linux-kernel - Re: [RFC PATCH 2/4] mm/mmu_notifier: use unsigned for event field in range struct

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:   Fri, 1 Feb 2019 20:13:40 -0500
From:   Andrea Arcangeli <aarcange@...hat.com>
To:     jglisse@...hat.com
Cc:     linux-mm@...ck.org, linux-kernel@...r.kernel.org,
        Peter Xu <peterx@...hat.com>,
        Andrew Morton <akpm@...ux-foundation.org>,
        Paolo Bonzini <pbonzini@...hat.com>,
        Radim Krčmář <rkrcmar@...hat.com>,
        kvm@...r.kernel.org
Subject: Re: [RFC PATCH 2/4] mm/mmu_notifier: use unsigned for event field in
 range struct

On Thu, Jan 31, 2019 at 01:37:04PM -0500, Jerome Glisse wrote:
> From: Jérôme Glisse <jglisse@...hat.com>
> 
> Use unsigned for event field in range struct so that we can also set
> flags with the event. This patch change the field and introduce the
> helper.
> 
> Signed-off-by: Jérôme Glisse <jglisse@...hat.com>
> Cc: Andrea Arcangeli <aarcange@...hat.com>
> Cc: Peter Xu <peterx@...hat.com>
> Cc: Andrew Morton <akpm@...ux-foundation.org>
> Cc: Paolo Bonzini <pbonzini@...hat.com>
> Cc: Radim Krčmář <rkrcmar@...hat.com>
> Cc: kvm@...r.kernel.org
> ---
>  include/linux/mmu_notifier.h | 8 ++++++--
>  1 file changed, 6 insertions(+), 2 deletions(-)
> 
> diff --git a/include/linux/mmu_notifier.h b/include/linux/mmu_notifier.h
> index be873c431886..d7a35975c2bd 100644
> --- a/include/linux/mmu_notifier.h
> +++ b/include/linux/mmu_notifier.h
> @@ -6,6 +6,7 @@
>  #include <linux/spinlock.h>
>  #include <linux/mm_types.h>
>  #include <linux/srcu.h>
> +#include <linux/log2.h>
>  
>  struct mmu_notifier;
>  struct mmu_notifier_ops;
> @@ -38,8 +39,11 @@ enum mmu_notifier_event {
>  	MMU_NOTIFY_PROTECTION_VMA,
>  	MMU_NOTIFY_PROTECTION_PAGE,
>  	MMU_NOTIFY_SOFT_DIRTY,
> +	MMU_NOTIFY_EVENT_MAX
>  };
>  
> +#define MMU_NOTIFIER_EVENT_BITS order_base_2(MMU_NOTIFY_EVENT_MAX)
> +
>  #ifdef CONFIG_MMU_NOTIFIER
>  
>  /*
> @@ -60,7 +64,7 @@ struct mmu_notifier_range {
>  	struct mm_struct *mm;
>  	unsigned long start;
>  	unsigned long end;
> -	enum mmu_notifier_event event;
> +	unsigned event;
>  	bool blockable;
>  };

This is only allocated in the stack, so saving RAM by mixing bitfields
with enum in the same 4 bytes to save 4 bytes isn't of maximum
priority.

A possibly cleaner way to save those 4 bytes without mixing enum with
bitfields by hand, is to add a "unsigned short flags" which will make
"event/flags/blockable" fit in the same 8 bytes (bool only needs 1
byte) as before the patch (the first bitfield can start from 0 then).

Yet another way is to drop blockable and convert it to a flag in
"unsigned int flags".