[PATCH] trace: fix return value of __setup handlers

[PATCH] trace: fix return value of __setup handlers

[Randy Dunlap]

__setup() handlers should generally return 1 to indicate that the
boot options have been handled.

Using invalid option values causes the entire kernel boot option
string to be reported as Unknown and added to init's environment
strings, polluting it.

  Unknown kernel command line parameters "BOOT_IMAGE=/boot/bzImage-517rc6
    kprobe_event=p,syscall_any,$arg1 trace_options=quiet
    trace_clock=jiffies", will be passed to user space.

 Run /sbin/init as init process
   with arguments:
     /sbin/init
   with environment:
     HOME=/
     TERM=linux
     BOOT_IMAGE=/boot/bzImage-517rc6
     kprobe_event=p,syscall_any,$arg1
     trace_options=quiet
     trace_clock=jiffies

Return 1 from the __setup() handlers so that init's environment is not
polluted with kernel boot options.

Fixes: 7bcfaf54f591 ("tracing: Add trace_options kernel command line parameter")
Fixes: e1e232ca6b8f ("tracing: Add trace_clock=<clock> kernel parameter")
Fixes: 970988e19eb0 ("tracing/kprobe: Add kprobe_event= boot parameter")
Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Reported-by: Igor Zhbanov <i.zhbanov@omprussia.ru>
Link: lore.kernel.org/r/64644a2f-4a20-bab3-1e15-3b2cdd0defe3@omprussia.ru
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Ingo Molnar <mingo@redhat.com>
---
"trace_clock=" reports invalid parameter usage later, when it tries
to use the value:
  Trace clock jiffies not defined, going back to default

 kernel/trace/trace.c        |    4 ++--
 kernel/trace/trace_kprobe.c |    2 +-
 2 files changed, 3 insertions(+), 3 deletions(-)

--- linux-next-20220302.orig/kernel/trace/trace.c
+++ linux-next-20220302/kernel/trace/trace.c
@@ -235,7 +235,7 @@ static char trace_boot_options_buf[MAX_T
 static int __init set_trace_boot_options(char *str)
 {
 	strlcpy(trace_boot_options_buf, str, MAX_TRACER_SIZE);
-	return 0;
+	return 1;
 }
 __setup("trace_options=", set_trace_boot_options);
 
@@ -246,7 +246,7 @@ static int __init set_trace_boot_clock(c
 {
 	strlcpy(trace_boot_clock_buf, str, MAX_TRACER_SIZE);
 	trace_boot_clock = trace_boot_clock_buf;
-	return 0;
+	return 1;
 }
 __setup("trace_clock=", set_trace_boot_clock);
 
--- linux-next-20220302.orig/kernel/trace/trace_kprobe.c
+++ linux-next-20220302/kernel/trace/trace_kprobe.c
@@ -32,7 +32,7 @@ static int __init set_kprobe_boot_events
 	strlcpy(kprobe_boot_events_buf, str, COMMAND_LINE_SIZE);
 	disable_tracing_selftest("running kprobe events");
 
-	return 0;
+	return 1;
 }
 __setup("kprobe_event=", set_kprobe_boot_events);
 

Re: [PATCH] trace: fix return value of __setup handlers

[Steven Rostedt]

On Wed,  2 Mar 2022 19:17:44 -0800
Randy Dunlap <rdunlap@infradead.org> wrote:

> __setup() handlers should generally return 1 to indicate that the
> boot options have been handled.
> 
> Using invalid option values causes the entire kernel boot option
> string to be reported as Unknown and added to init's environment
> strings, polluting it.
> 
>   Unknown kernel command line parameters "BOOT_IMAGE=/boot/bzImage-517rc6
>     kprobe_event=p,syscall_any,$arg1 trace_options=quiet
>     trace_clock=jiffies", will be passed to user space.
> 
>  Run /sbin/init as init process
>    with arguments:
>      /sbin/init
>    with environment:
>      HOME=/
>      TERM=linux
>      BOOT_IMAGE=/boot/bzImage-517rc6
>      kprobe_event=p,syscall_any,$arg1
>      trace_options=quiet
>      trace_clock=jiffies
> 
> Return 1 from the __setup() handlers so that init's environment is not
> polluted with kernel boot options.
> 
> Fixes: 7bcfaf54f591 ("tracing: Add trace_options kernel command line parameter")
> Fixes: e1e232ca6b8f ("tracing: Add trace_clock=<clock> kernel parameter")
> Fixes: 970988e19eb0 ("tracing/kprobe: Add kprobe_event= boot parameter")
> Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
> Reported-by: Igor Zhbanov <i.zhbanov@omprussia.ru>
> Link: lore.kernel.org/r/64644a2f-4a20-bab3-1e15-3b2cdd0defe3@omprussia.ru
> Cc: Steven Rostedt <rostedt@goodmis.org>
> Cc: Masami Hiramatsu <mhiramat@kernel.org>
> Cc: Ingo Molnar <mingo@redhat.com>
> ---
> "trace_clock=" reports invalid parameter usage later, when it tries
> to use the value:
>   Trace clock jiffies not defined, going back to default

That's because there's no "jiffies" clock. You want to try "trace_clock=uptime".

Thanks, I'll add this.

