Don't consider 7th bit of signal to calculate signal for Linux platform. #549
Conversation
It's used for PTRACE_O_SYSGOOD, and creates a bad signal if on.
|
We need a way to return whether the bit is set though |
| @@ -81,7 +81,7 @@ mod status { | |||
| } | |||
|
|
|||
| pub fn stop_signal(status: i32) -> Signal { | |||
| Signal::from_c_int((status & 0xFF00) >> 8).unwrap() | |||
| Signal::from_c_int((status & 0x7F00) >> 8).unwrap() | |||
There was a problem hiding this comment.
Can you add a comment explaining the mask?
|
Need to double check the cross-platform-ness of this fix, but it seems reasonable. Could you add something to the changelog for this? |
|
I'll add the comments/changelog entry when I get a chance (soon). I think the presence of a set 7th bit should be returned in the WaitStatus somehow, so that the caller can read it. What do you think would be the best way to represent this? I feel like adding it as a PtraceEvent would be the most straightforward to implement in a platform-specific way (both are exclusive to Linux). The architecture availability of this could be managed through enabling/disabling a particular value of the event enum used in PtraceEvent |
|
I'm not sure yet, as I'm not really familiar with |
|
@chaosagent We'd like to close this issue in favor of #566. Can you see if that PR works for your needs as well and seems like the correct path forward? |
|
I believe this was addressed by #566, so I'm closing this one. Please reopen or file a new one if not. |
It's used for PTRACE_O_SYSGOOD, and creates a bad signal if on.