Skip to contentSkip to conversation

Migrating the libhugetlbfs/testcases/alloc-instantiate-race.c test

PATCH In review Samir Mulani 2 participants 2 messages Oct 25, 2024 – Nov 8, 2024

lore
Share & tools
lore
#0 lore raw

PATCH v2 1/1 Migrating the libhugetlbfs/testcases/alloc-instantiate-race.c test

This test is designed to detect a kernel allocation race introduced with hugepage demand-faulting. The problem is that no lock is held between allocating a hugepage and instantiating it in the pagetables or page cache index. In between the two, the (huge) page is cleared, so there's substantial time. Thus two processes can race instantiating the (same) last available hugepage - one will fail on the allocation, and thus cause an OOM fault even though the page it actually wants is being instantiated by the other racing process.

---
v2: --Addressed the below requested changes

1. Added static keyword to global variable declarations for better encapsulation.
2. Removed test name from the test description for conciseness.
3. Implemented support for dynamic CPU mask allocation using CPU_ALLOC.
4. Enhanced error handling by using TERRNO for clearer error reporting.
5. Specified .min_cpus = 2 in tst_test to ensure the test runs only when at least two CPUs are online.
6. Integrated sched_getaffinity() to retrieve the IDs of available online CPUs.
7. Adjusted exit status handling in run_race() as per review recommendations.
8. Removed unnecessary TINFO messages to streamline output.
9. Ran make check and addressed all flagged issues.

---
 runtest/hugetlb                               |   1 +
 testcases/kernel/mem/.gitignore               |   1 +
 .../kernel/mem/hugetlb/hugemmap/hugemmap36.c  | 272 ++++++++++++++++++
 3 files changed, 274 insertions(+)
 create mode 100644 testcases/kernel/mem/hugetlb/hugemmap/hugemmap36.c
runtest/hugetlb +1 -0
index f294e9aaa..cc1e56f16 100644
@@ -35,6 +35,7 @@ hugemmap29 hugemmap29
 hugemmap30 hugemmap30
 hugemmap31 hugemmap31
 hugemmap32 hugemmap32
+hugemmap36 hugemmap36
 hugemmap05_1 hugemmap05 -m
 hugemmap05_2 hugemmap05 -s
 hugemmap05_3 hugemmap05 -s -m
testcases/kernel/mem/.gitignore +1 -0
index d88484fa1..8a0fcbdf5 100644
@@ -34,6 +34,7 @@
 /hugetlb/hugemmap/hugemmap30
 /hugetlb/hugemmap/hugemmap31
 /hugetlb/hugemmap/hugemmap32
+/hugetlb/hugemmap/hugemmap36
 /hugetlb/hugeshmat/hugeshmat01
 /hugetlb/hugeshmat/hugeshmat02
 /hugetlb/hugeshmat/hugeshmat03
testcases/kernel/mem/hugetlb/hugemmap/hugemmap36.c new file +272 -0

272 lines not on this page · raw message

Signature · 5 lines
2.31.1


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

Hi Samir,

Not a whole review, just few notes.

nit: missing static:

$ make check-hugemmap36
hugemmap36.c:28:15: warning: Symbol 'totpages' has no prototype or library ('tst_') prefix. Should it be static?
hugemmap36.c:91:6: warning: Symbol 'check_online_cpus' has no prototype or library ('tst_') prefix. Should it be static?

...

runtest/hugetlb hugemmap29 hugemmap29 · line 38

Quoted 9 lines
runtest/hugetlb +1 -0
index f294e9aaa..cc1e56f16 100644
@@ -35,6 +35,7 @@ hugemmap29 hugemmap29
 hugemmap30 hugemmap30
 hugemmap31 hugemmap31
 hugemmap32 hugemmap32
+hugemmap36 hugemmap36

You added hugemmap36 without -m ... parameter, but that fails.
I also wonder if we should test both private and shared or just one. Also, if one more important, maybe it could be a default (thus no mandatory to pass it with -m ...) and the other one optional.

