[PATCH v11 0/4] misc: fastrpc: Add polling mode support

[PATCH v11 0/4] misc: fastrpc: Add polling mode support

[Ekansh Gupta]

This patch series adds polling mode feature that have been missing in
upstream FastRPC driver.

- Add changes to move fdlist to ctx structure to avoid code duplicacy.
- Update context mask to support polling mode.
- Add changes to support polling feature.

Userspace change: https://github.com/qualcomm/fastrpc/pull/258
Patch [v10]: https://lore.kernel.org/all/20260424095903.1622565-1-ekansh.gupta@oss.qualcomm.com/

Changes in v11:
  - Moved back to read_poll*.
  - Improved error handling.

Changes in v10:
  - Add milos and sar2130p to fastrpc_poll_supported_machines. 
  - Updated comment for supported platform list.

Changes in v9:
  - Added platform support check.
  - Moved to readl as per Luben's suggestion.
  - Cleaned up fastrpc_wait_for_completion().

Changes in v8:
  - Added more comments.

Changes in v7:
  - Fixed warnings.
  - Fixed commit text.
  - Addressed clean-up comments.

Changes in v6:
  - Fixed poll memory calculation.
  - Added few formatting changes.

Changes in v5:
  - Add more details in commit text.

Changes in v4:
  - Replace hardcoded ctxid mask with GENMASK.
  - Fixed commit text.

Changes in v3:
  - Resolve compilation warning.

Changes in v2:
  - Added comments and fixed commit text.
  - Defined context id position as a macro.
  - Added new IOCTL to control polling mode as always enabling
    it might cause excess power consumption.
  - Cleaned up polling mode implementation.

Ekansh Gupta (4):
  misc: fastrpc: Move fdlist to invoke context structure
  misc: fastrpc: Replace hardcoded ctxid mask with GENMASK
  misc: fastrpc: Expand context ID mask for DSP polling mode support
  misc: fastrpc: Add polling mode support for fastRPC driver

 drivers/misc/fastrpc.c      | 189 ++++++++++++++++++++++++++++++++----
 include/uapi/misc/fastrpc.h |  29 ++++++
 2 files changed, 198 insertions(+), 20 deletions(-)

-- 
2.34.1

[PATCH v11 1/4] misc: fastrpc: Move fdlist to invoke context structure

[Ekansh Gupta]

The fdlist is currently part of the meta buffer which is set during
fastrpc_get_args(), this fdlist is getting recalculated during
fastrpc_put_args().

Move fdlist to the invoke context structure to improve maintainability
and reduce redundancy. This centralizes its handling and simplifies
meta buffer preparation and reading logic.

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Signed-off-by: Ekansh Gupta <ekansh.gupta@oss.qualcomm.com>
---
 drivers/misc/fastrpc.c | 13 ++++---------
 1 file changed, 4 insertions(+), 9 deletions(-)

diff --git a/drivers/misc/fastrpc.c b/drivers/misc/fastrpc.c
index 1080f9acf70a..a9f507a88c67 100644
--- a/drivers/misc/fastrpc.c
+++ b/drivers/misc/fastrpc.c
@@ -233,6 +233,7 @@ struct fastrpc_invoke_ctx {
 	int pid;
 	int client_id;
 	u32 sc;
+	u64 *fdlist;
 	u32 *crc;
 	u64 ctxid;
 	u64 msg_sz;
@@ -1016,6 +1017,7 @@ static int fastrpc_get_args(u32 kernel, struct fastrpc_invoke_ctx *ctx)
 	rpra = ctx->buf->virt;
 	list = fastrpc_invoke_buf_start(rpra, ctx->nscalars);
 	pages = fastrpc_phy_page_start(list, ctx->nscalars);
+	ctx->fdlist = (u64 *)(pages + ctx->nscalars);
 	args = (uintptr_t)ctx->buf->virt + metalen;
 	rlen = pkt_size - metalen;
 	ctx->rpra = rpra;
@@ -1118,18 +1120,11 @@ static int fastrpc_put_args(struct fastrpc_invoke_ctx *ctx,
 	union fastrpc_remote_arg *rpra = ctx->rpra;
 	struct fastrpc_user *fl = ctx->fl;
 	struct fastrpc_map *mmap = NULL;
-	struct fastrpc_invoke_buf *list;
-	struct fastrpc_phy_page *pages;
-	u64 *fdlist;
-	int i, inbufs, outbufs, handles;
+	u64 *fdlist = ctx->fdlist;
+	int i, inbufs;
 	int ret = 0;
 
 	inbufs = REMOTE_SCALARS_INBUFS(ctx->sc);
-	outbufs = REMOTE_SCALARS_OUTBUFS(ctx->sc);
-	handles = REMOTE_SCALARS_INHANDLES(ctx->sc) + REMOTE_SCALARS_OUTHANDLES(ctx->sc);
-	list = fastrpc_invoke_buf_start(rpra, ctx->nscalars);
-	pages = fastrpc_phy_page_start(list, ctx->nscalars);
-	fdlist = (uint64_t *)(pages + inbufs + outbufs + handles);
 
 	for (i = inbufs; i < ctx->nbufs; ++i) {
 		if (!ctx->maps[i]) {
-- 
2.34.1

Claude review: misc: fastrpc: Move fdlist to invoke context structure

[Claude Code Review Bot]

Patch Review

**Overall: Looks good.**

This patch moves the `fdlist` pointer from being recalculated in `fastrpc_put_args()` to being stored in `struct fastrpc_invoke_ctx`. The pointer is computed once in `fastrpc_get_args()` and reused later.

