Re: [RFC PATCH-tip v4 07/10] locking/rwsem: Change RWSEM_WAITING_BIAS for better disambiguation
From: Wanpeng Li <hidden>
Date: 2016-08-22 02:15:34
Also in:
linux-arch, linux-s390, lkml
From: Wanpeng Li <hidden>
Date: 2016-08-22 02:15:34
Also in:
linux-arch, linux-s390, lkml
2016-08-20 0:21 GMT+08:00 Waiman Long [off-list ref]:
On 08/19/2016 01:57 AM, Wanpeng Li wrote:quoted
2016-08-19 5:11 GMT+08:00 Waiman Long[off-list ref]:quoted
When the count value is in between 0 and RWSEM_WAITING_BIAS, there are 2 possibilities. Either a writer is present and there is no waitercount = 0xffff0001quoted
or there are waiters and readers. There is no easy way tocount = 0xffff000X However, RWSEM_WAITING_BIAS is equal to 0xffff0000, so both these two cases are beyond RWSEM_WAITING_BIAS, right? Regards, Wanpeng LiPerhaps I should make it clear that I am talking from a signed quantity point of view (it is an atomic_long_t). So RWSEM_WAITING_BIAS < RWSEM_ACTIVE_WRITE_BIAS < 0 Hope this clarify your question.
Yeah, thank you. :) Regards, Wanpeng Li