ring buffer: propagate __rb_map_vma return value to caller

[PATCH] ring buffer: propagate __rb_map_vma return value to caller

Posted by Ankit Khushwaha 4 months ago

The return value from `__rb_map_vma()`, which rejects writable or
executable mappings (VM_WRITE, VM_EXEC, or !VM_MAYSHARE), was being
ignored. As a result the caller of `__rb_map_vma` always returned 0 
even when the mapping had actually failed, allowing it to proceed
with an invalid VMA.

Reported-by: syzbot+ddc001b92c083dbf2b97@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?id=194151be8eaebd826005329b2e123aecae714bdb
Signed-off-by: Ankit Khushwaha <ankitkhushwaha.linux@gmail.com>
---

#syz test: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git 4ff71af020ae

---
 kernel/trace/ring_buffer.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c
index 43460949ad3f..4efb90364f48 100644
--- a/kernel/trace/ring_buffer.c
+++ b/kernel/trace/ring_buffer.c
@@ -7271,6 +7271,8 @@ int ring_buffer_map(struct trace_buffer *buffer, int cpu,
 		cpu_buffer->subbuf_ids = NULL;
 		rb_free_meta_page(cpu_buffer);
 		atomic_dec(&cpu_buffer->resize_disabled);
+		/* VM failed to be mapped */
+		return err;
 	}
 
 	return 0;
-- 
2.51.0

Re: [PATCH] ring buffer: propagate __rb_map_vma return value to caller

Posted by Steven Rostedt 4 months ago

On Tue,  7 Oct 2025 22:42:56 +0530
Ankit Khushwaha <ankitkhushwaha.linux@gmail.com> wrote:

Note, tracing subsystem expects the subject to start with a capital letter:

 ring buffer: Propagate __rb_map_vma return value to caller

> The return value from `__rb_map_vma()`, which rejects writable or
> executable mappings (VM_WRITE, VM_EXEC, or !VM_MAYSHARE), was being
> ignored. As a result the caller of `__rb_map_vma` always returned 0 
> even when the mapping had actually failed, allowing it to proceed
> with an invalid VMA.
> 
> Reported-by: syzbot+ddc001b92c083dbf2b97@syzkaller.appspotmail.com
> Closes: https://syzkaller.appspot.com/bug?id=194151be8eaebd826005329b2e123aecae714bdb
> Signed-off-by: Ankit Khushwaha <ankitkhushwaha.linux@gmail.com>
> ---
> 
> #syz test: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git 4ff71af020ae
> 
> ---
>  kernel/trace/ring_buffer.c | 2 ++
>  1 file changed, 2 insertions(+)
> 
> diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c
> index 43460949ad3f..4efb90364f48 100644
> --- a/kernel/trace/ring_buffer.c
> +++ b/kernel/trace/ring_buffer.c
> @@ -7271,6 +7271,8 @@ int ring_buffer_map(struct trace_buffer *buffer, int cpu,
>  		cpu_buffer->subbuf_ids = NULL;
>  		rb_free_meta_page(cpu_buffer);
>  		atomic_dec(&cpu_buffer->resize_disabled);
> +		/* VM failed to be mapped */

No need to add the comment. It's obvious what happened.

> +		return err;

Don't return here.

>  	}
>  
>  	return 0;

Change this to:

	return err;

as after that if statement, err will be 0 on success or the value you want
to return.

-- Steve

Re: [syzbot] [trace?] WARNING in tracing_buffers_mmap_close

Posted by syzbot 4 months ago

Hello,

syzbot has tested the proposed patch and the reproducer did not trigger any issue:

Reported-by: syzbot+ddc001b92c083dbf2b97@syzkaller.appspotmail.com
Tested-by: syzbot+ddc001b92c083dbf2b97@syzkaller.appspotmail.com

Tested on:

commit:         4ff71af0 Merge tag 'net-6.17-rc8' of git://git.kernel...
git tree:       https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git
console output: https://syzkaller.appspot.com/x/log.txt?x=17676892580000
kernel config:  https://syzkaller.appspot.com/x/.config?x=595e5938a1dd5b4e
dashboard link: https://syzkaller.appspot.com/bug?extid=ddc001b92c083dbf2b97
compiler:       gcc (Debian 12.2.0-14+deb12u1) 12.2.0, GNU ld (GNU Binutils for Debian) 2.40
patch:          https://syzkaller.appspot.com/x/patch.diff?x=1249d1e2580000

Note: testing is done by a robot and is best-effort only.