The correctness relies on `ctx->nscalars` equaling `inbufs + outbufs + handles` — which is confirmed by the `REMOTE_SCALARS_LENGTH` macro definition:

```c
ctx->fdlist = (u64 *)(pages + ctx->nscalars);
```

This is equivalent to the original:
```c
fdlist = (uint64_t *)(pages + inbufs + outbufs + handles);
```

The cleanup of the now-unused `outbufs`, `handles`, `list`, and `pages` variables in `fastrpc_put_args()` is correct. One minor nit: the patch also removes the blank line after `int ret = 0;` — this creates an empty blank line before the `for` loop, which is fine but cosmetic.

**No issues found.**

---

---
Generated by Claude Code Patch Reviewer

[PATCH v11 2/4] misc: fastrpc: Replace hardcoded ctxid mask with GENMASK

[Ekansh Gupta]

Replace the hardcoded context ID mask (0xFF0) with GENMASK(11, 4) to
improve readability and follow kernel bitfield conventions. Use
FIELD_PREP and FIELD_GET instead of manual shifts for setting and
extracting ctxid values.

Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Signed-off-by: Ekansh Gupta <ekansh.gupta@oss.qualcomm.com>
---
 drivers/misc/fastrpc.c | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/drivers/misc/fastrpc.c b/drivers/misc/fastrpc.c
index a9f507a88c67..3f5d5d73be5a 100644
--- a/drivers/misc/fastrpc.c
+++ b/drivers/misc/fastrpc.c
@@ -23,6 +23,7 @@
 #include <uapi/misc/fastrpc.h>
 #include <linux/of_reserved_mem.h>
 #include <linux/bits.h>
+#include <linux/bitops.h>
 
 #define ADSP_DOMAIN_ID (0)
 #define MDSP_DOMAIN_ID (1)
@@ -37,7 +38,7 @@
 #define FASTRPC_CTX_MAX (256)
 #define FASTRPC_INIT_HANDLE	1
 #define FASTRPC_DSP_UTILITIES_HANDLE	2
-#define FASTRPC_CTXID_MASK (0xFF0)
+#define FASTRPC_CTXID_MASK GENMASK(11, 4)
 #define INIT_FILELEN_MAX (2 * 1024 * 1024)
 #define INIT_FILE_NAMELEN_MAX (128)
 #define FASTRPC_DEVICE_NAME	"fastrpc"
@@ -515,7 +516,7 @@ static void fastrpc_context_free(struct kref *ref)
 		fastrpc_buf_free(ctx->buf);
 
 	spin_lock_irqsave(&cctx->lock, flags);
-	idr_remove(&cctx->ctx_idr, ctx->ctxid >> 4);
+	idr_remove(&cctx->ctx_idr, FIELD_GET(FASTRPC_CTXID_MASK, ctx->ctxid));
 	spin_unlock_irqrestore(&cctx->lock, flags);
 
 	kfree(ctx->maps);
@@ -649,7 +650,7 @@ static struct fastrpc_invoke_ctx *fastrpc_context_alloc(
 		spin_unlock_irqrestore(&cctx->lock, flags);
 		goto err_idr;
 	}
-	ctx->ctxid = ret << 4;
+	ctx->ctxid = FIELD_PREP(FASTRPC_CTXID_MASK, ret);
 	spin_unlock_irqrestore(&cctx->lock, flags);
 
 	kref_init(&ctx->refcount);
@@ -2508,7 +2509,7 @@ static int fastrpc_rpmsg_callback(struct rpmsg_device *rpdev, void *data,
 	if (len < sizeof(*rsp))
 		return -EINVAL;
 
-	ctxid = ((rsp->ctx & FASTRPC_CTXID_MASK) >> 4);
+	ctxid = FIELD_GET(FASTRPC_CTXID_MASK, rsp->ctx);
 
 	spin_lock_irqsave(&cctx->lock, flags);
 	ctx = idr_find(&cctx->ctx_idr, ctxid);
-- 
2.34.1

Claude review: misc: fastrpc: Replace hardcoded ctxid mask with GENMASK

[Claude Code Review Bot]

Patch Review

**Overall: Looks good.** Clean mechanical conversion.

The patch adds `#include <linux/bitops.h>` (though `<linux/bitfield.h>` would be more precise for `FIELD_PREP`/`FIELD_GET` — but `bitops.h` does pull in what's needed via `bits.h` which was already included).

Conversions are correct:
- `ctx->ctxid >> 4` → `FIELD_GET(FASTRPC_CTXID_MASK, ctx->ctxid)` — correct, GENMASK(11,4) with FIELD_GET extracts bits [11:4] shifted down.
- `ret << 4` → `FIELD_PREP(FASTRPC_CTXID_MASK, ret)` — correct.
- `(rsp->ctx & FASTRPC_CTXID_MASK) >> 4` → `FIELD_GET(FASTRPC_CTXID_MASK, rsp->ctx)` — correct.

**Minor nit:** The added include is `linux/bitops.h` but `FIELD_PREP`/`FIELD_GET` are from `linux/bitfield.h`. This likely works because of transitive includes, but `linux/bitfield.h` would be the canonical include for these macros.

