Weird PCMCIA behavior
Note: this is the same laptop mentioned above in the Maestro 2E thread above, so the same weird-ass power behavior applies here.
When on AC power, everything is dandy with PCMCIA. When on battery power, PCMCIA device detection fails, emanating a lower than normal pitched beep, followed by an even lower beep. However, if the apm module is inserted, this makes it behave properly, but ONLY if the apm module was compiled with 'Make CPU idle calls when idle'. Yeah, I know it's ucked fup.
Some possibly relevant details:
2.4.{14,15-pre{3,5}}
pcmcia-cs-3.1.29
lspci -vv:
00:0b.0 CardBus bridge: Toshiba America Info Systems ToPIC97 (rev 05)
Subsystem: Toshiba America Info Systems: Unknown device 0001
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: 0
Interrupt: pin A routed to IRQ 11
Region 0: Memory at 10000000 (32-bit, non-prefetchable) [size=4K]
Bus: primary=00, secondary=14, subordinate=14, sec-latency=0
Memory window 0: 10400000-107ff000 (prefetchable)
Memory window 1: 10800000-10bff000
I/O window 0: 00004000-000040ff
I/O window 1: 00004400-000044ff
BridgeCtl: Parity- SERR- ISA- VGA- MAbort- >Reset- 16bInt+ PostWrite+
16-bit legacy interface ports at 0001
Signature · 1 line
Josh Litherland ([email protected])
Quoted 11 lines · Faux Pas III
Note: this is the same laptop mentioned above in the Maestro 2E thread above, so the same weird-ass power behavior applies here.
When on AC power, everything is dandy with PCMCIA. When on battery power, PCMCIA device detection fails, emanating a lower than normal pitched beep, followed by an even lower beep. However, if the apm module is inserted, this makes it behave properly, but ONLY if the apm module was compiled with 'Make CPU idle calls when idle'. Yeah, I know it's ucked fup.
Quoted 4 lines
Some possibly relevant details:
2.4.{14,15-pre{3,5}}
pcmcia-cs-3.1.29
pcmcia-cs problems are reported to http://pcmcia-cs.sourceforge.net/
We encourage you to use the kernel's cardbus code instead :) (CONFIG_PCMCIA and CONFIG_CARDBUS)
As a side note, with kernel cardbus support, you should no longer need external utilities or external drivers. It should Just Work(tm).
Signature · 3 lines
Jeff Garzik | Only so many songs can be sung Building 1024 | with two lips, two lungs, and one tongue. MandrakeSoft | - nomeansno
Quoted 3 lines · Jeff Garzik
pcmcia-cs problems are reported to http://pcmcia-cs.sourceforge.net/
We encourage you to use the kernel's cardbus code instead :) (CONFIG_PCMCIAandCONFIG_CARDBUS)
That's what I'm using to get these errors. I think, although I haven't fully tested yet, that pcmcia-cs's core works fine in all power states.
Quoted 2 lines
As a side note, with kernel cardbus support, you should no longer need external utilities or external drivers. It should Just Work(tm).
Do I not still need the cardmgr and all that rot from pcmcia-cs ? That's what I've been using for detection, bind cards to specific modules, etc.
Signature · 1 line
Josh Litherland ([email protected])
Quoted 7 lines · Faux Pas III
Quoted 3 lines · Jeff Garzik
pcmcia-cs problems are reported to http://pcmcia-cs.sourceforge.net/
We encourage you to use the kernel's cardbus code instead :) (CONFIG_PCMCIAandCONFIG_CARDBUS)
Quoted 8 lines
That's what I'm using to get these errors. I think, although I haven't fully tested yet, that pcmcia-cs's core works fine in all power states.
Quoted 2 lines
As a side note, with kernel cardbus support, you should no longer need external utilities or external drivers. It should Just Work(tm).
Do I not still need the cardmgr and all that rot from pcmcia-cs ? That's what I've been using for detection, bind cards to specific modules, etc.
Nope. CardBus looks like hotplug PCI to the kernel, so all normal PCI drivers automagically work as CardBus drivers. You actually need no userspace tools at all..
Also note there is a kernel cardbus PM fix in the latest 2.4.15-preXX patches...
Signature · 3 lines
Jeff Garzik | Only so many songs can be sung Building 1024 | with two lips, two lungs, and one tongue. MandrakeSoft | - nomeansno
Quoted 3 lines · Jeff Garzik
Nope. CardBus looks like hotplug PCI to the kernel, so all normal PCI drivers automagically work as CardBus drivers. You actually need no userspace tools at all..
In my undertanding, if you compile drivers for your cardbus and for attached peripherals as modules (and that make sense for cardbus hardware), then you need the 'hotplug' userspace tool.
This tool is informed by the kernel that a new device is in and then is responsible to insmod the correct driver(s) (mainly by looking in modules.pcimap and friends) and to setup things like setup a network when a network card is inserted ...
Is that wrong ?
Christophe
Signature · 2 lines
Christophe Barbé <[email protected]> GnuPG FingerPrint: E0F6 FADF 2A5C F072 6AF8 F67A 8F45 2F1E D72C B41E