Hi,
I am using CCS 3.3, XDS510usb emulator and the target board has C5510 DSPs.
My setup has been working fine for the past few months. Recently I have been
having issues loading my code on to the DSPs. I get the following error:
Trouble Writing Register: Error 0x00000004/-1024 Error during: Register,
PTI_ERR_PREV_IROP_CMD Error Occured at 0x00000000
I thought it was an emulator/hardware error, so I used another emulator and
changed the target board. But the error still persists. Have anyone faced
similar issues?
Thanks,
VJ
Issue loading code on to the DSP.
Started by ●June 23, 2009
Reply by ●June 24, 20092009-06-24
Vivekjayadev-
> I am using CCS 3.3, XDS510usb emulator and the target board has C5510 DSPs.
>
> My setup has been working fine for the past few months. Recently I have
> been having issues loading my code on to the DSPs. I get the following error:
>
> Trouble Writing Register: Error 0x00000004/-1024 Error during: Register,
> PTI_ERR_PREV_IROP_CMD Error Occured at 0x00000000
>
> I thought it was an emulator/hardware error, so I used another emulator and
> changed the target board. But the error still persists. Have anyone faced
> similar issues?
How many 5510 DSPs on your board? How many in the JTAG chain?
-Jeff
> I am using CCS 3.3, XDS510usb emulator and the target board has C5510 DSPs.
>
> My setup has been working fine for the past few months. Recently I have
> been having issues loading my code on to the DSPs. I get the following error:
>
> Trouble Writing Register: Error 0x00000004/-1024 Error during: Register,
> PTI_ERR_PREV_IROP_CMD Error Occured at 0x00000000
>
> I thought it was an emulator/hardware error, so I used another emulator and
> changed the target board. But the error still persists. Have anyone faced
> similar issues?
How many 5510 DSPs on your board? How many in the JTAG chain?
-Jeff
Reply by ●June 24, 20092009-06-24
Vivekjayadev-
> I have 4 DSPs on my target board and 8 JTAG devices in the scan chain.
Ok. What are the non-DSP JTAG devices? We've had trouble before getting non-TI
devices to "peacefully co-exist" in a CCS + emulator scan chain. Not that it's
causing a problem in your application, I'm just curious...
> After poking at all the initializations, I realised that the gel file is not
> getting loaded properly. It picks the right gel file (c5510.gel) but it does not
> load that correctly. So, I tried a CPU reset in the GEL configuration and then it
> allows me to load the code. I have to do this every time I open up CCS. For now I
> have a work around, but I am looking for a permanent fix.
When you say "does not load correctly", can you be more specific? Is there an error
message? If you look in the .gel file after CCS runs, is part of the GEL code
missing? One thing you might try is:
-make your own version, say vj_c5510.gel
-add some status print messages
-modify CCS Setup to point to the new
.gel file
Then, when you first run CCS, you should be able see your print messages in the Gel
status window. If the code is only making it part way or something else is wrong,
this might be a way to debug.
-Jeff
> On Wed, Jun 24, 2009 at 10:53 AM, Jeff Brower wrote:
>
> Vivekjayadev-
>
> > I am using CCS 3.3, XDS510usb emulator and the target board has C5510
> DSPs.
> >
> > My setup has been working fine for the past few months. Recently I have
>
> > been having issues loading my code on to the DSPs. I get the following
> error:
> >
> > Trouble Writing Register: Error 0x00000004/-1024 Error during:
> Register,
> > PTI_ERR_PREV_IROP_CMD Error Occured at 0x00000000
> >
> > I thought it was an emulator/hardware error, so I used another emulator
> and
> > changed the target board. But the error still persists. Have anyone
> faced
> > similar issues?
>
> How many 5510 DSPs on your board? How many in the JTAG chain?
>
> -Jeff
>
> I have 4 DSPs on my target board and 8 JTAG devices in the scan chain.
Ok. What are the non-DSP JTAG devices? We've had trouble before getting non-TI
devices to "peacefully co-exist" in a CCS + emulator scan chain. Not that it's
causing a problem in your application, I'm just curious...
