[prev in list] [next in list] [prev in thread] [next in thread] 

List:       linux-tegra
Subject:    Re: [PATCH] soc/tegra: pmc: Don't allocate struct tegra_powergate on stack
From:       Jon Hunter <jonathanh () nvidia ! com>
Date:       2017-04-24 10:55:55
Message-ID: d947516f-7090-d539-6a10-f8348551fc0f () nvidia ! com
[Download RAW message or body]


On 17/04/17 06:50, Viresh Kumar wrote:
> On 21-03-17, 16:09, Viresh Kumar wrote:
> > On 21-03-17, 10:37, Jon Hunter wrote:
> > > 
> > > On 21/03/17 05:24, Viresh Kumar wrote:
> > > > The size of the struct tegra_powergate is quite big and if any more
> > > > fields are added to the internal genpd structure, following warnings are
> > > > thrown:
> > > > 
> > > > drivers/soc/tegra/pmc.c:577:1: warning: the frame size of 1176 bytes is \
> > > > larger than 1024 bytes [-Wframe-larger-than=]
> > > 
> > > Hmmm ... AFAICT the size of the tegra_powergate struct is 312 bytes
> > > (based upon next-20170321) and so it looks like something massive needs
> > > to be added to the genpd struct to blow this up to over 1024 bytes. Are
> > > there some genpd changes in-flight that are causing this?
> > 
> > https://marc.info/?l=linux-kernel&m=149000247329743&w=2
> > 
> > This is up for discussion right now though and we don't know if it
> > will surely get merged or not.
> 
> @Jon: Regardless of the above series, do you want this patch to be merged as it
> will still be better to avoid keeping large structures on stack.

Given that it is currently much less than the default threshold, it
seems ok to me as-is. However, if it looks like you patch to add the
device struct to the gpd struct is going to be accepted, then it is fine
with me. Maybe we should wait for you patch to be accepted then this can
be applied as a fix.

Jon

-- 
nvpublic
--
To unsubscribe from this list: send the line "unsubscribe linux-tegra" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html


[prev in list] [next in list] [prev in thread] [next in thread]