From mboxrd@z Thu Jan 1 00:00:00 1970
Return-Path: <michael.qiu@intel.com>
Received: from mga01.intel.com (mga01.intel.com [192.55.52.88])
by dpdk.org (Postfix) with ESMTP id 3FCC72A58
for <dev@dpdk.org>; Tue, 14 Jul 2015 07:07:18 +0200 (CEST)
Received: from fmsmga001.fm.intel.com ([10.253.24.23])
by fmsmga101.fm.intel.com with ESMTP; 13 Jul 2015 22:07:17 -0700
X-ExtLoop1: 1
X-IronPort-AV: E=Sophos;i="5.15,469,1432623600"; d="scan'208";a="746615938"
Received: from kmsmsx154.gar.corp.intel.com ([172.21.73.14])
by fmsmga001.fm.intel.com with ESMTP; 13 Jul 2015 22:07:16 -0700
Received: from shsmsx102.ccr.corp.intel.com (10.239.4.154) by
KMSMSX154.gar.corp.intel.com (172.21.73.14) with Microsoft SMTP Server (TLS)
id 14.3.224.2; Tue, 14 Jul 2015 13:03:07 +0800
Received: from shsmsx101.ccr.corp.intel.com ([169.254.1.246]) by
shsmsx102.ccr.corp.intel.com ([169.254.2.165]) with mapi id 14.03.0224.002;
Tue, 14 Jul 2015 13:03:05 +0800
From: "Qiu, Michael" <michael.qiu@intel.com>
To: Tetsuya Mukawa <mukawa@igel.co.jp>, "dev@dpdk.org" <dev@dpdk.org>
Thread-Topic: [PATCH] eal: fix vfio device never works.
Thread-Index: AQHQu6MJLsBHm5qwrE2iTGtyZxkNXQ==
Date: Tue, 14 Jul 2015 05:03:04 +0000
Message-ID: <533710CFB86FA344BFBF2D6802E60286046BA583@SHSMSX101.ccr.corp.intel.com>
References: <1436514439-4893-1-git-send-email-michael.qiu@intel.com>
<1436596189-1831-1-git-send-email-mukawa@igel.co.jp>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach:
X-MS-TNEF-Correlator:
x-originating-ip: [10.239.127.40]
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
Subject: Re: [dpdk-dev] [PATCH] eal: fix vfio device never works.
X-BeenThere: dev@dpdk.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: patches and discussions about DPDK <dev.dpdk.org>
List-Unsubscribe: <http://dpdk.org/ml/options/dev>,
<mailto:dev-request@dpdk.org?subject=unsubscribe>
List-Archive: <http://dpdk.org/ml/archives/dev/>
List-Post: <mailto:dev@dpdk.org>
List-Help: <mailto:dev-request@dpdk.org?subject=help>
List-Subscribe: <http://dpdk.org/ml/listinfo/dev>,
<mailto:dev-request@dpdk.org?subject=subscribe>
X-List-Received-Date: Tue, 14 Jul 2015 05:07:19 -0000
Hi, Tetsuya=0A=
=0A=
So this patch make linux app back to original?=0A=
=0A=
Thanks,=0A=
Michael=0A=
=0A=
On 7/11/2015 2:30 PM, Tetsuya Mukawa wrote:=0A=
> The patch fixes vfio initialization issue introduced by below patch.=0A=
> - Commit 35b3313e322b ("pci: merge mapping functions for linux and bsd")=
=0A=
>=0A=
> Root cause is that VFIO_PRESENT is inaccessible in eal common level.=0A=
> To fix it, remove pci_map/unmap_device from common code, then implement=
=0A=
> in linux and bsd code.=0A=
>=0A=
> Reported-by: Michael Qiu <michael.qiu@intel.com>=0A=
> Signed-off-by: Tetsuya Mukawa <mukawa@igel.co.jp>=0A=
> ---=0A=
> lib/librte_eal/bsdapp/eal/eal_pci.c | 39 ++++++++++++++++++++++++=0A=
> lib/librte_eal/common/eal_common_pci.c | 55 ----------------------------=
------=0A=
> lib/librte_eal/common/eal_private.h | 18 +++++++++++=0A=
> lib/librte_eal/linuxapp/eal/eal_pci.c | 50 ++++++++++++++++++++++++++++=
+++=0A=
> 4 files changed, 107 insertions(+), 55 deletions(-)=0A=