> After poking at all the initializations, I realised that the gel file is not
> getting loaded properly. It picks the right gel file (c5510.gel) but it does not
> load that correctly. So, I tried a CPU reset in the GEL configuration and then it
> allows me to load the code. I have to do this every time I open up CCS. For now I
> have a work around, but I am looking for a permanent fix.
When you say "does not load correctly", can you be more specific? Is there an error
message? If you look in the .gel file after CCS runs, is part of the GEL code
missing? One thing you might try is:
-make your own version, say vj_c5510.gel
-add some status print messages
-modify CCS Setup to point to the new
.gel file
Then, when you first run CCS, you should be able see your print messages in the Gel
status window. If the code is only making it part way or something else is wrong,
this might be a way to debug.
-Jeff
> On Wed, Jun 24, 2009 at 10:53 AM, Jeff Brower wrote:
>
> Vivekjayadev-
>
> > I am using CCS 3.3, XDS510usb emulator and the target board has C5510
> DSPs.
> >
> > My setup has been working fine for the past few months. Recently I have
>
> > been having issues loading my code on to the DSPs. I get the following
> error:
> >
> > Trouble Writing Register: Error 0x00000004/-1024 Error during:
> Register,
> > PTI_ERR_PREV_IROP_CMD Error Occured at 0x00000000
> >
> > I thought it was an emulator/hardware error, so I used another emulator
> and
> > changed the target board. But the error still persists. Have anyone
> faced
> > similar issues?
>
> How many 5510 DSPs on your board? How many in the JTAG chain?
>
> -Jeff
>
Reply by ●June 28, 20092009-06-28
I have 4 DSPs on my target board and 8 JTAG devices in the scan chain.
After poking at all the initializations, I realised that the gel file is not
getting loaded properly. It picks the right gel file (c5510.gel) but it does
not load that correctly. So, I tried a CPU reset in the GEL configuration
and then it allows me to load the code. I have to do this every time I open
up CCS.
For now I have a work around, but I am looking for a permanent fix.
- Thanks,
VJ
On Wed, Jun 24, 2009 at 10:53 AM, Jeff Brower wrote:
> Vivekjayadev-
>
> > I am using CCS 3.3, XDS510usb emulator and the target board has C5510
> DSPs.
> >
> > My setup has been working fine for the past few months. Recently I have
> > been having issues loading my code on to the DSPs. I get the following
> error:
> >
> > Trouble Writing Register: Error 0x00000004/-1024 Error during: Register,
> > PTI_ERR_PREV_IROP_CMD Error Occured at 0x00000000
> >
> > I thought it was an emulator/hardware error, so I used another emulator
> and
> > changed the target board. But the error still persists. Have anyone faced
> > similar issues?
>
> How many 5510 DSPs on your board? How many in the JTAG chain?
>
> -Jeff
>
After poking at all the initializations, I realised that the gel file is not
getting loaded properly. It picks the right gel file (c5510.gel) but it does
not load that correctly. So, I tried a CPU reset in the GEL configuration
and then it allows me to load the code. I have to do this every time I open
up CCS.
For now I have a work around, but I am looking for a permanent fix.
- Thanks,
VJ
On Wed, Jun 24, 2009 at 10:53 AM, Jeff Brower wrote:
> Vivekjayadev-
>
> > I am using CCS 3.3, XDS510usb emulator and the target board has C5510
> DSPs.
> >
> > My setup has been working fine for the past few months. Recently I have
> > been having issues loading my code on to the DSPs. I get the following
> error:
> >
> > Trouble Writing Register: Error 0x00000004/-1024 Error during: Register,
> > PTI_ERR_PREV_IROP_CMD Error Occured at 0x00000000
> >
> > I thought it was an emulator/hardware error, so I used another emulator
> and
> > changed the target board. But the error still persists. Have anyone faced
> > similar issues?
>
> How many 5510 DSPs on your board? How many in the JTAG chain?
>
> -Jeff
>
Reply by ●June 28, 20092009-06-28
The other 4 JTAG devices are 4 8-bit bypass devices. My set up has always had 4
c5510 DSPs and 4 Bypass devices, it was working fine that way.
When you say "does not load correctly", can you be more specific? Is
there an error message?
