Re: [PATCH v2 6/9] coresight: Store in-connections as well as out-connections

From: Mike Leach
Date: Tue Mar 21 2023 - 13:56:57 EST

Next message: Laurent Pinchart: "Re: [PATCH v2 6/6] media: starfive: Add Starfive Camera Subsystem driver"
Previous message: Slade Watkins: "Re: [PATCH 5.4 00/57] 5.4.238-rc2 review"
In reply to: Suzuki K Poulose: "Re: [PATCH v2 6/9] coresight: Store in-connections as well as out-connections"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

Hi James

On Thu, 16 Mar 2023 at 20:23, Suzuki K Poulose <suzuki.poulose@xxxxxxx> wrote:
>
> On 10/03/2023 16:06, James Clark wrote:
> > This will allow CATU to get its associated ETR in a generic way where
> > currently the enable path has some hard coded searches which avoid
> > the need to store input connections.
> >
> > Signed-off-by: James Clark <james.clark@xxxxxxx>
> > ---
> > drivers/hwtracing/coresight/coresight-core.c | 56 +++++++++++++++--
> > .../hwtracing/coresight/coresight-platform.c | 61 ++++++++++++++++---
> > drivers/hwtracing/coresight/coresight-sysfs.c | 1 -
> > include/linux/coresight.h | 25 ++++++++
> > 4 files changed, 130 insertions(+), 13 deletions(-)
> >
> > diff --git a/drivers/hwtracing/coresight/coresight-core.c b/drivers/hwtracing/coresight/coresight-core.c
> > index f457914e445e..a8ba7493c09a 100644
> > --- a/drivers/hwtracing/coresight/coresight-core.c
> > +++ b/drivers/hwtracing/coresight/coresight-core.c
> > @@ -59,6 +59,7 @@ const u32 coresight_barrier_pkt[4] = {0x7fffffff, 0x7fffffff, 0x7fffffff, 0x7fff
> > EXPORT_SYMBOL_GPL(coresight_barrier_pkt);
> >
> > static const struct cti_assoc_op *cti_assoc_ops;
> > +static int coresight_fixup_inputs(struct coresight_device *csdev);
> >
> > ssize_t coresight_simple_show_pair(struct device *_dev,
> > struct device_attribute *attr, char *buf)
> > @@ -1369,6 +1370,35 @@ static int coresight_fixup_orphan_conns(struct coresight_device *csdev)
> > csdev, coresight_orphan_match);
> > }
> >
> > +/*
> > + * Device connections are discovered before one/both devices have been created,
> > + * so inputs must be added later.
> > + */
> > +static int coresight_fixup_inputs(struct coresight_device *csdev)
> > +{
> > + int i, ret = 0;
> > + struct coresight_connection *out_conn;
> > + struct coresight_connection in_conn;
> > +
> > + for (i = 0; i < csdev->pdata->nr_outconns; i++) {
> > + out_conn = &csdev->pdata->out_conns[i];
> > + if (!out_conn->remote_dev || !out_conn->remote_dev->pdata)
> > + continue;
> > +
> > + /* Reverse local/remote relationships for inputs */
> > + in_conn.remote_dev = csdev;
> > + in_conn.remote_port = out_conn->port;
> > + in_conn.port = out_conn->remote_port;
> > + in_conn.remote_fwnode = csdev->dev.fwnode;
> > + ret = coresight_add_in_conn(out_conn->remote_dev->dev.parent,
> > + out_conn->remote_dev->pdata,
> > + &in_conn);
> > + if (ret)
> > + return ret;
> > + }
> > +
> > + return 0;
> > +}
> >
