Re: [PATCH 07/11] rvtrace: Add trace ramsink driver

From: Bo Gan

Date: Fri Oct 10 2025 - 20:38:39 EST

On 10/8/25 03:35, Mayuresh Chitale wrote:

On Tue, Oct 7, 2025 at 1:21 PM Bo Gan <ganboing@xxxxxxxxx <mailto:ganboing@xxxxxxxxx>> wrote:
>
> On 10/1/25 23:07, Anup Patel wrote:
> > From: Mayuresh Chitale <mchitale@xxxxxxxxxxxxxxxx <mailto:mchitale@xxxxxxxxxxxxxxxx>>
> >
> > Add initial implementation of RISC-V trace ramsink driver. The ramsink
> > is defined in the RISC-V Trace Control Interface specification.
> >
> > Co-developed-by: Anup Patel <apatel@xxxxxxxxxxxxxxxx <mailto:apatel@xxxxxxxxxxxxxxxx>>
> > Signed-off-by: Anup Patel <apatel@xxxxxxxxxxxxxxxx <mailto:apatel@xxxxxxxxxxxxxxxx>>
> > Signed-off-by: Mayuresh Chitale <mchitale@xxxxxxxxxxxxxxxx <mailto:mchitale@xxxxxxxxxxxxxxxx>>
> > ---
> > drivers/hwtracing/rvtrace/Kconfig | 8 +
> > drivers/hwtracing/rvtrace/Makefile | 1 +
> > drivers/hwtracing/rvtrace/rvtrace-ramsink.c | 198 ++++++++++++++++++++
> > 3 files changed, 207 insertions(+)
> > create mode 100644 drivers/hwtracing/rvtrace/rvtrace-ramsink.c
> >
> > diff --git a/drivers/hwtracing/rvtrace/Kconfig b/drivers/hwtracing/rvtrace/Kconfig
> > index ba35c05f3f54..aef7e9989165 100644
> > --- a/drivers/hwtracing/rvtrace/Kconfig
> > +++ b/drivers/hwtracing/rvtrace/Kconfig
> > @@ -21,3 +21,11 @@ config RVTRACE_ENCODER
> > default y
> > help
> > This driver provides support for RISC-V Trace Encoder component.
> > +
> > +config RVTRACE_RAMSINK
> > + tristate "RISC-V Trace Ramsink driver"
> > + depends on RVTRACE
> > + default y
> > + help
> > + This driver provides support for Risc-V E-Trace Ramsink
> > + component.
> > diff --git a/drivers/hwtracing/rvtrace/Makefile b/drivers/hwtracing/rvtrace/Makefile
> > index f320693a1fc5..122e575da9fb 100644
> > --- a/drivers/hwtracing/rvtrace/Makefile
> > +++ b/drivers/hwtracing/rvtrace/Makefile
> > @@ -3,3 +3,4 @@
> > obj-$(CONFIG_RVTRACE) += rvtrace.o
> > rvtrace-y := rvtrace-core.o rvtrace-platform.o
> > obj-$(CONFIG_RVTRACE_ENCODER) += rvtrace-encoder.o
> > +obj-$(CONFIG_RVTRACE_RAMSINK) += rvtrace-ramsink.o
> > diff --git a/drivers/hwtracing/rvtrace/rvtrace-ramsink.c b/drivers/hwtracing/rvtrace/rvtrace-ramsink.c
> > new file mode 100644
> > index 000000000000..7bd0cf1e4dfd
> > --- /dev/null
> > +++ b/drivers/hwtracing/rvtrace/rvtrace-ramsink.c
> > @@ -0,0 +1,198 @@
> > +// SPDX-License-Identifier: GPL-2.0
> > +/*
> > + * Copyright (c) 2025 Ventana Micro Systems Inc.
> > + */
> > +
> > +#include <linux/device.h>
> > +#include <linux/io.h>
> > +#include <linux/of.h>
> > +#include <linux/of_graph.h>
> > +#include <linux/platform_device.h>
> > +#include <linux/property.h>
> > +#include <linux/dma-mapping.h>
> > +#include <linux/rvtrace.h>
> > +#include <linux/types.h>
> > +#include <linux/sizes.h>
> > +
> > +#define RVTRACE_RAMSINK_STARTLOW_OFF 0x010
> > +#define RVTRACE_RAMSINK_STARTHIGH_OFF 0x014
> > +#define RVTRACE_RAMSINK_LIMITLOW_OFF 0x018
> > +#define RVTRACE_RAMSINK_LIMITHIGH_OFF 0x01c
> > +#define RVTRACE_RAMSINK_WPLOW_OFF 0x020
> > +#define RVTRACE_RAMSINK_WPHIGH_OFF 0x024
> > +#define RVTRACE_RAMSINK_RPLOW_OFF 0x028
> > +#define RVTRACE_RAMSINK_RPHIGH_OFF 0x02c
> > +
> > +struct rvtrace_ramsink_priv {
> > + size_t size;
> > + void *va;
> > + dma_addr_t start;
> > + dma_addr_t end;
> > + /* WP from prev iteration */
> > + dma_addr_t prev_head;
> > +};
> > +
> > +struct trace_buf {
> > + void *base;
> > + size_t size;
> > + int cur, len;
> > +};
> > +
> > +static void tbuf_to_pbuf_copy(struct trace_buf *src, struct trace_buf *dst)
> > +{
