From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mout-p-202.mailbox.org (mout-p-202.mailbox.org [80.241.56.172]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id B7AFC24292F for ; Mon, 24 Feb 2025 09:57:33 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=80.241.56.172 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1740391055; cv=none; b=bau74A7vvnTiCem+QbGauITGdPvrWDHj6O8Rruoed6LBmfzjMdK+Q79HCUQousm0rEIFL3lxhKFNKUntOLGTqE8pofzWsrzuPqTFZfmkt10vazcFKCyi7QCvnoNfHJP3ME6OMOPQ7/yYeupg60CbqvHSbUgrtXAFpwnT3aZCTVs= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1740391055; c=relaxed/simple; bh=zZ4LSD0pf5gem00PnlfjoiKkJZOWfQSUlOEEwXRRwAo=; h=Message-ID:Subject:From:To:Cc:Date:In-Reply-To:References: Content-Type:MIME-Version; b=J/6ZpSUKxKM1lYleXtxfN8xcW0h7vK/Ca9mm5XpGX2Ek8EiRoMLN5REjgx4t7YWhHiaTRafzULkSy3WPQfAbxXlDcI0BBHlpV70f9vNYXlRvLG+M1R81KJqLJYVGRvH5Cg70/FjeSa0RLVZ443ZVojYoKE6kcnizZYWfYOYYvLM= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dmarc=pass (p=reject dis=none) header.from=mailbox.org; spf=pass smtp.mailfrom=mailbox.org; dkim=pass (2048-bit key) header.d=mailbox.org header.i=@mailbox.org header.b=bPP8dfzG; arc=none smtp.client-ip=80.241.56.172 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=reject dis=none) header.from=mailbox.org Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=mailbox.org Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=mailbox.org header.i=@mailbox.org header.b="bPP8dfzG" Received: from smtp202.mailbox.org (smtp202.mailbox.org [IPv6:2001:67c:2050:b231:465::202]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (4096 bits) server-digest SHA256) (No client certificate requested) by mout-p-202.mailbox.org (Postfix) with ESMTPS id 4Z1bkf0DrSz9spX; Mon, 24 Feb 2025 10:57:30 +0100 (CET) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=mailbox.org; s=mail20150812; t=1740391050; h=from:from:reply-to:reply-to:subject:subject:date:date: message-id:message-id:to:to:cc:cc:mime-version:mime-version: content-type:content-type: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=ZVke/a1muIuPPw9tqzQVhkbMsCi2gU1HQNx6O8LlpO0=; b=bPP8dfzG1KdKoo4XbbnX0Egqkj/tZSMiFQtHy0HvtRoDc1AE3cT7JliMii9B6hAOhVOblf dM7uLyZnW/0Q/zmfi3RhUY5+XHeCFuCfNRpTwnnk+rCzBHW4JTB5CcAv74yZkt9ZVWI6dJ BMCH5t719k17DKhmW5PQJl7ec77gcQdbfka5T4+QJDdY2gIqODr1uWlXA5I2ZQ2zeze2GI S2GTSWjkvKDG8RQAGVqsvyyIr+evhi7laaLgrGdDr1FhaxKfJnnkEcVVZhw9PI6DgHPE46 95+Kpar4PR98oI4e9W9SlPpTO+IwujfdjeSmSzWZkadPJyIc8PTjD3UtKOLgwA== Message-ID: <21a944a054790c0a5b542183b7d5fa54483232be.camel@mailbox.org> Subject: Re: [PATCH] drm/scheduler: Fix mem leak when last_scheduled signaled From: Philipp Stanner Reply-To: phasta@kernel.org To: phasta@kernel.org, qianyi liu , Danilo Krummrich , Matthew Brost , Christian =?ISO-8859-1?Q?K=F6nig?= , Maarten Lankhorst , Maxime Ripard , Thomas Zimmermann , David Airlie , Daniel Vetter Cc: dri-devel@lists.freedesktop.org, linux-kernel@vger.kernel.org Date: Mon, 24 Feb 2025 10:57:26 +0100 In-Reply-To: <3b369e1a49b354852f361b103999673e4f7906a9.camel@mailbox.org> References: <20250221062702.1293754-1-liuqianyi125@gmail.com> <3b369e1a49b354852f361b103999673e4f7906a9.camel@mailbox.org> Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 X-MBO-RS-META: o6us1zn17yc464ke5wp6ap13n3xd8wwq X-MBO-RS-ID: 181b0d227478099764b On Mon, 2025-02-24 at 10:52 +0100, Philipp Stanner wrote: > Hello, >=20 > subject line: please write "drm/sched" instead of "drm/scheduler". It > has become the norm >=20 > On Fri, 2025-02-21 at 14:27 +0800, qianyi liu wrote: > > Problem: If prev(last_scheduled) was already signaled I encountred > > a >=20 > prev(last_scheduled) almost reads like a function call. Maybe write > "prev / last_scheduled"? >=20 > > memory leak in drm_sched_entity_fini. This is because the > > prev(last_scheduled) fence is not free properly. >=20 > s/free/freed >=20 > >=20 > > Fix: Balance the prev(last_scheduled) fence refcnt when > > dma_fence_add_callback failed. > >=20 > > Signed-off-by: qianyi liu > > --- > > =C2=A0drivers/gpu/drm/scheduler/sched_entity.c | 7 +++++-- > > =C2=A01 file changed, 5 insertions(+), 2 deletions(-) > >=20 > > diff --git a/drivers/gpu/drm/scheduler/sched_entity.c > > b/drivers/gpu/drm/scheduler/sched_entity.c > > index 69bcf0e99d57..1c0c14bcf726 100644 > > --- a/drivers/gpu/drm/scheduler/sched_entity.c > > +++ b/drivers/gpu/drm/scheduler/sched_entity.c > > @@ -259,9 +259,12 @@ static void drm_sched_entity_kill(struct > > drm_sched_entity *entity) > > =C2=A0 struct drm_sched_fence *s_fence =3D job->s_fence; > > =C2=A0 > > =C2=A0 dma_fence_get(&s_fence->finished); > > - if (!prev || dma_fence_add_callback(prev, &job- > > > finish_cb, > > - =C2=A0=C2=A0 > > drm_sched_entity_kill_jobs_cb)) > > + if (!prev || > > + =C2=A0=C2=A0=C2=A0 dma_fence_add_callback(prev, &job->finish_cb, > > + =C2=A0=C2=A0 > > drm_sched_entity_kill_jobs_cb)) { > > + dma_fence_put(prev); >=20 > But now the fence will also be put when prev =3D=3D NULL. Is that > intentional? It doesn't seem correct to me from looking at the commit > message, which states "Balance [=E2=80=A6] refcnt when dma_fence_add_call= back > failed" >=20 > It didn't get clear to me immediately which dma_fence_get() your new > dma_fence_put() balances. Can you ellaborate on that or maybe write a > comment? >=20 > But also be handy of could share the kmemleak trace. Argh. -> "It would also be handy if you could share the kmemleak trace" I should drink less=E2=80=A6 P. >=20 >=20 > Thanks > P. >=20 > > =C2=A0 drm_sched_entity_kill_jobs_cb(NULL, &job- > > > finish_cb); > > + } > > =C2=A0 > > =C2=A0 prev =3D &s_fence->finished; > > =C2=A0 } >=20