From: Jon Hunter <jonathanh@nvidia.com>
To: Krishna Reddy <vdumpa@nvidia.com>
Cc: snikam@nvidia.com, nicoleotsuka@gmail.com, mperttunen@nvidia.com,
bhuntsman@nvidia.com, will@kernel.org,
linux-kernel@vger.kernel.org, praithatha@nvidia.com,
talho@nvidia.com, iommu@lists.linux-foundation.org,
nicolinc@nvidia.com, linux-tegra@vger.kernel.org,
yhsu@nvidia.com, treding@nvidia.com, robin.murphy@arm.com,
linux-arm-kernel@lists.infradead.org, bbiswas@nvidia.com
Subject: Re: [PATCH v8 1/3] iommu/arm-smmu: add NVIDIA implementation for dual ARM MMU-500 usage
Date: Tue, 30 Jun 2020 09:19:15 +0100 [thread overview]
Message-ID: <53bfa5c8-c32d-6fa3-df60-a18ab33ca1c2@nvidia.com> (raw)
In-Reply-To: <20200630001051.12350-2-vdumpa@nvidia.com>
On 30/06/2020 01:10, Krishna Reddy wrote:
> NVIDIA's Tegra194 SoC uses two ARM MMU-500s together to interleave
> IOVA accesses across them.
> Add NVIDIA implementation for dual ARM MMU-500s and add new compatible
> string for Tegra194 SoC SMMU topology.
There is no description here of the 3rd SMMU that you mention below.
I think that we should describe the full picture here.
> Signed-off-by: Krishna Reddy <vdumpa@nvidia.com>
> ---
> MAINTAINERS | 2 +
> drivers/iommu/Makefile | 2 +-
> drivers/iommu/arm-smmu-impl.c | 3 +
> drivers/iommu/arm-smmu-nvidia.c | 196 ++++++++++++++++++++++++++++++++
> drivers/iommu/arm-smmu.h | 1 +
> 5 files changed, 203 insertions(+), 1 deletion(-)
> create mode 100644 drivers/iommu/arm-smmu-nvidia.c
>
> diff --git a/MAINTAINERS b/MAINTAINERS
> index 7b5ffd646c6b9..64c37dbdd4426 100644
> --- a/MAINTAINERS
> +++ b/MAINTAINERS
> @@ -16808,8 +16808,10 @@ F: drivers/i2c/busses/i2c-tegra.c
>
> TEGRA IOMMU DRIVERS
> M: Thierry Reding <thierry.reding@gmail.com>
> +R: Krishna Reddy <vdumpa@nvidia.com>
> L: linux-tegra@vger.kernel.org
> S: Supported
> +F: drivers/iommu/arm-smmu-nvidia.c
> F: drivers/iommu/tegra*
>
> TEGRA KBC DRIVER
> diff --git a/drivers/iommu/Makefile b/drivers/iommu/Makefile
> index 342190196dfb0..2b8203db73ec3 100644
> --- a/drivers/iommu/Makefile
> +++ b/drivers/iommu/Makefile
> @@ -15,7 +15,7 @@ obj-$(CONFIG_AMD_IOMMU) += amd/iommu.o amd/init.o amd/quirks.o
> obj-$(CONFIG_AMD_IOMMU_DEBUGFS) += amd/debugfs.o
> obj-$(CONFIG_AMD_IOMMU_V2) += amd/iommu_v2.o
> obj-$(CONFIG_ARM_SMMU) += arm_smmu.o
> -arm_smmu-objs += arm-smmu.o arm-smmu-impl.o arm-smmu-qcom.o
> +arm_smmu-objs += arm-smmu.o arm-smmu-impl.o arm-smmu-nvidia.o arm-smmu-qcom.o
> obj-$(CONFIG_ARM_SMMU_V3) += arm-smmu-v3.o
> obj-$(CONFIG_DMAR_TABLE) += intel/dmar.o
> obj-$(CONFIG_INTEL_IOMMU) += intel/iommu.o intel/pasid.o
> diff --git a/drivers/iommu/arm-smmu-impl.c b/drivers/iommu/arm-smmu-impl.c
> index c75b9d957b702..70f7318017617 100644
> --- a/drivers/iommu/arm-smmu-impl.c
> +++ b/drivers/iommu/arm-smmu-impl.c
> @@ -171,6 +171,9 @@ struct arm_smmu_device *arm_smmu_impl_init(struct arm_smmu_device *smmu)
> if (of_property_read_bool(np, "calxeda,smmu-secure-config-access"))
> smmu->impl = &calxeda_impl;
>
> + if (of_device_is_compatible(smmu->dev->of_node, "nvidia,tegra194-smmu"))
> + return nvidia_smmu_impl_init(smmu);
> +
> if (of_device_is_compatible(np, "qcom,sdm845-smmu-500") ||
> of_device_is_compatible(np, "qcom,sc7180-smmu-500"))
> return qcom_smmu_impl_init(smmu);
> diff --git a/drivers/iommu/arm-smmu-nvidia.c b/drivers/iommu/arm-smmu-nvidia.c
> new file mode 100644
> index 0000000000000..1124f0ac1823a
> --- /dev/null
> +++ b/drivers/iommu/arm-smmu-nvidia.c
> @@ -0,0 +1,196 @@
> +// SPDX-License-Identifier: GPL-2.0-only
> +// NVIDIA ARM SMMU v2 implementation quirks
> +// Copyright (C) 2019-2020 NVIDIA CORPORATION. All rights reserved.
