On Thu, 11 May 2017, Kamil Rytarowski wrote:
On 11.05.2017 15:17, Taylor R Campbell wrote:Date: Thu, 11 May 2017 20:35:19 +0800 (+08) From: Paul Goyette <paul%whooppee.com@localhost> On Thu, 11 May 2017, Kengo NAKAHARA wrote:(1) Why splsoftserial() is required instead of kpreempt_disable()? localcount_drain() uses xc_broadcast(0, ...), that is, it uses low priority xcall. Low priority xcall would be done by kthread context, so I think kpreempt_disable() would be sufficient to prevent localcount_drain() xcall running.I think you are correct. Taylor, do you agree?Yes, I think this is fine. I probably chose splsoftserial because I was thinking of pserialize(9).While there, locking.9 is begging for being updated for new APIs.
Well, it looks to me like almost everything is listed there, except for psref(9) (and now, of course, localcount(9)).
It's interesting that you added info for pserialize(9) but did not add psref(9)!
Anyway, I will make a note to add a paragraph for localcount(9). +------------------+--------------------------+----------------------------+ | Paul Goyette | PGP Key fingerprint: | E-mail addresses: | | (Retired) | FA29 0E3B 35AF E8AE 6651 | paul at whooppee dot com | | Kernel Developer | 0786 F758 55DE 53BA 7731 | pgoyette at netbsd dot org | +------------------+--------------------------+----------------------------+