| Message ID | 1385048414-9812-2-git-send-email-dbaryshkov@gmail.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
On Thu, Nov 21, 2013 at 4:40 PM, Dmitry Eremin-Solenikov <dbaryshkov@gmail.com> wrote: > +static bool h3100_lcd_request(void) > +{ > + static bool h3100_lcd_ok; > + int rc; > + > + if (h3100_lcd_ok) > + return true; > + > + rc = gpio_request_array(h3100_lcd_gpio, ARRAY_SIZE(h3100_lcd_gpio)); > + if (rc) > + pr_err("%s: can't request GPIOs\n", __func__); > + else > + h3100_lcd_ok = true; > + > + return h3100_lcd_ok; > +} Hm hm this design pattern is somewhat strange, a run-once construct, but OK then. Acked-by: Linus Walleij <linus.walleij@linaro.org> Yours, Linus Walleij
On Tue, Nov 26, 2013 at 12:52 PM, Linus Walleij <linus.walleij@linaro.org> wrote: > On Thu, Nov 21, 2013 at 4:40 PM, Dmitry Eremin-Solenikov > <dbaryshkov@gmail.com> wrote: > >> +static bool h3100_lcd_request(void) >> +{ >> + static bool h3100_lcd_ok; >> + int rc; >> + >> + if (h3100_lcd_ok) >> + return true; >> + >> + rc = gpio_request_array(h3100_lcd_gpio, ARRAY_SIZE(h3100_lcd_gpio)); >> + if (rc) >> + pr_err("%s: can't request GPIOs\n", __func__); >> + else >> + h3100_lcd_ok = true; >> + >> + return h3100_lcd_ok; >> +} > > Hm hm this design pattern is somewhat strange, a run-once > construct, but OK then. > Acked-by: Linus Walleij <linus.walleij@linaro.org> Another possibility would be to extend sa11x0-fb driver to also provide lcd_init and lcd_shutdown callbacks (to request GPIOs). Would it be better to do so? It is really unfortunate that lcd_class drivers are not directly controlled from fbdev drivers. Hopefully CDF can solve that.
On Tue, Nov 26, 2013 at 10:07 AM, Dmitry Eremin-Solenikov <dbaryshkov@gmail.com> wrote: > On Tue, Nov 26, 2013 at 12:52 PM, Linus Walleij > <linus.walleij@linaro.org> wrote: >> On Thu, Nov 21, 2013 at 4:40 PM, Dmitry Eremin-Solenikov >> <dbaryshkov@gmail.com> wrote: >> >>> +static bool h3100_lcd_request(void) >>> +{ >>> + static bool h3100_lcd_ok; >>> + int rc; >>> + >>> + if (h3100_lcd_ok) >>> + return true; >>> + >>> + rc = gpio_request_array(h3100_lcd_gpio, ARRAY_SIZE(h3100_lcd_gpio)); >>> + if (rc) >>> + pr_err("%s: can't request GPIOs\n", __func__); >>> + else >>> + h3100_lcd_ok = true; >>> + >>> + return h3100_lcd_ok; >>> +} >> >> Hm hm this design pattern is somewhat strange, a run-once >> construct, but OK then. >> Acked-by: Linus Walleij <linus.walleij@linaro.org> > > Another possibility would be to extend sa11x0-fb driver to also > provide lcd_init and lcd_shutdown callbacks (to request GPIOs). > Would it be better to do so? > > It is really unfortunate that lcd_class drivers are not directly controlled > from fbdev drivers. Hopefully CDF can solve that. Hm it looks like there are no good solutions as long as LCDs are no separate devices. :-( Nominally you would use the new gpiod* interface and tie the GPIOs to a struct device * in this case the LCD, and since that currently has no struct device, this is OK for now... Yours, Linus Walleij
diff --git a/arch/arm/mach-sa1100/h3100.c b/arch/arm/mach-sa1100/h3100.c index b8f2b15..5b78c9f 100644 --- a/arch/arm/mach-sa1100/h3100.c +++ b/arch/arm/mach-sa1100/h3100.c @@ -28,15 +28,35 @@ /* * helper for sa1100fb */ +static struct gpio h3100_lcd_gpio[] = { + { H3100_GPIO_LCD_3V_ON, GPIOF_OUT_INIT_LOW, "LCD 3V" }, + { H3XXX_EGPIO_LCD_ON, GPIOF_OUT_INIT_LOW, "LCD ON" }, +}; + +static bool h3100_lcd_request(void) +{ + static bool h3100_lcd_ok; + int rc; + + if (h3100_lcd_ok) + return true; + + rc = gpio_request_array(h3100_lcd_gpio, ARRAY_SIZE(h3100_lcd_gpio)); + if (rc) + pr_err("%s: can't request GPIOs\n", __func__); + else + h3100_lcd_ok = true; + + return h3100_lcd_ok; +} + static void h3100_lcd_power(int enable) { - if (!gpio_request(H3XXX_EGPIO_LCD_ON, "LCD ON")) { - gpio_set_value(H3100_GPIO_LCD_3V_ON, enable); - gpio_direction_output(H3XXX_EGPIO_LCD_ON, enable); - gpio_free(H3XXX_EGPIO_LCD_ON); - } else { - pr_err("%s: can't request H3XXX_EGPIO_LCD_ON\n", __func__); - } + if (!h3100_lcd_request()) + return; + + gpio_set_value(H3100_GPIO_LCD_3V_ON, enable); + gpio_set_value(H3XXX_EGPIO_LCD_ON, enable); } static struct sa1100fb_mach_info h3100_lcd_info = { @@ -91,7 +111,6 @@ static struct gpio_default_state h3100_default_gpio[] = { { H3XXX_GPIO_COM_DCD, GPIO_MODE_IN, "COM DCD" }, { H3XXX_GPIO_COM_CTS, GPIO_MODE_IN, "COM CTS" }, { H3XXX_GPIO_COM_RTS, GPIO_MODE_OUT0, "COM RTS" }, - { H3100_GPIO_LCD_3V_ON, GPIO_MODE_OUT0, "LCD 3v" }, }; static void __init h3100_mach_init(void)
As GPIOs are going to move to platform device, there is no guarantee that they will be available at init_machine time. Request all GPIOs directly in lcd_power callback and not at init_machine time. Signed-off-by: Dmitry Eremin-Solenikov <dbaryshkov@gmail.com> --- arch/arm/mach-sa1100/h3100.c | 35 +++++++++++++++++++++++++++-------- 1 file changed, 27 insertions(+), 8 deletions(-)