[dpdk-dev] pci_unbind.py failure
Hi Jyotiswarup, Based on my experience and little knowledge on the code, rte_pmd_init_all registers your poll-mode driver (em in this case) and sets the device initialization function. rte_eal_pci_probe will then probe for pci devices and then initialize their corresponding drivers. These initialization will increment the number of usable ports (e.g. e1000) and also the value that is returned by rte_eth_dev_count. I think that without calling rte_pmd_init_all, rte_eal_pci_probe will fail unless you have a specialize routine that sets the initialization function of your driver. Cheers, Pepe On Wed, Nov 13, 2013 at 11:00 PM, Jyotiswarup Raiturkar < jyotisr5 at googlemail.com> wrote: > Hi all > > Hi > > I'm using vmware player , i've using ethernet1.virtualDev = "e1000" > > Yes I do rte_eal_init() and then rte_eal_pci_probe(). I'm not using > rte_pmd_init_all() ; not sure what it does.. Also FYI i'm using DPDK > 1.4.1r0 > > I've finally got real hardware too, will try on the same and update the > thread.. > > Regards > Jyoti > > > On Wed, Nov 13, 2013 at 8:27 PM, Vladimir Medvedkin >wrote: > > > Hi all, > > > > I have faced with similar problem on my real environment with 82599 nic. > > Looks like in some cases rte_eth_dev_count() return 0 instead of real > > number of igb_uio binded ports. After restarting app several times > > rte_eth_dev_count() return real number of ports and continue execute > > normaly. > > > > > > 2013/11/13 Daniel Kaminsky > > > > > Hi Jyotiswarup, > > > > > > Did you initialized all the relevant parts before > > > (rte_eal_init(), rte_pmd_init_all() and don't forget > > rte_eal_pci_probe())? > > > > > > Regards, > > > Daniel > > > > > > > > > On Wed, Nov 13, 2013 at 1:27 PM, Jose Gavine Cueto > > >wrote: > > > > > > > Hi, > > > > > > > > How are you using it? I've successfully used it with vbox. > > > > > > > > Cheers > > > > On Nov 13, 2013 7:17 PM, "Jyotiswarup Raiturkar" < > > > jyotisr5 at googlemail.com> > > > > wrote: > > > > > > > > > hi > > > > > > > > > > I got my application running inside a vm (vmplayer) where the VM > > > > emulates a > > > > > e1000 NIC (82545EM). But rte_eth_dev_count() seems to return 0.. > From > > > the > > > > > website looks like it's a supported NIC . My lspci and pci_ubind > > status > > > > is > > > > > below. Any pointers? > > > > > > > > > > # ./tools/pci_unbind.py --status > > > > > > > > > > Network devices using IGB_UIO driver > > > > > > > > > > :02:06.0 '82545EM Gigabit Ethernet Controller (Copper)' > > drv=igb_uio > > > > > unused=e1000 > > > > > > > > > > Network devices using kernel driver > > > > > === > > > > > :02:01.0 '79c970 [PCnet32 LANCE]' if=eth1 drv=pcnet32 unused= > > > > *Active* > > > > > > > > > > Other network devices > > > > > = > > > > > > > > > > > > > > > > > > > > # lspci -vt > > > > > -[:00]-+-00.0 Intel Corporation 440BX/ZX/DX - 82443BX/ZX/DX > Host > > > > > bridge > > > > >+-01.0-[01]-- > > > > >+-07.0 Intel Corporation 82371AB/EB/MB PIIX4 ISA > > > > >+-07.1 Intel Corporation 82371AB/EB/MB PIIX4 IDE > > > > >+-07.3 Intel Corporation 82371AB/EB/MB PIIX4 ACPI > > > > >+-07.7 VMware Virtual Machine Communication Interface > > > > >+-0f.0 VMware SVGA II Adapter > > > > >+-10.0 LSI Logic / Symbios Logic 53c1030 PCI-X > Fusion-MPT > > > > Dual > > > > > Ultra320 SCSI > > > > >+-11.0-[02]--+-00.0 VMware USB1.1 UHCI Controller > > > > >|+-01.0 Advanced Micro Devices [AMD] 79c970 > > > > > [PCnet32 LANCE] > > > > >|+-02.0 Ensoniq ES1371 [AudioPCI-97] > > > > >|+-03.0 VMware USB2 EHCI Controller > > > > >|+-05.0 VMware Device 07e0 > > > > >|\-06.0 Intel Corporation 82545EM Gigabit > > > > Ethernet > > > > > Controller (Copper) > > > > >+-15.0-[03]-- > > > > >+-15.1-[04]-- > > > > >+-15.2-[05]-- > > > > >+-15.3-[06]-- > > > > >+-15.4-[07]-- > > > > >+-15.5-[08]-- > > > > >+-15.6-[09]-- > > > > >+-15.7-[0a]-- > > > > >+-16.0-[0b]-- > > > > >+-16.1-[0c]-- > > > > >+-16.2-[0d]-- > > > > >+-16.3-[0e]-- > > > > >+-16.4-[0f]-- > > > > >+-16.5-[10]-- > > > > >+-16.6-[11]-- > > > > >+-16.7-[12]-- > > > > >+-17.0-[13]-- > > > > >+-17.1-[14]-- > > > > >+-17.2-[15]-- > > > > >+-17.3-[16]-- > > > > >+-17.4-[17]-- > > > > >+-17.5-[18]-- > > > > >+-17.6-[19]-- > > > > >+-17.7-[1a]-- > > > > >+-18.0-[1b]-- > > > > >+-18.1-[1c]-- > > > > >+-18.2-[1d]-- > > > > >+-18.3-[1e]
[dpdk-dev] Cannot run l3fwd with the problem of " nb_tx_queues"
Hi Everyone: I meet a problem with running the DPDK1.3 example l3fwd in the VMware. I have two 82545EM NIC to run the l3fwd and the OS is Cent6.4. But l3fwd cannot be configured port correctly .I start the l3fwd with following command: ./build/l3fwd -c 0x3 -n 2 -- -p 0x3 --config="(0,0,0),(1,0,1)" However, I get following errors with debug info: Initializing port 0 ... Creating queues: nb_rxq=1 nb_txq=2... PMD: rte_eth_dev_configure: ethdev port_id=0 nb_tx_queues=2 > 1 EAL: Error - exiting with code: 1 Cause: Cannot configure device: err=-22, port=0 And I have known that the reason for this problem produced in the **rte_eth_dev_configure function**. However, I do not kown how to fix it. I would really appreciate If someone could provide some help to solve this issue.
