**** BEGIN LOGGING AT Wed Jul 05 02:59:56 2006
Jul 05 05:58:41 <jr--> rwhitby. problematic part is in ixp400_eth.c in section that configures ports and mii stuff. to me it looks like only first match will be included if kernel is compiled for multiple machine ids. it's already like that so if you set CONFIG_ARCH_IXDP425 and CONFIG_MACH_IXDPG425 on config it will fail.
Jul 05 05:59:58 <rwhitby> Ah, yes.  That would be true.  Is the config for the fsg the same as the nslu2?
Jul 05 06:00:24 <jr--> no it's not. fsg has multiple ethernet interfaces
Jul 05 06:00:48 <rwhitby> and they're connected directly to the ixp processor, not through a switch chip?
Jul 05 06:03:10 <jr--> thru switch.
Jul 05 06:05:25 <jr--> i followed traces on pcb to verify which external port is connected to which switch chip pins and checked datasheets. so that part should be correct. however i'm not sure if rest if correct. mii-tool shows ok info and all ports work so it can't be that much off. it's different from any existing ixp4xx platform so it i think it needs fsg specific patch to ixp400_eth.c
Jul 05 06:08:01 <rwhitby> But the ixp400_eth only pertains to the on-chip ethernet port, which presumably is being used to talk to the switch chip.
Jul 05 06:08:37 <jr--> these two parts: static int phyAddresses[IXP425_ETH_ACC_MII_MAX_ADDR] and static phy_cfg_t default_phy_cfg[] must be changed.
Jul 05 06:08:58 <jr--> first is same as IXDPG425. second is fsg specific
Jul 05 06:09:06 <rwhitby> are there module params for that?
Jul 05 06:09:49 <jr--> hmm. didn't even check. i do know that i need to pas no_phy_scan=1 (or something like that) when loading. otherwise it fails.
Jul 05 06:12:41 <jr--> looks like there's no suitable module parameters
Jul 05 06:21:14 <rwhitby> jr--: patch it so it works, and then we'll work out how to incorporate that patch with the other machine id's later.
Jul 05 06:21:23 <rwhitby> back later
Jul 05 06:21:28 <jr--> there's another file with similar problem IxEthMii.c that's part of ixp4xx-csr. since switch chip user (RTL8305SB) can't be autodetected it's hardcoded on intel source. montejade uses same chip.
Jul 05 06:21:47 <jr--> rwhitby. ok. thanks.
Jul 05 06:26:35 <jr--> nslu2-linux wiki mentions "gpio" util. is it available for download somewhere?
Jul 05 08:39:40 <gilligan_> hi
Jul 05 10:18:14 <placid_> jr--: howdy, where did you see gpio reference on the wiki? plenty of mentions of hardware gpio
Jul 05 10:38:14 <jr--> placid. http://www.nslu2-linux.org/wiki/NAS100d/GPIOConnections
Jul 05 11:14:55 <CIA-5> 03rwhitby * r3547 10optware/trunk/Makefile: Makefile: added option for native compilation on ds101g+
Jul 05 11:24:26 <cyanescent> anyone tried compiling mailman ?
Jul 05 15:16:18 <CIA-5> 03oleo * r3548 10optware/trunk/ (6 files in 2 dirs): buildroot: initial import of uClibc toolchain for mipsel
Jul 05 16:09:56 <CIA-5> 03bzhou * r3549 10optware/trunk/make/mailman.mk: mailman now passes optware-check
Jul 05 16:11:34 <CIA-5> 03bzhou * r3550 10optware/trunk/Makefile: added mailman, it builds ok, needs runtime testing
Jul 05 16:53:43 <jimthfreeman> hi
Jul 05 16:53:50 <jimthfreeman> have a question related to ARM devices, what JTAG debugger/emulator is considered verry good at this moment?
Jul 05 18:31:23 <CIA-5> 03gda * r3551 10optware/trunk/make/bzip2.mk: bzip2: bzcat was missing\!
Jul 05 19:14:52 <gda_> re
Jul 05 19:15:37 <gda_> I don't know what is different, but today motion does find the libmysqlclient.so with
Jul 05 19:15:47 <gda_> MOTION_LDFLAGS="-Wl,-rpath,/usr/syno/mysql/lib/mysql"
Jul 05 20:35:01 <CIA-5> 03gda * r3552 10optware/trunk/ (6 files in 2 dirs): motion: new motion detection
Jul 05 20:37:01 <CIA-5> 03gda * r3553 10optware/trunk/Makefile: motion: ready for testing
Jul 05 20:47:12 <gda_> any cyrus-imap users here?
Jul 05 20:48:29 <gda_> is there a reason why the saslauthd is not part of the ipk?
Jul 05 20:59:03 <gda_> stupid me, saslauthd is of course in cyrus-sasl
**** ENDING LOGGING AT Thu Jul 06 02:59:57 2006