---

---
Generated by Claude Code Patch Reviewer

[PATCH v11 3/4] misc: fastrpc: Expand context ID mask for DSP polling mode support

[Ekansh Gupta]

Current FastRPC context uses a 12-bit mask:
  [ID(8 bits)][PD type(4 bits)] = GENMASK(11, 4)

This works for normal calls but fails for DSP polling mode.
Polling mode expects a 16-bit layout:
  [15:8] = context ID (8 bits)
  [7:5]  = reserved
  [4]    = async mode bit
  [3:0]  = PD type (4 bits)

If async bit (bit 4) is set, DSP disables polling. With current
mask, odd IDs can set this bit, causing DSP to skip poll updates.

Update FASTRPC_CTXID_MASK to GENMASK(15, 8) so IDs occupy upper
byte and lower byte is left for DSP flags and PD type.

Reserved bits remain unused. This change is compatible with
polling mode and does not break non-polling behavior.

Bit layout:
  [15:8] = CCCCCCCC (context ID)
  [7:5]  = xxx (reserved)
  [4]    = A (async mode)
  [3:0]  = PPPP (PD type)

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Signed-off-by: Ekansh Gupta <ekansh.gupta@oss.qualcomm.com>
---
 drivers/misc/fastrpc.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/misc/fastrpc.c b/drivers/misc/fastrpc.c
index 3f5d5d73be5a..c4a3547a5c7f 100644
--- a/drivers/misc/fastrpc.c
+++ b/drivers/misc/fastrpc.c
@@ -38,7 +38,7 @@
 #define FASTRPC_CTX_MAX (256)
 #define FASTRPC_INIT_HANDLE	1
 #define FASTRPC_DSP_UTILITIES_HANDLE	2
-#define FASTRPC_CTXID_MASK GENMASK(11, 4)
+#define FASTRPC_CTXID_MASK GENMASK(15, 8)
 #define INIT_FILELEN_MAX (2 * 1024 * 1024)
 #define INIT_FILE_NAMELEN_MAX (128)
 #define FASTRPC_DEVICE_NAME	"fastrpc"
-- 
2.34.1

Claude review: misc: fastrpc: Expand context ID mask for DSP polling mode support

[Claude Code Review Bot]

Patch Review

**Overall: Looks good.** The commit message clearly explains the rationale.

The change from `GENMASK(11, 4)` to `GENMASK(15, 8)` shifts the context ID to the upper byte, leaving bits [7:0] for DSP flags (async mode bit, PD type). The commit message documents the bit layout well.

One question worth considering: this changes the wire format of context IDs sent to the DSP. The commit message states "This change is compatible with polling mode and does not break non-polling behavior" — this relies on the DSP firmware being able to handle both the old GENMASK(11,4) layout and the new GENMASK(15,8) layout. Since this is v11, this backward-compatibility question has presumably been addressed in prior review rounds.

**No issues found.**

---

---
Generated by Claude Code Patch Reviewer

[PATCH v11 4/4] misc: fastrpc: Add polling mode support for fastRPC driver

[Ekansh Gupta]

For any remote call to DSP, after sending an invocation message,
fastRPC driver waits for glink response and during this time the
CPU can go into low power modes. This adds latency to overall fastrpc
call as CPU wakeup and scheduling latencies are included. Add polling
mode support with which fastRPC driver will poll continuously on a
memory after sending a message to remote subsystem which will eliminate
CPU wakeup and scheduling latencies and reduce fastRPC overhead. In case
poll timeout happens, the call will fallback to normal RPC mode.  Poll
mode can be enabled by user by using FASTRPC_IOCTL_SET_OPTION ioctl
request with FASTRPC_POLL_MODE request id.

Signed-off-by: Ekansh Gupta <ekansh.gupta@oss.qualcomm.com>
---
 drivers/misc/fastrpc.c      | 167 ++++++++++++++++++++++++++++++++++--
 include/uapi/misc/fastrpc.h |  29 +++++++
 2 files changed, 189 insertions(+), 7 deletions(-)

diff --git a/drivers/misc/fastrpc.c b/drivers/misc/fastrpc.c
index c4a3547a5c7f..aab70947482d 100644
--- a/drivers/misc/fastrpc.c
+++ b/drivers/misc/fastrpc.c
@@ -24,6 +24,8 @@
 #include <linux/of_reserved_mem.h>
 #include <linux/bits.h>
 #include <linux/bitops.h>
+#include <linux/compiler.h>
+#include <linux/iopoll.h>
 
 #define ADSP_DOMAIN_ID (0)
 #define MDSP_DOMAIN_ID (1)
@@ -38,6 +40,12 @@
 #define FASTRPC_CTX_MAX (256)
 #define FASTRPC_INIT_HANDLE	1
 #define FASTRPC_DSP_UTILITIES_HANDLE	2
+/*
+ * Maximum handle value for static handles.
+ * Static handles are pre-defined, fixed numeric values statically assigned
+ * in the IDL file or FastRPC framework.
+ */
+#define FASTRPC_MAX_STATIC_HANDLE (20)
 #define FASTRPC_CTXID_MASK GENMASK(15, 8)
 #define INIT_FILELEN_MAX (2 * 1024 * 1024)
 #define INIT_FILE_NAMELEN_MAX (128)
@@ -106,6 +114,12 @@
 
