Date: Tue, 17 Jul 2007 13:01:56 +0200 From: Patrick Oeschger <paketix@bluewin.ch> To: freebsd-net@freebsd.org Subject: em(4)/82571eb: fifo not dma'ed to host memory Message-ID: <20070717110156.GA37722@hbnnbsddev03.sharedtcs.net>
next in thread | raw e-mail | index | archive | help
bug description for 7.0-CURRENT - em(4) driver version 6.5.3
- frame reception is not possible (missed packets counter incrementing)
- frame transmission is not possible (driver watchdog fires)
bug exists also on 6.2-RELENG - em(4) driver version 6.2.9
bug does *not* exist on 6.1-RELENG - em(4) driver version 3.2.18
jack vogel told me that the i82571 can be tricky when having a bunch
of them - i have 8 such interfaces on the motherboard so i disabled
7 of them with a modified em_probe (if_em.c)
running the driver on just one interface does not change the behaviour
- init of the interface during verbose boot looks fine
- bringing the interface to upness looks fine too
- RDH0/RDT0 registers do *not* increment after frame reception
- frames suspected *not* getting dma'ed from fifo to host memory
- RX interrupt suspected *not* being sent to the driver
see dmesg and debug ouput for futher details...
anybody having some experience with this issue?
any advice to track this further down?
thanks, pat
diff for em_probe subr to ensure that only one interface is init'ed:
freebsd70# diff -u if_em.c.orig if_em.c
--- if_em.c.orig Tue Jul 17 11:18:18 2007
+++ if_em.c Tue Jul 17 06:34:48 2007
@@ -379,6 +379,10 @@
INIT_DEBUGOUT("em_probe: begin");
+ // prevent init of all em(4) devices except pci9/0/0
+ if(pci_get_bus(dev) != 0x09 || pci_get_slot(dev) != 0x00 ||
+ pci_get_function(dev) != 0x00)
+ return (ENXIO);
+
pci_vendor_id = pci_get_vendor(dev);
if (pci_vendor_id != EM_VENDOR_ID)
return (ENXIO);
excerpt of verbose dmesg looks fine for em0:
pcib9: <PCI-PCI bridge> irq 19 at device 11.0 on pci2
pcib9: secondary bus 9
pcib9: subordinate bus 9
pcib9: I/O decode 0x6000-0x6fff
pcib9: memory decode 0xfd700000-0xfd7fffff
pcib9: prefetched decode 0xfce00000-0xfcefffff
pci9: <PCI bus> on pcib9
pci9: physical bus=9
found-> vendor=0x8086, dev=0x105e, revid=0x06
bus=9, slot=0, func=0
class=02-00-00, hdrtype=0x00, mfdev=1
cmdreg=0x0000, statreg=0x0010, cachelnsz=16 (dwords)
lattimer=0x00 (0 ns), mingnt=0x00 (0 ns), maxlat=0x00 (0 ns)
intpin=a, irq=255
powerspec 2 supports D0 D3 current D0
MSI supports 1 message, 64 bit
map[10]: type Memory, range 32, base 0, size 17, memory disabled
map[14]: type Memory, range 32, base 0, size 17, memory disabled
map[18]: type I/O Port, range 32, base 0, size 5, port disabled
found-> vendor=0x8086, dev=0x105e, revid=0x06
bus=9, slot=0, func=1
class=02-00-00, hdrtype=0x00, mfdev=1
cmdreg=0x0000, statreg=0x0010, cachelnsz=16 (dwords)
lattimer=0x00 (0 ns), mingnt=0x00 (0 ns), maxlat=0x00 (0 ns)
intpin=b, irq=255
powerspec 2 supports D0 D3 current D0
MSI supports 1 message, 64 bit
map[10]: type Memory, range 32, base 0, size 17, memory disabled
map[14]: type Memory, range 32, base 0, size 17, memory disabled
