OS/2, eCS & ArcaOS - Technical > Storage
DISKIO - updated version in need of TEST
Dariusz Piatkowski:
Hello Roberto!
Thank you for testing and the feedback.
--- Quote from: roberto on December 16, 2024, 06:58:56 pm ---My test:
--- Code: ---[D:\diskio125]diskio
DISKIO - Fixed Disk Benchmark, Version 1.25
Dhrystone 2.1 C benchmark routines (C) 1988 Reinhold P. Weicker
Dhrystone benchmark for this CPU: 3307018 runs/sec
--- End code ---
--- End quote ---
Hmm, seems like low CPU score, what are you running? If that's a fast CPU I think we may be getting an overflow in the benchmark code, on my TO-DO list.
--- Quote from: roberto on December 16, 2024, 06:58:56 pm ---
--- Code: ---Hard disk 1: 255 sides, 14593 cylinders, 63 sectors per track = 114470 MB
Drive cache/bus transfer rate: 101023 k/sec
Data transfer rate on cylinder 0 : 116343 k/sec
Data transfer rate on cylinder 14591 : 241957 k/sec
CPU usage by full speed disk transfers: 0%
Average latency time: 0.1 ms
Average data access time: 0.1 ms
Multithreaded disk I/O
1 thread: 116510 k/sec, 0% CPU usage
2 threads: 109296 k/sec, 0% CPU usage
3 threads: 108393 k/sec, 0% CPU usage
4 threads: 108952 k/sec, 0% CPU usage
5 threads: 109710 k/sec, 0% CPU usage
6 threads: 110312 k/sec, 0% CPU usage
7 threads: 110657 k/sec, 0% CPU usage
8 threads: 110428 k/sec, 0% CPU usage
--- End code ---
--- End quote ---
So CPU is an 8-core chip, yes?
Interesting that its showing '0% CPU usage', I mean great, but that just seem like maybe we have a side-effect of showing whole numbers only and no decimal places for the 'CPU Usage'.
Can you get me a debug-run on just that drive please?
--- Quote from: roberto on December 16, 2024, 06:58:56 pm ---
--- Code: ---Hard disk 3: 12 sides, 512 cylinders, 9 sectors per track = 27 MB
Drive cache/bus transfer rate: Disk read error.
Data transfer rate on cylinder 0 : Disk read error.
Data transfer rate on cylinder 510 : Disk read error.
CPU usage by full speed disk transfers: Disk read error.
Average latency time: Disk read error.
Average data access time: Disk read error.
Multithreaded disk I/O
1 thread: Disk read error.
SYS1808:
The process has stopped. The software diagnostic
code (exception code) is 009B.
--- End code ---
--- End quote ---
Yup, USB devices...I am looking at ways of checking whether the storage devices that are identified by DISKIO at the start are actually usable, before the attempt to benchmark them actually starts.
Dave Yeo:
Hi Dariusz, I get results more like Roberto, though it usually doesn't get as far here before crashing,
--- Code: ---DISKIO - Fixed Disk Benchmark, Version 1.25
(C) 1994-1998 Kai Uwe Rommel
(C) 2004 madded2
2020-2024 Dariusz Piatkowski
Number of fixed disks: 16
Number of CD-ROM drives: 2
Dhrystone 2.1 C benchmark routines (C) 1988 Reinhold P. Weicker
Dhrystone benchmark for this CPU: 148719 runs/sec
Hard disk 1: 255 sides, 60801 cylinders, 63 sectors per track = 476937 MB
Drive cache/bus transfer rate: 166465 k/sec
Data transfer rate on cylinder 0 : 332980 k/sec
Data transfer rate on cylinder 60799 : 331016 k/sec
CPU usage by full speed disk transfers: 8%
Average latency time: 0.0 ms
Average data access time: Disk read error.
Multithreaded disk I/O
1 thread:
--- End code ---
is logged and then
--- Code: ---SYS1808:
The process has stopped. The software diagnostic
code (exception code) is 009B.
--- End code ---
The debug build exits here after the CPU benchmark, which varies widely on a quiet system and once locked the computer hard.
