Re: [PATCH] seqlock: serialize against writers
From: Andi Kleen <hidden>
Date: 2008-08-29 17:06:48
Also in:
lkml
From: Andi Kleen <hidden>
Date: 2008-08-29 17:06:48
Also in:
lkml
No idea on the problem itself because I don't understand it, just:
Given that, we either need to address the const requirement for userspace, alter userspace's usage, or find another way to address the deadlock. Any ideas?
In the worst case you can always disable the vsyscall gtod() and thus the need for a user space safe seqlock_t, but it will hurt some workloads significantly. -Andi