When I bring up an instance of CCS, I do not see a Disassembly window (Assembly code). But after I do a CPU reset the disassebly window comes up. That makes me wonder if the gel file is getting loaded at the right time.
Vivekjayadev-
> I have 4 DSPs on my target board and 8 JTAG devices in the scan chain.
Ok. What are the non-DSP JTAG devices? We've had trouble before getting
non-TI
devices to "peacefully co-exist" in a CCS + emulator scan chain. Not
that it's
causing a problem in your application, I'm just curious...
> After poking at all the initializations, I realised that the gel file is
not
> getting loaded properly. It picks the right gel file (c5510.gel) but it
does not
> load that correctly. So, I tried a CPU reset in the GEL configuration and
then it
> allows me to load the code. I have to do this every time I open up CCS. For
now I
> have a work around, but I am looking for a permanent fix.
When you say "does not load correctly", can you be more specific? Is
there an error
message? If you look in the .gel file after CCS runs, is part of the GEL code
missing? One thing you might try is:
-make your own version, say vj_c5510.gel
-add some status print messages
-modify CCS Setup to point to the new
.gel file
Then, when you first run CCS, you should be able see your print messages in the
Gel
status window. If the code is only making it part way or something else is
wrong,
this might be a way to debug.
-Jeff
> On Wed, Jun 24, 2009 at 10:53 AM, Jeff Brower
wrote:
>
> Vivekjayadev-
>
> > I am using CCS 3.3, XDS510usb emulator and the target board has
C5510
> DSPs.
> >
> > My setup has been working fine for the past few months. Recently
I have
>
> > been having issues loading my code on to the DSPs. I get the
following
> error:
> >
> > Trouble Writing Register: Error 0x00000004/-1024 Error during:
> Register,
> > PTI_ERR_PREV_IROP_CMD Error Occured at 0x00000000
> >
> > I thought it was an emulator/hardware error, so I used another
emulator
> and
> > changed the target board. But the error still persists. Have
anyone
> faced
> > similar issues?
>
> How many 5510 DSPs on your board? How many in the JTAG chain?
>
> -Jeff
>
When you say "does not load correctly", can you be more specific? Is
there an error message?
When I bring up an instance of CCS, I do not see a Disassembly window (Assembly code). But after I do a CPU reset the disassebly window comes up. That makes me wonder if the gel file is getting loaded at the right time.
Vivekjayadev-
> I have 4 DSPs on my target board and 8 JTAG devices in the scan chain.
Ok. What are the non-DSP JTAG devices? We've had trouble before getting
non-TI
devices to "peacefully co-exist" in a CCS + emulator scan chain. Not
that it's
causing a problem in your application, I'm just curious...
> After poking at all the initializations, I realised that the gel file is
not
> getting loaded properly. It picks the right gel file (c5510.gel) but it
does not
> load that correctly. So, I tried a CPU reset in the GEL configuration and
then it
> allows me to load the code. I have to do this every time I open up CCS. For
now I
> have a work around, but I am looking for a permanent fix.
When you say "does not load correctly", can you be more specific? Is
there an error
message? If you look in the .gel file after CCS runs, is part of the GEL code
missing? One thing you might try is:
-make your own version, say vj_c5510.gel
-add some status print messages
-modify CCS Setup to point to the new
.gel file
Then, when you first run CCS, you should be able see your print messages in the
Gel
status window. If the code is only making it part way or something else is
wrong,
this might be a way to debug.
-Jeff
> On Wed, Jun 24, 2009 at 10:53 AM, Jeff Brower
wrote:
>
> Vivekjayadev-
>
> > I am using CCS 3.3, XDS510usb emulator and the target board has
C5510
> DSPs.
> >
> > My setup has been working fine for the past few months. Recently
I have
>
> > been having issues loading my code on to the DSPs. I get the
following
> error:
> >
> > Trouble Writing Register: Error 0x00000004/-1024 Error during:
> Register,
> > PTI_ERR_PREV_IROP_CMD Error Occured at 0x00000000
> >
> > I thought it was an emulator/hardware error, so I used another
emulator
> and
> > changed the target board. But the error still persists. Have
anyone
> faced
> > similar issues?
>
> How many 5510 DSPs on your board? How many in the JTAG chain?