This is a I5-7500@3.5Ghz with Nvme and SSD drives.
Perhaps a SMP issue?
Running the original also crashes,
--- Code: ---DISKIO - Fixed Disk Benchmark, Version 1.18z
(C) 1994-1998 Kai Uwe Rommel
(C) 2004 madded2
Number of fixed disks: 16
Number of CD-ROM drives: 2
Dhrystone 2.1 C benchmark routines (C) 1988 Reinhold P. Weicker
Dhrystone benchmark for this CPU: 7461002 runs/sec
Hard disk 1: 255 sides, 60801 cylinders, 63 sectors per track = 476937 MB
Drive cache/bus transfer rate: 150517 k/sec
Data transfer rate on cylinder 0 : -155262 k/sec
Data transfer rate on cylinder 60799: -144451 k/sec
CPU usage by full speed disk transfers: 44%
Average data access time: Disk read error.
Multithreaded disk I/O (4 threads): SYS1808:
The process has stopped. The software diagnostic
code (exception code) is 009B.
--- End code ---
An earlier build from you got further,
--- Code: ---DISKIO - Fixed Disk Benchmark, Version 1.20
(C) 1994-1998 Kai Uwe Rommel
(C) 2004 madded2
2020 Dariusz Piatkowski
Number of fixed disks: 16
Number of CD-ROM drives: 2
Dhrystone 2.1 C benchmark routines (C) 1988 Reinhold P. Weicker
Dhrystone benchmark for this CPU: 8660078 runs/sec
Hard disk 1: 255 sides, 60801 cylinders, 63 sectors per track = 476937 MB
Drive cache/bus transfer rate: 146900 k/sec
Data transfer rate on cylinder 0 : 285680 k/sec
Data transfer rate on cylinder 60799: 287686 k/sec
CPU usage by full speed disk transfers: 50%
Average latency time: 0.0 ms
Average data access time: Disk read error.
Multithreaded disk I/O (4 threads):
221010 k/sec, 46% CPU usage
Hard disk 2: 255 sides, 583 cylinders, 63 sectors per track = 4573 MB
Drive cache/bus transfer rate: 24411 k/sec
Data transfer rate on cylinder 0 : 22839 k/sec
Data transfer rate on cylinder 581 : 19798 k/sec
CPU usage by full speed disk transfers: 48%
Average latency time: 0.1 ms
Average data access time: Disk read error.
Multithreaded disk I/O (4 threads):
24381 k/sec, 50% CPU usage
Hard disk 3: 255 sides, 60322 cylinders, 127 sectors per track = 953871 MB
Drive cache/bus transfer rate: 39907 k/sec
Data transfer rate on cylinder 0 : Disk read error.
Data transfer rate on cylinder 60320: 433087 k/sec
CPU usage by full speed disk transfers: Disk read error.
Average latency time: 0.0 ms
Average data access time: Disk read error.
Multithreaded disk I/O (4 threads): Disk read error.
40497 k/sec, 62% CPU usage
Hard disk 4: 255 sides, 60322 cylinders, 127 sectors per track = 953871 MB
Drive cache/bus transfer rate: 48704 k/sec
Data transfer rate on cylinder 0 : Disk read error.
Data transfer rate on cylinder 60320: 610275 k/sec
CPU usage by full speed disk transfers: Disk read error.
Average latency time:
SYS1808:
The process has stopped. The software diagnostic
code (exception code) is 009B.
--- End code ---
There's entries in popuplog.os2 for each crash.
Dave Yeo:
I remembered that you don't run AcpiDaemon, so I killed it and retested, got further before crashing. Then tried without redirecting stdout, stderr (diskio 2>&1 | tee diskio.log) and it finished, of course the results scrolled off the screen
Even doing "diskio > diskio.log" crashes as above.
Running without AcpiDaemon here also sees the CPU temp at about 65C compared to 38C with it running and testing diskio.
