| Message ID | 20250224183938.3800317-4-niklas.soderlund+renesas@ragnatech.se (mailing list archive) |
|---|---|
| State | New |
| Headers | show |
| Series | media: rcar-vin: Unify notifiers and enable MC on Gen2 | expand |
Hi Niklas, On Mon, 24 Feb 2025 at 19:40, Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se> wrote: > Prepare to move Gen2 and earlier models to media controller by > generating a unique VIN group id for each VIN instance. On Gen3 and Gen4 > it is important to have a specific id in the group as media graph routes > depend on this. On Gen2 and earlier models all that will matter is to > have a unique id in the range. > > Break out the id generation to a own function keeping the logic for Gen3 > and Gen4 while generating a sequential id for Gen2 models. > > Signed-off-by: Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se> Thanks for your patch! > --- a/drivers/media/platform/renesas/rcar-vin/rcar-core.c > +++ b/drivers/media/platform/renesas/rcar-vin/rcar-core.c > @@ -114,23 +114,41 @@ static void rvin_group_release(struct kref *kref) > mutex_unlock(&rvin_group_lock); > } > > +static int rvin_group_get_id(struct rvin_dev *vin) > +{ > + struct device_node *np; > + unsigned int count; > + u32 id; > + > + switch (vin->info->model) { > + case RCAR_GEN3: > + if (!of_property_read_u32(vin->dev->of_node, "renesas,id", &id)) > + return id; > + break; Please insert a blank line here. > + default: > + count = 0; > + for_each_matching_node(np, vin->dev->driver->of_match_table) { This is a rather expensive operation. What about calling ida_alloc() instead? And probably moving the code to obtain the ID to rcar_vin_probe()? Gr{oetje,eeting}s, Geert
Hi Geert Thanks for your comments. On 2025-02-27 14:46:24 +0100, Geert Uytterhoeven wrote: > Hi Niklas, > > On Mon, 24 Feb 2025 at 19:40, Niklas Söderlund > <niklas.soderlund+renesas@ragnatech.se> wrote: > > Prepare to move Gen2 and earlier models to media controller by > > generating a unique VIN group id for each VIN instance. On Gen3 and Gen4 > > it is important to have a specific id in the group as media graph routes > > depend on this. On Gen2 and earlier models all that will matter is to > > have a unique id in the range. > > > > Break out the id generation to a own function keeping the logic for Gen3 > > and Gen4 while generating a sequential id for Gen2 models. > > > > Signed-off-by: Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se> > > Thanks for your patch! > > > --- a/drivers/media/platform/renesas/rcar-vin/rcar-core.c > > +++ b/drivers/media/platform/renesas/rcar-vin/rcar-core.c > > @@ -114,23 +114,41 @@ static void rvin_group_release(struct kref *kref) > > mutex_unlock(&rvin_group_lock); > > } > > > > +static int rvin_group_get_id(struct rvin_dev *vin) > > +{ > > + struct device_node *np; > > + unsigned int count; > > + u32 id; > > + > > + switch (vin->info->model) { > > + case RCAR_GEN3: > > + if (!of_property_read_u32(vin->dev->of_node, "renesas,id", &id)) > > + return id; > > + break; > > Please insert a blank line here. > > > + default: > > + count = 0; > > + for_each_matching_node(np, vin->dev->driver->of_match_table) { > > This is a rather expensive operation. > What about calling ida_alloc() instead? That is a good idea, did not know about this. I opted to use ida_alloc_range() to make sure the ID is in supported range. Thanks for the tip. > And probably moving the code to obtain the ID to rcar_vin_probe()? Good idea! > > Gr{oetje,eeting}s, > > Geert > > -- > Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org > > In personal conversations with technical people, I call myself a hacker. But > when I'm talking to journalists I just say "programmer" or something like that. > -- Linus Torvalds
diff --git a/drivers/media/platform/renesas/rcar-vin/rcar-core.c b/drivers/media/platform/renesas/rcar-vin/rcar-core.c index 91e871580e70..e3d2df6cd54b 100644 --- a/drivers/media/platform/renesas/rcar-vin/rcar-core.c +++ b/drivers/media/platform/renesas/rcar-vin/rcar-core.c @@ -114,23 +114,41 @@ static void rvin_group_release(struct kref *kref) mutex_unlock(&rvin_group_lock); } +static int rvin_group_get_id(struct rvin_dev *vin) +{ + struct device_node *np; + unsigned int count; + u32 id; + + switch (vin->info->model) { + case RCAR_GEN3: + if (!of_property_read_u32(vin->dev->of_node, "renesas,id", &id)) + return id; + break; + default: + count = 0; + for_each_matching_node(np, vin->dev->driver->of_match_table) { + if (np == vin->dev->of_node) + return count; + count++; + } + break; + } + + vin_err(vin, "Can't figure out group id\n"); + + return -EINVAL; +} + static int rvin_group_get(struct rvin_dev *vin, int (*link_setup)(struct rvin_group *), const struct media_device_ops *ops) { struct rvin_group *group; - u32 id; - int ret; + int id, ret; - /* Make sure VIN id is present and sane */ - ret = of_property_read_u32(vin->dev->of_node, "renesas,id", &id); - if (ret) { - vin_err(vin, "%pOF: No renesas,id property found\n", - vin->dev->of_node); - return -EINVAL; - } - - if (id >= RCAR_VIN_NUM) { + id = rvin_group_get_id(vin); + if (id < 0 || id >= RCAR_VIN_NUM) { vin_err(vin, "%pOF: Invalid renesas,id '%u'\n", vin->dev->of_node, id); return -EINVAL;
Prepare to move Gen2 and earlier models to media controller by generating a unique VIN group id for each VIN instance. On Gen3 and Gen4 it is important to have a specific id in the group as media graph routes depend on this. On Gen2 and earlier models all that will matter is to have a unique id in the range. Break out the id generation to a own function keeping the logic for Gen3 and Gen4 while generating a sequential id for Gen2 models. Signed-off-by: Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se> --- .../platform/renesas/rcar-vin/rcar-core.c | 40 ++++++++++++++----- 1 file changed, 29 insertions(+), 11 deletions(-)