weird pcmcia problem
Running 2.6.0-test3 (both with and without your recent yenta socket patches) pcmcia cards present during boot don't show up until they are removed and reinserted. Once reinserted, they work fine. This only seems to occur if yenta_socket is build into the kernel; if support is modular, cards appear to be recognized properly at boot time. I am running on a thinkpad t21. Let me know if I can help with this problem in any way... thanks
-nld
messages from boot:
Yenta: CardBus bridge found at 0000:00:02.0 [1014:0130]
Yenta: Using INTVAL to route CSC interrupts to PCI
Yenta: Routing CardBus interrupts to PCI
Yenta: ISA IRQ list 04b8, PCI irq11
Socket status: 30000010
Yenta: CardBus bridge found at 0000:00:02.1 [1014:0130]
Yenta: Using INTVAL to route CSC interrupts to PCI
Yenta: Routing CardBus interrupts to PCI
Yenta: ISA IRQ list 04b8, PCI irq11
Socket status: 30000006
<later>
cs: warning: no high memory space available!
cs: unable to map card memory!
cs: unable to map card memory!
cs: IO port probe 0x0c00-0x0cff: clean.
cs: IO port probe 0x0800-0x08ff: clean.
cs: IO port probe 0x0100-0x04ff: excluding 0x170-0x177 0x370-0x377 0x3b8-0x3df 0x4d0-0x4d7
cs: IO port probe 0x0a00-0x0aff: clean.
Quoted 7 lines · Narayan Desai
Running 2.6.0-test3 (both with and without your recent yenta socket patches) pcmcia cards present during boot don't show up until they are removed and reinserted. Once reinserted, they work fine. This only seems to occur if
yenta_socketis build into the kernel; if support is modular, cards appear to be recognized properly at boot time. I am running on a thinkpad t21. Let me know if I can help with this problem in any way... thanks
I see the same problem. "cardctl eject; cardctl insert" is enough to get the card recognised for me.
-- Jamie
Quoted 7 lines · Narayan Desai
Running 2.6.0-test3 (both with and without your recent yenta socket patches) pcmcia cards present during boot don't show up until they are removed and reinserted. Once reinserted, they work fine. This only seems to occur if
yenta_socketis build into the kernel; if support is modular, cards appear to be recognized properly at boot time. I am running on a thinkpad t21. Let me know if I can help with this problem in any way... thanks
Same issue on 2.6.0-test3-mm2 on a Dell Latitude C840 with a TrueMobile 1150 wireless (uses orinoco_cs driver) - card is recognized at boot, and somewhat configured:
Aug 18 11:16:52 turing-police kernel: eth1: Station identity 001f:0001:0008:000a Aug 18 11:16:53 turing-police kernel: eth1: Looks like a Lucent/Agere firmware version 8.10 Aug 18 11:16:53 turing-police kernel: eth1: Ad-hoc demo mode supported Aug 18 11:16:53 turing-police kernel: eth1: IEEE standard IBSS ad-hoc mode supported Aug 18 11:16:53 turing-police kernel: eth1: WEP supported, 104-bit key Aug 18 11:16:53 turing-police kernel: eth1: MAC address 00:02:2D:5C:11:48 Aug 18 11:16:53 turing-police kernel: eth1: Station name "HERMES I" Aug 18 11:16:53 turing-police kernel: eth1: ready Aug 18 11:16:53 turing-police kernel: eth1: index 0x01: Vcc 3.3, irq 11, io 0x0100-0x013f
but it takes a 'cardctl eject 2;cardctl insert 2' to actually finish the job and get things started (actually popping the card out not required, and not possible, it's a built-in). Relevant .config:
CONFIG_YENTA=yCONFIG_HERMES=y
# CONFIG_PLX_HERMES is not set
# CONFIG_TMD_HERMES is not setCONFIG_PCI_HERMES=yCONFIG_PCMCIA_HERMES=y
I'm wondering if this is a cardmgr/hotplug bug - if they can't insmod the Yenta support, they don't realize that it could already be onboard so they don't even try to start the cards?