> > + int bytes_dst, bytes_src, bytes;
> > + void *dst_addr, *src_addr;
> > +
> > + while (src->size) {
> > + src_addr = src->base + src->cur;
> > + dst_addr = dst->base + dst->cur;
> > +
> > + if (dst->len - dst->cur < src->size)
> > + bytes_dst = dst->len - dst->cur;
> > + else
> > + bytes_dst = src->size;
> > + if (src->len - src->cur < src->size)
> > + bytes_src = src->len - src->cur;
> > + else
> > + bytes_src = src->size;
> > + bytes = bytes_dst < bytes_src ? bytes_dst : bytes_src;
> > + memcpy(dst_addr, src_addr, bytes);
> > + dst->cur = (dst->cur + bytes) % dst->len;
> > + src->cur = (src->cur + bytes) % src->len;
> > + src->size -= bytes;
> > + }
> > +}
> > +
> > +static size_t rvtrace_ramsink_copyto_auxbuf(struct rvtrace_component *comp,
> > + struct rvtrace_perf_auxbuf *buf)
> > +{
> > + struct rvtrace_ramsink_priv *priv = dev_get_drvdata(&comp->dev);
> > + struct trace_buf src, dst;
> > + u32 wp_low, wp_high;
> > + u64 buf_cur_head;
> > + size_t size;
> > +
> > + wp_low = rvtrace_read32(comp->pdata, RVTRACE_RAMSINK_WPLOW_OFF);
> > + wp_high = rvtrace_read32(comp->pdata, RVTRACE_RAMSINK_WPHIGH_OFF);
> > + buf_cur_head = (u64)(wp_high) << 32 | wp_low;
> > +
> > + if (buf_cur_head == priv->prev_head)
> > + return 0;
> > +
> > + dst.base = buf->base;
> > + dst.len = buf->length;
> > + dst.cur = buf->pos;
> > + dst.size = 0;
> > +
> > + src.base = priv->va;
> > + src.len = priv->end - priv->start;
> > + if (buf_cur_head > priv->prev_head) {
> > + src.size = buf_cur_head - priv->prev_head;
> > + } else {
> > + src.size = priv->end - priv->prev_head;
> > + src.size += buf_cur_head - priv->start;
> > + }
> > +
> > + src.cur = buf_cur_head - priv->start;
> > + size = src.size;
> > + tbuf_to_pbuf_copy(&src, &dst);
> > + buf->pos = dst.cur;
> > + priv->prev_head = buf_cur_head;
> > +
> > + return size;
> > +}
> > +
> > +static int rvtrace_ramsink_setup(struct rvtrace_component *comp)
> > +{
> > + struct rvtrace_ramsink_priv *priv;
> > +
> > + priv = devm_kzalloc(&comp->dev, sizeof(*priv), GFP_KERNEL);
> > + if (!priv)
> > + return -ENOMEM;
> > + dev_set_drvdata(&comp->dev, priv);
> > +
> > + priv->size = SZ_4M;
> Can we make this size dynamically determined? 4M seems inadequate. This is
> exceedingly so if the RAM sink is linked to a funnel, where you can have
> many harts dumping traces into this sink.
>
> > + priv->va = dma_alloc_coherent(&comp->dev, priv->size, &priv->start, GFP_KERNEL);
> > + if (!priv->va)
> > + return -ENOMEM;
> > + priv->end = priv->start + priv->size;
> > + priv->prev_head = priv->start;
> > +
> > + /* Setup ram sink addresses */
> > + rvtrace_write32(comp->pdata, lower_32_bits(priv->start), RVTRACE_RAMSINK_STARTLOW_OFF);
> > + rvtrace_write32(comp->pdata, upper_32_bits(priv->start), RVTRACE_RAMSINK_STARTHIGH_OFF);
> > + rvtrace_write32(comp->pdata, lower_32_bits(priv->start), RVTRACE_RAMSINK_WPLOW_OFF);
> > + rvtrace_write32(comp->pdata, upper_32_bits(priv->start), RVTRACE_RAMSINK_WPHIGH_OFF);
> > + /* Limit address needs to be set to end - 4 so that HW doesn't cause an overflow. */> + rvtrace_write32(comp->pdata, lower_32_bits(priv->end - 0x4), RVTRACE_RAMSINK_LIMITLOW_OFF);
> Should not hardcode 4 as the trace write width. Control Interface Spec
> chapter 7.1 has the proper instruction on how to probe for this width:
>
> "Not every value may be settable in trRamStart/Limit registers. Value
> written may be trimmed (for example aligned on a particular 2^N boundary)
> and a trace tool should verify values being written"
>
Thanks for the comments. I will incorporate those in v2.