>
> -Jeff
>
Reply by ●June 29, 20092009-06-29
Vivekjayadev-
> The other 4 JTAG devices are 4 8-bit bypass devices. My set up has
> always had 4 c5510 DSPs and 4 Bypass devices, it
> was working fine that way.
Ok, if the bypass devices are not other devices on the board (other chips), then that means you have v1.2 silicon 5510
devices. v1.2 has a bug that causes each DSP to show 2 JTAG devices, and the bypass configuration is a work-around
allowing CCS to function. If that's the case, then your chips are extremely old. How old is your board? As a note,
the current TI 5510 errata published on their website only covers 5510 v2.10.
> When you say "does not load correctly", can you be more specific? Is
> there an error message?
>
> When I bring up an instance of CCS, I do not see a Disassembly window
> (Assembly code). But after I do a CPU reset the
> disassebly window comes up. That makes me wonder if the gel file is
> getting loaded at the right time.
That sounds normal. But you should be able to see status messages from the GEL code when CCS first runs, that's why I
gave my suggestion below. Did you try it?
The fact a manual reset can fix things might indicate a small change in timing in your system that occurred recently.
Maybe your PC config changed, some rework was done to the DSP board, etc. Possibly the first time CCS runs the GEL
code does a Reset but it doesn't "take".
-Jeff
> Vivekjayadev-
>> I have 4 DSPs on my target board and 8 JTAG devices in the scan chain.
>
> Ok. What are the non-DSP JTAG devices? We've had trouble before getting
> non-TI
> devices to "peacefully co-exist" in a CCS + emulator scan chain. Not
> that it's
> causing a problem in your application, I'm just curious...
>> After poking at all the initializations, I realised that the gel file is
> not
>> getting loaded properly. It picks the right gel file (c5510.gel) but it
> does not
>> load that correctly. So, I tried a CPU reset in the GEL configuration and
> then it
>> allows me to load the code. I have to do this every time I open up CCS. For
> now I
>> have a work around, but I am looking for a permanent fix.
>
> When you say "does not load correctly", can you be more specific? Is
> there an error
> message? If you look in the .gel file after CCS runs, is part of the GEL code
> missing? One thing you might try is:
>
> -make your own version, say vj_c5510.gel
>
> -add some status print messages
>
> -modify CCS Setup to point to the new
> .gel file
>
> Then, when you first run CCS, you should be able see your print messages in the
> Gel
> status window. If the code is only making it part way or something else is
> wrong,
> this might be a way to debug.
>
> -Jeff
>> On Wed, Jun 24, 2009 at 10:53 AM, Jeff Brower
> wrote:
>>
>> Vivekjayadev-
>>
>> > I am using CCS 3.3, XDS510usb emulator and the target board has
> C5510
>> DSPs.
>> >
>> > My setup has been working fine for the past few months. Recently
> I have
>>
>> > been having issues loading my code on to the DSPs. I get the
> following
>> error:
>> >
>> > Trouble Writing Register: Error 0x00000004/-1024 Error during:
>> Register,
>> > PTI_ERR_PREV_IROP_CMD Error Occured at 0x00000000
>> >
>> > I thought it was an emulator/hardware error, so I used another
> emulator
>> and
>> > changed the target board. But the error still persists. Have
> anyone
>> faced
>> > similar issues?
>>
>> How many 5510 DSPs on your board? How many in the JTAG chain?
>>
>> -Jeff
> The other 4 JTAG devices are 4 8-bit bypass devices. My set up has
> always had 4 c5510 DSPs and 4 Bypass devices, it
> was working fine that way.
Ok, if the bypass devices are not other devices on the board (other chips), then that means you have v1.2 silicon 5510
devices. v1.2 has a bug that causes each DSP to show 2 JTAG devices, and the bypass configuration is a work-around
allowing CCS to function. If that's the case, then your chips are extremely old. How old is your board? As a note,
the current TI 5510 errata published on their website only covers 5510 v2.10.
> When you say "does not load correctly", can you be more specific? Is
> there an error message?
>
> When I bring up an instance of CCS, I do not see a Disassembly window
> (Assembly code). But after I do a CPU reset the
> disassebly window comes up. That makes me wonder if the gel file is
> getting loaded at the right time.