A message that is not in the archive (sent off-list or not yet received): [email protected]
Quoted 7 lines · On Tue, Aug 19, 2003 at 02:16:23PM -0400, [email protected] wrote:
Quoted 4 lines · On Tue, 19 Aug 2003 00:47:00 EDT, [email protected] said:
Quoted 3 lines · Narayan Desai
Running 2.6.0-test3 (both with and without your recent yenta socket patches) pcmcia cards present during boot don't show up until they are removed and reinserted. Once reinserted, they work fine. This only
Quoted 7 lines
Quoted 3 lines
Same issue on 2.6.0-test3-mm2 on a Dell Latitude C840 with a TrueMobile 1150 wireless (uses
orinoco_csdriver) - card is recognized at boot, and somewhat configured:Went to 2.6.0-test3-mm3, and the problem is resolved on my laptop. Not sure if Narayan's machine is using a different "recent Yenta socket patches" than what's in -mm3, or if there's something ELSE that made the difference.
That wasn't expected.
Can you provide all the following information please:
- make/model of machine
- type of cardbus bridge (from lspci)
- type of card (pcmcia or cardbus)
- make/model of card
- full kernel dmesg (including yenta, card services messages)
- cardmgr messages from system log
thx.
Signature · 2 lines
Russell King ([email protected]) The developer of ARM Linux http://www.arm.linux.org.uk/personal/aboutme.html
Quoted 1 line · Russell King
That wasn't expected.
Well, I hadn't tried any Yenta patch on top of -mm2, so I wasn't sure..
Quoted 3 lines
Can you provide all the following information please:
- make/model of machine
Dell Latitude C840
Quoted 1 line
- type of cardbus bridge (from lspci)
02:01.0 CardBus bridge: Texas Instruments PCI4451 PC card Cardbus Controller
Subsystem: Dell Computer Corporation: Unknown device 00d5
Quoted 1 line
- type of card (pcmcia or cardbus)
% cardctl ident 2
product info: "Dell", "TrueMobile 1150 Series PC Card", "Version 01.01", ""
manfid: 0x0156, 0x0002
function: 6 (network)
Quoted 3 lines
- make/model of card
- full kernel dmesg (including yenta, card services messages)
Aug 19 14:04:19 turing-police kernel: Yenta: CardBus bridge found at 0000:02:01.0 [1028:00d5] Aug 19 14:04:19 turing-police kernel: Yenta: Using CSCINT to route CSC interrupts to PCI Aug 19 14:04:19 turing-police kernel: Yenta: Routing CardBus interrupts to PCI Aug 19 14:04:19 turing-police kernel: Yenta: ISA IRQ list 00b8, PCI irq9 Aug 19 14:04:19 turing-police kernel: Socket status: 30000020 Aug 19 14:04:19 turing-police kernel: PCI: Enabling device 0000:02:01.1 (0000 -> 0002) Aug 19 14:04:19 turing-police kernel: Yenta: CardBus bridge found at 0000:02:01.1 [1028:00d5] Aug 19 14:04:19 turing-police kernel: Yenta: Using CSCINT to route CSC interrupts to PCI Aug 19 14:04:19 turing-police kernel: Yenta: Routing CardBus interrupts to PCI Aug 19 14:04:19 turing-police kernel: Yenta: ISA IRQ list 00b8, PCI irq9 Aug 19 14:04:19 turing-police kernel: Socket status: 30000006 Aug 19 14:04:19 turing-police kernel: Yenta: CardBus bridge found at 0000:02:03.0 [12a3:ab01] Aug 19 14:04:19 turing-police kernel: Yenta: Enabling burst memory read transactions Aug 19 14:04:19 turing-police kernel: Yenta: Using CSCINT to route CSC interrupts to PCI Aug 19 14:04:20 turing-police kernel: Yenta: Routing CardBus interrupts to PCI Aug 19 14:04:20 turing-police kernel: Yenta: ISA IRQ list 0000, PCI irq9 Aug 19 14:04:20 turing-police kernel: Socket status: 30000010
