From patchwork Thu Sep 26 17:20:59 2013 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Guennadi Liakhovetski X-Patchwork-Id: 2949811 Return-Path: X-Original-To: patchwork-linux-sh@patchwork.kernel.org Delivered-To: patchwork-parsemail@patchwork2.web.kernel.org Received: from mail.kernel.org (mail.kernel.org [198.145.19.201]) by patchwork2.web.kernel.org (Postfix) with ESMTP id 40D81BFF0B for ; Thu, 26 Sep 2013 17:22:14 +0000 (UTC) Received: from mail.kernel.org (localhost [127.0.0.1]) by mail.kernel.org (Postfix) with ESMTP id CCBF82025B for ; Thu, 26 Sep 2013 17:22:12 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id 4D8132028D for ; Thu, 26 Sep 2013 17:22:11 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753610Ab3IZRWK (ORCPT ); Thu, 26 Sep 2013 13:22:10 -0400 Received: from moutng.kundenserver.de ([212.227.126.186]:51747 "EHLO moutng.kundenserver.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753557Ab3IZRVH (ORCPT ); Thu, 26 Sep 2013 13:21:07 -0400 Received: from axis700.grange (dslb-094-221-123-102.pools.arcor-ip.net [94.221.123.102]) by mrelayeu.kundenserver.de (node=mreu1) with ESMTP (Nemesis) id 0MhgyF-1VCp1q1Knj-00Mcn2; Thu, 26 Sep 2013 19:21:01 +0200 Received: from 6a.grange (6a.grange [192.168.1.11]) by axis700.grange (Postfix) with ESMTPS id D44DF40BB6; Thu, 26 Sep 2013 19:21:00 +0200 (CEST) Received: from lyakh by 6a.grange with local (Exim 4.72) (envelope-from ) id 1VPFFM-0003ma-FS; Thu, 26 Sep 2013 19:21:00 +0200 From: Guennadi Liakhovetski To: linux-kernel@vger.kernel.org Cc: Magnus Damm , linux-sh@vger.kernel.org, Linus Walleij , Laurent Pinchart , Simon Horman , Guennadi Liakhovetski Subject: [PATCH v2 4/5] ARM: shmobile: r8a7790: add CPUFreq clock support Date: Thu, 26 Sep 2013 19:20:59 +0200 Message-Id: <1380216060-14506-5-git-send-email-g.liakhovetski@gmx.de> X-Mailer: git-send-email 1.7.2.5 In-Reply-To: <1380216060-14506-1-git-send-email-g.liakhovetski@gmx.de> References: <1380216060-14506-1-git-send-email-g.liakhovetski@gmx.de> X-Provags-ID: V02:K0:S6zIWDeGkbPQ8wW+ZIn8Udf+6nuCwHMoXrO2obmNdem gpALTrIF+vrDXPkt8pmonlQSX0HsKN7TVo/M9DDYGV6dEgDGyO LDIhWMMhiIrcAJo/4yrfv+NhHGoKZJ/pb3YZLij81b9bRa8ALs pumG+bxUksnARtc3PZY6HkzR7yhF5ZQViVwfMVz2pBv31pJ1Z5 CZAyBPU478ywTmDdJFPs9mpi5dE8dCDG+43HDhEmEHlrsC6aZM Fz4EtsxD5DMfkalpWzhp4WwLoYw44PV+8T290BvH4RA+o2pt5Q 9c8oGzUgK5zmwXBysXTR61XG00kBcj5qcY6amL2aBGKkivb/Bu 2Oh6eoSkuG/uks4w4W5/yPQPqyIWfS6fEmCS+bVOzzz8gPXMpi pgU5mF7gB+W/g== Sender: linux-sh-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-sh@vger.kernel.org X-Spam-Status: No, score=-9.3 required=5.0 tests=BAYES_00,FREEMAIL_FROM, RCVD_IN_DNSWL_HI, RP_MATCHES_RCVD, UNPARSEABLE_RELAY autolearn=ham version=3.3.1 X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on mail.kernel.org X-Virus-Scanned: ClamAV using ClamSMTP Add support for the Z clock on r8a7790, driving the four SoC's CA15 cores, and its parent - PLL0. This is required for CPUFreq support on this SoC. Signed-off-by: Guennadi Liakhovetski --- v2: changed clock alias to "cpu0" instead of deprecated "cpufreq-cpu0" to comply with