> +
> +#include <linux/bitfield.h>
> +#include <linux/delay.h>
> +#include <linux/of.h>
> +#include <linux/platform_device.h>
> +#include <linux/slab.h>
> +
> +#include "arm-smmu.h"
> +
> +/*
> + * Tegra194 has three ARM MMU-500 Instances.
> + * Two of them are used together for interleaved IOVA accesses and
> + * used by non-isochronous HW devices for SMMU translations.
> + * Third one is used for SMMU translations from isochronous HW devices.
> + * It is possible to use this implementation to program either
> + * all three or two of the instances identically as desired through
> + * DT node.
> + *
> + * Programming all the three instances identically comes with redundant TLB
> + * invalidations as all three never need to be TLB invalidated for a HW device.
> + *
> + * When Linux kernel supports multiple SMMU devices, the SMMU device used for
> + * isochornous HW devices should be added as a separate ARM MMU-500 device
> + * in DT and be programmed independently for efficient TLB invalidates.
> + */
> +#define MAX_SMMU_INSTANCES 3
> +
> +#define TLB_LOOP_TIMEOUT_IN_US 1000000 /* 1s! */
> +#define TLB_SPIN_COUNT 10
> +
> +struct nvidia_smmu {
> + struct arm_smmu_device smmu;
> + unsigned int num_inst;
> + void __iomem *bases[MAX_SMMU_INSTANCES];
> +};
> +
> +static inline struct nvidia_smmu *to_nvidia_smmu(struct arm_smmu_device *smmu)
> +{
> + return container_of(smmu, struct nvidia_smmu, smmu);
> +}
> +
> +static inline void __iomem *nvidia_smmu_page(struct arm_smmu_device *smmu,
> + unsigned int inst, int page)
If you run checkpatch --strict on these you will get a lot of ...
CHECK: Alignment should match open parenthesis
#116: FILE: drivers/iommu/arm-smmu-nvidia.c:46:
+static inline void __iomem *nvidia_smmu_page(struct arm_smmu_device *smmu,
+ unsigned int inst, int page)
We should fix these.
> +{
> + struct nvidia_smmu *nvidia_smmu = to_nvidia_smmu(smmu);
> +
> + if (!nvidia_smmu->bases[0])
> + nvidia_smmu->bases[0] = smmu->base;
> +
> + return nvidia_smmu->bases[inst] + (page << smmu->pgshift);
> +}
> +
> +static u32 nvidia_smmu_read_reg(struct arm_smmu_device *smmu,
> + int page, int offset)
> +{
> + void __iomem *reg = nvidia_smmu_page(smmu, 0, page) + offset;
> +
> + return readl_relaxed(reg);
> +}
> +
> +static void nvidia_smmu_write_reg(struct arm_smmu_device *smmu,
> + int page, int offset, u32 val)
> +{
> + unsigned int i;
> + struct nvidia_smmu *nvidia_smmu = to_nvidia_smmu(smmu);
> +
> + for (i = 0; i < nvidia_smmu->num_inst; i++) {
> + void __iomem *reg = nvidia_smmu_page(smmu, i, page) + offset;
Personally, I would declare 'reg' outside of the loop as I feel it will make
the code cleaner and easier to read.
