From: Dave Hansen <dave@sr71.net>
To: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org, linux-mm@kvack.org, x86@kernel.org,
dave.hansen@linux.intel.com
Subject: Re: [PATCH 10/34] x86, pkeys: arch-specific protection bitsy
Date: Tue, 8 Dec 2015 08:34:41 -0800 [thread overview]
Message-ID: <566706A1.3040906@sr71.net> (raw)
In-Reply-To: <alpine.DEB.2.11.1512081523180.3595@nanos>
On 12/08/2015 07:15 AM, Thomas Gleixner wrote:
> On Thu, 3 Dec 2015, Dave Hansen wrote:
>>
>> +static inline int vma_pkey(struct vm_area_struct *vma)
>
> Shouldn't this return something unsigned?
Ingo had asked that we use 'int' in the syscalls at some point. We also
use a -1 to mean "no pkey set" (to differentiate it from pkey=0) at
least at the very top of the syscall level.
>> +{
>> + u16 pkey = 0;
>> +#ifdef CONFIG_X86_INTEL_MEMORY_PROTECTION_KEYS
>> + unsigned long vma_pkey_mask = VM_PKEY_BIT0 | VM_PKEY_BIT1 |
>> + VM_PKEY_BIT2 | VM_PKEY_BIT3;
>> + /*
>> + * ffs is one-based, not zero-based, so bias back down by 1.
>> + */
>> + int vm_pkey_shift = __builtin_ffsl(vma_pkey_mask) - 1;
>
> Took me some time to figure out that this will resolve to a compile
> time constant (hopefully). Is there a reason why we don't have a
> VM_PKEY_SHIFT constant in the header file which makes that code just
> simple and intuitive?
All of the VM_* flags are #defined as bitmaps directly and don't define
shifts:
#define VM_MAYWRITE 0x00000020
#define VM_MAYEXEC 0x00000040
#define VM_MAYSHARE 0x00000080
...
So to get a shift we've either got to do a ffs somewhere, or we have to
define the VM_PKEY_BIT*'s differently from all of the other VM_* flags.
Or, we do something along the lines of:
#define VM_PKEY_BIT0 0x100000000UL
#define __VM_PKEY_SHIFT (32)
and we run a small risk that somebody will desynchronize the shift and
the bit definition.
We only need this shift in this *one* place, so that's why I opted for
the local variable and ffs.
>> + /*
>> + * gcc generates better code if we do this rather than:
>> + * pkey = (flags & mask) >> shift
>> + */
>> + pkey = (vma->vm_flags >> vm_pkey_shift) &
>> + (vma_pkey_mask >> vm_pkey_shift);
>
> My gcc (4.9) does it the other way round for whatever reason.
I'll go recheck.
WARNING: multiple messages have this Message-ID (diff)
From: Dave Hansen <dave@sr71.net>
To: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org, linux-mm@kvack.org, x86@kernel.org,
dave.hansen@linux.intel.com
Subject: Re: [PATCH 10/34] x86, pkeys: arch-specific protection bitsy
Date: Tue, 8 Dec 2015 08:34:41 -0800 [thread overview]
Message-ID: <566706A1.3040906@sr71.net> (raw)
In-Reply-To: <alpine.DEB.2.11.1512081523180.3595@nanos>
On 12/08/2015 07:15 AM, Thomas Gleixner wrote:
> On Thu, 3 Dec 2015, Dave Hansen wrote:
>>
>> +static inline int vma_pkey(struct vm_area_struct *vma)
>
> Shouldn't this return something unsigned?
Ingo had asked that we use 'int' in the syscalls at some point. We also
use a -1 to mean "no pkey set" (to differentiate it from pkey=0) at
least at the very top of the syscall level.
>> +{
>> + u16 pkey = 0;
>> +#ifdef CONFIG_X86_INTEL_MEMORY_PROTECTION_KEYS
>> + unsigned long vma_pkey_mask = VM_PKEY_BIT0 | VM_PKEY_BIT1 |
>> + VM_PKEY_BIT2 | VM_PKEY_BIT3;
>> + /*
>> + * ffs is one-based, not zero-based, so bias back down by 1.
