[dpdk-dev] [PATCH] vhost example doc update
add vhost user documentation fix some minor issues Signed-off-by: Huawei Xie --- doc/guides/sample_app_ug/vhost.rst | 123 - 1 file changed, 94 insertions(+), 29 deletions(-) diff --git a/doc/guides/sample_app_ug/vhost.rst b/doc/guides/sample_app_ug/vhost.rst index fa53db6..4f28abd 100644 --- a/doc/guides/sample_app_ug/vhost.rst +++ b/doc/guides/sample_app_ug/vhost.rst @@ -45,7 +45,7 @@ Background Virtio networking (virtio-net) was developed as the Linux* KVM para-virtualized method for communicating network packets between host and guest. It was found that virtio-net performance was poor due to context switching and packet copying between host, guest, and QEMU. -The following figure shows the system architecture for a virtio- based networking (virtio-net). +The following figure shows the system architecture for a virtio-based networking (virtio-net). .. _figure_16: @@ -89,20 +89,34 @@ Sample Code Overview The DPDK vhost-net sample code demonstrates KVM (QEMU) offloading the servicing of a Virtual Machine's (VM's) virtio-net devices to a DPDK-based application in place of the kernel's vhost-net module. -The DPDK vhost-net sample code is a simple packet switching application with the following features: +The DPDK vhost-net sample code is based on vhost library. Vhost library is developed for user space ethernet switch to +easily integrate with vhost functionality. + +The vhost library implements the following features: * Management of virtio-net device creation/destruction events. -* Mapping of the VM's physical memory into the DPDK vhost-net sample code's address space. +* Mapping of the VM's physical memory into the DPDK vhost-net's address space. * Triggering/receiving notifications to/from VMs via eventfds. * A virtio-net back-end implementation providing a subset of virtio-net features. +There are two vhost implementations in vhost library, vhost cuse and vhost user. In vhost cuse, a character device driver is implemented to +receive and process vhost requests through ioctl messages. In vhost user, a socket server is created to received vhost requests through +socket messages. Most of the messages share the same handler routine. + +.. note:: +**Any vhost cuse specific requirement in the following sections will be emphasized**. + +Two impelmentations are turned on and off statically through configure file. Only one implementation could be turned on. They don't co-exist in current implementation. + +The vhost sample code application is a simple packet switching application with the following feature: + * Packet switching between virtio-net devices and the network interface card, including using VMDQs to reduce the switching that needs to be performed in software. -The following figure shows the architecture of the Vhost sample application. +The following figure shows the architecture of the Vhost sample application based on vhost-cuse. .. _figure_18: @@ -131,15 +145,19 @@ The example in this section have been validated with the following distributions * Fedora* 19 +* Fedora* 20 + Prerequisites - This section lists prerequisite packages that must be installed. -Installing Packages on the Host -~~~ +Installing Packages on the Host(vhost cuse required) + -The vhost sample code uses the following packages; fuse, fuse-devel, and kernel- modules-extra. +The vhost cuse code uses the following packages; fuse, fuse-devel, and kernel-modules-extra. +The vhost user code don't rely on those modules as eventfds are already installed into vhost process through +unix domain socket. #. Install Fuse Development Libraries and headers: @@ -153,6 +171,11 @@ The vhost sample code uses the following packages; fuse, fuse-devel, and kernel- yum -y install kernel-modules-extra +QEMU simulator +~~ + +For vhost user, qemu 2.2 is required. + Setting up the Execution Environment @@ -202,7 +225,7 @@ In this section, we create a second hugetlbs mount point to allocate hugepages f .. code-block:: console -echo 256 > /sys/kernel/mm/hugepages/hugepages-2048kB/ nr_hugepages +echo 256 > /sys/kernel/mm/hugepages/hugepages-2048kB/nr_hugepages #. Mount hugetlbs at a separate mount point for 2 MB pages: @@ -251,6 +274,8 @@ at the command line as follows. Observe that in the example, "-device" and "-netdev" are repeated for two virtio-net devices. +For vhost cuse: + .. code-block:: console user at target:~$ qemu-system-x86_64 ... \ @@ -259,15 +284,46 @@ at the command line as follows. -netdev tap,id=hostnet2,vhost=on,vhostfd= \ -device virtio-net-pci, netdev=hostnet2,id=net1 +For vhost user: + +.. code-block:: console + +user at target:~$ qemu-system-x86_64 ... \ +-chardev socket,id=char1,path= \ +-netdev
[dpdk-dev] [PATCH] vhost library doc update
add vhost user documentation Signed-off-by: Huawei Xie --- doc/guides/prog_guide/vhost_lib.rst | 52 ++--- 1 file changed, 42 insertions(+), 10 deletions(-) diff --git a/doc/guides/prog_guide/vhost_lib.rst b/doc/guides/prog_guide/vhost_lib.rst index 0b6eda7..ab35b74 100644 --- a/doc/guides/prog_guide/vhost_lib.rst +++ b/doc/guides/prog_guide/vhost_lib.rst @@ -31,25 +31,28 @@ Vhost Library = -The vhost cuse (cuse: user space character device driver) library implements a -vhost cuse driver. It also creates, manages and destroys vhost devices for -corresponding virtio devices in the guest. Vhost supported vSwitch could register -callbacks to this library, which will be called when a vhost device is activated -or deactivated by guest virtual machine. +The vhost library implements a user space vhost driver. It supports both vhost-cuse +(cuse: user space character device) and vhost-user(user space socket server). +It also creates, manages and destroys vhost devices for corresponding virtio +devices in the guest. Vhost supported vSwitch could register callbacks to this +library, which will be called when a vhost device is activated or deactivated +by guest virtual machine. Vhost API Overview -- * Vhost driver registration - rte_vhost_driver_register registers the vhost cuse driver into the system. - Character device file will be created in the /dev directory. + rte_vhost_driver_register registers the vhost driver into the system. + For vhost-cuse, character device file will be created under the /dev directory. Character device name is specified as the parameter. + For vhost-user, a unix domain socket server will be created with the parameter as + the local socket path. * Vhost session start rte_vhost_driver_session_start starts the vhost session loop. - Vhost cuse session is an infinite blocking loop. + Vhost session is an infinite blocking loop. Put the session in a dedicate DPDK thread. * Callback register @@ -73,6 +76,8 @@ Vhost API Overview Vhost Implementation +Vhost cuse implementation +~ When vSwitch registers the vhost driver, it will register a cuse device driver into the system and creates a character device file. This cuse driver will receive vhost open/release/IOCTL message from QEMU simulator. @@ -89,13 +94,40 @@ which means vhost could access the shared virtio ring and the guest physical address specified in the entry of the ring. The guest virtual machine tells the vhost whether the virtio device is ready -for processing or is de-activated through VHOST_SET_BACKEND message. +for processing or is de-activated through VHOST_NET_SET_BACKEND message. The registered callback from vSwitch will be called. When the release call is released, vhost will destroy the device. +Vhost user implementation +~ +When vSwitch registers a vhost driver, it will create a unix domain socket server +into the system. This server will listen for a connection and process the vhost message from +QEMU simulator. + +When there is a new socket connection, it means a new virtio device has been created in +the guest virtual machine, and the vhost driver will create a vhost device for this virtio device. + +For messages with a file descriptor, the file descriptor could be directly used in the vhost +process as it is already installed by unix domain socket. + * VHOST_SET_MEM_TABLE + * VHOST_SET_VRING_KICK + * VHOST_SET_VRING_CALL + * VHOST_SET_LOG_FD + * VHOST_SET_VRING_ERR + +For VHOST_SET_MEM_TABLE message, QEMU will send us information for each memory region and its +file descriptor in the ancillary data of the message. The fd is used to map that region. + +There is no VHOST_NET_SET_BACKEND message as in vhost cuse to signal us whether virtio device +is ready or should be stopped. +VHOST_SET_VRING_KICK is used as the signal to put the vhost device onto data plane. +VHOST_GET_VRING_BASE is used as the signal to remove vhost device from data plane. + +When the socket connection is closed, vhost will destroy the device. + Vhost supported vSwitch reference - -For how to support vhost in vSwitch, please refer to vhost example in the +For more vhost details and how to support vhost in vSwitch, please refer to vhost example in the DPDK Sample Applications Guide. -- 1.8.1.4
[dpdk-dev] [PATCH 00/10] eal: rte_tailq api cleanup
On 2015/03/12 7:25, Mcnamara, John wrote: > > From: David Marchand [mailto:david.marchand at 6wind.com] > Sent: Wednesday, March 11, 2015 9:30 PM > To: Mcnamara, John > Cc: Tetsuya Mukawa; dev at dpdk.org > Subject: Re: [dpdk-dev] [PATCH 00/10] eal: rte_tailq api cleanup > >>> I just tested the patch with testpmd and it fixes the previous issue. >> Ok, thanks, just for the record, can you add your setup ? >> linux / bsd, uio / vfio (if linux), which pmd ? > > Hi David, > > I always complain when people don't send enough details in bug reports or > updates, so sorry about that. > > Here we go, it is pretty vanilla but I sure you will get other data points by > morning: > > $ uname -a > Linux sie-lab-214-036.ir.intel.com 3.6.10-4.fc18.x86_64 > #1 SMP Tue Dec 11 18:01:27 UTC 2012 x86_64 x86_64 x86_64 GNU/Linux > > UIO with testpmd and skeleton forwarding app > > John. > -- > Hi David, I've checked the patch on following environments, and make sure it works fine. - igb_uio: - vfio: - uio_pci_generic Tested on Ubuntu14.04 (3.13.0-30-generic) with e1000 PMD. - nic_uio Tested on FreeBSD 10.1-RELEASE #0 r274401 with e1000 PMD. Thanks, Tetsuya
[dpdk-dev] [PATCH 00/10] eal: rte_tailq api cleanup
On 2015/03/12 11:05, Tetsuya Mukawa wrote: > On 2015/03/12 7:25, Mcnamara, John wrote: >> From: David Marchand [mailto:david.marchand at 6wind.com] >> Sent: Wednesday, March 11, 2015 9:30 PM >> To: Mcnamara, John >> Cc: Tetsuya Mukawa; dev at dpdk.org >> Subject: Re: [dpdk-dev] [PATCH 00/10] eal: rte_tailq api cleanup >> I just tested the patch with testpmd and it fixes the previous issue. >>> Ok, thanks, just for the record, can you add your setup ? >>> linux / bsd, uio / vfio (if linux), which pmd ? >> Hi David, >> >> I always complain when people don't send enough details in bug reports or >> updates, so sorry about that. >> >> Here we go, it is pretty vanilla but I sure you will get other data points >> by morning: >> >> $ uname -a >> Linux sie-lab-214-036.ir.intel.com 3.6.10-4.fc18.x86_64 >> #1 SMP Tue Dec 11 18:01:27 UTC 2012 x86_64 x86_64 x86_64 GNU/Linux >> >> UIO with testpmd and skeleton forwarding app >> >> John. >> -- >> > Hi David, > > I've checked the patch on following environments, and make sure it works > fine. > > - igb_uio: > - vfio: > - uio_pci_generic > Tested on Ubuntu14.04 (3.13.0-30-generic) with e1000 PMD. > > - nic_uio > Tested on FreeBSD 10.1-RELEASE #0 r274401 with e1000 PMD. > > Thanks, > Tetsuya > Sorry for my bad English. I've already made sure above. Everything works fine with the patch on my environments. Tetsuya
[dpdk-dev] [RFC] af_packet: support port hotplug
Hi John.
On 2015/03/11 3:36, John W. Linville wrote:
> This patch adds finalization code to free resources allocated by
> the PMD. This is based on earlier patches for other PMDs by Tetsuya
> Mukawa , with corrections related to data->name.
>
> Signed-off-by: John W. Linville
> Cc: Tetsuya Mukawa
> ---
> lib/librte_pmd_af_packet/rte_eth_af_packet.c | 56
>
> 1 file changed, 56 insertions(+)
>
> diff --git a/lib/librte_pmd_af_packet/rte_eth_af_packet.c
> b/lib/librte_pmd_af_packet/rte_eth_af_packet.c
> index 80e9bdf7f819..57998ab19c96 100644
> --- a/lib/librte_pmd_af_packet/rte_eth_af_packet.c
> +++ b/lib/librte_pmd_af_packet/rte_eth_af_packet.c
> @@ -399,6 +399,13 @@ static struct eth_dev_ops ops = {
> .stats_reset = eth_stats_reset,
> };
>
> +static struct eth_driver rte_af_packet_pmd = {
> + .pci_drv = {
> + .name = "rte_af_packet_pmd",
> + .drv_flags = RTE_PCI_DRV_DETACHABLE,
> + },
> +};
> +
> /*
> * Opens an AF_PACKET socket
> */
> @@ -653,6 +660,10 @@ rte_pmd_init_internals(const char *name,
> if (*eth_dev == NULL)
> goto error;
>
> + /* check length of device name */
> + if ((strlen((*eth_dev)->data->name) + 1) > sizeof(data->name))
> + goto error;
> +
> /*
>* now put it all together
>* - store queue data in internals,
> @@ -669,12 +680,14 @@ rte_pmd_init_internals(const char *name,
> data->nb_tx_queues = (uint16_t)nb_queues;
> data->dev_link = pmd_link;
> data->mac_addrs = &(*internals)->eth_addr;
> + strncpy(data->name, (*eth_dev)->data->name,
> strlen((*eth_dev)->data->name));
Above line is over 80 characters.
(I am not sure we need to keep it, but may be good.)
>
> pci_dev->numa_node = numa_node;
>
> (*eth_dev)->data = data;
> (*eth_dev)->dev_ops = &ops;
> (*eth_dev)->pci_dev = pci_dev;
> + (*eth_dev)->driver = &rte_af_packet_pmd;
>
> return 0;
>
> @@ -835,10 +848,53 @@ rte_pmd_af_packet_devinit(const char *name, const char
> *params)
> return 0;
> }
>
> +static int
> +rte_pmd_af_packet_devuninit(const char *name)
> +{
> + struct rte_eth_dev *eth_dev = NULL;
> + struct pmd_internals *internals;
> + struct tpacket_req req;
> +
> + unsigned q;
> +
> + RTE_LOG(INFO, PMD, "Closing AF_PACKET ethdev on numa socket %u\n",
> + rte_socket_id());
> +
> + if (name == NULL)
> + return -1;
> +
> + /* retrieve ethdev entry */
> + eth_dev = rte_eth_dev_allocated(name);
> + if (eth_dev == NULL)
> + return -1;
> +
> + internals = eth_dev->data->dev_private;
> + req = internals->req;
> +
> + for (q = 0; q < internals->nb_queues; q++) {
> + munmap(internals->rx_queue[q].map,
> +2 * req.tp_block_size * req.tp_block_nr);
> + if (internals->rx_queue[q].rd)
> + rte_free(internals->rx_queue[q].rd);
> + if (internals->tx_queue[q].rd)
> + rte_free(internals->tx_queue[q].rd);
> + }
> +
> + rte_free(internals);
> + rte_free(eth_dev->data);
> + rte_free(eth_dev->pci_dev);
> +
> + rte_eth_dev_release_port(eth_dev);
> +
> +
> + return 0;
> +}
> +
> static struct rte_driver pmd_af_packet_drv = {
> .name = "eth_af_packet",
> .type = PMD_VDEV,
> .init = rte_pmd_af_packet_devinit,
> + .uninit = rte_pmd_af_packet_devuninit,
> };
>
> PMD_REGISTER_DRIVER(pmd_af_packet_drv);
Except above, I guess it's good.
I've checked followings.
- Code review
- No missing symbol
- No compile error with icc
- No compile error with gcc-4.6
- No compile error with gcc-4.7
- No compile error with gcc-4.8
- No compile error with gcc-4.9
- checkpatch.pl
Thanks,
Tetsuya
[dpdk-dev] [PATCH 00/10] eal: rte_tailq api cleanup
Hello John, Tetsuya, On Thu, Mar 12, 2015 at 3:05 AM, Tetsuya Mukawa wrote: > On 2015/03/12 7:25, Mcnamara, John wrote: > > Here we go, it is pretty vanilla but I sure you will get other data > points by morning: > > > > $ uname -a > > Linux sie-lab-214-036.ir.intel.com 3.6.10-4.fc18.x86_64 > > #1 SMP Tue Dec 11 18:01:27 UTC 2012 x86_64 x86_64 x86_64 GNU/Linux > > > > UIO with testpmd and skeleton forwarding app > > I've checked the patch on following environments, and make sure it works > fine. > > - igb_uio: > - vfio: > - uio_pci_generic > Tested on Ubuntu14.04 (3.13.0-30-generic) with e1000 PMD. > > - nic_uio > Tested on FreeBSD 10.1-RELEASE #0 r274401 with e1000 PMD. > Ok, good. Big thanks to both of you :-). The only case remaining would be ivshmem. -- David Marchand
[dpdk-dev] [PATCH] doc: add l2fwd-jobstats user guide
> -Original Message- > From: Wodkowski, PawelX > Sent: Wednesday, March 11, 2015 9:42 AM > To: dev at dpdk.org > Cc: Butler, Siobhan A > Subject: [PATCH] doc: add l2fwd-jobstats user guide > > Signed-off-by: Pawel Wodkowski > --- > doc/guides/sample_app_ug/index.rst| 1 + > doc/guides/sample_app_ug/l2_forward_job_stats.rst | 637 > ++ > 2 files changed, 638 insertions(+) > create mode 100644 doc/guides/sample_app_ug/l2_forward_job_stats.rst > > diff --git a/doc/guides/sample_app_ug/index.rst > b/doc/guides/sample_app_ug/index.rst > index 5720181..c89a2f0 100644 > --- a/doc/guides/sample_app_ug/index.rst > +++ b/doc/guides/sample_app_ug/index.rst > @@ -47,6 +47,7 @@ Sample Applications User Guide > ipv4_multicast > ip_reassembly > kernel_nic_interface > +l2_forward_job_stats > l2_forward_real_virtual > l3_forward > l3_forward_power_man > diff --git a/doc/guides/sample_app_ug/l2_forward_job_stats.rst > b/doc/guides/sample_app_ug/l2_forward_job_stats.rst > new file mode 100644 > index 000..76cea71 > --- /dev/null > +++ b/doc/guides/sample_app_ug/l2_forward_job_stats.rst > @@ -0,0 +1,637 @@ > +.. BSD LICENSE > +Copyright(c) 2010-2015 Intel Corporation. All rights reserved. > +All rights reserved. > + > +Redistribution and use in source and binary forms, with or without > +modification, are permitted provided that the following conditions > +are met: > + > +* Redistributions of source code must retain the above copyright > +notice, this list of conditions and the following disclaimer. > +* Redistributions in binary form must reproduce the above copyright > +notice, this list of conditions and the following disclaimer in > +the documentation and/or other materials provided with the > +distribution. > +* Neither the name of Intel Corporation nor the names of its > +contributors may be used to endorse or promote products derived > +from this software without specific prior written permission. > + > +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND > CONTRIBUTORS > +"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT > NOT > +LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND > FITNESS FOR > +A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE > COPYRIGHT > +OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, > INCIDENTAL, > +SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT > NOT > +LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS > OF USE, > +DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED > AND ON ANY > +THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT > +(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF > THE USE > +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH > DAMAGE. > + > +L2 Forwarding Sample Application (in Real and Virtualized Environments) > with core load statistics. > += > == > +=== > + > +The L2 Forwarding sample application is a simple example of packet > +processing using the Data Plane Development Kit (DPDK) which also takes > +advantage of Single Root I/O Virtualization (SR-IOV) features in a > virtualized > environment. > + > +.. note:: > + > +This application is a variation of L2 Forwarding sample application. It > demonstrate possible > +scheme of job stats library usage therefore some parts of this document > is identical with original > +L2 forwarding application. > + > +Overview > + > + > +The L2 Forwarding sample application, which can operate in real and > +virtualized environments, performs L2 forwarding for each packet that is > received. > +The destination port is the adjacent port from the enabled portmask, > +that is, if the first four ports are enabled (portmask 0xf), ports 1 > +and 2 forward into each other, and ports 3 and 4 forward into each other. > +Also, the MAC addresses are affected as follows: > + > +* The source MAC address is replaced by the TX port MAC address > + > +* The destination MAC address is replaced by 02:00:00:00:00:TX_PORT_ID > + > +This application can be used to benchmark performance using a traffic- > generator, as shown in the Figure 3. > + > +The application can also be used in a virtualized environment as shown in > Figure 4. > + > +The L2 Forwarding application can also be used as a starting point for > developing a new application based on the DPDK. > + > +.. _figure_3: > + > +**Figure 3. Performance Benchmark Setup (Basic Environment)** > + > +.. image4_png has been replaced > + > +|l2_fwd_benchmark_setup| > + > +.. _figure_4: > + > +**Figure 4. Performance Benchmark Setup (Virtualized Environment)** > + > +.. image5_png has been renamed > + > +|l2_fwd_virtenv_benchmark_setup| > + > +Virtual Function Setup Instructions > +~~
[dpdk-dev] [PATCH] eal: fix tailq init for uio/vfio resources
2015-03-11 18:25, David Marchand:
> Commit a2348166ea18 ("tailq: move to dynamic tailq") introduced a bug in
> uio/vfio resources list init.
>
> These resources list were pointed at through a pointer initialised only once
> but
> too early in the eal init (before tailqs init).
>
> Fix this by "resolving" this pointer when used (which is well after tailqs
> init).
>
> Fixes: a2348166ea18 ("tailq: move to dynamic tailq")
> Reported-by: Marvin Liu
> Reported-by: Tetsuya Mukawa
> Signed-off-by: David Marchand
> ---
>
> This patch is an alternative to the patch proposed by Marvin.
> It has been tested only on Linux and with uio driver.
Applied, thanks
[dpdk-dev] SIMD checksum
Howdy, I am writing to share some SIMD (SSE2 and AVX2) IP checksum routines. The commit log for rte_ip.h said that this was an area of future interest for DPDK. Code: https://github.com/lukego/snabbswitch/blob/ipchecksum-simd/src/lib/checksum.c Feedback welcome. We are currently reviewing and integrating this ourselves. Great if it is of use to other people too. The performance seems to be better and I see this as very valuable for reducing variance between offloaded and non-offloaded configurations. (I would like to replace NIC offload with SIMD offload completely to simplify the end-user's mental model but it remains to be seen if I will get away with this.) Sorry that this is not sent as a pull request. We use DPDK as a reference implementation but not as a software dependency. The rest of our implementation (tests, runtime CPU feature dispatching) would not fit into the DPDK code base and would have to be ported. In the perfect universe I would love to see useful routines like this living in a small repository of their own that everyboy could share. "Small and stand-alone subroutines for userspace networking." Perhaps we will collectively break up our big libraries like this over time. (I do really appreciate the fact that many DPDK library routines are easy to excerpt. You can see that we are using DPDK's IP checksum as the non-SIMD fallback, albeit we forked it to fit it into our project.) Cheers, -Luke
[dpdk-dev] [PATCH 00/10] eal: rte_tailq api cleanup
Hi David, I have verified this patch working with ivshmem and vfio driver. Testing environment: Commit: 0a530f0d58b0304d3f7515bb14f81fd2b6e6931e OS: Fedora20(3.11.10-301.fc20.x86_64) GCC: gcc version 4.8.3 20140911 QEMU: 1.5.2 with additional ivshmem patch NIC: Intel Corporation 82599ES 10-Gigabit SFI/SFP+ Network Connection [8086:10fb] > -Original Message- > From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of David Marchand > Sent: Thursday, March 12, 2015 1:44 PM > To: Tetsuya Mukawa; Mcnamara, John > Cc: dev at dpdk.org > Subject: Re: [dpdk-dev] [PATCH 00/10] eal: rte_tailq api cleanup > > Hello John, Tetsuya, > > On Thu, Mar 12, 2015 at 3:05 AM, Tetsuya Mukawa wrote: > > > On 2015/03/12 7:25, Mcnamara, John wrote: > > > Here we go, it is pretty vanilla but I sure you will get other data > > points by morning: > > > > > > $ uname -a > > > Linux sie-lab-214-036.ir.intel.com 3.6.10-4.fc18.x86_64 > > > #1 SMP Tue Dec 11 18:01:27 UTC 2012 x86_64 x86_64 x86_64 GNU/Linux > > > > > > UIO with testpmd and skeleton forwarding app > > > > I've checked the patch on following environments, and make sure it works > > fine. > > > > - igb_uio: > > - vfio: > > - uio_pci_generic > > Tested on Ubuntu14.04 (3.13.0-30-generic) with e1000 PMD. > > > > - nic_uio > > Tested on FreeBSD 10.1-RELEASE #0 r274401 with e1000 PMD. > > > > Ok, good. > Big thanks to both of you :-). > > The only case remaining would be ivshmem. > > > -- > David Marchand
[dpdk-dev] [PATCH v2 00/15] eal: Port Hotplug support for BSD
This patch series adds port hotplug support to BSD. Also, the patches consolidates pci uio functions of linuxapp and bsdapp. Following patches cleanup pci uio code to consolidate later. eal: Fix cording style of eal_pci.c and eal_pci_uio.c eal: Close file descriptor of uio configuration eal: Fix memory leak of pci_uio_map_resource() eal: Fix needless incrementation of pci_map_addr eal/bsdapp: Change names of pci related data structure eal: Use map_idx in pci_uio_map_resource() of bsdapp to work same as linuxapp eal: Fix interface of pci_map_resource() of bsdapp eal: Add pci_uio_alloc_uio_resource() eal: Add pci_uio_map_uio_resource_by_index() Following patches are actually for functions consolidation. eal: Consolidate pci_map and mapped_pci_resource of linuxapp and bsdapp eal: Consolidate rte_eal_pci_probe/close_one_driver() of linuxapp and bsdapp eal: Consolidate pci_map/unmap_device() of linuxapp and bsdapp eal: Consolidate pci_map/unmap_resource() of linuxapp and bsdapp eal: Consolidate pci uio functions of linuxapp and bsdapp The last patch removes CONFIG_RTE_LIBRTE_EAL_HOTPLUG configuration from Linux and BSD. And port hotplug is enabled as default with both Linux and BSD. eal: Enable Port Hotplug as default in Linux and BSD PATCH v2 changes - Consolidate pci uio functions of linuxapp and bsdapp. Tetsuya Mukawa (15): eal: Fix cording style of eal_pci.c and eal_pci_uio.c eal: Close file descriptor of uio configuration eal: Fix memory leak of pci_uio_map_resource() eal: Fix needless incrementation of pci_map_addr eal/bsdapp: Change names of pci related data structure eal: Use map_idx in pci_uio_map_resource() of bsdapp to work same as linuxapp eal: Fix interface of pci_map_resource() of bsdapp eal: Add pci_uio_alloc_uio_resource() eal: Add pci_uio_map_uio_resource_by_index() eal: Consolidate pci_map and mapped_pci_resource of linuxapp and bsdapp eal: Consolidate rte_eal_pci_probe/close_one_driver() of linuxapp and bsdapp eal: Consolidate pci_map/unmap_device() of linuxapp and bsdapp eal: Consolidate pci_map/unmap_resource() of linuxapp and bsdapp eal: Consolidate pci uio functions of linuxapp and bsdapp eal: Enable Port Hotplug as default in Linux and BSD config/common_bsdapp | 6 - config/common_linuxapp | 5 - lib/librte_eal/bsdapp/eal/Makefile | 1 + lib/librte_eal/bsdapp/eal/eal_pci.c| 292 ++--- .../bsdapp/eal/include/exec-env/rte_interrupts.h | 1 + lib/librte_eal/bsdapp/eal/rte_eal_version.map | 6 + lib/librte_eal/common/eal_common_dev.c | 2 - lib/librte_eal/common/eal_common_pci.c | 234 - lib/librte_eal/common/eal_common_pci_uio.c | 224 lib/librte_eal/common/eal_private.h| 57 +++- lib/librte_eal/common/include/rte_pci.h| 42 ++- lib/librte_eal/linuxapp/eal/Makefile | 1 + lib/librte_eal/linuxapp/eal/eal_pci.c | 236 + lib/librte_eal/linuxapp/eal/eal_pci_init.h | 39 +-- lib/librte_eal/linuxapp/eal/eal_pci_uio.c | 263 +-- lib/librte_ether/rte_ethdev.c | 22 +- 16 files changed, 702 insertions(+), 729 deletions(-) create mode 100644 lib/librte_eal/common/eal_common_pci_uio.c -- 1.9.1
[dpdk-dev] [PATCH v2 01/15] eal: Fix cording style of eal_pci.c and eal_pci_uio.c
This patch fixes cording style of below files in linuxapp and bsdapp.
