| Message ID | 20191021105822.20271-5-lee.jones@linaro.org (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | Simplify MFD Core | expand |
On Mon, Oct 21, 2019 at 11:58:17AM +0100, Lee Jones wrote: > CS5535 is the only user of mfd_clone_cell(). It makes more sense to > register child devices in the traditional way and remove the quite > bespoke mfd_clone_cell() call from the MFD API. > > Signed-off-by: Lee Jones <lee.jones@linaro.org> > --- > drivers/mfd/cs5535-mfd.c | 24 ++++++++++++++++++------ > 1 file changed, 18 insertions(+), 6 deletions(-) > > diff --git a/drivers/mfd/cs5535-mfd.c b/drivers/mfd/cs5535-mfd.c > index 053e33447808..96a99ac13384 100644 > --- a/drivers/mfd/cs5535-mfd.c > +++ b/drivers/mfd/cs5535-mfd.c > @@ -57,9 +57,17 @@ static struct mfd_cell cs5535_mfd_cells[] = { > }, > }; > > -static const char *olpc_acpi_clones[] = { > - "olpc-xo1-pm-acpi", > - "olpc-xo1-sci-acpi" > +static struct mfd_cell cs5535_olpc_mfd_cells[] = { > + { > + .name = "olpc-xo1-pm-acpi", > + .num_resources = 1, > + .resources = &cs5535_mfd_resources[ACPI_BAR], > + }, > + { > + .name = "olpc-xo1-sci-acpi", > + .num_resources = 1, > + .resources = &cs5535_mfd_resources[ACPI_BAR], > + }, Is the cs5535-acpi cell actually used by anything? I think it was only ever used as a template and can be removed; I didn't spot any driver that uses it. PS If the cell were removed then my review comment on the previous patch becomes moot ;-) > }; > > static int cs5535_mfd_probe(struct pci_dev *pdev, > @@ -105,10 +113,14 @@ static int cs5535_mfd_probe(struct pci_dev *pdev, > goto err_remove_devices; > } > > - err = mfd_clone_cell("cs5535-acpi", olpc_acpi_clones, > - ARRAY_SIZE(olpc_acpi_clones)); > + err = mfd_add_devices(&pdev->dev, PLATFORM_DEVID_NONE, > + cs5535_olpc_mfd_cells, > + ARRAY_SIZE(cs5535_olpc_mfd_cells), > + NULL, 0, NULL); > if (err) { > - dev_err(&pdev->dev, "Failed to clone MFD cell\n"); > + dev_err(&pdev->dev, > + "Failed to add CS5532 OLPC sub-devices: %d\n", > + err); > goto err_release_acpi; > } > } > -- > 2.17.1 >
On Mon, 21 Oct 2019, Daniel Thompson wrote: > On Mon, Oct 21, 2019 at 11:58:17AM +0100, Lee Jones wrote: > > CS5535 is the only user of mfd_clone_cell(). It makes more sense to > > register child devices in the traditional way and remove the quite > > bespoke mfd_clone_cell() call from the MFD API. > > > > Signed-off-by: Lee Jones <lee.jones@linaro.org> > > --- > > drivers/mfd/cs5535-mfd.c | 24 ++++++++++++++++++------ > > 1 file changed, 18 insertions(+), 6 deletions(-) > > > > diff --git a/drivers/mfd/cs5535-mfd.c b/drivers/mfd/cs5535-mfd.c > > index 053e33447808..96a99ac13384 100644 > > --- a/drivers/mfd/cs5535-mfd.c > > +++ b/drivers/mfd/cs5535-mfd.c > > @@ -57,9 +57,17 @@ static struct mfd_cell cs5535_mfd_cells[] = { > > }, > > }; > > > > -static const char *olpc_acpi_clones[] = { > > - "olpc-xo1-pm-acpi", > > - "olpc-xo1-sci-acpi" > > +static struct mfd_cell cs5535_olpc_mfd_cells[] = { > > + { > > + .name = "olpc-xo1-pm-acpi", > > + .num_resources = 1, > > + .resources = &cs5535_mfd_resources[ACPI_BAR], > > + }, > > + { > > + .name = "olpc-xo1-sci-acpi", > > + .num_resources = 1, > > + .resources = &cs5535_mfd_resources[ACPI_BAR], > > + }, > > Is the cs5535-acpi cell actually used by anything? I think it was only > ever used as a template and can be removed; I didn't spot any driver that > uses it. I did think about this, but I assumed removing it at this stage would make the resource matching below more convoluted. I'll take another look at see what I can do. > PS If the cell were removed then my review comment on the previous patch > becomes moot ;-) > > > > }; > > > > static int cs5535_mfd_probe(struct pci_dev *pdev, > > @@ -105,10 +113,14 @@ static int cs5535_mfd_probe(struct pci_dev *pdev, > > goto err_remove_devices; > > } > > > > - err = mfd_clone_cell("cs5535-acpi", olpc_acpi_clones, > > - ARRAY_SIZE(olpc_acpi_clones)); > > + err = mfd_add_devices(&pdev->dev, PLATFORM_DEVID_NONE, > > + cs5535_olpc_mfd_cells, > > + ARRAY_SIZE(cs5535_olpc_mfd_cells), > > + NULL, 0, NULL); > > if (err) { > > - dev_err(&pdev->dev, "Failed to clone MFD cell\n"); > > + dev_err(&pdev->dev, > > + "Failed to add CS5532 OLPC sub-devices: %d\n", > > + err); > > goto err_release_acpi; > > } > > }
diff --git a/drivers/mfd/cs5535-mfd.c b/drivers/mfd/cs5535-mfd.c index 053e33447808..96a99ac13384 100644 --- a/drivers/mfd/cs5535-mfd.c +++ b/drivers/mfd/cs5535-mfd.c @@ -57,9 +57,17 @@ static struct mfd_cell cs5535_mfd_cells[] = { }, }; -static const char *olpc_acpi_clones[] = { - "olpc-xo1-pm-acpi", - "olpc-xo1-sci-acpi" +static struct mfd_cell cs5535_olpc_mfd_cells[] = { + { + .name = "olpc-xo1-pm-acpi", + .num_resources = 1, + .resources = &cs5535_mfd_resources[ACPI_BAR], + }, + { + .name = "olpc-xo1-sci-acpi", + .num_resources = 1, + .resources = &cs5535_mfd_resources[ACPI_BAR], + }, }; static int cs5535_mfd_probe(struct pci_dev *pdev, @@ -105,10 +113,14 @@ static int cs5535_mfd_probe(struct pci_dev *pdev, goto err_remove_devices; } - err = mfd_clone_cell("cs5535-acpi", olpc_acpi_clones, - ARRAY_SIZE(olpc_acpi_clones)); + err = mfd_add_devices(&pdev->dev, PLATFORM_DEVID_NONE, + cs5535_olpc_mfd_cells, + ARRAY_SIZE(cs5535_olpc_mfd_cells), + NULL, 0, NULL); if (err) { - dev_err(&pdev->dev, "Failed to clone MFD cell\n"); + dev_err(&pdev->dev, + "Failed to add CS5532 OLPC sub-devices: %d\n", + err); goto err_release_acpi; } }
CS5535 is the only user of mfd_clone_cell(). It makes more sense to register child devices in the traditional way and remove the quite bespoke mfd_clone_cell() call from the MFD API. Signed-off-by: Lee Jones <lee.jones@linaro.org> --- drivers/mfd/cs5535-mfd.c | 24 ++++++++++++++++++------ 1 file changed, 18 insertions(+), 6 deletions(-)