Re: CONFIG_VMAP_STACK, on-stack struct, and wake_up_bit

[Bob Peterson]

----- Original Message -----
| On Wed, Oct 26, 2016 at 10:15 AM, Linus Torvalds
| <torvalds@xxxxxxxxxxxxxxxxxxxx> wrote:
| >
| > Oh, and the patch is obviously entirely untested. I wouldn't want to
| > ruin my reputation by *testing* the patches I send out. What would be
| > the fun in that?
| 
| So I tested it. It compiles, and it actually also solves the
| performance problem I was complaining about a couple of weeks ago with
| "unlock_page()" having an insane 3% CPU overhead when doing lots of
| small script ("make -j16 test" in the git tree for those that weren't
| involved in the original thread three weeks ago).
| 
| So quite frankly, I'll just commit it. It should fix the new problem
| with gfs2 and CONFIG_VMAP_STACK, and I see no excuse for the crazy
| zone stuff considering how harmful it is to everybody else.
| 
| I expect that when the NUMA people complain about page locking (if
| they ever even notice), PeterZ will stand up like the hero he is, and
| say "look here, I can solve this for you".
| 
|                     Linus
| 
I can test it for you, if you give me about an hour.

Bob Peterson