That sounds normal. But you should be able to see status messages from the GEL code when CCS first runs, that's why I
gave my suggestion below. Did you try it?
The fact a manual reset can fix things might indicate a small change in timing in your system that occurred recently.
Maybe your PC config changed, some rework was done to the DSP board, etc. Possibly the first time CCS runs the GEL
code does a Reset but it doesn't "take".
-Jeff
> Vivekjayadev-
>> I have 4 DSPs on my target board and 8 JTAG devices in the scan chain.
>
> Ok. What are the non-DSP JTAG devices? We've had trouble before getting
> non-TI
> devices to "peacefully co-exist" in a CCS + emulator scan chain. Not
> that it's
> causing a problem in your application, I'm just curious...
>> After poking at all the initializations, I realised that the gel file is
> not
>> getting loaded properly. It picks the right gel file (c5510.gel) but it
> does not
>> load that correctly. So, I tried a CPU reset in the GEL configuration and
> then it
>> allows me to load the code. I have to do this every time I open up CCS. For
> now I
>> have a work around, but I am looking for a permanent fix.
>
> When you say "does not load correctly", can you be more specific? Is
> there an error
> message? If you look in the .gel file after CCS runs, is part of the GEL code
> missing? One thing you might try is:
>
> -make your own version, say vj_c5510.gel
>
> -add some status print messages
>
> -modify CCS Setup to point to the new
> .gel file
>
> Then, when you first run CCS, you should be able see your print messages in the
> Gel
> status window. If the code is only making it part way or something else is
> wrong,
> this might be a way to debug.
>
> -Jeff
>> On Wed, Jun 24, 2009 at 10:53 AM, Jeff Brower
> wrote:
>>
>> Vivekjayadev-
>>
>> > I am using CCS 3.3, XDS510usb emulator and the target board has
> C5510
>> DSPs.
>> >
>> > My setup has been working fine for the past few months. Recently
> I have
>>
>> > been having issues loading my code on to the DSPs. I get the
> following
>> error:
>> >
>> > Trouble Writing Register: Error 0x00000004/-1024 Error during:
>> Register,
>> > PTI_ERR_PREV_IROP_CMD Error Occured at 0x00000000
>> >
>> > I thought it was an emulator/hardware error, so I used another
> emulator
>> and
>> > changed the target board. But the error still persists. Have
> anyone
>> faced
>> > similar issues?
>>
>> How many 5510 DSPs on your board? How many in the JTAG chain?
>>
>> -Jeff
Reply by ●July 2, 20092009-07-02
The target board has been around for a while. I will try to find more
information about the DSPs on the website.
I did try to add a couple of status messages in the gel file and was able to see them in the status window.
I downloaded a new c5510.gel file from the website and redirected CCS to pick the new one. And the issue seems to have resolved. It might have been something with the old gel file or manually resetting the chip.
Thanks for the feedback...
PS: It almost takes 4 days for my message to get posted on the forum. Is it normal.
I did try to add a couple of status messages in the gel file and was able to see them in the status window.
I downloaded a new c5510.gel file from the website and redirected CCS to pick the new one. And the issue seems to have resolved. It might have been something with the old gel file or manually resetting the chip.
Thanks for the feedback...
PS: It almost takes 4 days for my message to get posted on the forum. Is it normal.
Reply by ●July 2, 20092009-07-02
Vivekjayadev-
> The target board has been around for a while. I will try to find
> more information about the DSPs on the website.
>
> I did try to add a couple of status messages in the gel file and
> was able to see them in the status window.
Ok sounds good. That's kind of like a "built in self test" -- useful to make sure
things are working Ok before you try to download and run code.
> I downloaded a new c5510.gel file from the website and redirected
> CCS to pick the new one. And the issue seems to have resolved. It
> might have been something with the old gel file or manually
> resetting the chip.
Hmm... that's good, but for some reason I'm not sure your problem won't come back.
Anyway, Ok for the time being.
> Thanks for the feedback...
>
> PS: It almost takes 4 days for my message to get posted on the forum. Is it normal.
No it's not normal. The Yahoo DSP groups are moderated, but normally it's same-day.
