Content-Type: text/plain;
        charset=us-ascii
Mime-Version: 1.0 (Mac OS X Mail 14.0 \(3654.120.0.1.13\))
Subject: Re: [PATCH v2 0/3] Support register names of all architectures
From:   Athira Rajeev <atrajeev@linux.vnet.ibm.com>
In-Reply-To: <20211207180653.1147374-1-german.gomez@arm.com>
Date:   Tue, 14 Dec 2021 14:20:26 +0530
Cc:     Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        linux-perf-users@vger.kernel.org,
        Arnaldo Carvalho de Melo <acme@kernel.org>,
        John Garry <john.garry@huawei.com>,
        Will Deacon <will@kernel.org>,
        Mathieu Poirier <mathieu.poirier@linaro.org>,
        Leo Yan <leo.yan@linaro.org>,
        Mark Rutland <mark.rutland@arm.com>,
        Alexander Shishkin <alexander.shishkin@linux.intel.com>,
        Jiri Olsa <jolsa@redhat.com>,
        Namhyung Kim <namhyung@kernel.org>,
        linux-arm-kernel@lists.infradead.org, linux-csky@vger.kernel.org,
        linux-riscv@lists.infradead.org
Content-Transfer-Encoding: quoted-printable
Message-Id: <EAC4AC1D-A305-42E2-B22F-8B64B0968841@linux.vnet.ibm.com>
References: <20211207180653.1147374-1-german.gomez@arm.com>
To:     German Gomez <german.gomez@arm.com>
Precedence: bulk


> On 07-Dec-2021, at 11:36 PM, German Gomez <german.gomez@arm.com> =
wrote:
>=20
> The following changeset applies some corrections to the way system
> registers are processed and presented when reading perf.data files =
using
> the various perf tools.
>=20
> The commit message from [3/3] shows how register names aren't =
correctly
> presented when performing x-arch analysis of perf.data files =
(recording
> in one arch, then reading the file from a different arch).
>=20
>  - [PATCH 1/3] Fixes a potential out-of-bounds access when reading the
>    values of the registers in the perf.data file.
>  - [PATCH 2/3] Fixes an issue of ARM and ARM64 registers having the
>    same enum name.
>  - [PATCH 3/3] Refactors the function "perf_reg_name" declared in the
>   "tools/perf/util/perf_regs.h" header, in order to support every =
arch.
>=20
> Thanks,
> German

Looks good to me. Tested this patchset in powerpc by capturing regs in =
powerpc and doing
perf report to read the data from x86.

Reviewed-and-Tested-by: Athira Rajeev <atrajeev@linux.vnet.ibm.com>
>=20
> --
> Changes since v1
>=20
>  - Added "Reported-by" tags.
>  - Removed [PATCH 2/4] because it's not needed (suggested by Athira
>    Rajeev).
>  - Removed [PATCH 3/4] which created additional header files with the
>    register names of every arch.
>  - Introduced [PATCH 2/3] to deal with ARM and ARM64 registers having =
the
>    same enum name across "/tools/perf/".
>  - Reworked the refactor of "perf_reg_name" function (now implemented =
in
>    perf_regs.c, rather than in the header file) in [PATCH 3/3].
>=20
> German Gomez (3):
>  perf tools: Prevent out-of-bounds access to registers
>  perf tools: Rename perf_event_arm_regs for ARM64 registers
>  perf tools: Support register names from all archs
>=20
> tools/perf/arch/arm/include/perf_regs.h       |  42 --
> tools/perf/arch/arm64/include/perf_regs.h     |  78 +-
> tools/perf/arch/csky/include/perf_regs.h      |  82 ---
> tools/perf/arch/mips/include/perf_regs.h      |  69 --
> tools/perf/arch/powerpc/include/perf_regs.h   |  66 --
> tools/perf/arch/riscv/include/perf_regs.h     |  74 --
> tools/perf/arch/s390/include/perf_regs.h      |  78 --
> tools/perf/arch/x86/include/perf_regs.h       |  82 ---
> tools/perf/builtin-script.c                   |  18 +-
> tools/perf/util/event.h                       |   5 +-
> tools/perf/util/libunwind/arm64.c             |   2 +
> tools/perf/util/perf_regs.c                   | 671 +++++++++++++++++-
> tools/perf/util/perf_regs.h                   |  10 +-
> .../scripting-engines/trace-event-python.c    |  10 +-
> tools/perf/util/session.c                     |  25 +-
> 15 files changed, 709 insertions(+), 603 deletions(-)
>=20
> --=20
> 2.25.1
>=20