* Re: [PATCH v3 2/3] iommu/mediatek: Add enable IOMMU SMC command for INFRA master
[not found] ` <20220919082611.19824-3-chengci.xu@mediatek.com>
@ 2022-09-20 9:01 ` AngeloGioacchino Del Regno
0 siblings, 0 replies; 3+ messages in thread
From: AngeloGioacchino Del Regno @ 2022-09-20 9:01 UTC (permalink / raw)
To: Chengci.Xu, Yong Wu, Joerg Roedel, Will Deacon, Robin Murphy,
Rob Herring, Krzysztof Kozlowski, Matthias Brugger
Cc: iommu, linux-mediatek, devicetree, linux-kernel,
linux-arm-kernel, Project_Global_Chrome_Upstream_Group
Il 19/09/22 10:26, Chengci.Xu ha scritto:
> The register which can enable IOMMU for INFRA master should be setted
> in secure world for security concerns. Therefore, we add a SMC command
> for INFRA master to enable/disable INFRA IOMMU in ATF. This function is
> prepared for MT8188.
>
> Signed-off-by: Chengci.Xu <chengci.xu@mediatek.com>
> ---
> drivers/iommu/mtk_iommu.c | 21 +++++++++++++++++++--
> include/soc/mediatek/smi.h | 1 +
> 2 files changed, 20 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/iommu/mtk_iommu.c b/drivers/iommu/mtk_iommu.c
> index 552e4eb8c610..8b8a289bab2c 100644
> --- a/drivers/iommu/mtk_iommu.c
> +++ b/drivers/iommu/mtk_iommu.c
> @@ -3,6 +3,7 @@
> * Copyright (c) 2015-2016 MediaTek Inc.
> * Author: Yong Wu <yong.wu@mediatek.com>
> */
> +#include <linux/arm-smccc.h>
> #include <linux/bitfield.h>
> #include <linux/bug.h>
> #include <linux/clk.h>
> @@ -28,6 +29,7 @@
> #include <linux/slab.h>
> #include <linux/spinlock.h>
> #include <linux/soc/mediatek/infracfg.h>
> +#include <linux/soc/mediatek/mtk_sip_svc.h>
> #include <asm/barrier.h>
> #include <soc/mediatek/smi.h>
>
> @@ -138,6 +140,7 @@
> #define PM_CLK_AO BIT(15)
> #define IFA_IOMMU_PCIE_SUPPORT BIT(16)
> #define PGTABLE_PA_35_EN BIT(17)
> +#define CFG_IFA_MASTER_IN_ATF BIT(18)
>
> #define MTK_IOMMU_HAS_FLAG_MASK(pdata, _x, mask) \
> ((((pdata)->flags) & (mask)) == (_x))
> @@ -553,7 +556,20 @@ static int mtk_iommu_config(struct mtk_iommu_data *data, struct device *dev,
> larb_mmu->mmu |= MTK_SMI_MMU_EN(portid);
> else
> larb_mmu->mmu &= ~MTK_SMI_MMU_EN(portid);
> - } else if (MTK_IOMMU_IS_TYPE(data->plat_data, MTK_IOMMU_TYPE_INFRA)) {
> + } else if (MTK_IOMMU_IS_TYPE(data->plat_data, MTK_IOMMU_TYPE_INFRA) &&
> + MTK_IOMMU_HAS_FLAG(data->plat_data, CFG_IFA_MASTER_IN_ATF)) {
> + struct arm_smccc_res res;
> +
> + arm_smccc_smc(MTK_SIP_KERNEL_IOMMU_CONTROL,
> + IOMMU_ATF_CMD_CONFIG_INFRA_IOMMU,
> + portid, enable, 0, 0, 0, 0, &res);
> + if (res.a0 != 0) {
> + dev_err(dev, "%s iommu(%s) inframaster %d fail(%ld).\n",
> + enable ? "enable" : "disable",
> + dev_name(data->dev), portid, res.a0);
> + ret = -EINVAL;
> + }
> + } else {
This one is opening a big window for future mistakes.
I think that the only way that you have to do this is...
} else if (MTK_IOMMU_IS_TYPE(data->plat_data, MTK_IOMMU_TYPE_INFRA)) {
if (MTK_IOMMU_HAS_FLAG(data->plat_data, CFG_IFA_MASTER_IN_ATF)) {
arm_smcc_smc ....
} else {
peri_mmuen_msk ....
}
}
Otherwise, to reduce indentation, you'd have to do something like...
static int mtk_iommu_config_one(struct mtk_iommu_data *data, struct device *dev,
bool enable, u32 regionid, u32 larbid, u32 portid)
{
struct mtk_smi_larb_iommu *larb_mmu;
............
}
static int mtk_iommu_config(struct .............)
{
vars, etc....
for (i = 0; i < fwspec->num_ids; ++i)
mtk_iommu_config_one(data, dev, enable, regionid,
MTK_M4U_TO_LARB(...), MTK_M4U_TO_PORT(...));
}
or....
static int mtk_iommu_config_one_infra(struct ....)
{
if (is atf) ....
else ....
}
Your choice.
Regards,
Angelo
^ permalink raw reply [flat|nested] 3+ messages in thread