- eal_pci.c
- eal_pci_uio.c
Signed-off-by: Tetsuya Mukawa
---
lib/librte_eal/bsdapp/eal/eal_pci.c | 67 ++-
lib/librte_eal/linuxapp/eal/eal_pci.c | 32 +--
lib/librte_eal/linuxapp/eal/eal_pci_uio.c | 37 ++---
3 files changed, 80 insertions(+), 56 deletions(-)
diff --git a/lib/librte_eal/bsdapp/eal/eal_pci.c
b/lib/librte_eal/bsdapp/eal/eal_pci.c
index fe3ef86..cbd0a4e 100644
--- a/lib/librte_eal/bsdapp/eal/eal_pci.c
+++ b/lib/librte_eal/bsdapp/eal/eal_pci.c
@@ -142,8 +142,9 @@ pci_map_resource(void *requested_addr, const char *devname,
off_t offset,
MAP_SHARED, fd, offset);
close(fd);
if (mapaddr == MAP_FAILED ||
- (requested_addr != NULL && mapaddr != requested_addr)) {
- RTE_LOG(ERR, EAL, "%s(): cannot mmap(%s(%d), %p, 0x%lx, 0x%lx):"
+ (requested_addr != NULL && mapaddr != requested_addr)) {
+ RTE_LOG(ERR, EAL,
+ "%s(): cannot mmap(%s(%d), %p, 0x%lx, 0x%lx):"
" %s (%p)\n", __func__, devname, fd, requested_addr,
(unsigned long)size, (unsigned long)offset,
strerror(errno), mapaddr);
@@ -161,9 +162,10 @@ fail:
static int
pci_uio_map_secondary(struct rte_pci_device *dev)
{
-size_t i;
-struct uio_resource *uio_res;
- struct uio_res_list *uio_res_list = RTE_TAILQ_CAST(rte_uio_tailq.head,
uio_res_list);
+ size_t i;
+ struct uio_resource *uio_res;
+ struct uio_res_list *uio_res_list =
+ RTE_TAILQ_CAST(rte_uio_tailq.head, uio_res_list);
TAILQ_FOREACH(uio_res, uio_res_list, next) {
@@ -179,10 +181,10 @@ pci_uio_map_secondary(struct rte_pci_device *dev)
!= uio_res->maps[i].addr) {
RTE_LOG(ERR, EAL,
"Cannot mmap device resource\n");
- return (-1);
+ return -1;
}
}
- return (0);
+ return 0;
}
RTE_LOG(ERR, EAL, "Cannot find resource for device\n");
@@ -201,7 +203,8 @@ pci_uio_map_resource(struct rte_pci_device *dev)
uint64_t pagesz;
struct rte_pci_addr *loc = &dev->addr;
struct uio_resource *uio_res;
- struct uio_res_list *uio_res_list = RTE_TAILQ_CAST(rte_uio_tailq.head,
uio_res_list);
+ struct uio_res_list *uio_res_list =
+ RTE_TAILQ_CAST(rte_uio_tailq.head, uio_res_list);
struct uio_map *maps;
dev->intr_handle.fd = -1;
@@ -209,14 +212,16 @@ pci_uio_map_resource(struct rte_pci_device *dev)
/* secondary processes - use already recorded details */
if (rte_eal_process_type() != RTE_PROC_PRIMARY)
- return (pci_uio_map_secondary(dev));
+ return pci_uio_map_secondary(dev);
snprintf(devname, sizeof(devname), "/dev/uio at pci:%u:%u:%u",
dev->addr.bus, dev->addr.devid, dev->addr.function);
if (access(devname, O_RDWR) < 0) {
- RTE_LOG(WARNING, EAL, " "PCI_PRI_FMT" not managed by UIO
driver, "
- "skipping\n", loc->domain, loc->bus,
loc->devid, loc->function);
+ RTE_LOG(WARNING, EAL,
+ " "PCI_PRI_FMT" not managed by UIO driver, "
+ "skipping\n", loc->domain, loc->bus,
+ loc->devid, loc->function);
return 1;
}
@@ -233,7 +238,7 @@ pci_uio_map_resource(struct rte_pci_device *dev)
if ((uio_res = rte_zmalloc("UIO_RES", sizeof (*uio_res), 0)) == NULL) {
RTE_LOG(ERR, EAL,
"%s(): cannot store uio mmap details\n", __func__);
- return (-1);
+ return -1;
}
snprintf(uio_res->path, sizeof(uio_res->path), "%s", devname);
@@ -248,7 +253,8 @@ pci_uio_map_resource(struct rte_pci_device *dev)
j = uio_res->nb_maps;
/* skip empty BAR */
- if ((phaddr = dev->mem_resource[i].phys_addr) == 0)
+ phaddr = dev->mem_resource[i].phys_addr;
+ if (phaddr == 0)
continue;
/* if matching map is found, then use it */
@@ -261,7 +267,7 @@ pci_uio_map_resource(struct rte_pci_device *dev)
(size_t)maps[j].size)
) == NULL) {
rte_free(uio_res);
- return (-1);
+ return -1;
}
maps[j].addr = mapaddr;
@@ -271,7 +277,7 @@ pci_uio_map_resource(struct rte_pci_device *dev)
TAILQ_INSERT_TAIL(uio_res_list, uio_res, next);
- return (0);
+
[dpdk-dev] [PATCH v2 02/15] eal: Close file descriptor of uio configuration
When pci_uio_unmap_resource() is called, a file descriptor that is used for uio configuration should be closed. Signed-off-by: Tetsuya Mukawa --- lib/librte_eal/linuxapp/eal/eal_pci_uio.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/lib/librte_eal/linuxapp/eal/eal_pci_uio.c b/lib/librte_eal/linuxapp/eal/eal_pci_uio.c index 6f229d6..77bb5ed 100644 --- a/lib/librte_eal/linuxapp/eal/eal_pci_uio.c +++ b/lib/librte_eal/linuxapp/eal/eal_pci_uio.c @@ -462,6 +462,8 @@ pci_uio_unmap_resource(struct rte_pci_device *dev) /* close fd if in primary process */ close(dev->intr_handle.fd); + if (dev->intr_handle.uio_cfg_fd >= 0) + close(dev->intr_handle.uio_cfg_fd); dev->intr_handle.fd = -1; dev->intr_handle.type = RTE_INTR_HANDLE_UNKNOWN; -- 1.9.1
[dpdk-dev] [PATCH v2 03/15] eal: Fix memory leak of pci_uio_map_resource()
When pci_map_resource() is failed but path is allocated correctly,
path won't be freed. Also, when open() is failed, uio_res won't be freed.
This patch fixes these.
Signed-off-by: Tetsuya Mukawa
---
lib/librte_eal/linuxapp/eal/eal_pci_uio.c | 20 +---
1 file changed, 13 insertions(+), 7 deletions(-)
diff --git a/lib/librte_eal/linuxapp/eal/eal_pci_uio.c
b/lib/librte_eal/linuxapp/eal/eal_pci_uio.c
index 77bb5ed..901f277 100644
--- a/lib/librte_eal/linuxapp/eal/eal_pci_uio.c
+++ b/lib/librte_eal/linuxapp/eal/eal_pci_uio.c
@@ -350,6 +350,11 @@ pci_uio_map_resource(struct rte_pci_device *dev)
loc->domain, loc->bus, loc->devid,
loc->function, i);
+ /* allocate memory to keep path */
+ maps[map_idx].path = rte_malloc(NULL, strlen(devname) + 1, 0);
+ if (maps[map_idx].path == NULL)
+ goto fail0;
+
/*
* open resource file, to mmap it
*/
@@ -357,7 +362,7 @@ pci_uio_map_resource(struct rte_pci_device *dev)
if (fd < 0) {
RTE_LOG(ERR, EAL, "Cannot open %s: %s\n",
devname, strerror(errno));
- return -1;
+ goto fail0;
}
/* try mapping somewhere close to the end of hugepages */
@@ -372,14 +377,9 @@ pci_uio_map_resource(struct rte_pci_device *dev)
pci_map_addr = RTE_PTR_ADD(mapaddr,
(size_t)dev->mem_resource[i].len);
- maps[map_idx].path = rte_malloc(NULL, strlen(devname) + 1, 0);
- if (maps[map_idx].path == NULL)
- fail = 1;
-
if (fail) {
- rte_free(uio_res);
close(fd);
- return -1;
+ goto fail1;
}
close(fd);
@@ -397,6 +397,12 @@ pci_uio_map_resource(struct rte_pci_device *dev)
TAILQ_INSERT_TAIL(uio_res_list, uio_res, next);
return 0;
+
+fail1:
+ rte_free(maps[map_idx].path);
+fail0:
+ rte_free(uio_res);
+ return -1;
}
#ifdef RTE_LIBRTE_EAL_HOTPLUG
--
1.9.1
[dpdk-dev] [PATCH v2 04/15] eal: Fix needless incrementation of pci_map_addr
When pci_map_resource() is failed, mapaddr will be MAP_FAILED.
In this case, pci_map_addr should not be incremented.
The patch fixes it. Also, fix below.
- To shrink code, move close().
- Remove fail variable.
Signed-off-by: Tetsuya Mukawa
---
lib/librte_eal/linuxapp/eal/eal_pci_uio.c | 10 ++
1 file changed, 2 insertions(+), 8 deletions(-)
diff --git a/lib/librte_eal/linuxapp/eal/eal_pci_uio.c
b/lib/librte_eal/linuxapp/eal/eal_pci_uio.c
index 901f277..2741c62 100644
--- a/lib/librte_eal/linuxapp/eal/eal_pci_uio.c
+++ b/lib/librte_eal/linuxapp/eal/eal_pci_uio.c
@@ -337,7 +337,6 @@ pci_uio_map_resource(struct rte_pci_device *dev)
maps = uio_res->maps;
for (i = 0, map_idx = 0; i != PCI_MAX_RESOURCE; i++) {
int fd;
- int fail = 0;
/* skip empty BAR */
phaddr = dev->mem_resource[i].phys_addr;
@@ -371,18 +370,13 @@ pci_uio_map_resource(struct rte_pci_device *dev)
mapaddr = pci_map_resource(pci_map_addr, fd, 0,
(size_t)dev->mem_resource[i].len, 0);
+ close(fd);
if (mapaddr == MAP_FAILED)
- fail = 1;
+ goto fail1;
pci_map_addr = RTE_PTR_ADD(mapaddr,
(size_t)dev->mem_resource[i].len);
- if (fail) {
- close(fd);
- goto fail1;
- }
- close(fd);
-
maps[map_idx].phaddr = dev->mem_resource[i].phys_addr;
maps[map_idx].size = dev->mem_resource[i].len;
maps[map_idx].addr = mapaddr;
--
1.9.1
[dpdk-dev] [PATCH v2 05/15] eal/bsdapp: Change names of pci related data structure
To merge pci code of linuxapp and bsdapp, this patch changes names
like below.
- uio_map to pci_map
- uio_resource to mapped_pci_resource
- uio_res_list to mapped_pci_res_list
Also, add 'path' variable to pci_map of bsdapp.
Signed-off-by: Tetsuya Mukawa
---
lib/librte_eal/bsdapp/eal/eal_pci.c | 25 +
1 file changed, 13 insertions(+), 12 deletions(-)
diff --git a/lib/librte_eal/bsdapp/eal/eal_pci.c
b/lib/librte_eal/bsdapp/eal/eal_pci.c
index cbd0a4e..7fcb15d 100644
--- a/lib/librte_eal/bsdapp/eal/eal_pci.c
+++ b/lib/librte_eal/bsdapp/eal/eal_pci.c
@@ -83,8 +83,9 @@
* enabling bus master.
*/
-struct uio_map {
+struct pci_map {
void *addr;
+ char *path;
uint64_t offset;
uint64_t size;
uint64_t phaddr;
@@ -94,16 +95,16 @@ struct uio_map {
* For multi-process we need to reproduce all PCI mappings in secondary
* processes, so save them in a tailq.
*/
-struct uio_resource {
- TAILQ_ENTRY(uio_resource) next;
+struct mapped_pci_resource {
+ TAILQ_ENTRY(mapped_pci_resource) next;
struct rte_pci_addr pci_addr;
char path[PATH_MAX];
size_t nb_maps;
- struct uio_map maps[PCI_MAX_RESOURCE];
+ struct pci_map maps[PCI_MAX_RESOURCE];
};
-TAILQ_HEAD(uio_res_list, uio_resource);
+TAILQ_HEAD(mapped_pci_res_list, mapped_pci_resource);
static struct rte_tailq_elem rte_uio_tailq = {
.name = "UIO_RESOURCE_LIST",
@@ -163,9 +164,9 @@ static int
pci_uio_map_secondary(struct rte_pci_device *dev)
{
size_t i;
- struct uio_resource *uio_res;
- struct uio_res_list *uio_res_list =
- RTE_TAILQ_CAST(rte_uio_tailq.head, uio_res_list);
+ struct mapped_pci_resource *uio_res;
+ struct mapped_pci_res_list *uio_res_list =
+ RTE_TAILQ_CAST(rte_uio_tailq.head, mapped_pci_res_list);
TAILQ_FOREACH(uio_res, uio_res_list, next) {
@@ -202,10 +203,10 @@ pci_uio_map_resource(struct rte_pci_device *dev)
uint64_t offset;
uint64_t pagesz;
struct rte_pci_addr *loc = &dev->addr;
- struct uio_resource *uio_res;
- struct uio_res_list *uio_res_list =
- RTE_TAILQ_CAST(rte_uio_tailq.head, uio_res_list);
- struct uio_map *maps;
+ struct mapped_pci_resource *uio_res;
+ struct mapped_pci_res_list *uio_res_list =
+ RTE_TAILQ_CAST(rte_uio_tailq.head, mapped_pci_res_list);
+ struct pci_map *maps;
dev->intr_handle.fd = -1;
dev->intr_handle.type = RTE_INTR_HANDLE_UNKNOWN;
--
1.9.1
[dpdk-dev] [PATCH v2 06/15] eal: Use map_idx in pci_uio_map_resource() of bsdapp to work same as linuxapp
This patch changes code that maps pci resources in bsdapp.
Linuxapp has almost same code. To consolidate both, fix implementation
of bsdapp to work same as linuxapp.
Signed-off-by: Tetsuya Mukawa
---
lib/librte_eal/bsdapp/eal/eal_pci.c | 24
1 file changed, 12 insertions(+), 12 deletions(-)
diff --git a/lib/librte_eal/bsdapp/eal/eal_pci.c
b/lib/librte_eal/bsdapp/eal/eal_pci.c
index 7fcb15d..bd0f8cd 100644
--- a/lib/librte_eal/bsdapp/eal/eal_pci.c
+++ b/lib/librte_eal/bsdapp/eal/eal_pci.c
@@ -196,7 +196,7 @@ pci_uio_map_secondary(struct rte_pci_device *dev)
static int
pci_uio_map_resource(struct rte_pci_device *dev)
{
- int i, j;
+ int i, map_idx;
char devname[PATH_MAX]; /* contains the /dev/uioX */
void *mapaddr;
uint64_t phaddr;
@@ -250,9 +250,8 @@ pci_uio_map_resource(struct rte_pci_device *dev)
pagesz = sysconf(_SC_PAGESIZE);
maps = uio_res->maps;
- for (i = uio_res->nb_maps = 0; i != PCI_MAX_RESOURCE; i++) {
+ for (i = 0, map_idx = 0; i != PCI_MAX_RESOURCE; i++) {
- j = uio_res->nb_maps;
/* skip empty BAR */
phaddr = dev->mem_resource[i].phys_addr;
if (phaddr == 0)
@@ -260,22 +259,23 @@ pci_uio_map_resource(struct rte_pci_device *dev)
/* if matching map is found, then use it */
offset = i * pagesz;
- maps[j].offset = offset;
- maps[j].phaddr = dev->mem_resource[i].phys_addr;
- maps[j].size = dev->mem_resource[i].len;
- if (maps[j].addr != NULL ||
- (mapaddr = pci_map_resource(NULL, devname, (off_t)offset,
- (size_t)maps[j].size)
- ) == NULL) {
+ maps[map_idx].offset = offset;
+ maps[map_idx].phaddr = dev->mem_resource[i].phys_addr;
+ maps[map_idx].size = dev->mem_resource[i].len;
+ mapaddr = pci_map_resource(NULL, devname, (off_t)offset,
+ (size_t)maps[map_idx].size);
+ if ((maps[map_idx].addr != NULL) || (mapaddr == NULL)) {
rte_free(uio_res);
return -1;
}
- maps[j].addr = mapaddr;
- uio_res->nb_maps++;
+ maps[map_idx].addr = mapaddr;
+ map_idx++;
dev->mem_resource[i].addr = mapaddr;
}
+ uio_res->nb_maps = map_idx;
+
TAILQ_INSERT_TAIL(uio_res_list, uio_res, next);
return 0;
--
1.9.1
[dpdk-dev] [PATCH v2 07/15] eal: Fix interface of pci_map_resource() of bsdapp
The function is implemented in both linuxapp and bsdapp, but interface
is different. The patch fixes the function of bsdapp to do same as
linuxapp. After applying it, file descriptor should be opened and
closed out of pci_map_resource().
Signed-off-by: Tetsuya Mukawa
---
lib/librte_eal/bsdapp/eal/eal_pci.c | 117 +++---
lib/librte_eal/linuxapp/eal/eal_pci_uio.c | 5 +-
2 files changed, 78 insertions(+), 44 deletions(-)
diff --git a/lib/librte_eal/bsdapp/eal/eal_pci.c
b/lib/librte_eal/bsdapp/eal/eal_pci.c
index bd0f8cd..c92c624 100644
--- a/lib/librte_eal/bsdapp/eal/eal_pci.c
+++ b/lib/librte_eal/bsdapp/eal/eal_pci.c
@@ -100,7 +100,7 @@ struct mapped_pci_resource {
struct rte_pci_addr pci_addr;
char path[PATH_MAX];
- size_t nb_maps;
+ int nb_maps;
struct pci_map maps[PCI_MAX_RESOURCE];
};
@@ -122,48 +122,31 @@ pci_unbind_kernel_driver(struct rte_pci_device *dev
__rte_unused)
/* map a particular resource from a file */
static void *
-pci_map_resource(void *requested_addr, const char *devname, off_t offset,
-size_t size)
+pci_map_resource(void *requested_addr, int fd, off_t offset, size_t size,
+int additional_flags)
{
- int fd;
void *mapaddr;
- /*
-* open devname, to mmap it
-*/
- fd = open(devname, O_RDWR);
- if (fd < 0) {
- RTE_LOG(ERR, EAL, "Cannot open %s: %s\n",
- devname, strerror(errno));
- goto fail;
- }
-
/* Map the PCI memory resource of device */
mapaddr = mmap(requested_addr, size, PROT_READ | PROT_WRITE,
- MAP_SHARED, fd, offset);
- close(fd);
- if (mapaddr == MAP_FAILED ||
- (requested_addr != NULL && mapaddr != requested_addr)) {
+ MAP_SHARED | additional_flags, fd, offset);
+ if (mapaddr == MAP_FAILED) {
RTE_LOG(ERR, EAL,
- "%s(): cannot mmap(%s(%d), %p, 0x%lx, 0x%lx):"
- " %s (%p)\n", __func__, devname, fd, requested_addr,
+ "%s(): cannot mmap(%d, %p, 0x%lx, 0x%lx): %s (%p)\n",
+ __func__, fd, requested_addr,
(unsigned long)size, (unsigned long)offset,
strerror(errno), mapaddr);
- goto fail;
+ } else {
+ RTE_LOG(DEBUG, EAL, " PCI memory mapped at %p\n", mapaddr);
}
- RTE_LOG(DEBUG, EAL, " PCI memory mapped at %p\n", mapaddr);
-
return mapaddr;
-
-fail:
- return NULL;
}
static int
pci_uio_map_secondary(struct rte_pci_device *dev)
{
- size_t i;
+ int i, fd;
struct mapped_pci_resource *uio_res;
struct mapped_pci_res_list *uio_res_list =
RTE_TAILQ_CAST(rte_uio_tailq.head, mapped_pci_res_list);
@@ -171,19 +154,42 @@ pci_uio_map_secondary(struct rte_pci_device *dev)
TAILQ_FOREACH(uio_res, uio_res_list, next) {
/* skip this element if it doesn't match our PCI address */
- if (memcmp(&uio_res->pci_addr, &dev->addr, sizeof(dev->addr)))
+ if (rte_eal_compare_pci_addr(&uio_res->pci_addr, &dev->addr))
continue;
for (i = 0; i != uio_res->nb_maps; i++) {
- if (pci_map_resource(uio_res->maps[i].addr,
-uio_res->path,
-(off_t)uio_res->maps[i].offset,
-(size_t)uio_res->maps[i].size)
- != uio_res->maps[i].addr) {
- RTE_LOG(ERR, EAL,
- "Cannot mmap device resource\n");
+ /*
+* open devname, to mmap it
+*/
+ fd = open(uio_res->maps[i].path, O_RDWR);
+ if (fd < 0) {
+ RTE_LOG(ERR, EAL, "Cannot open %s: %s\n",
+ uio_res->maps[i].path, strerror(errno));
+ return -1;
+ }
+
+ void *mapaddr = pci_map_resource(uio_res->maps[i].addr,
+ fd, (off_t)uio_res->maps[i].offset,
+ (size_t)uio_res->maps[i].size, 0);
+ if (mapaddr != uio_res->maps[i].addr) {
+ if (mapaddr == MAP_FAILED)
+ RTE_LOG(ERR, EAL,
+ "Cannot mmap device resource "
+ "file %s: %s\n",
+ uio_res->maps[i].path,
+ strerror(errno));
+
[dpdk-dev] [PATCH v2 08/15] eal: Add pci_uio_alloc_uio_resource()
This patch adds a new function called pci_uio_alloc_uio_resource().
The function hides how to prepare uio resource in linuxapp and bsdapp.
With the function, pci_uio_map_resource() will be more abstracted.
Signed-off-by: Tetsuya Mukawa
---
lib/librte_eal/bsdapp/eal/eal_pci.c | 61 --
lib/librte_eal/linuxapp/eal/eal_pci_uio.c | 63 ---
2 files changed, 83 insertions(+), 41 deletions(-)
diff --git a/lib/librte_eal/bsdapp/eal/eal_pci.c
b/lib/librte_eal/bsdapp/eal/eal_pci.c
index c92c624..f88cd15 100644
--- a/lib/librte_eal/bsdapp/eal/eal_pci.c
+++ b/lib/librte_eal/bsdapp/eal/eal_pci.c
@@ -198,28 +198,17 @@ pci_uio_map_secondary(struct rte_pci_device *dev)
return 1;
}
-/* map the PCI resource of a PCI device in virtual memory */
static int
-pci_uio_map_resource(struct rte_pci_device *dev)
+pci_uio_alloc_uio_resource(struct rte_pci_device *dev,
+ struct mapped_pci_resource **uio_res)
{
- int i, map_idx;
char devname[PATH_MAX]; /* contains the /dev/uioX */
- void *mapaddr;
- uint64_t phaddr;
- uint64_t offset;
- uint64_t pagesz;
- struct rte_pci_addr *loc = &dev->addr;
- struct mapped_pci_resource *uio_res;
- struct mapped_pci_res_list *uio_res_list =
- RTE_TAILQ_CAST(rte_uio_tailq.head, mapped_pci_res_list);
- struct pci_map *maps;
+ struct rte_pci_addr *loc;
- dev->intr_handle.fd = -1;
- dev->intr_handle.type = RTE_INTR_HANDLE_UNKNOWN;
+ if ((dev == NULL) || (uio_res == NULL))
+ return -1;
- /* secondary processes - use already recorded details */
- if (rte_eal_process_type() != RTE_PROC_PRIMARY)
- return pci_uio_map_secondary(dev);
+ loc = &dev->addr;
snprintf(devname, sizeof(devname), "/dev/uio at pci:%u:%u:%u",
dev->addr.bus, dev->addr.devid, dev->addr.function);
@@ -242,18 +231,50 @@ pci_uio_map_resource(struct rte_pci_device *dev)
dev->intr_handle.type = RTE_INTR_HANDLE_UIO;
/* allocate the mapping details for secondary processes*/
- if ((uio_res = rte_zmalloc("UIO_RES", sizeof (*uio_res), 0)) == NULL) {
+ *uio_res = rte_zmalloc("UIO_RES", sizeof(**uio_res), 0);
+ if (*uio_res == NULL) {
RTE_LOG(ERR, EAL,
"%s(): cannot store uio mmap details\n", __func__);
return -1;
}
- snprintf(uio_res->path, sizeof(uio_res->path), "%s", devname);
- memcpy(&uio_res->pci_addr, &dev->addr, sizeof(uio_res->pci_addr));
+ snprintf((*uio_res)->path, sizeof((*uio_res)->path), "%s", devname);
+ memcpy(&(*uio_res)->pci_addr,
+ &dev->addr, sizeof((*uio_res)->pci_addr));
+
+ return 0;
+}
+
+/* map the PCI resource of a PCI device in virtual memory */
+static int
+pci_uio_map_resource(struct rte_pci_device *dev)
+{
+ int i, map_idx, ret;
+ char *devname;
+ void *mapaddr;
+ uint64_t phaddr;
+ uint64_t offset;
+ uint64_t pagesz;
+ struct mapped_pci_resource *uio_res = NULL;
+ struct mapped_pci_res_list *uio_res_list =
+ RTE_TAILQ_CAST(rte_uio_tailq.head, mapped_pci_res_list);
+ struct pci_map *maps;
+
+ dev->intr_handle.fd = -1;
+ dev->intr_handle.type = RTE_INTR_HANDLE_UNKNOWN;
+
+ /* secondary processes - use already recorded details */
+ if (rte_eal_process_type() != RTE_PROC_PRIMARY)
+ return pci_uio_map_secondary(dev);
+ /* allocate uio resource */
+ ret = pci_uio_alloc_uio_resource(dev, &uio_res);
+ if ((ret != 0) || (uio_res == NULL))
+ return ret;
/* Map all BARs */
pagesz = sysconf(_SC_PAGESIZE);
+ devname = uio_res->path;
maps = uio_res->maps;
for (i = 0, map_idx = 0; i != PCI_MAX_RESOURCE; i++) {
diff --git a/lib/librte_eal/linuxapp/eal/eal_pci_uio.c
b/lib/librte_eal/linuxapp/eal/eal_pci_uio.c
index f100c1a..099c413 100644
--- a/lib/librte_eal/linuxapp/eal/eal_pci_uio.c
+++ b/lib/librte_eal/linuxapp/eal/eal_pci_uio.c
@@ -262,30 +262,20 @@ pci_get_uio_dev(struct rte_pci_device *dev, char *dstbuf,
return uio_num;
}
-/* map the PCI resource of a PCI device in virtual memory */
-int
-pci_uio_map_resource(struct rte_pci_device *dev)
+static int
+pci_uio_alloc_uio_resource(struct rte_pci_device *dev,
+ struct mapped_pci_resource **uio_res)
{
- int i, map_idx;
char dirname[PATH_MAX];
char cfgname[PATH_MAX];
char devname[PATH_MAX]; /* contains the /dev/uioX */
- void *mapaddr;
int uio_num;
- uint64_t phaddr;
- struct rte_pci_addr *loc = &dev->addr;
- struct mapped_pci_resource *uio_res;
- struct mapped_pci_res_list *uio_res_list =
- RTE_TAILQ_CAST(rte_uio_tailq.head, mapped_pci_res_list);
- struct pci_
[dpdk-dev] [PATCH v2 09/15] eal: Add pci_uio_map_uio_resource_by_index()
This patch adds a new function called pci_uio_map_resource_by_index().
