On Wed, 2013-08-21 at 09:30 -0700, Linus Torvalds wrote:
> I will be reverting the whole soft-dirty mess. I thought the
> bit-mapping games it played were already too complicated (the patch to
> pgtable-2level.h in commit 41bb3476b361 just makes me want to barf and
> came in very late, so I'm not
On Wed, 2013-08-21 at 09:30 -0700, Linus Torvalds wrote:
I will be reverting the whole soft-dirty mess. I thought the
bit-mapping games it played were already too complicated (the patch to
pgtable-2level.h in commit 41bb3476b361 just makes me want to barf and
came in very late, so I'm not
On Thu, Aug 22, 2013 at 03:37:33PM +0900, Minchan Kim wrote:
> >
> > Look, good thing is that 7th bit also available on the 4level pages
> > (ie x86-64) without additional code modification, that's why I picked
> > it in first place. I prepared the patch locally which doesn't use
> > pse bit for
On 08/22/2013 03:33 PM, Jan Beulich wrote:
>> From: Cyrill Gorcunov
>> Subject: [PATCH] mm: Make sure _PAGE_SWP_SOFT_DIRTY bit is not set on
>> present pte
>>
>> _PAGE_SOFT_DIRTY bit should never be set on present pte so add
>> VM_BUG_ON to catch any potential future abuse.
>>
>> Also add a
>>> On 22.08.13 at 13:27, Cyrill Gorcunov wrote:
> On Thu, Aug 22, 2013 at 08:27:45AM +0100, Jan Beulich wrote:
>> >>> On 22.08.13 at 09:03, Cyrill Gorcunov wrote:
>> > Ok, how about this?
>> >
>> > static inline pte_t pte_swp_mksoft_dirty(pte_t pte)
>> > {
>> >BUG_ON(pte_present(pte));
>>
On Thu, Aug 22, 2013 at 08:27:45AM +0100, Jan Beulich wrote:
> >>> On 22.08.13 at 09:03, Cyrill Gorcunov wrote:
> > Ok, how about this?
> >
> > static inline pte_t pte_swp_mksoft_dirty(pte_t pte)
> > {
> > BUG_ON(pte_present(pte));
> > return pte_set_flags(pte, _PAGE_SWP_SOFT_DIRTY);
> >
On 08/22/2013 01:32 PM, David Vrabel wrote:
> On 22/08/13 00:04, Linus Torvalds wrote:
>> On Wed, Aug 21, 2013 at 12:03 PM, Cyrill Gorcunov wrote:
>>>
>>> I personally don't see bug here because
>>>
>>> - this swapped page soft dirty bit is set for non-present entries only,
>>>never for
On 22/08/13 00:04, Linus Torvalds wrote:
> On Wed, Aug 21, 2013 at 12:03 PM, Cyrill Gorcunov wrote:
>>
>> I personally don't see bug here because
>>
>> - this swapped page soft dirty bit is set for non-present entries only,
>>never for present ones, just at moment we form swap pte entry
>>
>>> On 21.08.13 at 19:28, Andy Lutomirski wrote:
> On Wed, Aug 21, 2013 at 6:48 AM, David Vrabel wrote:
>> All,
>>
>> 179ef71c (mm: save soft-dirty bits on swapped pages) introduces a new
>> PTE bit on x86 _PTE_SWP_SOFT_DIRTY which has the same value as _PTE_PSE
>> and _PTE_PAT.
>>
>> With a Xen
>>> On 22.08.13 at 01:04, Linus Torvalds wrote:
> On Wed, Aug 21, 2013 at 12:03 PM, Cyrill Gorcunov wrote:
>>
>> I personally don't see bug here because
>>
>> - this swapped page soft dirty bit is set for non-present entries only,
>>never for present ones, just at moment we form swap pte
>>> On 22.08.13 at 09:03, Cyrill Gorcunov wrote:
> Ok, how about this?
