[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20220616164012.uuok46yozncbwe7u@black.fi.intel.com>
Date: Thu, 16 Jun 2022 19:40:12 +0300
From: "Kirill A. Shutemov" <kirill.shutemov@...ux.intel.com>
To: Peter Zijlstra <peterz@...radead.org>
Cc: Dave Hansen <dave.hansen@...ux.intel.com>,
Andy Lutomirski <luto@...nel.org>, x86@...nel.org,
Kostya Serebryany <kcc@...gle.com>,
Andrey Ryabinin <ryabinin.a.a@...il.com>,
Andrey Konovalov <andreyknvl@...il.com>,
Alexander Potapenko <glider@...gle.com>,
Dmitry Vyukov <dvyukov@...gle.com>,
"H . J . Lu" <hjl.tools@...il.com>,
Andi Kleen <ak@...ux.intel.com>,
Rick Edgecombe <rick.p.edgecombe@...el.com>,
linux-mm@...ck.org, linux-kernel@...r.kernel.org, namit@...are.com
Subject: Re: [PATCHv3 4/8] x86/mm: Handle LAM on context switch
On Thu, Jun 16, 2022 at 11:08:07AM +0200, Peter Zijlstra wrote:
> Either use that one spare byte, or find room elsewhere I suppose.
Okay, I will put into the byte after invalidate_other and modify
tlbstate_lam_cr3_mask() and set_tlbstate_lam_cr3_mask() to shift it
accordingly.
It looks like this:
struct tlb_state {
struct mm_struct * loaded_mm; /* 0 8 */
union {
struct mm_struct * last_user_mm; /* 8 8 */
unsigned long last_user_mm_spec; /* 8 8 */
}; /* 8 8 */
union {
struct mm_struct * last_user_mm; /* 0 8 */
unsigned long last_user_mm_spec; /* 0 8 */
};
u16 loaded_mm_asid; /* 16 2 */
u16 next_asid; /* 18 2 */
bool invalidate_other; /* 20 1 */
u8 lam; /* 21 1 */
unsigned short user_pcid_flush_mask; /* 22 2 */
unsigned long cr4; /* 24 8 */
struct tlb_context ctxs[6]; /* 32 96 */
/* size: 128, cachelines: 2, members: 9 */
};
--
Kirill A. Shutemov
Powered by blists - more mailing lists