[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20180723094746.447v2jnegqlosi2f@kshutemo-mobl1>
Date: Mon, 23 Jul 2018 12:47:46 +0300
From: "Kirill A. Shutemov" <kirill@...temov.name>
To: Dave Hansen <dave.hansen@...el.com>
Cc: "Kirill A. Shutemov" <kirill.shutemov@...ux.intel.com>,
Ingo Molnar <mingo@...hat.com>, x86@...nel.org,
Thomas Gleixner <tglx@...utronix.de>,
"H. Peter Anvin" <hpa@...or.com>,
Tom Lendacky <thomas.lendacky@....com>,
Kai Huang <kai.huang@...ux.intel.com>,
Jacob Pan <jacob.jun.pan@...ux.intel.com>,
linux-kernel@...r.kernel.org, linux-mm@...ck.org
Subject: Re: [PATCHv5 11/19] x86/mm: Implement vma_keyid()
On Wed, Jul 18, 2018 at 04:40:14PM -0700, Dave Hansen wrote:
> > --- a/arch/x86/mm/mktme.c
> > +++ b/arch/x86/mm/mktme.c
> > @@ -1,3 +1,4 @@
> > +#include <linux/mm.h>
> > #include <asm/mktme.h>
> >
> > phys_addr_t mktme_keyid_mask;
> > @@ -37,3 +38,14 @@ struct page_ext_operations page_mktme_ops = {
> > .need = need_page_mktme,
> > .init = init_page_mktme,
> > };
> > +
> > +int vma_keyid(struct vm_area_struct *vma)
> > +{
> > + pgprotval_t prot;
> > +
> > + if (!mktme_enabled())
> > + return 0;
> > +
> > + prot = pgprot_val(vma->vm_page_prot);
> > + return (prot & mktme_keyid_mask) >> mktme_keyid_shift;
> > +}
>
> I'm a bit surprised this isn't inlined. Not that function calls are
> expensive, but we *could* entirely avoid them using the normal pattern of:
>
> // In the header:
> static inline vma_keyid(...)
> {
> if (!mktme_enabled())
> return 0;
>
> return __vma_keyid(...); // <- the .c file version
> }
Okay. I'll do this. But it would be a macros. <asm/mktme.h> gets included
very early. We cannot really use jump label code there directly.
--
Kirill A. Shutemov
Powered by blists - more mailing lists