>=0A=
> diff --git a/lib/librte_eal/bsdapp/eal/eal_pci.c b/lib/librte_eal/bsdapp/=
eal/eal_pci.c=0A=
> index 12f39d9..ed31222 100644=0A=
> --- a/lib/librte_eal/bsdapp/eal/eal_pci.c=0A=
> +++ b/lib/librte_eal/bsdapp/eal/eal_pci.c=0A=
> @@ -92,6 +92,45 @@ pci_unbind_kernel_driver(struct rte_pci_device *dev __=
rte_unused)=0A=
> return -ENOTSUP;=0A=
> }=0A=
> =0A=
> +/* Map pci device */=0A=
> +int=0A=
> +pci_map_device(struct rte_pci_device *dev)=0A=
> +{=0A=
> + int ret =3D -1;=0A=
> +=0A=
> + /* try mapping the NIC resources */=0A=
> + switch (dev->kdrv) {=0A=
> + case RTE_KDRV_NIC_UIO:=0A=
> + /* map resources for devices that use uio */=0A=
> + ret =3D pci_uio_map_resource(dev);=0A=
> + break;=0A=
> + default:=0A=
> + RTE_LOG(DEBUG, EAL,=0A=
> + " Not managed by a supported kernel driver, skipped\n");=0A=
> + ret =3D 1;=0A=
> + break;=0A=
> + }=0A=
> +=0A=
> + return ret;=0A=
> +}=0A=
> +=0A=
> +/* Unmap pci device */=0A=
> +void=0A=
> +pci_unmap_device(struct rte_pci_device *dev)=0A=
> +{=0A=
> + /* try unmapping the NIC resources */=0A=
> + switch (dev->kdrv) {=0A=
> + case RTE_KDRV_NIC_UIO:=0A=
> + /* unmap resources for devices that use uio */=0A=
> + pci_uio_unmap_resource(dev);=0A=
> + break;=0A=
> + default:=0A=
> + RTE_LOG(DEBUG, EAL,=0A=
> + " Not managed by a supported kernel driver, skipped\n");=0A=
> + break;=0A=
> + }=0A=
> +}=0A=
> +=0A=
> void=0A=
> pci_uio_free_resource(struct rte_pci_device *dev,=0A=
> struct mapped_pci_resource *uio_res)=0A=
> diff --git a/lib/librte_eal/common/eal_common_pci.c b/lib/librte_eal/comm=
on/eal_common_pci.c=0A=
> index 3805aed..60e40e0 100644=0A=
> --- a/lib/librte_eal/common/eal_common_pci.c=0A=
> +++ b/lib/librte_eal/common/eal_common_pci.c=0A=
> @@ -137,61 +137,6 @@ pci_unmap_resource(void *requested_addr, size_t size=
)=0A=
> requested_addr);=0A=
> }=0A=
> =0A=
> -/* Map pci device */=0A=
> -static int=0A=
> -pci_map_device(struct rte_pci_device *dev)=0A=
> -{=0A=
> - int ret =3D -1;=0A=
> -=0A=
> - /* try mapping the NIC resources using VFIO if it exists */=0A=
> - switch (dev->kdrv) {=0A=
> - case RTE_KDRV_VFIO:=0A=
> -#ifdef VFIO_PRESENT=0A=
> - if (pci_vfio_is_enabled())=0A=
> - ret =3D pci_vfio_map_resource(dev);=0A=
> -#endif=0A=
> - break;=0A=
> - case RTE_KDRV_IGB_UIO:=0A=
> - case RTE_KDRV_UIO_GENERIC:=0A=
> - case RTE_KDRV_NIC_UIO:=0A=
> - /* map resources for devices that use uio */=0A=
> - ret =3D pci_uio_map_resource(dev);=0A=
> - break;=0A=
> - default:=0A=
> - RTE_LOG(DEBUG, EAL, " Not managed by a supported kernel driver,"=0A=
> - " skipped\n");=0A=
> - ret =3D 1;=0A=
> - break;=0A=
> - }=0A=
> -=0A=
> - return ret;=0A=
> -}=0A=
> -=0A=
> -/* Unmap pci device */=0A=
> -static void=0A=
> -pci_unmap_device(struct rte_pci_device *dev)=0A=
> -{=0A=
> - if (dev =3D=3D NULL)=0A=
> - return;=0A=
> -=0A=
> - /* try unmapping the NIC resources using VFIO if it exists */=0A=
> - switch (dev->kdrv) {=0A=
> - case RTE_KDRV_VFIO:=0A=
> - RTE_LOG(ERR, EAL, "Hotplug doesn't support vfio yet\n");=0A=
> - break;=0A=
> - case RTE_KDRV_IGB_UIO:=0A=
> - case RTE_KDRV_UIO_GENERIC:=0A=
> - case RTE_KDRV_NIC_UIO:=0A=