>
> static inline pte_t pte_swp_mksoft_dirty(pte_t pte)
> {
> BUG_ON(pte_present(pte));
> return pte_set_flags(pte, _PAGE_SWP_SOFT_DIRTY);
> }
Sure, fine with me. Perhaps VM_BUG_ON() or some other similar
On Thu, Aug 22, 2013 at 07:56:26AM +0100, Jan Beulich wrote:
> >>> On 21.08.13 at 18:19, Cyrill Gorcunov wrote:
> > On Wed, Aug 21, 2013 at 05:03:13PM +0100, Jan Beulich wrote:
> >> >
> >> > Only to non-present ptes, as far as I know.
> >>
> >> That's not really any guarantee. And the accessor
>>> On 21.08.13 at 18:19, Cyrill Gorcunov wrote:
> On Wed, Aug 21, 2013 at 05:03:13PM +0100, Jan Beulich wrote:
>> >
>> > Only to non-present ptes, as far as I know.
>>
>> That's not really any guarantee. And the accessor functions also
>> don't check that they'd be used on non-present PTEs
On 08/22/2013 04:51 AM, Dave Jones wrote:
> On Wed, Aug 21, 2013 at 04:04:54PM -0700, Linus Torvalds wrote:
> > On Wed, Aug 21, 2013 at 12:03 PM, Cyrill Gorcunov
> wrote:
> > >
> > > I personally don't see bug here because
> > >
> > > - this swapped page soft dirty bit is set for
Hello Cyrill,
On Thu, Aug 22, 2013 at 09:49:19AM +0400, Cyrill Gorcunov wrote:
> On Wed, Aug 21, 2013 at 04:42:53PM -0700, Andi Kleen wrote:
> > Cyrill Gorcunov writes:
> > >
> > > Hi all, I worked on patch which would not touch PSE bit for dirty page
> > > tracking and the result is not that
Hello Cyrill,
On Thu, Aug 22, 2013 at 09:49:19AM +0400, Cyrill Gorcunov wrote:
On Wed, Aug 21, 2013 at 04:42:53PM -0700, Andi Kleen wrote:
Cyrill Gorcunov gorcu...@gmail.com writes:
Hi all, I worked on patch which would not touch PSE bit for dirty page
tracking and the result is not
On 08/22/2013 04:51 AM, Dave Jones wrote:
On Wed, Aug 21, 2013 at 04:04:54PM -0700, Linus Torvalds wrote:
On Wed, Aug 21, 2013 at 12:03 PM, Cyrill Gorcunov gorcu...@gmail.com
wrote:
I personally don't see bug here because
- this swapped page soft dirty bit is set for
On 21.08.13 at 18:19, Cyrill Gorcunov gorcu...@gmail.com wrote:
On Wed, Aug 21, 2013 at 05:03:13PM +0100, Jan Beulich wrote:
Only to non-present ptes, as far as I know.
That's not really any guarantee. And the accessor functions also
don't check that they'd be used on non-present PTEs
On Thu, Aug 22, 2013 at 07:56:26AM +0100, Jan Beulich wrote:
On 21.08.13 at 18:19, Cyrill Gorcunov gorcu...@gmail.com wrote:
On Wed, Aug 21, 2013 at 05:03:13PM +0100, Jan Beulich wrote:
Only to non-present ptes, as far as I know.
That's not really any guarantee. And the accessor
On 22.08.13 at 09:03, Cyrill Gorcunov gorcu...@gmail.com wrote:
Ok, how about this?