> > static int coresight_fixup_device_conns(struct coresight_device *csdev)
> > {
> > @@ -1427,11 +1457,20 @@ static int coresight_remove_match(struct device *dev, void *data)
>
> Removing the connection should be much simpler now with the inports
> connection tracking.
>
> i.e., coresight_remove_conns() need not iterate over all the devices on
> the coresight bus. We had to do that to find all the devices connecting
> to the removed device.
>
> Instead with the in_conns we could readily get the csdev instance and
> remove the connection from those devices.
>
> Similarly for the out_conns can give you the remove device for removing
> the in_conns on that device.
>
> i.e,
>
> static void coresight_remove_conns(csdev)
> {
> struct coresight_device *remote;
>
> for (i = 0; i < csdev->pdata->nr_outconns; i++) {
> remote = csdev->data->out_conns[i].remote_dev;
>
> if (!remote)
> continue;
>
> for (j = 0; j < remote->pdata->nr_inconns; j++) {
> if (remote->pdata->in_conns[i].remote_dev == csdev)
> /* Clear in_conns[i] on remote */
>
> }
> }
>
> Similarly for in_conns
> for (i = 0; i < csdev->pdata->nr_inconns; i++) {
> remote = csdev->data->in_conns[i].remote_dev;
>
> if (!remote)
> continue;
>
> for (j = 0; j < remote->pdata->nr_outconns; j++) {
> if (remote->pdata->out_conns[i].remote_dev == csdev)
> /* Clear out_conns[i] on remote */
> }
>
>
> > */
> > fwnode_handle_put(conn->remote_fwnode);
> > conn->remote_fwnode = NULL;
> > + conn->remote_dev = NULL;
> > + /* No need to continue */
> > + break;
> > + }
> > + }
> > + for (i = 0; i < iterator->pdata->nr_inconns; i++) {
> > + conn = &iterator->pdata->in_conns[i];
> > + if (csdev == conn->remote_dev) {
> > + conn->remote_fwnode = NULL;
> > + conn->remote_dev = NULL;
> > /* No need to continue */
> > break;
> > }
> > }
> > -
> > /*
> > * Returning '0' ensures that all known component on the
> > * bus will be checked.
> > @@ -1552,21 +1591,28 @@ void coresight_release_platform_data(struct coresight_device *csdev,
> >
> > for (i = 0; i < pdata->nr_outconns; i++) {
> > /* If we have made the links, remove them now */
> > - if (csdev && conns[i].remote_dev)
> > + if (csdev && conns[i].remote_dev) {
> > coresight_remove_links(csdev, &conns[i]);
> > + conns[i].remote_dev = NULL;
> > + }
> > +
> > /*
> > * Drop the refcount and clear the handle as this device
> > * is going away
> > */
> > if (conns[i].remote_fwnode) {
> > fwnode_handle_put(conns[i].remote_fwnode);
> > - pdata->out_conns[i].remote_fwnode = NULL;
> > + conns[i].remote_fwnode = NULL;
> > }
> > }
> > + for (i = 0; i < pdata->nr_inconns; i++) {
> > + pdata->in_conns[i].remote_dev = NULL;
> > + pdata->in_conns[i].remote_fwnode = NULL;
> > + }
> > +
> > if (csdev)
> > coresight_remove_conns_sysfs_group(csdev);
> > }
> > -
> > struct coresight_device *coresight_register(struct coresight_desc *desc)
> > {
> > int ret;
> > @@ -1659,6 +1705,8 @@ struct coresight_device *coresight_register(struct coresight_desc *desc)
