[Bug 265714] igc(4) drops link under high traffic

[email protected]
Newsgroups gmane.os.freebsd.devel.net
Message-ID <[email protected]/bugzilla/>
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=265714

--- Comment #88 from [email protected] ---
I have a "microPC" that has two of these ports in it:

igc0: <Intel(R) Ethernet Controller I226-V> mem
0x80c00000-0x80cfffff,0x80d00000-0x80d03fff at device 0.0 on pci1
igc0: EEPROM V2.17-0 eTrack 0x80000303
igc0: Using 1024 TX descriptors and 1024 RX descriptors
igc0: Using 4 RX queues 4 TX queues
igc0: Using MSI-X interrupts with 5 vectors
igc0: Ethernet address: c8:ff:bf:05:95:e2
igc0: netmap queues/slots: TX 4/1024, RX 4/1024
igc1: <Intel(R) Ethernet Controller I226-V> mem
0x80900000-0x809fffff,0x80a00000-0x80a03fff at device 0.0 on pci3
igc1: EEPROM V2.17-0 eTrack 0x80000303
igc1: Using 1024 TX descriptors and 1024 RX descriptors
igc1: Using 4 RX queues 4 TX queues
igc1: Using MSI-X interrupts with 5 vectors
igc1: Ethernet address: c8:ff:bf:05:95:e3
igc1: netmap queues/slots: TX 4/1024, RX 4/1024

This is a 15.1-STABLE build which I just refreshed today (so it has everything
as of this morning) and I cannot reproduce hangs or other significant
misbehavior with iperf3 running against a server with a Mellanox card on the
other end through a 2.5Gbps connection (said Mellanox card is on 10G into the
switch) with either single-stream or 4-stream parallel (e.g. -P 4)

This DID used to produce trouble some time back but with the current patches it
does not and throughput is stable at 2.35Gbps -- for what its worth.  The
machine itself is an N150 CPU Chineesium machine that used to be all over the
place at very reasonable cost but the RAM insanity since the New Year has
basically killed all of those off (prices have more than doubled and the "tiny
box" ones have gone away entirely, which for this application is extremely
annoying.)

CPU: Intel(R) N150 (806.40-MHz K8-class CPU)
  Origin="GenuineIntel"  Id=0xb06e0  Family=0x6  Model=0xbe  Stepping=0
 
Features=0xbfebfbff<FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,MCA,CMOV,PAT,PSE36,CLFLUSH,DTS,ACPI,MMX,FXSR,SSE,SSE2,SS,HTT,TM,PBE>
 
Features2=0x7ffafbff<SSE3,PCLMULQDQ,DTES64,MON,DS_CPL,VMX,SMX,EST,TM2,SSSE3,SDBG,FMA,CX16,xTPR,PDCM,PCID,SSE4.1,SSE4.2,x2APIC,MOVBE,POPCNT,TSCDLT,AESNI,XSAVE,OSXSAVE,AVX,F16C,RDRAND>
  AMD Features=0x2c100800<SYSCALL,NX,Page1GB,RDTSCP,LM>
  AMD Features2=0x121<LAHF,ABM,Prefetch>
  Structured Extended
Features=0x239ca7eb<FSGSBASE,TSCADJ,BMI1,AVX2,FDPEXC,SMEP,BMI2,ERMS,INVPCID,NFPUSG,PQE,RDSEED,ADX,SMAP,CLFLUSHOPT,CLWB,PROCTRACE,SHA>
  Structured Extended
Features2=0x98c007bc<UMIP,PKU,OSPKE,WAITPKG,GFNI,VAES,VPCLMULQDQ,RDPID,MOVDIRI,MOVDIR64B>
  Structured Extended
Features3=0xfc184410<FSRM,MD_CLEAR,IBT,IBPB,STIBP,L1DFL,ARCH_CAP,CORE_CAP,SSBD>
  Structured Extended Features4=0x810
  XSAVE Features=0xf<XSAVEOPT,XSAVEC,XINUSE,XSAVES>
  IA32_ARCH_CAPS=0x1580fd6b<RDCL_NO,IBRS_ALL,SKIP_L1DFL_VME,MDS_NO,TAA_NO>
  VT-x: PAT,HLT,MTF,PAUSE,EPT,UG,VPID,VID,PostIntr
  TSC: P-state invariant, performance statistics
real memory  = 12884901888 (12288 MB)
avail memory = 12126695424 (11564 MB)
Event timer "LAPIC" quality 600
ACPI APIC Table: <ALASKA A M I >


The accumulated stats show LOTS of retries but no stalls or other misbehavior.

igc box as client:

[ ID] Interval           Transfer     Bitrate         Retr
[  5]   0.00-600.00 sec  54.8 GBytes   785 Mbits/sec  7447            sender
[  5]   0.00-600.02 sec  54.8 GBytes   785 Mbits/sec                  receiver
[  7]   0.00-600.00 sec  27.6 GBytes   395 Mbits/sec  10970            sender
[  7]   0.00-600.02 sec  27.6 GBytes   395 Mbits/sec                  receiver
[  9]   0.00-600.00 sec  27.2 GBytes   390 Mbits/sec  10970            sender
[  9]   0.00-600.02 sec  27.2 GBytes   390 Mbits/sec                  receiver
[ 11]   0.00-600.00 sec  54.8 GBytes   785 Mbits/sec  7446            sender
[ 11]   0.00-600.02 sec  54.8 GBytes   785 Mbits/sec                  receiver
[SUM]   0.00-600.00 sec   164 GBytes  2.35 Gbits/sec  36833             sender
[SUM]   0.00-600.02 sec   164 GBytes  2.35 Gbits/sec                  receiver

Same box as server (client is said Mellanox-equipped box coming this way):

- - - - - - - - - - - - - - - - - - - - - - - - -
[ ID] Interval           Transfer     Bitrate         Retr
[  5]   0.00-600.00 sec  42.0 GBytes   602 Mbits/sec  145070            sender
[  5]   0.00-600.00 sec  42.0 GBytes   602 Mbits/sec                  receiver
[  7]   0.00-600.00 sec  42.9 GBytes   614 Mbits/sec  168729            sender
[  7]   0.00-600.00 sec  42.9 GBytes   614 Mbits/sec                  receiver
[  9]   0.00-600.00 sec  42.5 GBytes   609 Mbits/sec  104962            sender
[  9]   0.00-600.00 sec  42.5 GBytes   609 Mbits/sec                  receiver
[ 11]   0.00-600.00 sec  36.7 GBytes   525 Mbits/sec  138302            sender
[ 11]   0.00-600.00 sec  36.7 GBytes   525 Mbits/sec                  receiver
[SUM]   0.00-600.00 sec   164 GBytes  2.35 Gbits/sec  557063             sender
[SUM]   0.00-600.00 sec   164 GBytes  2.35 Gbits/sec                  receiver

iperf Done.

FreeBSD 15.1-STABLE stable/15-n284864-6b25004d1d18 GENERIC

Note that I do not have flow control enabled on the switch and it shows zero
errors after this set of tests on either the client or server.

-- 
You are receiving this mail because:
You are the assignee for the bug.
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.