>> + */
>> + int vm_pkey_shift = __builtin_ffsl(vma_pkey_mask) - 1;
>
> Took me some time to figure out that this will resolve to a compile
> time constant (hopefully). Is there a reason why we don't have a
> VM_PKEY_SHIFT constant in the header file which makes that code just
> simple and intuitive?
All of the VM_* flags are #defined as bitmaps directly and don't define
shifts:
#define VM_MAYWRITE 0x00000020
#define VM_MAYEXEC 0x00000040
#define VM_MAYSHARE 0x00000080
...
So to get a shift we've either got to do a ffs somewhere, or we have to
define the VM_PKEY_BIT*'s differently from all of the other VM_* flags.
Or, we do something along the lines of:
#define VM_PKEY_BIT0 0x100000000UL
#define __VM_PKEY_SHIFT (32)
and we run a small risk that somebody will desynchronize the shift and
the bit definition.
We only need this shift in this *one* place, so that's why I opted for
the local variable and ffs.
>> + /*
>> + * gcc generates better code if we do this rather than:
>> + * pkey = (flags & mask) >> shift
>> + */
>> + pkey = (vma->vm_flags >> vm_pkey_shift) &
>> + (vma_pkey_mask >> vm_pkey_shift);
>
> My gcc (4.9) does it the other way round for whatever reason.
I'll go recheck.
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
next prev parent reply other threads:[~2015-12-08 16:34 UTC|newest]
Thread overview: 145+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-12-04 1:14 [PATCH 00/34] x86: Memory Protection Keys (v5) Dave Hansen
2015-12-04 1:14 ` Dave Hansen
2015-12-04 1:14 ` Dave Hansen
2015-12-04 1:14 ` [PATCH 01/34] mm, gup: introduce concept of "foreign" get_user_pages() Dave Hansen
2015-12-04 1:14 ` Dave Hansen
2015-12-04 1:14 ` [PATCH 02/34] x86, fpu: add placeholder for Processor Trace XSAVE state Dave Hansen
2015-12-04 1:14 ` Dave Hansen
2015-12-04 1:14 ` [PATCH 03/34] x86, pkeys: Add Kconfig option Dave Hansen
2015-12-04 1:14 ` Dave Hansen
2015-12-04 1:14 ` [PATCH 04/34] x86, pkeys: cpuid bit definition Dave Hansen
2015-12-04 1:14 ` Dave Hansen
2015-12-04 1:14 ` [PATCH 05/34] x86, pkeys: define new CR4 bit Dave Hansen
2015-12-04 1:14 ` Dave Hansen
2015-12-04 1:14 ` [PATCH 06/34] x86, pkeys: add PKRU xsave fields and data structure(s) Dave Hansen
2015-12-04 1:14 ` Dave Hansen
2015-12-04 1:14 ` [PATCH 07/34] x86, pkeys: PTE bits for storing protection key Dave Hansen
2015-12-04 1:14 ` Dave Hansen
2015-12-04 1:14 ` [PATCH 08/34] x86, pkeys: new page fault error code bit: PF_PK Dave Hansen
2015-12-04 1:14 ` Dave Hansen
2015-12-04 1:14 ` [PATCH 09/34] x86, pkeys: store protection in high VMA flags Dave Hansen
2015-12-04 1:14 ` Dave Hansen
2015-12-08 14:17 ` Thomas Gleixner
2015-12-08 14:17 ` Thomas Gleixner
2015-12-04 1:14 ` [PATCH 10/34] x86, pkeys: arch-specific protection bits Dave Hansen
2015-12-04 1:14 ` Dave Hansen
2015-12-08 15:15 ` [PATCH 10/34] x86, pkeys: arch-specific protection bitsy Thomas Gleixner