The function hides how to map uio resource in linuxapp and bsdapp.
With the function, pci_uio_map_resource() will be more abstracted.
Signed-off-by: Tetsuya Mukawa
---
lib/librte_eal/bsdapp/eal/eal_pci.c | 108 +++---
lib/librte_eal/linuxapp/eal/eal_pci_uio.c | 124 +-
2 files changed, 135 insertions(+), 97 deletions(-)
diff --git a/lib/librte_eal/bsdapp/eal/eal_pci.c
b/lib/librte_eal/bsdapp/eal/eal_pci.c
index f88cd15..3e63835 100644
--- a/lib/librte_eal/bsdapp/eal/eal_pci.c
+++ b/lib/librte_eal/bsdapp/eal/eal_pci.c
@@ -245,20 +245,73 @@ pci_uio_alloc_uio_resource(struct rte_pci_device *dev,
return 0;
}
-/* map the PCI resource of a PCI device in virtual memory */
static int
-pci_uio_map_resource(struct rte_pci_device *dev)
+pci_uio_map_uio_resource_by_index(struct rte_pci_device *dev, int res_idx,
+ struct mapped_pci_resource *uio_res, int map_idx)
{
- int i, map_idx, ret;
+ int fd;
char *devname;
void *mapaddr;
- uint64_t phaddr;
uint64_t offset;
uint64_t pagesz;
+ struct pci_map *maps;
+
+ if ((dev == NULL) || (uio_res == NULL) || (uio_res->path == NULL))
+ return -1;
+
+ maps = uio_res->maps;
+ devname = uio_res->path;
+ pagesz = sysconf(_SC_PAGESIZE);
+
+ /* allocate memory to keep path */
+ maps[map_idx].path = rte_malloc(NULL, strlen(devname) + 1, 0);
+ if (maps[map_idx].path == NULL) {
+ RTE_LOG(ERR, EAL, "Cannot allocate memory for "
+ "path: %s\n", strerror(errno));
+ return -1;
+ }
+
+ /*
+* open resource file, to mmap it
+*/
+ fd = open(devname, O_RDWR);
+ if (fd < 0) {
+ RTE_LOG(ERR, EAL, "Cannot open %s: %s\n",
+ devname, strerror(errno));
+ goto fail;
+ }
+
+ /* if matching map is found, then use it */
+ offset = res_idx * pagesz;
+ mapaddr = pci_map_resource(NULL, fd, (off_t)offset,
+ (size_t)dev->mem_resource[res_idx].len, 0);
+ close(fd);
+ if (mapaddr == NULL)
+ goto fail;
+
+ maps[map_idx].phaddr = dev->mem_resource[res_idx].phys_addr;
+ maps[map_idx].size = dev->mem_resource[res_idx].len;
+ maps[map_idx].addr = mapaddr;
+ maps[map_idx].offset = offset;
+ strcpy(maps[map_idx].path, devname);
+ dev->mem_resource[res_idx].addr = mapaddr;
+
+ return 0;
+
+fail:
+ rte_free(maps[map_idx].path);
+ return -1;
+}
+
+/* map the PCI resource of a PCI device in virtual memory */
+static int
+pci_uio_map_resource(struct rte_pci_device *dev)
+{
+ int i, map_idx, ret;
+ uint64_t phaddr;
struct mapped_pci_resource *uio_res = NULL;
struct mapped_pci_res_list *uio_res_list =
RTE_TAILQ_CAST(rte_uio_tailq.head, mapped_pci_res_list);
- struct pci_map *maps;
dev->intr_handle.fd = -1;
dev->intr_handle.type = RTE_INTR_HANDLE_UNKNOWN;
@@ -273,51 +326,18 @@ pci_uio_map_resource(struct rte_pci_device *dev)
return ret;
/* Map all BARs */
- pagesz = sysconf(_SC_PAGESIZE);
- devname = uio_res->path;
-
- maps = uio_res->maps;
for (i = 0, map_idx = 0; i != PCI_MAX_RESOURCE; i++) {
- int fd;
-
/* skip empty BAR */
phaddr = dev->mem_resource[i].phys_addr;
if (phaddr == 0)
continue;
- /* allocate memory to keep path */
- maps[map_idx].path = rte_malloc(NULL, strlen(devname) + 1, 0);
- if (maps[map_idx].path == NULL) {
- RTE_LOG(ERR, EAL, "Cannot allocate memory for "
- "path: %s\n", strerror(errno));
- goto fail0;
- }
-
- /*
-* open resource file, to mmap it
-*/
- fd = open(devname, O_RDWR);
- if (fd < 0) {
- RTE_LOG(ERR, EAL, "Cannot open %s: %s\n",
- devname, strerror(errno));
- goto fail0;
- }
+ ret = pci_uio_map_uio_resource_by_index(dev, i,
+ uio_res, map_idx);
+ if (ret != 0)
+ goto fail;
- /* if matching map is found, then use it */
- offset = i * pagesz;
- mapaddr = pci_map_resource(NULL, fd, (off_t)offset,
- (size_t)dev->mem_resource[i].len, 0);
- close(fd);
- if (mapaddr == NULL)
- goto fail1;
-
- maps[map_idx].phaddr = dev->mem_resource[
[dpdk-dev] [PATCH v2 10/15] eal: Consolidate pci_map and mapped_pci_resource of linuxapp and bsdapp
This patch consolidates below structures, and defines them in common code.
- struct pci_map
- strucy mapped_pci_resouces
Signed-off-by: Tetsuya Mukawa
---
lib/librte_eal/bsdapp/eal/eal_pci.c| 24
lib/librte_eal/common/include/rte_pci.h| 28
lib/librte_eal/linuxapp/eal/eal_pci_init.h | 23 ---
3 files changed, 28 insertions(+), 47 deletions(-)
diff --git a/lib/librte_eal/bsdapp/eal/eal_pci.c
b/lib/librte_eal/bsdapp/eal/eal_pci.c
index 3e63835..67ad88f 100644
--- a/lib/librte_eal/bsdapp/eal/eal_pci.c
+++ b/lib/librte_eal/bsdapp/eal/eal_pci.c
@@ -82,30 +82,6 @@
* network card, only providing access to PCI BAR to applications, and
* enabling bus master.
*/
-
-struct pci_map {
- void *addr;
- char *path;
- uint64_t offset;
- uint64_t size;
- uint64_t phaddr;
-};
-
-/*
- * For multi-process we need to reproduce all PCI mappings in secondary
- * processes, so save them in a tailq.
- */
-struct mapped_pci_resource {
- TAILQ_ENTRY(mapped_pci_resource) next;
-
- struct rte_pci_addr pci_addr;
- char path[PATH_MAX];
- int nb_maps;
- struct pci_map maps[PCI_MAX_RESOURCE];
-};
-
-TAILQ_HEAD(mapped_pci_res_list, mapped_pci_resource);
-
static struct rte_tailq_elem rte_uio_tailq = {
.name = "UIO_RESOURCE_LIST",
};
diff --git a/lib/librte_eal/common/include/rte_pci.h
b/lib/librte_eal/common/include/rte_pci.h
index 995f814..fc64601 100644
--- a/lib/librte_eal/common/include/rte_pci.h
+++ b/lib/librte_eal/common/include/rte_pci.h
@@ -219,6 +219,34 @@ struct rte_pci_driver {
/** Device driver supports detaching capability */
#define RTE_PCI_DRV_DETACHABLE 0x0010
+/**
+ * A structure describing a PCI mapping.
+ */
+struct pci_map {
+ void *addr;
+ char *path;
+ uint64_t offset;
+ uint64_t size;
+ uint64_t phaddr;
+};
+
+/**
+ * A structure describing a mapped PCI resource.
+ * For multi-process we need to reproduce all PCI mappings in secondary
+ * processes, so save them in a tailq.
+ */
+struct mapped_pci_resource {
+ TAILQ_ENTRY(mapped_pci_resource) next;
+
+ struct rte_pci_addr pci_addr;
+ char path[PATH_MAX];
+ int nb_maps;
+ struct pci_map maps[PCI_MAX_RESOURCE];
+};
+
+/** mapped pci device list */
+TAILQ_HEAD(mapped_pci_res_list, mapped_pci_resource);
+
/**< Internal use only - Macro used by pci addr parsing functions **/
#define GET_PCIADDR_FIELD(in, fd, lim, dlm) \
do { \
diff --git a/lib/librte_eal/linuxapp/eal/eal_pci_init.h
b/lib/librte_eal/linuxapp/eal/eal_pci_init.h
index aa7b755..d9d1878 100644
--- a/lib/librte_eal/linuxapp/eal/eal_pci_init.h
+++ b/lib/librte_eal/linuxapp/eal/eal_pci_init.h
@@ -36,29 +36,6 @@
#include "eal_vfio.h"
-struct pci_map {
- void *addr;
- char *path;
- uint64_t offset;
- uint64_t size;
- uint64_t phaddr;
-};
-
-/*
- * For multi-process we need to reproduce all PCI mappings in secondary
- * processes, so save them in a tailq.
- */
-struct mapped_pci_resource {
- TAILQ_ENTRY(mapped_pci_resource) next;
-
- struct rte_pci_addr pci_addr;
- char path[PATH_MAX];
- int nb_maps;
- struct pci_map maps[PCI_MAX_RESOURCE];
-};
-
-TAILQ_HEAD(mapped_pci_res_list, mapped_pci_resource);
-
/*
* Helper function to map PCI resources right after hugepages in virtual memory
*/
--
1.9.1
[dpdk-dev] [PATCH v2 11/15] eal: Consolidate rte_eal_pci_probe/close_one_driver() of linuxapp and bsdapp
This patch consolidates below functions, and implements these in common
eal code.
- rte_eal_pci_probe_one_driver()
- rte_eal_pci_close_one_driver()
Because pci_map_device() is only implemented in linuxapp, the patch
implements it in bsdapp too. This implemented function will be merged to
linuxapp one with later patch.
Signed-off-by: Tetsuya Mukawa
---
lib/librte_eal/bsdapp/eal/eal_pci.c| 79 ++---
lib/librte_eal/common/eal_common_pci.c | 138 +
lib/librte_eal/common/eal_private.h| 21 ++---
lib/librte_eal/linuxapp/eal/eal_pci.c | 153 ++---
4 files changed, 162 insertions(+), 229 deletions(-)
diff --git a/lib/librte_eal/bsdapp/eal/eal_pci.c
b/lib/librte_eal/bsdapp/eal/eal_pci.c
index 67ad88f..3778763 100644
--- a/lib/librte_eal/bsdapp/eal/eal_pci.c
+++ b/lib/librte_eal/bsdapp/eal/eal_pci.c
@@ -88,7 +88,7 @@ static struct rte_tailq_elem rte_uio_tailq = {
EAL_REGISTER_TAILQ(rte_uio_tailq)
/* unbind kernel driver for this device */
-static int
+int
pci_unbind_kernel_driver(struct rte_pci_device *dev __rte_unused)
{
RTE_LOG(ERR, EAL, "RTE_PCI_DRV_FORCE_UNBIND flag is not implemented "
@@ -429,6 +429,13 @@ skipdev:
return 0;
}
+/* Map pci device */
+int
+pci_map_device(struct rte_pci_device *dev)
+{
+ return pci_uio_map_resource(dev);
+}
+
/*
* Scan the content of the PCI bus, and add the devices in the devices
* list. Call pci_scan_one() for each pci entry found.
@@ -479,76 +486,6 @@ error:
return -1;
}
-/*
- * If vendor/device ID match, call the devinit() function of the
- * driver.
- */
-int
-rte_eal_pci_probe_one_driver(struct rte_pci_driver *dr, struct rte_pci_device
*dev)
-{
- struct rte_pci_id *id_table;
- int ret;
-
- for (id_table = dr->id_table ; id_table->vendor_id != 0; id_table++) {
-
- /* check if device's identifiers match the driver's ones */
- if (id_table->vendor_id != dev->id.vendor_id &&
- id_table->vendor_id != PCI_ANY_ID)
- continue;
- if (id_table->device_id != dev->id.device_id &&
- id_table->device_id != PCI_ANY_ID)
- continue;
- if (id_table->subsystem_vendor_id !=
- dev->id.subsystem_vendor_id &&
- id_table->subsystem_vendor_id != PCI_ANY_ID)
- continue;
- if (id_table->subsystem_device_id !=
- dev->id.subsystem_device_id &&
- id_table->subsystem_device_id != PCI_ANY_ID)
- continue;
-
- struct rte_pci_addr *loc = &dev->addr;
-
- RTE_LOG(DEBUG, EAL,
- "PCI device "PCI_PRI_FMT" on NUMA socket %i\n",
- loc->domain, loc->bus, loc->devid, loc->function,
- dev->numa_node);
-
- RTE_LOG(DEBUG, EAL,
- " probe driver: %x:%x %s\n", dev->id.vendor_id,
- dev->id.device_id, dr->name);
-
- /* no initialization when blacklisted, return without error */
- if (dev->devargs != NULL &&
- dev->devargs->type == RTE_DEVTYPE_BLACKLISTED_PCI) {
-
- RTE_LOG(DEBUG, EAL,
- " Device is blacklisted, not initializing\n");
- return 0;
- }
-
- if (dr->drv_flags & RTE_PCI_DRV_NEED_MAPPING) {
- /* map resources for devices that use igb_uio */
- ret = pci_uio_map_resource(dev);
- if (ret != 0)
- return ret;
- } else if (dr->drv_flags & RTE_PCI_DRV_FORCE_UNBIND &&
- rte_eal_process_type() == RTE_PROC_PRIMARY) {
- /* unbind current driver */
- if (pci_unbind_kernel_driver(dev) < 0)
- return -1;
- }
-
- /* reference driver structure */
- dev->driver = dr;
-
- /* call the driver devinit() function */
- return dr->devinit(dr, dev);
- }
- /* return positive value if driver is not found */
- return 1;
-}
-
/* Init the PCI EAL subsystem */
int
rte_eal_pci_init(void)
diff --git a/lib/librte_eal/common/eal_common_pci.c
b/lib/librte_eal/common/eal_common_pci.c
index 808b87b..538b2d9 100644
--- a/lib/librte_eal/common/eal_common_pci.c
+++ b/lib/librte_eal/common/eal_common_pci.c
@@ -99,6 +99,144 @@ static struct rte_devargs *pci_devargs_lookup(struct
rte_pci_device *dev)
}
/*
+ * If vendor/device ID match, call the devinit() function of the
+ * driver.
+ */
+static int
+rte_eal_pci_probe_one_driver(struct rte_pci_driver *dr,
+ struct rte_pci_device *dev)
+{
+ int ret;
+ struct rte_pci_id *id_table;
[dpdk-dev] [PATCH v2 12/15] eal: Consolidate pci_map/unmap_device() of linuxapp and bsdapp
The patch consolidates below functions, and implemented in common
eal code.
- pci_map_device()
- pci_unmap_device()
Signed-off-by: Tetsuya Mukawa
---
lib/librte_eal/bsdapp/eal/eal_pci.c | 12 +++
lib/librte_eal/common/eal_common_pci.c | 57 +
lib/librte_eal/common/eal_private.h | 4 +--
lib/librte_eal/common/include/rte_pci.h | 1 +
lib/librte_eal/linuxapp/eal/eal_pci.c | 55 ---
lib/librte_ether/rte_ethdev.c | 1 +
6 files changed, 65 insertions(+), 65 deletions(-)
diff --git a/lib/librte_eal/bsdapp/eal/eal_pci.c
b/lib/librte_eal/bsdapp/eal/eal_pci.c
index 3778763..1e42e42 100644
--- a/lib/librte_eal/bsdapp/eal/eal_pci.c
+++ b/lib/librte_eal/bsdapp/eal/eal_pci.c
@@ -280,7 +280,7 @@ fail:
}
/* map the PCI resource of a PCI device in virtual memory */
-static int
+int
pci_uio_map_resource(struct rte_pci_device *dev)
{
int i, map_idx, ret;
@@ -364,6 +364,9 @@ pci_scan_one(int dev_pci_fd, struct pci_conf *conf)
/* FreeBSD has no NUMA support (yet) */
dev->numa_node = 0;
+ /* FreeBSD has only one pass through driver */
+ dev->pt_driver = RTE_PT_NIC_UIO;
+
/* parse resources */
switch (conf->pc_hdr & PCIM_HDRTYPE) {
case PCIM_HDRTYPE_NORMAL:
@@ -429,13 +432,6 @@ skipdev:
return 0;
}
-/* Map pci device */
-int
-pci_map_device(struct rte_pci_device *dev)
-{
- return pci_uio_map_resource(dev);
-}
-
/*
* Scan the content of the PCI bus, and add the devices in the devices
* list. Call pci_scan_one() for each pci entry found.
diff --git a/lib/librte_eal/common/eal_common_pci.c
b/lib/librte_eal/common/eal_common_pci.c
index 538b2d9..726026f 100644
--- a/lib/librte_eal/common/eal_common_pci.c
+++ b/lib/librte_eal/common/eal_common_pci.c
@@ -98,6 +98,63 @@ static struct rte_devargs *pci_devargs_lookup(struct
rte_pci_device *dev)
return NULL;
}
+/* Map pci device */
+static int
+pci_map_device(struct rte_pci_device *dev)
+{
+ int ret = -1;
+
+ /* try mapping the NIC resources using VFIO if it exists */
+ switch (dev->pt_driver) {
+ case RTE_PT_VFIO:
+#ifdef VFIO_PRESENT
+ if (pci_vfio_is_enabled())
+ ret = pci_vfio_map_resource(dev);
+#endif
+ break;
+ case RTE_PT_IGB_UIO:
+ case RTE_PT_UIO_GENERIC:
+ case RTE_PT_NIC_UIO:
+ /* map resources for devices that use uio */
+ ret = pci_uio_map_resource(dev);
+ break;
+ default:
+ RTE_LOG(DEBUG, EAL, " Not managed by known pt driver,"
+ " skipped\n");
+ ret = 1;
+ break;
+ }
+
+ return ret;
+}
+
+#ifdef RTE_LIBRTE_EAL_HOTPLUG
+/* Unmap pci device */
+static void
+pci_unmap_device(struct rte_pci_device *dev)
+{
+ if (dev == NULL)
+ return;
+
+ /* try unmapping the NIC resources using VFIO if it exists */
+ switch (dev->pt_driver) {
+ case RTE_PT_VFIO:
+ RTE_LOG(ERR, EAL, "Hotplug doesn't support vfio yet\n");
+ break;
+ case RTE_PT_IGB_UIO:
+ case RTE_PT_UIO_GENERIC:
+ case RTE_PT_NIC_UIO:
+ /* unmap resources for devices that use uio */
+ pci_uio_unmap_resource(dev);
+ break;
+ default:
+ RTE_LOG(DEBUG, EAL, " Not managed by known pt driver,"
+ " skipped\n");
+ break;
+ }
+}
+#endif /* RTE_LIBRTE_EAL_HOTPLUG */
+
/*
* If vendor/device ID match, call the devinit() function of the
* driver.
diff --git a/lib/librte_eal/common/eal_private.h
b/lib/librte_eal/common/eal_private.h
index fe2c596..badb55c 100644
--- a/lib/librte_eal/common/eal_private.h
+++ b/lib/librte_eal/common/eal_private.h
@@ -171,14 +171,14 @@ int pci_unbind_kernel_driver(struct rte_pci_device *dev);
* @return
* 0 on success, negative on error
*/
-int pci_map_device(struct rte_pci_device *dev);
+int pci_uio_map_resource(struct rte_pci_device *dev);
/**
* Unmap this device
*
* This function is private to EAL.
*/
-void pci_unmap_device(struct rte_pci_device *dev);
+void pci_uio_unmap_resource(struct rte_pci_device *dev);
/**
* Init tail queues for non-EAL library structures. This is to allow
diff --git a/lib/librte_eal/common/include/rte_pci.h
b/lib/librte_eal/common/include/rte_pci.h
index fc64601..47345b8 100644
--- a/lib/librte_eal/common/include/rte_pci.h
+++ b/lib/librte_eal/common/include/rte_pci.h
@@ -147,6 +147,7 @@ enum rte_pt_driver {
RTE_PT_IGB_UIO = 1,
RTE_PT_VFIO = 2,
RTE_PT_UIO_GENERIC = 3,
+ RTE_PT_NIC_UIO = 4,
};
/**
diff --git a/lib/librte_eal/linuxapp/eal/eal_pci.c
b/lib/librte_eal/linuxapp/eal/eal_pci.c
index dbc2548..6a8153c 100644
--- a/lib/librte_eal/linuxapp/eal/eal_pci.c
+++ b/lib/librte_eal/linuxapp/eal/eal
[dpdk-dev] [PATCH v2 13/15] eal: Consolidate pci_map/unmap_resource() of linuxapp and bsdapp
The patch consolidates below functions, and implemented in common
eal code.
- pci_map_resource()
- pci_unmap_resource()
Signed-off-by: Tetsuya Mukawa
---
lib/librte_eal/bsdapp/eal/eal_pci.c| 24 -
lib/librte_eal/common/eal_common_pci.c | 43 ++
lib/librte_eal/common/include/rte_pci.h| 11
lib/librte_eal/linuxapp/eal/eal_pci.c | 42 -
lib/librte_eal/linuxapp/eal/eal_pci_init.h | 5
5 files changed, 54 insertions(+), 71 deletions(-)
diff --git a/lib/librte_eal/bsdapp/eal/eal_pci.c
b/lib/librte_eal/bsdapp/eal/eal_pci.c
index 1e42e42..d7d6439 100644
--- a/lib/librte_eal/bsdapp/eal/eal_pci.c
+++ b/lib/librte_eal/bsdapp/eal/eal_pci.c
@@ -46,7 +46,6 @@
#include
#include
#include
-#include
#include
#include
#include
@@ -96,29 +95,6 @@ pci_unbind_kernel_driver(struct rte_pci_device *dev
__rte_unused)
return -ENOTSUP;
}
-/* map a particular resource from a file */
-static void *
-pci_map_resource(void *requested_addr, int fd, off_t offset, size_t size,
-int additional_flags)
-{
- void *mapaddr;
-
- /* Map the PCI memory resource of device */
- mapaddr = mmap(requested_addr, size, PROT_READ | PROT_WRITE,
- MAP_SHARED | additional_flags, fd, offset);
- if (mapaddr == MAP_FAILED) {
- RTE_LOG(ERR, EAL,
- "%s(): cannot mmap(%d, %p, 0x%lx, 0x%lx): %s (%p)\n",
- __func__, fd, requested_addr,
- (unsigned long)size, (unsigned long)offset,
- strerror(errno), mapaddr);
- } else {
- RTE_LOG(DEBUG, EAL, " PCI memory mapped at %p\n", mapaddr);
- }
-
- return mapaddr;
-}
-
static int
pci_uio_map_secondary(struct rte_pci_device *dev)
{
diff --git a/lib/librte_eal/common/eal_common_pci.c
b/lib/librte_eal/common/eal_common_pci.c
index 726026f..6d98194 100644
--- a/lib/librte_eal/common/eal_common_pci.c
+++ b/lib/librte_eal/common/eal_common_pci.c
@@ -67,6 +67,7 @@
#include
#include
#include
+#include
#include
#include
@@ -98,6 +99,48 @@ static struct rte_devargs *pci_devargs_lookup(struct
rte_pci_device *dev)
return NULL;
}
+/* map a particular resource from a file */
+void *
+pci_map_resource(void *requested_addr, int fd, off_t offset, size_t size,
+int additional_flags)
+{
+ void *mapaddr;
+
+ /* Map the PCI memory resource of device */
+ mapaddr = mmap(requested_addr, size, PROT_READ | PROT_WRITE,
+ MAP_SHARED | additional_flags, fd, offset);
+ if (mapaddr == MAP_FAILED) {
+ RTE_LOG(ERR, EAL,
+ "%s(): cannot mmap(%d, %p, 0x%lx, 0x%lx): %s (%p)\n",
+ __func__, fd, requested_addr,
+ (unsigned long)size, (unsigned long)offset,
+ strerror(errno), mapaddr);
+ } else {
+ RTE_LOG(DEBUG, EAL, " PCI memory mapped at %p\n", mapaddr);
+ }
+
+ return mapaddr;
+}
+
+#ifdef RTE_LIBRTE_EAL_HOTPLUG
+/* unmap a particular resource */
+void
+pci_unmap_resource(void *requested_addr, size_t size)
+{
+ if (requested_addr == NULL)
+ return;
+
+ /* Unmap the PCI memory resource of device */
+ if (munmap(requested_addr, size)) {
+ RTE_LOG(ERR, EAL, "%s(): cannot munmap(%p, 0x%lx): %s\n",
+ __func__, requested_addr, (unsigned long)size,
+ strerror(errno));
+ } else
+ RTE_LOG(DEBUG, EAL, " PCI memory unmapped at %p\n",
+ requested_addr);
+}
+#endif /* RTE_LIBRTE_EAL_HOTPLUG */
+
/* Map pci device */
static int
pci_map_device(struct rte_pci_device *dev)
diff --git a/lib/librte_eal/common/include/rte_pci.h
b/lib/librte_eal/common/include/rte_pci.h
index 47345b8..d3b883e 100644
--- a/lib/librte_eal/common/include/rte_pci.h
+++ b/lib/librte_eal/common/include/rte_pci.h
@@ -363,8 +363,19 @@ int rte_eal_pci_scan(void);
*/
int rte_eal_pci_probe(void);
+/**
+ * Map pci resouce.
