| 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
| ||
|
Message-ID: <ed32a4fd-973a-ec59-c695-11411dd47dd1@microchip.com>
Date: Wed, 27 Nov 2019 10:59:20 +0000
From: <Nicolas.Ferre@...rochip.com>
To: <alexandre.belloni@...tlin.com>, <Claudiu.Beznea@...rochip.com>
CC: <linux@...linux.org.uk>, <Ludovic.Desroches@...rochip.com>,
<sre@...nel.org>, <linux-arm-kernel@...ts.infradead.org>,
<linux-kernel@...r.kernel.org>, <linux-pm@...r.kernel.org>
Subject: Re: [PATCH v2 02/17] ARM: at91: pm: move SAM9X60's PM under its own
SoC config flag
On 27/11/2019 at 11:07, Alexandre Belloni wrote:
> On 27/11/2019 08:06:47+0000, Claudiu.Beznea@...rochip.com wrote:
>>
>>
>> On 26.11.2019 23:28, Alexandre Belloni wrote:
>>> EXTERNAL EMAIL: Do not click links or open attachments unless you know the content is safe
>>>
>>> On 26/11/2019 15:12:06+0200, Claudiu Beznea wrote:
>>>> Move SAM9X60's PM part under SoC config flag. This allows the building
>>>> of SAM9X60 platform withouth depending on CONFIG_SOC_AT91SAM9 flag,
>>>> allowing us to select only necessary config flags for SAM9X60.
>>>>
>>>
>>> I'm really wondering, how much space does that really save?
>>>
>>> The net benefit seems to be very small...
>>
>> Not that much, indeed. We want to be independent of SOC_AT91SAM9.
>>
>
> The question is why? I don't see the technical benefit but I
> definitively see the maintenance burden of having two separate configs
> doing almost the same thing.
The AT91SAM9 config embeds a bunch of earlier drivers/clock definitions
that are not needed anymore in the new SAM9X60. Likewise, some sam9x60
new things are not needed at all for the older sam9 series.
There is somehow a generation gap between them...
I would like that we preserve the possibility to only embed the sam9x60
alone in a tailored kernel configuration, basically how it is done for
our SAMA5s. I know that we're talking about ~100s of KB here, but
however, it's easy to do now and it could make a difference when
targeting low spec systems.
Maintaining the SAMA5 as separate config options never proved us to be
difficult to do.
Best regards,
Nicolas
>>>> Signed-off-by: Claudiu Beznea <claudiu.beznea@...rochip.com>
>>>> ---
>>>> arch/arm/mach-at91/Makefile | 1 +
>>>> arch/arm/mach-at91/at91sam9.c | 18 ------------------
>>>> arch/arm/mach-at91/pm.c | 2 +-
>>>> arch/arm/mach-at91/sam9x60.c | 34 ++++++++++++++++++++++++++++++++++
>>>> 4 files changed, 36 insertions(+), 19 deletions(-)
>>>> create mode 100644 arch/arm/mach-at91/sam9x60.c
>>>>
>>>> diff --git a/arch/arm/mach-at91/Makefile b/arch/arm/mach-at91/Makefile
>>>> index de64301dcff2..f565490f1b70 100644
>>>> --- a/arch/arm/mach-at91/Makefile
>>>> +++ b/arch/arm/mach-at91/Makefile
>>>> @@ -6,6 +6,7 @@
>>>> # CPU-specific support
>>>> obj-$(CONFIG_SOC_AT91RM9200) += at91rm9200.o
>>>> obj-$(CONFIG_SOC_AT91SAM9) += at91sam9.o
