[haiku-development] State of ppc
- From: Steven Hoefel <stevenhoefel@xxxxxxxxxxx>
- To: <haiku-development@xxxxxxxxxxxxx>
- Date: Sun, 9 May 2010 21:38:46 +1000
To all,
I've been working recently on getting to know haiku-ppc as I've just come
across an old 1.25ghz, 256mb RAM eMac. Firstly I had a look at ticket #1048 and
determined the hanging issue; which Ingo quickly fixed... now I'm getting a
"Panic double lock of mutex" and would love pointers as to what to start
looking in to.
I've got a few things that need to be worked on, and need to know where/how to
start looking/working... I'm keen to see this working.
1) USB keyboard in the kernel debugger doesn't work. The USB 2.0 ports on the
eMac could be through a hub, of which I believe is not currently supported? As
this it the only method of input, I therefore cannot delve in to debugging. I
have been thinking of hard coding commands into the debugger as it first
launches... but I need to know what to look for and where.
2) After Ingo fixed the previous bug, I now get the following
exception:...Haiku revision: 36760heap_add_area: area -1 added to small heap
0x85000000 - usable range 0x8500d000 - 0x85800000heap_add_area: area -1 added
to medium heap 0x85800000 - usable range 0x85801000 - 0x85ccc000heap_add_area:
area -1 added to large heap 0x85cccccc - usable range 0x85cce000 -
0x86000000reserve_boot_loader_ranges(): Skipping range: 0x00000000,
12288reserve_boot_loader_ranges(): Skipping range: 0x00102000,
180224reserve_boot_loader_ranges(): Skipping range: 0x0012f000,
45056reserve_boot_loader_ranges(): Skipping range: 0x00200000, 2097152PANIC:
_mutex_lock(): double lock of 0x0000001c by thread 0Welcome to Kernel Debugging
Land...Running on CPU 0kdebug >
I've been digging through the code with dprintfs and have found that it
executes through:Line 310 of generic_vm_physical_page_mapper.cpp to
vm_create_null_area()Line 1623 of vm.cpp to VMCacheFactory::CreateNullCache().
Line 1482 of VMCache.cpp where it tries to instantiate the VMNullCache class.
This class inherits from VMCache which simply sets its internal fCacheRef to
null on construction. From this I can only assume that the malloc process
attempts to lock something and I haven't looked in to that yet...
Anyway, any help is greatly appreciated!
Thanks, Steven.
_________________________________________________________________
Need a new place to live? Find it on Domain.com.au
http://clk.atdmt.com/NMN/go/157631292/direct/01/
Other related posts:
