Test problems found by FTOREV23

Discussion about writing software for VS1005 and the VSOS Operating System. Also posts about VS1005-related hardware design and device drivers should be posted here.
Post Reply
winwan
Senior User
Posts: 86
Joined: Fri 2014-07-04 15:51

Test problems found by FTOREV23

Post by winwan »

I plugged 23S040 RAM into the hardware system,The system works in 32BIT mode

Code: Select all

Installed system devices:
S:  896K SPI Flash c213, handled by FAT.
Load drivers, config 0...
Driver: run... setclock 90
Driver: auxi2sm... 
Driver: run... auinput -b32
Driver: run... auoutput -b32
Driver: AUXPLAY... 
Driver: UARTIN... 
Driver: S:SHELL.AP3... 
VSOS SHELL

Code: Select all

S:>DRIVER +FTOREV23
VS23S0x0 size 512 KiB
S:>SETREV
ROOM:
  (-r) First reflection:     42 ms
  (-s) Room size       :   600 cm (1-32767)
  (-t) Reverb time     :   2600 ms (1-65535)
  (-f) Room softness   :  32767    (0=hard-65535=soft)
  (-d) Dry gain        :    768    (0-32767, 1024=1)
  (-w) Wet gain        :    256    (0-32767, 1024=1)
       Sample rate     :  48000 Hz
       Delay pairs     :      8
       Ext. mem. size  : 524288 bytes
       Ext. mem. read  : 0x4138 FTOREV23::ReadVS23S0x0
       Ext. mem. write : 0x4144 FTOREV23::WriteVS23S0x0

Code: Select all

S:>SETREV -s900 -v
ROOM:
  (-r) First reflection:     42 ms
  (-s) Room size       :    900 cm (1-32767)  /// :arrow:  :arrow: When the ROOM SIZE is less than 10000, the sound is normal
  (-t) Reverb time     :   2600 ms (1-65535)
  (-f) Room softness   :  32767    (0=hard-65535=soft)
  (-d) Dry gain        :    768    (0-32767, 1024=1)
  (-w) Wet gain        :    256    (0-32767, 1024=1)
       Sample rate     :  48000 Hz
       Delay pairs     :      8
       Ext. mem. size  : 524288 bytes
       Ext. mem. read  : 0x4138 FTOREV23::ReadVS23S0x0
       Ext. mem. write : 0x4144 FTOREV23::WriteVS23S0x0
When the ROOM SIZE is less than 10000, the sound is normal, but when the parameter is increased to more than 10000, there will be BA BA noise, and the larger the parameter, the more obvious

Code: Select all

S:>SETREV -s1000 -v
ROOM:
  (-r) First reflection:     42 ms
  (-s) Room size       :   2000 cm (1-32767) ///the larger the parameter, the more obvious
  (-t) Reverb time     :   2600 ms (1-65535)
  (-f) Room softness   :  32767    (0=hard-65535=soft)
  (-d) Dry gain        :    768    (0-32767, 1024=1)
  (-w) Wet gain        :    256    (0-32767, 1024=1)
       Sample rate     :  48000 Hz
       Delay pairs     :      8
       Ext. mem. size  : 524288 bytes
       Ext. mem. read  : 0x4138 FTOREV23::ReadVS23S0x0
       Ext. mem. write : 0x4144 FTOREV23::WriteVS23S0x0
Hannu
VLSI Staff
Posts: 546
Joined: Mon 2016-05-30 11:54
Location: Finland
Contact:

Re: Test problems found by FTOREV23

Post by Hannu »

Hi,

I can't tell straight away what would be the problem.

Please record a sample which has this noise and other without it and upload so we can investigate further. Preferably the source should be same.
winwan
Senior User
Posts: 86
Joined: Fri 2014-07-04 15:51

Re: Test problems found by FTOREV23

Post by winwan »

This is the recording file, there are three segments, respectively ROOM SIZE = 0; ROOM SIZE = 900; ROOM_SIZE = 6000;
org_file.wav
(18.01 MiB) Downloaded 273 times
ftorev23_room_size900.wav
(17.79 MiB) Downloaded 275 times
ftorev23_room_size6000.wav
(17.58 MiB) Downloaded 262 times
The running configuration is as follows:

Code: Select all

Installed system devices:
S:  896K SPI Flash c213, handled by FAT.
Load drivers, config 0...
Driver: run... setclock 90
Driver: auxi2sm... 
Driver: run... auinput -b32
Driver: run... auoutput -b32
Driver: AUXPLAY... 
Driver: UARTIN... 
Driver: S:SHELL.AP3... 
VSOS SHELL
S:>driver +ftorev23
VS23S0x0 size 512 KiB
S:>setrev -s900
S:>setrev -s6000

Code: Select all

S:>setrev
ROOM:
  (-r) First reflection:     42 ms
  (-s) Room size       :   6000 cm (1-32767)
  (-t) Reverb time     :   2600 ms (1-65535)
  (-f) Room softness   :  32767    (0=hard-65535=soft)
  (-d) Dry gain        :    768    (0-32767, 1024=1)
  (-w) Wet gain        :    256    (0-32767, 1024=1)
       Sample rate     :  48000 Hz
       Delay pairs     :      8
       Ext. mem. size  : 524288 bytes
       Ext. mem. read  : 0x4138 FTOREV23::ReadVS23S0x0
       Ext. mem. write : 0x4144 FTOREV23::WriteVS23S0x0
User avatar
Henrik
VLSI Staff
Posts: 1295
Joined: Tue 2010-06-22 14:10

Re: Test problems found by FTOREV23

Post by Henrik »

Hello!

