Hello Aishwarya! Thanks for the bug report! I tried this with the aarch64-linux-gnu-gcc-13 cross compiler on Debian, on next-20231215, but I can not reproduce it. On Fri, Dec 15, 2023 at 12:52:19PM +0000, Aishwarya TCV wrote: > On 08/12/2023 15:51, Günther Noack wrote: > > Exercises Landlock's IOCTL feature in different combinations of > > handling and permitting the rights LANDLOCK_ACCESS_FS_IOCTL, > > LANDLOCK_ACCESS_FS_READ_FILE, LANDLOCK_ACCESS_FS_WRITE_FILE and > > LANDLOCK_ACCESS_FS_READ_DIR, and in different combinations of using > > files and directories. > > > > Signed-off-by: Günther Noack <gnoack@xxxxxxxxxx> > > --- > > tools/testing/selftests/landlock/fs_test.c | 431 ++++++++++++++++++++- > > Hi Günther, > > When building kselftest against next-master the below build error is > observed. A bisect (full log > below) identified this patch as introducing the failure. > > Full log from a failure: > https://storage.kernelci.org/next/master/next-20231215/arm64/defconfig+kselftest/gcc-10/logs/kselftest.log > > ----- > make[4]: Entering directory > '/tmp/kci/linux/tools/testing/selftests/landlock' > aarch64-linux-gnu-gcc -Wall -O2 -isystem > /tmp/kci/linux/build/usr/include base_test.c -lcap -o > /tmp/kci/linux/build/kselftest/landlock/base_test > aarch64-linux-gnu-gcc -Wall -O2 -isystem > /tmp/kci/linux/build/usr/include fs_test.c -lcap -o > /tmp/kci/linux/build/kselftest/landlock/fs_test > In file included from /tmp/kci/linux/build/usr/include/linux/fs.h:19, > from fs_test.c:12: > /usr/include/aarch64-linux-gnu/sys/mount.h:35:3: error: expected The IOCTL patch set has introduced an "#include <linux/fs.h>" at the top of selftests/landlock/fs_test.c (that file includes some IOCTL command numbers), but that should in my mind normally be safe to include? I'm surprised that according to the log, fs.h line 19 is including sys/mount.h instead of linux/mount.h...? This line says “#include <linux/mount.h>”? > identifier before numeric constant > 35 | MS_RDONLY = 1, /* Mount read-only. */ > | ^~~~~~~~~ > In file included from common.h:19, > from fs_test.c:27: If you have more leads or more concrete reproduction steps, I'd be interested to know. Otherwise, I'd have to just hope that it'll work better on the next attempt...? Thanks, —Günther
