Page 1 of 1
(solved) Hieronymus not working on Sam440 ?
Posted: Sun Aug 16, 2015 11:32 am
by ChrisH
I tried
Hieronymus on my Sam440, but all it outputs is this obscure error message:
Failed to acquire samples, the interrupt server does not work or interrupt context is null
Does it work on anyone else's Sam440? Any idea what the problem could be?
EDIT: It works when I use a different boot partition, so looks like something's borked on my first boot partition.
Re: (solved) Hieronymus not working on Sam440 ?
Posted: Sun Aug 16, 2015 12:59 pm
by corto
ChrisH wrote:I tried
Hieronymus on my Sam440, but all it outputs is this obscure error message:
Failed to acquire samples, the interrupt server does not work or interrupt context is null
Does it work on anyone else's Sam440? Any idea what the problem could be?
EDIT: It works when I use a different boot partition, so looks like something's borked on my first boot partition.
Great to see you try Hieronymus!
Can you confirm you use the last version (0.22) from os4depot?
What is the format of your first boot partition?
Don't you boot 2 versions of the system?
I would like to understand the problem you faced. And if you have some questions or comments, you're welcome!
Re: (solved) Hieronymus not working on Sam440 ?
Posted: Sun Aug 16, 2015 6:15 pm
by ChrisH
corto wrote:Can you confirm you use the last version (0.22) from os4depot?
Yup.
What is the format of your first boot partition?
Don't you boot 2 versions of the system?
It's failing on AmigaOS4.1 Update 4 (!), but works on AmigaOS4.1 Final Edition. However, both boot partitions are *extremely* customised (so that they can share exactly the same settings, etc), and it is quite possible I broke something on the older (Update 4) partition. I should probably do a clean install on that partition...
I would like to understand the problem you faced. And if you have some questions or comments, you're welcome!
I guess there's no chance of you getting Hieronymus working on the X1000 ? That's my main machine.
P.S. Hieronymus already confirmed my suspicion about obscenely slow speed being due to a particular algorithm (procedure), something I could NOT do using GCC's profiling mode (-pg), so Hieronymus has done it's job
Re: (solved) Hieronymus not working on Sam440 ?
Posted: Sun Aug 16, 2015 10:13 pm
by corto
Don't you boot 2 versions of the system?
It's failing on AmigaOS4.1 Update 4 (!), but works on AmigaOS4.1 Final Edition. However, both boot partitions are *extremely* customised (so that they can share exactly the same settings, etc), and it is quite possible I broke something on the older (Update 4) partition. I should probably do a clean install on that partition...
No, that's not necessary. Hieronymus started to work on Sam440 with a given version of the rtg.library (I will be able to provide the exact version if necessary).
I modified the sources and with the version 0.22, Hieronymus should have printed the required version of rtg.library (that certainly came out around the update 5).
I guess there's no chance of you getting Hieronymus working on the X1000 ? That's my main machine.
P.S. Hieronymus already confirmed my suspicion about obscenely slow speed being due to a particular algorithm (procedure), something I could NOT do using GCC's profiling mode (-pg), so Hieronymus has done it's job
About the port to the X1000, the hope came when I acquired a machine and then I faced a problem. But recently, I thought about an idea to circumvent it. I will have to test that.
I'm very pleased to see Hieronymus helped you!
Re: (solved) Hieronymus not working on Sam440 ?
Posted: Tue Sep 01, 2015 10:13 pm
by ChrisH
corto wrote:I'm very pleased to see Hieronymus helped you!
Something else which would really help (and I'm surprised it isn't already implemented) is for Functions to be sorted by their Count, so that the most-frequent functions are at the top (as typically these are what you are interested in).
Re: (solved) Hieronymus not working on Sam440 ?
Posted: Thu Sep 03, 2015 12:38 pm
by corto
ChrisH wrote:corto wrote:I'm very pleased to see Hieronymus helped you!
