From: Mostafa Saleh <smostafa@google.com>
To: Jason Gunthorpe <jgg@nvidia.com>
Cc: iommu@lists.linux.dev, Jonathan Hunter <jonathanh@nvidia.com>,
Joerg Roedel <joro@8bytes.org>,
linux-arm-kernel@lists.infradead.org,
linux-tegra@vger.kernel.org, Robin Murphy <robin.murphy@arm.com>,
Thierry Reding <thierry.reding@kernel.org>,
Krishna Reddy <vdumpa@nvidia.com>, Will Deacon <will@kernel.org>,
David Matlack <dmatlack@google.com>,
Pasha Tatashin <pasha.tatashin@soleen.com>,
patches@lists.linux.dev, Samiullah Khawaja <skhawaja@google.com>
Subject: Re: [PATCH 1/9] iommu/arm-smmu-v3: Add struct arm_smmu_cmd to represent the HW format command
Date: Thu, 7 May 2026 09:19:14 +0000 [thread overview]
Message-ID: <afxZEgun2BjHJ-WJ@google.com> (raw)
In-Reply-To: <1-v1-b7dc0a0d4aa0+3723d-smmu_no_cmdq_ent_jgg@nvidia.com>
On Fri, May 01, 2026 at 11:29:10AM -0300, Jason Gunthorpe wrote:
> Like STE/CD, add a wrapper struct around the u64 array to represent the
> already FIELD_PREP'd command data. Unlike the STE/CD this is a u64
> because the command submission path will have the swap to le64.
>
> This makes the API clearer when a u64 is referring to a formatted
> command and makes the following changes easier to follow.
>
> Move the command constants out of the struct and into an enum alongside
> the rest of the constants defining the HW format so the entire HW format
> is self contained and independent of struct arm_smmu_cmdq_ent.
>
> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Reviewed-by: Mostafa Saleh <smostafa@google.com>
Thanks,
Mostafa
> ---
> drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h | 48 ++++++++++++---------
> 1 file changed, 28 insertions(+), 20 deletions(-)
>
> diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h
> index ef42df4753ec4d..092179f689e9f1 100644
> --- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h
> +++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h
> @@ -390,6 +390,10 @@ static inline unsigned int arm_smmu_cdtab_l2_idx(unsigned int ssid)
>
> #define CMDQ_PROD_OWNED_FLAG Q_OVERFLOW_FLAG
>
> +struct arm_smmu_cmd {
> + u64 data[CMDQ_ENT_DWORDS];
> +};
> +
> /*
> * This is used to size the command queue and therefore must be at least
> * BITS_PER_LONG so that the valid_map works correctly (it relies on the
> @@ -426,6 +430,8 @@ static inline unsigned int arm_smmu_cdtab_l2_idx(unsigned int ssid)
> #define CMDQ_ATC_1_SIZE GENMASK_ULL(5, 0)
> #define CMDQ_ATC_1_ADDR_MASK GENMASK_ULL(63, 12)
>
> +#define ATC_INV_SIZE_ALL 52
> +
> #define CMDQ_PRI_0_SSID GENMASK_ULL(31, 12)
> #define CMDQ_PRI_0_SID GENMASK_ULL(63, 32)
> #define CMDQ_PRI_1_GRPID GENMASK_ULL(8, 0)
