lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Date:   Thu, 17 Dec 2020 11:07:45 +1030
From:   "Andrew Jeffery" <andrew@...id.au>
To:     billy_tsai@...eedtech.com
Cc:     BMC-SW@...eedtech.com, andrew@...id.au, joel@....id.au,
        linus.walleij@...aro.org, linux-arm-kernel@...ts.infradead.org,
        linux-aspeed@...ts.ozlabs.org, linux-gpio@...r.kernel.org,
        linux-kernel@...r.kernel.org, openbmc@...ts.ozlabs.org
Subject: Re: [PATCH v2] driver: aspeed: g6: Fix PWMG0 pinctrl setting

> The SCU offset for signal PWM8 in group PWM8G0 is wrong, fix it from
> SCU414 to SCU4B4.
> Besides that, When PWM8~15 of PWMG0 set it needs to clear SCU414 bits
> at the same time.

FYI, we don't need to explicitly clear SCU414[...] as part of the PWM mux 
configuration as the these bits are cleared as part of disabling the SD1* 
signal state on each pin[1]. You should be able to confirm this by compiling 
with CONFIG_DEBUG_PINCTRL=y and "debug" on the kernel commandline.

That said, it would be neat if we had some kunit tests to exercise all this, 
but it's not something I've thought deeply about.

[1] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/pinctrl/aspeed/pinctrl-aspeed.c?h=v5.10#n248

>
> Fixes: 2eda1cdec49f ("pinctrl: aspeed: Add AST2600 pinmux support")
>
> Signed-off-by: Billy Tsai <billy_tsai@...eedtech.com>
> ---
>  drivers/pinctrl/aspeed/pinctrl-aspeed-g6.c | 24 ++++++++++++++--------
>  1 file changed, 16 insertions(+), 8 deletions(-)
>
> diff --git a/drivers/pinctrl/aspeed/pinctrl-aspeed-g6.c b/drivers/pinctrl/aspeed/pinctrl-aspeed-g6.c
> index b673a44ffa3b..1dfb12a5b2ce 100644
> --- a/drivers/pinctrl/aspeed/pinctrl-aspeed-g6.c
> +++ b/drivers/pinctrl/aspeed/pinctrl-aspeed-g6.c
> @@ -367,49 +367,57 @@ FUNC_GROUP_DECL(RMII4, F24, E23, E24, E25, C25, C24, B26, B25, B24);
>
>  #define D22 40
>  SIG_EXPR_LIST_DECL_SESG(D22, SD1CLK, SD1, SIG_DESC_SET(SCU414, 8));
> -SIG_EXPR_LIST_DECL_SEMG(D22, PWM8, PWM8G0, PWM8, SIG_DESC_SET(SCU414, 8));
> +SIG_EXPR_LIST_DECL_SEMG(D22, PWM8, PWM8G0, PWM8, SIG_DESC_SET(SCU4B4, 8),

Good catch, looks like a copy/paste fail on my part :)

> +SIG_DESC_CLEAR(SCU414, 8));

As above, this should be unnecessary.

Can you confirm and remove the CLEAR()s for v3?