-- Steve

> 
>  kernel/trace/trace.c        |    4 ++--
>  kernel/trace/trace_kprobe.c |    2 +-
>  2 files changed, 3 insertions(+), 3 deletions(-)
> 
> --- linux-next-20220302.orig/kernel/trace/trace.c
> +++ linux-next-20220302/kernel/trace/trace.c
> @@ -235,7 +235,7 @@ static char trace_boot_options_buf[MAX_T
>  static int __init set_trace_boot_options(char *str)
>  {
>  	strlcpy(trace_boot_options_buf, str, MAX_TRACER_SIZE);
> -	return 0;
> +	return 1;
>  }
>  __setup("trace_options=", set_trace_boot_options);
>  
> @@ -246,7 +246,7 @@ static int __init set_trace_boot_clock(c
>  {
>  	strlcpy(trace_boot_clock_buf, str, MAX_TRACER_SIZE);
>  	trace_boot_clock = trace_boot_clock_buf;
> -	return 0;
> +	return 1;
>  }
>  __setup("trace_clock=", set_trace_boot_clock);
>  
> --- linux-next-20220302.orig/kernel/trace/trace_kprobe.c
> +++ linux-next-20220302/kernel/trace/trace_kprobe.c
> @@ -32,7 +32,7 @@ static int __init set_kprobe_boot_events
>  	strlcpy(kprobe_boot_events_buf, str, COMMAND_LINE_SIZE);
>  	disable_tracing_selftest("running kprobe events");
>  
> -	return 0;
> +	return 1;
>  }
>  __setup("kprobe_event=", set_kprobe_boot_events);
>  

Re: [PATCH] trace: fix return value of __setup handlers

[Masami Hiramatsu]

On Wed,  2 Mar 2022 19:17:44 -0800
Randy Dunlap <rdunlap@infradead.org> wrote:

> __setup() handlers should generally return 1 to indicate that the
> boot options have been handled.
> 
> Using invalid option values causes the entire kernel boot option
> string to be reported as Unknown and added to init's environment
> strings, polluting it.
> 
>   Unknown kernel command line parameters "BOOT_IMAGE=/boot/bzImage-517rc6
>     kprobe_event=p,syscall_any,$arg1 trace_options=quiet
>     trace_clock=jiffies", will be passed to user space.
> 
>  Run /sbin/init as init process
>    with arguments:
>      /sbin/init
>    with environment:
>      HOME=/
>      TERM=linux
>      BOOT_IMAGE=/boot/bzImage-517rc6
>      kprobe_event=p,syscall_any,$arg1
>      trace_options=quiet
>      trace_clock=jiffies
> 
> Return 1 from the __setup() handlers so that init's environment is not
> polluted with kernel boot options.

Oops, thanks for pointing it out.

Acked-by: Masami Hiramatsu <mhiramat@kernel.org>

> 
> Fixes: 7bcfaf54f591 ("tracing: Add trace_options kernel command line parameter")
> Fixes: e1e232ca6b8f ("tracing: Add trace_clock=<clock> kernel parameter")
> Fixes: 970988e19eb0 ("tracing/kprobe: Add kprobe_event= boot parameter")
> Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
> Reported-by: Igor Zhbanov <i.zhbanov@omprussia.ru>
> Link: lore.kernel.org/r/64644a2f-4a20-bab3-1e15-3b2cdd0defe3@omprussia.ru
> Cc: Steven Rostedt <rostedt@goodmis.org>
> Cc: Masami Hiramatsu <mhiramat@kernel.org>
> Cc: Ingo Molnar <mingo@redhat.com>
> ---
> "trace_clock=" reports invalid parameter usage later, when it tries
> to use the value:
>   Trace clock jiffies not defined, going back to default
> 
>  kernel/trace/trace.c        |    4 ++--
>  kernel/trace/trace_kprobe.c |    2 +-
>  2 files changed, 3 insertions(+), 3 deletions(-)
> 
> --- linux-next-20220302.orig/kernel/trace/trace.c
> +++ linux-next-20220302/kernel/trace/trace.c
> @@ -235,7 +235,7 @@ static char trace_boot_options_buf[MAX_T
>  static int __init set_trace_boot_options(char *str)
>  {
>  	strlcpy(trace_boot_options_buf, str, MAX_TRACER_SIZE);
> -	return 0;
> +	return 1;
>  }
>  __setup("trace_options=", set_trace_boot_options);
>  
> @@ -246,7 +246,7 @@ static int __init set_trace_boot_clock(c
>  {
>  	strlcpy(trace_boot_clock_buf, str, MAX_TRACER_SIZE);
>  	trace_boot_clock = trace_boot_clock_buf;
> -	return 0;
> +	return 1;
>  }
>  __setup("trace_clock=", set_trace_boot_clock);
>  
> --- linux-next-20220302.orig/kernel/trace/trace_kprobe.c
> +++ linux-next-20220302/kernel/trace/trace_kprobe.c
> @@ -32,7 +32,7 @@ static int __init set_kprobe_boot_events
>  	strlcpy(kprobe_boot_events_buf, str, COMMAND_LINE_SIZE);
>  	disable_tracing_selftest("running kprobe events");
>  
> -	return 0;
> +	return 1;
>  }
>  __setup("kprobe_event=", set_kprobe_boot_events);
>  


-- 
Masami Hiramatsu <mhiramat@kernel.org>