Aug 19 14:04:24 turing-police kernel: cs: memory probe 0x0c0000-0x0fffff: excluding 0xc0000-0xcffff 0xe0000-0xfffff Aug 19 14:04:24 turing-police kernel: eth1: Station identity 001f:0001:0008:000a Aug 19 14:04:24 turing-police kernel: eth1: Looks like a Lucent/Agere firmware version 8.10 Aug 19 14:04:24 turing-police kernel: eth1: Ad-hoc demo mode supported Aug 19 14:04:24 turing-police kernel: eth1: IEEE standard IBSS ad-hoc mode supported Aug 19 14:04:24 turing-police kernel: eth1: WEP supported, 104-bit key Aug 19 14:04:25 turing-police kernel: eth1: MAC address 00:02:2D:5C:11:48 Aug 19 14:04:25 turing-police kernel: eth1: Station name "HERMES I" Aug 19 14:04:25 turing-police kernel: eth1: ready Aug 19 14:04:25 turing-police kernel: eth1: index 0x01: Vcc 3.3, irq 9, io 0x0100-0x013f Aug 19 14:04:25 turing-police kernel: eth1: New link status: Connected (0001) Aug 19 14:04:25 turing-police kernel: eth1: no IPv6 routers present
Aug 19 14:34:14 turing-police dhcpcd[746]: sending DHCP_REQUEST for 198.82.168.169 to 198.82.247.67 Aug 19 14:34:14 turing-police dhcpcd[746]: sendto: Socket operation on non-socket Aug 19 14:34:14 turing-police dhcpcd[746]: broadcasting DHCP_REQUEST for 198.82.168.169 Aug 19 14:34:14 turing-police dhcpcd[746]: sendto: Socket operation on non-socket Aug 19 14:34:14 turing-police dhcpcd[746]: dhcpStop: ioctl SIOCSIFADDR: Inappropriate ioctl for device Aug 19 14:34:14 turing-police dhcpcd[746]: dhcpStop: ioctl SIOCSIFFLAGS: Inappropriate ioctl for device Aug 19 14:34:14 turing-police dhcpcd[746]: broadcasting DHCP_DISCOVER Aug 19 14:34:14 turing-police dhcpcd[746]: broadcastAddr option is missing in DHCP server response. Assuming 198.82.168.255 Aug 19 14:34:14 turing-police dhcpcd[746]: broadcasting second DHCP_DISCOVER
Aug 19 14:34:14 turing-police dhcpcd[746]: DHCP_OFFER received from (198.82.247.67) Aug 19 14:34:14 turing-police dhcpcd[746]: broadcasting DHCP_REQUEST for 198.82.168.169
Quoted 1 line
- cardmgr messages from system log
Aug 19 14:04:13 turing-police pcmcia: cardmgr. Aug 19 14:04:13 turing-police cardmgr[657]: starting, version is 3.1.31 Aug 19 14:04:13 turing-police rc: Starting pcmcia: succeeded Aug 19 14:04:13 turing-police modprobe: FATAL: Module /dev/cm_657_2 not found. Aug 19 14:04:13 turing-police modprobe: FATAL: Module /dev/cm_657_5 not found. Aug 19 14:04:13 turing-police modprobe: FATAL: Module /dev/cm_657_8 not found. Aug 19 14:04:13 turing-police modprobe: FATAL: Module /dev/cm_657_11 not found. Aug 19 14:04:13 turing-police cardmgr[657]: watching 3 sockets Aug 19 14:04:13 turing-police cardmgr[657]: Card Services release does not match Aug 19 14:04:13 turing-police cardmgr[657]: socket 0: CardBus hotplug device Aug 19 14:04:14 turing-police cardmgr[657]: socket 2: Intersil PRISM2 11 Mbps Wireless Adapter Aug 19 14:04:14 turing-police cardmgr[657]: executing: 'modprobe hermes' Aug 19 14:04:14 turing-police cardmgr[657]: + FATAL: Module hermes not found. Aug 19 14:04:14 turing-police cardmgr[657]: modprobe exited with status 1 Aug 19 14:04:14 turing-police cardmgr[657]: module /lib/modules/2.6.0-test3-mm3/pcmcia/hermes.o not available Aug 19 14:04:14 turing-police cardmgr[657]: executing: 'modprobe orinoco' Aug 19 14:04:14 turing-police cardmgr[657]: + FATAL: Module orinoco not found. Aug 19 14:04:14 turing-police cardmgr[657]: modprobe exited with status 1 Aug 19 14:04:14 turing-police cardmgr[657]: module /lib/modules/2.6.0-test3-mm3/pcmcia/orinoco.o not available Aug 19 14:04:15 turing-police cardmgr[657]: executing: './network start eth1' Aug 19 14:04:15 turing-police /etc/hotplug/net.agent: invoke ifup eth1
Looking at an earlier boot, the only difference seems to be that net.agent is now invoking 'ifup'...
