From mboxrd@z Thu Jan 1 00:00:00 1970 Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.remote-ip=100.103.45.18 ARC-Seal: i=1; d=subspace.kernel.org; s=arc20261015; a=rsa-sha256; cv=none; t=1795010658; b=o0vL2d3X0aHGDRBOVOzTq8g3tWDFOu9gU6eAP5a+r+1C3muDe+nqf/DipSt3W7Uk/qWf HT3jTR14QXFqEPuCYHgPfsjd87RY52xMuqHDVpBVcONcSSmV+2FQvp+6+CgWl3fff/31q juz50dyQ+mVcRPoWopobn8Z+Q5GGmmQAhv7702499L61BInnH6DlSorThu31UGQJUQo3q 9ihBqtgZs0mR2zZsnXCKiUSzgSruG9My9C2lvmPKABslszFl6yZjxsc+Jy1tKcFSg4c6r aPWsh4C3ItDz++9y6FGpvqdv8fvFl4VWkyLOae9XZjrhKbVN4MEB1X8wfdMw0UO4+xA== ARC-Message-Signature: i=1; d=subspace.kernel.org; s=arc20261015; a=rsa-sha256; c=relaxed/relaxed; t=1795010658; h=Received:DKIM-Signature:Message-ID:From:Subject:Reply-To:To:Cc: In-Reply-To:References:Content-Type:Content-Transfer-Encoding:Date; bh=EeQyz2mYY5toUiEDRmwERA2gZdwxOVr0NH74i92r4J4=; b=HzWVr7KEv46SiGbNZYKcZMfA8McqyD4OpiTRnO4Prl/De7X+EprSBw1C4Yk8DrP1AbQM qF3ySeSfAVLC94eH0H3UlpaCJLPBG3mZgk6ns8coGcDJqzCECfEbNWDMcSzw901pvhAX+ BF4m/NSuEf+xqGG4Jb1Tw2zk6SJzbT8sEK3RWwdpsrzfpDGYcpXdIsXXqm85/zhmv3ijy DYY1EhaxpVWvy4c+EruCF3cScQA87i965/g9xipFidrxYhuaf7IlE8eceErrfstP7fxJG aEs9BVrDyH1aPgeGcnD+YSycgBJms90HhxHJj0KhKZ/21XQkXKqCnLcl6zIhPVz0tdQ== ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dkim=pass header.d=kernel.org header.i=@kernel.org header.a=rsa-sha256 header.s=k20260515 header.b=mk5Hs3et; arc=none smtp.remote-ip=100.103.45.18 Received: from smtp.kernel.org (quasi.space.kernel.org [100.103.45.18]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature ECDSA (prime256v1)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id C3972606A33E for ; Wed, 18 Nov 2026 14:04:18 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.a=rsa-sha256 header.s=k20260515 header.b=mk5Hs3et Received: by smtp.kernel.org (Postfix) with ESMTPSA id 66B251F000C5; Wed, 18 Nov 2026 14:04:18 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=kernel.org; s=k20260515; t=1795010658; bh=EeQyz2mYY5toUiEDRmwERA2gZdwxOVr0NH74i92r4J4=; h=From:Subject:Reply-To:To:Cc:In-Reply-To:References:Date; b=mk5Hs3etPe6yuWf2Wzwhym1IAVUN6eZaTrn+320ifwUskp1rHkNWyw4fkp0I899Nu N9dYrPUSz6nh0SExvlXdg8p+iBmAn9wUztr1+X5h6sBi+tlrIywRpES+a563z9wdxG PvNfi22Pa7C8KoH+bEiXBX6VDW09DM/3RfQwI8xd0qLfkj6hoFMi6kaFWOwG5Yxw2E stz777ygwkjCyQ54omKHcPVav0gj8tiY1sDNLSpdLDOkMQxB2dVfmFjrzpQH7L/bAE kF7iDHdbzN1oR5sSlN/9d0StKbp23vapMJlRwwJDAS4EvqeBxsquqZ8iRgyBnPxJ9p ILz6+KFNu7PsQ== Message-ID: From: sashiko-bot@kernel.org Subject: Re: [PATCH rdma-next 07/11] RDMA/mlx5: Select the QP core device through its PD Reply-To: sashiko-reviews@lists.linux.dev To: "Edward Srouji" Cc: linux-rdma@vger.kernel.org In-Reply-To: <20261117-mlx5-sd-prep-v1-7-8efd124c7895@nvidia.com> References: <20261117-mlx5-sd-prep-v1-0-8efd124c7895@nvidia.com> <20261117-mlx5-sd-prep-v1-7-8efd124c7895@nvidia.com> Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable Date: Wed, 18 Nov 2026 14:04:18 +0000 X-Mailing-List: linux-rdma@vger.kernel.org List-Id: List-Post: List-Help: List-Subscribe: List-Unsubscribe: List-Owner: Precedence: list Thank you for your contribution! Sashiko AI review found no new issues in this patch, but noticed 1 related pre-existing issue it does not introduce: - [Medium] RDMA/mlx5: invalid qp_type check for DCI in configure_requester_scat_cqe() [1] -- commit fd0bed19b1450b572f53f3dc5f8db9cab84d4594 Author: Akiva Goldberger RDMA/mlx5: Select the QP core device through its PD This commit resolves the mlx5_core_dev from the QP or its PD across the QP create, modify, query, and work request posting paths instead of accessing dev->mdev. > diff --git a/drivers/infiniband/hw/mlx5/qp.c b/drivers/infiniband/hw/mlx5= /qp.c > index 2709e5465d331..1135e710e3398 100644 > --- a/drivers/infiniband/hw/mlx5/qp.c > +++ b/drivers/infiniband/hw/mlx5/qp.c [ ... ] > @@ -1957,11 +1970,11 @@ static int create_rss_raw_qp_tir(struct mlx5_ib_d= ev *dev, struct ib_pd *pd, > return err; > } > =20 > -static void configure_requester_scat_cqe(struct mlx5_ib_dev *dev, > - struct mlx5_ib_qp *qp, > +static void configure_requester_scat_cqe(struct mlx5_ib_qp *qp, > struct ib_qp_init_attr *init_attr, > void *qpc) > { > + struct mlx5_core_dev *mdev =3D qp_to_mdev(qp); > int scqe_sz; > bool allow_scat_cqe =3D false; > =20 > @@ -1977,7 +1990,7 @@ static void configure_requester_scat_cqe(struct mlx= 5_ib_dev *dev, > } > =20 > if (init_attr->qp_type !=3D MLX5_IB_QPT_DCI || > - MLX5_CAP_GEN(dev->mdev, dc_req_scat_data_cqe)) > + MLX5_CAP_GEN(mdev, dc_req_scat_data_cqe)) > MLX5_SET(qpc, qpc, cs_req, MLX5_REQ_SCAT_DATA32_CQE); > } [Severity: Medium] This problem wasn't introduced by this patch, but should this check qp->type !=3D MLX5_IB_QPT_DCI instead of init_attr->qp_type [1]? Since 2fdddbd5c966 ("RDMA/mlx5: Move DRIVER QP flags check into separate function"), process_vendor_flags() sets qp->type to MLX5_IB_QPT_DCI while leaving init_attr->qp_type as IB_QPT_DRIVER. The condition init_attr->qp_type !=3D MLX5_IB_QPT_DCI is always true, so creating a DCI QP with scatter CQE enabled and a 64-byte send CQE on hardware without dc_req_scat_data_cqe support sets cs_req to MLX5_REQ_SCAT_DATA32_CQE and causes the firmware CREATE_QP command to fail. [1] https://sashiko.dev/#/bug/linux-c5534546-da56-41f8-bc32-0fc827b3c22e --=20 Sashiko AI review =C2=B7 https://sashiko.dev/#/patchset/20261117-mlx5-sd-pr= ep-v1-0-8efd124c7895@nvidia.com?part=3D7