 #define miscdev_to_fdevice(d) container_of(d, struct fastrpc_device, miscdev)
 
+/* Poll response number from remote processor for call completion */
+#define FASTRPC_POLL_RESPONSE (0xdecaf)
+
+/* Polling mode timeout limit */
+#define FASTRPC_POLL_MAX_TIMEOUT_US (10000)
+
 struct fastrpc_phy_page {
 	dma_addr_t addr;	/* dma address */
 	u64 size;		/* size of contiguous region */
@@ -236,8 +250,14 @@ struct fastrpc_invoke_ctx {
 	u32 sc;
 	u64 *fdlist;
 	u32 *crc;
+	/* Poll memory that DSP updates */
+	u32 *poll_addr;
 	u64 ctxid;
 	u64 msg_sz;
+	/* work done status flag */
+	bool is_work_done;
+	/* process updates poll memory instead of glink response */
+	bool is_polled;
 	struct kref refcount;
 	struct list_head node; /* list of ctxs */
 	struct completion work;
@@ -263,6 +283,7 @@ struct fastrpc_soc_data {
 	u32 sid_pos;
 	u32 dma_addr_bits_cdsp;
 	u32 dma_addr_bits_default;
+	bool poll_mode_supported;
 };
 
 struct fastrpc_channel_ctx {
@@ -285,6 +306,7 @@ struct fastrpc_channel_ctx {
 	struct list_head invoke_interrupted_mmaps;
 	bool secure;
 	bool unsigned_support;
+	bool poll_mode_supported;
 	u64 dma_mask;
 	const struct fastrpc_soc_data *soc_data;
 };
@@ -308,6 +330,8 @@ struct fastrpc_user {
 	int client_id;
 	int pd;
 	bool is_secure_dev;
+	/* Flags poll mode state */
+	bool poll_mode;
 	/* Lock for lists */
 	spinlock_t lock;
 	/* lock for allocations */
@@ -923,7 +947,8 @@ static int fastrpc_get_meta_size(struct fastrpc_invoke_ctx *ctx)
 		sizeof(struct fastrpc_invoke_buf) +
 		sizeof(struct fastrpc_phy_page)) * ctx->nscalars +
 		sizeof(u64) * FASTRPC_MAX_FDLIST +
-		sizeof(u32) * FASTRPC_MAX_CRCLIST;
+		sizeof(u32) * FASTRPC_MAX_CRCLIST +
+		sizeof(u32);
 
 	return size;
 }
@@ -1019,6 +1044,9 @@ static int fastrpc_get_args(u32 kernel, struct fastrpc_invoke_ctx *ctx)
 	list = fastrpc_invoke_buf_start(rpra, ctx->nscalars);
 	pages = fastrpc_phy_page_start(list, ctx->nscalars);
 	ctx->fdlist = (u64 *)(pages + ctx->nscalars);
+	ctx->poll_addr = (u32 *)((uintptr_t)ctx->fdlist + sizeof(u64) * FASTRPC_MAX_FDLIST +
+			     sizeof(u32) * FASTRPC_MAX_CRCLIST);
+
 	args = (uintptr_t)ctx->buf->virt + metalen;
 	rlen = pkt_size - metalen;
 	ctx->rpra = rpra;
@@ -1188,6 +1216,71 @@ static int fastrpc_invoke_send(struct fastrpc_session_ctx *sctx,
 
 }
 
+static u32 fastrpc_read_poll_addr(struct fastrpc_invoke_ctx *ctx)
+{
+	dma_rmb();
+	return READ_ONCE(*ctx->poll_addr);
+}
+
+static int poll_for_remote_response(struct fastrpc_invoke_ctx *ctx)
+{
+	u32 val;
+	int ret;
+
+	/*
+	 * Poll until DSP writes FASTRPC_POLL_RESPONSE into *ctx->poll_addr
+	 * or until another path marks the work done.
+	 */
+	ret = read_poll_timeout_atomic(fastrpc_read_poll_addr, val,
+				       (val == FASTRPC_POLL_RESPONSE) || ctx->is_work_done, 1,
+				       FASTRPC_POLL_MAX_TIMEOUT_US, false, ctx);
+
+	if (!ret && val == FASTRPC_POLL_RESPONSE) {
+		/*
+		 * DSP writes FASTRPC_POLL_RESPONSE to signal successful
+		 * completion via the poll path.
+		 */
+		ctx->is_work_done = true;
+		ctx->retval = 0;
+	}
+
+	if (ret == -ETIMEDOUT)
+		ret = -EIO;
+
+	return ret;
+}
+
+static inline int fastrpc_wait_for_response(struct fastrpc_invoke_ctx *ctx,
+					    u32 kernel)
+{
+	int err = 0;
+
+	if (kernel) {
+		if (!wait_for_completion_timeout(&ctx->work, 10 * HZ))
+			err = -ETIMEDOUT;
+	} else {
+		err = wait_for_completion_interruptible(&ctx->work);
+	}
+
+	return err;
+}
+
+static int fastrpc_wait_for_completion(struct fastrpc_invoke_ctx *ctx,
+				       u32 kernel)
+{
+	int err;
+
+	if (ctx->is_polled) {
+		err = poll_for_remote_response(ctx);
+		if (!err)
+			return 0;
+		/* If polling timed out or failed, move to normal response mode */
+		ctx->is_polled = false;
+	}
+
+	return fastrpc_wait_for_response(ctx, kernel);
+}
+
 static int fastrpc_internal_invoke(struct fastrpc_user *fl,  u32 kernel,
 				   u32 handle, u32 sc,
 				   struct fastrpc_invoke_args *args)
@@ -1223,13 +1316,14 @@ static int fastrpc_internal_invoke(struct fastrpc_user *fl,  u32 kernel,
 	if (err)
 		goto bail;
 