I have downloaded your files and looking at the issue.

I will write here when I know what the issue is.

One question:
If you run SetClock without any parameters, what is the output?

Code: Select all

S:>setclock
Kind regards,
- Henrik
Good signatures never die. They just fade away.
winwan
Senior User
Posts: 86
Joined: Fri 2014-07-04 15:51

Re: Test problems found by FTOREV23

Post by winwan »

I uploaded AUXPLAY 32BIT
auxplay.dl3
(1.74 KiB) Downloaded 248 times

Code: Select all

 
  fwrite(buf, sizeof(s_int32)*2 /*stereo */, SAMPLES, stdaudioout);
This FTREV23.C file lines 82-83

Code: Select all

u_int32 sampleRate = 48000UL;
s_int16 bits = 32;

Code: Select all

S:>setclock
SetClock running on VS1005g, clocks:
  XTALI 12.288MHz, CLKI 92.160MHz, limit 93.000MHz, src PLL
  RAM Delay 1, POR on
  CVDD 1.950V(25), IOVDD 3.30V(25), AVDD 3.60V(28)
  UART nominal 115200 bps, real 114841 bps (0.3% error), reg 0x006b
  SPI0(I) 23.0 Mbit/s, SPI1 46.1 Mbit/s, NAND 11.5 MByte/s, SD 5.8 MByte/s
Uptime 5:11.7, time counter interrupt 1000.0 times/s
Laser fuse(63:0) = 0x6b11:4222:7156:0418, crc ok
Serial Flash:
  RDID: manufacturer c2 (C2), type 20 (20), density 14 (14)
  SCUR: 0x00. Factory lock off, user lock off
  Field 0, 0x2022 0x0715 0x7156 0x2000 opCode 1 = TrimData
    CRC29:         0x00220715, ok
    USB Trim:      0x7153
    High speed:    no
    Mem Delay:     1
    Int Flash:     2.8 V
    VCORE Trim:    +0
    Customer code: 0
    Unused: 0
  Field 1, 0x5cfc 0x5dbb 0x001f 0xdaf1 opCode 2 = SerialNumber
    CRC29:         0x1cfc5dbb, ok
    Serial number: 0x001fdaf1
  Field 2, 0xffff 0xffff 0xffff 0xffff opCode 7 = Unused
  Field 3, 0xffff 0xffff 0xffff 0xffff opCode 7 = Unused
  Field 4, 0xffff 0xffff 0xffff 0xffff opCode 7 = Unused
  Field 5, 0xffff 0xffff 0xffff 0xffff opCode 7 = Unused
  Field 6, 0xffff 0xffff 0xffff 0xffff opCode 7 = Unused
winwan
Senior User
Posts: 86
Joined: Fri 2014-07-04 15:51

Re: Test problems found by FTOREV23

Post by winwan »

I retained the AUXPLAY monitoring function, running FTOREV and FTOREV23 is compared as follows, both modules are running at 32BIT, FTOREV23 runs with overflow data, FTOREV can uninstall normally, while FTOREV23 uninstalls error system crash

Code: Select all

S:>driver +ftorev
S:>
S:>driver -ftorev
S:>
S:>driver +ftorev23
VS23S0x0 size 512 KiB
S:>
S:>AUXSPLAY: In overflow +29
AUXSPLAY: Out underflow +28
AUXSPLAY: In overflow +3
AUXSPLAY: Out underflow +4
driver -ftorev23
E'Call to dropped lib! encnt=0, last vector 0x10'
ZeroPtrCall from 7889(0x1ed1) KERNELNoSym
i0=64519(0xfc07) UNASSIGNED=0x0000(0)

Task 0x20a5, priority 2, in RUNNING, name "AUXPLAY"
  State: 3 (TS_READY)
  Stack: Start 0x20b4, size 0x100, in use 0x57, max used 0xff (0x1 free)
  Stack Trace: current PC 0x4b66, Tasks:ZeroPtrCall from 1858(0x0742) KERNELE'Call to dropped lib! encnt=3, last vector 0x10'
ZeroPtrCall from 7889(0x1ed1) KERNEL::�����������������������������[0x24]
i0=64519(0xfc07) Tasks:0:568=0x0000(0)

Task 0x0000, priority 0, in RUNNING, name "3"
  State: 0 (TS_INVALID)
  Stack: Start 0x0000, size 0x0, in use 0x0, max used 0x0 (0x0 free)
  Stack Trace: current PC 0x4b66, Tasks:ZeroPtrCall from 15051(0x3acb) SHELL:
I think the noise is also to blame,

The FTIREV23 was also unable to function properly and died as soon as it started up

Code: Select all

S:>driver +ftirev23
VS23S0x0 size 512 KiB
Hannu
VLSI Staff
Posts: 546
Joined: Mon 2016-05-30 11:54
Location: Finland
Contact:

Re: Test problems found by FTOREV23

Post by Hannu »

The crash is simply because you removed ftorev before auxplay and while doing that the stdaudioout which auxplay uses points to somewhere where it shouldn't because there isn't code anymore.

The overflows you are probably coming from too small buffers in input or output side. Another reason can be that you don't high enough clock frequency.
Post Reply