From mboxrd@z Thu Jan  1 00:00:00 1970
From: tglx@linutronix.de (Thomas Gleixner)
Date: Thu, 24 Feb 2011 20:56:45 +0100 (CET)
Subject: [PATCH 4/4] msm: scm: Get cacheline size from CTR
In-Reply-To: <4D66B505.6010702@codeaurora.org>
References: <1298573085-23217-1-git-send-email-sboyd@codeaurora.org>
	<1298573085-23217-5-git-send-email-sboyd@codeaurora.org>
	<alpine.LFD.2.00.1102241959290.2701@localhost6.localdomain6>
	<4D66B505.6010702@codeaurora.org>
Message-ID: <alpine.LFD.2.00.1102242056010.2701@localhost6.localdomain6>
To: linux-arm-kernel@lists.infradead.org
List-Id: linux-arm-kernel.lists.infradead.org

On Thu, 24 Feb 2011, Stephen Boyd wrote:

> On 02/24/2011 11:01 AM, Thomas Gleixner wrote:
> > On Thu, 24 Feb 2011, Stephen Boyd wrote:
> >
> >>
> >>  /**
> >>   * scm_call() - Send an SCM command
> >>   * @svc_id: service identifier
> >> @@ -243,11 +248,13 @@ int scm_call(u32 svc_id, u32 cmd_id, const void *cmd_buf, size_t cmd_len,
> >>  	do {
> >>  		u32 start = (u32)rsp;
> >>  		u32 end = (u32)scm_get_response_buffer(rsp) + resp_len;
> >> -		start &= ~(CACHELINESIZE - 1);
> >> +		u32 cacheline_size = dcache_line_size();
> >
> > And why do you want to do that on every scm_call() invocation and on
> > every loop of that code? If your dcache_line_size() changes at
> > runtime, then you might have other problems.
> 
> I definitely don't want to do it for every loop. I'm fine with getting
> it every scm_call() invocation though.
> 
> For now, I'll pull the end and cacheline_size variables out of the
> do-while loop.

Why not do it correct right away and retrieve it in an __init
function?

Thanks,

	tglx