-	if (kernel) {
-		if (!wait_for_completion_timeout(&ctx->work, 10 * HZ))
-			err = -ETIMEDOUT;
-	} else {
-		err = wait_for_completion_interruptible(&ctx->work);
-	}
+	/*
+	 * Set message context as polled if the call is for a user PD
+	 * dynamic module and user has enabled poll mode.
+	 */
+	if (handle > FASTRPC_MAX_STATIC_HANDLE && fl->pd == USER_PD && fl->poll_mode)
+		ctx->is_polled = true;
 
+	err = fastrpc_wait_for_completion(ctx, kernel);
 	if (err)
 		goto bail;
 
@@ -1813,6 +1907,35 @@ static int fastrpc_get_info_from_kernel(struct fastrpc_ioctl_capability *cap,
 	return 0;
 }
 
+static int fastrpc_set_option(struct fastrpc_user *fl, char __user *argp)
+{
+	struct fastrpc_ioctl_set_option opt = {0};
+	int i;
+
+	if (!fl->cctx->poll_mode_supported)
+		return -EOPNOTSUPP;
+
+	if (copy_from_user(&opt, argp, sizeof(opt)))
+		return -EFAULT;
+
+	for (i = 0; i < ARRAY_SIZE(opt.reserved); i++) {
+		if (opt.reserved[i] != 0)
+			return -EINVAL;
+	}
+
+	if (opt.request_id != FASTRPC_POLL_MODE)
+		return -EINVAL;
+
+	if (opt.value == FASTRPC_POLL_MODE_ENABLE)
+		fl->poll_mode = true;
+	else if (opt.value == FASTRPC_POLL_MODE_DISABLE)
+		fl->poll_mode = false;
+	else
+		return -EINVAL;
+
+	return 0;
+}
+
 static int fastrpc_get_dsp_info(struct fastrpc_user *fl, char __user *argp)
 {
 	struct fastrpc_ioctl_capability cap = {0};
@@ -2168,6 +2291,9 @@ static long fastrpc_device_ioctl(struct file *file, unsigned int cmd,
 	case FASTRPC_IOCTL_MEM_UNMAP:
 		err = fastrpc_req_mem_unmap(fl, argp);
 		break;
+	case FASTRPC_IOCTL_SET_OPTION:
+		err = fastrpc_set_option(fl, argp);
+		break;
 	case FASTRPC_IOCTL_GET_DSP_INFO:
 		err = fastrpc_get_dsp_info(fl, argp);
 		break;
@@ -2323,6 +2449,7 @@ static const struct fastrpc_soc_data kaanapali_soc_data = {
 	.sid_pos = 56,
 	.dma_addr_bits_cdsp = 34,
 	.dma_addr_bits_default = 32,
+	.poll_mode_supported = true,
 };
 
 static const struct fastrpc_soc_data default_soc_data = {
@@ -2331,6 +2458,29 @@ static const struct fastrpc_soc_data default_soc_data = {
 	.dma_addr_bits_default = 32,
 };
 
+/*
+ * Exception list for older platforms that use default_soc_data but whose
+ * DSP firmware supports FastRPC polling mode.
+ *
+ * NOTE: This list is intentionally closed.
+ * Do NOT add new platforms here. New SoCs must advertise polling mode
+ * support via their soc_data.
+ */
+
+static const struct of_device_id fastrpc_poll_supported_machines[] __maybe_unused = {
+	{ .compatible = "qcom,milos" },
+	{ .compatible = "qcom,qcs8300" },
+	{ .compatible = "qcom,sa8775p" },
+	{ .compatible = "qcom,sar2130p" },
+	{ .compatible = "qcom,sm8450" },
+	{ .compatible = "qcom,sm8550" },
+	{ .compatible = "qcom,sm8650" },
+	{ .compatible = "qcom,sm8750" },
+	{ .compatible = "qcom,x1e80100" },
+	{ .compatible = "qcom,x1p42100" },
+	{},
+};
+
 static int fastrpc_rpmsg_probe(struct rpmsg_device *rpdev)
 {
 	struct device *rdev = &rpdev->dev;
@@ -2397,6 +2547,8 @@ static int fastrpc_rpmsg_probe(struct rpmsg_device *rpdev)
 	secure_dsp = !(of_property_read_bool(rdev->of_node, "qcom,non-secure-domain"));
 	data->secure = secure_dsp;
 	data->soc_data = soc_data;
+	data->poll_mode_supported = soc_data->poll_mode_supported ||
+		of_machine_get_match(fastrpc_poll_supported_machines);
 
 	switch (domain_id) {
 	case ADSP_DOMAIN_ID:
@@ -2521,6 +2673,7 @@ static int fastrpc_rpmsg_callback(struct rpmsg_device *rpdev, void *data,
 	}
 
 	ctx->retval = rsp->retval;
+	ctx->is_work_done = true;
 	complete(&ctx->work);
 
