On Tue, May 14, 2019 at 12:43:11PM +0200, Cyril Hrubis wrote:

Hi!

quoted

quoted

There is a problem with the stack size accounting during execve when
there is no stack limit:

$ ulimit -s
8192
$ ./hello.ilp32 
Hello World!
$ ulimit -s unlimited
$ ./hello.ilp32 
Segmentation fault
$ strace ./hello.ilp32 
execve("./hello.ilp32", ["./hello.ilp32"], 0xfffff10548f0 /* 77 vars */) = -1 ENOMEM (Cannot allocate memory)

+++ killed by SIGSEGV +++

Segmentation fault (core dumped)

Andreas.

Thanks Andreas, I will take a look. Do we have such test in LTP?

So the problem was in not converting new compat-sensitive code:

diff --git a/arch/arm64/include/asm/processor.h b/arch/arm64/include/asm/processor.h
index 5bdf357169d8..c509f83fa506 100644
--- a/arch/arm64/include/asm/processor.h
+++ b/arch/arm64/include/asm/processor.h

@@ -63,7 +63,7 @@
 				TASK_SIZE_32 : TASK_SIZE_64)
 #define TASK_SIZE_OF(tsk)	(is_compat_thread(tsk) ? \
 				TASK_SIZE_32 : TASK_SIZE_64)
-#define DEFAULT_MAP_WINDOW	(test_thread_flag(TIF_32BIT) ? \
+#define DEFAULT_MAP_WINDOW	(is_compat_task() ? \
 				TASK_SIZE_32 : DEFAULT_MAP_WINDOW_64)
 #else
 #define TASK_SIZE		TASK_SIZE_64

The fix is incorporated in ilp32-5.1.1:
https://github.com/norov/linux/tree/ilp32-5.1.1

We do have a test that we can run a binary with very small stack size
i.e. 512kB but there does not seem to be anything that would catch this
specific problem.

Can you please open an issue and describe how to reproduce the problem
at our github tracker:

https://github.com/linux-test-project/ltp/issues

Then we can create testcase based on that reproducer later on.

-- 
Cyril Hrubis
chrubis@suse.cz

OK, I'll do.

Yury