I am thinking of having the programming speed for 150-250KB/s so program the flash will take about half a minute and in case of the FTA it will be much less if we skip the blank sectors.slugworth wrote:The default settings for the st20 burner is a 4k buffer,which was ideal for the sst39vf3201 since it has 4k blocks.It programs at about 1meg per minute,so the 4meg flash takes around 4 minutes.It erases the entire chip at once,so that is fast.
Dcu3 jtag
-
- Junior Member
- Posts: 8979
- Joined: Mon Jul 18, 2005 9:33 pm
-
- Junior Member
- Posts: 267
- Joined: Sun Feb 17, 2008 1:13 pm
eeprom
Sti5105 pins to eeprom from a vestel schematic showing the sti5105 pinouts.
On a magnum3300 and skyview1000 the eeprom is a 24lc64
The chip on the right is the sti5105
On a magnum3300 and skyview1000 the eeprom is a 24lc64
The chip on the right is the sti5105
You do not have the required permissions to view the files attached to this post.
-
- Junior Member
- Posts: 8979
- Joined: Mon Jul 18, 2005 9:33 pm
Update for DCU3 development.
I can now set my own trap. There is a little bit inconsistence. Once that is fixed then I can have USB JTAG NT support DCU3 (Sti5105).
-
- Junior Member
- Posts: 1
- Joined: Wed Mar 25, 2009 11:33 pm
Will Sti5107 also be supported soon? I have an box with this cpu and cannot dump flash using Jkeys. Understanding that it is already DCU3 I've changed the start address to 40000000 but still would not dump. I get DCU peek error upon connection.
BTW the flash is a 32Mbit m28w320FS Top boot. I've read the datasheet and it says that is it has "Krypto" technology. I'm not sure if its "read protection" was activated, that is why I am having DCU peek errors.
BTW the flash is a 32Mbit m28w320FS Top boot. I've read the datasheet and it says that is it has "Krypto" technology. I'm not sure if its "read protection" was activated, that is why I am having DCU peek errors.
-
- Junior Member
- Posts: 8979
- Joined: Mon Jul 18, 2005 9:33 pm
The DCU3 has completely different instruction set.
Like We used to use "diff" and this command is not there for ST20 C1 (5105).
Due the the sudden move to N3 I might stop this development for DCU3 for a while.
What I have archieved is I can stop the CPU inject the trap code and detect the flash. The problem is that the trap code is not stable, some flags in the core crashed (or interrupted the trap). If that has passed then it will be quite simple to program the flash (a lot faster than the old USB JTAG that uses sprogram). But for now I might switch the task.
Like We used to use "diff" and this command is not there for ST20 C1 (5105).
Due the the sudden move to N3 I might stop this development for DCU3 for a while.
What I have archieved is I can stop the CPU inject the trap code and detect the flash. The problem is that the trap code is not stable, some flags in the core crashed (or interrupted the trap). If that has passed then it will be quite simple to program the flash (a lot faster than the old USB JTAG that uses sprogram). But for now I might switch the task.
-
- Junior Member
- Posts: 267
- Joined: Sun Feb 17, 2008 1:13 pm
response
I believe a "brain dead" receiver reacts differently.
I assume the receiver you play with is functioning normally.
I was after a dcu3 jtag to repair a dead receiver,which I finally did via the st20 toolset route.Dcu3 jtag once N3 takes over will just appeal to cloners or possible playing with receiver ram "on the fly" like I did with the 301.013 receiver with bev in the N1 days with jtango.I found the channel maps in ram and changed the service byte in ram so I could use dishnet firmware to watch bev which was unheard of at the time.I changed the service byte from 96 to 01 to watch bev channels but you had to refresh ram every 45 minutes to keep the channels from dropping off.Bev caught on and went to multiple channel maps in ram and moved them around,so a simple jtango script wouldn't work anymore to refresh ram.N2 put an end to that fun.
I assume the receiver you play with is functioning normally.
I was after a dcu3 jtag to repair a dead receiver,which I finally did via the st20 toolset route.Dcu3 jtag once N3 takes over will just appeal to cloners or possible playing with receiver ram "on the fly" like I did with the 301.013 receiver with bev in the N1 days with jtango.I found the channel maps in ram and changed the service byte in ram so I could use dishnet firmware to watch bev which was unheard of at the time.I changed the service byte from 96 to 01 to watch bev channels but you had to refresh ram every 45 minutes to keep the channels from dropping off.Bev caught on and went to multiple channel maps in ram and moved them around,so a simple jtango script wouldn't work anymore to refresh ram.N2 put an end to that fun.
-
- Junior Member
- Posts: 267
- Joined: Sun Feb 17, 2008 1:13 pm
uctap
For jkeys the uctap wiring seems better than the standard 9-11-13-15-19Hanukkah wrote:Will Sti5107 also be supported soon? I have an box with this cpu and cannot dump flash using Jkeys. Understanding that it is already DCU3 I've changed the start address to 40000000 but still would not dump. I get DCU peek error upon connection.
You use 9-11-13-15-17 and tie pin19 high.
Who is online
Users browsing this forum: No registered users and 7 guests