 	/*
diff --git a/include/uapi/misc/fastrpc.h b/include/uapi/misc/fastrpc.h
index c6e2925f47e6..ba1ea5ed426c 100644
--- a/include/uapi/misc/fastrpc.h
+++ b/include/uapi/misc/fastrpc.h
@@ -16,6 +16,7 @@
 #define FASTRPC_IOCTL_INIT_CREATE_STATIC _IOWR('R', 9, struct fastrpc_init_create_static)
 #define FASTRPC_IOCTL_MEM_MAP		_IOWR('R', 10, struct fastrpc_mem_map)
 #define FASTRPC_IOCTL_MEM_UNMAP		_IOWR('R', 11, struct fastrpc_mem_unmap)
+#define FASTRPC_IOCTL_SET_OPTION	_IOWR('R', 12, struct fastrpc_ioctl_set_option)
 #define FASTRPC_IOCTL_GET_DSP_INFO	_IOWR('R', 13, struct fastrpc_ioctl_capability)
 
 /**
@@ -67,6 +68,28 @@ enum fastrpc_proc_attr {
 /* Fastrpc attribute for memory protection of buffers */
 #define FASTRPC_ATTR_SECUREMAP	(1)
 
+/**
+ * FASTRPC_POLL_MODE - Enable/disable poll mode for FastRPC invocations
+ *
+ * Poll mode is an optimization that allows the CPU to poll shared memory
+ * for completion instead of waiting for an interrupt-based response.
+ * This reduces latency for fast-completing operations.
+ *
+ * Restrictions:
+ * - Only supported for USER_PD (User Protection Domain)
+ * - Only applies to dynamic modules (handle > 20)
+ * - Static modules always use interrupt-based completion
+ *
+ * Values:
+ * - 0: Disable poll mode (use interrupt-based completion)
+ * - 1: Enable poll mode (poll shared memory for completion)
+ */
+#define FASTRPC_POLL_MODE	(1)
+
+/* Values for FASTRPC_POLL_MODE request */
+#define FASTRPC_POLL_MODE_DISABLE	0
+#define FASTRPC_POLL_MODE_ENABLE	1
+
 struct fastrpc_invoke_args {
 	__u64 ptr;
 	__u64 length;
@@ -133,6 +156,12 @@ struct fastrpc_mem_unmap {
 	__s32 reserved[5];
 };
 
+struct fastrpc_ioctl_set_option {
+	__u32 request_id;	/* Request type (e.g., FASTRPC_POLL_MODE) */
+	__u32 value;	/* Request-specific value */
+	__s32 reserved[6];
+};
+
 struct fastrpc_ioctl_capability {
 	__u32 unused; /* deprecated, ignored by the kernel */
 	__u32 attribute_id;
-- 
2.34.1

Re: [PATCH v11 4/4] misc: fastrpc: Add polling mode support for fastRPC driver

[Dmitry Baryshkov]

On Wed, May 20, 2026 at 12:20:47PM +0530, Ekansh Gupta wrote:
> For any remote call to DSP, after sending an invocation message,
> fastRPC driver waits for glink response and during this time the
> CPU can go into low power modes. This adds latency to overall fastrpc
> call as CPU wakeup and scheduling latencies are included. Add polling
> mode support with which fastRPC driver will poll continuously on a
> memory after sending a message to remote subsystem which will eliminate
> CPU wakeup and scheduling latencies and reduce fastRPC overhead. In case
> poll timeout happens, the call will fallback to normal RPC mode.  Poll
> mode can be enabled by user by using FASTRPC_IOCTL_SET_OPTION ioctl
> request with FASTRPC_POLL_MODE request id.
> 
> Signed-off-by: Ekansh Gupta <ekansh.gupta@oss.qualcomm.com>
> ---
>  drivers/misc/fastrpc.c      | 167 ++++++++++++++++++++++++++++++++++--
>  include/uapi/misc/fastrpc.h |  29 +++++++
>  2 files changed, 189 insertions(+), 7 deletions(-)
> 
> @@ -1813,6 +1907,35 @@ static int fastrpc_get_info_from_kernel(struct fastrpc_ioctl_capability *cap,
>  	return 0;
>  }
>  
> +static int fastrpc_set_option(struct fastrpc_user *fl, char __user *argp)
> +{
> +	struct fastrpc_ioctl_set_option opt = {0};
> +	int i;
> +
> +	if (!fl->cctx->poll_mode_supported)
> +		return -EOPNOTSUPP;

This is being handled too early. What if the user passes any other
option?

> +
> +	if (copy_from_user(&opt, argp, sizeof(opt)))
> +		return -EFAULT;
> +
> +	for (i = 0; i < ARRAY_SIZE(opt.reserved); i++) {
> +		if (opt.reserved[i] != 0)
> +			return -EINVAL;
> +	}
> +
> +	if (opt.request_id != FASTRPC_POLL_MODE)
> +		return -EINVAL;
> +
> +	if (opt.value == FASTRPC_POLL_MODE_ENABLE)
> +		fl->poll_mode = true;
> +	else if (opt.value == FASTRPC_POLL_MODE_DISABLE)
> +		fl->poll_mode = false;
> +	else
> +		return -EINVAL;
> +
> +	return 0;
> +}
> +

-- 
With best wishes
Dmitry

Re: [PATCH v11 4/4] misc: fastrpc: Add polling mode support for fastRPC driver

[Ekansh Gupta]