> +
> + writel_relaxed(val, reg);
> + }
> +}
> +
> +static u64 nvidia_smmu_read_reg64(struct arm_smmu_device *smmu,
> + int page, int offset)
> +{
> + void __iomem *reg = nvidia_smmu_page(smmu, 0, page) + offset;
> +
> + return readq_relaxed(reg);
> +}
> +
> +static void nvidia_smmu_write_reg64(struct arm_smmu_device *smmu,
> + int page, int offset, u64 val)
> +{
> + unsigned int i;
> + struct nvidia_smmu *nvidia_smmu = to_nvidia_smmu(smmu);
> +
> + for (i = 0; i < nvidia_smmu->num_inst; i++) {
> + void __iomem *reg = nvidia_smmu_page(smmu, i, page) + offset;
> +
> + writeq_relaxed(val, reg);
> + }
> +}
> +
> +static void nvidia_smmu_tlb_sync(struct arm_smmu_device *smmu, int page,
> + int sync, int status)
> +{
> + unsigned int delay;
> +
> + arm_smmu_writel(smmu, page, sync, 0);
> +
> + for (delay = 1; delay < TLB_LOOP_TIMEOUT_IN_US; delay *= 2) {
So we are doubling the delay every time? Is this better than just using
the same on each loop?
> + unsigned int spin_cnt;
> +
> + for (spin_cnt = TLB_SPIN_COUNT; spin_cnt > 0; spin_cnt--) {
> + u32 val = 0;
> + unsigned int i;
> + struct nvidia_smmu *nvidia_smmu = to_nvidia_smmu(smmu);
Why not do this once at the beginning of the function?
> +
> + for (i = 0; i < nvidia_smmu->num_inst; i++) {
> + void __iomem *reg =
> + nvidia_smmu_page(smmu, i, page) + status;
> +
> + val |= readl_relaxed(reg);
> + }
> +
> + if (!(val & ARM_SMMU_sTLBGSTATUS_GSACTIVE))
> + return;
> +
> + cpu_relax();
> + }
> +
> + udelay(delay);
> + }
> +
> + dev_err_ratelimited(smmu->dev,
> + "TLB sync timed out -- SMMU may be deadlocked\n");
> +}
> +
> +static int nvidia_smmu_reset(struct arm_smmu_device *smmu)
> +{
> + unsigned int i;
> +
> + for (i = 0; i < to_nvidia_smmu(smmu)->num_inst; i++) {
> + u32 val;
> + void __iomem *reg = nvidia_smmu_page(smmu, i, ARM_SMMU_GR0) +
> + ARM_SMMU_GR0_sGFSR;
I feel that declaring variables here clutters the code.
> +
> + /* clear global FSR */
> + val = readl_relaxed(reg);
> + writel_relaxed(val, reg);
> + }
> +
> + return 0;
> +}
> +
> +static const struct arm_smmu_impl nvidia_smmu_impl = {
> + .read_reg = nvidia_smmu_read_reg,
> + .write_reg = nvidia_smmu_write_reg,
> + .read_reg64 = nvidia_smmu_read_reg64,
> + .write_reg64 = nvidia_smmu_write_reg64,
> + .reset = nvidia_smmu_reset,
> + .tlb_sync = nvidia_smmu_tlb_sync,
> +};
> +
> +struct arm_smmu_device *nvidia_smmu_impl_init(struct arm_smmu_device *smmu)
> +{
> + unsigned int i;
> + struct nvidia_smmu *nvidia_smmu;
> + struct platform_device *pdev = to_platform_device(smmu->dev);
> +
> + nvidia_smmu = devm_kzalloc(smmu->dev, sizeof(*nvidia_smmu), GFP_KERNEL);
> + if (!nvidia_smmu)
> + return ERR_PTR(-ENOMEM);
> +
> + nvidia_smmu->smmu = *smmu;
> + /* Instance 0 is ioremapped by arm-smmu.c after this function returns */
> + nvidia_smmu->num_inst = 1;
> +
> + for (i = 1; i < MAX_SMMU_INSTANCES; i++) {
> + struct resource *res;
> +
> + res = platform_get_resource(pdev, IORESOURCE_MEM, i);
> + if (!res)
> + break;
> +
> + nvidia_smmu->bases[i] = devm_ioremap_resource(smmu->dev, res);
> + if (IS_ERR(nvidia_smmu->bases[i]))
> + return ERR_CAST(nvidia_smmu->bases[i]);
> +
> + nvidia_smmu->num_inst++;
> + }
> +
> + nvidia_smmu->smmu.impl = &nvidia_smmu_impl;
> + /*
> + * Free the arm_smmu_device struct allocated in arm-smmu.c.
> + * Once this function returns, arm-smmu.c would use arm_smmu_device
> + * allocated as part of nvidia_smmu struct.
> + */
> + devm_kfree(smmu->dev, smmu);
Why don't we just store the pointer of the smmu struct passed to this function
in the nvidia_smmu struct and then we do not need to free this here. In other
words make ...
struct nvidia_smmu {
struct arm_smmu_device *smmu;
unsigned int num_inst;
void __iomem *bases[MAX_SMMU_INSTANCES];
};
This seems more appropriate, than copying the struct and freeing memory
allocated else-where.