Now to figure out why the 100mbit Ethernet won't answer ARP if the wireless card is up, I suspect it's related to that other thread... :)
Quoted 3 lines · Narayan Desai
cs: warning: no high memory space available!
cs: unable to map card memory!
cs: unable to map card memory!
Hmm, this looks suspicious. For some reason, we can't find any memory resources to map the CIS, so we can't identify the card type, which means we'll probably identify the card as a memory card.
Signature · 2 lines
Russell King ([email protected]) The developer of ARM Linux http://www.arm.linux.org.uk/personal/aboutme.html
thank god its not just my t21 :)
this is the same issue that I've had since somewhere between 2.5.70 2.5.75 (i didn't use any of the ones in between)
Russell - sorry for no tbugging you after OLS :)
sven
Quoted 8 lines · Russell King
Quoted 3 lines · Narayan Desai
cs: warning: no high memory space available!
cs: unable to map card memory!
cs: unable to map card memory!Hmm, this looks suspicious. For some reason, we can't find any memory resources to map the CIS, so we can't identify the card type, which means we'll probably identify the card as a memory card.
Quoted 7 lines · Narayan Desai
Running 2.6.0-test3 (both with and without your recent yenta socket patches) pcmcia cards present during boot don't show up until they are removed and reinserted. Once reinserted, they work fine. This only seems to occur if
yenta_socketis build into the kernel; if support is modular, cards appear to be recognized properly at boot time. I am running on a thinkpad t21. Let me know if I can help with this problem in any way... thanks
As a general thing, when people report this problem (or any other problem), can they please include at least the following details please:
- make/model of machine
- type of cardbus bridge (from lspci)
- type of card (pcmcia or cardbus)
- make/model of card
- full kernel dmesg (including yenta, card services messages)
- cardmgr messages from system log
Signature · 2 lines
Russell King ([email protected]) The developer of ARM Linux http://www.arm.linux.org.uk/personal/aboutme.html
On Tue, 2003-08-19 at 21:45, Russell King asked for:
Quoted 1 line
- make/model of machine
IBM thinkpad t21
Quoted 1 line
- type of cardbus bridge (from lspci)
00:02.0 CardBus bridge: Texas Instruments PCI1450 (rev 03)
Subsystem: IBM: Unknown device 0130
Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop-
ParErr- Stepping- SERR- FastB2B-
Status: Cap+ 66Mhz- UDF- FastB2B- ParErr- DEVSEL=medium >TAbort-
<TAbort- <MAbort- >SERR- <PERR-
Latency: 168, cache line size 08
Interrupt: pin A routed to IRQ 9
Region 0: Memory at 50000000 (32-bit, non-prefetchable)
[size=4K]
Bus: primary=00, secondary=02, subordinate=05, sec-latency=176
Memory window 0: 20000000-203ff000 (prefetchable)
Memory window 1: 20400000-207ff000
I/O window 0: 00001400-000014ff
I/O window 1: 00002400-000024ff
BridgeCtl: Parity- SERR- ISA- VGA- MAbort- >Reset+ 16bInt+PostWrite+
16-bit legacy interface ports at 0001
00:02.1 CardBus bridge: Texas Instruments PCI1450 (rev 03)
Subsystem: IBM: Unknown device 0130
Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop-
ParErr- Stepping- SERR- FastB2B-
Status: Cap+ 66Mhz- UDF- FastB2B- ParErr- DEVSEL=medium >TAbort-
<TAbort- <MAbort- >SERR- <PERR-
Latency: 168, cache line size 08
Interrupt: pin B routed to IRQ 9
Region 0: Memory at 50100000 (32-bit, non-prefetchable)
[size=4K]
Bus: primary=00, secondary=06, subordinate=09, sec-latency=176
Memory window 0: 20800000-20bff000 (prefetchable)
Memory window 1: 20c00000-20fff000
I/O window 0: 00002800-000028ff
I/O window 1: 00002c00-00002cff
BridgeCtl: Parity- SERR- ISA- VGA- MAbort- >Reset+ 16bInt+PostWrite+
16-bit legacy interface ports at 0001
Quoted 1 line
- type of card (pcmcia or cardbus)
cardbus
Quoted 1 line
- make/model of card
cisco aironet 340
Quoted 1 line
- full kernel dmesg (including yenta, card services messages)
Linux Kernel Card Services 3.1.22
options: [pci] [cardbus] [pm]
Intel PCIC probe: not found.
