Problems about the mainline linux and this repo

[gtxzsxxk]

Are you still maintaining the mainline linux's fbtft drivers? I see that there are quite a few differences between the mainline linux and this repo. Take a probable mistake in the mainline fbtft-core.c for example.

mainline code

static void fbtft_reset(struct fbtft_par *par)
{
    if (!par->gpio.reset)
        return;

    fbtft_par_dbg(DEBUG_RESET, par, "%s()\n", __func__);

    gpiod_set_value_cansleep(par->gpio.reset, 1);
    usleep_range(20, 40);
    gpiod_set_value_cansleep(par->gpio.reset, 0);
    msleep(120);

    gpiod_set_value_cansleep(par->gpio.cs, 1);  /* Activate chip */
}

this repo

void fbtft_reset(struct fbtft_par *par)
{
    if (par->gpio.reset == -1)
        return;
    fbtft_par_dbg(DEBUG_RESET, par, "%s()\n", __func__);
    gpio_set_value(par->gpio.reset, 0);
    udelay(20);
    gpio_set_value(par->gpio.reset, 1);
    mdelay(120);
}

Note that for most tft-lcds the reset logic is negative, i.e. when the reset pin is low-level, the lcd will reset. And the mainline code is causing problems. The comment is activate chip but what is it doing? It is resetting the lcds because it keeps the reset pin low and it is disabling the lcds too because it keeps cs high. I believe the mainline code is having some typo because when I make the par->gpio.cs into par->gpio.reset every thing begins to work.

[notro]

Maintainer status from the frontpage wiki: fbtft is currently listed as orphaned in MAINTANERS so it doesn't have someone looking after it, at least officially.

Status of the code in this repo: https://github.com/notro/fbtft/blob/master/README#L4-L6

Unfortunately the reset polarity was changed in this commit: https://github.com/torvalds/linux/commit/ec03c2104365ead0a33627c05e685093eed3eaef

the cs gpio is used for chip select on a parallell bus.

[gtxzsxxk]

Your comment explains. Thanks.