On 20-05-2026 19:06, Dmitry Baryshkov wrote:
> On Wed, May 20, 2026 at 12:20:47PM +0530, Ekansh Gupta wrote:
>> For any remote call to DSP, after sending an invocation message,
>> fastRPC driver waits for glink response and during this time the
>> CPU can go into low power modes. This adds latency to overall fastrpc
>> call as CPU wakeup and scheduling latencies are included. Add polling
>> mode support with which fastRPC driver will poll continuously on a
>> memory after sending a message to remote subsystem which will eliminate
>> CPU wakeup and scheduling latencies and reduce fastRPC overhead. In case
>> poll timeout happens, the call will fallback to normal RPC mode.  Poll
>> mode can be enabled by user by using FASTRPC_IOCTL_SET_OPTION ioctl
>> request with FASTRPC_POLL_MODE request id.
>>
>> Signed-off-by: Ekansh Gupta <ekansh.gupta@oss.qualcomm.com>
>> ---
>>  drivers/misc/fastrpc.c      | 167 ++++++++++++++++++++++++++++++++++--
>>  include/uapi/misc/fastrpc.h |  29 +++++++
>>  2 files changed, 189 insertions(+), 7 deletions(-)
>>
>> @@ -1813,6 +1907,35 @@ static int fastrpc_get_info_from_kernel(struct fastrpc_ioctl_capability *cap,
>>  	return 0;
>>  }
>>  
>> +static int fastrpc_set_option(struct fastrpc_user *fl, char __user *argp)
>> +{
>> +	struct fastrpc_ioctl_set_option opt = {0};
>> +	int i;
>> +
>> +	if (!fl->cctx->poll_mode_supported)
>> +		return -EOPNOTSUPP;
> 
> This is being handled too early. What if the user passes any other
> option?
okay, right. I was handling this with only poll mode as an option, I'll
move this after request_id check.>
>> +
>> +	if (copy_from_user(&opt, argp, sizeof(opt)))
>> +		return -EFAULT;
>> +
>> +	for (i = 0; i < ARRAY_SIZE(opt.reserved); i++) {
>> +		if (opt.reserved[i] != 0)
>> +			return -EINVAL;
>> +	}
>> +
>> +	if (opt.request_id != FASTRPC_POLL_MODE)
>> +		return -EINVAL;
>> +
>> +	if (opt.value == FASTRPC_POLL_MODE_ENABLE)
>> +		fl->poll_mode = true;
>> +	else if (opt.value == FASTRPC_POLL_MODE_DISABLE)
>> +		fl->poll_mode = false;
>> +	else
>> +		return -EINVAL;
>> +
>> +	return 0;
>> +}
>> +
> 

Claude review: misc: fastrpc: Add polling mode support for fastRPC driver

[Claude Code Review Bot]

Patch Review

**Overall: Needs attention on several points.**

#### 1. Concurrency issue with `is_work_done` (important)

`ctx->is_work_done` is written from two concurrent contexts without synchronization:
- The rpmsg callback (interrupt context): `ctx->is_work_done = true;` then `complete(&ctx->work);`
- The polling loop: reads `ctx->is_work_done` in `read_poll_timeout_atomic`, writes `ctx->is_work_done = true` in `poll_for_remote_response`

```c
ctx->is_work_done = true;    // in rpmsg_callback
complete(&ctx->work);
```

```c
ret = read_poll_timeout_atomic(fastrpc_read_poll_addr, val,
    (val == FASTRPC_POLL_RESPONSE) || ctx->is_work_done, ...);
if (!ret && val == FASTRPC_POLL_RESPONSE) {
    ctx->is_work_done = true;
    ctx->retval = 0;
}
```

`is_work_done` is a plain `bool` — it should be accessed with `READ_ONCE`/`WRITE_ONCE` or be an `atomic_t` to avoid data races. The compiler could cache the value in a register during the poll loop, causing the callback's write to be missed. While `read_poll_timeout_atomic` does use `READ_ONCE` on the poll value, the `ctx->is_work_done` read in the condition expression may not get the same treatment.

#### 2. Memory barrier placement in `fastrpc_read_poll_addr`

```c
static u32 fastrpc_read_poll_addr(struct fastrpc_invoke_ctx *ctx)
{
    dma_rmb();
    return READ_ONCE(*ctx->poll_addr);
}
```

The `dma_rmb()` is called before every single read inside `read_poll_timeout_atomic`. While technically correct for ordering, this is a busy-poll loop with 1 µs delay — the barrier on every iteration may add unnecessary overhead. More importantly, the `dma_rmb()` ensures ordering of DMA reads, but a matching `dma_wmb()` on the DSP side is assumed. This is fine as a convention, just worth noting.

#### 3. Race between poll completion and rpmsg callback

When `poll_for_remote_response` succeeds (val == FASTRPC_POLL_RESPONSE), it sets:
```c
ctx->is_work_done = true;
ctx->retval = 0;
```

But the rpmsg callback may still fire later (the DSP sends both a poll response AND an rpmsg callback), and overwrite `ctx->retval` with `rsp->retval`:
```c
ctx->retval = rsp->retval;
ctx->is_work_done = true;
complete(&ctx->work);
```

Then `schedule_work(&ctx->put_work)` is called in the callback, which may race with the normal cleanup path after `fastrpc_internal_invoke` returns. The `put_work` path does `fastrpc_context_put()` which decrements the refcount, so there shouldn't be a use-after-free if refcounting is correct — but this deserves verification. Does the context hold enough refcount references to survive both the poll completion path and the deferred callback?