Something else which would really help (and I'm surprised it isn't already implemented) is for Functions to be sorted by their Count, so that the most-frequent functions are at the top (as typically these are what you are interested in).
Sure, it would be useful ... I think it is not implemented yet because not so fun and creative to do (but I created a ticket about that one year ago in my bugtracker).
But recently, I worked on several points:
- I've added stats provided by the performance monitor for a given delay (if you know the Linux tool called "perf"). That's done on the MicroAOne and when it's done with a good software design and working, I will add the X1000 support.
- About the chance to have the profiling working on the X1000, as you asked: After new investigations, I found that the performance monitor was unfortunately not usable for sampling. So the other chance will be to have a low-level and periodic interrupt source.
Re: Hieronymus not working on Sam440
Posted: Mon Jul 10, 2017 11:10 am
by capehill
On my Sam440 Hieronymus 0.22 crashes. Stacktrace points towards elf.library. Anybody else with the problem?
Code: Select all
Crash log for task "hieronymus"
Generated by GrimReaper 53.19
Crash occured in module elf.library.kmod at address 0x01576614
Type of crash: DSI (Data Storage Interrupt) exception
Alert number: 0x80000003
Register dump:
GPR (General Purpose Registers):
0: 0156E504 5C148EE0 00000000 5C148E50 5B6C4E2C 5C149028 00000000 06180842
8: 01DCD9A2 00000000 5C148E70 00001000 4A444242 58F4FB10 80000004 80000001
16: 80000002 58F40000 58F40000 58F50000 58F4759C 58F483BC 00000000 80000004
24: 00000001 5B6C4E2C 5C148E50 5B6C4DB0 5C148FB0 00000000 00000000 00000000
FPR (Floating Point Registers, NaN = Not a Number):
0: nan 791 804 208
4: 2 271.8 78.8 208
8: 2 128 271.8 255
12: nan 65536 0 0
16: 0 0 0 0
20: 0 1.17664e-38 0 0
24: 0 0 0 0
28: 0 0 0 0
FPSCR (Floating Point Status and Control Register): 0x82000000
SPRs (Special Purpose Registers):
Machine State (msr) : 0x0002F030
Condition (cr) : 0x90380000
Instruction Pointer (ip) : 0x01576614
Xtended Exception (xer) : 0x00000000
Count (ctr) : 0x00000000
Link (lr) : 0x00000000
DSI Status (dsisr) : 0x0000FF01
Data Address (dar) : 0x00040100
680x0 emulated registers:
DATA: 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
ADDR: 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
FPU0: 0 0 0 0
FPU4: 0 0 0 0
Symbol info:
Instruction pointer 0x01576614 belongs to module "elf.library.kmod" (HUNK/Kickstart)
Stack trace:
native kernel module elf.library.kmod+0x00008254
native kernel module elf.library.kmod+0x00008248
native kernel module elf.library.kmod+0x000004d0
native kernel module elf.library.kmod+0x000005e0
hieronymus:Bosch_StopAcquisition()+0x3fc (section 1 @ 0xBA8)
hieronymus:main()+0x490 (section 1 @ 0x13E4)
native kernel module newlib.library.kmod+0x00001ea0
native kernel module newlib.library.kmod+0x00002d84
native kernel module newlib.library.kmod+0x00003074
hieronymus:_start()+0x170 (section 1 @ 0x16C)
native kernel module dos.library.kmod+0x00026698
native kernel module kernel+0x0003e9a8
native kernel module kernel+0x0003e9f0
PPC disassembly:
0157660c: 7c7a1b78 mr r26,r3
01576610: 813c0048 lwz r9,72(r28)
*01576614: a1490030 lhz r10,48(r9)
01576618: 2f8a0000 cmpwi cr7,r10,0
0157661c: 419e0054 beq- cr7,0x1576670
System information:
CPU
Model: AMCC PPC440EP V1.3