Toshiba T1000
Posted: Tue 07 Nov, 2017 10:50 pm
I'm currently working on the Toshiba T1000, the T3100e's little brother. This has a number of unique features, and I'd welcome some advice regarding a couple:
i) The T1000 has an optional 768k memory expansion, which is battery-backed and can be configured as a persistent RAMdisk. The question is how I should attempt to implement this in PCEM -- as a very small hard drive with a fixed format, or as a very big NVRAM?
ii) I'm experiencing lockups in screen repainting. Looking in the debugger, there appears to be a deadlock: the video blit thread is waiting for the wake_blit_thread event:
while the main thread is waiting for the buffer_not_in_use event:
It only seems to happen when using my T1000 video output code, and I'm not sure why since it's pretty much the same as the T3100e video code, which doesn't lock up as far as I can see.
I've attached the T1000 patch as it stands, in case anyone wants to look over my code and point out what I'm doing wrong. This is definitely work in progress and not ready to merge yet, even if the project wasn't currently in feature freeze.
i) The T1000 has an optional 768k memory expansion, which is battery-backed and can be configured as a persistent RAMdisk. The question is how I should attempt to implement this in PCEM -- as a very small hard drive with a fixed format, or as a very big NVRAM?
ii) I'm experiencing lockups in screen repainting. Looking in the debugger, there appears to be a deadlock: the video blit thread is waiting for the wake_blit_thread event:
Code: Select all
thread_wait_event(blit_data.wake_blit_thread, -1);
Code: Select all
thread_wait_event(blit_data.buffer_not_in_use, 1);
I've attached the T1000 patch as it stands, in case anyone wants to look over my code and point out what I'm doing wrong. This is definitely work in progress and not ready to merge yet, even if the project wasn't currently in feature freeze.