All of lore.kernel.org
 help / color / mirror / Atom feed
* Porting Question - new board. Need some help.
@ 2003-09-11  5:42 Narendra Sankar
  0 siblings, 0 replies; only message in thread
From: Narendra Sankar @ 2003-09-11  5:42 UTC (permalink / raw)
  To: linux-mips

Hi everyone
I am helping a friend port the kernel to a new board that he has built
with a PMC R5231A cpu. The system controller is a custom design (it has
a standard 16550A uart and the interrupts are directly routed to the
cpu). I am currently working off linux_2_4_21 checked out from CVS.
Following the instructions from Jun Sun
(http://linux.junsun.net/porting-howto/) I get to the following point
where things seem to break -

-----------------------------------------------------------
CPU revision is: 00002831
FPU revision is: 00002830
Primary instruction cache 32kB, physically tagged, 2-way, linesize 32
bytes. 
Primary data cache 32kB 2-way, linesize 32 bytes.
Linux version 2.4.21 (naren@laptop) (gcc version 2.95.4 20010319
(prerelease)) 3 
Determined physical RAM map:
 memory: 04000000 @ 00000000 (usable)
Initial ramdisk at: 0x80276000 (1765376 bytes)
On node 0 totalpages: 16384
zone(0): 16384 pages.
zone(1): 0 pages.
zone(2): 0 pages.
Kernel command line: console=ttyS0,115200 kgdb=ttyS0,115200
ramdisk_size=8192 root=/dev/ram*
Calibrating delay loop... 748.74 BogoMIPS 
Memory: 61092k/65536k available (705k kernel code, 4444k reserved, 1848k
data, ) 
Dentry cache hash table entries: 8192 (order: 4, 65536 bytes)
Inode cache hash table entries: 4096 (order: 3, 32768 bytes) 
Mount cache hash table entries: 512 (order: 0, 4096 bytes) 
Buffer-cache hash table entries: 4096 (order: 2, 16384 bytes) 
Page-cache hash table entries: 16384 (order: 4, 65536 bytes) 
Checking for 'wait' instruction... available. 
POSIX conformance testing by UNIFIX 
Linux NET4.0 for Linux 2.4 Based upon Swansea University Computer
Society NET3.039 
Starting kswapd 
pty: 256 Unix98 ptys configured 
Serial driver version 5.05c(2001-07-08) with no serial options enabled
Generic MIPS RTC Driver v1.0 
RAMDISK driver initialized: 16 RAM disks of 8192K size 1024 blocksize
loop: loaded (max 8 devices) 
RAMDISK: Compressed image found at block 0
Freeing initrd memory: 1724k freed 
VFS: Mounted root (ext2 filesystem) readonly. 
Freeing unused kernel memory: 48k freed 
EXT2-fs warning: mounting unchecked fs, running e2fsck is recommended
-------------------------------------------------------

I am using the big-endian ramdisk from the debian mips distribution for
test purposes. However I do not get to any sort of console prompt. The
system just seems to hang at that point. (Using the QED ITE board, I get
to the point where the debian setup starts.)

The toolchain used to build the kernel is the mips-linux toolchain from
ftp.linux-mips.org-

Reading specs from /usr/lib/gcc-lib/mips-linux/2.95.4/specs
gcc version 2.95.4 20010319 (prerelease)

Trying to debug using kgdb, I can break right before the execve of
/sbin/init. However I cannot step or debug beyond that point. (Using gdb
5.3 built with target=mips-linux). So I know that I get to the exec of
init. However beyond that I do not know what is happening. Using printks
I determined that the system at least enters load_elf_binary.

Anyone have any ideas where I can look for what is going wrong? Also I
am wondering why I cannot use gdb to step into execve? I followed the
porting guide instructions in order to set up kgdb. I can break and step
before the call to init but nothing after that.


Thanks a lot for any help

Naren Sankar

^ permalink raw reply	[flat|nested] only message in thread

only message in thread, other threads:[~2003-09-11  5:42 UTC | newest]

Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2003-09-11  5:42 Porting Question - new board. Need some help Narendra Sankar

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.