From mboxrd@z Thu Jan 1 00:00:00 1970 From: zhangfei Subject: Re: [PATCH v2 3/3] mmc: dw_mmc: k3: support hi6220 Date: Thu, 14 May 2015 13:52:07 +0800 Message-ID: <55543807.5010700@linaro.org> References: <1430985401-12202-1-git-send-email-zhangfei.gao@linaro.org> <1430985401-12202-4-git-send-email-zhangfei.gao@linaro.org> <5553CD68.8000101@samsung.com> Mime-Version: 1.0 Content-Type: text/plain; charset=windows-1252; format=flowed Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <5553CD68.8000101@samsung.com> Sender: linux-mmc-owner@vger.kernel.org To: Jaehoon Chung , ulf.hansson@linaro.org Cc: linux-mmc@vger.kernel.org, devicetree@vger.kernel.org, "Jorge A. Ramirez-Ortiz" , Dan Yuan List-Id: devicetree@vger.kernel.org On 05/14/2015 06:17 AM, Jaehoon Chung wrote: > On 05/07/2015 04:56 PM, Zhangfei Gao wrote: >> Support hi6220, tested on hikey board >> emmc: support hs >> sd: support hs, sdr12, sdr25 >> >> Signed-off-by: Jorge A. Ramirez-Ortiz >> Signed-off-by: Dan Yuan >> Signed-off-by: Zhangfei Gao >> --- >> drivers/mmc/host/dw_mmc-k3.c | 101 ++++++++++++++++++++++++++++++++++++++++++- >> 1 file changed, 99 insertions(+), 2 deletions(-) >> >> diff --git a/drivers/mmc/host/dw_mmc-k3.c b/drivers/mmc/host/dw_mmc-k3.c >> index 650f9cc..bc16bee 100644 >> --- a/drivers/mmc/host/dw_mmc-k3.c >> +++ b/drivers/mmc/host/dw_mmc-k3.c >> @@ -8,16 +8,28 @@ >> * (at your option) any later version. >> */ >> >> -#include >> -#include >> #include >> +#include >> #include >> #include >> +#include >> #include >> +#include >> +#include >> +#include >> >> #include "dw_mmc.h" >> #include "dw_mmc-pltfm.h" >> >> +#define AO_SCTRL_SEL18 BIT(10) >> +#define AO_SCTRL_CTRL3 0x40C > > how about adding the comment? Sure > >> + >> +#define SDMMC_CMD_DISABLE_BOOT BIT(26) > > I can't find where this define is used. My bad, the usage is removed for simple, will remove this as well > >> + >> +struct k3_priv { >> + struct regmap *reg; >> +}; >> + >> static void dw_mci_k3_set_ios(struct dw_mci *host, struct mmc_ios *ios) >> { >> int ret; >> @@ -33,8 +45,93 @@ static const struct dw_mci_drv_data k3_drv_data = { >> .set_ios = dw_mci_k3_set_ios, >> }; >> >> +static int dw_mci_hi6220_parse_dt(struct dw_mci *host) >> +{ >> + struct k3_priv *priv; >> + >> + priv = devm_kzalloc(host->dev, sizeof(*priv), GFP_KERNEL); >> + if (!priv) >> + return -ENOMEM; >> + >> + priv->reg = syscon_regmap_lookup_by_phandle(host->dev->of_node, >> + "hisilicon,peripheral-syscon"); >> + if (IS_ERR(priv->reg)) >> + priv->reg = NULL; >> + >> + host->priv = priv; >> + return 0; >> +} >> + >> +static int dw_mci_hi6220_switch_voltage(struct mmc_host *mmc, struct mmc_ios *ios) >> +{ >> + struct dw_mci_slot *slot = mmc_priv(mmc); >> + struct k3_priv *priv; >> + struct dw_mci *host; >> + int min_uv, max_uv; >> + int ret; >> + >> + host = slot->host; >> + priv = host->priv; >> + >> + if (!priv || !priv->reg) >> + return 0; > > Is it right that return to 0? Since only sd need voltage switch and the register AO_SCTRL only takes care of sd. While emmc directly use 1.8v single voltage, so directly return 0. > >> + >> + if (ios->signal_voltage == MMC_SIGNAL_VOLTAGE_330) { >> + ret = regmap_update_bits(priv->reg, AO_SCTRL_CTRL3, >> + AO_SCTRL_SEL18, 0); >> + min_uv = 3000000; >> + max_uv = 3000000; >> + } else if (ios->signal_voltage == MMC_SIGNAL_VOLTAGE_180) { >> + ret = regmap_update_bits(priv->reg, AO_SCTRL_CTRL3, >> + AO_SCTRL_SEL18, AO_SCTRL_SEL18); >> + min_uv = 1800000; >> + max_uv = 1800000; > > max_uv and min_uv are fixed as 3.0v and 1.8V? Double confirmed with hisilicon, their controller has to be set 3v or 1.8v for io voltage. Otherwise it can not work. Thanks