linux-kernel - Re: [PATCH] mmc: meson-gx: Remove setting of CMD_CFG

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] [day] [month] [year] [list]
Message-ID: <7c81e31f-3ece-4a95-80db-dfe7fcd4de0a@amlogic.com>
Date:   Thu, 26 Oct 2023 15:40:31 +0800
From:   Rong Chen <Rong.Chen@...ogic.com>
To:     Jerome Brunet <jbrunet@...libre.com>,
        Ulf Hansson <ulf.hansson@...aro.org>,
        Neil Armstrong <narmstrong@...libre.com>,
        Kevin Hilman <khilman@...libre.com>,
        Martin Blumenstingl <martin.blumenstingl@...glemail.com>
Cc:     linux-mmc <linux-mmc@...r.kernel.org>,
        linux-arm-kernel <linux-arm-kernel@...ts.infradead.org>,
        linux-amlogic <linux-amlogic@...ts.infradead.org>,
        linux-kernel <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] mmc: meson-gx: Remove setting of CMD_CFG_ERROR

Hi Jerome,

I tested on meson-g12a-u200 and meson-sm1-sei610 board successfully.

*meson-g12a-u200:*

[ 1.308436] mmc1: new HS200 MMC card at address 0001 [ 1.309223] 
mmcblk1: mmc1:0001 AJTD4R 14.6 GiB [ 1.314724] mmcblk1: p1 p2 p3 p4 p5 
p6 p7 p8 p9 p10 p11 p12 p13 p14 p15 p16 [ 1.321724] mmcblk1boot0: 
mmc1:0001 AJTD4R 4.00 MiB [ 1.325727] mmcblk1boot1: mmc1:0001 AJTD4R 
4.00 MiB [ 1.330427] mmcblk1rpmb: mmc1:0001 AJTD4R 4.00 MiB, chardev 
(511:0) [ 11.368251] platform cpufreq-dt: deferred probe pending [ 
29.695649] mmc0: new high speed SDHC card at address 0007 [ 29.696402] 
mmcblk0: mmc0:0007 SD16G 14.5 GiB [ 29.708053] mmcblk0: p1 [ 48.132967] 
mmc0: card 0007 removed

*meson-sm1-sei610:*

[ 1.815416] mmc1: new HS200 MMC card at address 0001 iled:[ 1.819542] 
mmcblk1: mmc1:0001 016G30 14.7 GiB No such file or d[ 1.826207] 
mmcblk1boot0: mmc1:0001 016G30 4.00 MiB [ 1.831692] mmcblk1boot1: 
mmc1:0001 016G30 4.00 MiB [ 1.836250] mmcblk1rpmb: mmc1:0001 016G30 4.00 
MiB, chardev (511:0)

[ 5.857437] mmc0: new high speed SDHC card at address 0007

[ 5.858046] mmcblk0: mmc0:0007 SD16G 14.5 GiB [ 5.869727] mmcblk0: p1

[ 9.958449] mmc0: card 0007 removed


在 2023/9/27 17:30, Jerome Brunet 写道:
> [ EXTERNAL EMAIL ]
>
> On Wed 27 Sep 2023 at 16:51, "Rong.Chen@...ogic.com" <Rong.Chen@...ogic.com> wrote:
>
>> Hi Jerome,
>>
>> Socs containing previous controllers are gxl, gxm, gxbb，g12b，g12a，sm1，axg, a1，s4 and t7.
> That is mostly all the SoC we know of in mainline ... This why the way
> you explained is confusing.
>
> And below, you say 'on latest controllers', okay but which ones ?
>
>> The bit CMD_CFG_ERROR is used for selecting SG LINK of data address on latest controllers，but
>> data address don't use SG LINK at present，so it will cause an IRQ and stop working on latest controllers.
>> Therefore, remove this setting.
> Not totally sure what SG LINK but okay.
>
> While the change itself is fine you really need reformulate the description.
> Based on your explanation, something like:
> ---
> For the t7 and older SoC families, the CMD_CFG_ERROR has no effect.
> Starting from SoC family XYZ, it will select SG LINK on error but this
> is causing a problem with the data address ... blablabla
> ---
>
> Please expand a bit on this, try to make it clearer.
> Also, please insert the proper fixes tag
>
> Last, make sure tp reply with plain text emails, otherwise lists will
> drop your reply and please do not top post.
>
> Thx
>
>> Thanks.
>> -------------------------------------------------------------------------------------------------------------------------
>> Rong.Chen@...ogic.com
>>
>>
>>   From: Jerome Brunet
>>   Date: 2023-09-27 15:43
>>   To: Rong.Chen; Ulf Hansson; Neil Armstrong; Kevin Hilman; Martin Blumenstingl
>>   CC: linux-mmc; linux-arm-kernel; linux-amlogic; linux-kernel
>>   Subject: Re: [PATCH] mmc: meson-gx: Remove setting of CMD_CFG_ERROR
>>   [You don't often get email from jbrunet@...libre.com. Learn why this is important at
>>   https://aka.ms/LearnAboutSenderIdentification ]
>>
>>   [ EXTERNAL EMAIL ]
>>
>>   s
>>   On Wed 27 Sep 2023 at 15:25, "Rong.Chen" <Rong.Chen@...ogic.com> wrote:
>>
>>   > From: Rong Chen <rong.chen@...ogic.com>
>>   >
>>   > Setting bit CMD_CFG_ERROR has non effects on previous controllers.
>>
>>   Which previous controllers ? v2, v3 ? Can you list the SoCs please ?
>>
>>   > However, it will cause an IRQ and stop working on latest controllers,
>>
>>   Isn't it desirable to stop on error ? seems like a sane thing to do.
>>
>>   > This bit is used for selecting SG LINK of data address.
>>   > Don't set the bit CMD_CFG_ERROR anymore to resolve this issuse.
>>
>>   I'm not sure I understand the actual issue being resolved.
>>   Could you provide more details please ?
>>
>>   >
>>   > Therefore, remove this setting.
>>   >
>>   > Signed-off-by: Rong Chen <rong.chen@...ogic.com>
>>   > ---
>>   >  drivers/mmc/host/meson-gx-mmc.c | 1 -
>>   >  1 file changed, 1 deletion(-)
>>   >
>>   > diff --git a/drivers/mmc/host/meson-gx-mmc.c b/drivers/mmc/host/meson-gx-mmc.c
>>   > index 9837dab096e6..c7c067b9415a 100644
>>   > --- a/drivers/mmc/host/meson-gx-mmc.c
>>   > +++ b/drivers/mmc/host/meson-gx-mmc.c
>>   > @@ -801,7 +801,6 @@ static void meson_mmc_start_cmd(struct mmc_host *mmc, struct mmc_command *cmd)
>>   >
>>   >       cmd_cfg |= FIELD_PREP(CMD_CFG_CMD_INDEX_MASK, cmd->opcode);
>>   >       cmd_cfg |= CMD_CFG_OWNER;  /* owned by CPU */
>>   > -     cmd_cfg |= CMD_CFG_ERROR; /* stop in case of error */
>>   >
>>   >       meson_mmc_set_response_bits(cmd, &cmd_cfg);