linux-kernel - Re: [PATCH] ARM: davinci: turn off DDR PHY when entering deep sleep

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

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <CAM8FrKJijXr1nJFhJZDSufHmfRQXL1RPwUA7OsSTVAiGPhECuA@mail.gmail.com>
Date:	Thu, 24 May 2012 13:46:49 +0200
From:	Marcus Folkesson <marcus.folkesson@...il.com>
To:	Sekhar Nori <nsekhar@...com>
Cc:	Kevin Hilman <khilman@...com>,
	Russell King <linux@....linux.org.uk>,
	Nicolas Pitre <nicolas.pitre@...aro.org>,
	Catalin Marinas <catalin.marinas@....com>,
	linux-arm-kernel@...ts.infradead.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] ARM: davinci: turn off DDR PHY when entering deep sleep

Hi Sekhar,

2012/5/21 Sekhar Nori <nsekhar@...com>:
> Hi Marcus,
>
> Thanks for the patch.
>
> On 5/10/2012 11:22 PM, Marcus Folkesson wrote:
>> Deepsleep is the most power saving state on the davinci-platform.
>> The DDR PHY consumes about 25mW and should be turned off in this low
>> power state.
>>
>> Signed-off-by: Marcus Folkesson <marcus.folkesson@...il.com>
>> ---
>>  arch/arm/mach-davinci/include/mach/memory.h |    2 ++
>>  arch/arm/mach-davinci/sleep.S               |   10 ++++++++++
>>  2 files changed, 12 insertions(+), 0 deletions(-)
>>
>> diff --git a/arch/arm/mach-davinci/include/mach/memory.h b/arch/arm/mach-davinci/include/mach/memory.h
>> index 7873194..1b1b8ac 100644
>> --- a/arch/arm/mach-davinci/include/mach/memory.h
>> +++ b/arch/arm/mach-davinci/include/mach/memory.h
>
> This is moved to arch/arm/mach-davinci/include/mach/ddr2.h in current
> kernel.
>
>> @@ -36,6 +36,8 @@
>>  #define DDR2_MCLKSTOPEN_BIT  BIT(30)
>>  #define DDR2_LPMODEN_BIT     BIT(31)
>>
>> +#define DDR2_DRPHYC1R_OFFSET         0xe4
>> +#define DDR_PWRDNEN_BIT              BIT(6)
>>  /*
>>   * Increase size of DMA-consistent memory region
>>   */
>> diff --git a/arch/arm/mach-davinci/sleep.S b/arch/arm/mach-davinci/sleep.S
>> index 5f1e045..30713b2 100644
>> --- a/arch/arm/mach-davinci/sleep.S
>> +++ b/arch/arm/mach-davinci/sleep.S
>> @@ -57,6 +57,11 @@ ENTRY(davinci_cpu_suspend)
>>
>>       ldmia   r0, {r0-r4}
>>
>> +     /* Turn PHY off */
>> +     ldr     ip, [r0, #DDR2_DRPHYC1R_OFFSET]
>> +     orr     ip, ip, #DDR_PWRDNEN_BIT
>> +     str     ip, [r0, #DDR2_DRPHYC1R_OFFSET]
>
> Current TRM (section 14.2.13.1) specifies that this bit be set during
> DDR initialization sequence itself (done in UBL or U-Boot/SPL). I am
> checking with folks from the TI design team on whether it can be done
> later on as part of the DeepSleep sequence.
>
> It looks like IOPWRDN bit in VTPIO_CTL also needs to be set for this
> configuration take effect. You probably did not have to do it because
> the bootloader you are using already has this set?
>
> How much testing has this patch undergone? Have you tested it across
> multiple suspend-resume cycles? How much does the power consumed by DDR
> PHY go down by (and which type of DDR)?

You are right.

I now see that it is possible to set all those bits in the bootloader
and let the PHY automatically go down in idle.
This patch is therefore unnecessary.

Thank you all for the attention!


Best regards
Marcus Folkesson
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/