static inline pte_t pte_swp_mksoft_dirty(pte_t pte)
{
BUG_ON(pte_present(pte));
return pte_set_flags(pte, _PAGE_SWP_SOFT_DIRTY);
}
Sure, fine with me. Perhaps VM_BUG_ON() or some other similar
On 22.08.13 at 01:04, Linus Torvalds torva...@linux-foundation.org wrote:
On Wed, Aug 21, 2013 at 12:03 PM, Cyrill Gorcunov gorcu...@gmail.com wrote:
I personally don't see bug here because
- this swapped page soft dirty bit is set for non-present entries only,
never for present ones,
On 21.08.13 at 19:28, Andy Lutomirski l...@amacapital.net wrote:
On Wed, Aug 21, 2013 at 6:48 AM, David Vrabel david.vra...@citrix.com wrote:
All,
179ef71c (mm: save soft-dirty bits on swapped pages) introduces a new
PTE bit on x86 _PTE_SWP_SOFT_DIRTY which has the same value as _PTE_PSE
On 22/08/13 00:04, Linus Torvalds wrote:
On Wed, Aug 21, 2013 at 12:03 PM, Cyrill Gorcunov gorcu...@gmail.com wrote:
I personally don't see bug here because
- this swapped page soft dirty bit is set for non-present entries only,
never for present ones, just at moment we form swap pte
On 08/22/2013 01:32 PM, David Vrabel wrote:
On 22/08/13 00:04, Linus Torvalds wrote:
On Wed, Aug 21, 2013 at 12:03 PM, Cyrill Gorcunov gorcu...@gmail.com wrote:
I personally don't see bug here because
- this swapped page soft dirty bit is set for non-present entries only,
never for
On Thu, Aug 22, 2013 at 08:27:45AM +0100, Jan Beulich wrote:
On 22.08.13 at 09:03, Cyrill Gorcunov gorcu...@gmail.com wrote:
Ok, how about this?
static inline pte_t pte_swp_mksoft_dirty(pte_t pte)
{
BUG_ON(pte_present(pte));
return pte_set_flags(pte, _PAGE_SWP_SOFT_DIRTY);
On 22.08.13 at 13:27, Cyrill Gorcunov gorcu...@gmail.com wrote:
On Thu, Aug 22, 2013 at 08:27:45AM +0100, Jan Beulich wrote:
On 22.08.13 at 09:03, Cyrill Gorcunov gorcu...@gmail.com wrote:
Ok, how about this?
static inline pte_t pte_swp_mksoft_dirty(pte_t pte)
{
On 08/22/2013 03:33 PM, Jan Beulich wrote:
From: Cyrill Gorcunov gorcu...@gmail.com
Subject: [PATCH] mm: Make sure _PAGE_SWP_SOFT_DIRTY bit is not set on
present pte
_PAGE_SOFT_DIRTY bit should never be set on present pte so add
VM_BUG_ON to catch any potential future abuse.
Also add a
On Thu, Aug 22, 2013 at 03:37:33PM +0900, Minchan Kim wrote:
Look, good thing is that 7th bit also available on the 4level pages
(ie x86-64) without additional code modification, that's why I picked
it in first place. I prepared the patch locally which doesn't use
pse bit for tracking
On Wed, Aug 21, 2013 at 04:42:53PM -0700, Andi Kleen wrote:
> Cyrill Gorcunov writes:
> >
> > Hi all, I worked on patch which would not touch PSE bit for dirty page
> > tracking and the result is not that good:
> >
> > - 2level pages now always page dirty if page is swapped in and out, because
>
On Wed, Aug 21, 2013 at 08:51:15PM -0400, Dave Jones wrote:
>
> Could this explain what I'm seeing in another thread ?
> https://lkml.org/lkml/2013/8/7/27
Don't think so. This code is merged in -rc6, while your report is saying the
kernel version is -rc4 (also this feature requires
On Wed, Aug 21, 2013 at 04:04:54PM -0700, Linus Torvalds wrote:
> On Wed, Aug 21, 2013 at 12:03 PM, Cyrill Gorcunov wrote:
> >
> > I personally don't see bug here because
> >
> > - this swapped page soft dirty bit is set for non-present entries only,
> >never for present ones, just at
Cyrill Gorcunov writes:
>
> Hi all, I worked on patch which would not touch PSE bit for dirty page
> tracking and the result is not that good:
>
> - 2level pages now always page dirty if page is swapped in and out, because
>there is no space left in PTE (other than PSE bit)
Maybe just don't
On Wed, Aug 21, 2013 at 09:30:03AM -0700, Linus Torvalds wrote:
> Quite frankly, unless I see a patch later today that is
>
> (a) obvious
> (b) explains what is going on
> (c) tested
>
> I will be reverting the whole soft-dirty mess. I thought the
> bit-mapping games it played were already
On Wed, Aug 21, 2013 at 12:03 PM, Cyrill Gorcunov wrote:
>
> I personally don't see bug here because
>
> - this swapped page soft dirty bit is set for non-present entries only,
>never for present ones, just at moment we form swap pte entry
>
> - i don't find any code which would test for
On 08/21/2013 11:07 PM, Andy Lutomirski wrote:
> On Wed, Aug 21, 2013 at 12:03 PM, Cyrill Gorcunov wrote:
>> On Wed, Aug 21, 2013 at 08:50:26PM +0200, H. Peter Anvin wrote:
>>> But is there a manifest bug or not? What is the deal with Xen?
>>>
>>
>> I personally don't see bug here because
>>
>>
On Wed, Aug 21, 2013 at 12:07:13PM -0700, Andy Lutomirski wrote:
> On Wed, Aug 21, 2013 at 12:03 PM, Cyrill Gorcunov wrote:
> > On Wed, Aug 21, 2013 at 08:50:26PM +0200, H. Peter Anvin wrote:
> >> But is there a manifest bug or not? What is the deal with Xen?
> >>
> >
> > I personally don't see
On Wed, Aug 21, 2013 at 12:03 PM, Cyrill Gorcunov wrote:
> On Wed, Aug 21, 2013 at 08:50:26PM +0200, H. Peter Anvin wrote:
>> But is there a manifest bug or not? What is the deal with Xen?
>>
>
> I personally don't see bug here because
>
> - this swapped page soft dirty bit is set for
On Wed, Aug 21, 2013 at 08:50:26PM +0200, H. Peter Anvin wrote:
> But is there a manifest bug or not? What is the deal with Xen?
>
I personally don't see bug here because
- this swapped page soft dirty bit is set for non-present entries only,
never for present ones, just at moment we form
But is there a manifest bug or not? What is the deal with Xen?
Cyrill Gorcunov wrote:
>On Wed, Aug 21, 2013 at 09:25:47PM +0400, Cyrill Gorcunov wrote:
>> >
>> > However, I do find the use of PTE bits in this way somewhat
>fragile.
>> > What other potential corner cases might still remain that
On Wed, Aug 21, 2013 at 09:25:47PM +0400, Cyrill Gorcunov wrote:
> >
> > However, I do find the use of PTE bits in this way somewhat fragile.
> > What other potential corner cases might still remain that will require
> > further games with PTE bits?
>
> OK, so this is not a bug finally. The
On Wed, Aug 21, 2013 at 6:48 AM, David Vrabel wrote:
> All,
>
> 179ef71c (mm: save soft-dirty bits on swapped pages) introduces a new
> PTE bit on x86 _PTE_SWP_SOFT_DIRTY which has the same value as _PTE_PSE
> and _PTE_PAT.
>
> With a Xen PV guest, the use of the _PTE_PAT will result in the page
On Wed, Aug 21, 2013 at 05:56:08PM +0100, David Vrabel wrote:
> >
> > And I asked David to point me how it happens, because I don't
> > understand at which point pse bit get analized when page is
> > not present.
>
> As Jan said, we're concerned that the bit was being used on present PTEs
> and
On 21/08/13 17:19, Cyrill Gorcunov wrote:
> On Wed, Aug 21, 2013 at 05:03:13PM +0100, Jan Beulich wrote:
>>>
>>> Only to non-present ptes, as far as I know.
>>
>> That's not really any guarantee. And the accessor functions also
>> don't check that they'd be used on non-present PTEs only.
>
>
On Wed, Aug 21, 2013 at 09:30:03AM -0700, Linus Torvalds wrote:
> On Wed, Aug 21, 2013 at 7:19 AM, Cyrill Gorcunov wrote:
> > On Wed, Aug 21, 2013 at 04:11:25PM +0200, H. Peter Anvin wrote:
> >>
> >> Eep. This should be reverted, indeed. This isn't a manifest bug on !Xen
> >> but we
> >> have
On Wed, Aug 21, 2013 at 7:19 AM, Cyrill Gorcunov wrote:
> On Wed, Aug 21, 2013 at 04:11:25PM +0200, H. Peter Anvin wrote:
>>
>> Eep. This should be reverted, indeed. This isn't a manifest bug on !Xen
>> but we
>> have gotten requests for WT support which would mean adding in the PAT but
>>
On Wed, Aug 21, 2013 at 05:03:13PM +0100, Jan Beulich wrote:
> >
> > Only to non-present ptes, as far as I know.
>
> That's not really any guarantee. And the accessor functions also
> don't check that they'd be used on non-present PTEs only.
Wait. This _PAGE_SWP_SOFT_DIRTY bit (which is in real
>>> On 21.08.13 at 17:42, Cyrill Gorcunov wrote:
> On Wed, Aug 21, 2013 at 03:53:36PM +0100, Jan Beulich wrote:
>> >>> On 21.08.13 at 16:12, Cyrill Gorcunov wrote:
>> > David, could you please explain, Xen keeps and analyze _PTE_PAT bit
>> > for ptes which are not present?
>>
>> No, the problem
On Wed, Aug 21, 2013 at 03:53:36PM +0100, Jan Beulich wrote:
> >>> On 21.08.13 at 16:12, Cyrill Gorcunov wrote:
> > On Wed, Aug 21, 2013 at 02:48:20PM +0100, David Vrabel wrote:
> >> All,
> >>
> >> 179ef71c (mm: save soft-dirty bits on swapped pages) introduces a new
> >> PTE bit on x86
Only WB pages should be swappable, but even so, the cacheability should be in
the vma.
Jan Beulich wrote:
On 21.08.13 at 16:12, Cyrill Gorcunov wrote:
>> On Wed, Aug 21, 2013 at 02:48:20PM +0100, David Vrabel wrote:
>>> All,
>>>
>>> 179ef71c (mm: save soft-dirty bits on swapped pages)
>>> On 21.08.13 at 16:12, Cyrill Gorcunov wrote:
> On Wed, Aug 21, 2013 at 02:48:20PM +0100, David Vrabel wrote:
>> All,
>>
>> 179ef71c (mm: save soft-dirty bits on swapped pages) introduces a new
>> PTE bit on x86 _PTE_SWP_SOFT_DIRTY which has the same value as _PTE_PSE
>> and _PTE_PAT.
>>
>>
On Wed, Aug 21, 2013 at 04:22:21PM +0200, H. Peter Anvin wrote:
> OK now I'm confused. I guess I shouldn't comment while on vacation
> and cache cold on everything.
I rather think I'm missing something, that's why I asked David how
this featue affects non present pte.
--
To unsubscribe from this
Good question...
Cyrill Gorcunov wrote:
>On Wed, Aug 21, 2013 at 02:48:20PM +0100, David Vrabel wrote:
>> All,
>>
>> 179ef71c (mm: save soft-dirty bits on swapped pages) introduces a new
>> PTE bit on x86 _PTE_SWP_SOFT_DIRTY which has the same value as
>_PTE_PSE
>> and _PTE_PAT.
>>
>> With a
OK now I'm confused. I guess I shouldn't comment while on vacation and cache
cold on everything.
Cyrill Gorcunov wrote:
>On Wed, Aug 21, 2013 at 04:11:25PM +0200, H. Peter Anvin wrote:
>>
>> Eep. This should be reverted, indeed. This isn't a manifest bug on
>!Xen but we
>> have gotten
On Wed, Aug 21, 2013 at 04:11:25PM +0200, H. Peter Anvin wrote:
>
> Eep. This should be reverted, indeed. This isn't a manifest bug on !Xen but
> we
> have gotten requests for WT support which would mean adding in the PAT but
> again.
Please no, letme fix it. That's what I'm having in mind:
Eep. This should be reverted, indeed. This isn't a manifest bug on !Xen but
we have gotten requests for WT support which would mean adding in the PAT but
again.
David Vrabel wrote:
>All,
>
>179ef71c (mm: save soft-dirty bits on swapped pages) introduces a new
>PTE bit on x86
On Wed, Aug 21, 2013 at 02:48:20PM +0100, David Vrabel wrote:
> All,
>
> 179ef71c (mm: save soft-dirty bits on swapped pages) introduces a new
> PTE bit on x86 _PTE_SWP_SOFT_DIRTY which has the same value as _PTE_PSE
> and _PTE_PAT.
>
> With a Xen PV guest, the use of the _PTE_PAT will result in
On 8/21/2013 9:48 AM, David Vrabel wrote:
All,
179ef71c (mm: save soft-dirty bits on swapped pages) introduces a new
PTE bit on x86 _PTE_SWP_SOFT_DIRTY which has the same value as _PTE_PSE
and _PTE_PAT.
With a Xen PV guest, the use of the _PTE_PAT will result in the page
having unexpected
All,
179ef71c (mm: save soft-dirty bits on swapped pages) introduces a new
PTE bit on x86 _PTE_SWP_SOFT_DIRTY which has the same value as _PTE_PSE
and _PTE_PAT.
With a Xen PV guest, the use of the _PTE_PAT will result in the page
having unexpected cachability which will introduce a range of
All,
179ef71c (mm: save soft-dirty bits on swapped pages) introduces a new
PTE bit on x86 _PTE_SWP_SOFT_DIRTY which has the same value as _PTE_PSE
and _PTE_PAT.
With a Xen PV guest, the use of the _PTE_PAT will result in the page
having unexpected cachability which will introduce a range of
On 8/21/2013 9:48 AM, David Vrabel wrote:
All,
179ef71c (mm: save soft-dirty bits on swapped pages) introduces a new
PTE bit on x86 _PTE_SWP_SOFT_DIRTY which has the same value as _PTE_PSE
and _PTE_PAT.
With a Xen PV guest, the use of the _PTE_PAT will result in the page
having unexpected
On Wed, Aug 21, 2013 at 02:48:20PM +0100, David Vrabel wrote:
All,
179ef71c (mm: save soft-dirty bits on swapped pages) introduces a new
PTE bit on x86 _PTE_SWP_SOFT_DIRTY which has the same value as _PTE_PSE
and _PTE_PAT.
With a Xen PV guest, the use of the _PTE_PAT will result in the
Eep. This should be reverted, indeed. This isn't a manifest bug on !Xen but
we have gotten requests for WT support which would mean adding in the PAT but
again.
David Vrabel david.vra...@citrix.com wrote:
All,
179ef71c (mm: save soft-dirty bits on swapped pages) introduces a new
PTE bit on
On Wed, Aug 21, 2013 at 04:11:25PM +0200, H. Peter Anvin wrote:
Eep. This should be reverted, indeed. This isn't a manifest bug on !Xen but
we
have gotten requests for WT support which would mean adding in the PAT but
again.
Please no, letme fix it. That's what I'm having in mind: don't
OK now I'm confused. I guess I shouldn't comment while on vacation and cache
cold on everything.
Cyrill Gorcunov gorcu...@gmail.com wrote:
On Wed, Aug 21, 2013 at 04:11:25PM +0200, H. Peter Anvin wrote:
Eep. This should be reverted, indeed. This isn't a manifest bug on
!Xen but we
have
Good question...
Cyrill Gorcunov gorcu...@gmail.com wrote:
On Wed, Aug 21, 2013 at 02:48:20PM +0100, David Vrabel wrote:
All,
179ef71c (mm: save soft-dirty bits on swapped pages) introduces a new
PTE bit on x86 _PTE_SWP_SOFT_DIRTY which has the same value as
_PTE_PSE
and _PTE_PAT.
