Re: [PATCH] soundwire: debugfs: use controller id instead of link_id

From: Vinod Koul
Date: Mon Feb 01 2021 - 05:15:33 EST


On 21-01-21, 17:23, Srinivas Kandagatla wrote:
>
>
> On 21/01/2021 15:12, Pierre-Louis Bossart wrote:
> >
> >
> > On 1/21/21 6:03 AM, Srinivas Kandagatla wrote:
> > >
> > >
> > > On 19/01/2021 19:09, Pierre-Louis Bossart wrote:
> > > >
> > > > > currently we have
> > > > > /sys/kernel/debug/soundwire/master-*
> > > > >
> > > > > Are you suggesting that we have something like this:
> > > > >
> > > > > /sys/kernel/debug/soundwire/xyz-controller/master-<LINK-ID> ??
> > > >
> > > > Yes this is what I was thinking about.
> > >
> > > Vinod/Pierre,
> > >
> > > One Question here,
> > >
> > > Why is link_id part of "struct sdw_bus", should it not be part of
> > > "struct sdw_master_device " ?
> > >
> > > Given that "There is one Link per each Master"
> >
> > it's true that link == master == bus at the concept level.
> >
> > but we have an existing code base with different structures and we can't
> > break too many things at once.
> >
> > In the existing flow, the 'bus' is created and setup first, the
> > sdw_bus_master_add() routine takes a 'bus' argument, and the link_id is
> > already set. This routine only creates a device and in the rest of the
> > code we keep using the 'bus' pointer, so there's no real short-term
> > scope for moving the information into the 'sdw_master_device' structure
> > - that would be a lot of surgery when nothing is really broken.
>
> I totally agree!
>
> If I understand it correctly in Intel case there will be only one Link ID
> per bus.

Yes IIUC there would be one link id per bus.

the ida approach gives us unique id for each master,bus I would like to
propose using that everywhere

>
>
> Does this change look good to you?
>
> ---------------->cut<---------------
>
> diff --git a/drivers/soundwire/debugfs.c b/drivers/soundwire/debugfs.c
> index b6cad0d59b7b..f22868614f09 100644
> --- a/drivers/soundwire/debugfs.c
> +++ b/drivers/soundwire/debugfs.c
> @@ -19,13 +19,14 @@ void sdw_bus_debugfs_init(struct sdw_bus *bus)
> return;
>
> /* create the debugfs master-N */
> + bus->controller_debugfs = debugfs_create_dir(dev_name(bus->dev),
> sdw_debugfs_root);
> snprintf(name, sizeof(name), "master-%d", bus->link_id);
> - bus->debugfs = debugfs_create_dir(name, sdw_debugfs_root);
> + bus->debugfs = debugfs_create_dir(name, bus->controller_debugfs);
> }
>
> void sdw_bus_debugfs_exit(struct sdw_bus *bus)
> {
> - debugfs_remove_recursive(bus->debugfs);
> + debugfs_remove_recursive(bus->controller_debugfs);
> }
>
> #define RD_BUF (3 * PAGE_SIZE)
> diff --git a/include/linux/soundwire/sdw.h b/include/linux/soundwire/sdw.h
> index b198f471bea8..242bde30d8bd 100644
> --- a/include/linux/soundwire/sdw.h
> +++ b/include/linux/soundwire/sdw.h
> @@ -877,6 +877,7 @@ struct sdw_bus {
> struct sdw_master_prop prop;
> struct list_head m_rt_list;
> #ifdef CONFIG_DEBUG_FS
> + struct dentry *controller_debugfs;
> struct dentry *debugfs;
> #endif
> struct sdw_defer defer_msg;
>
> ---------------->cut<---------------
>
> With this change I get something like this on my board:
>
> ~# find /sys/kernel/debug/soundwire/
> /sys/kernel/debug/soundwire/
> /sys/kernel/debug/soundwire/sdw-master-2
> /sys/kernel/debug/soundwire/sdw-master-2/master-0
> /sys/kernel/debug/soundwire/sdw-master-2/master-0/sdw:0:217:2110:0:4
> /sys/kernel/debug/soundwire/sdw-master-2/master-0/sdw:0:217:2110:0:4/registers
> /sys/kernel/debug/soundwire/sdw-master-2/master-0/sdw:0:217:2110:0:3
> /sys/kernel/debug/soundwire/sdw-master-2/master-0/sdw:0:217:2110:0:3/registers
> /sys/kernel/debug/soundwire/sdw-master-1
> /sys/kernel/debug/soundwire/sdw-master-1/master-0
> /sys/kernel/debug/soundwire/sdw-master-1/master-0/sdw:0:217:10d:0:3
> /sys/kernel/debug/soundwire/sdw-master-1/master-0/sdw:0:217:10d:0:3/registers
> /sys/kernel/debug/soundwire/sdw-master-0
> /sys/kernel/debug/soundwire/sdw-master-0/master-0
> /sys/kernel/debug/soundwire/sdw-master-0/master-0/sdw:0:217:10d:0:4
> /sys/kernel/debug/soundwire/sdw-master-0/master-0/sdw:0:217:10d:0:4/registers
>
>
>
> Thanks,
> srini

--
~Vinod