Rare project developers

[Gawen Davey <>]

Additionally, the BSP build script should have -DTHRIFT_ENABLED=off in the cmake invocation as well.

Sorry there's a lot going on at the moment and struggling a bit to keep up with everything.

---

From: mc36 <>
Sent: 22 June 2022 11:44
To: Gawen Davey <>; Frédéric LOUI <>; Alexander Gall <>
Cc: <>; Alexander Jeffries <>
Subject: Re: Access

 

hi,
after applying your patch, recompiled everything and given it a try... i still get segfault and bfswd does not start...
thanks,
cs


mc36@stordis:~$ ./s.sh
[sudo] password for mc36:
256
i2c-1   i2c             igb BB                                  I2C adapter
i2c-2   smbus           SMBus I801 adapter at f000              SMBus adapter
i2c-0   i2c             igb BB                                  I2C adapter
Using SDE /home/mc36/bf-sde-9.9.0/
Using SDE_INSTALL /home/mc36/bf-sde-9.9.0//install/
Setting up DMA Memory Pool
Using TARGET_CONFIG_FILE /home/mc36/bf-sde-9.9.0//install//share/p4/targets/tofino/bf_router.conf
Using PATH /home/mc36/bf-sde-9.9.0//install//bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin
Using LD_LIBRARY_PATH /usr/local/lib:/home/mc36/bf-sde-9.9.0//install//lib:
checking for bf_sysfs_fname /sys/class/bf/bf0/device/dev_add
kernel mode packet driver present, forcing kernel_pkt option!
bf_switchd: Install dir: /home/mc36/bf-sde-9.9.0//install/
bf_switchd: system services initialized
bf_switchd: loading conf_file /home/mc36/bf-sde-9.9.0//install//share/p4/targets/tofino/bf_router.conf...
bf_switchd: processing device configuration...
Configuration for dev_id 0
   Family        : tofino
   pci_sysfs_str : /sys/devices/pci0000:00/0000:00:03.0/0000:05:00.0
   pci_int_mode  : 0
   sds_fw_path   : share/tofino_sds_fw/avago/firmware
bf_switchd: processing P4 configuration...
P4 profile for dev_id 0
num P4 programs 1
   p4_name: bf_router
   p4_pipeline_name: pipe
     libpd:
     libpdthrift:
     context: /home/mc36/bf-sde-9.9.0//install//bf_router.tofino/pipe/context.json
     config: /home/mc36/bf-sde-9.9.0//install//bf_router.tofino/pipe/tofino.bin
   Pipes in scope [0 1 2 3 ]
   diag:
   accton diag:
   Agent[0]: /home/mc36/bf-sde-9.9.0//install//lib/libpltfm_mgr.so
bf_switchd: library /home/mc36/bf-sde-9.9.0//install//lib/libpltfm_mgr.so loaded
bf_switchd: agent[0] initialized

Platform Manager: Starting BF2556X-1T
Platform Manager: Initialising SMB (i2c) subsystem
Platform SMB: Detecting system device
Platform SMB: System i2c bus found at /dev/i2c-2
Platform Manager: Initialising QSFP subsystem
Platform Manager: Initialising IPC (socket) subsystem
Platform IPC: Socket at /tmp/tofino_usocket ready
Platform Manager: Initialising IPMI (sensor) subsystem

Device 0: Operational mode set to ASIC
Initialized the device types using platforms infra API
ASIC detected at PCI /sys/class/bf/bf0/device
ASIC pci device id is 16 (0x0010)
Skipped pkt-mgr init
Starting PD-API RPC server on port 9090
bf_switchd: drivers initialized
\
Platform Manager: Initialising LED subsystem
Platform Manager: Initialising MDIO subsystem
Platform Manager: Starting health monitor

