On Sun, 2004-02-29 at 11:43, digger vermont wrote: > It comes from pcmcia-cs. This is all without any card in the slot. > 2.6.3-ben2:
Weird... Make sure you are properly using the userland pcmcia-cs compiled for kernel drivers and not its own userland stuff. > Feb 28 18:33:08 [sudo] digger : TTY=pts/2 ; PWD=/home/digger ; USER=root > ; COMMAND=/etc/init.d/pcmcia restart > Feb 28 18:33:08 [cardmgr] exiting > Feb 28 18:33:08 [cardmgr] watching 1 socket > Feb 28 18:33:08 [kernel] cs: IO port probe 0x0c00-0x0cff:<7>IN from bad > port c00 at e2109c38 > Feb 28 18:33:08 [kernel] IN from bad port c01 at e2109c5c > Feb 28 18:33:08 [kernel] cs: IO port probe 0x0800-0x08ff:<7>IN from bad > port 800 at e2109c38 > Feb 28 18:33:08 [kernel] IN from bad port 801 at e2109c5c > Feb 28 18:33:08 [kernel] IN from bad port 808 at e2109c38 > Feb 28 18:33:08 [kernel] cs: IO port probe 0x0100-0x04ff:<7>IN from bad > port 100 at e2109c38 > Feb 28 18:33:08 [kernel] IN from bad port 101 at e2109c5c > Feb 28 18:33:08 [kernel] IN from bad port 3b8 at e2109c38 > Feb 28 18:33:08 [kernel] IN from bad port 3c8 at e2109d1c > Feb 28 18:33:08 [kernel] IN from bad port 3d0 at e2109d1c > Feb 28 18:33:08 [cardmgr] starting, version is 3.2.5 I don't see why the PCMCIA driver would probe those ports with nothing plugged... > lsmod > Module Size Used by > radeon 132204 2 > ds 15076 0 > yenta_socket 16288 0 > pcmcia_core 75256 2 ds,yenta_socket The above sounds fine. yenta_socket is what you want indeed, just make sure it's not trying to load another host driver during boot... > cat /etc/pcmcia/config.opts > # > # Local PCMCIA Configuration File > > include port 0x100-0x4ff, port 0x800-0x8ff, port 0xc00-0xcff > include memory 0x80080000-0x8008ffff That doesn't sound too bad neither... Actually, you can probably define a single range of 0...0xfff for IOs ... Things definitely don't behave this way for me (2.6.3-ben2 with debian sid) on my tipb. Ben.
