Change Details

While waiting for input in the read builtin, if select() is interrupted but there is no pending signal, we act like we timed out, and return the same status as if we had been interrupted by SIGALRM, instead of looping until we actually do time out. * Improve validation of the timeout value. * Replace the single select() call with a ppoll() loop. * Add a test case that verifies that `read -t 0` behaves correctly when input is available; we already have one that verifies that it does when there isn't. PR: 290844 MFC after: 1 week

While waiting for input in the read builtin, if select() is interrupted but there is no pending signal, we act like we timed out, and return the same status as if we had been interrupted by SIGALRM, instead of looping until we actually do time out. * Replace the single select() call with a ppoll() loop. * Add a test case that verifies that `read -t 0` behaves correctly when input is available; we already have one that verifies that it does when there isn't. * Improve validation of the timeout value. We now accept things like "1h30m15s", which we didn't use to, but not "5sec", which we did. PR: 290844 MFC after: 1 week

While waiting for input in the read builtin, if select() is interrupted but there is no pending signal, we act like we timed out, and return the same status as if we had been interrupted by SIGALRM, instead of looping until we actually do time out. * Improve validation of the timeout value. * Replace the single select() call with a ppoll() loop. * Add a test case that verifies that `read -t 0` behaves correctly when input is available; we already have one that verifies that it does when there isn't. * Improve validation of the timeout value. We now accept things like "1h30m15s", which we didn't use to, but not "5sec", which we did. PR: 290844 MFC after: 1 week