map[18]: type I/O Port, range 32, base 0, size 5, port disabled
em0: <Intel(R) PRO/1000 Network Connection Version - 6.5.3>
at device 0.0 on pci9
pcib9: em0 requested memory range 0xfd700000-0xfd7fffff: good
pcib2: em0 requested memory range 0xfd700000-0xfd7fffff: good
pcib1: em0 requested memory range 0xfd700000-0xfd7fffff: good
em0: Lazy allocation of 0x20000 bytes rid 0x10 type 3 at 0xfd700000
pcib1: matched entry for 1.0.INTD
pcib1: slot 0 INTD hardwired to IRQ 19
pcib2: slot 11 INTA is routed to irq 19
pcib9: slot 0 INTA is routed to irq 19
em0: bpf attached
em0: Ethernet address: 00:10:f3:0c:5b:2a
ioapic0: routing intpin 19 (PCI IRQ 19) to vector 49
em0: [FILTER]
pci9: <network, ethernet> at device 0.1 (no driver attached)
bringing em0 to upness seems to work fine:
printf's inserted into subroutines em_handle_link and em_rxeof (if_em.c)
freebsd70# ifconfig em0 up
em0: Link is Down
em_handle_link
em_rxeof
em_handle_link
em_rxeof
em0: Link is up 1000 Mbps Full Duplex
freebsd70# ifconfig em0
em0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> metric 0
mtu 1500
options=18b<RXCSUM,TXCSUM,VLAN_MTU,VLAN_HWCSUM,TSO4>
ether 00:10:f3:0c:5b:2a
media: Ethernet autoselect (1000baseTX <full-duplex>)
status: active
interface stats:
good packets increment but no em_rxeof/interrupts generated
fifo is full after 85 received frames
memory and missed packets are incrementing because they cant be
stored in fifo (fifo exhausted)
freebsd70# sysctl dev.em.0.stats=1
em0: Excessive collisions = 0
em0: Sequence errors = 0
em0: Defer count = 0
em0: Missed Packets = 144
em0: Receive No Buffers = 0
em0: Receive Length Errors = 0
em0: Receive errors = 0
em0: Crc errors = 0
em0: Alignment errors = 0
em0: Carrier extension errors = 0
em0: RX overruns = 0
em0: watchdog timeouts = 0
em0: XON Rcvd = 0
em0: XON Xmtd = 0
em0: XOFF Rcvd = 0
em0: XOFF Xmtd = 0
em0: Good Packets Rcvd = 85
em0: Good Packets Xmtd = 0
em0: TSO Contexts Xmtd = 0
em0: TSO Contexts Failed = 0
interface debug_info:
output for registers RDBAH0/RDBAL0/RDLEN0/RDH0/RDT0/RXDCTL added
RDH0/RDT0 do not increment after frame reception...
cp. 6.1-RELENG: RDH0/RDT0 incremented after frame reception
RDT0 is always 'RDH0 - 1'
freebsd70# sysctl dev.em.0.debug_info=1
em0: Adapter hardware address = 0xc3b69a1c
em0: CTRL = 0x81c0241 RCTL = 0x8002
em0: RDBAH0 = 0x0 RDBAL0 = 0x1458000
em0: RDLEN0 = 0x1000 RDH0 = 0x0
em0: RDT0 = 0xff RXDCTL = 0x10000
em0: Packet buffer = Tx=16k Rx=32k
em0: Flow control watermarks high = 30720 low = 29220
em0: tx_int_delay = 66, tx_abs_int_delay = 66
em0: rx_int_delay = 0, rx_abs_int_delay = 66
em0: fifo workaround = 0, fifo_reset_count = 0
em0: hw tdh = 0, hw tdt = 0
em0: hw rdh = 0, hw rdt = 255
em0: Num Tx descriptors avail = 256
em0: Tx Descriptors not avail1 = 0
em0: Tx Descriptors not avail2 = 0
em0: Std mbuf failed = 0
em0: Std mbuf cluster failed = 0
em0: Driver dropped packets = 0
em0: Driver tx dma failure in encap = 0
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20070717110156.GA37722>