bf_switchd: dev_id 0 initialized

bf_switchd: initialized 1 devices
2022-06-22 12:42:12.185073 BF_SYS ERROR - /home/mc36/bf-sde-9.9.0//install//lib/libdriver.so(+0x1c53bf2) [0x7f15747cabf2]
2022-06-22 12:42:12.185127 BF_SYS ERROR - /home/mc36/bf-sde-9.9.0//install//lib/libdriver.so(bf_switchd_lib_init+0x12ff) [0x7f15747d0dff]
2022-06-22 12:42:12.185148 BF_SYS ERROR - bf_switchd(+0x3773) [0x558c1b10d773]
2022-06-22 12:42:12.185165 BF_SYS ERROR - /lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0xea) [0x7f157275ad0a]
2022-06-22 12:42:12.185180 BF_SYS ERROR - bf_switchd(+0x398a) [0x558c1b10d98a]
/home/mc36/bf-sde-9.9.0//run_switchd.sh: line 218:  1092 Segmentation fault      sudo env "SDE=$SDE" "SDE_INSTALL=$SDE_INSTALL" $ASAN_ON_ERROR "PATH=$PATH"
"LD_LIBRARY_PATH=$LD_LIBRARY_PATH" $DBG bf_switchd "$SERVER_LISTEN_LOCAL_ONLY" --install-dir $SDE_INSTALL --conf-file $TARGET_CONFIG_FILE "--init-mode=$INIT_MODE" $SKIP_HLD_STR
$SKIP_P4_STR $SKIP_PORT_ADD_STR $STS_PORT_STR $KERNEL_PKT_STR $SHELL_NO_WAIT $@
mc36@stordis:~$
mc36@stordis:~$
mc36@stordis:~$ sudo dmesg | tail -n 50
[   21.682010] i2c /dev entries driver
[   21.713375] igb: Intel(R) Gigabit Ethernet Network Driver
[   21.719450] igb: Copyright (c) 2007-2014 Intel Corporation.
[   22.086426] igb 0000:08:00.0: added PHC on eth0
[   22.090376] igb 0000:08:00.0 enp8s0f0: renamed from eth0
[   22.091505] igb 0000:08:00.0: Intel(R) Gigabit Ethernet Network Connection
[   22.105133] igb 0000:08:00.0: enp8s0f0: (PCIe:5.0Gb/s:Width x1) 84:c7:8f:02:20:61
[   22.113580] igb 0000:08:00.0: enp8s0f0: PBA No: 106300-000
[   22.119720] igb 0000:08:00.0: Using MSI-X interrupts. 8 rx queue(s), 8 tx queue(s)
[   22.490289] igb 0000:08:00.1: added PHC on eth0
[   22.494004] igb 0000:08:00.1 enp8s0f1: renamed from eth0
[   22.495370] igb 0000:08:00.1: Intel(R) Gigabit Ethernet Network Connection
[   22.508998] igb 0000:08:00.1: enp8s0f1: (PCIe:5.0Gb/s:Width x1) 84:c7:8f:02:20:62
[   22.517446] igb 0000:08:00.1: enp8s0f1: PBA No: 106300-000
[   22.523582] igb 0000:08:00.1: Using MSI-X interrupts. 8 rx queue(s), 8 tx queue(s)
[   22.532840] igb 0000:0a:00.0: PHY reset is blocked due to SOL/IDER session.
[   22.565111] pps pps0: new PPS source ptp2
[   22.568073] igb 0000:0a:00.0 enp10s0: renamed from eth0
[   22.569754] igb 0000:0a:00.0: added PHC on enp10s0
[   22.580807] igb 0000:0a:00.0: Intel(R) Gigabit Ethernet Network Connection
[   22.588505] igb 0000:0a:00.0: enp10s0: (PCIe:2.5Gb/s:Width x1) 84:c7:8f:02:20:60
[   22.596826] igb 0000:0a:00.0: enp10s0: PBA No: 000300-000
[   22.602870] igb 0000:0a:00.0: Using MSI-X interrupts. 4 rx queue(s), 4 tx queue(s)
[   22.631318] i801_smbus 0000:00:1f.3: SMBus using PCI interrupt
[   22.638472] i2c i2c-2: 2/4 memory slots populated (from DMI)
[   22.672393] igb 0000:0a:00.0 enp10s0: igb: enp10s0 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: RX
[   22.683389] IPv6: ADDRCONF(NETDEV_CHANGE): enp10s0: link becomes ready
[   22.928611] bf_kpkt: loading out-of-tree module taints kernel.
[   22.965925] bf_kpkt: module verification failed: signature and/or required key missing - tainting kernel
[   22.986580] bf_kpkt: Intel(R) Switch ASIC Linux Packet Driver - version 1.0
[   22.994366] bf_kpkt: Copyright (c) 2015-2022 Intel Corporation.
[   23.001003] bf device class created
[   23.004901] Use MSI interrupt
[   23.008215] kpkt_mode 1 hd_room 32 dr_int_en 1 rx_count 256
[   23.014506] bf 0000:05:00.0: enabling device (0100 -> 0102)
[   23.020908] pmd_set_huge: Cannot satisfy [mem 0x383ff8000000-0x383ff8200000] with a huge-page mapping due to MTRR override.
[   23.033696] bf using 2 MSI irq from 104
[   23.033859] BF allocating 2 MSI vectors from  104
[   23.039121] bf 0000:05:00.0: bf device 0 registered with irq 104
[   23.045826] bf probe ok
[   23.048559] bf_kpkt: NAPI Net Driver,version 9.9.0-22-cpr (01May2017)

