From patchwork Thu Mar 12 18:28:08 2009 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Kevin Hilman X-Patchwork-Id: 11489 Received: from vger.kernel.org (vger.kernel.org [209.132.176.167]) by demeter.kernel.org (8.14.2/8.14.2) with ESMTP id n2CIWtgt013343 for ; Thu, 12 Mar 2009 18:32:57 GMT Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1757208AbZCLScu (ORCPT ); Thu, 12 Mar 2009 14:32:50 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1753497AbZCLScu (ORCPT ); Thu, 12 Mar 2009 14:32:50 -0400 Received: from mail-gx0-f163.google.com ([209.85.217.163]:60567 "EHLO mail-gx0-f163.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757255AbZCLScs (ORCPT ); Thu, 12 Mar 2009 14:32:48 -0400 Received: by mail-gx0-f163.google.com with SMTP id 7so458514gxk.13 for ; Thu, 12 Mar 2009 11:32:46 -0700 (PDT) Received: by 10.142.116.12 with SMTP id o12mr109307wfc.332.1236882765949; Thu, 12 Mar 2009 11:32:45 -0700 (PDT) Received: from localhost ([216.254.16.51]) by mx.google.com with ESMTPS id 30sm2081663wff.47.2009.03.12.11.32.43 (version=TLSv1/SSLv3 cipher=RC4-MD5); Thu, 12 Mar 2009 11:32:45 -0700 (PDT) From: Kevin Hilman To: linux-omap@vger.kernel.org Subject: [PATCH 58/86] [ARM] OMAP3 clock: optimize DPLL rate rounding algorithm Date: Thu, 12 Mar 2009 11:28:08 -0700 Message-Id: <1236882516-29403-59-git-send-email-khilman@deeprootsystems.com> X-Mailer: git-send-email 1.6.1.2 In-Reply-To: <1236882516-29403-58-git-send-email-khilman@deeprootsystems.com> References: <1236882516-29403-1-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-2-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-3-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-4-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-5-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-6-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-7-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-8-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-9-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-10-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-11-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-12-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-13-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-14-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-15-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-16-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-17-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-18-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-19-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-20-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-21-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-22-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-23-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-24-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-25-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-26-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-27-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-28-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-29-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-30-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-31-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-32-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-33-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-34-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-35-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-36-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-37-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-38-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-39-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-40-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-41-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-42-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-43-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-44-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-45-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-46-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-47-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-48-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-49-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-50-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-51-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-52-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-53-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-54-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-55-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-56-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-57-git-send-email-khilman@deeprootsystems.com> <1236882516-29403-58-git-send-email-khilman@deeprootsystems.com> Sender: linux-omap-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-omap@vger.kernel.org From: Paul Walmsley The previous DPLL rate rounding algorithm counted the divider (N) down from the maximum to 1. Since we currently use a broad DPLL rate tolerance, and lower N values are more power-efficient, we can often bypass several iterations through the loop by counting N upwards from 1. Peter de Schrijver put up with several test cycles of this patch - thanks Peter. linux-omap source commit is 6f6d82bb2f80fa20a841ac3e95a6f44a5a156188. Signed-off-by: Paul Walmsley Cc: Peter de Schrijver Signed-off-by: Tony Lindgren Signed-off-by: Russell King --- arch/arm/mach-omap2/clock.c | 35 ++++++++++++++++++----------------- 1 files changed, 18 insertions(+), 17 deletions(-) diff --git a/arch/arm/mach-omap2/clock.c b/arch/arm/mach-omap2/clock.c index 2f0eaaa..76e20bc 100644 --- a/arch/arm/mach-omap2/clock.c +++ b/arch/arm/mach-omap2/clock.c @@ -45,7 +45,7 @@ #define DPLL_MIN_DIVIDER 1 /* Possible error results from _dpll_test_mult */ -#define DPLL_MULT_UNDERFLOW (1 << 0) +#define DPLL_MULT_UNDERFLOW -1 /* * Scale factor to mitigate roundoff errors in DPLL rate rounding. @@ -826,7 +826,7 @@ static int _dpll_test_mult(int *m, int n, unsigned long *new_rate, unsigned long target_rate, unsigned long parent_rate) { - int flags = 0, carry = 0; + int r = 0, carry = 0; /* Unscale m and round if necessary */ if (*m % DPLL_SCALE_FACTOR >= DPLL_ROUNDING_VAL) @@ -847,13 +847,13 @@ static int _dpll_test_mult(int *m, int n, unsigned long *new_rate, if (*m < DPLL_MIN_MULTIPLIER) { *m = DPLL_MIN_MULTIPLIER; *new_rate = 0; - flags = DPLL_MULT_UNDERFLOW; + r = DPLL_MULT_UNDERFLOW; } if (*new_rate == 0) *new_rate = _dpll_compute_new_rate(parent_rate, *m, n); - return flags; + return r; } /** @@ -892,21 +892,27 @@ long omap2_dpll_round_rate(struct clk *clk, unsigned long target_rate) dd->last_rounded_rate = 0; - for (n = dd->max_divider; n >= DPLL_MIN_DIVIDER; n--) { + for (n = DPLL_MIN_DIVIDER; n <= dd->max_divider; n++) { /* Compute the scaled DPLL multiplier, based on the divider */ m = scaled_rt_rp * n; /* - * Since we're counting n down, a m overflow means we can - * can immediately skip to the next n + * Since we're counting n up, a m overflow means we + * can bail out completely (since as n increases in + * the next iteration, there's no way that m can + * increase beyond the current m) */ if (m > scaled_max_m) - continue; + break; r = _dpll_test_mult(&m, n, &new_rate, target_rate, clk->parent->rate); + /* m can't be set low enough for this n - try with a larger n */ + if (r == DPLL_MULT_UNDERFLOW) + continue; + e = target_rate - new_rate; pr_debug("clock: n = %d: m = %d: rate error is %d " "(new_rate = %ld)\n", n, m, e, new_rate); @@ -918,16 +924,11 @@ long omap2_dpll_round_rate(struct clk *clk, unsigned long target_rate) min_e_n = n; pr_debug("clock: found new least error %d\n", min_e); - } - /* - * Since we're counting n down, a m underflow means we - * can bail out completely (since as n decreases in - * the next iteration, there's no way that m can - * increase beyond the current m) - */ - if (r & DPLL_MULT_UNDERFLOW) - break; + /* We found good settings -- bail out now */ + if (min_e <= clk->dpll_data->rate_tolerance) + break; + } } if (min_e < 0) {