Or really, just clone devolving into fork.  This should not ever happen
in practice.  We do want to reserve calling cpu_clone_regs for the case
in which we are actually performing a clone.

Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
---
 linux-user/syscall.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/linux-user/syscall.c b/linux-user/syscall.c
index 96cd4bf86d..f7d0754c8d 100644
--- a/linux-user/syscall.c
+++ b/linux-user/syscall.c
@@ -5553,10 +5553,14 @@ static int do_fork(CPUArchState *env, unsigned int flags, abi_ulong newsp,
         pthread_mutex_destroy(&info.mutex);
         pthread_mutex_unlock(&clone_lock);
     } else {
-        /* if no CLONE_VM, we consider it is a fork */
+        /* If no CLONE_VM, we consider it is a fork.  */
         if (flags & CLONE_INVALID_FORK_FLAGS) {
             return -TARGET_EINVAL;
         }
+        /* As a fork, setting a new sp does not make sense.  */
+        if (newsp) {
+            return -TARGET_EINVAL;
+        }
 
         /* We can't support custom termination signals */
         if ((flags & CSIGNAL) != TARGET_SIGCHLD) {
@@ -5571,7 +5575,6 @@ static int do_fork(CPUArchState *env, unsigned int flags, abi_ulong newsp,
         ret = fork();
         if (ret == 0) {
             /* Child Process.  */
-            cpu_clone_regs(env, newsp);
             fork_end(1);
             /* There is a race condition here.  The parent process could
                theoretically read the TID in the child process before the child
-- 
2.17.1

On 5/9/19 8:27 PM, Richard Henderson wrote:
> Or really, just clone devolving into fork.  This should not ever happen
> in practice.  We do want to reserve calling cpu_clone_regs for the case
> in which we are actually performing a clone.
> 
> Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
> ---
>  linux-user/syscall.c | 7 +++++--
>  1 file changed, 5 insertions(+), 2 deletions(-)
> 
> diff --git a/linux-user/syscall.c b/linux-user/syscall.c
> index 96cd4bf86d..f7d0754c8d 100644
> --- a/linux-user/syscall.c
> +++ b/linux-user/syscall.c
> @@ -5553,10 +5553,14 @@ static int do_fork(CPUArchState *env, unsigned int flags, abi_ulong newsp,
>          pthread_mutex_destroy(&info.mutex);
>          pthread_mutex_unlock(&clone_lock);
>      } else {
> -        /* if no CLONE_VM, we consider it is a fork */
> +        /* If no CLONE_VM, we consider it is a fork.  */
>          if (flags & CLONE_INVALID_FORK_FLAGS) {
>              return -TARGET_EINVAL;
>          }
> +        /* As a fork, setting a new sp does not make sense.  */
> +        if (newsp) {
> +            return -TARGET_EINVAL;
> +        }

This causes failures for aarch64 and riscv.

We have to allow no-op setting of sp as well.
Other targets set newsp to 0 for in vfork.S in glibc.


r~