> > ret = coresight_create_conns_sysfs_group(csdev);
> > if (!ret)
> > ret = coresight_fixup_device_conns(csdev);
> > + if (!ret)
> > + ret = coresight_fixup_inputs(csdev);
> Don't we also need to fixup the "input" connections on this
> device ? It would be good move all of this connection stuff
> into a wrapper function.
>
> > if (!ret)
> > ret = coresight_fixup_orphan_conns(csdev);
> >
> > diff --git a/drivers/hwtracing/coresight/coresight-platform.c b/drivers/hwtracing/coresight/coresight-platform.c
> > index 16553f7dde12..20e3351cbdc2 100644
> > --- a/drivers/hwtracing/coresight/coresight-platform.c
> > +++ b/drivers/hwtracing/coresight/coresight-platform.c
> > @@ -20,8 +20,7 @@
> >
> > #include "coresight-priv.h"
> > /*
> > - * coresight_alloc_conns: Allocate connections record for each output
> > - * port from the device.
> > + * coresight_alloc_conns: Allocate connections record for each input/output device.
> > */
> > static int coresight_alloc_conns(struct device *dev,
> > struct coresight_platform_data *pdata)
> > @@ -33,7 +32,14 @@ static int coresight_alloc_conns(struct device *dev,
> > if (!pdata->out_conns)
> > return -ENOMEM;
> > }
> > -
> > + if (pdata->nr_inconns) {
> > + pdata->in_conns = devm_krealloc_array(dev, pdata->in_conns,
> > + pdata->nr_inconns,
> > + sizeof(*pdata->in_conns),
> > + GFP_KERNEL | __GFP_ZERO);
> > + if (!pdata->in_conns)
> > + return -ENOMEM;
> > + }
>
> devm_krealloc_array() skips any allocation if the requested new_size <=
> origin_size. So this should be fine for the out_conns, to not realloc
> the space again unnecessarily.
>
> > return 0;
> > }
> >
> > @@ -79,6 +85,45 @@ int coresight_add_conn(struct device *dev,
> > }
> > EXPORT_SYMBOL_GPL(coresight_add_conn);
>
> Should this be named as add_out_conn in the previous patch
> just to make it explicit and easier to read.
>
> >
> > +/*
> > + * Add a connection in the first free slot, or realloc
> > + * if there is no space.
> > + *
> > + * Do nothing if the connection already exists because inputs are
> > + * fixed up multiple times.
> > + */
> > +int coresight_add_in_conn(struct device *dev,
> > + struct coresight_platform_data *pdata,
>
> minor nit: Is there any reason why this can't simply be:
>
> int coresight_add_in_conn(struct coresight_device *csdev,
> struct coresight_connection *conn)
> {
> struct device *dev = csdev->dev.parent;
> struct coresight_platform_data *pdata = csdev->pdata;
>
> > + struct coresight_connection *conn)
>
>
> Suzuki
>
>
> > +{
> > + int ret;
> > + struct coresight_connection *free_conn = NULL;
> > + int i;
> > +
> > + /* Search for a free slot or exit if a duplicate is found */
> > + if (pdata->in_conns) {
> > + for (i = 0; i < pdata->nr_inconns; ++i) {
> > + if (!free_conn && !pdata->in_conns[i].remote_fwnode)
> > + free_conn = &pdata->in_conns[i];
> > + if (pdata->in_conns[i].remote_fwnode ==
> > + conn->remote_fwnode)
> > + return 0;
> > + }
> > + }
> > +

