From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756469Ab1EBBVq (ORCPT ); Sun, 1 May 2011 21:21:46 -0400 Received: from smtp-out.google.com ([216.239.44.51]:54582 "EHLO smtp-out.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753997Ab1EBBT5 (ORCPT ); Sun, 1 May 2011 21:19:57 -0400 DomainKey-Signature: a=rsa-sha1; s=beta; d=google.com; c=nofws; q=dns; h=from:to:cc:subject:date:message-id:x-mailer:in-reply-to:references; b=XRlkQnoKJRHq4cKpxM2Ad0AyqoSL41Y0kxqeDQ8q6i8/uxtPJqUuECklu9CpI4NdH lIBURAntJ1K1D33j0N7gg== From: Nikhil Rao To: Ingo Molnar , Peter Zijlstra , Mike Galbraith Cc: linux-kernel@vger.kernel.org, "Nikunj A. Dadhania" , Srivatsa Vaddagiri , Stephan Barwolf , Nikhil Rao Subject: [PATCH v1 10/19] sched: update wake_affine path to use u64, s64 for weights Date: Sun, 1 May 2011 18:19:08 -0700 Message-Id: <1304299157-25769-11-git-send-email-ncrao@google.com> X-Mailer: git-send-email 1.7.3.1 In-Reply-To: <1304299157-25769-1-git-send-email-ncrao@google.com> References: <1304299157-25769-1-git-send-email-ncrao@google.com> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Update the s64/u64 math in wake_affine() and effective_load() to handle increased resolution. Signed-off-by: Nikhil Rao --- kernel/sched_fair.c | 8 ++++---- 1 files changed, 4 insertions(+), 4 deletions(-) diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index 49e1eeb..1e011b1 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -1388,7 +1388,7 @@ static void task_waking_fair(struct rq *rq, struct task_struct *p) * of group shares between cpus. Assuming the shares were perfectly aligned one * can calculate the shift in shares. */ -static long effective_load(struct task_group *tg, int cpu, long wl, long wg) +static s64 effective_load(struct task_group *tg, int cpu, s64 wl, s64 wg) { struct sched_entity *se = tg->se[cpu]; @@ -1396,7 +1396,7 @@ static long effective_load(struct task_group *tg, int cpu, long wl, long wg) return wl; for_each_sched_entity(se) { - long lw, w; + s64 lw, w; tg = se->my_q->tg; w = se->my_q->load.weight; @@ -1409,7 +1409,7 @@ static long effective_load(struct task_group *tg, int cpu, long wl, long wg) wl += w; if (lw > 0 && wl < lw) - wl = (wl * tg->shares) / lw; + wl = div64_s64(wl * tg->shares, lw); else wl = tg->shares; @@ -1504,7 +1504,7 @@ static int wake_affine(struct sched_domain *sd, struct task_struct *p, int sync) if (balanced || (this_load <= load && - this_load + target_load(prev_cpu, idx) <= tl_per_task)) { + this_load + target_load(prev_cpu, idx) <= (u64)tl_per_task)) { /* * This domain has SD_WAKE_AFFINE and * p is cache cold in this domain, and -- 1.7.3.1