> @@ -447,6 +453,28 @@ static inline unsigned int arm_smmu_cdtab_l2_idx(unsigned int ssid)
> #define CMDQ_SYNC_0_MSIDATA GENMASK_ULL(63, 32)
> #define CMDQ_SYNC_1_MSIADDR_MASK GENMASK_ULL(51, 2)
>
> +enum arm_smmu_cmdq_opcode {
> + CMDQ_OP_PREFETCH_CFG = 0x1,
> + CMDQ_OP_CFGI_STE = 0x3,
> + CMDQ_OP_CFGI_ALL = 0x4,
> + CMDQ_OP_CFGI_CD = 0x5,
> + CMDQ_OP_CFGI_CD_ALL = 0x6,
> + CMDQ_OP_TLBI_NH_ALL = 0x10,
> + CMDQ_OP_TLBI_NH_ASID = 0x11,
> + CMDQ_OP_TLBI_NH_VA = 0x12,
> + CMDQ_OP_TLBI_NH_VAA = 0x13,
> + CMDQ_OP_TLBI_EL2_ALL = 0x20,
> + CMDQ_OP_TLBI_EL2_ASID = 0x21,
> + CMDQ_OP_TLBI_EL2_VA = 0x22,
> + CMDQ_OP_TLBI_S12_VMALL = 0x28,
> + CMDQ_OP_TLBI_S2_IPA = 0x2a,
> + CMDQ_OP_TLBI_NSNH_ALL = 0x30,
> + CMDQ_OP_ATC_INV = 0x40,
> + CMDQ_OP_PRI_RESP = 0x41,
> + CMDQ_OP_RESUME = 0x44,
> + CMDQ_OP_CMD_SYNC = 0x46,
> +};
> +
> /* Event queue */
> #define EVTQ_ENT_SZ_SHIFT 5
> #define EVTQ_ENT_DWORDS ((1 << EVTQ_ENT_SZ_SHIFT) >> 3)
> @@ -520,15 +548,10 @@ struct arm_smmu_cmdq_ent {
>
> /* Command-specific fields */
> union {
> - #define CMDQ_OP_PREFETCH_CFG 0x1
> struct {
> u32 sid;
> } prefetch;
>
> - #define CMDQ_OP_CFGI_STE 0x3
> - #define CMDQ_OP_CFGI_ALL 0x4
> - #define CMDQ_OP_CFGI_CD 0x5
> - #define CMDQ_OP_CFGI_CD_ALL 0x6
> struct {
> u32 sid;
> u32 ssid;
> @@ -538,16 +561,6 @@ struct arm_smmu_cmdq_ent {
> };
> } cfgi;
>
> - #define CMDQ_OP_TLBI_NH_ALL 0x10
> - #define CMDQ_OP_TLBI_NH_ASID 0x11
> - #define CMDQ_OP_TLBI_NH_VA 0x12
> - #define CMDQ_OP_TLBI_NH_VAA 0x13
> - #define CMDQ_OP_TLBI_EL2_ALL 0x20
> - #define CMDQ_OP_TLBI_EL2_ASID 0x21
> - #define CMDQ_OP_TLBI_EL2_VA 0x22
> - #define CMDQ_OP_TLBI_S12_VMALL 0x28
> - #define CMDQ_OP_TLBI_S2_IPA 0x2a
> - #define CMDQ_OP_TLBI_NSNH_ALL 0x30
> struct {
> u8 num;
> u8 scale;
> @@ -559,8 +572,6 @@ struct arm_smmu_cmdq_ent {
> u64 addr;
> } tlbi;
>
> - #define CMDQ_OP_ATC_INV 0x40
> - #define ATC_INV_SIZE_ALL 52
> struct {
> u32 sid;
> u32 ssid;
> @@ -569,7 +580,6 @@ struct arm_smmu_cmdq_ent {
> bool global;
> } atc;
>
> - #define CMDQ_OP_PRI_RESP 0x41
> struct {
> u32 sid;
> u32 ssid;
> @@ -577,14 +587,12 @@ struct arm_smmu_cmdq_ent {
> enum pri_resp resp;
> } pri;
>
> - #define CMDQ_OP_RESUME 0x44
> struct {
> u32 sid;
> u16 stag;
> u8 resp;
> } resume;
>
> - #define CMDQ_OP_CMD_SYNC 0x46
> struct {
> u64 msiaddr;
> } sync;
> --
> 2.43.0
>
next prev parent reply other threads:[~2026-05-07 9:19 UTC|newest]
Thread overview: 44+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-05-01 14:29 [PATCH 0/9] Remove SMMUv3 struct arm_smmu_cmdq_ent Jason Gunthorpe
2026-05-01 14:29 ` [PATCH 1/9] iommu/arm-smmu-v3: Add struct arm_smmu_cmd to represent the HW format command Jason Gunthorpe
2026-05-06 6:11 ` Nicolin Chen
