Summary: | [em] Intel S5000-based mobo, em driver does not attach to first ethernet controller | ||||||||
---|---|---|---|---|---|---|---|---|---|
Product: | Base System | Reporter: | Dmitry A.Deineka <ddeineka> | ||||||
Component: | kern | Assignee: | jfv | ||||||
Status: | Closed Overcome By Events | ||||||||
Severity: | Affects Only Me | CC: | sbruno | ||||||
Priority: | Normal | Keywords: | IntelNetworking | ||||||
Version: | 6.2-RELEASE | ||||||||
Hardware: | Any | ||||||||
OS: | Any | ||||||||
Attachments: |
|
Description
Dmitry A.Deineka
2007-11-08 14:30:01 UTC
This is like joke, but today we was try to add 4 more GB of memory to this machine. With 8Gb total memory all em's does not work with similar diagnostics (dmesg output): Nov 8 17:10:24 dbmaster kernel: Copyright (c) 1992-2007 The FreeBSD Project. Nov 8 17:10:24 dbmaster kernel: Copyright (c) 1979, 1980, 1983, 1986, 1988, 1989, 1991, 1992, 1993, 1994 Nov 8 17:10:24 dbmaster kernel: The Regents of the University of California. All rights reserved. Nov 8 17:10:24 dbmaster kernel: FreeBSD is a registered trademark of The FreeBSD Foundation. Nov 8 17:10:24 dbmaster kernel: FreeBSD 6.3-PRERELEASE #4: Thu Nov 8 15:57:41 EET 2007 Nov 8 17:10:24 dbmaster kernel: axl@dbmaster.itl.ua:/usr/obj/usr/src/sys/DBMASTER Nov 8 17:10:24 dbmaster kernel: ACPI APIC Table: <INTEL S5000PSL> Nov 8 17:10:24 dbmaster kernel: Timecounter "i8254" frequency 1193182 Hz quality 0 Nov 8 17:10:24 dbmaster kernel: CPU: Intel(R) Xeon(R) CPU E5310 @ 1.60GHz (1595.94-MHz K8-class CPU) Nov 8 17:10:24 dbmaster kernel: Origin = "GenuineIntel" Id = 0x6f7 Stepping = 7 Nov 8 17:10:24 dbmaster kernel: 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> Nov 8 17:10:24 dbmaster kernel: Features2=0x4e33d<SSE3,RSVD2,MON,DS_CPL,VMX,TM2,SSSE3,CX16,xTPR,PDCM,DCA> Nov 8 17:10:24 dbmaster kernel: AMD Features=0x20100800<SYSCALL,NX,LM> Nov 8 17:10:24 dbmaster kernel: AMD Features2=0x1<LAHF> Nov 8 17:10:24 dbmaster kernel: Cores per package: 4 Nov 8 17:10:24 dbmaster kernel: real memory = 10200547328 (9728 MB) Nov 8 17:10:24 dbmaster kernel: avail memory = 8266711040 (7883 MB) Nov 8 17:10:24 dbmaster kernel: ioapic0 <Version 2.0> irqs 0-23 on motherboard Nov 8 17:10:24 dbmaster kernel: ioapic1 <Version 2.0> irqs 24-47 on motherboard Nov 8 17:10:24 dbmaster kernel: lapic0: Forcing LINT1 to edge trigger Nov 8 17:10:24 dbmaster kernel: kbd1 at kbdmux0 Nov 8 17:10:24 dbmaster kernel: acpi0: <INTEL S5000PSL> on motherboard Nov 8 17:10:24 dbmaster kernel: acpi_hpet0: <High Precision Event Timer> iomem 0xfed00000-0xfed003ff on acpi0 Nov 8 17:10:24 dbmaster kernel: Timecounter "HPET" frequency 14318180 Hz quality 2000 Nov 8 17:10:24 dbmaster kernel: acpi0: Power Button (fixed) Nov 8 17:10:24 dbmaster kernel: Timecounter "ACPI-fast" frequency 3579545 Hz quality 1000 Nov 8 17:10:24 dbmaster kernel: acpi_timer0: <24-bit timer at 3.579545MHz> port 0x408-0x40b on acpi0 Nov 8 17:10:24 dbmaster kernel: cpu0: <ACPI CPU> on acpi0 Nov 8 17:10:24 dbmaster kernel: device_attach: acpi_perf0 attach returned 6 Nov 8 17:10:24 dbmaster kernel: device_attach: acpi_perf0 attach returned 6 Nov 8 17:10:24 dbmaster kernel: acpi_throttle0: <ACPI CPU Throttling> on cpu0 Nov 8 17:10:24 dbmaster kernel: acpi_button0: <Sleep Button> on acpi0 Nov 8 17:10:24 dbmaster kernel: acpi_button1: <Power Button> on acpi0 Nov 8 17:10:24 dbmaster kernel: pcib0: <ACPI Host-PCI bridge> port 0xca2,0xca3,0xcf8-0xcff on acpi0 Nov 8 17:10:24 dbmaster kernel: pci0: <ACPI PCI bus> on pcib0 Nov 8 17:10:24 dbmaster kernel: pcib1: <ACPI PCI-PCI bridge> at device 2.0 on pci0 Nov 8 17:10:24 dbmaster kernel: pci1: <ACPI PCI bus> on pcib1 Nov 8 17:10:24 dbmaster kernel: pcib2: <ACPI PCI-PCI bridge> irq 16 at device 0.0 on pci1 Nov 8 17:10:24 dbmaster kernel: pci2: <ACPI PCI bus> on pcib2 Nov 8 17:10:24 dbmaster