[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20140822143117.29474ac2ac8092e27d26660a@linux-foundation.org>
Date: Fri, 22 Aug 2014 14:31:17 -0700
From: Andrew Morton <akpm@...ux-foundation.org>
To: Zhaoxiu Zeng <zhaoxiu.zeng@...il.com>
Cc: Ingo Molnar <mingo@...nel.org>, George Spelvin <linux@...izon.com>,
David Howells <dhowells@...hat.com>,
Peter Zijlstra <peterz@...radead.org>,
AKASHI Takahiro <takahiro.akashi@...aro.org>,
Josh Boyer <jwboyer@...hat.com>, linux-kernel@...r.kernel.org
Subject: Re: [PATCH 1/1] GCD: add binary GCD algorithm
On Fri, 15 Aug 2014 20:49:16 +0800 Zhaoxiu Zeng <zhaoxiu.zeng@...il.com> wrote:
> Because some architectures (alpha, armv6, etc.) don't provide hardware division,
> the mod operation is slow! Binary GCD algorithm uses simple arithmetic operations,
> it replaces division with arithmetic shifts, comparisons, and subtraction.
I had a look around and it seems that most (all?) gcd() and lcd()
callers are on initialization-time slowpaths.
Do you know of a workload which will significantly benefit from this
change? If so, by how much?
Otherwise I don't think we can justify the additional maintenance
cost/risk, sorry.
And if we *do* decide to proceed with this patch, we should include a
patch which enables it on as many architectures as possible, so it gets
runtime tested.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists