* Re: [Discontig-devel] Re: [Linux-ia64] Re: 2.5.59 & mmap_sem
@ 2003-02-18 10:43 Andrew Morton
2003-02-24 10:14 ` Xavier Bru
` (2 more replies)
0 siblings, 3 replies; 4+ messages in thread
From: Andrew Morton @ 2003-02-18 10:43 UTC (permalink / raw)
To: linux-ia64
William Lee Irwin III <wli@holomorphy.com> wrote:
>
> On Tue, Feb 18, 2003 at 09:46:31AM +0100, Xavier Bru wrote:
> > Thanks for your answers.
> > You are right, we do not need a page structure for mapping /dev/mem in
> > IO space (I am not a mm expert :-).
> > Here after a possible patch that allows the Xserver running on the NUMA
> > platform. (We had the same problem on Azusa).
>
> Looks perfectly reasonable to me. Other discontig/NUMA implementations
> are likely to run into the same issue.
>
> You might want to send this on to lkml and cc: akpm, he's pretty much
> the final destination wrt. VM patches and 2.5.
>
I'd prefer to do this at the vma level if poss. Given that mmap_mem()
has set VM_RESERVED against the VMA, it should be sufficient to do
diff -puN mm/memory.c~a mm/memory.c
--- 25/mm/memory.c~a 2003-02-18 02:40:01.000000000 -0800
+++ 25-akpm/mm/memory.c 2003-02-18 02:40:18.000000000 -0800
@@ -211,6 +211,9 @@ int copy_page_range(struct mm_struct *ds
unsigned long cow;
struct pte_chain *pte_chain = NULL;
+ if (vma->vm_flags & (VM_RESERVED|VM_IO))
+ return 0;
+
if (is_vm_hugetlb_page(vma))
return copy_hugetlb_page_range(dst, src, vma);
diff -puN fs/hugetlbfs/inode.c~a fs/hugetlbfs/inode.c
--- 25/fs/hugetlbfs/inode.c~a 2003-02-18 02:42:47.000000000 -0800
+++ 25-akpm/fs/hugetlbfs/inode.c 2003-02-18 02:42:55.000000000 -0800
@@ -63,7 +63,7 @@ static int hugetlbfs_file_mmap(struct fi
down(&inode->i_sem);
UPDATE_ATIME(inode);
- vma->vm_flags |= VM_HUGETLB | VM_RESERVED;
+ vma->vm_flags |= VM_HUGETLB;
vma->vm_ops = &hugetlb_vm_ops;
ret = hugetlb_prefault(mapping, vma);
len = (loff_t)(vma->vm_end - vma->vm_start) +
_
The semantics (and usage!) of the VM_foo flags are rather vague. It needs
definiton, and an audit.
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [Discontig-devel] Re: [Linux-ia64] Re: 2.5.59 & mmap_sem
2003-02-18 10:43 [Discontig-devel] Re: [Linux-ia64] Re: 2.5.59 & mmap_sem Andrew Morton
@ 2003-02-24 10:14 ` Xavier Bru
2003-03-02 1:32 ` Andrew Morton
2003-03-03 15:37 ` Xavier Bru
2 siblings, 0 replies; 4+ messages in thread
From: Xavier Bru @ 2003-02-24 10:14 UTC (permalink / raw)
To: linux-ia64
Works OK for us.
Thanks
--
Sincères salutations.
_____________________________________________________________________
Xavier BRU BULL ISD/R&D/INTEL office: FREC B1-422
tel : +33 (0)4 76 29 77 45 http://www-frec.bull.fr
fax : +33 (0)4 76 29 77 70 mailto:Xavier.Bru@bull.net
addr: BULL, 1 rue de Provence, BP 208, 38432 Echirolles Cedex, FRANCE
_____________________________________________________________________
Andrew Morton writes:
> William Lee Irwin III <wli@holomorphy.com> wrote:
> >
> > On Tue, Feb 18, 2003 at 09:46:31AM +0100, Xavier Bru wrote:
> > > Thanks for your answers.
> > > You are right, we do not need a page structure for mapping /dev/mem in
> > > IO space (I am not a mm expert :-).
> > > Here after a possible patch that allows the Xserver running on the NUMA
> > > platform. (We had the same problem on Azusa).
> >
> > Looks perfectly reasonable to me. Other discontig/NUMA implementations
> > are likely to run into the same issue.
> >
> > You might want to send this on to lkml and cc: akpm, he's pretty much
> > the final destination wrt. VM patches and 2.5.
> >
>
> I'd prefer to do this at the vma level if poss. Given that mmap_mem()
> has set VM_RESERVED against the VMA, it should be sufficient to do
>
> diff -puN mm/memory.c~a mm/memory.c
> --- 25/mm/memory.c~a 2003-02-18 02:40:01.000000000 -0800
> +++ 25-akpm/mm/memory.c 2003-02-18 02:40:18.000000000 -0800
> @@ -211,6 +211,9 @@ int copy_page_range(struct mm_struct *ds
> unsigned long cow;
> struct pte_chain *pte_chain = NULL;
>
> + if (vma->vm_flags & (VM_RESERVED|VM_IO))
> + return 0;
> +
> if (is_vm_hugetlb_page(vma))
> return copy_hugetlb_page_range(dst, src, vma);
>
> diff -puN fs/hugetlbfs/inode.c~a fs/hugetlbfs/inode.c
> --- 25/fs/hugetlbfs/inode.c~a 2003-02-18 02:42:47.000000000 -0800
> +++ 25-akpm/fs/hugetlbfs/inode.c 2003-02-18 02:42:55.000000000 -0800
> @@ -63,7 +63,7 @@ static int hugetlbfs_file_mmap(struct fi
> down(&inode->i_sem);
>
> UPDATE_ATIME(inode);
> - vma->vm_flags |= VM_HUGETLB | VM_RESERVED;
> + vma->vm_flags |= VM_HUGETLB;
> vma->vm_ops = &hugetlb_vm_ops;
> ret = hugetlb_prefault(mapping, vma);
> len = (loff_t)(vma->vm_end - vma->vm_start) +
>
> _
>
>
> The semantics (and usage!) of the VM_foo flags are rather vague. It needs
> definiton, and an audit.
>
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [Discontig-devel] Re: [Linux-ia64] Re: 2.5.59 & mmap_sem
2003-02-18 10:43 [Discontig-devel] Re: [Linux-ia64] Re: 2.5.59 & mmap_sem Andrew Morton
2003-02-24 10:14 ` Xavier Bru
@ 2003-03-02 1:32 ` Andrew Morton
2003-03-03 15:37 ` Xavier Bru
2 siblings, 0 replies; 4+ messages in thread
From: Andrew Morton @ 2003-03-02 1:32 UTC (permalink / raw)
To: linux-ia64
On reflection, I think your approach is best - there is (some) value in being
able to inherit a /dev/mem mapping across fork(), and ceasing to do so may
break applications.
I recast your patch a little. Saves a few instructions and prevents a bogus
uninitialised variable warning.
diff -puN mm/memory.c~copy_page_range-invalid-page-fix mm/memory.c
--- 25/mm/memory.c~copy_page_range-invalid-page-fix 2003-03-01 17:26:03.000000000 -0800
+++ 25-akpm/mm/memory.c 2003-03-01 17:26:29.000000000 -0800
@@ -286,9 +286,11 @@ skip_copy_pte_range:
goto cont_copy_pte_range_noset;
}
pfn = pte_pfn(pte);
+ if (!pfn_valid(pfn)) {
+ set_pte(dst_pte, pte);
+ goto cont_copy_pte_range_noset;
+ }
page = pfn_to_page(pfn);
- if (!pfn_valid(pfn))
- goto cont_copy_pte_range;
if (PageReserved(page))
goto cont_copy_pte_range;
_
Xavier Bru <Xavier.Bru@bull.net> wrote:
>
> Works OK for us.
> Thanks
>
> --
>
> _____________________________________________________________________
>
> Xavier BRU BULL ISD/R&D/INTEL office: FREC B1-422
> tel : +33 (0)4 76 29 77 45 http://www-frec.bull.fr
> fax : +33 (0)4 76 29 77 70 mailto:Xavier.Bru@bull.net
> addr: BULL, 1 rue de Provence, BP 208, 38432 Echirolles Cedex, FRANCE
> _____________________________________________________________________
>
> Andrew Morton writes:
> > William Lee Irwin III <wli@holomorphy.com> wrote:
> > >
> > > On Tue, Feb 18, 2003 at 09:46:31AM +0100, Xavier Bru wrote:
> > > > Thanks for your answers.
> > > > You are right, we do not need a page structure for mapping /dev/mem in
> > > > IO space (I am not a mm expert :-).
> > > > Here after a possible patch that allows the Xserver running on the NUMA
> > > > platform. (We had the same problem on Azusa).
> > >
> > > Looks perfectly reasonable to me. Other discontig/NUMA implementations
> > > are likely to run into the same issue.