2015-12-08 15:15 ` Thomas Gleixner
2015-12-08 16:34 ` Dave Hansen [this message]
2015-12-08 16:34 ` Dave Hansen
2015-12-08 17:24 ` Thomas Gleixner
2015-12-08 17:24 ` Thomas Gleixner
2015-12-08 18:06 ` Dave Hansen
2015-12-08 18:29 ` Thomas Gleixner
2015-12-08 18:29 ` Thomas Gleixner
2015-12-08 18:35 ` Thomas Gleixner
2015-12-08 18:35 ` Thomas Gleixner
2015-12-04 1:14 ` [PATCH 11/34] x86, pkeys: pass VMA down in to fault signal generation code Dave Hansen
2015-12-04 1:14 ` Dave Hansen
2015-12-04 1:14 ` [PATCH 12/34] signals, pkeys: notify userspace about protection key faults Dave Hansen
2015-12-04 1:14 ` Dave Hansen
2015-12-04 1:14 ` [PATCH 13/34] x86, pkeys: fill in pkey field in siginfo Dave Hansen
2015-12-04 1:14 ` Dave Hansen
2015-12-04 1:14 ` [PATCH 14/34] x86, pkeys: add functions to fetch PKRU Dave Hansen
2015-12-04 1:14 ` Dave Hansen
2015-12-08 15:18 ` Thomas Gleixner
2015-12-08 15:18 ` Thomas Gleixner
2015-12-04 1:14 ` [PATCH 15/34] mm: factor out VMA fault permission checking Dave Hansen
2015-12-04 1:14 ` Dave Hansen
2015-12-08 17:26 ` Thomas Gleixner
2015-12-08 17:26 ` Thomas Gleixner
2015-12-04 1:14 ` [PATCH 16/34] x86, mm: simplify get_user_pages() PTE bit handling Dave Hansen
2015-12-04 1:14 ` Dave Hansen
2015-12-08 18:01 ` Thomas Gleixner
2015-12-08 18:01 ` Thomas Gleixner
2015-12-08 18:30 ` Dave Hansen
2015-12-08 18:30 ` Dave Hansen
2015-12-04 1:14 ` [PATCH 17/34] x86, pkeys: check VMAs and PTEs for protection keys Dave Hansen
2015-12-04 1:14 ` Dave Hansen
2015-12-08 18:11 ` Thomas Gleixner
2015-12-08 18:11 ` Thomas Gleixner
2015-12-04 1:14 ` [PATCH 18/34] mm: add gup flag to indicate "foreign" mm access Dave Hansen
2015-12-04 1:14 ` Dave Hansen
2015-12-04 1:14 ` [PATCH 19/34] x86, pkeys: optimize fault handling in access_error() Dave Hansen
2015-12-04 1:14 ` Dave Hansen
2015-12-08 18:14 ` Thomas Gleixner
2015-12-08 18:14 ` Thomas Gleixner
2015-12-04 1:14 ` [PATCH 20/34] x86, pkeys: differentiate instruction fetches Dave Hansen
2015-12-04 1:14 ` Dave Hansen
2015-12-08 18:17 ` Thomas Gleixner
2015-12-08 18:17 ` Thomas Gleixner
2015-12-04 1:14 ` [PATCH 21/34] x86, pkeys: dump PKRU with other kernel registers Dave Hansen
2015-12-04 1:14 ` Dave Hansen
2015-12-08 18:19 ` Thomas Gleixner
2015-12-08 18:19 ` Thomas Gleixner
2015-12-04 1:14 ` [PATCH 22/34] x86, pkeys: dump PTE pkey in /proc/pid/smaps Dave Hansen
2015-12-04 1:14 ` Dave Hansen
2015-12-08 18:20 ` Thomas Gleixner
2015-12-08 18:20 ` Thomas Gleixner
2015-12-04 1:14 ` [PATCH 23/34] x86, pkeys: add Kconfig prompt to existing config option Dave Hansen
2015-12-04 1:14 ` Dave Hansen
2015-12-08 18:21 ` Thomas Gleixner
2015-12-08 18:21 ` Thomas Gleixner
2015-12-04 1:14 ` [PATCH 24/34] mm, multi-arch: pass a protection key in to calc_vm_flag_bits() Dave Hansen
2015-12-04 1:14 ` Dave Hansen
2015-12-04 1:14 ` [PATCH 25/34] x86, pkeys: add arch_validate_pkey() Dave Hansen
2015-12-04 1:14 ` Dave Hansen
