linux-kernel - Re: [PATCH 1/7] mm/hugetlb: add folio support to hugetlb specific flag macros

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Thu, 1 Sep 2022 09:55:21 -0700
From:   Mike Kravetz <mike.kravetz@...cle.com>
To:     Sidhartha Kumar <sidhartha.kumar@...cle.com>
Cc:     Matthew Wilcox <willy@...radead.org>, linux-kernel@...r.kernel.org,
        linux-mm@...ck.org, akpm@...ux-foundation.org,
        songmuchun@...edance.com, vbabka@...e.cz,
        william.kucharski@...cle.com, dhowells@...hat.com,
        peterx@...hat.com, arnd@...db.de, ccross@...gle.com,
        hughd@...gle.com, ebiederm@...ssion.com
Subject: Re: [PATCH 1/7] mm/hugetlb: add folio support to hugetlb specific
 flag macros

On 08/30/22 11:09, Sidhartha Kumar wrote:
> 
> 
> On 8/29/22 8:33 PM, Matthew Wilcox wrote:
> > On Mon, Aug 29, 2022 at 04:00:08PM -0700, Sidhartha Kumar wrote:
> > >   #define TESTHPAGEFLAG(uname, flname)				\
> > > +static __always_inline								\
> > > +int folio_test_hugetlb_##flname(struct folio *folio)		\
> > One change I made was to have folio_test_foo() return bool instead of
> > int.  It helps the compiler really understand what's going on.  Maybe
> > some humans too ;-)
> > 
> 
> I went with returning an int to stay consistent with the page version
> of the macros which return an int. I'm fine with changing it to return
> a bool.

I believe the page test macros returned an int when I added the hugetlb
specific versions.  So, I just did the same.  Since they are now bool,
it makes sense to have these be consistent.
-- 
Mike Kravetz

> 
> > > +	{	void **private = &folio->private;		\
> > > +		return test_bit(HPG_##flname, (void *)((unsigned long)private));	\
> > I've made this tricky for you by making folio->private a void * instead
> > of the unsigned long in page.  Would this look better as ...
> > 
> > 	{							\
> > 		void *private = &folio->private;		\
> > 		return test_bit(HPG_##flname, private);		\
> > 
> > perhaps?
> 
> Ya this looks much better and passes the tests, will add to v2.
> 
> 
>