From: Matthias Brugger <matthias.bgg-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
To: weiyi.lu-NuS5LvNUpcJWk0Htik3J/w@public.gmane.org,
Stephen Boyd <sboyd-sgV2jX0FEOL9JmXXK+q4OQ@public.gmane.org>
Cc: James Liao <jamesjj.liao-NuS5LvNUpcJWk0Htik3J/w@public.gmane.org>,
Fan Chen <fan.chen-NuS5LvNUpcJWk0Htik3J/w@public.gmane.org>,
devicetree-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org,
linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
linux-mediatek-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org,
linux-clk-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
srv_heupstream-NuS5LvNUpcJWk0Htik3J/w@public.gmane.org
Subject: Re: [PATCH v1 6/9] soc: mediatek: add bus protection extend API
Date: Thu, 17 Aug 2017 12:57:44 +0200 [thread overview]
Message-ID: <a179cf17-0380-f47e-645e-92f38b0e1917@gmail.com> (raw)
In-Reply-To: <1502779370-30150-7-git-send-email-weiyi.lu-NuS5LvNUpcJWk0Htik3J/w@public.gmane.org>
On 08/15/2017 08:42 AM, weiyi.lu-NuS5LvNUpcJWk0Htik3J/w@public.gmane.org wrote:
> From: Weiyi Lu <weiyi.lu-NuS5LvNUpcJWk0Htik3J/w@public.gmane.org>
>
> MT2712 add "set/clear" bus control register to each control register set
> instead of providing only one "enable" control register, we could avoid
> the read-modify-write racing by using extend API with such new design.
> Also there exists the second bus control register set, we could not access
> this setting via original API rather than the extend API.
> By improving the mtk-infracfg bus protection implementation to
> support set/clear bus protection control method and second
> register set while power on/off scpsys.
>
> Signed-off-by: Weiyi Lu <weiyi.lu-NuS5LvNUpcJWk0Htik3J/w@public.gmane.org>
> ---
> drivers/soc/mediatek/mtk-infracfg.c | 90 +++++++++++++++++++++++++++++++++--
> drivers/soc/mediatek/mtk-scpsys.c | 36 ++++++++++++--
> include/linux/soc/mediatek/infracfg.h | 14 +++++-
> 3 files changed, 132 insertions(+), 8 deletions(-)
>
> diff --git a/drivers/soc/mediatek/mtk-infracfg.c b/drivers/soc/mediatek/mtk-infracfg.c
> index dba3055..63898e4 100644
> --- a/drivers/soc/mediatek/mtk-infracfg.c
> +++ b/drivers/soc/mediatek/mtk-infracfg.c
> @@ -17,9 +17,6 @@
> #include <linux/soc/mediatek/infracfg.h>
> #include <asm/processor.h>
>
> -#define INFRA_TOPAXI_PROTECTEN 0x0220
> -#define INFRA_TOPAXI_PROTECTSTA1 0x0228
> -
> /**
> * mtk_infracfg_set_bus_protection - enable bus protection
> * @regmap: The infracfg regmap
> @@ -89,3 +86,90 @@ int mtk_infracfg_clear_bus_protection(struct regmap *infracfg, u32 mask)
>
> return 0;
> }
> +
> +/**
> + * mtk_infracfg_set_bus_protection_ext - enable bus protection
> + * @regmap: The infracfg regmap
> + * @mask: The mask containing the protection bits to be enabled.
> + * @reg_set: The register used to enable protection bits.
> + * @reg_sta: The register used to check the protection bits are enabled.
> + * @reg_en: The register used to enable protection bits when there doesn't
> + * exist reg_set.
> + *
> + * This function enables the bus protection bits for disabled power
> + * domains so that the system does not hang when some unit accesses the
> + * bus while in power down.
> + */
> +int mtk_infracfg_set_bus_protection_ext(struct regmap *infracfg, u32 mask,
> + u32 reg_set, u32 reg_sta, u32 reg_en)
There is no need to add a new function. The only differences are:
1. it passes the register offsets instead of assuming default ones
2. it checks if reg_set/reg_clr and enables/disable.
We should change the original function instead of adding a new one.
From what I see, we are all done if we provide a boolean which tells infracfg
if it should use set/clear or update register.
For now we don't have any user of INFRA_TOPAXI_PROTECTSTA3 (same for
*PROTECTEN1*). Do you expect one in the near future? If so, we could pass a
flags value instead of bool which points that out. It seems there is no need for
now to pass the register value.
