code review 76080044: runtime: add stack growth tests

runtime: add stack growth tests
Also move generated code into a separate file,
because it's difficult to work with the file otherwise.

[dvyukov, 2014-03-14 15:13:35]

Hello golang-codereviews@googlegroups.com,

I'd like you to review this change to
https://dvyukov%40google.com@code.google.com/p/go/

[khr, 2014-03-17 21:50:44]

LGTM.

https://codereview.appspot.com/76080044/diff/30003/src/pkg/runtime/stack_test.go
File src/pkg/runtime/stack_test.go (right):

https://codereview.appspot.com/76080044/diff/30003/src/pkg/runtime/stack_test...
src/pkg/runtime/stack_test.go:179: _ = x
Why is this necessary?
Maybe you should check that x==i+1?

https://codereview.appspot.com/76080044/diff/30003/src/pkg/runtime/stack_test...
src/pkg/runtime/stack_test.go:194: _ = x
Ditto

[dvyukov, 2014-03-19 13:22:36]

https://codereview.appspot.com/76080044/diff/30003/src/pkg/runtime/stack_test.go
File src/pkg/runtime/stack_test.go (right):

https://codereview.appspot.com/76080044/diff/30003/src/pkg/runtime/stack_test...
src/pkg/runtime/stack_test.go:179: _ = x
On 2014/03/17 21:50:44, khr wrote:
> Why is this necessary?
> Maybe you should check that x==i+1?

Done.

https://codereview.appspot.com/76080044/diff/30003/src/pkg/runtime/stack_test...
src/pkg/runtime/stack_test.go:194: _ = x
On 2014/03/17 21:50:44, khr wrote:
> Ditto

Done.

[dvyukov, 2014-03-19 13:23:01]

*** Submitted as https://code.google.com/p/go/source/detail?r=b3ac9687b0c9 ***

runtime: add stack growth tests
Also move generated code into a separate file,
because it's difficult to work with the file otherwise.

LGTM=khr
R=golang-codereviews, khr
CC=golang-codereviews
https://codereview.appspot.com/76080044

[gobot, 2014-03-19 13:29:24]

This CL appears to have broken the freebsd-386 builder.
See http://build.golang.org/log/336f89da6acb8eb37c665d08208a20b0ccb48999

[0intro, 2014-03-19 16:10:07]

It seems the parallel runtime test is sometimes timing out on Plan 9 since the
introduction of the TestStackGrowth test. We're investigating it.

[dvyukov, 2014-03-19 16:17:19]

On Wed, Mar 19, 2014 at 8:10 PM,  <0intro@gmail.com> wrote:
> It seems the parallel runtime test is sometimes timing out on Plan 9
> since the introduction of the TestStackGrowth test. We're investigating
> it.

It's slow on beefy machines as well (2.3 seconds, while I usually aim
at 0.1s for -short mode).
We probably need to reduce constants in the test in -short mode.