With a
On Wed, Aug 21, 2013 at 04:22:21PM +0200, H. Peter Anvin wrote:
OK now I'm confused. I guess I shouldn't comment while on vacation
and cache cold on everything.
I rather think I'm missing something, that's why I asked David how
this featue affects non present pte.
--
To unsubscribe from this
On 21.08.13 at 16:12, Cyrill Gorcunov gorcu...@gmail.com wrote:
On Wed, Aug 21, 2013 at 02:48:20PM +0100, David Vrabel wrote:
All,
179ef71c (mm: save soft-dirty bits on swapped pages) introduces a new
PTE bit on x86 _PTE_SWP_SOFT_DIRTY which has the same value as _PTE_PSE
and _PTE_PAT.
Only WB pages should be swappable, but even so, the cacheability should be in
the vma.
Jan Beulich jbeul...@suse.com wrote:
On 21.08.13 at 16:12, Cyrill Gorcunov gorcu...@gmail.com wrote:
On Wed, Aug 21, 2013 at 02:48:20PM +0100, David Vrabel wrote:
All,
179ef71c (mm: save soft-dirty bits
On Wed, Aug 21, 2013 at 03:53:36PM +0100, Jan Beulich wrote:
On 21.08.13 at 16:12, Cyrill Gorcunov gorcu...@gmail.com wrote:
On Wed, Aug 21, 2013 at 02:48:20PM +0100, David Vrabel wrote:
All,
179ef71c (mm: save soft-dirty bits on swapped pages) introduces a new
PTE bit on x86
On 21.08.13 at 17:42, Cyrill Gorcunov gorcu...@gmail.com wrote:
On Wed, Aug 21, 2013 at 03:53:36PM +0100, Jan Beulich wrote:
On 21.08.13 at 16:12, Cyrill Gorcunov gorcu...@gmail.com wrote:
David, could you please explain, Xen keeps and analyze _PTE_PAT bit
for ptes which are not present?
On Wed, Aug 21, 2013 at 05:03:13PM +0100, Jan Beulich wrote:
Only to non-present ptes, as far as I know.
That's not really any guarantee. And the accessor functions also
don't check that they'd be used on non-present PTEs only.
Wait. This _PAGE_SWP_SOFT_DIRTY bit (which is in real PSE
On Wed, Aug 21, 2013 at 7:19 AM, Cyrill Gorcunov gorcu...@gmail.com wrote:
On Wed, Aug 21, 2013 at 04:11:25PM +0200, H. Peter Anvin wrote:
Eep. This should be reverted, indeed. This isn't a manifest bug on !Xen
but we
have gotten requests for WT support which would mean adding in the PAT
On Wed, Aug 21, 2013 at 09:30:03AM -0700, Linus Torvalds wrote:
On Wed, Aug 21, 2013 at 7:19 AM, Cyrill Gorcunov gorcu...@gmail.com wrote:
On Wed, Aug 21, 2013 at 04:11:25PM +0200, H. Peter Anvin wrote:
Eep. This should be reverted, indeed. This isn't a manifest bug on !Xen
but we
On 21/08/13 17:19, Cyrill Gorcunov wrote:
On Wed, Aug 21, 2013 at 05:03:13PM +0100, Jan Beulich wrote:
Only to non-present ptes, as far as I know.
That's not really any guarantee. And the accessor functions also
don't check that they'd be used on non-present PTEs only.
Wait. This
On Wed, Aug 21, 2013 at 05:56:08PM +0100, David Vrabel wrote:
And I asked David to point me how it happens, because I don't
understand at which point pse bit get analized when page is
not present.
As Jan said, we're concerned that the bit was being used on present PTEs
and not just
On Wed, Aug 21, 2013 at 6:48 AM, David Vrabel david.vra...@citrix.com wrote:
All,
179ef71c (mm: save soft-dirty bits on swapped pages) introduces a new
PTE bit on x86 _PTE_SWP_SOFT_DIRTY which has the same value as _PTE_PSE
and _PTE_PAT.