Rich Walsh:
--- Code: ---Number of fixed disks: 17
Number of CD-ROM drives: 1
Dhrystone 2.1 C benchmark routines (C) 1988 Reinhold P. Weicker
Dhrystone benchmark for this CPU: -567273196 runs/sec
Hard disk 1: 255 sides, 62261 cylinders, 63 sectors per track = 488390 MB
Drive cache/bus transfer rate:
SYS1808:
The process has stopped. The software diagnostic
code (exception code) is 009B.
--- End code ---
Even if it didn't crash with a "divide by zero" error (009B), 'Diskio' would be unlikely to provide useful info about my disks. Both of them are GPT and are normally *hidden* from the system by 'gpt.flt'.
What OS/2 sees are a bunch of _emulated_ MBR disks, one for each mounted partition plus a few spares that appear to be unpartitioned. Doing a speed test on a phony disk that refers to some partition somewhere on one of the physical disks wouldn't be overly informative. In any case, I don't think you can access them using the Physical Disk IOCTLs that 'Diskio' depends on, so we'll probably never know.
FWIW... GPT disks can be made visible, though I'm not sure how accessible they are. I look forward to trying 'Diskio' again with the disks visible once the current crop of issues is resolved.
Dave Yeo:
--- Quote from: Rich Walsh on December 17, 2024, 06:43:07 am ---
Even if it didn't crash with a "divide by zero" error (009B), 'Diskio' would be unlikely to provide useful info about my disks. Both of them are GPT and are normally *hidden* from the system by 'gpt.flt'.
What OS/2 sees are a bunch of _emulated_ MBR disks, one for each mounted partition plus a few spares that appear to be unpartitioned. Doing a speed test on a phony disk that refers to some partition somewhere on one of the physical disks wouldn't be overly informative. In any case, I don't think you can access them using the Physical Disk IOCTLs that 'Diskio' depends on, so we'll probably never know.
FWIW... GPT disks can be made visible, though I'm not sure how accessible they are. I look forward to trying 'Diskio' again with the disks visible once the current crop of issues is resolved.
--- End quote ---
Seems to kind of work on GPT disk here, each partition presents as a device the full size of the device, eg for my 1TB NVMe
--- Code: ---Hard disk 3: 255 sides, 60322 cylinders, 127 sectors per track = 953871 MB
Drive cache/bus transfer rate: 137487 k/sec
Data transfer rate on cylinder 0 : Disk read error.
Data transfer rate on cylinder 60320 : 71482 k/sec
CPU usage by full speed disk transfers: Disk read error.
Average latency time: 0.0 ms
Average data access time: Disk read error.
Multithreaded disk I/O
1 thread: 6654 k/sec, 28% CPU usage
2 threads: 762685 k/sec, 28% CPU usage
3 threads: Disk read error.
896092 k/sec, 20% CPU usage
4 threads: Disk read error.
1889132 k/sec, 26% CPU usage
Hard disk 4: 255 sides, 60322 cylinders, 127 sectors per track = 953871 MB
Drive cache/bus transfer rate: 148253 k/sec
Data transfer rate on cylinder 0 : Disk read error.
Data transfer rate on cylinder 60320 : 32923 k/sec
CPU usage by full speed disk transfers: Disk read error.
Average latency time: 0.0 ms
Average data access time: Disk read error.
Multithreaded disk I/O
1 thread: Disk read error.
158947 k/sec, 35% CPU usage
2 threads: Disk read error.
166172 k/sec, 35% CPU usage
3 threads: Disk read error.
156057 k/sec, 35% CPU usage
4 threads: Disk read error.
162560 k/sec, 34% CPU usage
Hard disk 5: 255 sides, 60322 cylinders, 127 sectors per track = 953871 MB
Drive cache/bus transfer rate: 149314 k/sec
Data transfer rate on cylinder 0 : Disk read error.
Data transfer rate on cylinder 60320 :
The external process was cancelled by a Ctrl+Break or another process.
--- End code ---
First try locked up the machine so I did CTRL-C when I had enough info. The numbers are quite a bit worse then the MBR SSD along with the Disk read errors which seem to vary.
Navigation
[0] Message Index
[#] Next page
[*] Previous page
Go to full version