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]
Date:   Wed, 14 Dec 2022 21:06:48 +0530
From:   Shazad Hussain <quic_shazhuss@...cinc.com>
To:     Brian Masney <bmasney@...hat.com>,
        Konrad Dybcio <konrad.dybcio@...aro.org>
CC:     <andersson@...nel.org>, <krzysztof.kozlowski+dt@...aro.org>,
        <robh+dt@...nel.org>, <johan+linaro@...nel.org>,
        <linux-arm-msm@...r.kernel.org>, <devicetree@...r.kernel.org>,
        <linux-kernel@...r.kernel.org>, <ahalaney@...hat.com>,
        <echanude@...hat.com>
Subject: Re: [PATCH 3/4] arm64: dts: qcom: sa8540p-ride: add qup1_i2c15 and
 qup2_i2c18 nodes



On 12/14/2022 6:00 PM, Brian Masney wrote:
> On Tue, Dec 13, 2022 at 03:48:27PM +0100, Konrad Dybcio wrote:
>>> +	qup1_i2c15_default: qup1-i2c15-state {
>>> +		mux-pins {
>>> +			pins = "gpio36", "gpio37";
>>> +			function = "qup15";
>>> +		};
>>> +
>>> +		config-pins {
>>> +			pins = "gpio36", "gpio37";
>>> +			drive-strength = <0x02>;
>>> +			bias-pull-up;
>>> +		};
>>> +	};
>>
>> You can drop mux/config-pins and have the pin properties live directly
>> under the qup1-i2cN-state node.
> 
> Hi Konrad (and Shazad below),
> 
> I need to enable 5 i2c buses (0, 1, 12, 15, 18) on this board. I tried
> the following combinations with the pin mapping configuration and the
> only one that seems to work reliably for me is what I originally had.
> 
> With the following, only 2 out of the 5 buses are detected. There's no
> i2c mesages in dmesg.
> 
>      i2c0_default: i2c0-default-state {
>          pins = "gpio135", "gpio136";
>          function = "qup15";
>      };
> 
> Next, I added a drive-strength and bias-pull-up. All 5 buses are
> detected. One bus throws read errors when I probe it with i2cdetect, two
> others 'i2cdetect -a -y $BUSNUM' takes ~5 seconds to run, and the

This I have also observed on downstream as well, where scanning all 
addresses takes some amount of time near to 5-6 seconds.

> remaining two are fast.
> 
>      i2c0_default: i2c0-default-state {
>          pins = "gpio135", "gpio136";
>          function = "qup15";
>          drive-strength = <2>;
>          bias-pull-up;
>      };
> 

This is the default config we should use.

> This is the style where i2cdetect seems to be happy for all 5 buses and
> is fast:
> 
>      i2c0_default: i2c0-default-state {
>          mux-pins {
>              pins = "gpio135", "gpio136";
>              function = "qup0";
>          };
> 
>          config-pins {
>              pins = "gpio135", "gpio136";
>              drive-strength = <2>;
>              bias-pull-up;
>          };
>      };
> 
> 
> Shazad: 'i2cdetect -a -y $BUSNUM) shows that all 5 buses have the same
> addresses listening. Is that expected? That seems a bit odd to me.
> 

Brian, even I haven't checked with all enabled, let me check this on 
other projects and with downstream as well and get back to you.

-Shazad

> [root@...alhost ~]# i2cdetect -a -y 0
> Warning: Can't use SMBus Quick Write command, will skip some addresses
>       0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f
> 00:
> 10:
> 20:
> 30: -- -- -- -- -- -- -- --
> 40:
> 50: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
> 60:
> 70:
> 
> I triple checked that I have the QUP pins defined correctly for the 5
> buses. I checked them against what's in the downstream kernel and I also
> checked them against what's in upstream's
> drivers/pinctrl/qcom/pinctrl-sc8280xp.c. This is the pin mapping that I
> have:
> 
>      i2c0: gpio135, gpio136
>      i2c1: gpio158, gpio159
>      i2c12: gpio0, gpio1
>      i2c15: gpio36, gpio37
>      i2c18: gpio66, gpio67
> 
> Brian
> 

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