changes, introduced in patches from Sudeep KarkadaNagesha cpufreq: cpufreq-cpu0: assign cpu_dev correctly to cpu0 device ARM: shmobile: change dev_id to cpu0 while registering cpu clock arch/arm/mach-shmobile/Kconfig | 2 + arch/arm/mach-shmobile/clock-r8a7790.c | 150 ++++++++++++++++++++++++++++++++ 2 files changed, 152 insertions(+), 0 deletions(-) diff --git a/arch/arm/mach-shmobile/Kconfig b/arch/arm/mach-shmobile/Kconfig index eda2857..7f08bca 100644 --- a/arch/arm/mach-shmobile/Kconfig +++ b/arch/arm/mach-shmobile/Kconfig @@ -100,6 +100,8 @@ config ARCH_R8A7790 select CPU_V7 select SH_CLK_CPG select RENESAS_IRQC + select ARCH_HAS_CPUFREQ + select ARCH_HAS_OPP config ARCH_R8A7791 bool "R-Car M2 (R8A77910)" diff --git a/arch/arm/mach-shmobile/clock-r8a7790.c b/arch/arm/mach-shmobile/clock-r8a7790.c index a64f965..062d997 100644 --- a/arch/arm/mach-shmobile/clock-r8a7790.c +++ b/arch/arm/mach-shmobile/clock-r8a7790.c @@ -54,9 +54,12 @@ #define SMSTPCR8 0xe6150990 #define SMSTPCR9 0xe6150994 +#define FRQCRB 0xE6150004 #define SDCKCR 0xE6150074 #define SD2CKCR 0xE6150078 #define SD3CKCR 0xE615007C +#define FRQCRC 0xE61500E0 +#define PLLECR 0xE61500D0 #define MMC0CKCR 0xE6150240 #define MMC1CKCR 0xE6150244 #define SSPCKCR 0xE6150248 @@ -85,6 +88,7 @@ static struct clk main_clk = { * clock ratio of these clock will be updated * on r8a7790_clock_init() */ +SH_FIXED_RATIO_CLK_SET(pll0_clk, main_clk, 1, 1); SH_FIXED_RATIO_CLK_SET(pll1_clk, main_clk, 1, 1); SH_FIXED_RATIO_CLK_SET(pll3_clk, main_clk, 1, 1); SH_FIXED_RATIO_CLK_SET(lb_clk, pll1_clk, 1, 1); @@ -113,15 +117,155 @@ SH_FIXED_RATIO_CLK_SET(zb3d2_clk, pll3_clk, 1, 8); SH_FIXED_RATIO_CLK_SET(ddr_clk, pll3_clk, 1, 8); SH_FIXED_RATIO_CLK_SET(mp_clk, pll1_div2_clk, 1, 15); +/* Locking not needed yet, only one clock is using FRQCR[BC] divisors so far */ +static atomic_t frqcr_lock; +#define CPG_MAP(o) ((o) - CPG_BASE + cpg_mapping.base) + +/* Several clocks need to access FRQCRB, have to lock */ +static bool frqcr_kick_check(struct clk *clk) +{ + return !(ioread32(CPG_MAP(FRQCRB)) & BIT(31)); +} + +static int frqcr_kick_do(struct clk *clk) +{ + int i; + + /* set KICK bit in FRQCRB to update hardware setting, check success */ + iowrite32(ioread32(CPG_MAP(FRQCRB)) | BIT(31), CPG_MAP(FRQCRB)); + for (i = 1000; i; i--) + if (ioread32(CPG_MAP(FRQCRB)) & BIT(31)) + cpu_relax(); + else + return 0; + + return -ETIMEDOUT; +} + +static int zclk_set_rate(struct clk *clk, unsigned long rate) +{ + void __iomem *frqcrc; + int ret; + unsigned long step, p_rate; + u32 val; + + if (!clk->parent || !__clk_get(clk->parent)) + return -ENODEV; + + if (!atomic_inc_and_test(&frqcr_lock) || !frqcr_kick_check(clk)) { + ret = -EBUSY; + goto done; + } + + /* + * Users are supposed to first call clk_set_rate() only with + * clk_round_rate() results. So, we don't fix wrong rates here, but + * guard against them anyway + */ + + p_rate = clk_get_rate(clk->parent); + if (rate == p_rate) { + val = 0; + } else { + step = DIV_ROUND_CLOSEST(p_rate, 32); + + if (rate > p_rate || rate < step) { + ret = -EINVAL; + goto