mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
* [PATCH 0/3] tools/nolibc: rlimit support
@ 2023-11-22 23:21 Thomas Weißschuh
  2023-11-22 23:21 ` [PATCH 1/3] tools/nolibc: drop custom definition of struct rusage Thomas Weißschuh
                   ` (2 more replies)
  0 siblings, 3 replies; 6+ messages in thread
From: Thomas Weißschuh @ 2023-11-22 23:21 UTC (permalink / raw)
  To: Willy Tarreau, Shuah Khan
  Cc: linux-kernel, linux-kselftest, Thomas Weißschuh

The series adds support for setrlimit/getrlimit.
Mainly to avoid spurious coredumps when running the tests under
qemu-user.

Signed-off-by: Thomas Weißschuh <linux@weissschuh.net>
---
Thomas Weißschuh (3):
      tools/nolibc: drop custom definition of struct rusage
      tools/nolibc: add support for getrlimit/setrlimit
      selftests/nolibc: disable coredump via setrlimit

 tools/include/nolibc/sys.h                   | 38 ++++++++++++++++++++++++++++
 tools/include/nolibc/types.h                 | 21 +--------------
 tools/testing/selftests/nolibc/nolibc-test.c | 31 +++++++++++++++++++++++
 3 files changed, 70 insertions(+), 20 deletions(-)
---
base-commit: 0dbd4651f3f80151910a36416fa0df28a10c3b0a
change-id: 20231122-nolibc-rlimit-bb5b1f264fc4

Best regards,
-- 
Thomas Weißschuh <linux@weissschuh.net>


^ permalink raw reply	[flat|nested] 6+ messages in thread

* [PATCH 1/3] tools/nolibc: drop custom definition of struct rusage
  2023-11-22 23:21 [PATCH 0/3] tools/nolibc: rlimit support Thomas Weißschuh
@ 2023-11-22 23:21 ` Thomas Weißschuh
  2023-11-22 23:21 ` [PATCH 2/3] tools/nolibc: add support for getrlimit/setrlimit Thomas Weißschuh
  2023-11-22 23:21 ` [PATCH 3/3] selftests/nolibc: disable coredump via setrlimit Thomas Weißschuh
  2 siblings, 0 replies; 6+ messages in thread
From: Thomas Weißschuh @ 2023-11-22 23:21 UTC (permalink / raw)
  To: Willy Tarreau, Shuah Khan
  Cc: linux-kernel, linux-kselftest, Thomas Weißschuh

A future commit will include linux/resource.h, which will conflict with
the private definition of struct rusage in nolibc.
Avoid the conflict by dropping the private definition and use the one
from the UAPI headers.

Signed-off-by: Thomas Weißschuh <linux@weissschuh.net>
---
 tools/include/nolibc/types.h | 21 +--------------------
 1 file changed, 1 insertion(+), 20 deletions(-)

diff --git a/tools/include/nolibc/types.h b/tools/include/nolibc/types.h
index ad0ddaa89e50..b26a5d0c417c 100644
--- a/tools/include/nolibc/types.h
+++ b/tools/include/nolibc/types.h
@@ -13,6 +13,7 @@
 #include <linux/stat.h>
 #include <linux/time.h>
 #include <linux/wait.h>
+#include <linux/resource.h>
 
 
 /* Only the generic macros and types may be defined here. The arch-specific
@@ -178,26 +179,6 @@ struct linux_dirent64 {
 	char           d_name[];
 };
 
-/* needed by wait4() */
-struct rusage {
-	struct timeval ru_utime;
-	struct timeval ru_stime;
-	long   ru_maxrss;
-	long   ru_ixrss;
-	long   ru_idrss;
-	long   ru_isrss;
-	long   ru_minflt;
-	long   ru_majflt;
-	long   ru_nswap;
-	long   ru_inblock;
-	long   ru_oublock;
-	long   ru_msgsnd;
-	long   ru_msgrcv;
-	long   ru_nsignals;
-	long   ru_nvcsw;
-	long   ru_nivcsw;
-};
-
 /* The format of the struct as returned by the libc to the application, which
  * significantly differs from the format returned by the stat() syscall flavours.
  */

