Re: vm_page_t related KBI [Was: Re: panic at vm_page_wire with FreeBSD 9.0 Beta 3]

On Mon, Nov 07, 2011 at 11:47:59AM -0800, mdf_at_FreeBSD.org wrote:
> On Mon, Nov 7, 2011 at 11:35 AM, Kostik Belousov <kostikbel_at_gmail.com> wrote:
> > On Mon, Nov 07, 2011 at 11:45:38AM -0600, Alan Cox wrote:
> >> Ok.  I'll offer one final suggestion.  Please consider an alternative
> >> suffix to "func".  Perhaps, "kbi" or "KBI".  In other words, something
> >> that hints at the function's reason for existing.
> >
> > Sure. Below is the extraction of only vm_page_lock() bits, together
> > with the suggested rename. When Attilio provides the promised simplification
> > of the mutex KPI, this can be reduced.
> >
> > diff --git a/sys/vm/vm_page.c b/sys/vm/vm_page.c
> > index 389aea5..ea4ea34 100644
> > --- a/sys/vm/vm_page.c
> > +++ b/sys/vm/vm_page.c
> > _at__at_ -2677,6 +2677,44 _at__at_ vm_page_test_dirty(vm_page_t m)
> >                vm_page_dirty(m);
> >  }
> >
> > +void
> > +vm_page_lock_KBI(vm_page_t m, const char *file, int line)
> > +{
> > +
> > +#if LOCK_DEBUG > 0 || defined(MUTEX_NOINLINE)
> > +       _mtx_lock_flags(vm_page_lockptr(m), 0, file, line);
> > +#else
> > +       __mtx_lock(vm_page_lockptr(m), 0, file, line);
> > +#endif
> > +}
> > +
> > +void
> > +vm_page_unlock_KBI(vm_page_t m, const char *file, int line)
> > +{
> > +
> > +#if LOCK_DEBUG > 0 || defined(MUTEX_NOINLINE)
> > +       _mtx_unlock_flags(vm_page_lockptr(m), 0, file, line);
> > +#else
> > +       __mtx_unlock(vm_page_lockptr(m), curthread, 0, file, line);
> > +#endif
> > +}
> > +
> > +int
> > +vm_page_trylock_KBI(vm_page_t m, const char *file, int line)
> > +{
> > +
> > +       return (_mtx_trylock(vm_page_lockptr(m), 0, file, line));
> > +}
> > +
> > +void
> > +vm_page_lock_assert_KBI(vm_page_t m, int a, const char *file, int line)
> > +{
> > +
> > +#ifdef INVARIANTS
> > +       _mtx_assert(vm_page_lockptr(m), a, file, line);
> > +#endif
> > +}
> > +
> >  int so_zerocp_fullpage = 0;
> >
> >  /*
> > diff --git a/sys/vm/vm_page.h b/sys/vm/vm_page.h
> > index 7099b70..a5604b7 100644
> > --- a/sys/vm/vm_page.h
> > +++ b/sys/vm/vm_page.h
> > _at__at_ -218,11 +218,23 _at__at_ extern struct vpglocks pa_lock[];
> >
> >  #define        PA_LOCK_ASSERT(pa, a)   mtx_assert(PA_LOCKPTR(pa), (a))
> >
> > +#ifdef KLD_MODULE
> > +#define        vm_page_lock(m)         vm_page_lock_KBI((m), LOCK_FILE, LOCK_LINE)
> > +#define        vm_page_unlock(m)       vm_page_unlock_KBI((m), LOCK_FILE, LOCK_LINE)
> > +#define        vm_page_trylock(m)      vm_page_trylock_KBI((m), LOCK_FILE, LOCK_LINE)
> > +#ifdef INVARIANTS
> > +#define        vm_page_lock_assert(m, a)       \
> > +    vm_page_lock_assert_KBI((m), (a), LOCK_FILE, LOCK_LINE)
> > +#else
> > +#define        vm_page_lock_assert(m, a)
> > +#endif
> > +#else  /* KLD_MODULE */
> >  #define        vm_page_lockptr(m)      (PA_LOCKPTR(VM_PAGE_TO_PHYS((m))))
> 
> Is it not possible to have vm_page_lockptr() be a function for the
> KLD_MODULE case?  Because then the vm_page_lock() functions and
> friends would all just use mtx_lock, etc., in both the KLD and non-KLD
> case.
> 
> Or am I missing something?
It is possible, but I tried to avoid exposing lockptr.
IMHO vm_page_lockptr() is an implementation detail.

Please also see my other response to Alan regarding the relocking macros.