done; + } + + val = 32 - rate / step; + } + + frqcrc = clk->mapped_reg + (FRQCRC - (u32)clk->enable_reg); + + iowrite32((ioread32(frqcrc) & ~(clk->div_mask << clk->enable_bit)) | + (val << clk->enable_bit), frqcrc); + + ret = frqcr_kick_do(clk); + +done: + atomic_dec(&frqcr_lock); + __clk_put(clk->parent); + return ret; +} + +static long zclk_round_rate(struct clk *clk, unsigned long rate) +{ + /* + * theoretical rate = parent rate * multiplier / 32, + * where 1 <= multiplier <= 32. Therefore we should do + * multiplier = rate * 32 / parent rate + * rounded rate = parent rate * multiplier / 32. + * However, multiplication before division won't fit in 32 bits, so + * we sacrifice some precision by first dividing and then multiplying. + * To find the nearest divisor we calculate both and pick up the best + * one. This avoids 64-bit arithmetics. + */ + unsigned long step, mul_min, mul_max, rate_min, rate_max; + + rate_max = clk_get_rate(clk->parent); + + /* output freq <= parent */ + if (rate >= rate_max) + return rate_max; + + step = DIV_ROUND_CLOSEST(rate_max, 32); + /* output freq >= parent / 32 */ + if (step >= rate) + return step; + + mul_min = rate / step; + mul_max = DIV_ROUND_UP(rate, step); + rate_min = step * mul_min; + if (mul_max == mul_min) + return rate_min; + + rate_max = step * mul_max; + + if (rate_max - rate < rate - rate_min) + return rate_max; + + return rate_min; +} + +static unsigned long zclk_recalc(struct clk *clk) +{ + void __iomem *frqcrc = FRQCRC - (u32)clk->enable_reg + clk->mapped_reg; + unsigned int max = clk->div_mask + 1; + unsigned long val = ((ioread32(frqcrc) >> clk->enable_bit) & + clk->div_mask); + + return DIV_ROUND_CLOSEST(clk_get_rate(clk->parent), max) * + (max - val); +} + +static struct sh_clk_ops zclk_ops = { + .recalc = zclk_recalc, + .set_rate = zclk_set_rate, + .round_rate = zclk_round_rate, +}; + +static struct clk z_clk = { + .parent = &pll0_clk, + .div_mask = 0x1f, + .enable_bit = 8, + /* We'll need to access FRQCRB and FRQCRC */ + .enable_reg = (void __iomem *)FRQCRB, + .ops = &zclk_ops, +}; + static struct clk *main_clks[] = { &extal_clk, &extal_div2_clk, &main_clk, + &pll0_clk, &pll1_clk, &pll1_div2_clk, &pll3_clk, &lb_clk, &qspi_clk, + &z_clk, &zg_clk, &zx_clk, &zs_clk, @@ -254,6 +398,9 @@ static struct clk_lookup lookups[] = { CLKDEV_CON_ID("qspi", &qspi_clk), CLKDEV_CON_ID("cp", &cp_clk), + /* CPU clock */ + CLKDEV_DEV_ID("cpu0", &z_clk), + /* DIV4 */ CLKDEV_CON_ID("sdh", &div4_clks[DIV4_SDH]), @@ -301,6 +448,7 @@ static struct clk_lookup lookups[] = { #define R8A7790_CLOCK_ROOT(e, m, p0, p1, p30, p31) \ extal_clk.rate = e * 1000 * 1000; \ main_clk.parent = m; \ + SH_CLK_SET_RATIO(&pll0_clk_ratio, p0 / 2, 1); \ SH_CLK_SET_RATIO(&pll1_clk_ratio, p1 / 2, 1); \ if (mode & MD(19)) \ SH_CLK_SET_RATIO(&pll3_clk_ratio, p31, 1); \ @@ -313,6 +461,8 @@ void __init r8a7790_clock_init(void) u32 mode = rcar_gen2_read_mode_pins(); int k, ret = 0; + atomic_set(&frqcr_lock, -1); + switch (mode & (MD(14) | MD(13))) { case 0: R8A7790_CLOCK_ROOT(15, &extal_clk, 172, 208, 106, 88);