Regards,
Matthias
> +{
> + unsigned long expired;
> + u32 val;
> + int ret;
> +
> + if (reg_set)
> + regmap_write(infracfg, reg_set, mask);
> + else
> + regmap_update_bits(infracfg, reg_en, mask, mask);
> +
> + expired = jiffies + HZ;
> +
> + while (1) {
> + ret = regmap_read(infracfg, reg_sta, &val);
> + if (ret)
> + return ret;
> +
> + if ((val & mask) == mask)
> + break;
> +
> + cpu_relax();
> + if (time_after(jiffies, expired))
> + return -EIO;
> + }
> +
> + return 0;
> +}
> +
> +/**
> + * mtk_infracfg_clear_bus_protection_ext - disable bus protection
> + * @regmap: The infracfg regmap
> + * @mask: The mask containing the protection bits to be disabled.
> + * @reg_clr: The register used to disable protection bits.
> + * @reg_sta: The register used to check the protection bits are disabled.
> + * @reg_en: The register used to disable protection bits when there doesn't
> + * exist reg_clr.
> + *
> + * This function disables the bus protection bits previously enabled with
> + * mtk_infracfg_set_bus_protection.
> + */
> +
> +int mtk_infracfg_clear_bus_protection_ext(struct regmap *infracfg, u32 mask,
> + u32 reg_clr, u32 reg_sta, u32 reg_en)
> +{
> + unsigned long expired;
> + int ret;
> +
> + if (reg_clr)
> + regmap_write(infracfg, reg_clr, mask);
> + else
> + regmap_update_bits(infracfg, reg_en, mask, 0);
> +
> + expired = jiffies + HZ;
> +
> + while (1) {
> + u32 val;
> +
> + ret = regmap_read(infracfg, reg_sta, &val);
> + if (ret)
> + return ret;
> +
> + if (!(val & mask))
> + break;
> +
> + cpu_relax();
> + if (time_after(jiffies, expired))
> + return -EIO;
> + }
> +
> + return 0;
> +}
> diff --git a/drivers/soc/mediatek/mtk-scpsys.c b/drivers/soc/mediatek/mtk-scpsys.c
> index ceb2cc4..b181ef2 100644
> --- a/drivers/soc/mediatek/mtk-scpsys.c
> +++ b/drivers/soc/mediatek/mtk-scpsys.c
> @@ -89,6 +89,13 @@ enum clk_id {
>
> #define MAX_CLKS 2
>
> +struct bus_prot_ext {
> + u32 set_ofs;
> + u32 clr_ofs;
> + u32 en_ofs;
> + u32 sta_ofs;
> +};
> +
> struct scp_domain_data {
> const char *name;
> u32 sta_mask;
> @@ -96,6 +103,7 @@ struct scp_domain_data {
> u32 sram_pdn_bits;
> u32 sram_pdn_ack_bits;
> u32 bus_prot_mask;
> + struct bus_prot_ext bp_ext;
> enum clk_id clk_id[MAX_CLKS];
> bool active_wakeup;
> };
> @@ -228,8 +236,18 @@ static int scpsys_power_on(struct generic_pm_domain *genpd)
> }
>
> if (scpd->data->bus_prot_mask) {
> - ret = mtk_infracfg_clear_bus_protection(scp->infracfg,
> - scpd->data->bus_prot_mask);
> + if (((scpd->data->bp_ext.set_ofs && scpd->data->bp_ext.clr_ofs)
> + || scpd->data->bp_ext.en_ofs)
> + && scpd->data->bp_ext.sta_ofs)
> + ret = mtk_infracfg_clear_bus_protection_ext(
> + scp->infracfg,
> + scpd->data->bus_prot_mask,
> + scpd->data->bp_ext.clr_ofs,
> + scpd->data->bp_ext.sta_ofs,
> + scpd->data->bp_ext.en_ofs);
> + else
> + ret = mtk_infracfg_clear_bus_protection(scp->infracfg,
> + scpd->data->bus_prot_mask);
> if (ret)
> goto err_pwr_ack;
> }
> @@ -263,8 +281,18 @@ static int scpsys_power_off(struct generic_pm_domain *genpd)
> int i;
>
> if (scpd->data->bus_prot_mask) {
> - ret = mtk_infracfg_set_bus_protection(scp->infracfg,
> - scpd->data->bus_prot_mask);
> + if (((scpd->data->bp_ext.set_ofs && scpd->data->bp_ext.clr_ofs)
> + || scpd->data->bp_ext.en_ofs)
> + && scpd->data->bp_ext.sta_ofs)
> + ret = mtk_infracfg_set_bus_protection_ext(
> + scp->infracfg,
> + scpd->data->bus_prot_mask,
> + scpd->data->bp_ext.set_ofs,
> + scpd->data->bp_ext.sta_ofs,
> + scpd->data->bp_ext.en_ofs);
> + else
> + ret = mtk_infracfg_set_bus_protection(scp->infracfg,
> + scpd->data->bus_prot_mask);
> if (ret)
> goto out;
> }