[   23.057882] bf_pci0 BF PCI-MAC is up
[   23.059777] bf 0000:05:00.0 ens1: renamed from bf_pci0
[   23.061901] bf_kpkt kernel processing enabled for dev 0 subdev_id 0
[   25.849946] bf_m_init 31
[   29.379258] bf_dev_add
[   29.383494] IPv6: ADDRCONF(NETDEV_CHANGE): ens1: link becomes ready
[   29.392419] bf_switchd[1093]: segfault at 0 ip 0000000000000000 sp 00007ffeb74bff18 error 14 in bf_switchd[558c1b10a000+3000]
[   29.405071] Code: Unable to access opcode bytes at RIP 0xffffffffffffffd6.
mc36@stordis:~$

On 6/22/22 11:08, Gawen Davey wrote:
> Yes, It will go on the portal. But it's beta, not ready for general release.
>
> With respect to the latest error, looks like SDE 9.9.0 removed some code that handled a case where a part number is not properly burned into the "efuse". This is not something that
> can be fixed by the BSP, and requires a patched bf-drivers package. So the SDE version as it comes from Intel won't ever work.
>
> I've got a script as a workaround. It unpacks the SDE and bf-drivers package, applies the necessary patch, and packages everything back up again. It expects the original SDE
> tarball to be in the same directory as itself, and will replace it in-situ.
>
> Probably don't need to create an entire SDE tarball, so the last three lines are probably unecessary.
>
> In any case, I've raised a ticket with Intel, but this certainly sets back my plans. I don't want to add more scripts and instructions to our documentation and downloads, so
> nothing will be going on the support portal until I hear back from Intel.
>
> #! /usr/bin/env bash
>
> set -eE
>
> script_dir="$( cd "$( dirname "${BASH_SOURCE[0]}" )" &> /dev/null && pwd )"
>
> cd ${script_dir}
>
> tar xvf bf-sde-9.9.0.tgz
> cd bf-sde-9.9.0/packages
> tar xvf bf-drivers-9.9.0.tgz
> rm bf-drivers-9.9.0.tgz
>
> patch -p0 <<'EOM'
> Index: bf-drivers-9.9.0/src/lld/lld_efuse_tof.c
> ===================================================================
> --- bf-drivers-9.9.0.orig/src/lld/lld_efuse_tof.c
> +++ bf-drivers-9.9.0/src/lld/lld_efuse_tof.c
> @@ -117,6 +117,15 @@ int lld_efuse_tof_load(bf_dev_id_t dev_i
>       dev_p->efuse_data.voltage_scaling =
>             extract_bit_fld_128(hi64, lo64, (245 - 128), (243 - 128));
> +      /* If either pipes or ports are disabled in efuse we'll pretend
> +    * this is a tofin-small */
> +   if ((dev_p->efuse_data.pipe_disable != 0) ||
> +         (dev_p->efuse_data.port_disable_map_lo != 0)) {
> +      /* This is a temporary hack since some of the Tofino lite chip dont have
> +          have the chip part number burned correctly in the efuse */
> +      dev_p->efuse_data.chip_part_number = BFN_PART_NBR_BFNT10032D;   // T-3.2-Half
> +   }
> +
>       if (dev_p->efuse_data.chip_part_number == BFN_PART_NBR_BFNT10032D) {
>          // T-3.2-Half
>          uint64_t p = dev_p->efuse_data.port_disable_map_lo;
> EOM
>
>
> tar cvf bf-drivers-9.9.0.tgz bf-drivers-9.9.0
> rm -rf bf-drivers-9.9.0
> cd ${script_dir}
> rm bf-sde-9.9.0.tgz
> tar cvf bf-sde-9.9.0.tgz bf-sde-9.9.0
>
>
>