tech-net archive
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: boot problems with bge(4)

On 06/07/11 11:51, SAITOH Masanobu wrote:

UNBLIEBABILE!!!

I've checked the latest Linux tg3 driver version 3.116j.
See line 14491-14492 in tg3.c.................

Could you test attached patch? It seems we have to do more work to
support BCM5785...


This is the first time the driver does not show
the "firmware handshake timed out" message.
I don't see it in further (re-)boots.

All my boot tests were successful to the login prompt
and haven't seen any 'nfs_boot timeout' failures
which is a good sign.

But yes, there is more work to do on the DMA side.

Christoph





(2011/06/07 17:33), Christoph Egger wrote:

On 06/06/11 20:33, SAITOH Masanobu wrote:

(2011/06/06 23:55), Christoph Egger wrote:

On 06/06/11 16:26, SAITOH Masanobu wrote:

Thank you Christoph.

I don't know why the following code fails:

           if (pci_get_capability(sc->sc_pc, sc->sc_pcitag,
PCI_CAP_PCIEXPRESS,
                   &sc->bge_pciecap, NULL) != 0) {
                   /* PCIe */
                   sc->bge_flags |= BGE_PCIE;
                   bge_set_max_readrq(sc);
           }


The driver iterates the while-loop in pci_get_capability() three times
and returns 0 then.

Christoph


I'm afraid that your system has problem in the pci config registers.


In pci_get_capability() I dumped the values of PCI_CAPLIST_CAP(reg)
in the while loop.

The output:
pci_get_capability: PCI_CAPLIST_CAP 0x1
pci_get_capability: PCI_CAPLIST_CAP 0x3
pci_get_capability: PCI_CAPLIST_CAP 0x9
pci_get_capability: PCI_CAPLIST_CAP 0x5


The first is PCI_CAP_PWRMGMT,
the secons is PCI_CAP_VPD,
the third is PCI_CAP_VENDSPEC
and the fourth is PCI_CAP_MSI.


Could you show me the message of "pcictl pci0 dump -d 20"? It seems
your are using NFS boot.


Yes.


Can you boot from a disk?


No, but sometimes it happens that the dhcp request succeeds. When that
happens the machine happens.

Here the output of
"pcictl pci0 dump -d 20 -f 6":

PCI configuration registers:
   Common header:
     0x00: 0x169914e4 0x00100006 0x02000001 0x00000010

     Vendor Name: Broadcom (0x14e4)
     Device Name: BCM5785G 10/100/1000 Ethernet (0x1699)
     Command register: 0x0006
       I/O space accesses: off
       Memory space accesses: on
       Bus mastering: on
       Special cycles: off
       MWI transactions: off
       Palette snooping: off
       Parity error checking: off
       Address/data stepping: off
       System error (SERR): off
       Fast back-to-back transactions: off
       Interrupt disable: off
     Status register: 0x0010
       Interrupt status: inactive
       Capability List support: on
       66 MHz capable: off
       User Definable Features (UDF) support: off
       Fast back-to-back capable: off
       Data parity error detected: off
       DEVSEL timing: fast (0x0)
       Slave signaled Target Abort: off
       Master received Target Abort: off
       Master received Master Abort: off
       Asserted System Error (SERR): off
       Parity error detected: off
     Class Name: network (0x02)
     Subclass Name: ethernet (0x00)
     Interface: 0x00
     Revision ID: 0x01
     BIST: 0x00
     Header Type: 0x00 (0x00)
     Latency Timer: 0x00
     Cache Line Size: 0x10

   Type 0 ("normal" device) header:
     0x10: 0xffb00004 0x00000000 0x00000000 0x00000000
     0x20: 0x00000000 0x00000000 0x00000000 0x169914e4
     0x30: 0x00000000 0x00000048 0x00000000 0x0000030a

     Base address register at 0x10
       type: 64-bit nonprefetchable memory
       base: 0x00000000ffb00000, not sized
     Base address register at 0x18
       not implemented(?)
     Base address register at 0x1c
       not implemented(?)
     Base address register at 0x20
       not implemented(?)
     Base address register at 0x24
       not implemented(?)
     Cardbus CIS Pointer: 0x00000000
     Subsystem vendor ID: 0x14e4
     Subsystem ID: 0x1699
     Expansion ROM Base Address: 0x00000000
     Capability list pointer: 0x48
     Reserved @ 0x38: 0x00000000
     Maximum Latency: 0x00
     Minimum Grant: 0x00
     Interrupt pin: 0x03 (pin C)
     Interrupt line: 0x0a