PCI: Found IRQ 9 for device 0000:00:02.0
PCI: Sharing IRQ 9 with 0000:00:05.0
PCI: Sharing IRQ 9 with 0000:01:00.0
Yenta: CardBus bridge found at 0000:00:02.0 [1014:0130]
Yenta IRQ list 00b8, PCI irq9
Socket status: 30000010
PCI: Found IRQ 9 for device 0000:00:02.1
Yenta: CardBus bridge found at 0000:00:02.1 [1014:0130]
Yenta IRQ list 00b8, PCI irq9
Socket status: 30000006
cs: IO port probe 0x0c00-0x0cff: clean.
cs: IO port probe 0x0800-0x08ff: clean.
cs: IO port probe 0x0100-0x04ff: excluding 0x378-0x37f 0x3c0-0x3df 0x4d0-0x4d7
cs: IO port probe 0x0a00-0x0aff: clean.
cs: memory probe 0xa0000000-0xa0ffffff: clean.
(then when i re-insert it)
airo: Doing fast bap_reads
airo: MAC enabled eth0 0:40:96:33:e:a4
eth0: index 0x05: Vcc 5.0, Vpp 5.0, irq 3, io 0x0100-0x013f
bad: scheduling while atomic!
Call Trace: [<c011a487>] schedule+0x3b7/0x3c0 [<e0852d3a>] sendcommand+0xaa/0xe0 [airo] [<e0852c5c>] issuecommand+0x5c/0x90 [airo] [<e085321a>] PC4500_accessrid+0x4a/0x90 [airo] [<c010b855>] error_code+0x2d/0x38 [<e08532c1>] PC4500_readrid+0x61/0x130 [airo] [<e0850661>] readStatsRid+0x31/0x50 [airo] [<e0850d17>] airo_read_stats+0x67/0x150 [airo] [<c01189c8>] pgd_alloc+0x18/0x20 [<c013745c>] find_get_page+0x2c/0x60 [<c013861e>] filemap_nopage+0x28e/0x320 [<c013b131>] buffered_rmqueue+0xd1/0x170 [<c017fd2c>] proc_alloc_inode+0x4c/0x80 [<c017fd2c>] proc_alloc_inode+0x4c/0x80 [<c016afec>] alloc_inode+0x1c/0x140 [<c013ae60>] __rmqueue+0xd0/0x120 [<c013b131>] buffered_rmqueue+0xd1/0x170 [<c0202466>] vsnprintf+0x246/0x470 [<e0850e16>] airo_get_stats+0x16/0x20 [airo] [<c0299bbb>] dev_seq_printf_stats+0xeb/0x100 [<c0299bf8>] dev_seq_show+0x28/0x90 [<c01705ad>] seq_read+0x1dd/0x310 [<c0152498>] vfs_read+0xb8/0x130 [<c0152742>] sys_read+0x42/0x70 [<c010ae4b>] syscall_call+0x7/0xb
bad: scheduling while atomic!
