic */
+static inline int is_hugetlb_gigantic(unsigned long hpage_size)
+{
+	return (hpage_size / getpagesize()) >> 11;
+}
+
 /*
  * to get the lower nine permission bits
  * from shmid_ds.ipc_perm.mode
Signature · 1 line
2.31.1
Signature · 5 lines
Cyril Hrubis
[email protected]

-- 
Mailing list info: https://lists.linux.it/listinfo/ltp
author
#2 lore raw

Hi Cyril,

Thanks for looking at it.
I agree to your comments, sent the v3. Also second commit tag was wrong. I changed it too.

~Tarun

Quoted 111 lines

Hi!

Quoted 22 lines
testcases/kernel/mem/hugetlb/hugemmap/hugemmap32.c new file +22 -0
+#define _GNU_SOURCE
+#include <dirent.h>
+
+#include <stdio.h>
+
+#include "hugetlb.h"
+
+#define PATH_GIGANTIC_HUGEPAGE "/sys/kernel/mm/hugepages"
+#define GIGANTIC_MIN_ORDER 10
+
+static int org_g_hpages;
+static char g_hpage_path[4096];
+
+static void run_test(void)
+{
+	if (FILE_PRINTF(g_hpage_path, "%d", 1))
+		tst_brk(TCONF, "Can't update the gigantic hugepages.");
+	SAFE_FILE_PRINTF(g_hpage_path, "%d", 0);
+
+	if (tst_taint_check())
+		tst_res(TFAIL, "Freeing Gigantic pages resulted in Bad Page State bug.");
+	else

testcases/kernel/mem/hugetlb/hugemmap/hugemmap32.c line 50

Quoted 10 lines
testcases/kernel/mem/hugetlb/hugemmap/hugemmap32.c new file +10 -0
+		tst_res(TPASS, "Successfully freed the gigantic hugepages");
+}
+
+static void setup(void)
+{
+	DIR *dir;
+	struct dirent *ent;
+	unsigned long hpage_size;
+
+	dir = SAFE_OPENDIR(PATH_GIGANTIC_HUGEPAGE);

^
This is very minor, but there is nothing gigantic about the path, that's just sysfs hugepates directory, so I suppose that it should be just PATH_HUGEPAGE

testcases/kernel/mem/hugetlb/hugemmap/hugemmap32.c line 52

Quoted 2 lines
testcases/kernel/mem/hugetlb/hugemmap/hugemmap32.c new file +2 -0
+	while ((ent = SAFE_READDIR(dir))) {
+		if (strstr(ent->d_name, "hugepages-") != NULL) {

Isn't the strstr() here reduntant?

I as far as I can tell if the line in sscanf() will not match the call will simply return 0.

Quoted 15 lines
testcases/kernel/mem/hugetlb/hugemmap/hugemmap32.c new file +15 -0
+			if ((sscanf(ent->d_name, "hugepages-%lukB", &hpage_size) == 1) &&
+				is_hugetlb_gigantic(hpage_size * 1024)) {
+				sprintf(g_hpage_path, "%s/%s/%s", PATH_GIGANTIC_HUGEPAGE,
+						ent->d_name, "nr_hugepages");
+				break;
+			}
+		}
+	}
+	if (!g_hpage_path[0])
+		tst_brk(TCONF, "Gigantic hugepages not supported");
+
+	SAFE_CLOSEDIR(dir);
+	SAFE_FILE_LINES_SCANF(g_hpage_path, "%d", &org_g_hpages);
+}
+

testcases/kernel/mem/hugetlb/hugemmap/hugemmap32.c line 77

Quoted 10 lines
testcases/kernel/mem/hugetlb/hugemmap/hugemmap32.c new file +10 -0
+static void cleanup(void)
+{
+	if (g_hpage_path[0])
+		SAFE_FILE_PRINTF(g_hpage_path, "%d", org_g_hpages);
+}
+
+static struct tst_test test = {
+	.tags = (struct tst_tag[]) {
+	    {"linux-git", "ba9c1201beaa"},
+	    {"linux-git", "7118fc2906e9"},

^
This has appears to be wrong. Shouldn't the last digit be 2 instead of 9?

Quoted 28 lines
testcases/kernel/mem/hugetlb/hugemmap/hugemmap32.c new file +8 -0
+	    {}
+	},
+	.needs_root = 1,
+	.setup = setup,
+	.cleanup = cleanup,
+	.test_all = run_test,
+	.taint_check = TST_TAINT_B,
+};
testcases/kernel/mem/hugetlb/lib/hugetlb.h +6 -0
index 241dab708..34fe08c24 100644
@@ -39,6 +39,12 @@
 # endif
 #endif
 
+/* Check if hugetlb page is gigantic */
+static inline int is_hugetlb_gigantic(unsigned long hpage_size)
+{
+	return (hpage_size / getpagesize()) >> 11;
+}
+
 /*
  * to get the lower nine permission bits
  * from shmid_ds.ipc_perm.mode
Signature · 1 line
2.31.1
Signature · 5 lines
Cyril Hrubis
[email protected]

-- 
Mailing list info: https://lists.linux.it/listinfo/ltp
Signature · 1 line
Mailing list info: https://lists.linux.it/listinfo/ltp