+ */
+void *pci_map_resource(void *requested_addr, int fd, off_t offset,
+ size_t size, int additional_flags);
+
#ifdef RTE_LIBRTE_EAL_HOTPLUG
/**
+ * Map pci resouce.
+ */
+void pci_unmap_resource(void *requested_addr, size_t size);
+
+/**
* Probe the single PCI device.
*
* Scan the content of the PCI bus, and find the pci device specified by pci
diff --git a/lib/librte_eal/linuxapp/eal/eal_pci.c
b/lib/librte_eal/linuxapp/eal/eal_pci.c
index 6a8153c..d072e98 100644
--- a/lib/librte_eal/linuxapp/eal/eal_pci.c
+++ b/lib/librte_eal/linuxapp/eal/eal_pci.c
@@ -33,7 +33,6 @@
#include
#include
-#include
#include
#include
@@ -142,47 +141,6 @@ pci_find_max_end_va(void)
return RTE_PTR_ADD(last->addr, last->len);
}
-
-/* map a particular resource from a file */
[dpdk-dev] [PATCH v2 14/15] eal: Consolidate pci uio functions of linuxapp and bsdapp
The patch consolidates below functions, and implement these
in eal_common_pci_uio.c.
- pci_uio_map_secondary()
- pci_uio_map_resource()
- pci_uio_unmap()
- pci_uio_find_resource()
- pci_uio_unmap_resource()
Signed-off-by: Tetsuya Mukawa
---
lib/librte_eal/bsdapp/eal/Makefile | 1 +
lib/librte_eal/bsdapp/eal/eal_pci.c| 111 +-
.../bsdapp/eal/include/exec-env/rte_interrupts.h | 1 +
lib/librte_eal/common/eal_common_pci_uio.c | 226 +
lib/librte_eal/common/eal_private.h| 36
lib/librte_eal/linuxapp/eal/Makefile | 1 +
lib/librte_eal/linuxapp/eal/eal_pci_init.h | 11 +-
lib/librte_eal/linuxapp/eal/eal_pci_uio.c | 184 +
8 files changed, 273 insertions(+), 298 deletions(-)
create mode 100644 lib/librte_eal/common/eal_common_pci_uio.c
diff --git a/lib/librte_eal/bsdapp/eal/Makefile
b/lib/librte_eal/bsdapp/eal/Makefile
index 2357cfa..1dec11e 100644
--- a/lib/librte_eal/bsdapp/eal/Makefile
+++ b/lib/librte_eal/bsdapp/eal/Makefile
@@ -68,6 +68,7 @@ SRCS-$(CONFIG_RTE_LIBRTE_EAL_BSDAPP) += eal_common_memzone.c
SRCS-$(CONFIG_RTE_LIBRTE_EAL_BSDAPP) += eal_common_log.c
SRCS-$(CONFIG_RTE_LIBRTE_EAL_BSDAPP) += eal_common_launch.c
SRCS-$(CONFIG_RTE_LIBRTE_EAL_BSDAPP) += eal_common_pci.c
+SRCS-$(CONFIG_RTE_LIBRTE_EAL_BSDAPP) += eal_common_pci_uio.c
SRCS-$(CONFIG_RTE_LIBRTE_EAL_BSDAPP) += eal_common_memory.c
SRCS-$(CONFIG_RTE_LIBRTE_EAL_BSDAPP) += eal_common_tailqs.c
SRCS-$(CONFIG_RTE_LIBRTE_EAL_BSDAPP) += eal_common_errno.c
diff --git a/lib/librte_eal/bsdapp/eal/eal_pci.c
b/lib/librte_eal/bsdapp/eal/eal_pci.c
index d7d6439..b6785d4 100644
--- a/lib/librte_eal/bsdapp/eal/eal_pci.c
+++ b/lib/librte_eal/bsdapp/eal/eal_pci.c
@@ -81,10 +81,6 @@
* network card, only providing access to PCI BAR to applications, and
* enabling bus master.
*/
-static struct rte_tailq_elem rte_uio_tailq = {
- .name = "UIO_RESOURCE_LIST",
-};
-EAL_REGISTER_TAILQ(rte_uio_tailq)
/* unbind kernel driver for this device */
int
@@ -95,62 +91,7 @@ pci_unbind_kernel_driver(struct rte_pci_device *dev
__rte_unused)
return -ENOTSUP;
}
-static int
-pci_uio_map_secondary(struct rte_pci_device *dev)
-{
- int i, fd;
- struct mapped_pci_resource *uio_res;
- struct mapped_pci_res_list *uio_res_list =
- RTE_TAILQ_CAST(rte_uio_tailq.head, mapped_pci_res_list);
-
- TAILQ_FOREACH(uio_res, uio_res_list, next) {
-
- /* skip this element if it doesn't match our PCI address */
- if (rte_eal_compare_pci_addr(&uio_res->pci_addr, &dev->addr))
- continue;
-
- for (i = 0; i != uio_res->nb_maps; i++) {
- /*
-* open devname, to mmap it
-*/
- fd = open(uio_res->maps[i].path, O_RDWR);
- if (fd < 0) {
- RTE_LOG(ERR, EAL, "Cannot open %s: %s\n",
- uio_res->maps[i].path, strerror(errno));
- return -1;
- }
-
- void *mapaddr = pci_map_resource(uio_res->maps[i].addr,
- fd, (off_t)uio_res->maps[i].offset,
- (size_t)uio_res->maps[i].size, 0);
- if (mapaddr != uio_res->maps[i].addr) {
- if (mapaddr == MAP_FAILED)
- RTE_LOG(ERR, EAL,
- "Cannot mmap device resource "
- "file %s: %s\n",
- uio_res->maps[i].path,
- strerror(errno));
- else
- RTE_LOG(ERR, EAL,
- "Cannot mmap device resource "
- "file %s to address: %p\n",
- uio_res->maps[i].path,
- uio_res->maps[i].addr);
-
- close(fd);
- return -1;
- }
- /* fd is not needed in slave process, close it */
- close(fd);
- }
- return 0;
- }
-
- RTE_LOG(ERR, EAL, "Cannot find resource for device\n");
- return 1;
-}
-
-static int
+int
pci_uio_alloc_uio_resource(struct rte_pci_device *dev,
struct mapped_pci_resource **uio_res)
{
@@ -197,7 +138,7 @@ pci_uio_alloc_uio_resource(struct rte_pci_device *dev,
return 0;
}
-static int
+int
pci_uio_map_uio_resource_by_index(struct rte_pci_device *dev, int res_idx,
st
[dpdk-dev] [PATCH v2 15/15] eal: Enable Port Hotplug as default in Linux and BSD
This patch removes CONFIG_RTE_LIBRTE_EAL_HOTPLUG option, and enables it
as default in both Linux and BSD.
Also, to support port hotplug, rte_eal_pci_scan() and below missing
symbols should be exported to ethdev library.
- rte_eal_parse_devargs_str()
- rte_eal_pci_close_one()
- rte_eal_pci_probe_one()
- rte_eal_pci_scan()
- rte_eal_vdev_init()
- rte_eal_vdev_uninit()
Signed-off-by: Tetsuya Mukawa
---
config/common_bsdapp | 6 --
config/common_linuxapp| 5 -
lib/librte_eal/bsdapp/eal/eal_pci.c | 6 +++---
lib/librte_eal/bsdapp/eal/rte_eal_version.map | 6 ++
lib/librte_eal/common/eal_common_dev.c| 2 --
lib/librte_eal/common/eal_common_pci.c| 8
lib/librte_eal/common/eal_common_pci_uio.c| 2 --
lib/librte_eal/common/include/rte_pci.h | 2 --
lib/librte_ether/rte_ethdev.c | 21 -
9 files changed, 9 insertions(+), 49 deletions(-)
diff --git a/config/common_bsdapp b/config/common_bsdapp
index 8ff4dc2..88b44e9 100644
--- a/config/common_bsdapp
+++ b/config/common_bsdapp
@@ -116,12 +116,6 @@ CONFIG_RTE_LIBRTE_EAL_BSDAPP=y
CONFIG_RTE_LIBRTE_EAL_LINUXAPP=n
#
-# Compile Environment Abstraction Layer to support hotplug
-# So far, Hotplug functions only support linux
-#
-CONFIG_RTE_LIBRTE_EAL_HOTPLUG=n
-
-#
# Compile Environment Abstraction Layer to support Vmware TSC map
#
CONFIG_RTE_LIBRTE_EAL_VMWARE_TSC_MAP_SUPPORT=y
diff --git a/config/common_linuxapp b/config/common_linuxapp
index 97f1c9e..f9c9780 100644
--- a/config/common_linuxapp
+++ b/config/common_linuxapp
@@ -114,11 +114,6 @@ CONFIG_RTE_PCI_MAX_READ_REQUEST_SIZE=0
CONFIG_RTE_LIBRTE_EAL_LINUXAPP=y
#
-# Compile Environment Abstraction Layer to support hotplug
-#
-CONFIG_RTE_LIBRTE_EAL_HOTPLUG=y
-
-#
# Compile Environment Abstraction Layer to support Vmware TSC map
#
CONFIG_RTE_LIBRTE_EAL_VMWARE_TSC_MAP_SUPPORT=y
diff --git a/lib/librte_eal/bsdapp/eal/eal_pci.c
b/lib/librte_eal/bsdapp/eal/eal_pci.c
index b6785d4..50c9544 100644
--- a/lib/librte_eal/bsdapp/eal/eal_pci.c
+++ b/lib/librte_eal/bsdapp/eal/eal_pci.c
@@ -305,8 +305,8 @@ skipdev:
* Scan the content of the PCI bus, and add the devices in the devices
* list. Call pci_scan_one() for each pci entry found.
*/
-static int
-pci_scan(void)
+int
+rte_eal_pci_scan(void)
{
int fd;
unsigned dev_count = 0;
@@ -362,7 +362,7 @@ rte_eal_pci_init(void)
if (internal_config.no_pci)
return 0;
- if (pci_scan() < 0) {
+ if (rte_eal_pci_scan() < 0) {
RTE_LOG(ERR, EAL, "%s(): Cannot scan PCI bus\n", __func__);
return -1;
}
diff --git a/lib/librte_eal/bsdapp/eal/rte_eal_version.map
b/lib/librte_eal/bsdapp/eal/rte_eal_version.map
index 67b6a6c..7e850a9 100644
--- a/lib/librte_eal/bsdapp/eal/rte_eal_version.map
+++ b/lib/librte_eal/bsdapp/eal/rte_eal_version.map
@@ -37,14 +37,20 @@ DPDK_2.0 {
rte_eal_lcore_role;
rte_eal_mp_remote_launch;
rte_eal_mp_wait_lcore;
+ rte_eal_parse_devargs_str;
+ rte_eal_pci_close_one;
rte_eal_pci_dump;
rte_eal_pci_probe;
+ rte_eal_pci_probe_one;
rte_eal_pci_register;
+ rte_eal_pci_scan;
rte_eal_pci_unregister;
rte_eal_process_type;
rte_eal_remote_launch;
rte_eal_tailq_lookup;
rte_eal_tailq_register;
+ rte_eal_vdev_init;
+ rte_eal_vdev_uninit;
rte_eal_wait_lcore;
rte_exit;
rte_get_hpet_cycles;
diff --git a/lib/librte_eal/common/eal_common_dev.c
b/lib/librte_eal/common/eal_common_dev.c
index 92a5a94..4089d66 100644
--- a/lib/librte_eal/common/eal_common_dev.c
+++ b/lib/librte_eal/common/eal_common_dev.c
@@ -125,7 +125,6 @@ rte_eal_dev_init(void)
return 0;
}
-#ifdef RTE_LIBRTE_EAL_HOTPLUG
int
rte_eal_vdev_uninit(const char *name)
{
@@ -151,4 +150,3 @@ rte_eal_vdev_uninit(const char *name)
RTE_LOG(ERR, EAL, "no driver found for %s\n", name);
return -EINVAL;
}
-#endif /* RTE_LIBRTE_EAL_HOTPLUG */
diff --git a/lib/librte_eal/common/eal_common_pci.c
b/lib/librte_eal/common/eal_common_pci.c
index 6d98194..126dd01 100644
--- a/lib/librte_eal/common/eal_common_pci.c
+++ b/lib/librte_eal/common/eal_common_pci.c
@@ -122,7 +122,6 @@ pci_map_resource(void *requested_addr, int fd, off_t
offset, size_t size,
return mapaddr;
}
-#ifdef RTE_LIBRTE_EAL_HOTPLUG
/* unmap a particular resource */
void
pci_unmap_resource(void *requested_addr, size_t size)
@@ -139,7 +138,6 @@ pci_unmap_resource(void *requested_addr, size_t size)
RTE_LOG(DEBUG, EAL, " PCI memory unmapped at %p\n",
requested_addr);
}
-#endif /* RTE_LIBRTE_EAL_HOTPLUG */
/* Map pci device */
static int
@@ -171,7 +169,6 @@ pci_map_device(struct rte_pci_device *dev)
return ret;
}
-#ifdef RTE_LIBRTE_EAL_HOTPLUG
/* Unmap pci device */
static
[dpdk-dev] [PATCH 1/3] BSD: Support Port Hotplug function
On 2015/03/09 11:23, Tetsuya Mukawa wrote: > On 2015/03/06 19:32, Iremonger, Bernard wrote: > >> Hi Tetsuya, >> >> The above functions being added to /lib/librte_eal/bsdapp/eal/eal_pci.c are >> very similar to the functions >> in lib/librte_eal/linuxapp/eal/eal_pci.c the main difference seems to be >> "struct uio_resource *uio_res" in bsd >> and "struct mapped_pci_resource *uio_res " in linux. >> Both of these structures are the same although named differently. >> There are at present only 3 references to "struct uio_resource" in >> librte_eal/bsdapp/eal/eal_pci.c. If "struct uio_resource" was renamed to" >> struct mapped_pci_resource " the functions being added to >> bsdapp/eal/eal_pci.c would be the same as the functions in >> linuxapp/eal/eal_pci.c. >> If both groups of functions were the same it would make sense to have only >> one copy of these functions which could be moved into >> librte_eal/common/eal_common_pci.c . >> >> There may be some extra work in doing this but I think it would make the >> code more maintainable in the future. >> What do you think? > Hi Bernard, > > Yes, I agree with it. > I will merge above, and submit again. > Thanks for suggestion. > > Regards, > Tetsuya > Hi Bernard, I've sent v2 patches. The patches depends on the latest master branch. So if you apply it, could you please pull first? Also, I tried to separate patches for reviewer, but if you feel it should be squashed more, please let me know. Thanks, Tetsuya
[dpdk-dev] [PATCH v7] af_packet: Fix some klocwork errors
On 3/11/2015 9:34 AM, Ouyang Changchun wrote:
> Fix possible memory leak issue: free kvlist before return;
> Fix possible resource lost issue: close qssockfd before return;
>
> Signed-off-by: Changchun Ouyang
> ---
> Change in v7:
> - Remove unnecessary '!= NULL' check before freeing it;
>
> Change in v6:
> - Refine exit point;
>
> Change in v5:
> - Initialize qsockfd with -1;
>
> Change in v4:
> - Check sockfd in internals->rx_queue against 0;
>
> Change in v3:
> - Also close sockets for all queues;
>
> Change in v2:
> - Make the error exit point a common path.
Acked-by: Michael Qiu
> lib/librte_pmd_af_packet/rte_eth_af_packet.c | 25 -
> 1 file changed, 16 insertions(+), 9 deletions(-)
>
> diff --git a/lib/librte_pmd_af_packet/rte_eth_af_packet.c
> b/lib/librte_pmd_af_packet/rte_eth_af_packet.c
> index 80e9bdf..2ac50ba 100644
> --- a/lib/librte_pmd_af_packet/rte_eth_af_packet.c
> +++ b/lib/librte_pmd_af_packet/rte_eth_af_packet.c
> @@ -442,7 +442,8 @@ rte_pmd_init_internals(const char *name,
> struct tpacket_req *req;
> struct pkt_rx_queue *rx_queue;
> struct pkt_tx_queue *tx_queue;
> - int rc, qsockfd, tpver, discard;
> + int rc, tpver, discard;
> + int qsockfd = -1;
> unsigned int i, q, rdsize;
> int fanout_arg __rte_unused, bypass __rte_unused;
>
> @@ -691,9 +692,14 @@ error:
> rte_free((*internals)->rx_queue[q].rd);
> if ((*internals)->tx_queue[q].rd)
> rte_free((*internals)->tx_queue[q].rd);
> + if (((*internals)->rx_queue[q].sockfd != 0) &&
> + ((*internals)->rx_queue[q].sockfd != qsockfd))
> + close((*internals)->rx_queue[q].sockfd);
> }
> rte_free(*internals);
> }
> + if (qsockfd != -1)
> + close(qsockfd);
> return -1;
> }
>
> @@ -802,7 +808,7 @@ int
> rte_pmd_af_packet_devinit(const char *name, const char *params)
> {
> unsigned numa_node;
> - int ret;
> + int ret = 0;
> struct rte_kvargs *kvlist;
> int sockfd = -1;
>
> @@ -811,8 +817,10 @@ rte_pmd_af_packet_devinit(const char *name, const char
> *params)
> numa_node = rte_socket_id();
>
> kvlist = rte_kvargs_parse(params, valid_arguments);
> - if (kvlist == NULL)
> - return -1;
> + if (kvlist == NULL) {
> + ret = -1;
> + goto exit;
> + }
>
> /*
>* If iface argument is passed we open the NICs and use them for
> @@ -823,16 +831,15 @@ rte_pmd_af_packet_devinit(const char *name, const char
> *params)
> ret = rte_kvargs_process(kvlist, ETH_AF_PACKET_IFACE_ARG,
>&open_packet_iface, &sockfd);
> if (ret < 0)
> - return -1;
> + goto exit;
> }
>
> ret = rte_eth_from_packet(name, &sockfd, numa_node, kvlist);
> close(sockfd); /* no longer needed */
>
> - if (ret < 0)
> - return -1;
> -
> - return 0;
> +exit:
> + rte_kvargs_free(kvlist);
> + return ret;
> }
>
> static struct rte_driver pmd_af_packet_drv = {
[dpdk-dev] [PATCH v2 01/15] eal: Fix cording style of eal_pci.c and eal_pci_uio.c
On Thu, Mar 12, 2015 at 07:17:40PM +0900, Tetsuya Mukawa wrote:
> This patch fixes cording style of below files in linuxapp and bsdapp.
> - eal_pci.c
> - eal_pci_uio.c
>
> Signed-off-by: Tetsuya Mukawa
Hi Tetsuya,
While there is some good cleanup here, I disagree with a number of the changes
made purely to the whitespace in the file. The style of using a double-indent
for line continuations is very widely used in DPDK code, much more so than the
style of lining things up with the previous line.
So ack to the changes removing unnecessary braces, and occasional splitting of
really long lines (though a few chars over 80 is ok). NAK to the whitespace
and indentation changes.
Regards,
/Bruce
> ---
> lib/librte_eal/bsdapp/eal/eal_pci.c | 67
> ++-
> lib/librte_eal/linuxapp/eal/eal_pci.c | 32 +--
> lib/librte_eal/linuxapp/eal/eal_pci_uio.c | 37 ++---
> 3 files changed, 80 insertions(+), 56 deletions(-)
>
> diff --git a/lib/librte_eal/bsdapp/eal/eal_pci.c
> b/lib/librte_eal/bsdapp/eal/eal_pci.c
> index fe3ef86..cbd0a4e 100644
> --- a/lib/librte_eal/bsdapp/eal/eal_pci.c
> +++ b/lib/librte_eal/bsdapp/eal/eal_pci.c
> @@ -142,8 +142,9 @@ pci_map_resource(void *requested_addr, const char
> *devname, off_t offset,
> MAP_SHARED, fd, offset);
> close(fd);
> if (mapaddr == MAP_FAILED ||
> - (requested_addr != NULL && mapaddr != requested_addr)) {
> - RTE_LOG(ERR, EAL, "%s(): cannot mmap(%s(%d), %p, 0x%lx, 0x%lx):"
> + (requested_addr != NULL && mapaddr != requested_addr)) {
> + RTE_LOG(ERR, EAL,
> + "%s(): cannot mmap(%s(%d), %p, 0x%lx, 0x%lx):"
> " %s (%p)\n", __func__, devname, fd, requested_addr,
> (unsigned long)size, (unsigned long)offset,
> strerror(errno), mapaddr);
> @@ -161,9 +162,10 @@ fail:
> static int
> pci_uio_map_secondary(struct rte_pci_device *dev)
> {
> -size_t i;
> -struct uio_resource *uio_res;
> - struct uio_res_list *uio_res_list = RTE_TAILQ_CAST(rte_uio_tailq.head,
> uio_res_list);
> + size_t i;
> + struct uio_resource *uio_res;
> + struct uio_res_list *uio_res_list =
> + RTE_TAILQ_CAST(rte_uio_tailq.head, uio_res_list);
>
> TAILQ_FOREACH(uio_res, uio_res_list, next) {
>
> @@ -179,10 +181,10 @@ pci_uio_map_secondary(struct rte_pci_device *dev)
> != uio_res->maps[i].addr) {
> RTE_LOG(ERR, EAL,
> "Cannot mmap device resource\n");
> - return (-1);
> + return -1;
> }
> }
> - return (0);
> + return 0;
> }
>
> RTE_LOG(ERR, EAL, "Cannot find resource for device\n");
> @@ -201,7 +203,8 @@ pci_uio_map_resource(struct rte_pci_device *dev)
> uint64_t pagesz;
> struct rte_pci_addr *loc = &dev->addr;
> struct uio_resource *uio_res;
> - struct uio_res_list *uio_res_list = RTE_TAILQ_CAST(rte_uio_tailq.head,
> uio_res_list);
> + struct uio_res_list *uio_res_list =
> + RTE_TAILQ_CAST(rte_uio_tailq.head, uio_res_list);
> struct uio_map *maps;
>
> dev->intr_handle.fd = -1;
> @@ -209,14 +212,16 @@ pci_uio_map_resource(struct rte_pci_device *dev)
>
> /* secondary processes - use already recorded details */
> if (rte_eal_process_type() != RTE_PROC_PRIMARY)
> - return (pci_uio_map_secondary(dev));
> + return pci_uio_map_secondary(dev);
>
> snprintf(devname, sizeof(devname), "/dev/uio at pci:%u:%u:%u",
> dev->addr.bus, dev->addr.devid, dev->addr.function);
>
> if (access(devname, O_RDWR) < 0) {
> - RTE_LOG(WARNING, EAL, " "PCI_PRI_FMT" not managed by UIO
> driver, "
> - "skipping\n", loc->domain, loc->bus,
> loc->devid, loc->function);
> + RTE_LOG(WARNING, EAL,
> + " "PCI_PRI_FMT" not managed by UIO driver, "
> + "skipping\n", loc->domain, loc->bus,
> + loc->devid, loc->function);
> return 1;
> }
>
> @@ -233,7 +238,7 @@ pci_uio_map_resource(struct rte_pci_device *dev)
> if ((uio_res = rte_zmalloc("UIO_RES", sizeof (*uio_res), 0)) == NULL) {
> RTE_LOG(ERR, EAL,
> "%s(): cannot store uio mmap details\n", __func__);
> - return (-1);
> + return -1;
> }
>
> snprintf(uio_res->path, sizeof(uio_res->path), "%s", devname);
> @@ -248,7 +253,8 @@ pci_uio_map_resource(struct rte_pci_device *dev)
>
> j = uio_res->nb_maps;
> /* skip empty BAR */
> - if ((phaddr = dev->mem_resource[i].phys_addr) == 0)
> + ph
[dpdk-dev] [PATCH v2 01/15] eal: Fix cording style of eal_pci.c and eal_pci_uio.c
On 3/12/2015 6:50 PM, Bruce Richardson wrote:
> On Thu, Mar 12, 2015 at 07:17:40PM +0900, Tetsuya Mukawa wrote:
>> This patch fixes cording style of below files in linuxapp and bsdapp.
>> - eal_pci.c
>> - eal_pci_uio.c
>>
>> Signed-off-by: Tetsuya Mukawa
> Hi Tetsuya,
>
> While there is some good cleanup here, I disagree with a number of the changes
> made purely to the whitespace in the file. The style of using a double-indent
> for line continuations is very widely used in DPDK code, much more so than the
> style of lining things up with the previous line.
Yes, but both style are seeing in dpdk, here the patch is using Tab +
whitespace, which is also
the linux kernel's style.
So is there any rule to allow only one style?
Mixed style is bad...
Thanks,
Michael
> So ack to the changes removing unnecessary braces, and occasional splitting of
> really long lines (though a few chars over 80 is ok). NAK to the whitespace
> and indentation changes.