As with the out conns this assumes there can never be a match after a free slot.

Mike

> > + if (!free_conn) {
> > + pdata->nr_inconns++;
> > + ret = coresight_alloc_conns(dev, pdata);
> > + if (ret)
> > + return ret;
> > + free_conn = &pdata->in_conns[pdata->nr_inconns - 1];
> > + }
> > +
> > + *free_conn = *conn;
> > + return 0;
> > +}
> > +EXPORT_SYMBOL_GPL(coresight_add_in_conn);
> > +
> > static struct device *
> > coresight_find_device_by_fwnode(struct fwnode_handle *fwnode)
> > {
> > @@ -249,7 +294,7 @@ static int of_coresight_get_cpu(struct device *dev)
> >
> > /*
> > * of_coresight_parse_endpoint : Parse the given output endpoint @ep
> > - * and fill the connection information in @conn
> > + * and fill the connection information in @in_conn and @out_conn
> > *
> > * Parses the local port, remote device name and the remote port.
> > *
> > @@ -333,14 +378,14 @@ static int of_get_coresight_platform_data(struct device *dev,
> > /* Get the number of input and output port for this component */
> > of_coresight_get_ports(node, &pdata->nr_inconns, &pdata->nr_outconns);
> >
> > - /* If there are no output connections, we are done */
> > - if (!pdata->nr_outconns)
> > - return 0;
> > -
> > ret = coresight_alloc_conns(dev, pdata);
> > if (ret)
> > return ret;
> >
> > + /* If there are no output connections, we are done */
> > + if (!pdata->nr_outconns)
> > + return 0;
> > +
> > parent = of_coresight_get_output_ports_node(node);
> > /*
> > * If the DT uses obsoleted bindings, the ports are listed
> > diff --git a/drivers/hwtracing/coresight/coresight-sysfs.c b/drivers/hwtracing/coresight/coresight-sysfs.c
> > index 3da9868d9237..2abf9639ac0f 100644
> > --- a/drivers/hwtracing/coresight/coresight-sysfs.c
> > +++ b/drivers/hwtracing/coresight/coresight-sysfs.c
> > @@ -202,5 +202,4 @@ void coresight_remove_links(struct coresight_device *orig,
> > devm_kfree(&orig->dev, conn->link->orig_name);
> > devm_kfree(&orig->dev, conn->link);
> > conn->link = NULL;
> > - conn->remote_dev = NULL;
> > }
> > diff --git a/include/linux/coresight.h b/include/linux/coresight.h
> > index 47fa58d6981d..fd268b24c761 100644
> > --- a/include/linux/coresight.h
> > +++ b/include/linux/coresight.h
> > @@ -110,6 +110,7 @@ struct coresight_platform_data {
> > int nr_inconns;
> > int nr_outconns;
> > struct coresight_connection *out_conns;
> > + struct coresight_connection *in_conns;
>
> Please document the field above the structure.
>
> Suzuki
>
>
> > };
> >
> > /**
> > @@ -177,6 +178,27 @@ struct coresight_desc {
> > * @remote_fwnode once the remote's coresight_device has
> > * been created.
> > * @link: Representation of the connection as a sysfs link.
> > + *
> > + * The full connection structure looks like this, where in_conns store references to
> > + * the parent device in the same remote_dev member as output connections.
> > + *
> > + * +-----------------------------+ +-----------------------------+
> > + * |coresight_device | |coresight_connection |
> > + * |-----------------------------| |-----------------------------|
> > + * ---->| | | |
> > + * | | | | remote_dev*|------
> > + * | |pdata->out_conns[nr_outconns]|----------->| | |
> > + * | | | | | |
> > + * | +-----------------------------+ +-----------------------------+ |
> > + * | |
> > + * | +-----------------------------+ +-----------------------------+ |
> > + * | |coresight_connection | |coresight_device | |
> > + * | |-----------------------------| |------------------------------ |
> > + * | | | | |<-----
> > + * -----|remote_dev* | | |
> > + * | |<-----------|pdata->in_conns[nr_inconns] |
> > + * | | | |
> > + * +-----------------------------+ +-----------------------------+
> > */
> > struct coresight_connection {
> > int port;
> > @@ -619,5 +641,8 @@ struct coresight_platform_data *coresight_get_platform_data(struct device *dev);
> > int coresight_add_conn(struct device *dev,
> > struct coresight_platform_data *pdata,
> > const struct coresight_connection *conn);
> > +int coresight_add_in_conn(struct device *dev,
> > + struct coresight_platform_data *pdata,
> > + struct coresight_connection *conn);
> >
> > #endif /* _LINUX_COREISGHT_H */
>

--
Mike Leach
Principal Engineer, ARM Ltd.
Manchester Design Centre. UK

Next message: Laurent Pinchart: "Re: [PATCH v2 6/6] media: starfive: Add Starfive Camera Subsystem driver"
Previous message: Slade Watkins: "Re: [PATCH 5.4 00/57] 5.4.238-rc2 review"
In reply to: Suzuki K Poulose: "Re: [PATCH v2 6/9] coresight: Store in-connections as well as out-connections"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]