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;
MIME-Version: 1.0
References: <20230703162509.77828-1-koba.ko@canonical.com> <4ec2b7d2-11a5-6ab6-087a-175ed31faca4@web.de>
 <SJ1PR11MB60839A0FC6B5E79E3E5A7997FC29A@SJ1PR11MB6083.namprd11.prod.outlook.com>
 <CAJB-X+X+09-m57JcZcb-_9dKUG3CtAbLXxGTEg7R0bB8pyJx9Q@mail.gmail.com>
 <CAJB-X+XVO29wVxVezjFrgCyXigqEJxAzb0K0wueXNto5K_x2tA@mail.gmail.com> <9c27530e-21f9-15ce-5116-5af5b0c25f53@web.de>
In-Reply-To: <9c27530e-21f9-15ce-5116-5af5b0c25f53@web.de>
From:   Koba Ko <koba.ko@canonical.com>
Date:   Tue, 4 Jul 2023 16:05:41 +0800
Message-ID: <CAJB-X+Wu-Zd6pCrK54aR9iaeS7PW2VmwB+Y+Qeci7Ut0YcdsRg@mail.gmail.com>
Subject: Re: [PATCH v2] EDAC/i10nm: shift exponent is negative
To:     Markus Elfring <Markus.Elfring@web.de>
Cc:     linux-edac@vger.kernel.org, kernel-janitors@vger.kernel.org,
        LKML <linux-kernel@vger.kernel.org>,
        Borislav Petkov <bp@alien8.de>,
        James Morse <james.morse@arm.com>,
        Mauro Carvalho Chehab <mchehab@kernel.org>,
        Robert Richter <rric@kernel.org>,
        Tony Luck <tony.luck@intel.com>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

On Tue, Jul 4, 2023 at 3:16=E2=80=AFPM Markus Elfring <Markus.Elfring@web.d=
e> wrote:
>
> > As per suggestions, i modified V3.
> > could you please take a look
> >
> > Subject: [PATCH][V3] EDAC/i10nm: shift exponent is negative
>
> Would you like to put the text =E2=80=9C[PATCH v4] EDAC/i10nm: Fix an ina=
ppropriate shift exponent=E2=80=9D
> into a subsequent patch?

I didn't send V3 so the suggestions could be put in V3.

>
> I would find further wording variants nicer.
>
>
> > Because failed to read from DIMM, get the negative value for shift
> > operation.
>
> A surprising value was determined after a read failure from a DIMM.
>
>
> =E2=80=A6
> > UBSAN complains this error
>
> Software analyses pointed a data processing issue out.
>
>
> > `UBSAN: shift-out-of-bounds in drivers/edac/skx_common.c:369:16
> >  shift exponent -66 is negative`
> >
> > when get rows, cols and ranks, the returned error value doesn't be
> > handled.
>
> A special value combination could not be handled so far.
>
>
> > check the return value is EINVAL, if yes, directly return 0 and
> > show error message explicitly.
>
> Check if an invalid value was detected by a call of the function =E2=80=
=9Cskx_get_dimm_attr=E2=80=9D.
>
> * Print a corresponding error message in this case.
>
> * Return zero then directly from the function =E2=80=9Cskx_get_dimm_info=
=E2=80=9D.
>
>
> =E2=80=A6
> @@ -351,6 +351,8 @@ int skx_get_dimm_info(u32 mtr, u32 mcmtr, u32 amap, s=
truct dimm_info *dimm,
>         ranks =3D numrank(mtr);
>         rows =3D numrow(mtr);
>         cols =3D imc->hbm_mc ? 6 : numcol(mtr);
> +       if (ranks =3D=3D -EINVAL || rows =3D=3D -EINVAL || cols =3D=3D -E=
INVAL)
> +               return 0;
> =E2=80=A6
>
>
> Can it be nicer to perform a check for such an error code directly
> after each variable assignment?
> (May this condition check be split?)
>
>
> > Fixes: 4ec656bdf43a13) EDAC, skx_edac: Add EDAC driver for Skylake
>
> Please properly apply parentheses and double quotes for this tag.
>
> See also:
> https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/D=
ocumentation/process/submitting-patches.rst?h=3Dv6.4#n145
>
>
> > V2 -> V3: simplify the summary and add 'Fixes:'
> > V1 -> V2: make error-print explicitly
>
> How do you think about to use more succinct version identifiers
> for such descriptions?
>
> V4:
> =E2=80=A6
>
> V3:
> =E2=80=A6
>
>
> Regards,
> Markus