Hi, Am 28.10.2010 02:58, schrieb Larry Finger: > On 10/27/2010 01:52 PM, Arnd Hannemann wrote: >> first of all: great! b43 sdio card works with open firmware! >> >> However, on "rmmod b43" I get the following backtraces >> with current Linus git (v2.6.36-6794-g12ba8d1): >> >> >> [ 831.696189] b43-phy1 debug: Removing Interface type 2 >> [ 831.698425] b43-phy1 debug: Wireless interface stopped >> [ 851.560519] ------------[ cut here ]------------ >> [ 851.560531] WARNING: at drivers/mmc/core/core.c:237 mmc_wait_for_cmd+0x88/0x90() >> [ 851.560534] Hardware name: 20552PG >> [ 851.560536] Modules linked in: b43(-) ssb mmc_block binfmt_misc rfcomm sco bnep ppdev l2cap ipt_MASQUERADE iptable_nat nf_nat nf_conntrack_ipv4 nf_defrag_ipv4 xt_state nf_conntrack ipt_REJECT xt_tcpudp iptable_filter ip_tables x_tables bridge stp kvm_intel kvm arc4 iwlagn snd_hda_codec_conexant snd_hda_intel snd_hda_codec iwlcore snd_hwdep snd_pcm thinkpad_acpi mac80211 snd_seq_midi snd_rawmidi snd_seq_midi_event snd_seq r852 joydev snd_timer sm_common pcmcia nand snd_seq_device cfg80211 sdhci_pci btusb psmouse tpm_tis yenta_socket nand_ids lp snd pcmcia_rsrc nand_ecc bluetooth sdhci tpm pcmcia_core parport mtd snd_page_alloc serio_raw tpm_bios soundcore nvram led_class sha256_generic aes_i586 aes_generic dm_crypt i915 drm_kms_helper drm ahci intel_agp i2c_algo_bit intel_gtt e1000e libahci video agpgart output >> [ 851.560620] Pid: 2504, comm: rmmod Not tainted 2.6.36-titan0+ #1 >> [ 851.560622] Call Trace: >> [ 851.560631] [<c014a102>] warn_slowpath_common+0x72/0xa0 >> [ 851.560636] [<c04d94c8>] ? mmc_wait_for_cmd+0x88/0x90 >> [ 851.560641] [<c04d94c8>] ? mmc_wait_for_cmd+0x88/0x90 >> [ 851.560645] [<c014a152>] warn_slowpath_null+0x22/0x30 >> [ 851.560649] [<c04d94c8>] mmc_wait_for_cmd+0x88/0x90 >> [ 851.560655] [<c0401585>] ? device_release+0x25/0x80 >> [ 851.560660] [<c04df210>] mmc_io_rw_direct_host+0xa0/0x150 >> [ 851.560665] [<c04df370>] mmc_io_rw_direct+0x30/0x40 >> [ 851.560669] [<c04e06e7>] sdio_disable_func+0x37/0xa0 >> [ 851.560683] [<f8dfcb80>] b43_sdio_remove+0x30/0x50 [b43] >> [ 851.560687] [<c04df8cc>] sdio_bus_remove+0x1c/0x60 >> [ 851.560692] [<c016d39f>] ? blocking_notifier_call_chain+0x1f/0x30 >> [ 851.560697] [<c0404991>] __device_release_driver+0x51/0xb0 >> [ 851.560701] [<c0404a7f>] driver_detach+0x8f/0xa0 >> [ 851.560705] [<c0403c83>] bus_remove_driver+0x63/0xa0 >> [ 851.560709] [<c0405039>] driver_unregister+0x49/0x80 >> [ 851.560713] [<c0405039>] ? driver_unregister+0x49/0x80 >> [ 851.560718] [<c04dfad7>] sdio_unregister_driver+0x17/0x20 >> [ 851.560727] [<f8dfcb42>] b43_sdio_exit+0x12/0x20 [b43] >> [ 851.560734] [<f8dfe76f>] b43_exit+0x17/0x3c [b43] >> [ 851.560740] [<c017fb8d>] sys_delete_module+0x13d/0x200 >> [ 851.560747] [<c01fd7d2>] ? do_munmap+0x212/0x300 >> [ 851.560752] [<c010311f>] sysenter_do_call+0x12/0x28 >> [ 851.560757] ---[ end trace 31e14488072d2f7d ]--- >> [ 851.560759] ------------[ cut here ]------------ > > I'm not an expert on the sdio interface, but I'll take a crack at this. > > The first WARN_ON comes because the host has not been claimed. Please try the > following patch so that we know if it is claimed when b43_sdio_remove() is > entered and ssb_unregister_bus() clears it, or it has not been claimed when the > routine is entered. If the latter, we will likely need to surround the > sdio_disable_func() calls with sdio_claim_host()...sdio_remove_host() statements. I get the following output from your patch when unloading b43: [ 1364.620044] b43-phy1 debug: Removing Interface type 2 [ 1364.621812] b43-phy1 debug: Wireless interface stopped [ 1372.440297] b43-sdio: before ssb_bus_unregister, claimed is 0 [ 1372.440345] b43-sdio: before sdio_disable_func, claimed is 0 [ 1372.440351] ------------[ cut here ]------------ [ 1372.440367] WARNING: at drivers/mmc/core/core.c:237 mmc_wait_for_cmd+0x88/0x90() Regards Arnd -- To unsubscribe from this list: send the line "unsubscribe linux-mmc" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
