Alan Donovan

That's a dead cmd/compile process: there's no command because argv has been destroyed along with the rest of the address space. Perhaps the go list parent simply hasn't called waitpid...

A simpler explanation is that gopls sent SIGINT, waited for 1s, then sent SIGKILL, and the child (go) exited during the 1s grace period, causing kill(2) to fail with ESRCH...

> the panic indicates that we waited up to 10 seconds after that kill. The error from Kill is merely logged. Good point. So the sequence of events that we...

> Is it possible that the dump shows a different go process? Seems improbable, but I feel like we've eliminated the impossible. We should log the pid of cmd.Process, like...

We still haven't seen a single log with both (a) the "error killing the Go command" message and (b) a pid showing that 'go' is the child of the test....

> [domink] As a tool author I'm in favour of this change, but I am conveniently ignoring that this is a backwards incompatible change. That's my feeling too. I wish...

Thanks for clarifying. If the driver can't support it, then please consider the alternative of specifically rejecting time.Duration from the generic conversion applied to other integer data types. The error...

> `time.Duration` is not a valid [`driver.Value`](https://golang.org/pkg/database/sql/driver/#Value) and doesn't implement the [`driver.Valuer`](https://golang.org/pkg/database/sql/driver/#Valuer) interface. You should not expect it to work in a special way. `database/sql` just use the native representation...

> I don't think that's a task for the MySQL driver. The same mistake could happen when using `database/sql` with any other SQL driver. So I don't think that's the...

> It's programmer's responsibility to know something under the hood and make sure their codes act as what they want, not driver. It's a car-driver's responsibility to drive safely, but...