Re: [PATCH v2 sl-b 1/5] mm: Add mem_dump_obj() to print source of memory block

2020-12-10 Thread Paul E. McKenney
On Thu, Dec 10, 2020 at 09:04:11PM +0900, Joonsoo Kim wrote: > On Tue, Dec 08, 2020 at 05:12:59PM -0800, paul...@kernel.org wrote: > > From: "Paul E. McKenney" > > > > There are kernel facilities such as per-CPU reference counts that give > > error messages in generic handlers or callbacks, whose

Re: [PATCH v2 sl-b 1/5] mm: Add mem_dump_obj() to print source of memory block

2020-12-10 Thread Paul E. McKenney
On Thu, Dec 10, 2020 at 11:48:26AM +0100, Vlastimil Babka wrote: > On 12/10/20 12:04 AM, Paul E. McKenney wrote: > >> > +/** > >> > + * kmem_valid_obj - does the pointer reference a valid slab object? > >> > + * @object: pointer to query. > >> > + * > >> > + * Return: %true if the pointer is to a n

Re: [PATCH v2 sl-b 1/5] mm: Add mem_dump_obj() to print source of memory block

2020-12-10 Thread Joonsoo Kim
On Tue, Dec 08, 2020 at 05:12:59PM -0800, paul...@kernel.org wrote: > From: "Paul E. McKenney" > > There are kernel facilities such as per-CPU reference counts that give > error messages in generic handlers or callbacks, whose messages are > unenlightening. In the case of per-CPU reference-count

Re: [PATCH v2 sl-b 1/5] mm: Add mem_dump_obj() to print source of memory block

2020-12-10 Thread Vlastimil Babka
On 12/10/20 12:04 AM, Paul E. McKenney wrote: >> > +/** >> > + * kmem_valid_obj - does the pointer reference a valid slab object? >> > + * @object: pointer to query. >> > + * >> > + * Return: %true if the pointer is to a not-yet-freed object from >> > + * kmalloc() or kmem_cache_alloc(), either %tr

Re: [PATCH v2 sl-b 1/5] mm: Add mem_dump_obj() to print source of memory block

2020-12-09 Thread Paul E. McKenney
On Wed, Dec 09, 2020 at 06:28:50PM +0100, Vlastimil Babka wrote: > On 12/9/20 2:12 AM, paul...@kernel.org wrote: > > From: "Paul E. McKenney" > > > > There are kernel facilities such as per-CPU reference counts that give > > error messages in generic handlers or callbacks, whose messages are > >

Re: [PATCH v2 sl-b 1/5] mm: Add mem_dump_obj() to print source of memory block

2020-12-09 Thread Paul E. McKenney
On Wed, Dec 09, 2020 at 05:53:06PM +, Christoph Hellwig wrote: > On Wed, Dec 09, 2020 at 06:57:02AM -0800, Paul E. McKenney wrote: > > On Wed, Dec 09, 2020 at 08:17:10AM +, Christoph Hellwig wrote: > > > Your two new exports don't actually seem to get used in modular code > > > at all in th

Re: [PATCH v2 sl-b 1/5] mm: Add mem_dump_obj() to print source of memory block

2020-12-09 Thread Christoph Hellwig
On Wed, Dec 09, 2020 at 06:57:02AM -0800, Paul E. McKenney wrote: > On Wed, Dec 09, 2020 at 08:17:10AM +, Christoph Hellwig wrote: > > Your two new exports don't actually seem to get used in modular code > > at all in this series. > > Indeed, and I either need to remove the exports or make my

Re: [PATCH v2 sl-b 1/5] mm: Add mem_dump_obj() to print source of memory block

2020-12-09 Thread Vlastimil Babka
On 12/9/20 2:12 AM, paul...@kernel.org wrote: > From: "Paul E. McKenney" > > There are kernel facilities such as per-CPU reference counts that give > error messages in generic handlers or callbacks, whose messages are > unenlightening. In the case of per-CPU reference-count underflow, this > is

Re: [PATCH v2 sl-b 1/5] mm: Add mem_dump_obj() to print source of memory block

2020-12-09 Thread Paul E. McKenney
On Wed, Dec 09, 2020 at 08:17:10AM +, Christoph Hellwig wrote: > Your two new exports don't actually seem to get used in modular code > at all in this series. Indeed, and I either need to remove the exports or make my test code in kernel/rcu/rcuscale.o suitable for upstreaming. Or find the ap

Re: [PATCH v2 sl-b 1/5] mm: Add mem_dump_obj() to print source of memory block

2020-12-09 Thread Christoph Hellwig
Your two new exports don't actually seem to get used in modular code at all in this series.

[PATCH v2 sl-b 1/5] mm: Add mem_dump_obj() to print source of memory block

2020-12-08 Thread paulmck
From: "Paul E. McKenney" There are kernel facilities such as per-CPU reference counts that give error messages in generic handlers or callbacks, whose messages are unenlightening. In the case of per-CPU reference-count underflow, this is not a problem when creating a new use of this facility bec