Also I get warning for shared:

./hugemmap36 -m shared
tst_hugepage.c:84: TINFO: 2 hugepage(s) reserved
tst_tmpdir.c:316: TINFO: Using /tmp/LTP_hugL2gsLX as tmpdir (tmpfs filesystem)
tst_test.c:1100: TINFO: Mounting none to /tmp/LTP_hugL2gsLX/hugetlbfs fstyp=hugetlbfs flags=0
tst_test.c:1890: TINFO: LTP version: 20240930-63-g6408294d8
tst_test.c:1894: TINFO: Tested kernel: 6.12.0-rc6-1.gb3de43a-default #1 SMP PREEMPT_DYNAMIC Mon Nov 4 00:37:44 UTC 2024 (b3de43a) x86_64 tst_test.c:1725: TINFO: Timeout per run is 0h 00m 30s
hugemmap36.c:241: TINFO: Mapping 1/2 pages..
hugemmap36.c:213: TINFO: instantiating..
hugemmap36.c:132: TINFO: Mapping final page..
hugemmap36.c:145: TINFO: Child 1 status: 0

hugemmap36.c:149: TINFO: Child 2 status: 0

hugemmap36.c:219: TPASS: Test done hugemmap36.c:251: TWARN: kill(3887,SIGKILL) failed: ESRCH (3) hugemmap36.c:253: TWARN: kill(3888,SIGKILL) failed: ESRCH (3)

Test is failing when run more than once:
./hugemmap36 -m private -i2
tst_hugepage.c:84: TINFO: 2 hugepage(s) reserved
tst_tmpdir.c:316: TINFO: Using /tmp/LTP_hugBynfWl as tmpdir (tmpfs filesystem)
tst_test.c:1100: TINFO: Mounting none to /tmp/LTP_hugBynfWl/hugetlbfs fstyp=hugetlbfs flags=0
tst_test.c:1890: TINFO: LTP version: 20240930-63-g6408294d8
tst_test.c:1894: TINFO: Tested kernel: 6.12.0-rc6-1.gb3de43a-default #1 SMP PREEMPT_DYNAMIC Mon Nov 4 00:37:44 UTC 2024 (b3de43a) x86_64 tst_test.c:1725: TINFO: Timeout per run is 0h 00m 30s
hugemmap36.c:241: TINFO: Mapping 1/2 pages..
hugemmap36.c:213: TINFO: instantiating..
hugemmap36.c:132: TINFO: Mapping final page..
hugemmap36.c:219: TPASS: Test done
hugemmap36.c:213: TINFO: instantiating..
hugemmap36.c:105: TBROK: At least 2 online CPUs are required : ECHILD (10)

Ideally it'd allow to run test more times with -iN or -IN.

Could you please rebase your branch before posting a new version? (conflict in the runtest and .gitignore)

Quoted 1 line
testcases/kernel/mem/hugetlb/hugemmap/hugemmap36.c +0 -0

...

testcases/kernel/mem/hugetlb/hugemmap/hugemmap36.c line 55

Quoted 13 lines
testcases/kernel/mem/hugetlb/hugemmap/hugemmap36.c +13 -0
+static int one_racer(void *p, int cpu,
+		     volatile int *mytrigger, volatile int *othertrigger)
+{
+	volatile int *pi = p;
+	cpu_set_t *cpuset;
+	size_t mask_size;
+	int err;
+
+	// Allocate CPU mask dynamically
+	cpuset = CPU_ALLOC(cpu + 1);
+	if (!cpuset)
+		tst_brk(TBROK | TERRNO, "CPU_ALLOC() failed");
+	// Get the required size for the allocated CPU set

nit: some comments for obvious things (like this one) are IMHO useless

Quoted 4 lines
testcases/kernel/mem/hugetlb/hugemmap/hugemmap36.c +4 -0
+	mask_size = CPU_ALLOC_SIZE(cpu + 1);
+
+	/* Split onto different CPUs to encourage the race */
+	CPU_ZERO_S(mask_size, cpuset);