>
> Regards,
> /Bruce
>
>> ---
>> lib/librte_eal/bsdapp/eal/eal_pci.c | 67
>> ++-
>> lib/librte_eal/linuxapp/eal/eal_pci.c | 32 +--
>> lib/librte_eal/linuxapp/eal/eal_pci_uio.c | 37 ++---
>> 3 files changed, 80 insertions(+), 56 deletions(-)
>>
>> diff --git a/lib/librte_eal/bsdapp/eal/eal_pci.c
>> b/lib/librte_eal/bsdapp/eal/eal_pci.c
>> index fe3ef86..cbd0a4e 100644
>> --- a/lib/librte_eal/bsdapp/eal/eal_pci.c
>> +++ b/lib/librte_eal/bsdapp/eal/eal_pci.c
>> @@ -142,8 +142,9 @@ pci_map_resource(void *requested_addr, const char
>> *devname, off_t offset,
>> MAP_SHARED, fd, offset);
>> close(fd);
>> if (mapaddr == MAP_FAILED ||
>> -(requested_addr != NULL && mapaddr != requested_addr)) {
>> -RTE_LOG(ERR, EAL, "%s(): cannot mmap(%s(%d), %p, 0x%lx, 0x%lx):"
>> +(requested_addr != NULL && mapaddr != requested_addr)) {
>> +RTE_LOG(ERR, EAL,
>> +"%s(): cannot mmap(%s(%d), %p, 0x%lx, 0x%lx):"
>> " %s (%p)\n", __func__, devname, fd, requested_addr,
>> (unsigned long)size, (unsigned long)offset,
>> strerror(errno), mapaddr);
>> @@ -161,9 +162,10 @@ fail:
>> static int
>> pci_uio_map_secondary(struct rte_pci_device *dev)
>> {
>> -size_t i;
>> -struct uio_resource *uio_res;
>> -struct uio_res_list *uio_res_list = RTE_TAILQ_CAST(rte_uio_tailq.head,
>> uio_res_list);
>> +size_t i;
>> +struct uio_resource *uio_res;
>> +struct uio_res_list *uio_res_list =
>> +RTE_TAILQ_CAST(rte_uio_tailq.head, uio_res_list);
>>
>> TAILQ_FOREACH(uio_res, uio_res_list, next) {
>>
>> @@ -179,10 +181,10 @@ pci_uio_map_secondary(struct rte_pci_device *dev)
>> != uio_res->maps[i].addr) {
>> RTE_LOG(ERR, EAL,
>> "Cannot mmap device resource\n");
>> -return (-1);
>> +return -1;
>> }
>> }
>> -return (0);
>> +return 0;
>> }
>>
>> RTE_LOG(ERR, EAL, "Cannot find resource for device\n");
>> @@ -201,7 +203,8 @@ pci_uio_map_resource(struct rte_pci_device *dev)
>> uint64_t pagesz;
>> struct rte_pci_addr *loc = &dev->addr;
>> struct uio_resource *uio_res;
>> -struct uio_res_list *uio_res_list = RTE_TAILQ_CAST(rte_uio_tailq.head,
>> uio_res_list);
>> +struct uio_res_list *uio_res_list =
>> +RTE_TAILQ_CAST(rte_uio_tailq.head, uio_res_list);
>> struct uio_map *maps;
>>
>> dev->intr_handle.fd = -1;
>> @@ -209,14 +212,16 @@ pci_uio_map_resource(struct rte_pci_device *dev)
>>
>> /* secondary processes - use already recorded details */
>> if (rte_eal_process_type() != RTE_PROC_PRIMARY)
>> -return (pci_uio_map_secondary(dev));
>> +return pci_uio_map_secondary(dev);
>>
>> snprintf(devname, sizeof(devname), "/dev/uio at pci:%u:%u:%u",
>> dev->addr.bus, dev->addr.devid, dev->addr.function);
>>
>> if (access(devname, O_RDWR) < 0) {
>> -RTE_LOG(WARNING, EAL, " "PCI_PRI_FMT" not managed by UIO
>> driver, "
>> -"skipping\n", loc->domain, loc->bus,
>> loc->devid, loc->function);
>> +RTE_LOG(WARNING, EAL,
>> +" "PCI_PRI_FMT" not managed by UIO driver, "
>> +"skipping\n", loc->domain, loc->bus,
>> +loc->devid, loc->function);
>> return 1;
>> }
>>
>> @@ -233,7 +238,7 @@ pci_uio_map_resource(struct rte_pci_device *dev)
>> if ((uio_res = rte_zmalloc("UIO_RES", sizeof (*uio_res), 0)) == NULL) {
>> RTE_LOG(ERR, EAL,
>> "%s(): cannot store uio mmap details\n", __func__);
>> -return (-1);
>> +return -1;
>>
[dpdk-dev] [PATCH] doc: remove copyright from base of page
Removing Intel copyright from base of page for each document. Signed-off-by: Siobhan Butler --- doc/guides/conf.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/doc/guides/conf.py b/doc/guides/conf.py index 385af03..fba34a6 100644 --- a/doc/guides/conf.py +++ b/doc/guides/conf.py @@ -32,7 +32,7 @@ import subprocess project = 'DPDK' -copyright = '2014, Intel' +html_show_copyright = False version = subprocess.check_output(["make","-sRrC","../../", "showversion"]) -- 1.8.3.1
[dpdk-dev] [PATCH v2 07/15] eal: Fix interface of pci_map_resource() of bsdapp
On Thu, Mar 12, 2015 at 07:17:46PM +0900, Tetsuya Mukawa wrote:
> The function is implemented in both linuxapp and bsdapp, but interface
> is different. The patch fixes the function of bsdapp to do same as
> linuxapp. After applying it, file descriptor should be opened and
> closed out of pci_map_resource().
>
> Signed-off-by: Tetsuya Mukawa
> ---
> lib/librte_eal/bsdapp/eal/eal_pci.c | 117
> +++---
> lib/librte_eal/linuxapp/eal/eal_pci_uio.c | 5 +-
> 2 files changed, 78 insertions(+), 44 deletions(-)
>
> diff --git a/lib/librte_eal/bsdapp/eal/eal_pci.c
> b/lib/librte_eal/bsdapp/eal/eal_pci.c
> index bd0f8cd..c92c624 100644
> --- a/lib/librte_eal/bsdapp/eal/eal_pci.c
> +++ b/lib/librte_eal/bsdapp/eal/eal_pci.c
..
> - "%s(): cannot mmap(%s(%d), %p, 0x%lx, 0x%lx):"
> - " %s (%p)\n", __func__, devname, fd, requested_addr,
> + "%s(): cannot mmap(%d, %p, 0x%lx, 0x%lx): %s (%p)\n",
> + __func__, fd, requested_addr,
> (unsigned long)size, (unsigned long)offset,
> strerror(errno), mapaddr);
> - goto fail;
> + } else {
> + RTE_LOG(DEBUG, EAL, " PCI memory mapped at %p\n", mapaddr);
> }
Coding style for DPDK is not to put braces around single-line blocks.
/Bruce
[dpdk-dev] [PATCH v2 01/15] eal: Fix cording style of eal_pci.c and eal_pci_uio.c
On Thu, Mar 12, 2015 at 10:57:18AM +, Qiu, Michael wrote:
> On 3/12/2015 6:50 PM, Bruce Richardson wrote:
> > On Thu, Mar 12, 2015 at 07:17:40PM +0900, Tetsuya Mukawa wrote:
> >> This patch fixes cording style of below files in linuxapp and bsdapp.
> >> - eal_pci.c
> >> - eal_pci_uio.c
> >>
> >> Signed-off-by: Tetsuya Mukawa
> > Hi Tetsuya,
> >
> > While there is some good cleanup here, I disagree with a number of the
> > changes
> > made purely to the whitespace in the file. The style of using a
> > double-indent
> > for line continuations is very widely used in DPDK code, much more so than
> > the
> > style of lining things up with the previous line.
>
> Yes, but both style are seeing in dpdk, here the patch is using Tab +
> whitespace, which is also
> the linux kernel's style.
>
> So is there any rule to allow only one style?
> Mixed style is bad...
>
> Thanks,
> Michael
No there is no hard rule, that I am aware of. While I prefer the double-indent
myself, that is beside the point that in the absense of a hard rule to be
applied
globally, fixing whitespace from style A to style B just increases the size
of the diff which makes it hard to see the real code changes. Even with a slight
mixing of the styles the code is readable enough as-is.
/Bruce
> > So ack to the changes removing unnecessary braces, and occasional splitting
> > of
> > really long lines (though a few chars over 80 is ok). NAK to the whitespace
> > and indentation changes.
> >
> > Regards,
> > /Bruce
> >
> >> ---
> >> lib/librte_eal/bsdapp/eal/eal_pci.c | 67
> >> ++-
> >> lib/librte_eal/linuxapp/eal/eal_pci.c | 32 +--
> >> lib/librte_eal/linuxapp/eal/eal_pci_uio.c | 37 ++---
> >> 3 files changed, 80 insertions(+), 56 deletions(-)
> >>
> >> diff --git a/lib/librte_eal/bsdapp/eal/eal_pci.c
> >> b/lib/librte_eal/bsdapp/eal/eal_pci.c
> >> index fe3ef86..cbd0a4e 100644
> >> --- a/lib/librte_eal/bsdapp/eal/eal_pci.c
> >> +++ b/lib/librte_eal/bsdapp/eal/eal_pci.c
> >> @@ -142,8 +142,9 @@ pci_map_resource(void *requested_addr, const char
> >> *devname, off_t offset,
> >>MAP_SHARED, fd, offset);
> >>close(fd);
> >>if (mapaddr == MAP_FAILED ||
> >> - (requested_addr != NULL && mapaddr != requested_addr)) {
> >> - RTE_LOG(ERR, EAL, "%s(): cannot mmap(%s(%d), %p, 0x%lx, 0x%lx):"
> >> + (requested_addr != NULL && mapaddr != requested_addr)) {
> >> + RTE_LOG(ERR, EAL,
> >> + "%s(): cannot mmap(%s(%d), %p, 0x%lx, 0x%lx):"
> >>" %s (%p)\n", __func__, devname, fd, requested_addr,
> >>(unsigned long)size, (unsigned long)offset,
> >>strerror(errno), mapaddr);
> >> @@ -161,9 +162,10 @@ fail:
> >> static int
> >> pci_uio_map_secondary(struct rte_pci_device *dev)
> >> {
> >> -size_t i;
> >> -struct uio_resource *uio_res;
> >> - struct uio_res_list *uio_res_list = RTE_TAILQ_CAST(rte_uio_tailq.head,
> >> uio_res_list);
> >> + size_t i;
> >> + struct uio_resource *uio_res;
> >> + struct uio_res_list *uio_res_list =
> >> + RTE_TAILQ_CAST(rte_uio_tailq.head, uio_res_list);
> >>
> >>TAILQ_FOREACH(uio_res, uio_res_list, next) {
> >>
> >> @@ -179,10 +181,10 @@ pci_uio_map_secondary(struct rte_pci_device *dev)
> >>!= uio_res->maps[i].addr) {
> >>RTE_LOG(ERR, EAL,
> >>"Cannot mmap device resource\n");
> >> - return (-1);
> >> + return -1;
> >>}
> >>}
> >> - return (0);
> >> + return 0;
> >>}
> >>
> >>RTE_LOG(ERR, EAL, "Cannot find resource for device\n");
> >> @@ -201,7 +203,8 @@ pci_uio_map_resource(struct rte_pci_device *dev)
> >>uint64_t pagesz;
> >>struct rte_pci_addr *loc = &dev->addr;
> >>struct uio_resource *uio_res;
> >> - struct uio_res_list *uio_res_list = RTE_TAILQ_CAST(rte_uio_tailq.head,
> >> uio_res_list);
> >> + struct uio_res_list *uio_res_list =
> >> + RTE_TAILQ_CAST(rte_uio_tailq.head, uio_res_list);
> >>struct uio_map *maps;
> >>
> >>dev->intr_handle.fd = -1;
> >> @@ -209,14 +212,16 @@ pci_uio_map_resource(struct rte_pci_device *dev)
> >>
> >>/* secondary processes - use already recorded details */
> >>if (rte_eal_process_type() != RTE_PROC_PRIMARY)
> >> - return (pci_uio_map_secondary(dev));
> >> + return pci_uio_map_secondary(dev);
> >>
> >>snprintf(devname, sizeof(devname), "/dev/uio at pci:%u:%u:%u",
> >>dev->addr.bus, dev->addr.devid, dev->addr.function);
> >>
> >>if (access(devname, O_RDWR) < 0) {
> >> - RTE_LOG(WARNING, EAL, " "PCI_PRI_FMT" not managed by UIO
> >> driver, "
> >> - "skipping\n", loc->domain, loc->bus,
> >> loc->devid, loc->
[dpdk-dev] [PATCH] doc: update release notes description for new sample apps Updated release notes release description notes: - added new sample applications to list: Link Bonding,Skeleton, Callbacks,
Signed-off-by: Siobhan Butler --- doc/guides/rel_notes/rel_description.rst | 16 +--- 1 file changed, 13 insertions(+), 3 deletions(-) diff --git a/doc/guides/rel_notes/rel_description.rst b/doc/guides/rel_notes/rel_description.rst index 49f32b9..2fb5379 100644 --- a/doc/guides/rel_notes/rel_description.rst +++ b/doc/guides/rel_notes/rel_description.rst @@ -1,5 +1,5 @@ .. BSD LICENSE -Copyright(c) 2010-2014 Intel Corporation. All rights reserved. +Copyright(c) 2010-2015 Intel Corporation. All rights reserved. All rights reserved. Redistribution and use in source and binary forms, with or without @@ -32,7 +32,7 @@ Description of Release == These release notes cover the new features, -fixed bugs and known issues for Data Plane Development Kit (DPDK) release version 1.7.0. +fixed bugs and known issues for Data Plane Development Kit (DPDK) release version 2.0. For instructions on compiling and running the release, see the *DPDK Getting Started Guide*. @@ -103,6 +103,8 @@ The following is a list of DPDK documents in the suggested reading order: * IP Reassembly +* IP Pipeline + * IP Fragmentation * IPv4 Multicast @@ -111,6 +113,8 @@ The following is a list of DPDK documents in the suggested reading order: * L2 Forwarding IVSHMEM +* L2 Forwarding Jobstats + * L3 Forwarding * L3 Forwarding with Access Control @@ -119,6 +123,8 @@ The following is a list of DPDK documents in the suggested reading order: * L3 Forwarding in a Virtualized Environment +* Link Bonding + * Link Status Interrupt * Load Balancing @@ -149,6 +155,10 @@ The following is a list of DPDK documents in the suggested reading order: * Distributor +* RX-TX Callbacks + +* Skeleton + In addition, there are some other applications that are built when the libraries are created. The source for these applications is in the DPDK/app directory and are called: @@ -163,6 +173,6 @@ The following is a list of DPDK documents in the suggested reading order: * The testpmd application provides a number of different packet throughput tests and examples of features such as how to use the Flow Director found in the Intel? 82599 10 Gigabit Ethernet Controller. -The testpmd application is documented in the *DPDK Testpmd Application Note* (525362). +The testpmd application is documented in the *DPDK Testpmd Application Note*. The test application is not currently documented. However, you should be able to run and use test application with the command line help that is provided in the application. -- 1.8.3.1
[dpdk-dev] [PATCH] doc: update release notes description for new sample apps Updated release notes release description notes: - added new sample applications to list: Link Bonding, Skeleton, Callbacks
Self-Nack - title mistake > -Original Message- > From: Butler, Siobhan A > Sent: Thursday, March 12, 2015 11:36 AM > To: dev at dpdk.org > Cc: Butler, Siobhan A > Subject: [PATCH] doc: update release notes description for new sample apps > Updated release notes release description notes: - added new sample > applications to list: Link Bonding,Skeleton, Callbacks, Jobstats - updated > copyright date to 2015 - updated release nu... > > Signed-off-by: Siobhan Butler > --- > doc/guides/rel_notes/rel_description.rst | 16 +--- > 1 file changed, 13 insertions(+), 3 deletions(-) > > diff --git a/doc/guides/rel_notes/rel_description.rst > b/doc/guides/rel_notes/rel_description.rst > index 49f32b9..2fb5379 100644 > --- a/doc/guides/rel_notes/rel_description.rst > +++ b/doc/guides/rel_notes/rel_description.rst > @@ -1,5 +1,5 @@ > .. BSD LICENSE > -Copyright(c) 2010-2014 Intel Corporation. All rights reserved. > +Copyright(c) 2010-2015 Intel Corporation. All rights reserved. > All rights reserved. > > Redistribution and use in source and binary forms, with or without @@ - > 32,7 +32,7 @@ Description of Release == > > These release notes cover the new features, -fixed bugs and known issues > for Data Plane Development Kit (DPDK) release version 1.7.0. > +fixed bugs and known issues for Data Plane Development Kit (DPDK) > release version 2.0. > > For instructions on compiling and running the release, see the *DPDK > Getting Started Guide*. > > @@ -103,6 +103,8 @@ The following is a list of DPDK documents in the > suggested reading order: > > * IP Reassembly > > +* IP Pipeline > + > * IP Fragmentation > > * IPv4 Multicast > @@ -111,6 +113,8 @@ The following is a list of DPDK documents in the > suggested reading order: > > * L2 Forwarding IVSHMEM > > +* L2 Forwarding Jobstats > + > * L3 Forwarding > > * L3 Forwarding with Access Control > @@ -119,6 +123,8 @@ The following is a list of DPDK documents in the > suggested reading order: > > * L3 Forwarding in a Virtualized Environment > > +* Link Bonding > + > * Link Status Interrupt > > * Load Balancing > @@ -149,6 +155,10 @@ The following is a list of DPDK documents in the > suggested reading order: > > * Distributor > > +* RX-TX Callbacks > + > +* Skeleton > + > In addition, there are some other applications that are built when the > libraries are created. > The source for these applications is in the DPDK/app directory and are > called: > > @@ -163,6 +173,6 @@ The following is a list of DPDK documents in the > suggested reading order: > * The testpmd application provides a number of different packet > throughput tests and examples of features such as > how to use the Flow Director found in the Intel? 82599 10 Gigabit > Ethernet Controller. > > -The testpmd application is documented in the *DPDK Testpmd Application > Note* (525362). > +The testpmd application is documented in the *DPDK Testpmd > Application Note*. > The test application is not currently documented. > However, you should be able to run and use test application with the > command line help that is provided in the application. > -- > 1.8.3.1
[dpdk-dev] [PATCH] doc: remove copyright from base of page
> -Original Message- > From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Siobhan Butler > Sent: Thursday, March 12, 2015 10:59 AM > To: dev at dpdk.org > Subject: [dpdk-dev] [PATCH] doc: remove copyright from base of page > > Removing Intel copyright from base of page for each document. > > Signed-off-by: Siobhan Butler Acked-by: Pablo de Lara
[dpdk-dev] [PATCH v2] doc: update release notes description for new sample apps
Updated release notes release description notes: - added new sample applications to list: Link Bonding,Skeleton, Callbacks, Jobstats - updated copyright date to 2015 - updated release number form 1.7.0 to 2.0 Signed-off-by: Siobhan Butler --- doc/guides/rel_notes/rel_description.rst | 16 +--- 1 file changed, 13 insertions(+), 3 deletions(-) diff --git a/doc/guides/rel_notes/rel_description.rst b/doc/guides/rel_notes/rel_description.rst index 49f32b9..2fb5379 100644 --- a/doc/guides/rel_notes/rel_description.rst +++ b/doc/guides/rel_notes/rel_description.rst @@ -1,5 +1,5 @@ .. BSD LICENSE -Copyright(c) 2010-2014 Intel Corporation. All rights reserved. +Copyright(c) 2010-2015 Intel Corporation. All rights reserved. All rights reserved. Redistribution and use in source and binary forms, with or without @@ -32,7 +32,7 @@ Description of Release == These release notes cover the new features, -fixed bugs and known issues for Data Plane Development Kit (DPDK) release version 1.7.0. +fixed bugs and known issues for Data Plane Development Kit (DPDK) release version 2.0. For instructions on compiling and running the release, see the *DPDK Getting Started Guide*. @@ -103,6 +103,8 @@ The following is a list of DPDK documents in the suggested reading order: * IP Reassembly +* IP Pipeline + * IP Fragmentation * IPv4 Multicast @@ -111,6 +113,8 @@ The following is a list of DPDK documents in the suggested reading order: * L2 Forwarding IVSHMEM +* L2 Forwarding Jobstats + * L3 Forwarding * L3 Forwarding with Access Control @@ -119,6 +123,8 @@ The following is a list of DPDK documents in the suggested reading order: * L3 Forwarding in a Virtualized Environment +* Link Bonding + * Link Status Interrupt * Load Balancing @@ -149,6 +155,10 @@ The following is a list of DPDK documents in the suggested reading order: * Distributor +* RX-TX Callbacks + +* Skeleton + In addition, there are some other applications that are built when the libraries are created. The source for these applications is in the DPDK/app directory and are called: @@ -163,6 +173,6 @@ The following is a list of DPDK documents in the suggested reading order: * The testpmd application provides a number of different packet throughput tests and examples of features such as how to use the Flow Director found in the Intel? 82599 10 Gigabit Ethernet Controller. -The testpmd application is documented in the *DPDK Testpmd Application Note* (525362). +The testpmd application is documented in the *DPDK Testpmd Application Note*. The test application is not currently documented. However, you should be able to run and use test application with the command line help that is provided in the application. -- 1.8.3.1
[dpdk-dev] [PATCH v4] pci: save list of detached devices, and re-probe during driver unload
Thank you.
Can you tell when will the patch be pushed to the source code?
-Original Message-
From: Bruce Richardson [mailto:bruce.richard...@intel.com]
Sent: 11 March 2015 12:26
To: Raz Amir
Cc: dev at dpdk.org
Subject: Re: [dpdk-dev] [PATCH v4] pci: save list of detached devices, and
re-probe during driver unload
On Thu, Mar 05, 2015 at 06:24:23PM +0200, Raz Amir wrote:
> Added code that saves the pointers to the detached devices, during
> driver loading, and during driver unloading, go over the list, and
> re-attach them by calling device_probe_and_attach on each device.
