-
Notifications
You must be signed in to change notification settings - Fork 3
ns16550a: Buy back 16088 asy can overrun TX on interrupt race #57
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
base: arm64-gate
Are you sure you want to change the base?
Conversation
pfmooney
left a comment
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
One small nit, but yeah, looks fine for the commit to bear my name. Thanks for checking.
| struct nsasyncline *nsasync = ns16550->ns16550_priv; | ||
| int ret_status = DDI_INTR_UNCLAIMED; | ||
| uchar_t interrupt_id, lsr; | ||
| uchar_t lsr; |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Move this inside the for loop and make it const there like in upstream?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Sure, updated.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks.
13f6382 to
774011c
Compare
|
Might want to wait until the 16107 fix lands |
(includes the later fix for 16107 asy VERIFY too strong in face of polledio)
774011c to
0329712
Compare
Done |
|
Thanks for waiting. Sorry about the extra trouble. |
|
@hadfl says this gets stally on rpi (though it doesn't seem to on qemu). That said, it seems to be stalling for interrupts, which we've seen go ok on qemu and not raspberry pi before. I don't immediately see anything I did wrong merging with Patrick, though? |
This takes Patrick's fix for illumos#16088 and puts it into ns16550 as well (which is an almost direct copy of asy for mysterious reasons, remember).
@hadfl @citrus-it It seems best to double check this on raspberry pi, though I still wish we knew why this wasn't just asy(4D) in the first place
@pfmooney can you confirm you're ok with my putting your name on this, since it's directly taken out of your fix?