I think some frequent posters are set for "auto clear" or similar so there is no
delay, and for new members there would be some delay, but again shouldn't be more
then a day. I don't know why the posts are slow lately.
-Jeff
> Vivekjayadev-
>
> > The other 4 JTAG devices are 4 8-bit bypass devices. My set up has
> > always had 4 c5510 DSPs and 4 Bypass devices, it
> > was working fine that way.
>
> Ok, if the bypass devices are not other devices on the board (other chips), then that means you have v1.2 silicon 5510
> devices. v1.2 has a bug that causes each DSP to show 2 JTAG devices, and the bypass configuration is a work-around
> allowing CCS to function. If that's the case, then your chips are extremely old. How old is your board? As a note,
> the current TI 5510 errata published on their website only covers 5510 v2.10.
>
> > When you say "does not load correctly", can you be more specific? Is
> > there an error message?
> >
> > When I bring up an instance of CCS, I do not see a Disassembly window
> > (Assembly code). But after I do a CPU reset the
> > disassebly window comes up. That makes me wonder if the gel file is
> > getting loaded at the right time.
>
> That sounds normal. But you should be able to see status messages from the GEL code when CCS first runs, that's why I
> gave my suggestion below. Did you try it?
>
> The fact a manual reset can fix things might indicate a small change in timing in your system that occurred recently.
> Maybe your PC config changed, some rework was done to the DSP board, etc. Possibly the first time CCS runs the GEL
> code does a Reset but it doesn't "take".
>
> -Jeff
>
> > Vivekjayadev-
> >> I have 4 DSPs on my target board and 8 JTAG devices in the scan chain.
> >
> > Ok. What are the non-DSP JTAG devices? We've had trouble before getting
> > non-TI
> > devices to "peacefully co-exist" in a CCS + emulator scan chain. Not
> > that it's
> > causing a problem in your application, I'm just curious...
> >> After poking at all the initializations, I realised that the gel file is
> > not
> >> getting loaded properly. It picks the right gel file (c5510.gel) but it
> > does not
> >> load that correctly. So, I tried a CPU reset in the GEL configuration and
> > then it
> >> allows me to load the code. I have to do this every time I open up CCS. For
> > now I
> >> have a work around, but I am looking for a permanent fix.
> >
> > When you say "does not load correctly", can you be more specific? Is
> > there an error
> > message? If you look in the .gel file after CCS runs, is part of the GEL code
> > missing? One thing you might try is:
> >
> > -make your own version, say vj_c5510.gel
> >
> > -add some status print messages
> >
> > -modify CCS Setup to point to the new
> > .gel file
> >
> > Then, when you first run CCS, you should be able see your print messages in the
> > Gel
> > status window. If the code is only making it part way or something else is
> > wrong,
> > this might be a way to debug.
> >
> > -Jeff
> >> On Wed, Jun 24, 2009 at 10:53 AM, Jeff Brower
> > wrote:
> >>
> >> Vivekjayadev-
> >>
> >> > I am using CCS 3.3, XDS510usb emulator and the target board has
> > C5510
> >> DSPs.
> >> >
> >> > My setup has been working fine for the past few months. Recently
> > I have
> >>
> >> > been having issues loading my code on to the DSPs. I get the
> > following
> >> error:
> >> >
> >> > Trouble Writing Register: Error 0x00000004/-1024 Error during:
> >> Register,
> >> > PTI_ERR_PREV_IROP_CMD Error Occured at 0x00000000
> >> >
> >> > I thought it was an emulator/hardware error, so I used another
> > emulator
> >> and
> >> > changed the target board. But the error still persists. Have
> > anyone
> >> faced
> >> > similar issues?
> >>
> >> How many 5510 DSPs on your board? How many in the JTAG chain?
> >>
> >> -Jeff
> The target board has been around for a while. I will try to find
> more information about the DSPs on the website.
>
> I did try to add a couple of status messages in the gel file and
> was able to see them in the status window.
Ok sounds good. That's kind of like a "built in self test" -- useful to make sure
things are working Ok before you try to download and run code.
> I downloaded a new c5510.gel file from the website and redirected
> CCS to pick the new one. And the issue seems to have resolved. It
> might have been something with the old gel file or manually
> resetting the chip.
