From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <igt-dev-bounces@lists.freedesktop.org>
Received: from mga17.intel.com (mga17.intel.com [192.55.52.151])
 by gabe.freedesktop.org (Postfix) with ESMTPS id 4F4AA7365E
 for <igt-dev@lists.freedesktop.org>; Tue,  2 Nov 2021 19:42:18 +0000 (UTC)
Date: Tue, 02 Nov 2021 12:40:14 -0700
Message-ID: <87zgqml529.wl-ashutosh.dixit@intel.com>
From: "Dixit, Ashutosh" <ashutosh.dixit@intel.com>
In-Reply-To: <20211102083941.9644-2-zbigniew.kempczynski@intel.com>
References: <20211102083941.9644-1-zbigniew.kempczynski@intel.com>
 <20211102083941.9644-2-zbigniew.kempczynski@intel.com>
MIME-Version: 1.0 (generated by SEMI-EPG 1.14.7 - "Harue")
Content-Type: text/plain; charset=ISO-8859-2
Content-Transfer-Encoding: quoted-printable
Subject: Re: [igt-dev] [PATCH i-g-t 1/1] tests/gem_sync: Use softpin path
 for no-reloc gens
List-Unsubscribe: <https://lists.freedesktop.org/mailman/options/igt-dev>,
 <mailto:igt-dev-request@lists.freedesktop.org?subject=unsubscribe>
List-Archive: <https://lists.freedesktop.org/archives/igt-dev>
List-Post: <mailto:igt-dev@lists.freedesktop.org>
List-Help: <mailto:igt-dev-request@lists.freedesktop.org?subject=help>
List-Subscribe: <https://lists.freedesktop.org/mailman/listinfo/igt-dev>,
 <mailto:igt-dev-request@lists.freedesktop.org?subject=subscribe>
Errors-To: igt-dev-bounces@lists.freedesktop.org
Sender: "igt-dev" <igt-dev-bounces@lists.freedesktop.org>
To: Zbigniew =?ISO-8859-2?Q?Kempczy=F1ski?= <zbigniew.kempczynski@intel.com>
Cc: igt-dev@lists.freedesktop.org
List-ID: <igt-dev@lists.freedesktop.org>

On Tue, 02 Nov 2021 01:39:41 -0700, Zbigniew Kempczy=F1ski wrote:
>
> Migrate to softpin with pre-warming batch or an allocator. Some tests
> which don't fork can use pre-warming batch which establishes offsets in
> the kernel before real work is performed. For such adding pinned flag
> + zeroing relocation is enough. For multiprocess scenarios we need to
> arbitrate and an allocator need to be used instead.
>
> v2: fix also tests which are not running on CI (Ashutosh)
>
> v3: use appropriate allocator handle (ahnd) within active_wakeup_ring
>     spinners (Ashutosh)

Reviewed-by: Ashutosh Dixit <ashutosh.dixit@intel.com>

One question below though.

> Signed-off-by: Zbigniew Kempczy=F1ski <zbigniew.kempczynski@intel.com>
> Cc: Ashutosh Dixit <ashutosh.dixit@intel.com>
> ---
>  tests/i915/gem_sync.c | 59 ++++++++++++++++++++++++++++++++++++++++---
>  1 file changed, 55 insertions(+), 4 deletions(-)
>
> diff --git a/tests/i915/gem_sync.c b/tests/i915/gem_sync.c
> index 0093ca25f..8c435845e 100644
> --- a/tests/i915/gem_sync.c
> +++ b/tests/i915/gem_sync.c
> @@ -237,6 +237,7 @@ wakeup_ring(int fd, const intel_ctx_t *ctx, unsigned =
ring,
>	    int timeout, int wlen)
>  {
>	struct intel_engine_data ied;
> +	uint64_t ahnd =3D get_reloc_ahnd(fd, ctx->id);

Is this really needed in the parent even when they are only used in the
child processes? This pattern is used everywhere in this patch. Is this a
requirement of the allocator API? Thanks.