| 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 linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Date: Fri, 22 Sep 2017 12:31:12 +0200
From: Nicolas Ferre <nicolas.ferre@...rochip.com>
To: Romain Izard <romain.izard.pro@...il.com>,
Alexandre Belloni <alexandre.belloni@...e-electrons.com>,
Boris Brezillon <boris.brezillon@...e-electrons.com>,
Michael Turquette <mturquette@...libre.com>,
Stephen Boyd <sboyd@...eaurora.org>,
Ludovic Desroches <ludovic.desroches@...rochip.com>,
Wenyou Yang <wenyou.yang@...el.com>,
Josh Wu <rainyfeeling@...look.com>,
David Woodhouse <dwmw2@...radead.org>,
Brian Norris <computersforpeace@...il.com>,
Marek Vasut <marek.vasut@...il.com>,
Cyrille Pitchen <cyrille.pitchen@...ev4u.fr>,
Thierry Reding <thierry.reding@...il.com>,
Richard Genoud <richard.genoud@...il.com>,
Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
Alan Stern <stern@...land.harvard.edu>
CC: <linux-clk@...r.kernel.org>, <linux-kernel@...r.kernel.org>,
<linux-mtd@...ts.infradead.org>, <linux-pwm@...r.kernel.org>,
<linux-serial@...r.kernel.org>, <linux-usb@...r.kernel.org>,
<linux-arm-kernel@...ts.infradead.org>,
Romain Izard <romain.izard@...ile-devices.fr>
Subject: Re: [PATCH v2 3/9] clk: at91: pmc: Support backup for programmable
clocks
On 15/09/2017 at 16:04, Romain Izard wrote:
> From: Romain Izard <romain.izard@...ile-devices.fr>
>
> When an AT91 programmable clock is declared in the device tree, register
> it into the Power Management Controller driver. On entering suspend mode,
> the driver saves and restores the Programmable Clock registers to support
> the backup mode for these clocks.
>
> Signed-off-by: Romain Izard <romain.izard.pro@...il.com>
Romain,
Some nitpicking and one comment. But on the overall patch, here is my:
Acked-by: Nicolas Ferre <nicolas.ferre@...rochip.com>
See below:
> ---
> Changes in v2:
> * register PCKs on clock startup
>
> drivers/clk/at91/clk-programmable.c | 2 ++
> drivers/clk/at91/pmc.c | 27 +++++++++++++++++++++++++++
> drivers/clk/at91/pmc.h | 2 ++
> 3 files changed, 31 insertions(+)
>
> diff --git a/drivers/clk/at91/clk-programmable.c b/drivers/clk/at91/clk-programmable.c
> index 85a449cf61e3..0e6aab1252fc 100644
> --- a/drivers/clk/at91/clk-programmable.c
> +++ b/drivers/clk/at91/clk-programmable.c
> @@ -204,6 +204,8 @@ at91_clk_register_programmable(struct regmap *regmap,
> if (ret) {
> kfree(prog);
> hw = ERR_PTR(ret);
Nit: "else" not needed.
> + } else {
> + pmc_register_pck(id);
> }
>
> return hw;
> diff --git a/drivers/clk/at91/pmc.c b/drivers/clk/at91/pmc.c
> index 07dc2861ad3f..3910b7537152 100644
> --- a/drivers/clk/at91/pmc.c
> +++ b/drivers/clk/at91/pmc.c
> @@ -22,6 +22,7 @@
> #include "pmc.h"
>
> #define PMC_MAX_IDS 128
> +#define PMC_MAX_PCKS 8
>
> int of_at91_get_clk_range(struct device_node *np, const char *propname,
> struct clk_range *range)
> @@ -50,6 +51,7 @@ EXPORT_SYMBOL_GPL(of_at91_get_clk_range);
> static struct regmap *pmcreg;
>
> static u8 registered_ids[PMC_MAX_IDS];
> +static u8 registered_pcks[PMC_MAX_PCKS];
>
> static struct
> {
> @@ -66,8 +68,10 @@ static struct
> u32 pcr[PMC_MAX_IDS];
> u32 audio_pll0;
> u32 audio_pll1;
> + u32 pckr[PMC_MAX_PCKS];
> } pmc_cache;
>
> +/* Clock ID 0 is invalid */
(read: so we can use the 0 value as an indicator that this place in the
table hasn't been filled, so unused)
> void pmc_register_id(u8 id)
> {
> int i;
> @@ -82,6 +86,21 @@ void pmc_register_id(u8 id)
> }
> }
>
> +/* Programmable Clock 0 is valid */
I understand the rationale behind these ^^ two comments, but I would
like that it's more explicit. Saying that you will store the pck id as
(id + 1) and that you would have to invert this operation while using
the stored id.
Maybe add a comment about this transformation to the struct definition
as well...
> +void pmc_register_pck(u8 pck)
> +{
> + int i;
> +
> + for (i = 0; i < PMC_MAX_PCKS; i++) {
> + if (registered_pcks[i] == 0) {
> + registered_pcks[i] = pck + 1;
> + break;
> + }
> + if (registered_pcks[i] == (pck + 1))
> + break;
> + }
> +}
> +
> static int pmc_suspend(void)
> {
> int i;
> @@ -103,6 +122,10 @@ static int pmc_suspend(void)
> regmap_read(pmcreg, AT91_PMC_PCR,
> &pmc_cache.pcr[registered_ids[i]]);
> }
> + for (i = 0; registered_pcks[i]; i++) {
> + u8 num = registered_pcks[i] - 1;
Nit: declaration are better made at the beginning of the function. This
lead to a checkpatch warning:
"WARNING: Missing a blank line after declarations"
> + regmap_read(pmcreg, AT91_PMC_PCKR(num), &pmc_cache.pckr[num]);
> + }
>
> return 0;
> }
> @@ -143,6 +166,10 @@ static void pmc_resume(void)
> pmc_cache.pcr[registered_ids[i]] |
> AT91_PMC_PCR_CMD);
> }
> + for (i = 0; registered_pcks[i]; i++) {
> + u8 num = registered_pcks[i] - 1;
Ditto
> + regmap_write(pmcreg, AT91_PMC_PCKR(num), pmc_cache.pckr[num]);
> + }
>
> if (pmc_cache.uckr & AT91_PMC_UPLLEN)
> mask |= AT91_PMC_LOCKU;
> diff --git a/drivers/clk/at91/pmc.h b/drivers/clk/at91/pmc.h
> index 858e8ef7e8db..d22b1fa9ecdc 100644
> --- a/drivers/clk/at91/pmc.h
> +++ b/drivers/clk/at91/pmc.h
> @@ -31,8 +31,10 @@ int of_at91_get_clk_range(struct device_node *np, const char *propname,
>
> #ifdef CONFIG_PM
> void pmc_register_id(u8 id);
> +void pmc_register_pck(u8 pck);
> #else
> static inline void pmc_register_id(u8 id) {}
> +static inline void pmc_register_pck(u8 pck) {}
> #endif
>
> #endif /* __PMC_H_ */
>
--
Nicolas Ferre
Powered by blists - more mailing lists