Hmm... that's good, but for some reason I'm not sure your problem won't come back.
Anyway, Ok for the time being.
> Thanks for the feedback...
>
> PS: It almost takes 4 days for my message to get posted on the forum. Is it normal.
No it's not normal. The Yahoo DSP groups are moderated, but normally it's same-day.
I think some frequent posters are set for "auto clear" or similar so there is no
delay, and for new members there would be some delay, but again shouldn't be more
then a day. I don't know why the posts are slow lately.
-Jeff
> Vivekjayadev-
>
> > The other 4 JTAG devices are 4 8-bit bypass devices. My set up has
> > always had 4 c5510 DSPs and 4 Bypass devices, it
> > was working fine that way.
>
> Ok, if the bypass devices are not other devices on the board (other chips), then that means you have v1.2 silicon 5510
> devices. v1.2 has a bug that causes each DSP to show 2 JTAG devices, and the bypass configuration is a work-around
> allowing CCS to function. If that's the case, then your chips are extremely old. How old is your board? As a note,
> the current TI 5510 errata published on their website only covers 5510 v2.10.
>
> > When you say "does not load correctly", can you be more specific? Is
> > there an error message?
> >
> > When I bring up an instance of CCS, I do not see a Disassembly window
> > (Assembly code). But after I do a CPU reset the
> > disassebly window comes up. That makes me wonder if the gel file is
> > getting loaded at the right time.
>
> That sounds normal. But you should be able to see status messages from the GEL code when CCS first runs, that's why I
> gave my suggestion below. Did you try it?
>
> The fact a manual reset can fix things might indicate a small change in timing in your system that occurred recently.
> Maybe your PC config changed, some rework was done to the DSP board, etc. Possibly the first time CCS runs the GEL
> code does a Reset but it doesn't "take".
>
> -Jeff
>
> > Vivekjayadev-
> >> I have 4 DSPs on my target board and 8 JTAG devices in the scan chain.
> >
> > Ok. What are the non-DSP JTAG devices? We've had trouble before getting
> > non-TI
> > devices to "peacefully co-exist" in a CCS + emulator scan chain. Not
> > that it's
> > causing a problem in your application, I'm just curious...
> >> After poking at all the initializations, I realised that the gel file is
> > not
> >> getting loaded properly. It picks the right gel file (c5510.gel) but it
> > does not
> >> load that correctly. So, I tried a CPU reset in the GEL configuration and
> > then it
> >> allows me to load the code. I have to do this every time I open up CCS. For
> > now I
> >> have a work around, but I am looking for a permanent fix.
> >
> > When you say "does not load correctly", can you be more specific? Is
> > there an error
> > message? If you look in the .gel file after CCS runs, is part of the GEL code
> > missing? One thing you might try is:
> >
> > -make your own version, say vj_c5510.gel
> >
> > -add some status print messages
> >
> > -modify CCS Setup to point to the new
> > .gel file
> >
> > Then, when you first run CCS, you should be able see your print messages in the
> > Gel
> > status window. If the code is only making it part way or something else is
> > wrong,
> > this might be a way to debug.
> >
> > -Jeff
> >> On Wed, Jun 24, 2009 at 10:53 AM, Jeff Brower
> > wrote:
> >>
> >> Vivekjayadev-
> >>
> >> > I am using CCS 3.3, XDS510usb emulator and the target board has
> > C5510
> >> DSPs.
> >> >
> >> > My setup has been working fine for the past few months. Recently
> > I have
> >>
> >> > been having issues loading my code on to the DSPs. I get the
> > following
> >> error:
> >> >
> >> > Trouble Writing Register: Error 0x00000004/-1024 Error during:
> >> Register,
> >> > PTI_ERR_PREV_IROP_CMD Error Occured at 0x00000000
> >> >
> >> > I thought it was an emulator/hardware error, so I used another
> > emulator
> >> and
> >> > changed the target board. But the error still persists. Have
> > anyone
> >> faced
> >> > similar issues?
> >>
> >> How many 5510 DSPs on your board? How many in the JTAG chain?
> >>
> >> -Jeff
