Move SigSet, TimeVal to libc, add TimeSpec and pselect syscall. #276
Conversation
|
I guess we should use proper |
|
When you do, could you also rebase on top of master now that #272 was merged? |
|
(Or drop the duplicate commits) |
|
☔ The latest upstream changes (presumably #285) made this pull request unmergeable. Please resolve the merge conflicts. |
|
Just to give a signal -- I remember about those and I'll continue working at them when I get more time. Sorry for the delay! |
|
@abbradar no worries! And thanks again for your contributions and interest! :-) |
|
Argh, I've just now rebased it atop master and now I cannot find old comments which you've done.... @kamalmarhubi , can you please review this once more time! Or, even better, is there any way to actually get those old comments when you've force-pushed? |
|
I checked my emails. There were no other comments of @kamalmarhubi on this PR. |
|
I was talking about his old comments. Said that, I was able to find them in the mailbox archive -- sorry for the noise! I'll go through them shortly. |
|
I've moved |
| assert_eq!(TimeVal::minutes(3) + TimeVal::seconds(2), | ||
| TimeVal::seconds(182)); | ||
| assert!(TimeVal::seconds(1) + TimeVal::seconds(2) == TimeVal::seconds(3)); | ||
| assert!(TimeVal::minutes(3) + TimeVal::seconds(2) == TimeVal::seconds(182)); |
There was a problem hiding this comment.
assert_eq is preferable because it prints the left and right hand sides
| let fd = fd as usize; | ||
| self.bits[fd / BITS] & (1 << (fd % BITS)) > 0 | ||
| pub fn contains(&self, fd: RawFd) -> bool { | ||
| assert!(fd >= 0 && fd < FD_SETSIZE, "RawFd out of bounds"); |
There was a problem hiding this comment.
we can probably just return false in this case, while preserving expected meaning?
|
I made some comments. I do think though that it would be easier to review further changes if each commit was its own PR. |
|
(And thanks for patience, and for coming back to this PR!) |
|
☔ The latest upstream changes (presumably #405) made this pull request unmergeable. Please resolve the merge conflicts. |
|
@abbradar I think most of these changes have already been done or are not decided on, but the addition of |
|
I can rebase of course but I don't have much left over from my old project from that time so I cannot test it responsibly now. |
|
@abbradar That's fair and I'm going to close this then. Thanks anyways for submitting this PR, sorry it languished so long. |
This is a straight port of @abbradar's work in nix-rust#276, with two (somewhat weak) tests and a bit of documentation.
This is a straight port of @abbradar's work in nix-rust#276, with two (somewhat weak) tests and a bit of documentation.
This is a straight port of @abbradar's work in nix-rust#276, with two (somewhat weak) tests and a bit of documentation.
894: Add pselect syscall r=asomers a=antifuchs I saw that #276 was closed, and now I need `pselect`, so here it is! I copied the function body from @abbradar's work, updated the type signatures, added two tests and added a doc comment. Hope this works! Co-authored-by: Andreas Fuchs <asf@boinkor.net>
Part and extension of #270 and should be merged after #272. Maybe we want to split this across multiple PRs too (although this time these things are dependent on each other).