Cheers,

Andrew

>  PIN_DECL_2(D22, GPIOF0, SD1CLK, PWM8);
>  GROUP_DECL(PWM8G0, D22);
>
>  #define E22 41
>  SIG_EXPR_LIST_DECL_SESG(E22, SD1CMD, SD1, SIG_DESC_SET(SCU414, 9));
> -SIG_EXPR_LIST_DECL_SEMG(E22, PWM9, PWM9G0, PWM9, SIG_DESC_SET(SCU4B4, 9));
> +SIG_EXPR_LIST_DECL_SEMG(E22, PWM9, PWM9G0, PWM9, SIG_DESC_SET(SCU4B4, 9),
> +SIG_DESC_CLEAR(SCU414, 9));
>  PIN_DECL_2(E22, GPIOF1, SD1CMD, PWM9);
>  GROUP_DECL(PWM9G0, E22);
>
>  #define D23 42
>  SIG_EXPR_LIST_DECL_SESG(D23, SD1DAT0, SD1, SIG_DESC_SET(SCU414, 10));
> -SIG_EXPR_LIST_DECL_SEMG(D23, PWM10, PWM10G0, PWM10, SIG_DESC_SET(SCU4B4, 10));
> +SIG_EXPR_LIST_DECL_SEMG(D23, PWM10, PWM10G0, PWM10, SIG_DESC_SET(SCU4B4, 10),
> +SIG_DESC_CLEAR(SCU414, 10));
>  PIN_DECL_2(D23, GPIOF2, SD1DAT0, PWM10);
>  GROUP_DECL(PWM10G0, D23);
>
>  #define C23 43
>  SIG_EXPR_LIST_DECL_SESG(C23, SD1DAT1, SD1, SIG_DESC_SET(SCU414, 11));
> -SIG_EXPR_LIST_DECL_SEMG(C23, PWM11, PWM11G0, PWM11, SIG_DESC_SET(SCU4B4, 11));
> +SIG_EXPR_LIST_DECL_SEMG(C23, PWM11, PWM11G0, PWM11, SIG_DESC_SET(SCU4B4, 11),
> +SIG_DESC_CLEAR(SCU414, 11));
>  PIN_DECL_2(C23, GPIOF3, SD1DAT1, PWM11);
>  GROUP_DECL(PWM11G0, C23);
>
>  #define C22 44
>  SIG_EXPR_LIST_DECL_SESG(C22, SD1DAT2, SD1, SIG_DESC_SET(SCU414, 12));
> -SIG_EXPR_LIST_DECL_SEMG(C22, PWM12, PWM12G0, PWM12, SIG_DESC_SET(SCU4B4, 12));
> +SIG_EXPR_LIST_DECL_SEMG(C22, PWM12, PWM12G0, PWM12, SIG_DESC_SET(SCU4B4, 12),
> +SIG_DESC_CLEAR(SCU414, 12));
>  PIN_DECL_2(C22, GPIOF4, SD1DAT2, PWM12);
>  GROUP_DECL(PWM12G0, C22);
>
>  #define A25 45
>  SIG_EXPR_LIST_DECL_SESG(A25, SD1DAT3, SD1, SIG_DESC_SET(SCU414, 13));
> -SIG_EXPR_LIST_DECL_SEMG(A25, PWM13, PWM13G0, PWM13, SIG_DESC_SET(SCU4B4, 13));
> +SIG_EXPR_LIST_DECL_SEMG(A25, PWM13, PWM13G0, PWM13, SIG_DESC_SET(SCU4B4, 13),
> +SIG_DESC_CLEAR(SCU414, 13));
>  PIN_DECL_2(A25, GPIOF5, SD1DAT3, PWM13);
>  GROUP_DECL(PWM13G0, A25);
>
>  #define A24 46
>  SIG_EXPR_LIST_DECL_SESG(A24, SD1CD, SD1, SIG_DESC_SET(SCU414, 14));
> -SIG_EXPR_LIST_DECL_SEMG(A24, PWM14, PWM14G0, PWM14, SIG_DESC_SET(SCU4B4, 14));
> +SIG_EXPR_LIST_DECL_SEMG(A24, PWM14, PWM14G0, PWM14, SIG_DESC_SET(SCU4B4, 14),
> +SIG_DESC_CLEAR(SCU414, 14));
>  PIN_DECL_2(A24, GPIOF6, SD1CD, PWM14);
>  GROUP_DECL(PWM14G0, A24);
>
>  #define A23 47
>  SIG_EXPR_LIST_DECL_SESG(A23, SD1WP, SD1, SIG_DESC_SET(SCU414, 15));
> -SIG_EXPR_LIST_DECL_SEMG(A23, PWM15, PWM15G0, PWM15, SIG_DESC_SET(SCU4B4, 15));
> +SIG_EXPR_LIST_DECL_SEMG(A23, PWM15, PWM15G0, PWM15, SIG_DESC_SET(SCU4B4, 15),
> +SIG_DESC_CLEAR(SCU414, 15));
>  PIN_DECL_2(A23, GPIOF7, SD1WP, PWM15);
>  GROUP_DECL(PWM15G0, A23);
>
> --
> 2.17.1

Powered by blists - more mailing lists