-- 
2.43.0


^ permalink raw reply	[flat|nested] 6+ messages in thread

* [PATCH 2/3] tools/nolibc: add support for getrlimit/setrlimit
  2023-11-22 23:21 [PATCH 0/3] tools/nolibc: rlimit support Thomas Weißschuh
  2023-11-22 23:21 ` [PATCH 1/3] tools/nolibc: drop custom definition of struct rusage Thomas Weißschuh
@ 2023-11-22 23:21 ` Thomas Weißschuh
  2023-11-26  9:28   ` Willy Tarreau
  2023-11-22 23:21 ` [PATCH 3/3] selftests/nolibc: disable coredump via setrlimit Thomas Weißschuh
  2 siblings, 1 reply; 6+ messages in thread
From: Thomas Weißschuh @ 2023-11-22 23:21 UTC (permalink / raw)
  To: Willy Tarreau, Shuah Khan
  Cc: linux-kernel, linux-kselftest, Thomas Weißschuh

The implementation uses the prlimit64 systemcall as that is available on
all architectures.

Signed-off-by: Thomas Weißschuh <linux@weissschuh.net>
---
 tools/include/nolibc/sys.h                   | 38 ++++++++++++++++++++++++++++
 tools/testing/selftests/nolibc/nolibc-test.c | 29 +++++++++++++++++++++
 2 files changed, 67 insertions(+)

diff --git a/tools/include/nolibc/sys.h b/tools/include/nolibc/sys.h
index 2f359cb03d10..dda9dffd1d74 100644
--- a/tools/include/nolibc/sys.h
+++ b/tools/include/nolibc/sys.h
@@ -21,6 +21,7 @@
 #include <linux/fcntl.h> /* for O_* and AT_* */
 #include <linux/stat.h>  /* for statx() */
 #include <linux/prctl.h>
+#include <linux/resource.h>
 
 #include "arch.h"
 #include "errno.h"
@@ -898,6 +899,43 @@ int reboot(int cmd)
 }
 
 
+/*
+ * int getrlimit(int resource, struct rlimit *rlim);
+ * int setrlimit(int resource, const struct rlimit *rlim);
+ */
+
+static __attribute__((unused))
+int sys_prlimit64(pid_t pid, int resource,
+		  const struct rlimit64 *new_limit, struct rlimit64 *old_limit)
+{
+	return my_syscall4(__NR_prlimit64, pid, resource, new_limit, old_limit);
+}
+
+static __attribute__((unused))
+int getrlimit(int resource, struct rlimit *rlim)
+{
+	struct rlimit64 rlim64;
+	int ret;
+
+	ret = __sysret(sys_prlimit64(0, resource, NULL, &rlim64));
+	rlim->rlim_cur = rlim64.rlim_cur;
+	rlim->rlim_max = rlim64.rlim_max;
+
+	return ret;
+}
+
+static __attribute__((unused))
+int setrlimit(int resource, const struct rlimit *rlim)
+{
+	struct rlimit64 rlim64 = {
+		.rlim_cur = rlim->rlim_cur,
+		.rlim_max = rlim->rlim_max,
+	};
+
+	return __sysret(sys_prlimit64(0, resource, &rlim64, NULL));
+}
+
+
 /*
  * int sched_yield(void);
  */
diff --git a/tools/testing/selftests/nolibc/nolibc-test.c b/tools/testing/selftests/nolibc/nolibc-test.c
index 2b71fb5fae4e..d07cebace107 100644
--- a/tools/testing/selftests/nolibc/nolibc-test.c
+++ b/tools/testing/selftests/nolibc/nolibc-test.c
@@ -22,6 +22,7 @@
 #include <sys/mount.h>
 #include <sys/prctl.h>
 #include <sys/reboot.h>
+#include <sys/resource.h>
 #include <sys/stat.h>
 #include <sys/syscall.h>
 #include <sys/sysmacros.h>
@@ -839,6 +840,33 @@ int test_pipe(void)
 	return !!memcmp(buf, msg, len);
 }
 