>
> Signed-off-by: Raz Amir
Acked-by: Bruce Richardson
> ---
> lib/librte_eal/bsdapp/nic_uio/nic_uio.c | 34
> -
> 1 file changed, 29 insertions(+), 5 deletions(-)
>
> diff --git a/lib/librte_eal/bsdapp/nic_uio/nic_uio.c
> b/lib/librte_eal/bsdapp/nic_uio/nic_uio.c
> index 5ae8560..7a4ed5d 100644
> --- a/lib/librte_eal/bsdapp/nic_uio/nic_uio.c
> +++ b/lib/librte_eal/bsdapp/nic_uio/nic_uio.c
> @@ -55,6 +55,9 @@ __FBSDID("$FreeBSD$");
>
> #define MAX_BARS (PCIR_MAX_BAR_0 + 1)
>
> +#define MAX_DETACHED_DEVICES 128
> +static device_t detached_devices[MAX_DETACHED_DEVICES] = {}; static
> +int num_detached = 0;
>
> struct nic_uio_softc {
> device_tdev_t;
> @@ -288,17 +291,38 @@ nic_uio_load(void)
> function = strtol(token, NULL, 10);
>
> dev = pci_find_bsf(bus, device, function);
> - if (dev != NULL)
> - for (i = 0; i < NUM_DEVICES; i++)
> - if (pci_get_vendor(dev) == devices[i].vend
&&
> - pci_get_device(dev) ==
devices[i].dev)
> - device_detach(dev);
> + if (dev == NULL)
> + continue;
> +
> + for (i = 0; i < NUM_DEVICES; i++)
> + if (pci_get_vendor(dev) == devices[i].vend &&
> + pci_get_device(dev) ==
devices[i].dev) {
> + if (num_detached <
MAX_DETACHED_DEVICES) {
> +
printf("nic_uio_load: detaching and storing dev=%p\n", dev);
> +
detached_devices[num_detached++] = dev;
> + }
> + else
> +
printf("nic_uio_load: reached MAX_DETACHED_DEVICES=%d. dev=%p won't be
reattached\n",
> +
MAX_DETACHED_DEVICES, dev);
> + device_detach(dev);
> + }
> }
> }
>
> static void
> nic_uio_unload(void)
> {
> + int i;
> + printf("nic_uio_unload: entered ... \n");
> +
> + for (i = 0; i < num_detached; i++) {
> + printf("nic_uio_unload: calling to device_probe_and_attach
for dev=%p...\n",
> + detached_devices[i]);
> + device_probe_and_attach(detached_devices[i]);
> + printf("nic_uio_unload: done.\n");
> + }
> +
> + printf("nic_uio_unload: leaving ... \n");
> }
>
> static int
> --
> 2.1.2
>
[dpdk-dev] [PATCH v7] af_packet: Fix some klocwork errors
On Thu, Mar 12, 2015 at 10:43:24AM +, Qiu, Michael wrote: > On 3/11/2015 9:34 AM, Ouyang Changchun wrote: > > Fix possible memory leak issue: free kvlist before return; > > Fix possible resource lost issue: close qssockfd before return; > > > > Signed-off-by: Changchun Ouyang > > --- > > Change in v7: > > - Remove unnecessary '!= NULL' check before freeing it; > > > > Change in v6: > > - Refine exit point; > > > > Change in v5: > > - Initialize qsockfd with -1; > > > > Change in v4: > > - Check sockfd in internals->rx_queue against 0; > > > > Change in v3: > > - Also close sockets for all queues; > > > > Change in v2: > > - Make the error exit point a common path. > > Acked-by: Michael Qiu Acked-by: Neil Horman
[dpdk-dev] Poll-mode driver in DPDK related to code
Hi Team , I am using DPDK-1.7.1 and trying to understand the logical flow of packet with poll-mode driver . How the packet is passing to user space --- > how the memory copy is doing from pkt buffer to user space . Code detail in DPDK code base . Suggestions are welcome on this . Regards Kuldeep The information contained in this electronic message and any attachments to this message are intended for the exclusive use of the addressee(s) and may contain proprietary, confidential or privileged information. If you are not the intended recipient, you should not disseminate, distribute or copy this e-mail. Please notify the sender immediately and destroy all copies of this message and any attachments. WARNING: Computer viruses can be transmitted via email. The recipient should check this email and any attachments for the presence of viruses. The company accepts no liability for any damage caused by any virus transmitted by this email. www.wipro.com
[dpdk-dev] [PATCH] lib/librte_vhost: add CONFIG_RTE_LIBRTE_VHOST_USER switch
Turn on CONFIG_RTE_LIBRTE_VHOST to enable vhost. vhost-user is turned on by default. Turn off CONFIG_RTE_LIBRTE_VHOST_USER to enable vhost-cuse implementation. Signed-off-by: Huawei Xie --- config/common_linuxapp| 4 +++- lib/librte_vhost/Makefile | 11 +-- 2 files changed, 12 insertions(+), 3 deletions(-) diff --git a/config/common_linuxapp b/config/common_linuxapp index 97f1c9e..09a58ac 100644 --- a/config/common_linuxapp +++ b/config/common_linuxapp @@ -414,10 +414,12 @@ CONFIG_RTE_KNI_VHOST_DEBUG_TX=n # # Compile vhost library -# fuse-devel is needed to run vhost. +# fuse-devel is needed to run vhost-cuse. # fuse-devel enables user space char driver development +# vhost-user is turned on by default. # CONFIG_RTE_LIBRTE_VHOST=n +CONFIG_RTE_LIBRTE_VHOST_USER=y CONFIG_RTE_LIBRTE_VHOST_DEBUG=n # diff --git a/lib/librte_vhost/Makefile b/lib/librte_vhost/Makefile index 52f6575..a8645a6 100644 --- a/lib/librte_vhost/Makefile +++ b/lib/librte_vhost/Makefile @@ -39,13 +39,20 @@ EXPORT_MAP := rte_vhost_version.map LIBABIVER := 1 CFLAGS += $(WERROR_FLAGS) -I$(SRCDIR) -O3 -D_FILE_OFFSET_BITS=64 -CFLAGS += -I vhost_cuse -lfuse +ifeq ($(CONFIG_RTE_LIBRTE_VHOST_USER),y) CFLAGS += -I vhost_user +else +CFLAGS += -I vhost_cuse -lfuse LDFLAGS += -lfuse +endif + # all source are stored in SRCS-y SRCS-$(CONFIG_RTE_LIBRTE_VHOST) := virtio-net.c vhost_rxtx.c -#SRCS-$(CONFIG_RTE_LIBRTE_VHOST) += vhost_cuse/vhost-net-cdev.c vhost_cuse/virtio-net-cdev.c vhost_cuse/eventfd_copy.c +ifeq ($(CONFIG_RTE_LIBRTE_VHOST_USER),y) SRCS-$(CONFIG_RTE_LIBRTE_VHOST) += vhost_user/vhost-net-user.c vhost_user/virtio-net-user.c vhost_user/fd_man.c +else +SRCS-$(CONFIG_RTE_LIBRTE_VHOST) += vhost_cuse/vhost-net-cdev.c vhost_cuse/virtio-net-cdev.c vhost_cuse/eventfd_copy.c +endif # install includes SYMLINK-$(CONFIG_RTE_LIBRTE_VHOST)-include += rte_virtio_net.h -- 1.8.1.4
[dpdk-dev] [PATCH v2 1/4] mk: Remove combined library and related options
Remove CONFIG_RTE_BUILD_COMBINE_LIBS and CONFIG_RTE_LIBNAME. Signed-off-by: Sergio Gonzalez Monroy --- config/common_bsdapp| 6 -- config/common_linuxapp | 6 -- config/defconfig_ppc_64-power8-linuxapp-gcc | 2 - lib/Makefile| 1 - mk/rte.app.mk | 12 mk/rte.lib.mk | 35 -- mk/rte.sdkbuild.mk | 3 - mk/rte.sharelib.mk | 101 mk/rte.vars.mk | 9 --- 9 files changed, 175 deletions(-) delete mode 100644 mk/rte.sharelib.mk diff --git a/config/common_bsdapp b/config/common_bsdapp index 8ff4dc2..7ee5ecf 100644 --- a/config/common_bsdapp +++ b/config/common_bsdapp @@ -79,12 +79,6 @@ CONFIG_RTE_FORCE_INTRINSICS=n CONFIG_RTE_BUILD_SHARED_LIB=n # -# Combine to one single library -# -CONFIG_RTE_BUILD_COMBINE_LIBS=n -CONFIG_RTE_LIBNAME=intel_dpdk - -# # Compile Environment Abstraction Layer # CONFIG_RTE_LIBRTE_EAL=y diff --git a/config/common_linuxapp b/config/common_linuxapp index 97f1c9e..ae13805 100644 --- a/config/common_linuxapp +++ b/config/common_linuxapp @@ -79,12 +79,6 @@ CONFIG_RTE_FORCE_INTRINSICS=n CONFIG_RTE_BUILD_SHARED_LIB=n # -# Combine to one single library -# -CONFIG_RTE_BUILD_COMBINE_LIBS=n -CONFIG_RTE_LIBNAME="intel_dpdk" - -# # Compile Environment Abstraction Layer # CONFIG_RTE_LIBRTE_EAL=y diff --git a/config/defconfig_ppc_64-power8-linuxapp-gcc b/config/defconfig_ppc_64-power8-linuxapp-gcc index d97a885..f1af518 100644 --- a/config/defconfig_ppc_64-power8-linuxapp-gcc +++ b/config/defconfig_ppc_64-power8-linuxapp-gcc @@ -39,8 +39,6 @@ CONFIG_RTE_ARCH_64=y CONFIG_RTE_TOOLCHAIN="gcc" CONFIG_RTE_TOOLCHAIN_GCC=y -CONFIG_RTE_LIBNAME="powerpc_dpdk" - # Note: Power doesn't have this support CONFIG_RTE_LIBRTE_EAL_VMWARE_TSC_MAP_SUPPORT=n diff --git a/lib/Makefile b/lib/Makefile index d94355d..c34cf2f 100644 --- a/lib/Makefile +++ b/lib/Makefile @@ -77,5 +77,4 @@ DIRS-$(CONFIG_RTE_LIBRTE_KNI) += librte_kni DIRS-$(CONFIG_RTE_LIBRTE_IVSHMEM) += librte_ivshmem endif -include $(RTE_SDK)/mk/rte.sharelib.mk include $(RTE_SDK)/mk/rte.subdir.mk diff --git a/mk/rte.app.mk b/mk/rte.app.mk index 63a41e2..e2baa49 100644 --- a/mk/rte.app.mk +++ b/mk/rte.app.mk @@ -61,12 +61,6 @@ ifeq ($(NO_AUTOLIBS),) LDLIBS += --whole-archive -ifeq ($(CONFIG_RTE_BUILD_COMBINE_LIBS),y) -LDLIBS += -l$(RTE_LIBNAME) -endif - -ifeq ($(CONFIG_RTE_BUILD_COMBINE_LIBS),n) - ifeq ($(CONFIG_RTE_LIBRTE_DISTRIBUTOR),y) LDLIBS += -lrte_distributor endif @@ -137,8 +131,6 @@ ifeq ($(CONFIG_RTE_LIBRTE_VHOST), y) LDLIBS += -lrte_vhost endif -endif # ! CONFIG_RTE_BUILD_COMBINE_LIBS - ifeq ($(CONFIG_RTE_LIBRTE_PMD_PCAP),y) LDLIBS += -lpcap endif @@ -153,8 +145,6 @@ endif LDLIBS += --start-group -ifeq ($(CONFIG_RTE_BUILD_COMBINE_LIBS),n) - ifeq ($(CONFIG_RTE_LIBRTE_KVARGS),y) LDLIBS += -lrte_kvargs endif @@ -253,8 +243,6 @@ endif endif # plugins -endif # ! CONFIG_RTE_BUILD_COMBINE_LIBS - LDLIBS += $(EXECENV_LDLIBS) LDLIBS += --end-group diff --git a/mk/rte.lib.mk b/mk/rte.lib.mk index 0d7482d..d96101a 100644 --- a/mk/rte.lib.mk +++ b/mk/rte.lib.mk @@ -87,24 +87,6 @@ O_TO_S_DO = @set -e; \ $(O_TO_S) && \ echo $(O_TO_S_CMD) > $(call exe2cmd,$(@)) -ifeq ($(RTE_BUILD_SHARED_LIB),n) -O_TO_C = $(AR) crus $(LIB_ONE) $(OBJS-y) -O_TO_C_STR = $(subst ','\'',$(O_TO_C)) #'# fix syntax highlight -O_TO_C_DISP = $(if $(V),"$(O_TO_C_STR)"," AR_C $(@)") -O_TO_C_DO = @set -e; \ - $(lib_dir) \ - $(copy_obj) -else -O_TO_C = $(LD) -shared $(OBJS-y) -o $(LIB_ONE) -O_TO_C_STR = $(subst ','\'',$(O_TO_C)) #'# fix syntax highlight -O_TO_C_DISP = $(if $(V),"$(O_TO_C_STR)"," LD_C $(@)") -O_TO_C_DO = @set -e; \ - $(lib_dir) \ - $(copy_obj) -endif - -copy_obj = cp -f $(OBJS-y) $(RTE_OUTPUT)/build/lib; -lib_dir = [ -d $(RTE_OUTPUT)/lib ] || mkdir -p $(RTE_OUTPUT)/lib; -include .$(LIB).cmd # @@ -129,15 +111,6 @@ endif $(depfile_missing),\ $(depfile_newer)),\ $(O_TO_S_DO)) - -ifeq ($(RTE_BUILD_COMBINE_LIBS),y) - $(if $(or \ -$(file_missing),\ -$(call cmdline_changed,$(O_TO_C_STR)),\ -$(depfile_missing),\ -$(depfile_newer)),\ -$(O_TO_C_DO)) -endif else $(LIB): $(OBJS-y) $(DEP_$(LIB)) FORCE @[ -d $(dir $@) ] || mkdir -p $(dir $@) @@ -153,14 +126,6 @@ $(LIB): $(OBJS-y) $(DEP_$(LIB)) FORCE $(depfile_missing),\ $(depfile_newer)),\ $(O_TO_A_DO)) -ifeq ($(RTE_BUILD_COMBINE_LIBS),y) - $(if $(or \ -$(file_missing),\ -$(call cmdline_changed,$(O_TO_C_STR)),\ -$(depfile_missing),\ -$(depfile_newer)),\ -$(O_TO_C_DO)) -endif endif # diff --git a/mk/rte.sdkbuild.mk b/mk/rte.sdkbuild.mk index 3154457..2b24e74 100644 --- a/mk/rte.sdkbuild.mk +++ b/mk/rte
[dpdk-dev] [PATCH v2 4/4] mk: update LDLIBS for app building
Given the circular dependencies between eal, malloc, mempool and ring libraries and that eal does not have proper DT_NEEDED entries, we work around it by forcing linking against mentioned libraries by preceding them with --no-as-needed flag when building shared libraries. This patch also does: - Set --start-group/--end-group and --whole-archive/--no-whole-archive flags only when linking against static DPDK libs. - Set --as-needed for all libraries but eal, malloc, mempool and ring when linking against shared DPDK libs. - Link against EXECENV_LIBS always with --as-needed flag. Signed-off-by: Sergio Gonzalez Monroy --- lib/librte_jobstats/Makefile | 2 ++ lib/librte_pmd_fm10k/Makefile | 2 ++ lib/librte_pmd_mlx4/Makefile | 2 ++ lib/librte_pmd_null/Makefile | 2 ++ lib/librte_reorder/Makefile | 2 ++ mk/rte.app.mk | 57 +++ 6 files changed, 40 insertions(+), 27 deletions(-) diff --git a/lib/librte_jobstats/Makefile b/lib/librte_jobstats/Makefile index 136a448..04589d4 100644 --- a/lib/librte_jobstats/Makefile +++ b/lib/librte_jobstats/Makefile @@ -41,6 +41,8 @@ EXPORT_MAP := rte_jobstats_version.map LIBABIVER := 1 +LDLIBS += -lrte_eal + # all source are stored in SRCS-y SRCS-$(CONFIG_RTE_LIBRTE_JOBSTATS) := rte_jobstats.c diff --git a/lib/librte_pmd_fm10k/Makefile b/lib/librte_pmd_fm10k/Makefile index 998bf23..618235a 100644 --- a/lib/librte_pmd_fm10k/Makefile +++ b/lib/librte_pmd_fm10k/Makefile @@ -79,6 +79,8 @@ $(foreach obj, $(BASE_DRIVER_OBJS), $(eval CFLAGS_$(obj)+=$(CFLAGS_BASE_DRIVER)) VPATH += $(RTE_SDK)/lib/librte_pmd_fm10k/base +LDLIBS += -lrte_eal -lrte_malloc -lethdev + # # all source are stored in SRCS-y # diff --git a/lib/librte_pmd_mlx4/Makefile b/lib/librte_pmd_mlx4/Makefile index 813..9193f05 100644 --- a/lib/librte_pmd_mlx4/Makefile +++ b/lib/librte_pmd_mlx4/Makefile @@ -58,6 +58,8 @@ CFLAGS += -Wno-error=cast-qual EXPORT_MAP := rte_pmd_mlx4_version.map LIBABIVER := 1 +LDLIBS += -lrte_eal -lrte_mbuf -lrte_mempool -lrte_malloc -lethdev -libverbs + # DEBUG which is usually provided on the command-line may enable # CONFIG_RTE_LIBRTE_MLX4_DEBUG. ifeq ($(DEBUG),1) diff --git a/lib/librte_pmd_null/Makefile b/lib/librte_pmd_null/Makefile index 6472015..c9bf1fd 100644 --- a/lib/librte_pmd_null/Makefile +++ b/lib/librte_pmd_null/Makefile @@ -43,6 +43,8 @@ EXPORT_MAP := rte_pmd_null_version.map LIBABIVER := 1 +LDLIBS += -lrte_eal -lrte_malloc -lethdev -lrte_kvargs + # # all source are stored in SRCS-y # diff --git a/lib/librte_reorder/Makefile b/lib/librte_reorder/Makefile index 0c01de1..2c8f774 100644 --- a/lib/librte_reorder/Makefile +++ b/lib/librte_reorder/Makefile @@ -41,6 +41,8 @@ EXPORT_MAP := rte_reorder_version.map LIBABIVER := 1 +LDLIBS += -lrte_eal -lrte_malloc + # all source are stored in SRCS-y SRCS-$(CONFIG_RTE_LIBRTE_REORDER) := rte_reorder.c diff --git a/mk/rte.app.mk b/mk/rte.app.mk index e2baa49..7e225fd 100644 --- a/mk/rte.app.mk +++ b/mk/rte.app.mk @@ -59,7 +59,30 @@ LDLIBS += -L$(RTE_SDK_BIN)/lib # ifeq ($(NO_AUTOLIBS),) -LDLIBS += --whole-archive +LDLIBS += --no-as-needed +ifneq ($(CONFIG_RTE_BUILD_SHARED_LIB),y) +LDLIBS += --start-group +endif + +ifeq ($(CONFIG_RTE_LIBRTE_EAL),y) +LDLIBS += -lrte_eal +endif + +ifeq ($(CONFIG_RTE_LIBRTE_MALLOC),y) +LDLIBS += -lrte_malloc +endif + +ifeq ($(CONFIG_RTE_LIBRTE_MEMPOOL),y) +LDLIBS += -lrte_mempool +endif + +ifeq ($(CONFIG_RTE_LIBRTE_RING),y) +LDLIBS += -lrte_ring +endif + +ifeq ($(CONFIG_RTE_BUILD_SHARED_LIB),y) +LDLIBS += --as-needed +endif ifeq ($(CONFIG_RTE_LIBRTE_DISTRIBUTOR),y) LDLIBS += -lrte_distributor @@ -127,15 +150,12 @@ LDLIBS += -lm LDLIBS += -lrt endif -ifeq ($(CONFIG_RTE_LIBRTE_VHOST), y) -LDLIBS += -lrte_vhost -endif - ifeq ($(CONFIG_RTE_LIBRTE_PMD_PCAP),y) LDLIBS += -lpcap endif ifeq ($(CONFIG_RTE_LIBRTE_VHOST),y) +LDLIBS += -lrte_vhost LDLIBS += -lfuse endif @@ -143,8 +163,6 @@ ifeq ($(CONFIG_RTE_LIBRTE_MLX4_PMD),y) LDLIBS += -libverbs endif -LDLIBS += --start-group - ifeq ($(CONFIG_RTE_LIBRTE_KVARGS),y) LDLIBS += -lrte_kvargs endif @@ -161,22 +179,6 @@ ifeq ($(CONFIG_RTE_LIBRTE_ETHER),y) LDLIBS += -lethdev endif -ifeq ($(CONFIG_RTE_LIBRTE_MALLOC),y) -LDLIBS += -lrte_malloc -endif - -ifeq ($(CONFIG_RTE_LIBRTE_MEMPOOL),y) -LDLIBS += -lrte_mempool -endif - -ifeq ($(CONFIG_RTE_LIBRTE_RING),y) -LDLIBS += -lrte_ring -endif - -ifeq ($(CONFIG_RTE_LIBRTE_EAL),y) -LDLIBS += -lrte_eal -endif - ifeq ($(CONFIG_RTE_LIBRTE_CMDLINE),y) LDLIBS += -lrte_cmdline endif @@ -196,6 +198,7 @@ endif ifeq ($(CONFIG_RTE_BUILD_SHARED_LIB),n) # plugins (link only if static libraries) +LDLIBS += --whole-archive ifeq ($(CONFIG_RTE_LIBRTE_VMXNET3_PMD),y) LDLIBS += -lrte_pmd_vmxnet3_uio @@ -241,14 +244,14 @@ ifeq ($(CONFIG_RTE_LIBRTE_PMD_AF_PACKET),y) LDLIBS += -lrte_pmd_af_packet endif +LDLIBS += --no-whole-archive +LDLIBS += --end-group +LDLIBS += --as-needed + endif # plugins LDLIBS += $(EXE
[dpdk-dev] [PATCH v2 0/4] Improve build process
This patch series improves the DPDK build system mostly for shared libraries (and a few nits for static libraries) with the following goals: - Remove config option to build a combined library. - For shared libraries, explicitly link against dependant libraries (adding entries to DT_NEEDED). - Update app linking flags for static/shared DPDK libs. v2: - Do not create a core library to solve circular dependencies between eal, malloc, mempool and ring libraries. Instead, add DT_NEEDED entries for all libraries but eal, then for application linking, always link against these libraries by preceding them with --no-as-needed flag. Sergio Gonzalez Monroy (4): mk: Remove combined library and related options lib: Set LDLIBS for each library mk: Use LDLIBS when linking shared libraries mk: update LDLIBS for app building config/common_bsdapp| 6 -- config/common_linuxapp | 6 -- config/defconfig_ppc_64-power8-linuxapp-gcc | 2 - lib/Makefile| 1 - lib/librte_acl/Makefile | 2 + lib/librte_cfgfile/Makefile | 2 + lib/librte_cmdline/Makefile | 2 + lib/librte_distributor/Makefile | 2 + lib/librte_ether/Makefile | 5 +- lib/librte_hash/Makefile| 2 + lib/librte_ip_frag/Makefile | 3 + lib/librte_ivshmem/Makefile | 2 + lib/librte_jobstats/Makefile| 2 + lib/librte_kni/Makefile | 2 + lib/librte_kvargs/Makefile | 2 + lib/librte_lpm/Makefile | 2 + lib/librte_malloc/Makefile | 2 + lib/librte_mbuf/Makefile| 2 + lib/librte_mempool/Makefile | 2 + lib/librte_meter/Makefile | 2 + lib/librte_pipeline/Makefile| 2 + lib/librte_pmd_af_packet/Makefile | 2 + lib/librte_pmd_bond/Makefile| 6 ++ lib/librte_pmd_e1000/Makefile | 2 + lib/librte_pmd_enic/Makefile| 3 + lib/librte_pmd_fm10k/Makefile | 2 + lib/librte_pmd_i40e/Makefile| 2 + lib/librte_pmd_ixgbe/Makefile | 2 + lib/librte_pmd_mlx4/Makefile| 2 + lib/librte_pmd_null/Makefile| 2 + lib/librte_pmd_pcap/Makefile| 2 + lib/librte_pmd_ring/Makefile| 4 +- lib/librte_pmd_virtio/Makefile | 2 + lib/librte_pmd_vmxnet3/Makefile | 2 + lib/librte_pmd_xenvirt/Makefile | 3 + lib/librte_port/Makefile| 4 ++ lib/librte_power/Makefile | 2 + lib/librte_reorder/Makefile | 2 + lib/librte_ring/Makefile| 2 + lib/librte_sched/Makefile | 2 + lib/librte_table/Makefile | 4 ++ lib/librte_timer/Makefile | 2 + lib/librte_vhost/Makefile | 7 +- mk/rte.app.mk | 63 - mk/rte.lib.mk | 50 +++--- mk/rte.sdkbuild.mk | 3 - mk/rte.sharelib.mk | 101 mk/rte.vars.mk | 9 --- 48 files changed, 131 insertions(+), 209 deletions(-) delete mode 100644 mk/rte.sharelib.mk -- 1.9.3
[dpdk-dev] [PATCH v2 3/4] mk: Use LDLIBS when linking shared libraries
This patch mainly makes use of the LDLIBS variable when linking shared libraries, setting proper DT_NEEDED entries. This patch also fixes a few nits like syntax highlighting, the command string (O_TO_S_CMD) used for linking shared libraries and the displayed of dependencies when debugging is enable (D). Signed-off-by: Sergio Gonzalez Monroy --- mk/rte.lib.mk | 15 ++- 1 file changed, 10 insertions(+), 5 deletions(-) diff --git a/mk/rte.lib.mk b/mk/rte.lib.mk index d96101a..603badf 100644 --- a/mk/rte.lib.mk +++ b/mk/rte.lib.mk @@ -62,16 +62,19 @@ build: _postbuild exe2cmd = $(strip $(call dotfile,$(patsubst %,%.cmd,$(1 +_LDLIBS := --as-needed $(LDLIBS) $(EXECENV_LDLIBS) --no-as-needed + ifeq ($(LINK_USING_CC),1) # Override the definition of LD here, since we're linking with CC LD := $(CC) $(CPU_CFLAGS) _CPU_LDFLAGS := $(call linkerprefix,$(CPU_LDFLAGS)) +_LDLIBS := $(call linkerprefix,$(_LDLIBS)) else _CPU_LDFLAGS := $(CPU_LDFLAGS) endif O_TO_A = $(AR) crus $(LIB) $(OBJS-y) -O_TO_A_STR = $(subst ','\'',$(O_TO_A)) #'# fix syntax highlight +O_TO_A_STR = $(subst ','\'',$(O_TO_A)) #')# fix syntax highlight O_TO_A_DISP = $(if $(V),"$(O_TO_A_STR)"," AR $(@)") O_TO_A_CMD = "cmd_$@ = $(O_TO_A_STR)" O_TO_A_DO = @set -e; \ @@ -79,9 +82,11 @@ O_TO_A_DO = @set -e; \ $(O_TO_A) && \ echo $(O_TO_A_CMD) > $(call exe2cmd,$(@)) -O_TO_S = $(LD) $(_CPU_LDFLAGS) -shared $(OBJS-y) -Wl,-soname,$(LIB) -o $(LIB) -O_TO_S_STR = $(subst ','\'',$(O_TO_S)) #'# fix syntax highlight +O_TO_S = $(LD) $(_CPU_LDFLAGS) -L $(RTE_OUTPUT)/lib -Wl,-soname,$(LIB) \ +-shared $(OBJS-y) $(_LDLIBS) -o $(LIB) +O_TO_S_STR = $(subst ','\'',$(O_TO_S)) #')# fix syntax highlight O_TO_S_DISP = $(if $(V),"$(O_TO_S_STR)"," LD $(@)") +O_TO_S_CMD = "cmd_$@ = $(O_TO_S_STR)" O_TO_S_DO = @set -e; \ echo $(O_TO_S_DISP); \ $(O_TO_S) && \ @@ -100,7 +105,7 @@ ifeq ($(LIBABIVER),) endif @[ -d $(dir $@) ] || mkdir -p $(dir $@) $(if $(D),\ - @echo -n "$< -> $@ " ; \ + @echo -n "$? -> $@ " ; \ echo -n "file_missing=$(call boolean,$(file_missing)) " ; \ echo -n "cmdline_changed=$(call boolean,$(call cmdline_changed,$(O_TO_S_STR))) " ; \ echo -n "depfile_missing=$(call boolean,$(depfile_missing)) " ; \ @@ -115,7 +120,7 @@ else $(LIB): $(OBJS-y) $(DEP_$(LIB)) FORCE @[ -d $(dir $@) ] || mkdir -p $(dir $@) $(if $(D),\ - @echo -n "$< -> $@ " ; \ + @echo -n "$? -> $@ " ; \ echo -n "file_missing=$(call boolean,$(file_missing)) " ; \ echo -n "cmdline_changed=$(call boolean,$(call cmdline_changed,$(O_TO_A_STR))) " ; \ echo -n "depfile_missing=$(call boolean,$(depfile_missing)) " ; \ -- 1.9.3
[dpdk-dev] [PATCH v2 2/4] lib: Set LDLIBS for each library
When creating shared libraries, each library will be linked against their dependant libraries, which are specified by new LDLIBS var. Given the circular dependencies between eal, malloc, mempool and ring, we work around it by not linking eal against its dependent libraries. Therefore, eal will not have proper DT_NEEDED entries and we will have to force link against them by preceding such libraries with --no-as-needed flag. This patch sets LDLIBS variable for each library but eal and updates DEPDIRS of some libraries. Signed-off-by: Sergio Gonzalez Monroy --- lib/librte_acl/Makefile | 2 ++ lib/librte_cfgfile/Makefile | 2 ++ lib/librte_cmdline/Makefile | 2 ++ lib/librte_distributor/Makefile | 2 ++ lib/librte_ether/Makefile | 5 - lib/librte_hash/Makefile | 2 ++ lib/librte_ip_frag/Makefile | 3 +++ lib/librte_ivshmem/Makefile | 2 ++ lib/librte_kni/Makefile | 2 ++ lib/librte_kvargs/Makefile| 2 ++ lib/librte_lpm/Makefile | 2 ++ lib/librte_malloc/Makefile| 2 ++ lib/librte_mbuf/Makefile | 2 ++ lib/librte_mempool/Makefile | 2 ++ lib/librte_meter/Makefile | 2 ++ lib/librte_pipeline/Makefile | 2 ++ lib/librte_pmd_af_packet/Makefile | 2 ++ lib/librte_pmd_bond/Makefile | 6 ++ lib/librte_pmd_e1000/Makefile | 2 ++ lib/librte_pmd_enic/Makefile | 3 +++ lib/librte_pmd_i40e/Makefile | 2 ++ lib/librte_pmd_ixgbe/Makefile | 2 ++ lib/librte_pmd_pcap/Makefile | 2 ++ lib/librte_pmd_ring/Makefile | 4 +++- lib/librte_pmd_virtio/Makefile| 2 ++ lib/librte_pmd_vmxnet3/Makefile | 2 ++ lib/librte_pmd_xenvirt/Makefile | 3 +++ lib/librte_port/Makefile | 4 lib/librte_power/Makefile | 2 ++ lib/librte_ring/Makefile | 2 ++ lib/librte_sched/Makefile | 2 ++ lib/librte_table/Makefile | 4 lib/librte_timer/Makefile | 2 ++ lib/librte_vhost/Makefile | 7 --- 34 files changed, 84 insertions(+), 5 deletions(-) diff --git a/lib/librte_acl/Makefile b/lib/librte_acl/Makefile index 68dc248..00f5e33 100644 --- a/lib/librte_acl/Makefile +++ b/lib/librte_acl/Makefile @@ -41,6 +41,8 @@ EXPORT_MAP := rte_acl_version.map LIBABIVER := 1 +LDLIBS += -lrte_eal -lrte_malloc + # all source are stored in SRCS-y SRCS-$(CONFIG_RTE_LIBRTE_ACL) += tb_mem.c diff --git a/lib/librte_cfgfile/Makefile b/lib/librte_cfgfile/Makefile index 032c240..babe7d1 100644 --- a/lib/librte_cfgfile/Makefile +++ b/lib/librte_cfgfile/Makefile @@ -43,6 +43,8 @@ EXPORT_MAP := rte_cfgfile_version.map LIBABIVER := 1 +LDLIBS += -lrte_eal + # # all source are stored in SRCS-y # diff --git a/lib/librte_cmdline/Makefile b/lib/librte_cmdline/Makefile index 719dff6..0f7935f 100644 --- a/lib/librte_cmdline/Makefile +++ b/lib/librte_cmdline/Makefile @@ -40,6 +40,8 @@ EXPORT_MAP := rte_cmdline_version.map LIBABIVER := 1 +LDLIBS += -lrte_eal + # all source are stored in SRCS-y SRCS-$(CONFIG_RTE_LIBRTE_CMDLINE) := cmdline.c SRCS-$(CONFIG_RTE_LIBRTE_CMDLINE) += cmdline_cirbuf.c diff --git a/lib/librte_distributor/Makefile b/lib/librte_distributor/Makefile index 4c9af17..b275491 100644 --- a/lib/librte_distributor/Makefile +++ b/lib/librte_distributor/Makefile @@ -41,6 +41,8 @@ EXPORT_MAP := rte_distributor_version.map LIBABIVER := 1 +LDLIBS += -lrte_eal -lrte_mbuf + # all source are stored in SRCS-y SRCS-$(CONFIG_RTE_LIBRTE_DISTRIBUTOR) := rte_distributor.c diff --git a/lib/librte_ether/Makefile b/lib/librte_ether/Makefile index c0e5768..a1059d7 100644 --- a/lib/librte_ether/Makefile +++ b/lib/librte_ether/Makefile @@ -43,6 +43,8 @@ EXPORT_MAP := rte_ether_version.map LIBABIVER := 1 +LDLIBS += -lrte_eal -lrte_mempool -lrte_ring -lrte_malloc + SRCS-y += rte_ethdev.c # @@ -53,6 +55,7 @@ SYMLINK-y-include += rte_ethdev.h SYMLINK-y-include += rte_eth_ctrl.h # this lib depends upon: -DEPDIRS-y += lib/librte_eal lib/librte_mempool lib/librte_ring lib/librte_mbuf +DEPDIRS-y += lib/librte_eal lib/librte_mempool lib/librte_ring +DEPDIRS-y += lib/librte_mbuf lib/librte_malloc include $(RTE_SDK)/mk/rte.lib.mk diff --git a/lib/librte_hash/Makefile b/lib/librte_hash/Makefile index 3696cb1..bc9bfc7 100644 --- a/lib/librte_hash/Makefile +++ b/lib/librte_hash/Makefile @@ -41,6 +41,8 @@ EXPORT_MAP := rte_hash_version.map LIBABIVER := 1 +LDLIBS += -lrte_eal -lrte_malloc + # all source are stored in SRCS-y SRCS-$(CONFIG_RTE_LIBRTE_HASH) := rte_hash.c SRCS-$(CONFIG_RTE_LIBRTE_HASH) += rte_fbk_hash.c diff --git a/lib/librte_ip_frag/Makefile b/lib/librte_ip_frag/Makefile index 9d06780..ee72ab4 100644 --- a/lib/librte_ip_frag/Makefile +++ b/lib/librte_ip_frag/Makefile @@ -41,6 +41,8 @@ EXPORT_MAP := rte_ipfrag_version.map LIBABIVER := 1 +LDLIBS += -lrte_eal -lrte_malloc -lethdev + #source files SRCS-$(CONFIG_RTE_LIBRTE_IP_FRAG) += rte_ipv4_fragmentation.c SRCS-$(CONFIG_RTE_LIBRTE_IP_FRAG) += rte_ipv6_fragmentation.
