Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
References: <20230108143843.2987732-1-trix@redhat.com>
 <CANn89iLFtrQm-E5BRwgKFw4xRZiOOdWg-WTFi5eZsg7ycq2szg@mail.gmail.com>
 <pj41zlpmbmba16.fsf@u570694869fb251.ant.amazon.com>
 <db824c89-13f2-3349-9dd0-0fb7559c6273@redhat.com>
User-agent: mu4e 1.6.10; emacs 28.0.91
From:   Shay Agroskin <shayagr@amazon.com>
To:     Tom Rix <trix@redhat.com>
CC:     Eric Dumazet <edumazet@google.com>, <akiyano@amazon.com>,
        <darinzon@amazon.com>, <ndagan@amazon.com>, <saeedb@amazon.com>,
        <davem@davemloft.net>, <kuba@kernel.org>, <pabeni@redhat.com>,
        <nathan@kernel.org>, <ndesaulniers@google.com>, <khalasa@piap.pl>,
        <wsa+renesas@sang-engineering.com>, <yuancan@huawei.com>,
        <tglx@linutronix.de>, <42.hyeyoo@gmail.com>,
        <netdev@vger.kernel.org>, <linux-kernel@vger.kernel.org>,
        <llvm@lists.linux.dev>
Subject: Re: [PATCH] net: ena: initialize dim_sample
Date:   Wed, 11 Jan 2023 10:46:50 +0200
In-Reply-To: <db824c89-13f2-3349-9dd0-0fb7559c6273@redhat.com>
Message-ID: <pj41zllem9bglr.fsf@u570694869fb251.ant.amazon.com>
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"; format=flowed
Content-Transfer-Encoding: quoted-printable
Precedence: bulk


Tom Rix <trix@redhat.com> writes:

> On 1/10/23 8:58 AM, Shay Agroskin wrote:
>>
>> Eric Dumazet <edumazet@google.com> writes:
>>
>>> On Sun, Jan 8, 2023 at 3:38 PM Tom Rix <trix@redhat.com>=20
>>> wrote:
>>>>
>>>> clang static analysis reports this problem
>>>> drivers/net/ethernet/amazon/ena/ena_netdev.c:1821:2: warning:
>>>> Passed-by-value struct
>>>> =C2=A0 argument contains uninitialized data (e.g., field:=20
>>>> 'comp_ctr')
>>>> [core.CallAndMessage]
>>>> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 net_dim(&ena_napi->dim, dim=
_sample);
>>>> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 ^~~~~~~~~~~~~~~~~~~~~~~~~~~=
~~~~~~~~
>>>>
>>>> net_dim can call dim_calc_stats() which uses the comp_ctr=20
>>>> element,
>>>> so it must be initialized.
>>>
>>> This seems to be a dim_update_sample() problem really, when=20
>>> comp_ctr
>>> has been added...
>>>
>>> Your patch works, but we could avoid pre-initializing=20
>>> dim_sample in
>>> all callers,
>>> then re-writing all but one field...
>>>
>>> diff --git a/include/linux/dim.h b/include/linux/dim.h
>>> index
>>> 6c5733981563eadf5f06c59c5dc97df961692b02..4604ced4517268ef8912cd8053ac8=
f4d2630f977
>>> 100644
>>> --- a/include/linux/dim.h
>>> +++ b/include/linux/dim.h
>>> @@ -254,6 +254,7 @@ dim_update_sample(u16 event_ctr, u64=20
>>> packets, u64
>>> bytes, struct dim_sample *s)
>>> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 s->pkt_ctr=C2=A0=C2=A0 =3D p=
ackets;
>>> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 s->byte_ctr=C2=A0 =3D bytes;
>>> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 s->event_ctr =3D event_ctr;
>>> +=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 s->comp_ctr=C2=A0 =3D 0;
>>> =C2=A0}
>>>
>>> =C2=A0/**
>>
>> Hi,
>>
>> I'd rather go with Eric's solution to this issue than zero the=20
>> whole
>> struct in ENA
>
> Please look at the other callers of dim_update_sample.=C2=A0 The=20
> common
> pattern is to initialize the struct.
>
> This alternative will work, but the pattern of initializing the=20
> struct
> the other (~20) callers should be refactored.
>
> Tom
>

While Eric's patch might be bigger if you also remove the=20
pre-initialization in the other drivers, the Linux code itself=20
would be smaller (granted not significantly) and
it make less room for pitfalls in adding DIM support in other=20
drivers.

Is there a good argument against using Eric's patch other than=20
'the other patch would be bigger' ?

Shay

>>
>> Thanks,
>> Shay
>>