2026-05-06 23:41 ` Samiullah Khawaja
2026-05-07 9:19 ` Mostafa Saleh [this message]
2026-05-08 7:29 ` Pranjal Shrivastava
2026-05-01 14:29 ` [PATCH 2/9] iommu/arm-smmu-v3: Use the HW arm_smmu_cmd in cmdq selection functions Jason Gunthorpe
2026-05-07 9:21 ` Mostafa Saleh
2026-05-08 15:49 ` Jason Gunthorpe
2026-05-08 7:47 ` Pranjal Shrivastava
2026-05-08 15:54 ` Jason Gunthorpe
2026-05-08 16:58 ` Pranjal Shrivastava
2026-05-01 14:29 ` [PATCH 3/9] iommu/arm-smmu-v3: Use the HW arm_smmu_cmd in cmdq submission functions Jason Gunthorpe
2026-05-07 9:21 ` Mostafa Saleh
2026-05-08 8:27 ` Pranjal Shrivastava
2026-05-08 16:00 ` Jason Gunthorpe
2026-05-08 17:00 ` Pranjal Shrivastava
2026-05-01 14:29 ` [PATCH 4/9] iommu/arm-smmu-v3: Convert arm_smmu_cmdq_batch cmds to struct arm_smmu_cmd Jason Gunthorpe
2026-05-07 9:22 ` Mostafa Saleh
2026-05-08 9:26 ` Pranjal Shrivastava
2026-05-01 14:29 ` [PATCH 5/9] iommu/arm-smmu-v3: Remove CMDQ_OP_CFGI_CD_ALL from arm_smmu_cmdq_build_cmd() Jason Gunthorpe
2026-05-07 9:22 ` Mostafa Saleh
2026-05-08 9:45 ` Pranjal Shrivastava
2026-05-08 16:02 ` Jason Gunthorpe
2026-05-08 17:17 ` Pranjal Shrivastava
2026-05-01 14:29 ` [PATCH 6/9] iommu/arm-smmu-v3: Directly encode simple commands Jason Gunthorpe
2026-05-07 9:22 ` Mostafa Saleh
2026-05-08 11:33 ` Pranjal Shrivastava
2026-05-08 17:37 ` Jason Gunthorpe
2026-05-08 20:09 ` Pranjal Shrivastava
2026-05-08 23:36 ` Jason Gunthorpe
2026-05-10 18:59 ` Pranjal Shrivastava
2026-05-01 14:29 ` [PATCH 7/9] iommu/arm-smmu-v3: Directly encode CMDQ_OP_ATC_INV Jason Gunthorpe
2026-05-07 9:23 ` Mostafa Saleh
2026-05-08 11:46 ` Pranjal Shrivastava
2026-05-09 16:54 ` Jason Gunthorpe
2026-05-01 14:29 ` [PATCH 8/9] iommu/arm-smmu-v3: Directly encode CMDQ_OP_SYNC Jason Gunthorpe
2026-05-07 9:23 ` Mostafa Saleh
2026-05-08 13:41 ` Pranjal Shrivastava
2026-05-01 14:29 ` [PATCH 9/9] iommu/arm-smmu-v3: Directly encode TLBI commands Jason Gunthorpe
2026-05-07 9:24 ` Mostafa Saleh
2026-05-08 14:00 ` Pranjal Shrivastava
2026-05-07 9:26 ` [PATCH 0/9] Remove SMMUv3 struct arm_smmu_cmdq_ent Mostafa Saleh
2026-05-08 14:03 ` Pranjal Shrivastava
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=afxZEgun2BjHJ-WJ@google.com \
--to=smostafa@google.com \
--cc=dmatlack@google.com \
--cc=iommu@lists.linux.dev \
--cc=jgg@nvidia.com \
--cc=jonathanh@nvidia.com \
--cc=joro@8bytes.org \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-tegra@vger.kernel.org \
--cc=pasha.tatashin@soleen.com \
--cc=patches@lists.linux.dev \
--cc=robin.murphy@arm.com \
--cc=skhawaja@google.com \
--cc=thierry.reding@kernel.org \
--cc=vdumpa@nvidia.com \
--cc=will@kernel.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