Re: [patch 2/2] clone: prepare to recycle CLONE_DETACHED andCLONE_STOPPED

From: Andrew Morton
Date: Thu Nov 08 2007 - 23:36:49 EST


> On Thu, 08 Nov 2007 13:31:43 -0800 akpm@xxxxxxxxxxxxxxxxxxxx wrote:
> From: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
>
> Ulrich says that we never used these clone flags and that nothing should be
> using them.
>
> As we're down to only a single bit left in clone's flags argument, let's add a
> warning to check that no userspace is actually using these. Hopefully we will
> be able to recycle them.
>
> Cc: Ulrich Drepper <drepper@xxxxxxxxxx>
> Cc: Ingo Molnar <mingo@xxxxxxx>
> Cc: Roland McGrath <roland@xxxxxxxxxx>
> Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
> ---
>
> kernel/fork.c | 16 ++++++++++++----
> 1 file changed, 12 insertions(+), 4 deletions(-)
>
> diff -puN kernel/fork.c~clone-prepare-to-recycle-clone_detached-and-clone_stopped kernel/fork.c
> --- a/kernel/fork.c~clone-prepare-to-recycle-clone_detached-and-clone_stopped
> +++ a/kernel/fork.c
> @@ -1420,10 +1420,18 @@ long do_fork(unsigned long clone_flags,
> int trace = 0;
> long nr;
>
> - if (unlikely(current->ptrace)) {
> - trace = fork_traceflag (clone_flags);
> - if (trace)
> - clone_flags |= CLONE_PTRACE;
> + /*
> + * We hope to recycle these flags after 2.6.26
> + */
> + if (unlikely(clone_flags & (CLONE_DETACHED|CLONE_STOPPED))) {
> + if (printk_ratelimit()) {
> + char comm[TASK_COMM_LEN];
> +
> + printk(KERN_INFO "fork(): process `%s' used deprecated "
> + "clone flags 0x%lx\n",
> + get_task_comm(comm, current),
> + clone_flags & (CLONE_DETACHED|CLONE_STOPPED));
> + }
> }
>
> p = copy_process(clone_flags, stack_start, regs, stack_size,

That was all screwed up. Better version:

From: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>

Ulrich says that we never used these clone flags and that nothing should be
using them.

As we're down to only a single bit left in clone's flags argument, let's add a
warning to check that no userspace is actually using these. Hopefully we will
be able to recycle them.

Roland said:

CLONE_STOPPED was previously used by some NTPL versions when under
thread_db (i.e. only when being actively debugged by gdb), but not for a
long time now, and it never worked reliably when it was used. Removing it
seems fine to me.

Cc: Ulrich Drepper <drepper@xxxxxxxxxx>
Cc: Ingo Molnar <mingo@xxxxxxx>
Cc: Roland McGrath <roland@xxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---

kernel/fork.c | 17 +++++++++++++++++
1 file changed, 17 insertions(+)

diff -puN kernel/fork.c~clone-prepare-to-recycle-clone_detached-and-clone_stopped kernel/fork.c
--- a/kernel/fork.c~clone-prepare-to-recycle-clone_detached-and-clone_stopped
+++ a/kernel/fork.c
@@ -1420,6 +1420,23 @@ long do_fork(unsigned long clone_flags,
int trace = 0;
long nr;

+ /*
+ * We hope to recycle these flags after 2.6.26
+ */
+ if (unlikely(clone_flags & (CLONE_DETACHED|CLONE_STOPPED))) {
+ static int __read_mostly count = 100;
+
+ if (count && printk_ratelimit()) {
+ char comm[TASK_COMM_LEN];
+
+ count--;
+ printk(KERN_INFO "fork(): process `%s' used deprecated "
+ "clone flags 0x%lx\n",
+ get_task_comm(comm, current),
+ clone_flags & (CLONE_DETACHED|CLONE_STOPPED));
+ }
+ }
+
if (unlikely(current->ptrace)) {
trace = fork_traceflag (clone_flags);
if (trace)
_

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/