Call Trace: [<c011a487>] schedule+0x3b7/0x3c0 [<e0852d3a>] sendcommand+0xaa/0xe0 [airo] [<e0852c5c>] issuecommand+0x5c/0x90 [airo] [<e085321a>] PC4500_accessrid+0x4a/0x90 [airo] [<e08532c1>] PC4500_readrid+0x61/0x130 [airo] [<e0850661>] readStatsRid+0x31/0x50 [airo] [<e0850d17>] airo_read_stats+0x67/0x150 [airo] [<c0202466>] vsnprintf+0x246/0x470 [<e0850e16>] airo_get_stats+0x16/0x20 [airo] [<c0299bbb>] dev_seq_printf_stats+0xeb/0x100 [<c0299bf8>] dev_seq_show+0x28/0x90 [<c01705ad>] seq_read+0x1dd/0x310 [<c0152498>] vfs_read+0xb8/0x130 [<c0152742>] sys_read+0x42/0x70 [<c010ae4b>] syscall_call+0x7/0xb
Quoted 1 line
- cardmgr messages from system log
Aug 20 09:37:19 sven cardmgr[322]: starting, version is 3.2.2 Aug 20 09:37:19 sven cardmgr[322]: socket 0: Anonymous Memory Aug 20 09:37:19 sven cardmgr[322]: executing: 'modprobe memory_cs' Aug 20 09:37:19 sven cardmgr[322]: + FATAL: Module memory_cs not found. Aug 20 09:37:19 sven cardmgr[322]: modprobe exited with status 1 Aug 20 09:37:19 sven cardmgr[322]: module /lib/modules/2.6.0-test3/pcmcia/memory_cs.o not available Aug 20 09:37:19 sven cardmgr[322]: bind 'memory_cs' to socket 0 failed: Invalid argument Aug 20 09:37:19 sven cardmgr[322]: socket 0: Anonymous Memory Aug 20 09:38:35 sven cardmgr[322]: executing: 'modprobe -r memory_cs' Aug 20 09:38:35 sven cardmgr[322]: + FATAL: Module memory_cs not found. Aug 20 09:38:35 sven cardmgr[322]: modprobe exited with status 1 Aug 20 09:38:44 sven cardmgr[322]: socket 0: Aironet PC4800 Aug 20 09:38:44 sven cardmgr[322]: executing: 'modprobe airo_cs' Aug 20 09:38:45 sven cardmgr[322]: executing: './network start eth0' Aug 20 09:38:45 sven cardmgr[322]: + Ignoring unknown interface eth0=eth0.
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
"Russell" == Russell King <[email protected]> writes:
Russell> On Mon, Aug 18, 2003 at 07:34:59PM -0500, Narayan Desai
Quoted 7 lines · Russell> wrote:
Quoted 7 lines
Running 2.6.0-test3 (both with and without your recent yenta socket patches) pcmcia cards present during boot don't show up until they are removed and reinserted. Once reinserted, they work fine. This only seems to occur if
yenta_socketis build into the kernel; if support is modular, cards appear to be recognized properly at boot time. I am running on a thinkpad t21. Let me know if I can help with this problem in any way... thanks
Russell> As a general thing, when people report this problem (or any Russell> other problem), can they please include at least the Russell> following details please:
Russell> - make/model of machine
Russell> - type of cardbus bridge (from lspci) Russell> - type of card (pcmcia or cardbus) Russell> - make/model of card
Russell> - full kernel dmesg (including yenta, card services
Russell> messages)
Russell> - cardmgr messages from system log
This is an IBM thinkpad t21. the cardbus bridge is:
00:02.0 CardBus bridge: Texas Instruments PCI1450 (rev 03) 00:02.1 CardBus bridge: Texas Instruments PCI1450 (rev 03)
The problem is occurring with an orinoco gold pcmcia card.
the unable to map cs memory errors were the only odd thing in the kernel messages, and it looks like your diagnosis was right on the money. i these cardmgr messages:
Aug 18 20:38:06 topaz cardmgr[374]: watching 2 sockets Aug 18 20:38:06 topaz cardmgr[375]: starting, version is 3.2.2 Aug 18 20:38:06 topaz cardmgr[375]: socket 0: Anonymous Memory Aug 18 20:38:06 topaz cardmgr[375]: executing: 'modprobe memory_cs' Aug 18 20:38:06 topaz cardmgr[375]: + FATAL: Module memory_cs not found. Aug 18 20:38:06 topaz cardmgr[375]: modprobe exited with status 1 Aug 18 20:38:06 topaz cardmgr[375]: module /lib/modules/2.6.0-test3-p2/pcmcia/memory_cs.o not available Aug 18 20:38:06 topaz cardmgr[375]: bind 'memory_cs' to socket 0 failed: Invalid argument
let me know if there is more info that would be useful.