>>>> +obj-$(CONFIG_SOC_SAM9X60) += sam9x60.o
>>>> obj-$(CONFIG_SOC_SAMA5) += sama5.o
>>>> obj-$(CONFIG_SOC_SAMV7) += samv7.o
>>>>
>>>> diff --git a/arch/arm/mach-at91/at91sam9.c b/arch/arm/mach-at91/at91sam9.c
>>>> index bf629c90c758..7e572189a5eb 100644
>>>> --- a/arch/arm/mach-at91/at91sam9.c
>>>> +++ b/arch/arm/mach-at91/at91sam9.c
>>>> @@ -31,21 +31,3 @@ DT_MACHINE_START(at91sam_dt, "Atmel AT91SAM9")
>>>> .init_machine = at91sam9_init,
>>>> .dt_compat = at91_dt_board_compat,
>>>> MACHINE_END
>>>> -
>>>> -static void __init sam9x60_init(void)
>>>> -{
>>>> - of_platform_default_populate(NULL, NULL, NULL);
>>>> -
>>>> - sam9x60_pm_init();
>>>> -}
>>>> -
>>>> -static const char *const sam9x60_dt_board_compat[] __initconst = {
>>>> - "microchip,sam9x60",
>>>> - NULL
>>>> -};
>>>> -
>>>> -DT_MACHINE_START(sam9x60_dt, "Microchip SAM9X60")
>>>> - /* Maintainer: Microchip */
>>>> - .init_machine = sam9x60_init,
>>>> - .dt_compat = sam9x60_dt_board_compat,
>>>> -MACHINE_END
>>>> diff --git a/arch/arm/mach-at91/pm.c b/arch/arm/mach-at91/pm.c
>>>> index d5af6aedc02c..56a6a49b19e2 100644
>>>> --- a/arch/arm/mach-at91/pm.c
>>>> +++ b/arch/arm/mach-at91/pm.c
>>>> @@ -805,7 +805,7 @@ void __init at91rm9200_pm_init(void)
>>>>
>>>> void __init sam9x60_pm_init(void)
>>>> {
>>>> - if (!IS_ENABLED(CONFIG_SOC_AT91SAM9))
>>>> + if (!IS_ENABLED(CONFIG_SOC_SAM9X60))
>>>> return;
>>>>
>>>> at91_pm_modes_init();
>>>> diff --git a/arch/arm/mach-at91/sam9x60.c b/arch/arm/mach-at91/sam9x60.c
>>>> new file mode 100644
>>>> index 000000000000..d8c739d25458
>>>> --- /dev/null
>>>> +++ b/arch/arm/mach-at91/sam9x60.c
>>>> @@ -0,0 +1,34 @@
>>>> +// SPDX-License-Identifier: GPL-2.0+
>>>> +/*
>>>> + * Setup code for SAM9X60.
>>>> + *
>>>> + * Copyright (C) 2019 Microchip Technology Inc. and its subsidiaries
>>>> + *
>>>> + * Author: Claudiu Beznea <claudiu.beznea@...rochip.com>
>>>> + */
>>>> +
>>>> +#include <linux/of.h>
>>>> +#include <linux/of_platform.h>
>>>> +
>>>> +#include <asm/mach/arch.h>
>>>> +#include <asm/system_misc.h>
>>>> +
>>>> +#include "generic.h"
>>>> +
>>>> +static void __init sam9x60_init(void)
>>>> +{
>>>> + of_platform_default_populate(NULL, NULL, NULL);
>>>> +
>>>> + sam9x60_pm_init();
>>>> +}
>>>> +
>>>> +static const char *const sam9x60_dt_board_compat[] __initconst = {
>>>> + "microchip,sam9x60",
>>>> + NULL
>>>> +};
>>>> +
>>>> +DT_MACHINE_START(sam9x60_dt, "Microchip SAM9X60")
>>>> + /* Maintainer: Microchip */
>>>> + .init_machine = sam9x60_init,
>>>> + .dt_compat = sam9x60_dt_board_compat,
>>>> +MACHINE_END
>>>> --
>>>> 2.7.4
>>>>
>>>
>>> --
>>> Alexandre Belloni, Bootlin
>>> Embedded Linux and Kernel engineering
>>> https://bootlin.com
>>>
>
> --
> Alexandre Belloni, Bootlin
> Embedded Linux and Kernel engineering
> https://bootlin.com
>
--
Nicolas Ferre
Powered by blists - more mailing lists