With a Xen PV guest, the use of the _PTE_PAT will
On Wed, Aug 21, 2013 at 09:25:47PM +0400, Cyrill Gorcunov wrote:
However, I do find the use of PTE bits in this way somewhat fragile.
What other potential corner cases might still remain that will require
further games with PTE bits?
OK, so this is not a bug finally. The problem is
But is there a manifest bug or not? What is the deal with Xen?
Cyrill Gorcunov gorcu...@gmail.com wrote:
On Wed, Aug 21, 2013 at 09:25:47PM +0400, Cyrill Gorcunov wrote:
However, I do find the use of PTE bits in this way somewhat
fragile.
What other potential corner cases might still
On Wed, Aug 21, 2013 at 08:50:26PM +0200, H. Peter Anvin wrote:
But is there a manifest bug or not? What is the deal with Xen?
I personally don't see bug here because
- this swapped page soft dirty bit is set for non-present entries only,
never for present ones, just at moment we form
On Wed, Aug 21, 2013 at 12:03 PM, Cyrill Gorcunov gorcu...@gmail.com wrote:
On Wed, Aug 21, 2013 at 08:50:26PM +0200, H. Peter Anvin wrote:
But is there a manifest bug or not? What is the deal with Xen?
I personally don't see bug here because
- this swapped page soft dirty bit is set for
On Wed, Aug 21, 2013 at 12:07:13PM -0700, Andy Lutomirski wrote:
On Wed, Aug 21, 2013 at 12:03 PM, Cyrill Gorcunov gorcu...@gmail.com wrote:
On Wed, Aug 21, 2013 at 08:50:26PM +0200, H. Peter Anvin wrote:
But is there a manifest bug or not? What is the deal with Xen?
I personally don't
On 08/21/2013 11:07 PM, Andy Lutomirski wrote:
On Wed, Aug 21, 2013 at 12:03 PM, Cyrill Gorcunov gorcu...@gmail.com wrote:
On Wed, Aug 21, 2013 at 08:50:26PM +0200, H. Peter Anvin wrote:
But is there a manifest bug or not? What is the deal with Xen?
I personally don't see bug here because
On Wed, Aug 21, 2013 at 12:03 PM, Cyrill Gorcunov gorcu...@gmail.com wrote:
I personally don't see bug here because
- this swapped page soft dirty bit is set for non-present entries only,
never for present ones, just at moment we form swap pte entry
- i don't find any code which would
On Wed, Aug 21, 2013 at 09:30:03AM -0700, Linus Torvalds wrote:
Quite frankly, unless I see a patch later today that is
(a) obvious
(b) explains what is going on
(c) tested
I will be reverting the whole soft-dirty mess. I thought the
bit-mapping games it played were already too
Cyrill Gorcunov gorcu...@gmail.com writes:
Hi all, I worked on patch which would not touch PSE bit for dirty page
tracking and the result is not that good:
- 2level pages now always page dirty if page is swapped in and out, because
there is no space left in PTE (other than PSE bit)
On Wed, Aug 21, 2013 at 04:04:54PM -0700, Linus Torvalds wrote:
On Wed, Aug 21, 2013 at 12:03 PM, Cyrill Gorcunov gorcu...@gmail.com wrote:
I personally don't see bug here because
- this swapped page soft dirty bit is set for non-present entries only,
never for present ones,
On Wed, Aug 21, 2013 at 08:51:15PM -0400, Dave Jones wrote:
Could this explain what I'm seeing in another thread ?
https://lkml.org/lkml/2013/8/7/27
Don't think so. This code is merged in -rc6, while your report is saying the
kernel version is -rc4 (also this feature requires
On Wed, Aug 21, 2013 at 04:42:53PM -0700, Andi Kleen wrote:
Cyrill Gorcunov gorcu...@gmail.com writes:
Hi all, I worked on patch which would not touch PSE bit for dirty page
tracking and the result is not that good:
- 2level pages now always page dirty if page is swapped in and out,
88 matches
Mail list logo