[dpdk-dev] could nvm_checksum_validate Error may cause non-functioning of DPDK.
Hi, I am trying to work on DPDK-1.5.0r0 with intel i350 dual port NIC with intel i3 processor on Fedora 18 (kernel-3.11.4). But after building dpdk when i try to run testpmd application its not working with some error as... "EAL: Ask a virtual area of 0x2097152 bytes EAL: Virtual area found at 0x7faddaa0 (size = 0x20) EAL: Requesting 128 pages of size 2MB from socket 0 EAL: TSC frequency is ~330 KHz EAL: Master core 0 is ready (tid=ec6e3880) EAL: Core 1 is ready (tid=da1fe700) EAL: PCI device :01:00.0 on NUMA socket -1 EAL: probe driver: 8086:1521 rte_igb_pmd EAL: PCI memory mapped at 0x7fadec07e000 EAL: PCI memory mapped at 0x7fadec6f1000 EAL: PCI device :01:00.1 on NUMA socket -1 EAL: probe driver: 8086:1521 rte_igb_pmd EAL: PCI memory mapped at 0x7fadeb90 EAL: PCI memory mapped at 0x7fadec6ed000 EAL: Error - exiting with code: 1 Cause: No probed ethernet devices - check that CONFIG_RTE_LIBRTE_IGB_PMD=y and that CONFIG_RTE_LIBRTE_EM_PMD=y and that CONFIG_RTE_LIBRTE_IXGBE_PMD=y in your configuration file " When i searched for kernel logs, it shows me as NVM_CHECKSUM_INVALID for this NIC. As per my knowledge its a EEPROM error in NIC card. So to work around it i changed the driver code given in dpdk-1.5.0r0 in dpdk-1.5.0r1/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_nvm.c to just return the success messages whenever anyone ask for nvm_checksum_validate. But still its not working & showing the above logs. Please help. Thanx.
[dpdk-dev] Cannot run l3fwd with the problem of " nb_tx_queues"
Hi, Your nic (port 0) is capable of 1 txq only see error: " port_id=0 nb_tx_queues=2 > 1" I think that caused the error. Cheers, Pepe On Thu, Nov 14, 2013 at 1:21 PM, yuxuan zhang wrote: > Hi Everyone: > > I meet a problem with running the DPDK1.3 example l3fwd in the VMware. I > have two 82545EM NIC to run the l3fwd and the OS is Cent6.4. But l3fwd > cannot be configured port correctly .I start the l3fwd with following > command: > > ./build/l3fwd -c 0x3 -n 2 -- -p 0x3 --config="(0,0,0),(1,0,1)" > > However, I get following errors with debug info: > > Initializing port 0 ... Creating queues: nb_rxq=1 nb_txq=2... PMD: > rte_eth_dev_configure: ethdev port_id=0 nb_tx_queues=2 > 1 > EAL: Error - exiting with code: 1 > Cause: Cannot configure device: err=-22, port=0 > > And I have known that the reason for this problem produced in the > **rte_eth_dev_configure function**. However, I do not kown how to fix it. > > I would really appreciate If someone could provide some help to solve this > issue. > -- To stop learning is like to stop loving.
[dpdk-dev] could nvm_checksum_validate Error may cause non-functioning of DPDK.
14/11/2013 06:33, ankit kumar : > I am trying to work on DPDK-1.5.0r0 with intel i350 dual port NIC [...] > EAL: Error - exiting with code: 1 > Cause: No probed ethernet devices - check that > CONFIG_RTE_LIBRTE_IGB_PMD=y and that CONFIG_RTE_LIBRTE_EM_PMD=y and > that CONFIG_RTE_LIBRTE_IXGBE_PMD=y in your configuration file [...] > When i searched for kernel logs, it shows me as NVM_CHECKSUM_INVALID > for this NIC. As per my knowledge its a EEPROM error in NIC card. So > to work around it i changed the driver code given in dpdk-1.5.0r0 in > dpdk-1.5.0r1/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_nvm.c The drivers in KNI are only used for ethtool operations with kernel. The PMD file you are looking for is lib/librte_pmd_e1000/e1000/e1000_nvm.c -- Thomas
