[PATCH v2 2/6] phy: qcom-qmp-ufs: Add PHY Configuration support for sm8750
neil.armstrong at linaro.org
neil.armstrong at linaro.org
Mon Apr 7 00:22:40 PDT 2025
Hi,
On 25/03/2025 08:37, Nitin Rawat wrote:
>
>
> On 3/25/2025 1:04 PM, neil.armstrong at linaro.org wrote:
>> Hi,
>>
>> On 25/03/2025 04:12, Nitin Rawat wrote:
>>>
>>>
>>> On 3/24/2025 11:40 PM, Neil Armstrong wrote:
>>>> Hi,
>>>>
>>>> On 10/03/2025 22:12, Melody Olvera wrote:
>>>>> From: Nitin Rawat <quic_nitirawa at quicinc.com>
>>>>>
>>>>> Add SM8750 specific register layout and table configs. The serdes
>>>>> TX RX register offset has changed for SM8750 and hence keep UFS
>>>>> specific serdes offsets in a dedicated header file.
>>>>>
>>>>> Reviewed-by: Neil Armstrong <neil.armstrong at linaro.org>
>>>>> Co-developed-by: Manish Pandey <quic_mapa at quicinc.com>
>>>>> Signed-off-by: Manish Pandey <quic_mapa at quicinc.com>
>>>>> Signed-off-by: Nitin Rawat <quic_nitirawa at quicinc.com>
>>>>> ---
>>>>> drivers/phy/qualcomm/phy-qcom-qmp-qserdes-com-v6.h | 7 +
>>>>> .../qualcomm/phy-qcom-qmp-qserdes-txrx-ufs-v7.h | 67 ++++++++
>>>>> drivers/phy/qualcomm/phy-qcom-qmp-ufs.c | 180 +++++++++ ++ +++++++++-
>>>>> 3 files changed, 246 insertions(+), 8 deletions(-)
>>>>>
>>>>
>>>> <snip>
>>>>
>>>> This change breaks UFS on the SM8550-HDK:
>>>>
>>>> [ 7.418161] qcom-qmp-ufs-phy 1d80000.phy: phy initialization timed-out
>>>> [ 7.427021] phy phy-1d80000.phy.0: phy poweron failed --> -110
>>>> [ 7.493514] ufshcd-qcom 1d84000.ufshc: Enabling the controller failed
>>>> ...
>>>
>>> Hi Neil,
>>>
>>> Thanks for testing and reporting.
>>> I did tested this patch on SM8750 MTP, SM8750 QRD, SM8650 MTP, SM8550 MTP and SM8850 QRD all of these have rate B and hence no issue.
>>>
>>> Unfortunately only SM8550 HDK platform which UFS4.0 and RateA couldn't get tested. As we know SM8550 with gear 5 only support rate A.
>>>
>>> I was applying rate B setting without checking for mode type. Since
>>> SM8550 is only platform which support only rate A with UFS4.0 . Hence
>>> this could be the issue.
>>>
>>> Meanwhile can you help test at your end with below change and let me if it resolves for you. I will also try at my end to test as well.
>>>
>>> =============================================================================
>>> diff --git a/drivers/phy/qualcomm/phy-qcom-qmp-ufs.c b/drivers/phy/ qualcomm/phy-qcom-qmp-ufs.c
>>> index 45b3b792696e..b33e2e2b5014 100644
>>> --- a/drivers/phy/qualcomm/phy-qcom-qmp-ufs.c
>>> +++ b/drivers/phy/qualcomm/phy-qcom-qmp-ufs.c
>>> @@ -1754,7 +1754,8 @@ static void qmp_ufs_init_registers(struct qmp_ufs *qmp, const struct qmp_phy_cfg
>>> qmp_ufs_init_all(qmp, &cfg->tbls_hs_overlay[i]);
>>> }
>>>
>>> - qmp_ufs_init_all(qmp, &cfg->tbls_hs_b);
>>> + if (qmp->mode == PHY_MODE_UFS_HS_B)
>>> + qmp_ufs_init_all(qmp, &cfg->tbls_hs_b);
>>> }
>>>
>>> =================================================================================
>>
>> With this change the UFS works again.
>
> Thanks Neil for quick revert. I'll raise the official change.
Any news of a fix here ?
Thanks,
Neil
>
> Regards,
> nitin
>
>>
>> Thanks,
>> Neil
>>
>>>
>>>
>>> Thanks,
>>> Nitin
>>>
>>>>
>>>> GIT bisect points to:
>>>> b02cc9a176793b207e959701af1ec26222093b05 is the first bad commit
>>>> Author: Nitin Rawat <quic_nitirawa at quicinc.com>
>>>> Date: Mon Mar 10 14:12:30 2025 -0700
>>>>
>>>> phy: qcom-qmp-ufs: Add PHY Configuration support for sm8750
>>>>
>>>> bisect log:
>>>> git bisect start 'ff7f9b199e3f' 'v6.14-rc1'
>>>> git bisect good 36c18c562846300d4e59f1a65008800b787f4fe4
>>>> git bisect good 85cf0293c3a75726e7bc54d3efdc5dc783debc07
>>>> git bisect good b2cd73e18cec75f917d14b9188f82a2fdef64ebe
>>>> git bisect bad b247639d33ad16ea76797268fd0eef08d8027dfd
>>>> git bisect good 9b3f2dfdad1cc0ab90a0fa371c8cbee08b2446e3
>>>> git bisect bad 8dc30c3e4cf8c4e370cf08bd09eb87b0deccd3de
>>>> git bisect bad 100aeb03a437f30300894091627e4406605ee3cb
>>>> git bisect bad b2a1a2ae7818c9d8da12bf7b1983c8b9f5fb712b
>>>> git bisect good 8f831f272b4c89aa13b45bd010c2c18ad97a3f1b
>>>> git bisect good e45cc62c23428eefbae18a9b4d88d10749741bdd
>>>> git bisect bad ebf198f17b5ac967db6256f4083bbcbdcc2a3100
>>>> git bisect good 12185bc38f7667b1d895b2165a8a47335a4cf31b
>>>> git bisect bad e46e59b77a9e6f322ef1ad08a8874211f389cf47
>>>> git bisect bad b02cc9a176793b207e959701af1ec26222093b05
>>>>
>>>> CI run: https://git.codelinaro.org/linaro/qcomlt/ci/staging/cdba- tester/-/jobs/229880#L1281
>>>>
>>>> #regzbot introduced: b02cc9a17679
>>>>
>>>> Neil
>>>
>>
>
More information about the linux-phy
mailing list