Re: rw_enter_diag() vs WITNESS

2021-02-16 Thread Visa Hankala
On Tue, Feb 16, 2021 at 10:07:27AM +0100, Martin Pieuchot wrote: > On 15/02/21(Mon) 16:58, Visa Hankala wrote: > > On Mon, Feb 15, 2021 at 11:37:45AM +0100, Martin Pieuchot wrote: > > > Diagnostic function rw_enter_diag() should be called before > > > WITNESS_CHECKORDER() to have proper locking/deb

Re: rw_enter_diag() vs WITNESS

2021-02-16 Thread Martin Pieuchot
On 15/02/21(Mon) 16:58, Visa Hankala wrote: > On Mon, Feb 15, 2021 at 11:37:45AM +0100, Martin Pieuchot wrote: > > Diagnostic function rw_enter_diag() should be called before > > WITNESS_CHECKORDER() to have proper locking/debugging information. > > > > In the case of 'locking against myself' it i

Re: rw_enter_diag() vs WITNESS

2021-02-15 Thread Visa Hankala
On Mon, Feb 15, 2021 at 11:37:45AM +0100, Martin Pieuchot wrote: > Diagnostic function rw_enter_diag() should be called before > WITNESS_CHECKORDER() to have proper locking/debugging information. > > In the case of 'locking against myself' it is currently impossible > to know where the lock has be

rw_enter_diag() vs WITNESS

2021-02-15 Thread Martin Pieuchot
Diagnostic function rw_enter_diag() should be called before WITNESS_CHECKORDER() to have proper locking/debugging information. In the case of 'locking against myself' it is currently impossible to know where the lock has been previously acquired. Diff below fixes that, ok? Index: kern/kern_rwloc