[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20201008075306.6j7sv3d4prn6lzed@linutronix.de>
Date:   Thu, 8 Oct 2020 09:53:06 +0200
From:   Sebastian Andrzej Siewior <bigeasy@...utronix.de>
To:     Maninder Singh <maninder1.s@...sung.com>
Cc:     linux@...linux.org.uk, ndesaulniers@...gle.com, caij2003@...il.com,
        tglx@...utronix.de, maz@...nel.org, valentin.schneider@....com,
        vincent.whitchurch@...s.com, nhuck@...gle.com,
        akpm@...ux-foundation.org, 0x7f454c46@...il.com, will@...nel.org,
        a.sahrawat@...sung.com, v.narang@...sung.com,
        linux-arm-kernel@...ts.infradead.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH 0/3] IRQ stack support for ARM
On 2020-10-08 12:45:30 [+0530], Maninder Singh wrote:
> Observed Stack Overflow on 8KB kernel stack on ARM specially 
> incase on network interrupts, which results in undeterministic behaviour. 
> So there is need for per cpu dedicated IRQ stack for ARM.
You could try to look where this stack overflow is coming from. If this
is limited to softirq processing/NAPI (since you mentioned network) you
could try implementing do_softirq_own_stack().
Sebastian
Powered by blists - more mailing lists