> diff --git a/include/linux/soc/mediatek/infracfg.h b/include/linux/soc/mediatek/infracfg.h
> index a5714e9..d3fbb45 100644
> --- a/include/linux/soc/mediatek/infracfg.h
> +++ b/include/linux/soc/mediatek/infracfg.h
> @@ -1,6 +1,15 @@
> #ifndef __SOC_MEDIATEK_INFRACFG_H
> #define __SOC_MEDIATEK_INFRACFG_H
>
> +#define INFRA_TOPAXI_PROTECTEN 0x0220
> +#define INFRA_TOPAXI_PROTECTSTA1 0x0228
> +#define INFRA_TOPAXI_PROTECTEN1 0x0250
> +#define INFRA_TOPAXI_PROTECTSTA3 0x0258
> +#define INFRA_TOPAXI_PROTECTEN_SET 0x0260
> +#define INFRA_TOPAXI_PROTECTEN_CLR 0x0264
> +#define INFRA_TOPAXI_PROTECTEN1_SET 0x0270
> +#define INFRA_TOPAXI_PROTECTEN1_CLR 0x0274
> +
> #define MT8173_TOP_AXI_PROT_EN_MCI_M2 BIT(0)
> #define MT8173_TOP_AXI_PROT_EN_MM_M0 BIT(1)
> #define MT8173_TOP_AXI_PROT_EN_MM_M1 BIT(2)
> @@ -22,5 +31,8 @@
>
> int mtk_infracfg_set_bus_protection(struct regmap *infracfg, u32 mask);
> int mtk_infracfg_clear_bus_protection(struct regmap *infracfg, u32 mask);
> -
> +int mtk_infracfg_set_bus_protection_ext(struct regmap *infracfg, u32 mask,
> + u32 reg_set, u32 reg_sta, u32 reg_en);
> +int mtk_infracfg_clear_bus_protection_ext(struct regmap *infracfg, u32 mask,
> + u32 reg_clr, u32 reg_sta, u32 reg_en);
> #endif /* __SOC_MEDIATEK_INFRACFG_H */
>
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
next prev parent reply other threads:[~2017-08-17 10:57 UTC|newest]
Thread overview: 14+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-08-15 6:42 [PATCH v1 0/9] Mediatek MT2712 clock and scpsys support weiyi.lu
2017-08-15 6:42 ` [PATCH v1 1/9] dt-bindings: ARM: Mediatek: Document bindings for MT2712 weiyi.lu
2017-08-17 21:04 ` Rob Herring
2017-08-15 6:42 ` [PATCH v1 2/9] clk: mediatek: Add dt-bindings for MT2712 clocks weiyi.lu
2017-08-15 6:42 ` [PATCH v1 3/9] clk: mediatek: Add MT2712 clock support weiyi.lu
[not found] ` <1502779370-30150-1-git-send-email-weiyi.lu-NuS5LvNUpcJWk0Htik3J/w@public.gmane.org>
2017-08-15 6:42 ` [PATCH v1 4/9] arm: dts: mt2712: Add clock controller device nodes weiyi.lu-NuS5LvNUpcJWk0Htik3J/w
2017-08-15 7:17 ` [PATCH v1 0/9] Mediatek MT2712 clock and scpsys support Sean Wang
2017-08-15 6:42 ` [PATCH v1 5/9] dt-bindings: soc: add MT2712 power dt-bindings weiyi.lu
[not found] ` <1502779370-30150-6-git-send-email-weiyi.lu-NuS5LvNUpcJWk0Htik3J/w@public.gmane.org>
2017-08-17 21:09 ` Rob Herring
2017-08-15 6:42 ` [PATCH v1 6/9] soc: mediatek: add bus protection extend API weiyi.lu
[not found] ` <1502779370-30150-7-git-send-email-weiyi.lu-NuS5LvNUpcJWk0Htik3J/w@public.gmane.org>
2017-08-17 10:57 ` Matthias Brugger [this message]
2017-08-15 6:42 ` [PATCH v1 7/9] soc: mediatek: add dependent clock jpgdec/audio for scpsys weiyi.lu
2017-08-15 6:42 ` [PATCH v1 8/9] soc: mediatek: add MT2712 scpsys support weiyi.lu
2017-08-15 6:42 ` [PATCH v1 9/9] arm: dts: Add power controller device node of MT2712 weiyi.lu
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=a179cf17-0380-f47e-645e-92f38b0e1917@gmail.com \
--to=matthias.bgg-re5jqeeqqe8avxtiumwx3w@public.gmane.org \
--cc=devicetree-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=fan.chen-NuS5LvNUpcJWk0Htik3J/w@public.gmane.org \
--cc=jamesjj.liao-NuS5LvNUpcJWk0Htik3J/w@public.gmane.org \
--cc=linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org \
--cc=linux-clk-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=linux-mediatek-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org \
--cc=sboyd-sgV2jX0FEOL9JmXXK+q4OQ@public.gmane.org \
--cc=srv_heupstream-NuS5LvNUpcJWk0Htik3J/w@public.gmane.org \
--cc=weiyi.lu-NuS5LvNUpcJWk0Htik3J/w@public.gmane.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).