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;
From:   "Fabio M. De Francesco" <fmdefrancesco@gmail.com>
To:     Jani Nikula <jani.nikula@linux.intel.com>,
        Joonas Lahtinen <joonas.lahtinen@linux.intel.com>,
        Rodrigo Vivi <rodrigo.vivi@intel.com>,
        Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>,
        David Airlie <airlied@gmail.com>,
        Daniel Vetter <daniel@ffwll.ch>,
        intel-gfx@lists.freedesktop.org, dri-devel@lists.freedesktop.org,
        linux-kernel@vger.kernel.org,
        Sumitra Sharma <sumitraartsy@gmail.com>,
        Thomas =?ISO-8859-1?Q?Hellstr=F6m_=28Intel=29?= 
        <thomas_os@shipmail.org>
Cc:     Ira Weiny <ira.weiny@intel.com>, Deepak R Varma <drv@mailo.com>,
        Sumitra Sharma <sumitraartsy@gmail.com>
Subject: Re: [PATCH v2] drm/i915: Replace kmap() with kmap_local_page()
Date:   Sat, 24 Jun 2023 02:10:56 +0200
Message-ID: <3747290.kQq0lBPeGt@suse>
In-Reply-To: <20230617180420.GA410966@sumitra.com>
References: <20230617180420.GA410966@sumitra.com>
MIME-Version: 1.0
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain; charset="utf-8"
Precedence: bulk

On sabato 17 giugno 2023 20:04:20 CEST Sumitra Sharma wrote:
> kmap() has been deprecated in favor of the kmap_local_page()
> due to high cost, restricted mapping space, the overhead of a
> global lock for synchronization, and making the process sleep
> in the absence of free slots.
>=20
> kmap_local_page() is faster than kmap() and offers thread-local
> and CPU-local mappings, take pagefaults in a local kmap region

NIT: _can_ take pagefaults in a local kmap region

> and preserves preemption by saving the mappings of outgoing tasks
> and restoring those of the incoming one during a context switch.
>=20
> The mapping is kept thread local in the function
> =E2=80=9Ci915_vma_coredump_create=E2=80=9D in i915_gpu_error.c
>=20
> Therefore, replace kmap() with kmap_local_page().
>=20
> Suggested-by: Ira Weiny <ira.weiny@intel.com>
>=20
> Signed-off-by: Sumitra Sharma <sumitraartsy@gmail.com>
> ---
>=20
> Changes in v2:
> 	- Replace kmap() with kmap_local_page().
> 	- Change commit subject and message.

With the changes that Ira suggested and the minor fix I'm proposing to the=
=20
commit message, it looks good to me too, so this patch is...=20

Reviewed-by: Fabio M. De Francesco <fmdefrancesco@gmail.com>

However, as far as I'm concerned, our nits don't necessarily require any ne=
wer=20
version, especially because Tvrtko has already sent this patch for their CI.

Thanks,

=46abio

P.S.: As Sumitra says both kmap() and kmap_local_page() allows preemption i=
n=20
non atomic context.=20

=46urthermore, Tvrtko confirmed that the pages can come from HIGHMEM, there=
fore=20
kmap_local_page for local temporary mapping is unavoidable.

Last thing... Thomas thinks he wants to make it run atomically (if I=20
understood one of his messages correctly). As I already responded, nothing=
=20
prevents someone does another patch just to disable preemption (or to enter=
=20
atomic context by other means) around the code marked by kmap_local_page() =
/=20
kunmap_local() because these functions work perfectly _also_ in atomic cont=
ext=20
(including interrupts). But this is not something that Sumitra should be=20
worried about.

>=20
>  drivers/gpu/drm/i915/i915_gpu_error.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
>=20
> diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c
> b/drivers/gpu/drm/i915/i915_gpu_error.c index f020c0086fbc..bc41500eedf5
> 100644
> --- a/drivers/gpu/drm/i915/i915_gpu_error.c
> +++ b/drivers/gpu/drm/i915/i915_gpu_error.c
> @@ -1164,9 +1164,9 @@ i915_vma_coredump_create(const struct intel_gt *gt,
>=20
>  			drm_clflush_pages(&page, 1);
>=20
> -			s =3D kmap(page);
> +			s =3D kmap_local_page(page);
>  			ret =3D compress_page(compress, s, dst, false);
> -			kunmap(page);
> +			kunmap_local(s);
>=20
>  			drm_clflush_pages(&page, 1);
>=20
> --
> 2.25.1