Skip to contentSkip to conversation

hugemmap36: Migrating the libhugetlbfs/testcases/truncate_above_4GB.c test v6

PATCH In review Pavithra 2 participants 2 messages Apr 4 – Jul 2

lore
Share & tools
lore

Pavithra

author
#0 lore raw

PATCH 1/1 hugemmap36: Migrating the libhugetlbfs/testcases/truncate_above_4GB.c test v6

Test for a bug where truncating hugepage files at offsets >= 4GB on 32-bit systems could incorrectly truncate lower pages due to a misconversion of hugetlb_vmtruncate_list to a prio_tree. This test verifies that:
- Pages below 4GB remain accessible after truncation at 4GB
- Pages at or above the truncation point correctly trigger SIGBUS

---
 runtest/hugetlb                               |   1 +
 testcases/kernel/mem/.gitignore               |   1 +
 .../kernel/mem/hugetlb/hugemmap/hugemmap36.c  | 153 ++++++++++++++++++
 3 files changed, 155 insertions(+)
 create mode 100644 testcases/kernel/mem/hugetlb/hugemmap/hugemmap36.c
runtest/hugetlb +1 -0
index 0896d3c94..bd40a7a30 100644
@@ -36,6 +36,7 @@ hugemmap30 hugemmap30
 hugemmap31 hugemmap31
 hugemmap32 hugemmap32
 hugemmap34 hugemmap34
+hugemmap36 hugemmap36
 hugemmap05_1 hugemmap05 -m
 hugemmap05_2 hugemmap05 -s
 hugemmap05_3 hugemmap05 -s -m
testcases/kernel/mem/.gitignore +1 -0
index b4455de51..2ddef6bf1 100644
@@ -36,6 +36,7 @@
 /hugetlb/hugemmap/hugemmap31
 /hugetlb/hugemmap/hugemmap32
 /hugetlb/hugemmap/hugemmap34
+/hugetlb/hugemmap/hugemmap36
 /hugetlb/hugeshmat/hugeshmat01
 /hugetlb/hugeshmat/hugeshmat02
 /hugetlb/hugeshmat/hugeshmat03
testcases/kernel/mem/hugetlb/hugemmap/hugemmap36.c new file +153 -0

153 lines not on this page · raw message

Signature · 5 lines
2.53.0


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

Hi Pavithra,

Quoted 1 line · Pavithra

hugemmap36: Migrating the libhugetlbfs/testcases/truncate_above_4GB.c test v6

The "v6" revision tag belongs in the email subject only, not in the committed subject line. The subject should also use imperative mood:

hugemmap36: Migrate the libhugetlbfs/testcases/truncate_above_4GB.c test

testcases/kernel/mem/hugetlb/hugemmap/hugemmap36.c line 41

Quoted 6 lines
testcases/kernel/mem/hugetlb/hugemmap/hugemmap36.c new file +6 -0
+static void sigbus_handler_fail(int signum LTP_ATTRIBUTE_UNUSED,
+				siginfo_t *si LTP_ATTRIBUTE_UNUSED,
+				void *uc LTP_ATTRIBUTE_UNUSED)
+{
+	siglongjmp(sig_escape, 17);
+}

This handler does not record the unexpected SIGBUS in any way. The caller uses it here:

testcases/kernel/mem/hugetlb/hugemmap/hugemmap36.c line 111

Quoted 6 lines
testcases/kernel/mem/hugetlb/hugemmap/hugemmap36.c new file +6 -0
+	SAFE_SIGACTION(SIGBUS, &sa_fail, NULL);
+	if (sigsetjmp(sig_escape, 1) == 0)
+		if (*pi != 1) {
+			tst_res(TFAIL, "Data 1 has changed!");
+			goto cleanup;
+		}

If SIGBUS fires on *pi - which is exactly the kernel bug being tested (the low-offset page was incorrectly truncated) - siglongjmp returns 17, so sigsetjmp(...) == 0 is false, the TFAIL block is skipped entirely, and execution falls through.

The code then reaches the sa_pass / *qi block, *qi triggers SIGBUS, test_pass is set to 1, and the test reports TPASS. The test passes when the kernel bug is present.

sigbus_handler_fail needs to set a flag (e.g. test_pass = -1) so that after the first sigsetjmp block the caller can detect the unexpected SIGBUS and report TFAIL before proceeding.

Quoted 3 lines

+ * Test for a bug where truncating hugepage files at offsets >= 4GB on
+ * 32-bit systems could incorrectly truncate lower pages due to a
+ * misconversion of hugetlb_vmtruncate_list to a prio_tree.

...

testcases/kernel/mem/hugetlb/hugemmap/hugemmap36.c line 146

Quoted 1 line
testcases/kernel/mem/hugetlb/hugemmap/hugemmap36.c new file +1 -0
+	.needs_root = 1,

The doc-comment block does not explain why root is required. Ground Rule 4 requires that the reason be stated there.

Verdict - Needs revision

---
Note:

The agent can sometimes produce false positives although often its findings are genuine. If you find issues with the review, please comment this email or ignore the suggestions.

Regards,
LTP AI Reviewer

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