800ns? It is not that bad is it?terriblefire wrote: ↑Sat Jan 02, 2021 6:36 pm Just FYI..
here is the times you're getting
bfe001.JPG
TF CD32 Riser Revision 2 Design Complete
Moderators: terriblefire, Terriblefire Moderator
- arkadiusz.makarenko
- Moderator Team
- Posts: 1208
- Joined: Wed Jun 19, 2019 7:36 am
- Location: Edinburgh
Re: TF CD32 Riser Revision 2 Design Complete
Do not trust people. They are capable of greatness.
~ Stanislaw Lem
~ Stanislaw Lem
-
- Moderator Team
- Posts: 5389
- Joined: Mon Aug 28, 2017 10:56 pm
- Location: Glasgow, UK
Re: TF CD32 Riser Revision 2 Design Complete
Will do. ignore my last. the speed is 7uS to complete a cycle with the ARM. Thats way way too slow.arkadiusz.makarenko wrote: ↑Sat Jan 02, 2021 6:59 pm Could you check this? @terriblefire
There are two changes, KBD_Clock is fixed to one setting Open Drain and never reconfigured. I checked on datawrites and seems to be ok.
second change is simple optimization -OFast insead of -ODebug maybe speed will be a little but better?
———
"It is not necessarily a supply voltage at no load, but the amount of current it can provide when touched that
indicates how much hurting you shall receive."
"It is not necessarily a supply voltage at no load, but the amount of current it can provide when touched that
indicates how much hurting you shall receive."
-
- Moderator Team
- Posts: 5389
- Joined: Mon Aug 28, 2017 10:56 pm
- Location: Glasgow, UK
Re: TF CD32 Riser Revision 2 Design Complete
New firmwareterriblefire wrote: ↑Sat Jan 02, 2021 7:22 pmWill do. ignore my last. the speed is 7uS to complete a cycle with the ARM. Thats way way too slow.arkadiusz.makarenko wrote: ↑Sat Jan 02, 2021 6:59 pm Could you check this? @terriblefire
There are two changes, KBD_Clock is fixed to one setting Open Drain and never reconfigured. I checked on datawrites and seems to be ok.
second change is simple optimization -OFast insead of -ODebug maybe speed will be a little but better?
this is what it looks like without a mouse plugged in
No change in reset.
———
"It is not necessarily a supply voltage at no load, but the amount of current it can provide when touched that
indicates how much hurting you shall receive."
"It is not necessarily a supply voltage at no load, but the amount of current it can provide when touched that
indicates how much hurting you shall receive."
- arkadiusz.makarenko
- Moderator Team
- Posts: 1208
- Joined: Wed Jun 19, 2019 7:36 am
- Location: Edinburgh
Re: TF CD32 Riser Revision 2 Design Complete
I have trimmed what I could in Interrupt routine quickly, and changed Open Drain to PushPull for KBD_Clock.
So I hope it will force to keep the reset line correctly?
Except that I have changes IRQ priorities, now it gives highest priority for Amiga, and lower for USB.
So I hope it will force to keep the reset line correctly?
Except that I have changes IRQ priorities, now it gives highest priority for Amiga, and lower for USB.
Do not trust people. They are capable of greatness.
~ Stanislaw Lem
~ Stanislaw Lem
-
- Moderator Team
- Posts: 5389
- Joined: Mon Aug 28, 2017 10:56 pm
- Location: Glasgow, UK
Re: TF CD32 Riser Revision 2 Design Complete
I have a suspicion the 020 reset line is getting stuck low. I will investigate.
Will try that tomorrow. I have noticed a few hangs when using a mouse.
Will try that tomorrow. I have noticed a few hangs when using a mouse.
———
"It is not necessarily a supply voltage at no load, but the amount of current it can provide when touched that
indicates how much hurting you shall receive."
"It is not necessarily a supply voltage at no load, but the amount of current it can provide when touched that
indicates how much hurting you shall receive."
- arkadiusz.makarenko
- Moderator Team
- Posts: 1208
- Joined: Wed Jun 19, 2019 7:36 am
- Location: Edinburgh
Re: TF CD32 Riser Revision 2 Design Complete
Double checked everything, and now I release CPU as early as technically possible.
I have added macros as well, lets see if this made any difference at all.
I have added macros as well, lets see if this made any difference at all.
Do not trust people. They are capable of greatness.
~ Stanislaw Lem
~ Stanislaw Lem
Re: TF CD32 Riser Revision 2 Design Complete
I just checked your latest firmware guys it looks like everything is working fine amiga as if it would start faster
keyboard did not start after inserting usb dongle
I had to turn off and on amiga Then everything start working properly
keyboard did not start after inserting usb dongle
I had to turn off and on amiga Then everything start working properly
- arkadiusz.makarenko
- Moderator Team
- Posts: 1208
- Joined: Wed Jun 19, 2019 7:36 am
- Location: Edinburgh
Re: TF CD32 Riser Revision 2 Design Complete
I have removed keyboard sync, this firmware should allow you use of keyboard straight away. But I am 99% certain that your issues are related to your mouse keyboard setup.
- Attachments
-
- TFRiserRev2F722.zip
- (21.69 KiB) Downloaded 144 times
Do not trust people. They are capable of greatness.
~ Stanislaw Lem
~ Stanislaw Lem
- arkadiusz.makarenko
- Moderator Team
- Posts: 1208
- Joined: Wed Jun 19, 2019 7:36 am
- Location: Edinburgh
Re: TF CD32 Riser Revision 2 Design Complete
I was thinking how I can speed up things even more.
There very little that can be done inside the routine (I will optimize it even more, later, as there is still room for improvement)
But main issue is using HAL libraries for handling EXTI. There is a lot of overhead there which can be optimized.
Main issue is that I use only one EXTI group per one pin/ INTSIG. I should be able to trim all the checks and few other bits as well, and speed up by setting everything directly on registers, but challenge is that some elements need to stay as they are for USB stack to work.
There very little that can be done inside the routine (I will optimize it even more, later, as there is still room for improvement)
But main issue is using HAL libraries for handling EXTI. There is a lot of overhead there which can be optimized.
Main issue is that I use only one EXTI group per one pin/ INTSIG. I should be able to trim all the checks and few other bits as well, and speed up by setting everything directly on registers, but challenge is that some elements need to stay as they are for USB stack to work.
Do not trust people. They are capable of greatness.
~ Stanislaw Lem
~ Stanislaw Lem
- arkadiusz.makarenko
- Moderator Team
- Posts: 1208
- Joined: Wed Jun 19, 2019 7:36 am
- Location: Edinburgh
Re: TF CD32 Riser Revision 2 Design Complete
All interrupts except RTC are handled directly.
I will now go through disassembly and will look to optimize routines even more.
I will now go through disassembly and will look to optimize routines even more.
- Attachments
-
- TFRiserRev2F722.zip
- (21.48 KiB) Downloaded 154 times
Do not trust people. They are capable of greatness.
~ Stanislaw Lem
~ Stanislaw Lem