+int test_rlimit(void)
+{
+	struct rlimit rlim = {
+		.rlim_cur = 1 << 20,
+		.rlim_max = 1 << 20,
+	};
+	int ret;
+
+	ret = setrlimit(RLIMIT_CORE, &rlim);
+	if (ret)
+		return -1;
+
+	rlim.rlim_cur = 0;
+	rlim.rlim_max = 0;
+
+	ret = getrlimit(RLIMIT_CORE, &rlim);
+	if (ret)
+		return -1;
+
+	if (rlim.rlim_cur != 1 << 20)
+		return -1;
+	if (rlim.rlim_max != 1 << 20)
+		return -1;
+
+	return 0;
+}
+
 
 /* Run syscall tests between IDs <min> and <max>.
  * Return 0 on success, non-zero on failure.
@@ -929,6 +957,7 @@ int run_syscall(int min, int max)
 		CASE_TEST(poll_fault);        EXPECT_SYSER(1, poll(NULL, 1, 0), -1, EFAULT); break;
 		CASE_TEST(prctl);             EXPECT_SYSER(1, prctl(PR_SET_NAME, (unsigned long)NULL, 0, 0, 0), -1, EFAULT); break;
 		CASE_TEST(read_badf);         EXPECT_SYSER(1, read(-1, &tmp, 1), -1, EBADF); break;
+		CASE_TEST(rlimit);            EXPECT_SYSZR(1, test_rlimit()); break;
 		CASE_TEST(rmdir_blah);        EXPECT_SYSER(1, rmdir("/blah"), -1, ENOENT); break;
 		CASE_TEST(sched_yield);       EXPECT_SYSZR(1, sched_yield()); break;
 		CASE_TEST(select_null);       EXPECT_SYSZR(1, ({ struct timeval tv = { 0 }; select(0, NULL, NULL, NULL, &tv); })); break;

-- 
2.43.0


^ permalink raw reply	[flat|nested] 6+ messages in thread

* [PATCH 3/3] selftests/nolibc: disable coredump via setrlimit
  2023-11-22 23:21 [PATCH 0/3] tools/nolibc: rlimit support Thomas Weißschuh
  2023-11-22 23:21 ` [PATCH 1/3] tools/nolibc: drop custom definition of struct rusage Thomas Weißschuh
  2023-11-22 23:21 ` [PATCH 2/3] tools/nolibc: add support for getrlimit/setrlimit Thomas Weißschuh
@ 2023-11-22 23:21 ` Thomas Weißschuh
  2 siblings, 0 replies; 6+ messages in thread
From: Thomas Weißschuh @ 2023-11-22 23:21 UTC (permalink / raw)
  To: Willy Tarreau, Shuah Khan
  Cc: linux-kernel, linux-kselftest, Thomas Weißschuh

qemu-user does has its own implementation of coredumping.
That implementation does not respect the call to
prctl(PR_SET_DUMPABLE, 0) in run_protection().
This leads to a coredump for every test run under qemu-user.

Use also setrlimit() to inhibit coredump creation which is respected by
qemu-user.

Link: https://lore.kernel.org/qemu-devel/20231115-qemu-user-dumpable-v1-2-edbe7f0fbb02@t-8ch.de/
Signed-off-by: Thomas Weißschuh <linux@weissschuh.net>
---
 tools/testing/selftests/nolibc/nolibc-test.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/tools/testing/selftests/nolibc/nolibc-test.c b/tools/testing/selftests/nolibc/nolibc-test.c
index d07cebace107..ccb8d017a3e7 100644
--- a/tools/testing/selftests/nolibc/nolibc-test.c
+++ b/tools/testing/selftests/nolibc/nolibc-test.c
@@ -1167,6 +1167,7 @@ static int run_protection(int min __attribute__((unused)),
 {
 	pid_t pid;
 	int llen = 0, status;
+	struct rlimit rlimit = { 0, 0 };
 
 	llen += printf("0 -fstackprotector ");
 
@@ -1198,6 +1199,7 @@ static int run_protection(int min __attribute__((unused)),
 		close(STDERR_FILENO);
 
 		prctl(PR_SET_DUMPABLE, 0, 0, 0, 0);
+		setrlimit(RLIMIT_CORE, &rlimit);
 		smash_stack();
 		return 1;
 

