Re: kmemleak: Early log buffer exceeded (525980) during boot

[Qian Cai]

On 11/10/18 at 11:59 AM, Catalin Marinas wrote:

> On Sat, Nov 10, 2018 at 10:08:10AM -0500, Qian Cai wrote:
> > On Nov 8, 2018, at 4:23 PM, Qian Cai <cai@xxxxxx> wrote:
> > > The maximum value for DEBUG_KMEMLEAK_EARLY_LOG_SIZE is only 40000, so it
> > > disables kmemleak every time on this aarch64 server running the latest mainline
> > > (b00d209).
> > > 
> > > # echo scan > /sys/kernel/debug/kmemleak 
> > > -bash: echo: write error: Device or resource busy
> > > 
> > > Any idea on how to enable kmemleak there?
> > 
> > I have managed to hard-code DEBUG_KMEMLEAK_EARLY_LOG_SIZE to 600000,
> 
> That's quite a high number, I wouldn't have thought it is needed.
> Basically the early log buffer is only used until the slub allocator
> gets initialised and kmemleak_init() is called from start_kernel(). I
> don't know what allocates that much memory so early.
> 
> What else is in your .config?
https://c.gmx.com/@642631272677512867/tqD5eulbQAC-1h-fkVe1Iw

Does the dmesg helps? 
https://paste.ubuntu.com/p/BnhvXXhn7k/
> 
> > diff --git a/mm/kmemleak.c b/mm/kmemleak.c
> > index 877de4fa0720..c10119102c10 100644
> > --- a/mm/kmemleak.c
> > +++ b/mm/kmemleak.c
> > @@ -280,7 +280,7 @@ struct early_log {
> >  
> >  /* early logging buffer and current position */
> >  static struct early_log
> > -       early_log[CONFIG_DEBUG_KMEMLEAK_EARLY_LOG_SIZE] __initdata;
> > +       early_log[600000] __initdata;
> 
> You don't need to patch the kernel, the config variable is there to be
> changed.
Right, but the maximum is only 40000 in kconfig, so anything bigger than that will be rejected.