From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <vdavydov.dev@gmail.com>
Date: Wed, 13 Mar 2019 17:17:24 +0300
From: Vladimir Davydov <vdavydov.dev@gmail.com>
Subject: Re: [PATCH 3/3] lib/core/fiber: Put watermarks into stack to track
 its usage
Message-ID: <20190313141724.wgt5xt3ey2shmtlo@esperanza>
References: <20190312224721.8053-1-gorcunov@gmail.com>
 <20190312224721.8053-4-gorcunov@gmail.com>
 <20190313135204.x7gn54iqiy5ulavn@esperanza>
 <20190313140054.GO10420@uranus.lan>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20190313140054.GO10420@uranus.lan>
To: Cyrill Gorcunov <gorcunov@gmail.com>
Cc: tml <tarantool-patches@freelists.org>
List-ID: <tarantool-patches.dev.tarantool.org>

On Wed, Mar 13, 2019 at 05:00:54PM +0300, Cyrill Gorcunov wrote:
> > > +
> > > +	if (!stack_has_wmark(fiber->stack_overflow_wmark)) {
> > > +		say_warn("stack usage is close to the limit of %zu bytes",
> > > +			 (size_t)FIBER_STACK_SIZE_DEFAULT);
> > 
> > IMO warning only when we are close to the limit isn't very useful.
> > Let's keep track of the max stack size instead and print it whenever
> > it's increased.
> 
> Could you please elaborate. I don't understand what you mean by
> keep tracking of max stack size. You propose to remember somewhere
> if this mark has been vanished?

Yeah, keep track of the max recorded stack size in a static variable
and keep advancing the higher watermark position whenever it gets
overwritten so that we can report max stack size used:

  I> max stack size is 73728 bytes
  ...
  I> max stack size is 112640 bytes

(the wording might need polishing)

> > > @@ -927,8 +1090,12 @@ cord_create(struct cord *cord, const char *name)
> > >  	/* Record stack extents */
> > >  	tt_pthread_attr_getstack(cord->id, &cord->sched.stack,
> > >  				 &cord->sched.stack_size);
> > > +	cord->sched.stack_overflow_wmark = cord->sched.stack;
> > > +	cord->sched.stack_shrink_wmark = cord->sched.stack;
> > 
> > I don't think we need to set the watermarks for the sched fiber,
> > even if ASAN is on.
> 
> Won't they left unitialized then? The only reason I put nils here
> is to force them have known values even if they are unused.

I'd set them to NULLs for both cases, i.e. outside ifdef.