   Capability register at 0x48
     type: 0x01 (Power Management, rev. 1.0)
   Capability register at 0x40
     type: 0x03 (VPD)
   Capability register at 0x60
     type: 0x09 (Vendor-specific)
   Capability register at 0x50
     type: 0x05 (MSI)

   PCI Message Signaled Interrupt
     Message Control register: 0x0080
       MSI Enabled: no
       Multiple Message Capable: no (1 vector)
       Multiple Message Enabled: off (1 vector)
       64 Bit Address Capable: yes
       Per-Vector Masking Capable: no
     Message Address (lower) register: 0xd101a084
     Message Address (upper) register: 0x23858888
     Message Data register: 0x00009118

   PCI Power Management Capabilities Register
     Capabilities register: 0xc003
       Version: 1.2
       PME# clock: off
       Device specific initialization: off
       3.3V auxiliary current: self-powered
       D1 power management state support: off
       D2 power management state support: off
       PME# support: 0x18
     Control/status register: 0x2108
       Power state: D0
       PCI Express reserved: off
       No soft reset: on
       PME# assertion enabled
       PME# status: off

   Device-dependent header:
     0x40: 0x80986003 0xffffffff 0xc0034001 0x0a002108
     0x50: 0x00800005 0xd101a084 0x23858888 0x00009118
     0x60: 0x006c5009 0xf7a0ffa2 0xf0410088 0x761b000f
     0x70: 0x00001292 0x00000000 0x00005804 0x00000000
     0x80: 0x00000000 0x083cb001 0x04130034 0x0008c082
     0x90: 0x01000609 0x00000000 0x00000000 0x0000004e
     0xa0: 0x00000000 0x0000014f 0x00000000 0x00000171
     0xb0: 0x00000000 0x00000000 0x00000000 0x05785041
     0xc0: 0x00000000 0x169914e4 0x0000000e 0x00000000
     0xd0: 0x00000000 0x00000000 0x00000000 0x00000000
     0xe0: 0x00000000 0x00000000 0x00000000 0x00000000
     0xf0: 0x00000000 0x00000000 0x00000000 0x00000000





(2011/06/06 22:13), Christoph Egger wrote:

On 06/03/11 18:37, Christoph Egger wrote:

On 03.06.11 17:55, SAITOH Masanobu wrote:

Hi Christoph.

BCM5785 should not be PCIX chip. It's PCIe chip.

I suspect that your have some local patches for if_bge.c
Could you mail it to me? Please.


Yes, I can do that on monday.


Here we go.

Christoph



Thanks.

(2011/06/03 21:58), Christoph Egger wrote:

On 06/03/11 14:18, Christoph Egger wrote:


Hi,


[...]
bge0 at pci0 dev 20 function 6: Broadcom BCM5785G Gigabit Ethernet
bge0: interrupting at ioapic1 pin 18, event channel 9
bge0: unable to find PCIX capability
bge0: firmware handshake timed out, val = 4b657654
bge0: firmware handshake timed out, val = 4b657654
bge0: ASIC unknown BCM5785 (0x5785041), Ethernet address
00:00:1a:1a:d8:ab
brgphy0 at bge0 phy 1: BCM5785 1000BASE-T media interface, rev. 3
brgphy0: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX,
1000baseT,
1000baseT-FD , auto
[...]

I have a local patch to attach brgphy on bge(4) rather ukphy(4)
...


bge0: firmware handshake timed out, val = 4b657654
bge0: firmware handshake timed out, val = 4b657654
bge0: discarding oversize frame (len=-4)
bge0: discarding oversize frame (len=-4)
nfs_boot: trying DHCP/BOOTP
bge0: watchdog timeout -- resetting
bge0: firmware handshake timed out, val = 4b657654
bge0: firmware handshake timed out, val = 4b657654
bge0: discarding oversize frame (len=-4)
bge0: discarding oversize frame (len=-4)
nfs_boot: timeout...
nfs_boot: timeout...


The "discarding oversize frame" message comes from if_ethersubr.c


I just figured out in bge_rxeof():

         m->m_pkthdr.len = m->m_len = cur_rx->bge_len -
ETHER_CRC_LEN;

cur_rx->bge_len is 0.

Christoph
Home | Main Index | Thread Index | Old Index