On 01/30/2019 02:30 PM, Alexei Starovoitov wrote:
> On Wed, Jan 30, 2019 at 11:15:30AM +0100, Peter Zijlstra wrote:
>> On Tue, Jan 29, 2019 at 08:04:56PM -0800, Alexei Starovoitov wrote:
>>> Lockdep warns about false positive:
>> This is not a false positive, and you probably also need to use
>> down_read_non_owner() to match this up_read_non_owner().
>>
>> {up,down}_read() and {up,down}_read_non_owner() are not only different
>> in the lockdep annotation; there is also optimistic spin stuff that
>> relies on 'owner' tracking.
> Can you point out in the code the spin bit?
> As far as I can see sem->owner is debug only feature.
> All owner checks are done under CONFIG_DEBUG_RWSEMS.No, sem->owner is mainly for performing optimistic spinning which is a performance feature to make rwsem writer-lock performs similar to mutex. The debugging part is just an add-on. It is not the reason for the presence of sem->owner. > Also there is no down_read_trylock_non_owner() at the moment. > We can argue about it for -next, but I'd rather silence lockdep > with this patch today. > We can add down_read_trylock_non_owner() if there is a need for it. It should be easy to do. Cheers, Longman down_read_trylock_non_owner(