> +
> + return &nvidia_smmu->smmu;
> +}
> diff --git a/drivers/iommu/arm-smmu.h b/drivers/iommu/arm-smmu.h
> index d172c024be618..8cf1511ed9874 100644
> --- a/drivers/iommu/arm-smmu.h
> +++ b/drivers/iommu/arm-smmu.h
> @@ -450,6 +450,7 @@ static inline void arm_smmu_writeq(struct arm_smmu_device *smmu, int page,
> arm_smmu_writeq((s), ARM_SMMU_CB((s), (n)), (o), (v))
>
> struct arm_smmu_device *arm_smmu_impl_init(struct arm_smmu_device *smmu);
> +struct arm_smmu_device *nvidia_smmu_impl_init(struct arm_smmu_device *smmu);
> struct arm_smmu_device *qcom_smmu_impl_init(struct arm_smmu_device *smmu);
>
> int arm_mmu500_reset(struct arm_smmu_device *smmu);
>
Cheers
Jon
--
nvpublic
_______________________________________________
iommu mailing list
iommu@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/iommu
next prev parent reply other threads:[~2020-06-30 8:19 UTC|newest]
Thread overview: 38+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-06-30 0:10 [PATCH v8 0/3] Nvidia Arm SMMUv2 Implementation Krishna Reddy
2020-06-30 0:10 ` [PATCH v8 1/3] iommu/arm-smmu: add NVIDIA implementation for dual ARM MMU-500 usage Krishna Reddy
2020-06-30 0:16 ` Nicolin Chen
2020-06-30 5:54 ` Pritesh Raithatha
2020-06-30 8:19 ` Jon Hunter [this message]
2020-06-30 14:53 ` Robin Murphy
2020-06-30 15:17 ` Jon Hunter
2020-07-01 18:18 ` Krishna Reddy
2020-07-01 18:56 ` Robin Murphy
2020-07-01 19:12 ` Krishna Reddy
2020-06-30 17:04 ` Krishna Reddy
2020-06-30 10:17 ` Jon Hunter
2020-06-30 16:23 ` Krishna Reddy
2020-06-30 16:32 ` Jon Hunter
2020-06-30 16:44 ` Jon Hunter
2020-06-30 17:16 ` Krishna Reddy
2020-06-30 19:03 ` Jon Hunter
2020-06-30 20:21 ` Krishna Reddy
2020-06-30 0:10 ` [PATCH v8 2/3] dt-bindings: arm-smmu: Add binding for Tegra194 SMMU Krishna Reddy
2020-06-30 6:01 ` Pritesh Raithatha
2020-06-30 8:21 ` Jon Hunter
2020-06-30 12:27 ` Robin Murphy
2020-07-01 18:28 ` Krishna Reddy
2020-07-01 18:47 ` Jon Hunter
2020-07-01 19:00 ` Krishna Reddy
2020-07-01 19:31 ` Jon Hunter
2020-07-01 19:39 ` Krishna Reddy
2020-07-02 16:05 ` Robin Murphy
2020-07-01 19:03 ` Robin Murphy
2020-06-30 0:10 ` [PATCH v8 3/3] iommu/arm-smmu: Add global/context fault implementation hooks Krishna Reddy
2020-06-30 0:19 ` Nicolin Chen
2020-06-30 5:58 ` Pritesh Raithatha
2020-06-30 8:37 ` Jon Hunter
2020-06-30 12:13 ` Robin Murphy
2020-06-30 12:42 ` Jon Hunter
2020-07-01 18:48 ` Krishna Reddy
2020-07-01 19:14 ` Robin Murphy
2020-07-01 19:22 ` Krishna Reddy
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=53bfa5c8-c32d-6fa3-df60-a18ab33ca1c2@nvidia.com \
--to=jonathanh@nvidia.com \
--cc=bbiswas@nvidia.com \
--cc=bhuntsman@nvidia.com \
--cc=iommu@lists.linux-foundation.org \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-tegra@vger.kernel.org \
--cc=mperttunen@nvidia.com \
--cc=nicoleotsuka@gmail.com \
--cc=nicolinc@nvidia.com \
--cc=praithatha@nvidia.com \
--cc=robin.murphy@arm.com \
--cc=snikam@nvidia.com \
--cc=talho@nvidia.com \
--cc=treding@nvidia.com \
--cc=vdumpa@nvidia.com \
--cc=will@kernel.org \
--cc=yhsu@nvidia.com \
/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).