[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <78cc83d10cb6d251e53f5947ddc52735@kernel.org>
Date: Wed, 01 Dec 2021 07:56:29 +0000
From: Marc Zyngier <maz@...nel.org>
To: Vladimir Murzin <vladimir.murzin@....com>
Cc: Mark Rutland <mark.rutland@....com>, linux-kernel@...r.kernel.org,
aou@...s.berkeley.edu, catalin.marinas@....com,
deanbo422@...il.com, green.hu@...il.com, guoren@...nel.org,
jonas@...thpole.se, kernelfans@...il.com,
linux-arm-kernel@...ts.infradead.org, linux@...linux.org.uk,
nickhu@...estech.com, palmer@...belt.com, paulmck@...nel.org,
paul.walmsley@...ive.com, peterz@...radead.org, shorne@...il.com,
stefan.kristiansson@...nalahti.fi, tglx@...utronix.de,
torvalds@...ux-foundation.org, tsbogend@...ha.franken.de,
vgupta@...nel.org, will@...nel.org
Subject: Re: [PATCH 09/15] irq: arm: perform irqentry in entry code
On 2021-11-30 08:49, Vladimir Murzin wrote:
> One month and a week later...
>
> I observe that in terms of performance
>
> MRS r0, ipsr
>
> is equivalent to readl_relaxed(BASEADDR_V7M_SCB + V7M_SCB_ICSR)
>
> MOV.W r3, #3758153728
> LDR.W r0, [r3, #3332]
>
> Old compilers can produce less performant sequence like
>
> LDR r3,0xbcc0
> ADD.W r3,r3,#0xaf00
> LDR r0,[r3,#0]
>
> So, what would be your preference?
If there is no significant overhead to reading the MMIO register
and that you see a benefit in enabling COMPILE_TEST, then this
probably is the way to go.
Thanks,
M.
--
Jazz is not dead. It just smells funny...
Powered by blists - more mailing lists