kmalloc cannot figure out whether sleeping is allowable or not

Posted Apr 1, 2008 0:02 UTC (Tue) by rusty (guest, #26)
Parent article: Atomic context and kernel API design

> In fact, kmalloc() cannot figure out on its own whether sleeping is
> allowable or not...

You are (of course) correct.  Damn, but it was such a beautiful example; 
the same one I used in 2003 in my OLS keynote, and noone spotted it then 
either.

Thanks,
Rusty.

kmalloc cannot figure out whether sleeping is allowable or not

Posted Apr 2, 2008 0:54 UTC (Wed) by roelofs (guest, #2599) [Link]

Damn, but it was such a beautiful example; the same one I used in 2003 in my OLS keynote, and noone spotted it then either.

At least you're in good company. ;-)

A somewhat analogous "discovery" involved the thread-(non)safety of the double-checked locking pattern in C++, which various people (possibly including Scott Effective C++ Meyers himself) espoused for some years prior to the publication of an article describing its problems [PDF].

Greg