-nld
Quoted 1 line · Russell King
- make/model of machine
IBM thinkpad t21
Quoted 1 line
- type of cardbus bridge (from lspci)
00:02.0 CardBus bridge: Texas Instruments PCI1450 (rev 03)
Subsystem: IBM: Unknown device 0130
Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop-
ParErr- Stepping- SERR- FastB2B-
Status: Cap+ 66Mhz- UDF- FastB2B- ParErr- DEVSEL=medium >TAbort-
<TAbort- <MAbort- >SERR- <PERR-
Latency: 168, cache line size 08
Interrupt: pin A routed to IRQ 9
Region 0: Memory at 50000000 (32-bit, non-prefetchable)
[size=4K]
Bus: primary=00, secondary=02, subordinate=05, sec-latency=176
Memory window 0: 20000000-203ff000 (prefetchable)
Memory window 1: 20400000-207ff000
I/O window 0: 00001400-000014ff
I/O window 1: 00002400-000024ff
BridgeCtl: Parity- SERR- ISA- VGA- MAbort- >Reset+ 16bInt+PostWrite+
16-bit legacy interface ports at 0001
00:02.1 CardBus bridge: Texas Instruments PCI1450 (rev 03)
Subsystem: IBM: Unknown device 0130
Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop-
ParErr- Stepping- SERR- FastB2B-
Status: Cap+ 66Mhz- UDF- FastB2B- ParErr- DEVSEL=medium >TAbort-
<TAbort- <MAbort- >SERR- <PERR-
Latency: 168, cache line size 08
Interrupt: pin B routed to IRQ 9
Region 0: Memory at 50100000 (32-bit, non-prefetchable)
[size=4K]
Bus: primary=00, secondary=06, subordinate=09, sec-latency=176
Memory window 0: 20800000-20bff000 (prefetchable)
Memory window 1: 20c00000-20fff000
I/O window 0: 00002800-000028ff
I/O window 1: 00002c00-00002cff
BridgeCtl: Parity- SERR- ISA- VGA- MAbort- >Reset+ 16bInt+PostWrite+
16-bit legacy interface ports at 0001
Quoted 3 lines
- type of card (pcmcia or cardbus)
- make/model of card
cisco aironet 340
Quoted 3 lines
- full kernel dmesg (including yenta, card services messages)
- cardmgr messages from system log
Quoted 12 lines · On Tue, 19 Aug 2003 12:45:47 +0100 Russell King <[email protected]> wrote:
Quoted 7 lines · Narayan Desai
Running 2.6.0-test3 (both with and without your recent yenta socket patches) pcmcia cards present during boot don't show up until they are removed and reinserted. Once reinserted, they work fine. This only seems to occur if
yenta_socketis build into the kernel; if support is modular, cards appear to be recognized properly at boot time. I am running on a thinkpad t21. Let me know if I can help with this problem in any way... thanksAs a general thing, when people report this problem (or any other problem), can they please include at least the following details please:
Quoted 6 lines
- make/model of machine
- type of cardbus bridge (from lspci)
- type of card (pcmcia or cardbus)
- make/model of card
- full kernel dmesg (including yenta, card services messages)
- cardmgr messages from system log
I see something similar on my Acer Aspire 1300XV.
On 2.4 everything seems to be working properly (except that i have to eject and insert it twice to get the card up and running (more people have reported this issue, I think). In 2.6 things aren't working that well..