[dpdk-dev] [PATCH] ethdev: additional parameter in RX callback
This patch is a minor extension to the recent patchset for RX/TX callbacks based on feedback from users implementing solutions based on it. The patch adds a new parameter to the RX callback to pass in the number of available RX packets in addition to the number of dequeued packets. This provides the RX callback functions with additional information that can be used to decide how packets from a burst are handled. The TX callback doesn't require this additional parameter so the RX and TX callbacks no longer have the same function parameters. As such the single RX/TX callback has been refactored into two separate callbacks. Since this is an API change we hope it can be included in 2.0.0 to avoid changing the API in a subsequent release. John McNamara (1): ethdev: added additional packet count parameter to RX callbacks examples/rxtx_callbacks/main.c |3 +- lib/librte_ether/rte_ethdev.c |8 ++-- lib/librte_ether/rte_ethdev.h | 74 ++-- 3 files changed, 54 insertions(+), 31 deletions(-) -- 1.7.4.1
[dpdk-dev] [PATCH] ethdev: added additional packet count parameter to RX callbacks
Added a parameter to the RX callback to pass in the number of
available RX packets in addition to the number of dequeued packets.
This provides the RX callback functions with additional information
that can be used to decide how packets from a burst are handled.
The TX callback doesn't require this additional parameter so the RX
and TX callbacks no longer have the same function parameters. As such
the single RX/TX callback has been refactored into two separate callbacks.
Signed-off-by: John McNamara
---
examples/rxtx_callbacks/main.c |3 +-
lib/librte_ether/rte_ethdev.c |8 ++--
lib/librte_ether/rte_ethdev.h | 74 ++--
3 files changed, 54 insertions(+), 31 deletions(-)
diff --git a/examples/rxtx_callbacks/main.c b/examples/rxtx_callbacks/main.c
index 9e5e68e..21117ce 100644
--- a/examples/rxtx_callbacks/main.c
+++ b/examples/rxtx_callbacks/main.c
@@ -61,7 +61,8 @@ static struct {
static uint16_t
add_timestamps(uint8_t port __rte_unused, uint16_t qidx __rte_unused,
- struct rte_mbuf **pkts, uint16_t nb_pkts, void *_ __rte_unused)
+ struct rte_mbuf **pkts, uint16_t nb_pkts,
+ uint16_t max_pkts __rte_unused, void *_ __rte_unused)
{
unsigned i;
uint64_t now = rte_rdtsc();
diff --git a/lib/librte_ether/rte_ethdev.c b/lib/librte_ether/rte_ethdev.c
index 03fce08..e142920 100644
--- a/lib/librte_ether/rte_ethdev.c
+++ b/lib/librte_ether/rte_ethdev.c
@@ -3533,7 +3533,7 @@ rte_eth_dev_filter_ctrl(uint8_t port_id, enum
rte_filter_type filter_type,
void *
rte_eth_add_rx_callback(uint8_t port_id, uint16_t queue_id,
- rte_rxtx_callback_fn fn, void *user_param)
+ rte_rx_callback_fn fn, void *user_param)
{
#ifndef RTE_ETHDEV_RXTX_CALLBACKS
rte_errno = ENOTSUP;
@@ -3553,7 +3553,7 @@ rte_eth_add_rx_callback(uint8_t port_id, uint16_t
queue_id,
return NULL;
}
- cb->fn = fn;
+ cb->fn.rx = fn;
cb->param = user_param;
cb->next = rte_eth_devices[port_id].post_rx_burst_cbs[queue_id];
rte_eth_devices[port_id].post_rx_burst_cbs[queue_id] = cb;
@@ -3562,7 +3562,7 @@ rte_eth_add_rx_callback(uint8_t port_id, uint16_t
queue_id,
void *
rte_eth_add_tx_callback(uint8_t port_id, uint16_t queue_id,
- rte_rxtx_callback_fn fn, void *user_param)
+ rte_tx_callback_fn fn, void *user_param)
{
#ifndef RTE_ETHDEV_RXTX_CALLBACKS
rte_errno = ENOTSUP;
@@ -3582,7 +3582,7 @@ rte_eth_add_tx_callback(uint8_t port_id, uint16_t
queue_id,
return NULL;
}
- cb->fn = fn;
+ cb->fn.tx = fn;
cb->param = user_param;
cb->next = rte_eth_devices[port_id].pre_tx_burst_cbs[queue_id];
rte_eth_devices[port_id].pre_tx_burst_cbs[queue_id] = cb;
diff --git a/lib/librte_ether/rte_ethdev.h b/lib/librte_ether/rte_ethdev.h
index 21aa359..e0b0b8d 100644
--- a/lib/librte_ether/rte_ethdev.h
+++ b/lib/librte_ether/rte_ethdev.h
@@ -1384,33 +1384,52 @@ struct eth_dev_ops {
};
/**
- * Function type used for callbacks for processing packets on RX and TX
+ * Function type used for RX packet processing packet callbacks.
*
- * If configured for RX, it is called with a burst of packets that have just
- * been received on the given port and queue. On TX, it is called with a burst
- * of packets immediately before those packets are put onto the hardware queue
- * for transmission.
+ * The callback function is called on RX with a burst of packets that have
+ * been received on the given port and queue.
*
* @param port
- * The ethernet port on which rx or tx is being performed
+ * The Ethernet port on which RX is being performed.
* @param queue
- * The queue on the ethernet port which is being used to receive or transmit
- * the packets.
+ * The queue on the Ethernet port which is being used to receive the packets.
* @param pkts
- * The burst of packets on which processing is to be done. On RX, these
- * packets have just been received. On TX, they are about to be transmitted.
+ * The burst of packets that have just been received.
* @param nb_pkts
- * The number of packets in the burst pointed to by "pkts"
+ * The number of packets in the burst pointed to by "pkts".
+ * @param max_pkts
+ * The max number of packets that can be stored in the "pkts" array.
* @param user_param
* The arbitrary user parameter passed in by the application when the
callback
* was originally configured.
* @return
- * The number of packets remaining in pkts are processing.
- * * On RX, this will be returned to the user as the return value from
- * rte_eth_rx_burst.
- * * On TX, this will be the number of packets actually written to the NIC.
+ * The number of packets returned to the user.
*/
-typedef uint16_t (*rte_rxtx_callback_fn)(uint8_t port, uint16_t queue,
+typedef uint16_t (*rte_rx_callback_fn)(uint8_t port, uint16_t queue,
+
[dpdk-dev] [RFC] af_packet: support port hotplug
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of John W. Linville
> Sent: Tuesday, March 10, 2015 6:36 PM
> To: dev at dpdk.org
> Subject: [dpdk-dev] [RFC] af_packet: support port hotplug
>
> This patch adds finalization code to free resources allocated by the PMD.
> This is based on earlier
> patches for other PMDs by Tetsuya Mukawa , with
> corrections related to data-
> >name.
>
> Signed-off-by: John W. Linville
> Cc: Tetsuya Mukawa
> ---
> lib/librte_pmd_af_packet/rte_eth_af_packet.c | 56
>
> 1 file changed, 56 insertions(+)
>
> diff --git a/lib/librte_pmd_af_packet/rte_eth_af_packet.c
> b/lib/librte_pmd_af_packet/rte_eth_af_packet.c
> index 80e9bdf7f819..57998ab19c96 100644
> --- a/lib/librte_pmd_af_packet/rte_eth_af_packet.c
> +++ b/lib/librte_pmd_af_packet/rte_eth_af_packet.c
> @@ -399,6 +399,13 @@ static struct eth_dev_ops ops = {
> .stats_reset = eth_stats_reset,
> };
>
> +static struct eth_driver rte_af_packet_pmd = {
> + .pci_drv = {
> + .name = "rte_af_packet_pmd",
> + .drv_flags = RTE_PCI_DRV_DETACHABLE,
> + },
> +};
> +
> /*
> * Opens an AF_PACKET socket
> */
> @@ -653,6 +660,10 @@ rte_pmd_init_internals(const char *name,
> if (*eth_dev == NULL)
> goto error;
>
> + /* check length of device name */
> + if ((strlen((*eth_dev)->data->name) + 1) > sizeof(data->name))
> + goto error;
> +
> /*
>* now put it all together
>* - store queue data in internals,
> @@ -669,12 +680,14 @@ rte_pmd_init_internals(const char *name,
> data->nb_tx_queues = (uint16_t)nb_queues;
> data->dev_link = pmd_link;
> data->mac_addrs = &(*internals)->eth_addr;
> + strncpy(data->name, (*eth_dev)->data->name,
> +strlen((*eth_dev)->data->name));
>
> pci_dev->numa_node = numa_node;
>
> (*eth_dev)->data = data;
> (*eth_dev)->dev_ops = &ops;
> (*eth_dev)->pci_dev = pci_dev;
> + (*eth_dev)->driver = &rte_af_packet_pmd;
>
> return 0;
>
> @@ -835,10 +848,53 @@ rte_pmd_af_packet_devinit(const char *name, const char
> *params)
> return 0;
> }
>
> +static int
> +rte_pmd_af_packet_devuninit(const char *name) {
> + struct rte_eth_dev *eth_dev = NULL;
> + struct pmd_internals *internals;
> + struct tpacket_req req;
> +
> + unsigned q;
> +
> + RTE_LOG(INFO, PMD, "Closing AF_PACKET ethdev on numa socket %u\n",
> + rte_socket_id());
> +
> + if (name == NULL)
> + return -1;
Hi Tetsuya, John,
Before detaching a port, the port must be stopped and closed.
The stop and close are only allowed for RTE_PROC_PRIMARY.
Should there be a check for process_type here?
if (rte_eal_process_type() != RTE_PROC_PRIMARY)
return -EPERM;
Regards,
Bernard
> +
> + /* retrieve ethdev entry */
> + eth_dev = rte_eth_dev_allocated(name);
> + if (eth_dev == NULL)
> + return -1;
> +
> + internals = eth_dev->data->dev_private;
> + req = internals->req;
> +
> + for (q = 0; q < internals->nb_queues; q++) {
> + munmap(internals->rx_queue[q].map,
> +2 * req.tp_block_size * req.tp_block_nr);
> + if (internals->rx_queue[q].rd)
> + rte_free(internals->rx_queue[q].rd);
> + if (internals->tx_queue[q].rd)
> + rte_free(internals->tx_queue[q].rd);
> + }
> +
> + rte_free(internals);
> + rte_free(eth_dev->data);
> + rte_free(eth_dev->pci_dev);
> +
> + rte_eth_dev_release_port(eth_dev);
> +
> +
> + return 0;
> +}
> +
> static struct rte_driver pmd_af_packet_drv = {
> .name = "eth_af_packet",
> .type = PMD_VDEV,
> .init = rte_pmd_af_packet_devinit,
> + .uninit = rte_pmd_af_packet_devuninit,
> };
>
> PMD_REGISTER_DRIVER(pmd_af_packet_drv);
> --
> 2.1.0
[dpdk-dev] [PATCH v7] af_packet: Fix some klocwork errors
On Thu, Mar 12, 2015 at 09:38:46AM -0400, Neil Horman wrote: > On Thu, Mar 12, 2015 at 10:43:24AM +, Qiu, Michael wrote: > > On 3/11/2015 9:34 AM, Ouyang Changchun wrote: > > > Fix possible memory leak issue: free kvlist before return; > > > Fix possible resource lost issue: close qssockfd before return; > > > > > > Signed-off-by: Changchun Ouyang > > > --- > > > Change in v7: > > > - Remove unnecessary '!= NULL' check before freeing it; > > > > > > Change in v6: > > > - Refine exit point; > > > > > > Change in v5: > > > - Initialize qsockfd with -1; > > > > > > Change in v4: > > > - Check sockfd in internals->rx_queue against 0; > > > > > > Change in v3: > > > - Also close sockets for all queues; > > > > > > Change in v2: > > > - Make the error exit point a common path. > > > > Acked-by: Michael Qiu > Acked-by: Neil Horman Acked-by: John W. Linville -- John W. LinvilleSomeday the world will need a hero, and you linville at tuxdriver.com might be all we have. Be ready.
[dpdk-dev] [PATCH] tools: remove chmod absolute path in setup.sh
setup.sh uses /usr/bin/chmod, but depending on distribution, it is not always
there.
For example, Ubuntu has /bin/chmod. Fix this by removing the absolute path,
like it is
done e.g. with grep.
---
tools/setup.sh | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/tools/setup.sh b/tools/setup.sh
index 245900f..ac438c2 100755
--- a/tools/setup.sh
+++ b/tools/setup.sh
@@ -220,7 +220,7 @@ load_vfio_module()
# make sure regular users can read /dev/vfio
echo "chmod /dev/vfio"
- sudo /usr/bin/chmod a+x /dev/vfio
+ sudo chmod a+x /dev/vfio
if [ $? -ne 0 ] ; then
echo "FAIL"
quit
@@ -278,7 +278,7 @@ set_vfio_permissions()
{
# make sure regular users can read /dev/vfio
echo "chmod /dev/vfio"
- sudo /usr/bin/chmod a+x /dev/vfio
+ sudo chmod a+x /dev/vfio
if [ $? -ne 0 ] ; then
echo "FAIL"
quit
@@ -287,7 +287,7 @@ set_vfio_permissions()
# make sure regular user can access everything inside /dev/vfio
echo "chmod /dev/vfio/*"
- sudo /usr/bin/chmod 0666 /dev/vfio/*
+ sudo chmod 0666 /dev/vfio/*
if [ $? -ne 0 ] ; then
echo "FAIL"
quit
--
1.9.1
[dpdk-dev] [PATCH] kni: fix compilation issue on kernel 3.19
Due to API changes in functions ndo_dflt_bridge_getlink and igb_ndo_fdb_add in kernel 3.19, DPDK would not build. This patch solves the problem, by checking the kernel version and adding the necessary new parameters Signed-off-by: Pablo de Lara --- lib/librte_eal/linuxapp/kni/ethtool/igb/igb_main.c |7 +++ lib/librte_eal/linuxapp/kni/ethtool/igb/kcompat.h |3 +++ 2 files changed, 10 insertions(+), 0 deletions(-) diff --git a/lib/librte_eal/linuxapp/kni/ethtool/igb/igb_main.c b/lib/librte_eal/linuxapp/kni/ethtool/igb/igb_main.c index a802a02..24b147d 100644 --- a/lib/librte_eal/linuxapp/kni/ethtool/igb/igb_main.c +++ b/lib/librte_eal/linuxapp/kni/ethtool/igb/igb_main.c @@ -2103,6 +2103,9 @@ static int igb_set_features(struct net_device *netdev, static int igb_ndo_fdb_add(struct ndmsg *ndm, struct nlattr *tb[], struct net_device *dev, const unsigned char *addr, +#ifdef HAVE_NDO_FDB_ADD_VID + u16 vid, +#endif u16 flags) #else static int igb_ndo_fdb_add(struct ndmsg *ndm, @@ -2259,7 +2262,11 @@ static int igb_ndo_bridge_getlink(struct sk_buff *skb, u32 pid, u32 seq, else mode = BRIDGE_MODE_VEPA; +#ifdef HAVE_NDO_FDB_ADD_VID + return ndo_dflt_bridge_getlink(skb, pid, seq, dev, mode, 0, 0); +#else return ndo_dflt_bridge_getlink(skb, pid, seq, dev, mode); +#endif /* HAVE_NDO_FDB_ADD_VID */ } #endif /* HAVE_BRIDGE_ATTRIBS */ #endif /* NTF_SELF */ diff --git a/lib/librte_eal/linuxapp/kni/ethtool/igb/kcompat.h b/lib/librte_eal/linuxapp/kni/ethtool/igb/kcompat.h index 1213cc6..5c799e9 100644 --- a/lib/librte_eal/linuxapp/kni/ethtool/igb/kcompat.h +++ b/lib/librte_eal/linuxapp/kni/ethtool/igb/kcompat.h @@ -3881,4 +3881,7 @@ skb_set_hash(struct sk_buff *skb, __u32 hash, __always_unused int type) #define HAVE_VF_MIN_MAX_TXRATE 1 #endif /* >= 3.16.0 */ +#if ( LINUX_VERSION_CODE >= KERNEL_VERSION(3,19,0) ) +#define HAVE_NDO_FDB_ADD_VID +#endif /* >= 3.19.0 */ #endif /* _KCOMPAT_H_ */ -- 1.7.4.1
[dpdk-dev] [PATCH v2 6/6] rte_sched: eliminate floating point in calculating byte clock
> -Original Message-
> From: Stephen Hemminger [mailto:stephen at networkplumber.org]
> Sent: Tuesday, March 10, 2015 4:14 PM
> To: Dumitrescu, Cristian
> Cc: dev at dpdk.org; Stephen Hemminger; Stephen Hemminger
> Subject: [PATCH v2 6/6] rte_sched: eliminate floating point in calculating
> byte
> clock
>
> From: Stephen Hemminger
>
> The old code was doing a floating point divide for each rte_dequeue()
> which is very expensive. Change to using fixed point scaled math instead.
> This improved performance from 5Gbit/sec to 10 Gbit/sec
>
> Signed-off-by: Stephen Hemminger
> ---
> v2 -- no changes
> despite objections, the performance observation is real
> on Intel(R) Core(TM) i7-3770 CPU
>
> lib/librte_sched/rte_sched.c | 14 ++
> 1 file changed, 10 insertions(+), 4 deletions(-)
>
> diff --git a/lib/librte_sched/rte_sched.c b/lib/librte_sched/rte_sched.c
> index 74d0e0a..522a647 100644
> --- a/lib/librte_sched/rte_sched.c
> +++ b/lib/librte_sched/rte_sched.c
> @@ -102,6 +102,9 @@
>
> #define RTE_SCHED_BMP_POS_INVALID UINT32_MAX
>
> +/* For cycles_per_byte calculation */
> +#define RTE_SCHED_TIME_SHIFT 20
> +
> struct rte_sched_subport {
> /* Token bucket (TB) */
> uint64_t tb_time; /* time of last update */
> @@ -239,7 +242,7 @@ struct rte_sched_port {
> uint64_t time_cpu_cycles; /* Current CPU time measured in CPU
> cyles */
> uint64_t time_cpu_bytes; /* Current CPU time measured in bytes
> */
> uint64_t time;/* Current NIC TX time measured in bytes
> */
> - double cycles_per_byte; /* CPU cycles per byte */
> + uint32_t cycles_per_byte; /* CPU cycles per byte (scaled) */
>
> /* Scheduling loop detection */
> uint32_t pipe_loop;
> @@ -657,7 +660,9 @@ rte_sched_port_config(struct
> rte_sched_port_params *params)
> port->time_cpu_cycles = rte_get_tsc_cycles();
> port->time_cpu_bytes = 0;
> port->time = 0;
> - port->cycles_per_byte = ((double) rte_get_tsc_hz()) / ((double)
> params->rate);
> +
> + port->cycles_per_byte = (rte_get_tsc_hz() <<
> RTE_SCHED_TIME_SHIFT)
> + / params->rate;
>
> /* Scheduling loop detection */
> port->pipe_loop = RTE_SCHED_PIPE_INVALID;
> @@ -2126,11 +2131,12 @@ rte_sched_port_time_resync(struct
> rte_sched_port *port)
> {
> uint64_t cycles = rte_get_tsc_cycles();
> uint64_t cycles_diff = cycles - port->time_cpu_cycles;
> - double bytes_diff = ((double) cycles_diff) / port->cycles_per_byte;
> + uint64_t bytes_diff = (cycles_diff << RTE_SCHED_TIME_SHIFT)
> + / port->cycles_per_byte;
>
> /* Advance port time */
> port->time_cpu_cycles = cycles;
> - port->time_cpu_bytes += (uint64_t) bytes_diff;
> + port->time_cpu_bytes += bytes_diff;
> if (port->time < port->time_cpu_bytes) {
> port->time = port->time_cpu_bytes;
> }
> --
> 2.1.4
Stephen,
We agreed during the previous round to look at 64-bit multiplication option,
but looks like this patch is identical to the previous one. Did you meet any
issues in implementing this approach? As stated before, I do not think this is
the best solution for the reasons listed previously, and this part of the code
is too sensitive to take the risk.
Since Thomas indicated these patches will be considered for 2.1 rather than 2.0
release, it looks like we have some time to refine this work. I would reiterate
the same proposal that Thomas made: re-submit the patches where we have
consensus, and keep this one out for the moment; you and me can sync up offline
and come back with an implementation proposal that would hopefully address the
previous concerns for 2.1 release. Would this work for you?
Thank you for your work and for your understanding!
Regards,
Cristian
[dpdk-dev] [PATCH] ethdev: additional parameter in RX callback
On Thu, Mar 12, 2015 at 04:54:27PM +, John McNamara wrote: > > This patch is a minor extension to the recent patchset for RX/TX callbacks > based on feedback from users implementing solutions based on it. > > The patch adds a new parameter to the RX callback to pass in the number of > available RX packets in addition to the number of dequeued packets. > This provides the RX callback functions with additional information > that can be used to decide how packets from a burst are handled. > > The TX callback doesn't require this additional parameter so the RX > and TX callbacks no longer have the same function parameters. As such > the single RX/TX callback has been refactored into two separate callbacks. > > Since this is an API change we hope it can be included in 2.0.0 to avoid > changing the API in a subsequent release. > > > John McNamara (1): > ethdev: added additional packet count parameter to RX callbacks > > examples/rxtx_callbacks/main.c |3 +- > lib/librte_ether/rte_ethdev.c |8 ++-- > lib/librte_ether/rte_ethdev.h | 74 > ++-- > 3 files changed, 54 insertions(+), 31 deletions(-) > > -- > 1.7.4.1 > > Well, we're well past the new feature phase of this cycle, so I would say NACK. I would also suggest that you don't need to modify ABI to accomodate this feature. Instead just document the pkts array to be terminated by a reserved value, so that the callback can determine its size dynamically. You could alternatively create a new api call that allows you to retrieve that information from the context of the callback. Neil
[dpdk-dev] [PATCH v2 5/6] rte_sched: allow reading statistics without clearing
> -Original Message-
> From: Stephen Hemminger [mailto:stephen at networkplumber.org]
> Sent: Tuesday, March 10, 2015 4:14 PM
> To: Dumitrescu, Cristian
> Cc: dev at dpdk.org; Stephen Hemminger; Stephen Hemminger
> Subject: [PATCH v2 5/6] rte_sched: allow reading statistics without clearing
>
> From: Stephen Hemminger
>
> The rte_sched statistics API should allow reading statistics without
> clearing. Make auto-clear optional.