> - /* unmap resources for devices that use uio */=0A=
> - pci_uio_unmap_resource(dev);=0A=
> - break;=0A=
> - default:=0A=
> - RTE_LOG(DEBUG, EAL, " Not managed by a supported kernel driver,"=0A=
> - " skipped\n");=0A=
> - break;=0A=
> - }=0A=
> -}=0A=
> -=0A=
> /*=0A=
> * If vendor/device ID match, call the devinit() function of the=0A=
> * driver.=0A=
> diff --git a/lib/librte_eal/common/eal_private.h b/lib/librte_eal/common/=
eal_private.h=0A=
> index e16bb68..73363f6 100644=0A=
> --- a/lib/librte_eal/common/eal_private.h=0A=
> +++ b/lib/librte_eal/common/eal_private.h=0A=
> @@ -165,6 +165,24 @@ struct rte_pci_device;=0A=
> int pci_unbind_kernel_driver(struct rte_pci_device *dev);=0A=
> =0A=
> /**=0A=
> + * Map this device=0A=
> + *=0A=
> + * This function is private to EAL.=0A=
> + *=0A=
> + * @return=0A=
> + * 0 on success, negative on error and positive if no driver=0A=
> + * is found for the device.=0A=
> + */=0A=
> +int pci_map_device(struct rte_pci_device *dev);=0A=
> +=0A=
> +/**=0A=
> + * Unmap this device=0A=
> + *=0A=
> + * This function is private to EAL.=0A=
> + */=0A=
> +void pci_unmap_device(struct rte_pci_device *dev);=0A=
> +=0A=
> +/**=0A=
> * Map the PCI resource of a PCI device in virtual memory=0A=
> *=0A=
> * This function is private to EAL.=0A=
> diff --git a/lib/librte_eal/linuxapp/eal/eal_pci.c b/lib/librte_eal/linux=
app/eal/eal_pci.c=0A=
> index 1d5a13b..9a28ede 100644=0A=
> --- a/lib/librte_eal/linuxapp/eal/eal_pci.c=0A=
> +++ b/lib/librte_eal/linuxapp/eal/eal_pci.c=0A=
> @@ -123,6 +123,56 @@ pci_get_kernel_driver_by_path(const char *filename, =
char *dri_name)=0A=
> return -1;=0A=
> }=0A=
> =0A=
> +/* Map pci device */=0A=
> +int=0A=
> +pci_map_device(struct rte_pci_device *dev)=0A=
> +{=0A=
> + int ret =3D -1;=0A=
> +=0A=
> + /* try mapping the NIC resources using VFIO if it exists */=0A=
> + switch (dev->kdrv) {=0A=
> + case RTE_KDRV_VFIO:=0A=
> +#ifdef VFIO_PRESENT=0A=
> + if (pci_vfio_is_enabled())=0A=
> + ret =3D pci_vfio_map_resource(dev);=0A=
> +#endif=0A=
> + break;=0A=
> + case RTE_KDRV_IGB_UIO:=0A=
> + case RTE_KDRV_UIO_GENERIC:=0A=
> + /* map resources for devices that use uio */=0A=
> + ret =3D pci_uio_map_resource(dev);=0A=
> + break;=0A=
> + default:=0A=
> + RTE_LOG(DEBUG, EAL,=0A=
> + " Not managed by a supported kernel driver, skipped\n");=0A=
> + ret =3D 1;=0A=
> + break;=0A=
> + }=0A=
> +=0A=
> + return ret;=0A=
> +}=0A=
> +=0A=
> +/* Unmap pci device */=0A=
> +void=0A=
> +pci_unmap_device(struct rte_pci_device *dev)=0A=
> +{=0A=
> + /* try unmapping the NIC resources using VFIO if it exists */=0A=
> + switch (dev->kdrv) {=0A=
> + case RTE_KDRV_VFIO:=0A=
> + RTE_LOG(ERR, EAL, "Hotplug doesn't support vfio yet\n");=0A=
> + break;=0A=
> + case RTE_KDRV_IGB_UIO:=0A=
> + case RTE_KDRV_UIO_GENERIC:=0A=
> + /* unmap resources for devices that use uio */=0A=
> + pci_uio_unmap_resource(dev);=0A=
> + break;=0A=
> + default:=0A=
> + RTE_LOG(DEBUG, EAL,=0A=
> + " Not managed by a supported kernel driver, skipped\n");=0A=
> + break;=0A=
> + }=0A=
> +}=0A=
> +=0A=
> void *=0A=
> pci_find_max_end_va(void)=0A=
> {=0A=
=0A=