/* insertion of card */
Aug 19 15:35:07 [cardmgr] socket 0: Anonymous Memory Aug 19 15:35:07 [cardmgr] executing: 'modprobe sram_mtd' Aug 19 15:35:07 [cardmgr] + FATAL: Module sram_mtd not found. Aug 19 15:35:07 [cardmgr] modprobe exited with status 1 Aug 19 15:35:07 [cardmgr] module /lib/modules/2.6.0-test3/pcmcia/sram_mtd.o not available Aug 19 15:35:07 [cardmgr] Common memory region at 0x4000: Generic or SRAM Aug 19 15:35:07 [cardmgr] Common memory region at 0x0: Generic or SRAM Aug 19 15:35:07 [cardmgr] bind MTD 'sram_mtd' to socket 0 failed: Invalid argument Aug 19 15:35:07 [cardmgr] executing: 'modprobe memory_cs' Aug 19 15:35:07 [cardmgr] + FATAL: Module memory_cs not found. Aug 19 15:35:07 [cardmgr] modprobe exited with status 1 Aug 19 15:35:07 [cardmgr] module /lib/modules/2.6.0-test3/pcmcia/memory_cs.o not available Aug 19 15:35:07 [cardmgr] bind 'memory_cs' to socket 0 failed: Invalid
argument
/* and here I eject: */
Aug 19 15:36:31 [cardmgr] unbind MTD 'sram_mtd' from socket 0 failed: No such device Aug 19 15:36:31 [cardmgr] executing: 'modprobe -r memory_cs' Aug 19 15:36:31 [cardmgr] + FATAL: Module memory_cs not found. Aug 19 15:36:31 [cardmgr] modprobe exited with status 1 Aug 19 15:36:31 [cardmgr] executing: 'modprobe -r sram_mtd' Aug 19 15:36:31 [cardmgr] + FATAL: Module sram_mtd not found. Aug 19 15:36:31 [cardmgr] modprobe exited with status 1
And it keeps doing that until I restart cardmgr. After that:
/* Restart of cardmgr */
Aug 19 15:37:52 [cardmgr] exiting Aug 19 15:37:54 [cardmgr] watching 1 sockets Aug 19 15:37:54 [kernel] cs: IO port probe 0x0c00-0x0cff: clean. Aug 19 15:37:54 [kernel] cs: IO port probe 0x0800-0x08ff: clean. Aug 19 15:37:54 [kernel] cs: IO port probe 0x0100-0x04ff: excluding 0x2c8-0x2cf 0x378-0x37f 0x4d0-0x4d7 Aug 19 15:37:54 [kernel] cs: IO port probe 0x0a00-0x0aff: clean. Aug 19 15:37:54 [cardmgr] starting, version is 3.2.4 Aug 19 15:37:59 [su(pam_unix)] session closed for user root Aug 19 15:39:13 [cardmgr] socket 0: NETGEAR MA401RA 11Mbps 802.11 WLAN Card Aug 19 15:39:13 [cardmgr] executing: 'modprobe orinoco_cs' Aug 19 15:39:14 [cardmgr] executing: './network start eth1'
And it's working properly. Except for the insert-twice-issue.. dmesg attached.
Here is lspci -vvv for the cardbus:
00:0a.0 CardBus bridge: O2 Micro, Inc. OZ6912 Cardbus Controller
Subsystem: Acer Incorporated [ALI]: Unknown device 001d
Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop-
ParErr- Stepping+ SERR- FastB2B-
Status: Cap+ 66Mhz- UDF- FastB2B- ParErr- DEVSEL=slow >TAbort-
<TAbort- <MAbort- >SERR- <PERR-
Latency: 168
Interrupt: pin A routed to IRQ 11
Region 0: Memory at 10000000 (32-bit, non-prefetchable) [size=4K]
Bus: primary=00, secondary=02, subordinate=05, sec-latency=176
Memory window 0: 10400000-107ff000 (prefetchable)
Memory window 1: 10800000-10bff000
I/O window 0: 00001000-000010ff
I/O window 1: 00001400-000014ff
BridgeCtl: Parity- SERR- ISA- VGA- MAbort- >Reset- 16bInt+PostWrite+
16-bit legacy interface ports at 0001
uname -a :
Linux vega 2.6.0-test3 #6 Fri Aug 15 13:00:17 CEST 2003 i686 mobile AMD Athlon(tm) XP 1400+ AuthenticAMD GNU/Linux
Thanks.
Signature · 1 line
Henrik Persson [email protected]