>
> Signed-off-by: Stephen Hemminger
> ---
> v2 -- change to flag for clear from separate read/clear
>
> app/test/test_sched.c| 4 ++--
> examples/qos_sched/stats.c | 16 +++-
> lib/librte_sched/rte_sched.c | 44 ++-
> -
> lib/librte_sched/rte_sched.h | 18 ++
> 4 files changed, 45 insertions(+), 37 deletions(-)
>
> diff --git a/app/test/test_sched.c b/app/test/test_sched.c
> index 60c62de..3c9c9e3 100644
> --- a/app/test/test_sched.c
> +++ b/app/test/test_sched.c
> @@ -208,13 +208,13 @@ test_sched(void)
>
> struct rte_sched_subport_stats subport_stats;
> uint32_t tc_ov;
> - rte_sched_subport_read_stats(port, SUBPORT, &subport_stats,
> &tc_ov);
> + rte_sched_subport_read_stats(port, SUBPORT, &subport_stats,
> &tc_ov, 1);
> #if 0
> TEST_ASSERT_EQUAL(subport_stats.n_pkts_tc[TC-1], 10, "Wrong
> subport stats\n");
> #endif
> struct rte_sched_queue_stats queue_stats;
> uint16_t qlen;
> - rte_sched_queue_read_stats(port, QUEUE, &queue_stats, &qlen);
> + rte_sched_queue_read_stats(port, QUEUE, &queue_stats, &qlen,
> 1);
> #if 0
> TEST_ASSERT_EQUAL(queue_stats.n_pkts, 10, "Wrong queue
> stats\n");
> #endif
> diff --git a/examples/qos_sched/stats.c b/examples/qos_sched/stats.c
> index b4db7b5..88caa54 100644
> --- a/examples/qos_sched/stats.c
> +++ b/examples/qos_sched/stats.c
> @@ -61,7 +61,7 @@ qavg_q(uint8_t port_id, uint32_t subport_id, uint32_t
> pipe_id, uint8_t tc, uint8
> average = 0;
>
> for (count = 0; count < qavg_ntimes; count++) {
> -rte_sched_queue_read_stats(port, queue_id, &stats, &qlen);
> +rte_sched_queue_read_stats(port, queue_id, &stats, &qlen, 1);
> average += qlen;
> usleep(qavg_period);
> }
> @@ -99,7 +99,9 @@ qavg_tcpipe(uint8_t port_id, uint32_t subport_id,
> uint32_t pipe_id, uint8_t tc)
> for (count = 0; count < qavg_ntimes; count++) {
> part_average = 0;
> for (i = 0; i < RTE_SCHED_QUEUES_PER_TRAFFIC_CLASS; i++) {
> -rte_sched_queue_read_stats(port, queue_id + (tc *
> RTE_SCHED_QUEUES_PER_TRAFFIC_CLASS + i), &stats, &qlen);
> +rte_sched_queue_read_stats(port,
> +queue_id + (tc *
> RTE_SCHED_QUEUES_PER_TRAFFIC_CLASS + i),
> +&stats, &qlen, 1);
> part_average += qlen;
> }
> average += part_average /
> RTE_SCHED_QUEUES_PER_TRAFFIC_CLASS;
> @@ -138,7 +140,8 @@ qavg_pipe(uint8_t port_id, uint32_t subport_id,
> uint32_t pipe_id)
> for (count = 0; count < qavg_ntimes; count++) {
> part_average = 0;
> for (i = 0; i < RTE_SCHED_TRAFFIC_CLASSES_PER_PIPE *
> RTE_SCHED_QUEUES_PER_TRAFFIC_CLASS; i++) {
> -rte_sched_queue_read_stats(port, queue_id + i,
> &stats,
> &qlen);
> +rte_sched_queue_read_stats(port, queue_id + i,
> +&stats, &qlen, 1);
> part_average += qlen;
> }
> average += part_average /
> (RTE_SCHED_TRAFFIC_CLASSES_PER_PIPE *
> RTE_SCHED_QUEUES_PER_TRAFFIC_CLASS);
> @@ -178,7 +181,9 @@ qavg_tcsubport(uint8_t port_id, uint32_t subport_id,
> uint8_t tc)
> queue_id = RTE_SCHED_TRAFFIC_CLASSES_PER_PIPE *
> RTE_SCHED_QUEUES_PER_TRAFFIC_CLASS * (subport_id *
> port_params.n_pipes_per_subport + i);
>
> for (j = 0; j < RTE_SCHED_QUEUES_PER_TRAFFIC_CLASS;
> j++) {
> -rte_sched_queue_read_stats(port, queue_id +
> (tc *
> RTE_SCHED_QUEUES_PER_TRAFFIC_CLASS + j), &stats, &qlen);
> +rte_sched_queue_read_stats(port,
> +queue_id + (tc *
> RTE_SCHED_QUEUES_PER_TRAFFIC_CLASS + j),
> +&stats, &qlen, 1);
> part_average += qlen;
> }
> }
> @@ -220,7 +225,8 @@ qavg_subport(uint8_t port_id, uint32_t subport_id)
> queue_id = RTE_SCHED_TRAFFIC_CLASSES_PER_PIPE *
> RTE_SCHED_QUEUES_PER_TRAFFIC_CLASS * (subport_id *
> port_params.n_pipes_per_subport + i);
>
> for (j = 0; j < RTE_SCHED_TRAFFIC_CLASS
[dpdk-dev] [PATCH v3 0/3]: bug fixes in the ixgbe PF PMD Rx flow
This series contains some bug fixes that were found during my work on the ixgbe LRO patches. Sending this series separately on Thomas request so that it may be integrated into the 2.0 release. New in v3: - Adjusted to the new structs naming in the master. - Fixed rx_bulk_alloc_allowed and rx_vec_allowed initialization: - Don't set them to FALSE in rte_eth_dev_stop() flow - the following rte_eth_dev_start() will need them. - Reset them to TRUE in rte_eth_dev_configure() and not in a probe() flow. This will ensure the proper behaviour if port is re-configured. - Rename: - ixgbe_rx_vec_condition_check() -> ixgbe_rx_vec_dev_conf_condition_check() - set_rx_function() -> ixgbe_set_rx_function() - Clean up the logic in ixgbe_set_rx_function(). - Define stubs with __attribute__((weak)) instead of using #ifdef's. - Styling: beautify ixgbe_rxtx.h a bit. New in v2: - Fixed a compilation failure. Vlad Zolotarov (3): ixgbe: Use the rte_le_to_cpu_xx()/rte_cpu_to_le_xx() when reading/setting HW ring descriptor fields ixgbe: Bug fix: Properly configure Rx CRC stripping for x540 devices ixgbe: Unify the rx_pkt_bulk callback initialization lib/librte_pmd_ixgbe/ixgbe/ixgbe_type.h | 2 + lib/librte_pmd_ixgbe/ixgbe_ethdev.c | 13 +- lib/librte_pmd_ixgbe/ixgbe_rxtx.c | 216 +--- lib/librte_pmd_ixgbe/ixgbe_rxtx.h | 28 - lib/librte_pmd_ixgbe/ixgbe_rxtx_vec.c | 2 +- 5 files changed, 183 insertions(+), 78 deletions(-) -- 2.1.0
[dpdk-dev] [PATCH v3 1/3] ixgbe: Use the rte_le_to_cpu_xx()/rte_cpu_to_le_xx() when reading/setting HW ring descriptor fields
Fixed the above in ixgbe_rx_alloc_bufs() and in ixgbe_recv_scattered_pkts().
Signed-off-by: Vlad Zolotarov
---
lib/librte_pmd_ixgbe/ixgbe_rxtx.c | 13 +++--
1 file changed, 7 insertions(+), 6 deletions(-)
diff --git a/lib/librte_pmd_ixgbe/ixgbe_rxtx.c
b/lib/librte_pmd_ixgbe/ixgbe_rxtx.c
index 96c4b98..f7c081f 100644
--- a/lib/librte_pmd_ixgbe/ixgbe_rxtx.c
+++ b/lib/librte_pmd_ixgbe/ixgbe_rxtx.c
@@ -1027,7 +1027,7 @@ ixgbe_rx_alloc_bufs(struct ixgbe_rx_queue *rxq)
struct ixgbe_rx_entry *rxep;
struct rte_mbuf *mb;
uint16_t alloc_idx;
- uint64_t dma_addr;
+ __le64 dma_addr;
int diag, i;
/* allocate buffers in bulk directly into the S/W ring */
@@ -1050,7 +1050,7 @@ ixgbe_rx_alloc_bufs(struct ixgbe_rx_queue *rxq)
mb->port = rxq->port_id;
/* populate the descriptors */
- dma_addr = (uint64_t)mb->buf_physaddr + RTE_PKTMBUF_HEADROOM;
+ dma_addr = rte_cpu_to_le_64(RTE_MBUF_DATA_DMA_ADDR_DEFAULT(mb));
rxdp[i].read.hdr_addr = dma_addr;
rxdp[i].read.pkt_addr = dma_addr;
}
@@ -1558,13 +1558,14 @@ ixgbe_recv_scattered_pkts(void *rx_queue, struct
rte_mbuf **rx_pkts,
first_seg->ol_flags = pkt_flags;
if (likely(pkt_flags & PKT_RX_RSS_HASH))
- first_seg->hash.rss = rxd.wb.lower.hi_dword.rss;
+ first_seg->hash.rss =
+ rte_le_to_cpu_32(rxd.wb.lower.hi_dword.rss);
else if (pkt_flags & PKT_RX_FDIR) {
first_seg->hash.fdir.hash =
- (uint16_t)((rxd.wb.lower.hi_dword.csum_ip.csum)
- & IXGBE_ATR_HASH_MASK);
+ rte_le_to_cpu_16(rxd.wb.lower.hi_dword.csum_ip.csum)
+ & IXGBE_ATR_HASH_MASK;
first_seg->hash.fdir.id =
- rxd.wb.lower.hi_dword.csum_ip.ip_id;
+ rte_le_to_cpu_16(rxd.wb.lower.hi_dword.csum_ip.ip_id);
}
/* Prefetch data of first segment, if configured to do so. */
--
2.1.0
[dpdk-dev] [PATCH v3 2/3] ixgbe: Bug fix: Properly configure Rx CRC stripping for x540 devices
According to x540 spec chapter 8.2.4.8.9 CRCSTRIP field of RDRXCTL should
be configured to the same value as HLREG0.RXCRCSTRP.
Clearing the RDRXCTL.RSCFRSTSIZE field for x540 is not required by the spec
but seems harmless.
Signed-off-by: Vlad Zolotarov
---
lib/librte_pmd_ixgbe/ixgbe_rxtx.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/lib/librte_pmd_ixgbe/ixgbe_rxtx.c
b/lib/librte_pmd_ixgbe/ixgbe_rxtx.c
index f7c081f..92be61e 100644
--- a/lib/librte_pmd_ixgbe/ixgbe_rxtx.c
+++ b/lib/librte_pmd_ixgbe/ixgbe_rxtx.c
@@ -3680,7 +3680,8 @@ ixgbe_dev_rx_init(struct rte_eth_dev *dev)
IXGBE_WRITE_REG(hw, IXGBE_RXCSUM, rxcsum);
- if (hw->mac.type == ixgbe_mac_82599EB) {
+ if (hw->mac.type == ixgbe_mac_82599EB ||
+ hw->mac.type == ixgbe_mac_X540) {
rdrxctl = IXGBE_READ_REG(hw, IXGBE_RDRXCTL);
if (dev->data->dev_conf.rxmode.hw_strip_crc)
rdrxctl |= IXGBE_RDRXCTL_CRCSTRIP;
--
2.1.0
[dpdk-dev] [PATCH v3 3/3] ixgbe: Unify the rx_pkt_bulk callback initialization
- Set the callback in a single function that is called from
ixgbe_dev_rx_init() for a primary process and from eth_ixgbe_dev_init()
for a secondary processes. This is instead of multiple, hard to track
places.
- Added ixgbe_hw.rx_bulk_alloc_allowed - see ixgbe_hw.rx_vec_allowed
description below.
- Added ixgbe_hw.rx_vec_allowed: like with Bulk Allocation, Vector Rx is
enabled or disabled on a per-port level. All queues have to meet the
appropriate
preconditions and if any of them doesn't - the feature has to be disabled.
Therefore ixgbe_hw.rx_vec_allowed will be updated during each queues
configuration
(rte_eth_rx_queue_setup()) and then used in rte_eth_dev_start() to
configure the
appropriate callbacks. The same happens with ixgbe_hw.rx_vec_allowed in a
Bulk Allocation
context.
- Bugs fixed:
- Vector scattered packets callback was called regardless the appropriate
preconditions:
- Vector Rx specific preconditions.
- Bulk Allocation preconditions.
- Vector Rx was enabled/disabled according to the last queue setting and
not
based on all queues setting (which may be different for each queue).
Signed-off-by: Vlad Zolotarov
---
New in v3:
- Adjusted to the new structs naming in the master.
- Fixed rx_bulk_alloc_allowed and rx_vec_allowed initialization:
- Don't set them to FALSE in rte_eth_dev_stop() flow - the following
rte_eth_dev_start() will need them.
- Reset them to TRUE in rte_eth_dev_configure() and not in a probe() flow.
This will ensure the proper behaviour if port is re-configured.
- Rename:
- ixgbe_rx_vec_condition_check() ->
ixgbe_rx_vec_dev_conf_condition_check()
- set_rx_function() -> ixgbe_set_rx_function()
- Clean up the logic in ixgbe_set_rx_function().
- Define stubs with __attribute__((weak)) instead of using #ifdef's.
- Styling: beautify ixgbe_rxtx.h a bit.
New in v2:
- Fixed an artifact caused by git rebasing that broke the compilation.
---
lib/librte_pmd_ixgbe/ixgbe/ixgbe_type.h | 2 +
lib/librte_pmd_ixgbe/ixgbe_ethdev.c | 13 ++-
lib/librte_pmd_ixgbe/ixgbe_rxtx.c | 200 ++--
lib/librte_pmd_ixgbe/ixgbe_rxtx.h | 28 -
lib/librte_pmd_ixgbe/ixgbe_rxtx_vec.c | 2 +-
5 files changed, 174 insertions(+), 71 deletions(-)
diff --git a/lib/librte_pmd_ixgbe/ixgbe/ixgbe_type.h
b/lib/librte_pmd_ixgbe/ixgbe/ixgbe_type.h
index c67d462..9a66370 100644
--- a/lib/librte_pmd_ixgbe/ixgbe/ixgbe_type.h
+++ b/lib/librte_pmd_ixgbe/ixgbe/ixgbe_type.h
@@ -3657,6 +3657,8 @@ struct ixgbe_hw {
bool force_full_reset;
bool allow_unsupported_sfp;
bool wol_enabled;
+ bool rx_bulk_alloc_allowed;
+ bool rx_vec_allowed;
};
#define ixgbe_call_func(hw, func, params, error) \
diff --git a/lib/librte_pmd_ixgbe/ixgbe_ethdev.c
b/lib/librte_pmd_ixgbe/ixgbe_ethdev.c
index e4edb01..92d75db 100644
--- a/lib/librte_pmd_ixgbe/ixgbe_ethdev.c
+++ b/lib/librte_pmd_ixgbe/ixgbe_ethdev.c
@@ -756,8 +756,8 @@ eth_ixgbe_dev_init(struct rte_eth_dev *eth_dev)
"Using default TX function.");
}
- if (eth_dev->data->scattered_rx)
- eth_dev->rx_pkt_burst = ixgbe_recv_scattered_pkts;
+ ixgbe_set_rx_function(eth_dev);
+
return 0;
}
pci_dev = eth_dev->pci_dev;
@@ -1429,12 +1429,21 @@ ixgbe_dev_configure(struct rte_eth_dev *dev)
{
struct ixgbe_interrupt *intr =
IXGBE_DEV_PRIVATE_TO_INTR(dev->data->dev_private);
+ struct ixgbe_hw *hw =
+ IXGBE_DEV_PRIVATE_TO_HW(dev->data->dev_private);
PMD_INIT_FUNC_TRACE();
/* set flag to update link status after init */
intr->flags |= IXGBE_FLAG_NEED_LINK_UPDATE;
+ /*
+* Initialize to TRUE. If any of Rx queues doesn't meet the bulk
+* allocation or vector Rx preconditions we will reset it.
+*/
+ hw->rx_bulk_alloc_allowed = true;
+ hw->rx_vec_allowed = true;
+
return 0;
}
diff --git a/lib/librte_pmd_ixgbe/ixgbe_rxtx.c
b/lib/librte_pmd_ixgbe/ixgbe_rxtx.c
index 92be61e..5b1ba82 100644
--- a/lib/librte_pmd_ixgbe/ixgbe_rxtx.c
+++ b/lib/librte_pmd_ixgbe/ixgbe_rxtx.c
@@ -2075,12 +2075,12 @@ check_rx_burst_bulk_alloc_preconditions(__rte_unused
struct ixgbe_rx_queue *rxq)
/* Reset dynamic ixgbe_rx_queue fields back to defaults */
static void
-ixgbe_reset_rx_queue(struct ixgbe_rx_queue *rxq)
+ixgbe_reset_rx_queue(struct ixgbe_hw *hw, struct ixgbe_rx_queue *rxq)
{
static const union ixgbe_adv_rx_desc zeroed_desc = { .read = {
.pkt_addr = 0}};
unsigned i;
- uint16_t len;
+ uint16_t len = rxq->nb_rx_desc;
/*
* By default, the Rx queue setup function allocates enough memory for
@@ -2092,14 +2092,9 @@ ixgbe_reset_rx_queue(struct ixgbe_
[dpdk-dev] [PATCH v2 5/6] rte_sched: allow reading statistics without clearing
On Thu, 12 Mar 2015 19:28:11 +
"Dumitrescu, Cristian" wrote:
> Hi Stephen,
>
> Thank you for adding flexibility over clearing the stats or not.
>
> I have one concern though: why change the stats read API to add a clear
> parameter rather than keep prototype for the stats functions unchanged and
> add the flag as part of the port creation parameters in struct
> rte_sched_port_params? This parameter could be saved into the internal struct
> rte_sched_port, which is passed (as a pointer) to the stats read functions.
> In my opinion, this approach is slightly more elegant and it keeps the
> changes to a minimum.
>
> int
> rte_sched_queue_read_stats(struct rte_sched_port *port,
> uint32_t queue_id,
> struct rte_sched_queue_stats *stats,
> uint16_t *qlen)
> {
> ...
> if (port->clear_stats_on_read)
> memset(...);
> }
>
> I made this suggestion during the previous round, but I did not get any
> opinion from you on it yet.
>
> Regards,
> Cristian
I rejected the config parameter idea because I don't like it is inconsistent
with other statistics in DPDK and in related software. There is not a
config parameter that changes what BSD or Linux kernel API does.
The only reason for keeping the read and clear in one operation is
because you like it, and there are somebody might have built code
that expects it.
Changing the function signature is a nice red flag so that people will
notice at change.
[dpdk-dev] [PATCH v2 6/6] rte_sched: eliminate floating point in calculating byte clock
On Thu, 12 Mar 2015 19:06:39 +
"Dumitrescu, Cristian" wrote:
>
>
> > -Original Message-
> > From: Stephen Hemminger [mailto:stephen at networkplumber.org]
> > Sent: Tuesday, March 10, 2015 4:14 PM
> > To: Dumitrescu, Cristian
> > Cc: dev at dpdk.org; Stephen Hemminger; Stephen Hemminger
> > Subject: [PATCH v2 6/6] rte_sched: eliminate floating point in calculating
> > byte
> > clock
> >
> > From: Stephen Hemminger
> >
> > The old code was doing a floating point divide for each rte_dequeue()
> > which is very expensive. Change to using fixed point scaled math instead.
> > This improved performance from 5Gbit/sec to 10 Gbit/sec
> >
> > Signed-off-by: Stephen Hemminger
> > ---
> > v2 -- no changes
> > despite objections, the performance observation is real
> > on Intel(R) Core(TM) i7-3770 CPU
> >
> > lib/librte_sched/rte_sched.c | 14 ++
> > 1 file changed, 10 insertions(+), 4 deletions(-)
> >
> > diff --git a/lib/librte_sched/rte_sched.c b/lib/librte_sched/rte_sched.c
> > index 74d0e0a..522a647 100644
> > --- a/lib/librte_sched/rte_sched.c
> > +++ b/lib/librte_sched/rte_sched.c
> > @@ -102,6 +102,9 @@
> >
> > #define RTE_SCHED_BMP_POS_INVALID UINT32_MAX
> >
> > +/* For cycles_per_byte calculation */
> > +#define RTE_SCHED_TIME_SHIFT 20
> > +
> > struct rte_sched_subport {
> > /* Token bucket (TB) */
> > uint64_t tb_time; /* time of last update */
> > @@ -239,7 +242,7 @@ struct rte_sched_port {
> > uint64_t time_cpu_cycles; /* Current CPU time measured in CPU
> > cyles */
> > uint64_t time_cpu_bytes; /* Current CPU time measured in bytes
> > */
> > uint64_t time;/* Current NIC TX time measured in bytes
> > */
> > - double cycles_per_byte; /* CPU cycles per byte */
> > + uint32_t cycles_per_byte; /* CPU cycles per byte (scaled) */
> >
> > /* Scheduling loop detection */
> > uint32_t pipe_loop;
> > @@ -657,7 +660,9 @@ rte_sched_port_config(struct
> > rte_sched_port_params *params)
> > port->time_cpu_cycles = rte_get_tsc_cycles();
> > port->time_cpu_bytes = 0;
> > port->time = 0;
> > - port->cycles_per_byte = ((double) rte_get_tsc_hz()) / ((double)
> > params->rate);
> > +
> > + port->cycles_per_byte = (rte_get_tsc_hz() <<
> > RTE_SCHED_TIME_SHIFT)
> > + / params->rate;
> >
> > /* Scheduling loop detection */
> > port->pipe_loop = RTE_SCHED_PIPE_INVALID;
> > @@ -2126,11 +2131,12 @@ rte_sched_port_time_resync(struct
> > rte_sched_port *port)
> > {
> > uint64_t cycles = rte_get_tsc_cycles();
> > uint64_t cycles_diff = cycles - port->time_cpu_cycles;
> > - double bytes_diff = ((double) cycles_diff) / port->cycles_per_byte;
> > + uint64_t bytes_diff = (cycles_diff << RTE_SCHED_TIME_SHIFT)
> > + / port->cycles_per_byte;
> >
> > /* Advance port time */
> > port->time_cpu_cycles = cycles;
> > - port->time_cpu_bytes += (uint64_t) bytes_diff;
> > + port->time_cpu_bytes += bytes_diff;
> > if (port->time < port->time_cpu_bytes) {
> > port->time = port->time_cpu_bytes;
> > }
> > --
> > 2.1.4
>
> Stephen,
>
> We agreed during the previous round to look at 64-bit multiplication option,
> but looks like this patch is identical to the previous one. Did you meet any
> issues in implementing this approach? As stated before, I do not think this
> is the best solution for the reasons listed previously, and this part of the
> code is too sensitive to take the risk.
>
> Since Thomas indicated these patches will be considered for 2.1 rather than
> 2.0 release, it looks like we have some time to refine this work. I would
> reiterate the same proposal that Thomas made: re-submit the patches where we
> have consensus, and keep this one out for the moment; you and me can sync up
> offline and come back with an implementation proposal that would hopefully
> address the previous concerns for 2.1 release. Would this work for you?
>
> Thank you for your work and for your understanding!
>
> Regards,
> Cristian
>
I haven't had time to do the analysis to figure out how to keep the maths
in range so that 64 bit multiplicative divide would work. When I first tried
it the issue was that the cycles_diff and bytes_diff would both have to
scaled so that they are 32 bit values.
Should be possible, might get back to it (or someone else might take up
the cause).
[dpdk-dev] [PATCH] ethdev: additional parameter in RX callback
> -Original Message- > From: Neil Horman [mailto:nhorman at tuxdriver.com] > Sent: Thursday, March 12, 2015 7:16 PM > To: Mcnamara, John > Cc: dev at dpdk.org > Subject: Re: [dpdk-dev] [PATCH] ethdev: additional parameter in RX > callback > > > Well, we're well past the new feature phase of this cycle, so I would say > NACK. Hi Neil, This is more of a bug fix than a feature request. In retrospect it was a mistake not to have the same function declaration in the callbacks as in the rte_eth_XX_burst functions since it was likely that the callbacks would need access to the same information. This patch is trying to fix that error before it is set in stone. John
[dpdk-dev] [PATCH] librte_pmd_bond: remove memory alloc for rte_pci_driver
eth_driver already contains rte_pci_driver data structure.
Allocating rte_pci_driver without referencing it after bond
creation causes memory leakage.
---
lib/librte_pmd_bond/rte_eth_bond_api.c | 11 ++-
1 file changed, 2 insertions(+), 9 deletions(-)
diff --git a/lib/librte_pmd_bond/rte_eth_bond_api.c
b/lib/librte_pmd_bond/rte_eth_bond_api.c
index 903b7c3..13f3941 100644
--- a/lib/librte_pmd_bond/rte_eth_bond_api.c
+++ b/lib/librte_pmd_bond/rte_eth_bond_api.c
@@ -230,11 +230,8 @@ rte_eth_bond_create(const char *name, uint8_t mode,
uint8_t socket_id)
goto err;
}
- pci_drv = rte_zmalloc_socket(name, sizeof(*pci_drv), 0, socket_id);
- if (pci_drv == NULL) {
- RTE_BOND_LOG(ERR, "Unable to malloc pci_drv on socket");
- goto err;
- }
+ pci_drv = ð_drv->pci_drv;
+
pci_id_table = rte_zmalloc_socket(name, sizeof(*pci_id_table), 0,
socket_id);
if (pci_id_table == NULL) {
RTE_BOND_LOG(ERR, "Unable to malloc pci_id_table on socket");
@@ -266,9 +263,7 @@ rte_eth_bond_create(const char *name, uint8_t mode, uint8_t
socket_id)
pci_dev->numa_node = socket_id;
pci_drv->name = driver_name;
- eth_drv->pci_drv = (struct rte_pci_driver)(*pci_drv);
eth_dev->driver = eth_drv;
-
eth_dev->data->dev_private = internals;
eth_dev->data->nb_rx_queues = (uint16_t)1;
eth_dev->data->nb_tx_queues = (uint16_t)1;
@@ -325,8 +320,6 @@ rte_eth_bond_create(const char *name, uint8_t mode, uint8_t
socket_id)
err:
if (pci_dev)
rte_free(pci_dev);
- if (pci_drv)
- rte_free(pci_drv);
if (pci_id_table)
rte_free(pci_id_table);
if (eth_drv)
--
1.9.1
[dpdk-dev] Appropriate DPDK data structures for TCP sockets
On Mon, Feb 23, 2015 at 11:51:46PM +0200, Avi Kivity wrote: > https://github.com/cloudius-systems/seastar Hi Avi and others, My code unintentionally ended up looking somewhat like a C version of your seastar C++ code, even though I didn't really look at yours too much when I coded mine as it was using a lot of hardcore C++ features I really don't have any clue about. :) Someday maybe we can all do a bake-off of tests of DPDK TCPs and Host TCPs and see what kind of stability, features, and performance we can get. I didn't use anything too crazy or high performance just yet... just rte_hash and some spinlocks... but I'm keeping all the collective advice in mind to figure out how I'm going to make all this stuff work right soon. Matthew.