testcases/kernel/mem/hugetlb/hugemmap/hugemmap36.c line 69

Quoted 10 lines
testcases/kernel/mem/hugetlb/hugemmap/hugemmap36.c +10 -0
+	CPU_SET_S(cpu, mask_size, cpuset);
+	// Set CPU affinity using the allocated mask size
+	err = sched_setaffinity(getpid(), mask_size, cpuset);
+	if (err == -1)
+		tst_brk(TBROK | TERRNO, "sched_setaffinity() failed");
+	/* Ready */
+	*mytrigger = 1;
+	/* Wait for the other trigger to be set */
+	while (!*othertrigger)
+		;

Shouldn't be here at least minimal usleep() to avoid burning CPU? Maybe not, because test is really fast.

testcases/kernel/mem/hugetlb/hugemmap/hugemmap36.c line 72

Quoted 3 lines
testcases/kernel/mem/hugetlb/hugemmap/hugemmap36.c +3 -0
+	/* Set the shared value */
+	*pi = 1;
+	// Free the dynamically allocated CPU set

Do we need to document this?

Quoted 3 lines
testcases/kernel/mem/hugetlb/hugemmap/hugemmap36.c +3 -0
+	CPU_FREE(cpuset);
+	return 0;
+}

...

testcases/kernel/mem/hugetlb/hugemmap/hugemmap36.c line 233

Quoted 12 lines
testcases/kernel/mem/hugetlb/hugemmap/hugemmap36.c +12 -0
+static void setup(void)
+{
+	totpages = SAFE_READ_MEMINFO(MEMINFO_HPAGE_FREE);
+	hpage_size = tst_get_hugepage_size();
+
+	if (str_op)
+		if (strcmp(str_op, "shared") == 0)
+			race_type = MAP_SHARED;
+		else if (strcmp(str_op, "private") == 0)
+			race_type = MAP_PRIVATE;
+		else
+			tst_res(TFAIL, "Usage:mmap<private|shared>");

It should tst_brk(TBROK, ...) on invalid parameter:
1) tst_brk() to quit the testing immediately
2) TBROK is more suitable than TFAIL in this case.
I also find "Usage:mmap<private|shared>" misleading - I first tried "mmapprivate".

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

Quoted 2 lines
testcases/kernel/mem/hugetlb/hugemmap/hugemmap36.c +2 -0
+	else
+		tst_res(TFAIL, "Usage:mmap<private|shared>");

As I said, one could of these be a default.

Quoted 14 lines
testcases/kernel/mem/hugetlb/hugemmap/hugemmap36.c +14 -0
+
+	fd_sync = tst_creat_unlinked(MNTPOINT, 0);
+	/* Get a shared normal page for synchronization */
+	q_sync = SAFE_MMAP(NULL, getpagesize(), PROT_READ|PROT_WRITE,
+			MAP_SHARED|MAP_ANONYMOUS, -1, 0);
+	tst_res(TINFO, "Mapping %ld/%ld pages.. ", totpages-1, totpages);
+	p_sync = SAFE_MMAP(NULL, (totpages-1)*hpage_size, PROT_READ|PROT_WRITE,
+			MAP_SHARED, fd_sync, 0);
+}
+
+static void cleanup(void)
+{
+	if (fd_sync >= 0)
+		SAFE_CLOSE(fd_sync);

testcases/kernel/mem/hugetlb/hugemmap/hugemmap36.c line 259

Quoted 10 lines
testcases/kernel/mem/hugetlb/hugemmap/hugemmap36.c +10 -0
+	if (child1)
+		SAFE_KILL(child1, SIGKILL);
+	if (child2)
+		SAFE_KILL(child2, SIGKILL);
+}
+
+
+static struct tst_test test = {
+	.options = (struct  tst_option[]){
+		{"m:", &str_op, "Usage:mmap<private|shared>"},

maybe:

		{"m:", &str_op, "Type of mmap() mapping <private|shared>"},

(see ./hugemmap36 -h for whole output)

Kind regards,
Petr

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