2015-12-08 18:39 ` Thomas Gleixner
2015-12-08 18:39 ` Thomas Gleixner
2015-12-04 1:15 ` [PATCH 26/34] mm: implement new mprotect_key() system call Dave Hansen
2015-12-04 1:15 ` Dave Hansen
2015-12-05 6:50 ` Michael Kerrisk (man-pages)
2015-12-05 6:50 ` Michael Kerrisk (man-pages)
2015-12-05 6:50 ` Michael Kerrisk (man-pages)
2015-12-07 16:44 ` Dave Hansen
2015-12-07 16:44 ` Dave Hansen
2015-12-09 11:08 ` Michael Kerrisk (man-pages)
2015-12-09 11:08 ` Michael Kerrisk (man-pages)
2015-12-09 15:48 ` Dave Hansen
2015-12-09 15:48 ` Dave Hansen
2015-12-09 16:45 ` Michael Kerrisk (man-pages)
2015-12-09 16:45 ` Michael Kerrisk (man-pages)
2015-12-09 16:45 ` Michael Kerrisk (man-pages)
2015-12-09 17:05 ` Dave Hansen
2015-12-09 17:05 ` Dave Hansen
2015-12-09 17:05 ` Dave Hansen
2015-12-11 20:13 ` Michael Kerrisk (man-pages)
2015-12-11 20:13 ` Michael Kerrisk (man-pages)
2015-12-04 1:15 ` [PATCH 27/34] x86, pkeys: make mprotect_key() mask off additional vm_flags Dave Hansen
2015-12-04 1:15 ` Dave Hansen
2015-12-08 18:41 ` Thomas Gleixner
2015-12-08 18:41 ` Thomas Gleixner
2015-12-04 1:15 ` [PATCH 28/34] x86: wire up mprotect_key() system call Dave Hansen
2015-12-04 1:15 ` Dave Hansen
2015-12-08 18:44 ` Thomas Gleixner
2015-12-08 18:44 ` Thomas Gleixner
2015-12-08 18:44 ` Thomas Gleixner
2015-12-08 19:06 ` Dave Hansen
2015-12-08 19:06 ` Dave Hansen
2015-12-08 20:38 ` Thomas Gleixner
2015-12-08 20:38 ` Thomas Gleixner
2015-12-08 20:38 ` Thomas Gleixner
2015-12-04 1:15 ` [PATCH 29/34] x86: separate out LDT init from context init Dave Hansen
2015-12-04 1:15 ` Dave Hansen
2015-12-08 18:45 ` Thomas Gleixner
2015-12-08 18:45 ` Thomas Gleixner
2015-12-04 1:15 ` [PATCH 30/34] x86, fpu: allow setting of XSAVE state Dave Hansen
2015-12-04 1:15 ` Dave Hansen
2015-12-08 18:48 ` Thomas Gleixner
2015-12-08 18:48 ` Thomas Gleixner
2015-12-04 1:15 ` [PATCH 31/34] x86, pkeys: allocation/free syscalls Dave Hansen
2015-12-04 1:15 ` Dave Hansen
2015-12-04 1:15 ` [PATCH 32/34] x86, pkeys: add pkey set/get syscalls Dave Hansen
2015-12-04 1:15 ` Dave Hansen
2015-12-04 1:15 ` [PATCH 33/34] x86, pkeys: actually enable Memory Protection Keys in CPU Dave Hansen
2015-12-04 1:15 ` Dave Hansen
2015-12-04 1:15 ` [PATCH 34/34] x86, pkeys: Documentation Dave Hansen
2015-12-04 1:15 ` Dave Hansen
2015-12-04 23:31 ` [PATCH 00/34] x86: Memory Protection Keys (v5) Andy Lutomirski
2015-12-04 23:31 ` Andy Lutomirski
2015-12-04 23:38 ` Dave Hansen
2015-12-04 23:38 ` Dave Hansen
2015-12-04 23:38 ` Dave Hansen
2015-12-11 20:16 ` Andy Lutomirski
2015-12-11 20:16 ` Andy Lutomirski
2015-12-11 20:16 ` Andy Lutomirski
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=566706A1.3040906@sr71.net \
--to=dave@sr71.net \
--cc=dave.hansen@linux.intel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=tglx@linutronix.de \
--cc=x86@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.