-- 
2.43.0


^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [PATCH 2/3] tools/nolibc: add support for getrlimit/setrlimit
  2023-11-22 23:21 ` [PATCH 2/3] tools/nolibc: add support for getrlimit/setrlimit Thomas Weißschuh
@ 2023-11-26  9:28   ` Willy Tarreau
  2023-11-26 10:42     ` Thomas Weißschuh
  0 siblings, 1 reply; 6+ messages in thread
From: Willy Tarreau @ 2023-11-26  9:28 UTC (permalink / raw)
  To: Thomas Weißschuh; +Cc: Shuah Khan, linux-kernel, linux-kselftest

Hi Thomas,

> +int test_rlimit(void)
> +{
> +	struct rlimit rlim = {
> +		.rlim_cur = 1 << 20,
> +		.rlim_max = 1 << 20,
> +	};
> +	int ret;
> +
> +	ret = setrlimit(RLIMIT_CORE, &rlim);
> +	if (ret)
> +		return -1;
> +
> +	rlim.rlim_cur = 0;
> +	rlim.rlim_max = 0;
> +
> +	ret = getrlimit(RLIMIT_CORE, &rlim);
> +	if (ret)
> +		return -1;
> +
> +	if (rlim.rlim_cur != 1 << 20)
> +		return -1;
> +	if (rlim.rlim_max != 1 << 20)
> +		return -1;

I think you should used two different values here for cur and max so
that you can also detect stupid API bugs such as a union being used
instead of a struct, or copy-pastes in the implementation etc. For
example using 1<<20 and 1<<21 should do the trick.

Otherwise Ack-by me for the whole series, of course.

Thanks,
Willy

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [PATCH 2/3] tools/nolibc: add support for getrlimit/setrlimit
  2023-11-26  9:28   ` Willy Tarreau
@ 2023-11-26 10:42     ` Thomas Weißschuh
  0 siblings, 0 replies; 6+ messages in thread
From: Thomas Weißschuh @ 2023-11-26 10:42 UTC (permalink / raw)
  To: Willy Tarreau; +Cc: Shuah Khan, linux-kernel, linux-kselftest

On 2023-11-26 10:28:28+0100, Willy Tarreau wrote:
> Hi Thomas,
> 
> > +int test_rlimit(void)
> > +{
> > +	struct rlimit rlim = {
> > +		.rlim_cur = 1 << 20,
> > +		.rlim_max = 1 << 20,
> > +	};
> > +	int ret;
> > +
> > +	ret = setrlimit(RLIMIT_CORE, &rlim);
> > +	if (ret)
> > +		return -1;
> > +
> > +	rlim.rlim_cur = 0;
> > +	rlim.rlim_max = 0;
> > +
> > +	ret = getrlimit(RLIMIT_CORE, &rlim);
> > +	if (ret)
> > +		return -1;
> > +
> > +	if (rlim.rlim_cur != 1 << 20)
> > +		return -1;
> > +	if (rlim.rlim_max != 1 << 20)
> > +		return -1;
> 
> I think you should used two different values here for cur and max so
> that you can also detect stupid API bugs such as a union being used
> instead of a struct, or copy-pastes in the implementation etc. For
> example using 1<<20 and 1<<21 should do the trick.

Good point, I incorporated the suggestion.

> Otherwise Ack-by me for the whole series, of course.

Thanks!

FYI I retested and pushed the series.


Thomas

^ permalink raw reply	[flat|nested] 6+ messages in thread

end of thread, other threads:[~2023-11-26 10:42 UTC | newest]

Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2023-11-22 23:21 [PATCH 0/3] tools/nolibc: rlimit support Thomas Weißschuh
2023-11-22 23:21 ` [PATCH 1/3] tools/nolibc: drop custom definition of struct rusage Thomas Weißschuh
2023-11-22 23:21 ` [PATCH 2/3] tools/nolibc: add support for getrlimit/setrlimit Thomas Weißschuh
2023-11-26  9:28   ` Willy Tarreau
2023-11-26 10:42     ` Thomas Weißschuh
2023-11-22 23:21 ` [PATCH 3/3] selftests/nolibc: disable coredump via setrlimit Thomas Weißschuh

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®