#### 4. `poll_addr` alignment and buffer layout

```c
ctx->poll_addr = (u32 *)((uintptr_t)ctx->fdlist + sizeof(u64) * FASTRPC_MAX_FDLIST +
                     sizeof(u32) * FASTRPC_MAX_CRCLIST);
```

The `sizeof(u32)` added to `fastrpc_get_meta_size` accounts for the poll address word. The fdlist is `u64[]` (8-byte aligned), FASTRPC_MAX_FDLIST=16, FASTRPC_MAX_CRCLIST=64, so the offset is `16*8 + 64*4 = 128 + 256 = 384` bytes from fdlist, which is 4-byte aligned. This is fine for a `u32*`.

However, this `poll_addr` is within the DMA-coherent buffer (`ctx->buf`). The DSP writes `FASTRPC_POLL_RESPONSE` to this address. Is there a guarantee that the DSP knows the layout and will write to this exact offset? The commit message and code don't document how the DSP discovers this address. Presumably this is part of the FastRPC protocol, but it would be good to document.

#### 5. UAPI design: `__IOWR` vs `__IOW`

```c
#define FASTRPC_IOCTL_SET_OPTION  _IOWR('R', 12, struct fastrpc_ioctl_set_option)
```

This uses `_IOWR` (read+write), but `fastrpc_set_option` only does `copy_from_user` — it never writes back to userspace. `_IOW` would be more appropriate. Using `_IOWR` is not incorrect but is misleading about the direction.

#### 6. UAPI: gap in ioctl numbering

The new ioctl uses number 12, and the existing `FASTRPC_IOCTL_GET_DSP_INFO` uses 13. This fills the gap (number 12 was previously unused), which is fine.

#### 7. `__maybe_unused` on `fastrpc_poll_supported_machines`

```c
static const struct of_device_id fastrpc_poll_supported_machines[] __maybe_unused = {
```

The `__maybe_unused` is needed because `of_machine_get_match()` is a no-op returning NULL when `CONFIG_OF` is disabled, and the compiler would warn about the unused array. This is correct.

#### 8. `is_polled` set after `fastrpc_invoke_send`

```c
if (err)
    goto bail;

if (handle > FASTRPC_MAX_STATIC_HANDLE && fl->pd == USER_PD && fl->poll_mode)
    ctx->is_polled = true;

err = fastrpc_wait_for_completion(ctx, kernel);
```

The `is_polled` flag is set *after* `fastrpc_invoke_send` has already sent the message. This means the DSP may have already started writing to `poll_addr` before the flag is checked. Since `poll_addr` is initialized to 0 (DMA buffer is zeroed) and the DSP writes `FASTRPC_POLL_RESPONSE` on completion, this ordering is safe — the poll loop will just pick up the response. But it does mean there's a small window where the rpmsg callback could arrive and set `is_work_done` before the poll even starts, which is handled correctly since the poll condition checks `ctx->is_work_done`.

#### 9. Error code conversion

```c
if (ret == -ETIMEDOUT)
    ret = -EIO;
```

Converting `ETIMEDOUT` to `EIO` in `poll_for_remote_response` means the poll timeout is not distinguishable from an actual I/O error. Since the caller falls back to the completion-based path on any error from polling, this is acceptable, but the `-EIO` return is never actually used since the fallback path overwrites `err`:

```c
err = poll_for_remote_response(ctx);
if (!err)
    return 0;
ctx->is_polled = false;
return fastrpc_wait_for_response(ctx, kernel);
```

The `-EIO` conversion is dead code since the error value is discarded. Consider just returning any non-zero value or keeping `ETIMEDOUT`.

#### 10. No locking on `fl->poll_mode`

`fl->poll_mode` is set by the ioctl path and read from `fastrpc_internal_invoke`. There's no locking between these. If one thread is doing `SET_OPTION` while another is invoking, there's a data race on `poll_mode`. Since it's a `bool` and the worst case is enabling/disabling polling for one invocation, this is benign in practice but should use `READ_ONCE`/`WRITE_ONCE` for correctness.

---

**Summary:** Patches 1-3 are clean and ready. Patch 4 needs attention on:
- The `is_work_done` data race (use `READ_ONCE`/`WRITE_ONCE` or `atomic_t`)
- The `fl->poll_mode` data race (minor, same fix)
- The dead `-EIO` conversion
- Consider `_IOW` instead of `_IOWR` for the set option ioctl

---
Generated by Claude Code Patch Reviewer

Claude review: misc: fastrpc: Add polling mode support

[Claude Code Review Bot]

Overall Series Review

Subject: misc: fastrpc: Add polling mode support
Author: Ekansh Gupta <ekansh.gupta@oss.qualcomm.com>
Patches: 7
Reviewed: 2026-05-25T22:12:29.109860

---

This v11 series adds polling mode support to the Qualcomm FastRPC driver across 4 patches: a preparatory cleanup (fdlist move), two context ID mask changes, and the actual polling feature. The series is well-structured with logical progression. Patches 1-3 have existing Reviewed-by tags. However, patch 4 (the main feature) has several concerns around concurrency, memory barrier correctness, and UAPI design that should be addressed before merge.

---
Generated by Claude Code Patch Reviewer