kernel: pcib3: <ACPI PCI-PCI bridge> irq 16 at device 0.0 on pci2 Nov 8 17:10:24 dbmaster kernel: pci3: <ACPI PCI bus> on pcib3 Nov 8 17:10:24 dbmaster kernel: pcib4: <ACPI PCI-PCI bridge> irq 18 at device 2.0 on pci2 Nov 8 17:10:24 dbmaster kernel: pci4: <ACPI PCI bus> on pcib4 Nov 8 17:10:24 dbmaster kernel: em0: <Intel(R) PRO/1000 Network Connection Version - 6.7.2> port 0x2020-0x203f mem 0xb8820000-0xb883ffff,0xb8400000-0xb87fffff irq 18 at device 0.0 on pci4 Nov 8 17:10:24 dbmaster kernel: em0: Setup of Shared code failed Nov 8 17:10:24 dbmaster kernel: device_attach: em0 attach returned 6 Nov 8 17:10:24 dbmaster kernel: em1: <Intel(R) PRO/1000 Network Connection Version - 6.7.2> port 0x2000-0x201f mem 0xb8800000-0xb881ffff,0xb8000000-0xb83fffff irq 19 at device 0.1 on pci4 Nov 8 17:10:24 dbmaster kernel: em1: Hardware Initialization Failed Nov 8 17:10:24 dbmaster kernel: em1: Unable to initialize the hardware Nov 8 17:10:24 dbmaster kernel: device_attach: em1 attach returned 5 and later: Nov 8 17:10:25 dbmaster kernel: em0: <Intel(R) PRO/1000 Network Connection Version - 6.7.2> port 0x2020-0x203f mem 0xb8820000-0xb883ffff,0xb8400000-0xb87fffff irq 18 at device 0.0 on pci4 Nov 8 17:10:25 dbmaster kernel: em0: Setup of Shared code failed Nov 8 17:10:25 dbmaster kernel: device_attach: em0 attach returned 6 Nov 8 17:10:25 dbmaster kernel: em1: <Intel(R) PRO/1000 Network Connection Version - 6.7.2> port 0x2000-0x201f mem 0xb8800000-0xb881ffff,0xb8000000-0xb83fffff irq 19 at device 0.1 on pci4 Nov 8 17:10:25 dbmaster kernel: em1: Setup of Shared code failed Nov 8 17:10:25 dbmaster kernel: device_attach: em1 attach returned 6 Nov 8 17:10:25 dbmaster kernel: em0: <Intel(R) PRO/1000 Network Connection Version - 6.7.2> port 0x2020-0x203f mem 0xb8820000-0xb883ffff,0xb8400000-0xb87fffff irq 18 at device 0.0 on pci4 Nov 8 17:10:25 dbmaster kernel: em0: Setup of Shared code failed Nov 8 17:10:25 dbmaster kernel: device_attach: em0 attach returned 6 Nov 8 17:10:25 dbmaster kernel: em1: <Intel(R) PRO/1000 Network Connection Version - 6.7.2> port 0x2000-0x201f mem 0xb8800000-0xb881ffff,0xb8000000-0xb83fffff irq 19 at device 0.1 on pci4 Nov 8 17:10:25 dbmaster kernel: em1: Setup of Shared code failed Nov 8 17:10:25 dbmaster kernel: device_attach: em1 attach returned 6 Nov 8 17:10:25 dbmaster kernel: em0: <Intel(R) PRO/1000 Network Connection Version - 6.7.2> port 0x2020-0x203f mem 0xb8820000-0xb883ffff,0xb8400000-0xb87fffff irq 18 at device 0.0 on pci4 Nov 8 17:10:25 dbmaster kernel: em0: Setup of Shared code failed Nov 8 17:10:25 dbmaster kernel: device_attach: em0 attach returned 6 Nov 8 17:10:25 dbmaster kernel: em1: <Intel(R) PRO/1000 Network Connection Version - 6.7.2> port 0x2000-0x201f mem 0xb8800000-0xb881ffff,0xb8000000-0xb83fffff irq 19 at device 0.1 on pci4 Nov 8 17:10:25 dbmaster kernel: em1: Setup of Shared code failed Nov 8 17:10:25 dbmaster kernel: device_attach: em1 attach returned 6 When extra 4Gb was removed from motherboard, only em0 was down, em1 works, like in first bugreport. Hope this additional information helps. Responsible Changed From-To: freebsd-bugs->jfv Hi Jack can you look into that. Same board, same issue but intermittent between reboots. DEVICE_POLLING enabled if that gives a clue. David Wassman, MCP Net+ IT Network Administrator Davis, Monk & Company (800) 344-5034 (352) 372-6300 (352) 375-1583 FAX The information contained in this electronic message is legally privileged and confidential under applicable law, and is intended only for the use of the individual or entity named above. If you are not the intended recipient, you are hereby notified that any dissemination, copying or disclosure of this communication is strictly prohibited. If you have received this communication in error, please notify Davis, Monk & Company (352) 372-6300 and