linux-kernel - Re: [PATCH 14/33] arm_mpam: Probe hardware to find the supported partid/pmg values

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: <0b4e74a8-6a8a-429d-92a7-35df8ef04e9c@arm.com>
Date: Thu, 13 Nov 2025 16:43:32 +0000
From: Ben Horgan <ben.horgan@....com>
To: Fenghua Yu <fenghuay@...dia.com>, james.morse@....com
Cc: amitsinght@...vell.com, baisheng.gao@...soc.com,
 baolin.wang@...ux.alibaba.com, bobo.shaobowang@...wei.com,
 carl@...amperecomputing.com, catalin.marinas@....com, dakr@...nel.org,
 dave.martin@....com, david@...hat.com, dfustini@...libre.com,
 gregkh@...uxfoundation.org, gshan@...hat.com, guohanjun@...wei.com,
 jeremy.linton@....com, jonathan.cameron@...wei.com, kobak@...dia.com,
 lcherian@...vell.com, lenb@...nel.org, linux-acpi@...r.kernel.org,
 linux-arm-kernel@...ts.infradead.org, linux-kernel@...r.kernel.org,
 lpieralisi@...nel.org, peternewman@...gle.com, quic_jiles@...cinc.com,
 rafael@...nel.org, robh@...nel.org, rohit.mathew@....com,
 scott@...amperecomputing.com, sdonthineni@...dia.com, sudeep.holla@....com,
 tan.shaopeng@...itsu.com, will@...nel.org, xhao@...ux.alibaba.com,
 Shaopeng Tan <tan.shaopeng@...fujitsu.com>
Subject: Re: [PATCH 14/33] arm_mpam: Probe hardware to find the supported
 partid/pmg values

Hi Fenghua,

On 11/13/25 03:50, Fenghua Yu wrote:
> Hi, Ben and James,
> 
> On 11/7/25 04:34, Ben Horgan wrote:
>> From: James Morse <james.morse@....com>
> [SNIP]
> 
>> +static struct mpam_msc_ris *mpam_get_or_create_ris(struct mpam_msc *msc,
>> +                           u8 ris_idx)
>> +{
>> +    int err;
>> +    struct mpam_msc_ris *ris;
>> +
>> +    lockdep_assert_held(&mpam_list_lock);
>> +
>> +    if (!test_bit(ris_idx, &msc->ris_idxs)) {
>> +        err = mpam_ris_create_locked(msc, ris_idx, MPAM_CLASS_UNKNOWN,
>> +                         0, 0);
>> +        if (err)
>> +            return ERR_PTR(err);
>> +    }
>> +
>> +    list_for_each_entry(ris, &msc->ris, msc_list) {
>> +        if (ris->ris_idx == ris_idx)
>> +            return ris;
>> +    }
>> +
>> +    return ERR_PTR(-ENOENT);
>> +}
>> +
>>   static int mpam_msc_hw_probe(struct mpam_msc *msc)
>>   {
>>       u64 idr;
>> +    u16 partid_max;
>> +    u8 ris_idx, pmg_max;
>> +    struct mpam_msc_ris *ris;
>>       struct device *dev = &msc->pdev->dev;
>>         lockdep_assert_held(&msc->probe_lock);
>> @@ -464,6 +564,40 @@ static int mpam_msc_hw_probe(struct mpam_msc *msc)
>>           return -EIO;
>>       }
>>   +    /* Grab an IDR value to find out how many RIS there are */
>> +    mutex_lock(&msc->part_sel_lock);
>> +    idr = mpam_msc_read_idr(msc);
>> +    mutex_unlock(&msc->part_sel_lock);
>> +
>> +    msc->ris_max = FIELD_GET(MPAMF_IDR_RIS_MAX, idr);
>> +
>> +    /* Use these values so partid/pmg always starts with a valid
>> value */
>> +    msc->partid_max = FIELD_GET(MPAMF_IDR_PARTID_MAX, idr);
>> +    msc->pmg_max = FIELD_GET(MPAMF_IDR_PMG_MAX, idr);
>> +
>> +    for (ris_idx = 0; ris_idx <= msc->ris_max; ris_idx++) {
>> +        mutex_lock(&msc->part_sel_lock);
>> +        __mpam_part_sel(ris_idx, 0, msc);
>> +        idr = mpam_msc_read_idr(msc);
>> +        mutex_unlock(&msc->part_sel_lock);
>> +
>> +        partid_max = FIELD_GET(MPAMF_IDR_PARTID_MAX, idr);
>> +        pmg_max = FIELD_GET(MPAMF_IDR_PMG_MAX, idr);
>> +        msc->partid_max = min(msc->partid_max, partid_max);
>> +        msc->pmg_max = min(msc->pmg_max, pmg_max);
>> +
>> +        mutex_lock(&mpam_list_lock);
>> +        ris = mpam_get_or_create_ris(msc, ris_idx);
>> +        mutex_unlock(&mpam_list_lock);
>> +        if (IS_ERR(ris))
>> +            return PTR_ERR(ris);
> 
> It's better to destroy ris's that were previously created before this
> failed ris? Otherwise, there is a memory leak for those allocated ris's?

This should be handled by mpam_disable() which is run on probe failure.

> 
> Thanks.
> 
> -Fenghua


Thanks,

Ben