@Mayuresh, I went over the spec again. I don't see the handling off wrap.
You need to detect if the buffer has wrapped around by testing bit 0 of
trRamWPLow. So:

a. trRamWPLow needs masking bit 0 before being used as address to read
trace data.
b. If wrapped, then the trace buffer is actually scattered in two parts:
[WP, Limit + width) and [Start, WP). During copying, you need to use
the right portion of the trace buffer.

ref: https://github.com/ganboing/riscv-trace-umd/blob/master/rvtrace/funnel.py#L260

BTW, again your mail wasn't delivered to the mailing-list for some reason
Please check: I couldn't find your name on

https://lists.infradead.org/pipermail/linux-riscv/2025-October/thread.html
https://lore.kernel.org/all/20251002060732.100213-1-apatel@xxxxxxxxxxxxxxxx/

> > + rvtrace_write32(comp->pdata, upper_32_bits(priv->end), RVTRACE_RAMSINK_LIMITHIGH_OFF);
> > +
> > + return 0;
> > +}
> > +
> > +static void rvtrace_ramsink_cleanup(struct rvtrace_component *comp)
> > +{
> > + struct rvtrace_ramsink_priv *priv = dev_get_drvdata(&comp->dev);
> > +
> > + dma_free_coherent(&comp->dev, priv->size, priv->va, priv->start);
> > +}
> > +
> > +static int rvtrace_ramsink_probe(struct rvtrace_component *comp)
> > +{
> > + int ret;
> > +
> > + ret = rvtrace_ramsink_setup(comp);
> > + if (ret)
> > + return dev_err_probe(&comp->dev, ret, "failed to setup ramsink.\n");
> > +
> > + ret = rvtrace_enable_component(comp);
> > + if (ret)
> > + return dev_err_probe(&comp->dev, ret, "failed to enable ramsink.\n");
> > +
> > + return ret;
> > +}
> > +
> > +static void rvtrace_ramsink_remove(struct rvtrace_component *comp)
> > +{
> > + int ret;
> > +
> > + ret = rvtrace_disable_component(comp);
> > + if (ret)
> > + dev_err(&comp->dev, "failed to disable ramsink.\n");
> > +
> > + rvtrace_ramsink_cleanup(comp);
> > +}
> > +
> > +static struct rvtrace_component_id rvtrace_ramsink_ids[] = {
> > + { .type = RVTRACE_COMPONENT_TYPE_RAMSINK,
> > + .version = rvtrace_component_mkversion(1, 0), },
> > + {},
> > +};
> > +
> > +static struct rvtrace_driver rvtrace_ramsink_driver = {
> > + .id_table = rvtrace_ramsink_ids,
> > + .copyto_auxbuf = rvtrace_ramsink_copyto_auxbuf,
> > + .probe = rvtrace_ramsink_probe,
> > + .remove = rvtrace_ramsink_remove,
> > + .driver = {
> > + .name = "rvtrace-ramsink",
> > + },
> > +};
> > +
> > +static int __init rvtrace_ramsink_init(void)
> > +{
> > + return rvtrace_register_driver(&rvtrace_ramsink_driver);
> > +}
> > +
> > +static void __exit rvtrace_ramsink_exit(void)
> > +{
> > + rvtrace_unregister_driver(&rvtrace_ramsink_driver);
> > +}
> > +
> > +module_init(rvtrace_ramsink_init);
> > +module_exit(rvtrace_ramsink_exit);
> > +
> > +/* Module information */
> > +MODULE_AUTHOR("Mayuresh Chitale <mchitale@xxxxxxxxxxxxxxxx <mailto:mchitale@xxxxxxxxxxxxxxxx>>");
> > +MODULE_DESCRIPTION("RISC-V Trace Ramsink Driver");
> > +MODULE_LICENSE("GPL");
>
> Bo