delete this communication immediately without reading it, making any copies of it or distributing it. Hello, I have recently installed FreeBSD: FreeBSD www.example.com 6.4-RELEASE FreeBSD 6.4-RELEASE #0: Mon Apr 20 22:13:11 PDT 2009 root@www.example.com:/usr/src/sys/amd64/compile/www_amd64 amd64 Onto an Intel s5000PSLSATAR server motherboard. I can confirm the findings of the original PR: em0 does indeed fail to initialize: em0: <Intel(R) PRO/1000 Network Connection Version - 6.7.3> port 0x4020-0x403f mem 0xbe820000-0xbe83ffff,0xbe400000-0xbe7fffff irq 18 at device 0.0 on pci4 em0: Using MSI interrupt em0: Hardware Initialization Failed em0: Unable to initialize the hardware device_attach: em0 attach returned 5 and further, this is indeed an intermittent problem - some boots produce both a working em0 and em1, and some boots produce only a working em1. The motherboard has a very recent (January 2009) BIOS installed. jfv: I am happy to donate to you (or the project) one of these s5000 motherboards. Would that speed testing and problem resolution ? ----- John Kozubik - john@kozubik.com - http://www.kozubik.com Hi John, First, can you please try this hack used by Paul Saab, it eliminates the problem he has, it is not an acceptable final fix to us, but it will let me know if you have the same issue. If it does get rid of the failure, then I will send you newer code to try. Cheers, Jack === e1000_80003es2lan.c ================================================================== --- e1000_80003es2lan.c (/freebsd/head/sys/dev/e1000) (revision 195218) +++ e1000_80003es2lan.c (/ps/head/sys/dev/e1000) (revision 195218) @@ -356,7 +356,7 @@ DEBUGFUNC("e1000_acquire_mac_csr_80003es2lan"); - mask = E1000_SWFW_CSR_SM; + mask = 0;// E1000_SWFW_CSR_SM; return e1000_acquire_swfw_sync_80003es2lan(hw, mask); } @@ -373,7 +373,7 @@ DEBUGFUNC("e1000_release_mac_csr_80003es2lan"); - mask = E1000_SWFW_CSR_SM; + mask = 0;// E1000_SWFW_CSR_SM; e1000_release_swfw_sync_80003es2lan(hw, mask); } On Thu, Apr 23, 2009 at 2:17 PM, John Kozubik <john@kozubik.com> wrote: > > Hello, > > I have recently installed FreeBSD: > > FreeBSD www.example.com 6.4-RELEASE FreeBSD 6.4-RELEASE #0: Mon Apr 20 > 22:13:11 PDT 2009 > root@www.example.com:/usr/src/sys/amd64/compile/www_amd64 amd64 > > Onto an Intel s5000PSLSATAR server motherboard. > > I can confirm the findings of the original PR: em0 does indeed fail to > initialize: > > em0: <Intel(R) PRO/1000 Network Connection Version - 6.7.3> port > 0x4020-0x403f mem 0xbe820000-0xbe83ffff,0xbe400000-0xbe7fffff irq 18 at > device 0.0 on pci4 > em0: Using MSI interrupt > em0: Hardware Initialization Failed > em0: Unable to initialize the hardware > device_attach: em0 attach returned 5 > > and further, this is indeed an intermittent problem - some boots produce > both a working em0 and em1, and some boots produce only a working em1. > > The motherboard has a very recent (January 2009) BIOS installed. > > jfv: I am happy to donate to you (or the project) one of these s5000 > motherboards. Would that speed testing and problem resolution ? > > > ----- > John Kozubik - john@kozubik.com - http://www.kozubik.com > I am having a very similar issue with FBSD 6.3 running on on the intel 5000PSL. Both em0 and em1 may disappear with the same "Unable to Initialize hardware" error message. The suggested patch in kern/117926 did not work in this case. Sticking printf's in the driver shows it failing to select the correct register page in the file e1000_80003es2lan.c. See e1000_read_phy_reg_gg82563_80003es2lan() where it calls ret_val = e1000_read_phy_reg_m88(hw, page_select, &temp) ). Very often the temp variable returns with an unexpected value. Thanks. This should be retested with the 10.2r beta and stable/10 if possible. If this is still a problem, please reopen this ticket. |