[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20180309134439.76b43chxnrw7okrw@lakrids.cambridge.arm.com>
Date: Fri, 9 Mar 2018 13:44:40 +0000
From: Mark Rutland <mark.rutland@....com>
To: Shanker Donthineni <shankerd@...eaurora.org>
Cc: Will Deacon <will.deacon@....com>,
Robin Murphy <robin.murphy@....com>,
linux-kernel <linux-kernel@...r.kernel.org>,
linux-arm-kernel <linux-arm-kernel@...ts.infradead.org>,
Catalin Marinas <catalin.marinas@....com>,
kvmarm <kvmarm@...ts.cs.columbia.edu>,
Marc Zyngier <marc.zyngier@....com>,
Vikram Sethi <vikrams@...eaurora.org>,
Philip Elcan <pelcan@...eaurora.org>
Subject: Re: [PATCH v7] arm64: Add support for new control bits CTR_EL0.DIC
and CTR_EL0.IDC
On Wed, Mar 07, 2018 at 09:00:08AM -0600, Shanker Donthineni wrote:
> static inline void __flush_icache_all(void)
> {
> - asm("ic ialluis");
> - dsb(ish);
> + /* Instruction cache invalidation is not required for I/D coherence? */
> + if (!cpus_have_const_cap(ARM64_HAS_CACHE_DIC)) {
> + asm("ic ialluis");
> + dsb(ish);
> + }
> }
I don't think we need the comment here. We don't have this in the other
cases we look at the ARM64_HAS_CACHE_{IDC,DIC} caps.
This would also be slightly nicer as an early return:
static inline void __flush_icache_all(void)
{
if (cpus_have_const_cap(ARM64_HAS_CACHE_DIC))
return;
asm("ic ialluis");
dsb(ish);
}
... which minimizes indentation, and the diffstat.
The rest looks fine to me, so with the above changes:
Reviewed-by: Mark Rutland <mark.rutland@....com>
Thanks,
Mark.
Powered by blists - more mailing lists