anyone used kgdb lately?
I've been trying to set up my APUS and my PC so I can run a kgdb kernel on
APUS to possibly figure out (or at least gather enough information so someone else can help) why my A4000T SCSI locks up randomly when using APUS. Note I've checked and rechecked termination and hardware issues and
am assuming that that's not the problem (for better or worse). I've built gdb-4.17
with target powerpc-linux and installed it on my PC. I've connected a null modem
cable between the PC and the Amiga and tested it using kermit. I used a non-stripped
powerpc kernel as input to gdb and started up the stripped version on the Amiga.
I put the command line option debug=/dev/ttyS0 in the boothack script and tried
to start the kernel. Here's where things go sour. The stripped kernel
doesn't start
at all. By reading the docs in arch/m68k/kernel/kgdb it seems to say that the kernel
should work normally if the above debug option is removed. I tried removing the
debug=/dev/ttyS0 option from the command line and still the kernel does not boot.
My question is, has anyone used the kgdb option on APUS or any PPC lately
and does it work? I've compiled several kernels (using the same source)
without kgdb support and they have worked fine. I assume The kgdb code tries to initize the serial port first thing. Is that a good place to start
troubleshooting?
Fred
[[ This message was sent via the linuxppc-dev mailing list. Replies are ]] [[ not forced back to the list, so be sure to Cc linuxppc-dev if your ]] [[ reply is of general interest. To unsubscribe from linuxppc-dev, send ]] [[ the message 'unsubscribe' to [email protected] ]]
new irq.c
Damn list won't let me send the file itself (too large) so it's at ftp://linuxppc.cs.nmt.edu/misc/irq.c
Please try it out.
[[ This message was sent via the linuxppc-dev mailing list. Replies are ]] [[ not forced back to the list, so be sure to Cc linuxppc-dev if your ]] [[ reply is of general interest. To unsubscribe from linuxppc-dev, send ]] [[ the message 'unsubscribe' to [email protected] ]]
Re: List limits (was new irq.c)
[ Cort Dougan writes: ]
Quoted 2 lines
Damn list won't let me send the file itself (too large)
Actually it's the damn listmaster. :-) But it's your (meaning all of you) list and my policies are there to try to protect your mailboxes from the list. Given the number of subscribers that complain to me that they can't handle the volume of mail and the number of bounces from mailboxes exceeding quotas, I believe some sort of message size limit is desirable. Currently that limit is 40,000 bytes for the entire message. Is that too small?
Message size really doesn't address the larger issue of mailing files to a mail list. Do all the 1100 or so recipients of linuxppc-dev and linuxppc-dev-digest *really* want to receive every file/mega-patch potentially mailed to the list? Although all may want to know what's going on (or they wouldn't be subscribed), I suspect very few are eager to receive every file to build into their system, but every one of them pays for those few to receive the files if they're mailed, nonetheless. On the other hand, it's no stretch to claim that a development list exists to help exchange experimental code and bug fixes, but can't that be done more reliably by posting URLs where files can be retrieved via ftp?
I administer these lists to try to contribute to something I feel is important, so I'm open to (reasonable) suggestions. If you wish to express an opinion, I encourage you to respond to me directly since this has little to do with Linux development, but use your judgement if you think your response merits discussion. Now where did I put my asbestos underwear...?
Signature · 6 lines
Dave Wolfe [[ This message was sent via the linuxppc-dev mailing list. Replies are ]] [[ not forced back to the list, so be sure to Cc linuxppc-dev if your ]] [[ reply is of general interest. To unsubscribe from linuxppc-dev, send ]] [[ the message 'unsubscribe' to [email protected] ]]