> > >
> > > You might want to send this on to lkml and cc: akpm, he's pretty much
> > > the final destination wrt. VM patches and 2.5.
> > >
> >
> > I'd prefer to do this at the vma level if poss. Given that mmap_mem()
> > has set VM_RESERVED against the VMA, it should be sufficient to do
> >
> > diff -puN mm/memory.c~a mm/memory.c
> > --- 25/mm/memory.c~a 2003-02-18 02:40:01.000000000 -0800
> > +++ 25-akpm/mm/memory.c 2003-02-18 02:40:18.000000000 -0800
> > @@ -211,6 +211,9 @@ int copy_page_range(struct mm_struct *ds
> > unsigned long cow;
> > struct pte_chain *pte_chain = NULL;
> >
> > + if (vma->vm_flags & (VM_RESERVED|VM_IO))
> > + return 0;
> > +
> > if (is_vm_hugetlb_page(vma))
> > return copy_hugetlb_page_range(dst, src, vma);
> >
> > diff -puN fs/hugetlbfs/inode.c~a fs/hugetlbfs/inode.c
> > --- 25/fs/hugetlbfs/inode.c~a 2003-02-18 02:42:47.000000000 -0800
> > +++ 25-akpm/fs/hugetlbfs/inode.c 2003-02-18 02:42:55.000000000 -0800
> > @@ -63,7 +63,7 @@ static int hugetlbfs_file_mmap(struct fi
> > down(&inode->i_sem);
> >
> > UPDATE_ATIME(inode);
> > - vma->vm_flags |= VM_HUGETLB | VM_RESERVED;
> > + vma->vm_flags |= VM_HUGETLB;
> > vma->vm_ops = &hugetlb_vm_ops;
> > ret = hugetlb_prefault(mapping, vma);
> > len = (loff_t)(vma->vm_end - vma->vm_start) +
> >
> > _
> >
> >
> > The semantics (and usage!) of the VM_foo flags are rather vague. It needs
> > definiton, and an audit.
> >
>
>
> -------------------------------------------------------
> This sf.net email is sponsored by:ThinkGeek
> Welcome to geek heaven.
> http://thinkgeek.com/sf
> _______________________________________________
> Discontig-devel mailing list
> Discontig-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/discontig-devel
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [Discontig-devel] Re: [Linux-ia64] Re: 2.5.59 & mmap_sem
2003-02-18 10:43 [Discontig-devel] Re: [Linux-ia64] Re: 2.5.59 & mmap_sem Andrew Morton
2003-02-24 10:14 ` Xavier Bru
2003-03-02 1:32 ` Andrew Morton
@ 2003-03-03 15:37 ` Xavier Bru
2 siblings, 0 replies; 4+ messages in thread
From: Xavier Bru @ 2003-03-03 15:37 UTC (permalink / raw)
To: linux-ia64
I tested this one OK on 2.5.60.
Thanks.
Xavier
--
Sincères salutations.
_____________________________________________________________________
Xavier BRU BULL ISD/R&D/INTEL office: FREC B1-422
tel : +33 (0)4 76 29 77 45 http://www-frec.bull.fr
fax : +33 (0)4 76 29 77 70 mailto:Xavier.Bru@bull.net
addr: BULL, 1 rue de Provence, BP 208, 38432 Echirolles Cedex, FRANCE
_____________________________________________________________________
Andrew Morton writes:
>
> On reflection, I think your approach is best - there is (some) value in being
> able to inherit a /dev/mem mapping across fork(), and ceasing to do so may
> break applications.
>
> I recast your patch a little. Saves a few instructions and prevents a bogus
> uninitialised variable warning.
>
>
> diff -puN mm/memory.c~copy_page_range-invalid-page-fix mm/memory.c
> --- 25/mm/memory.c~copy_page_range-invalid-page-fix 2003-03-01 17:26:03.000000000 -0800
> +++ 25-akpm/mm/memory.c 2003-03-01 17:26:29.000000000 -0800
> @@ -286,9 +286,11 @@ skip_copy_pte_range:
> goto cont_copy_pte_range_noset;
> }
> pfn = pte_pfn(pte);
> + if (!pfn_valid(pfn)) {
> + set_pte(dst_pte, pte);
> + goto cont_copy_pte_range_noset;
> + }
> page = pfn_to_page(pfn);
> - if (!pfn_valid(pfn))
> - goto cont_copy_pte_range;
> if (PageReserved(page))
> goto cont_copy_pte_range;
>
>
> _
>
> Xavier Bru <Xavier.Bru@bull.net> wrote:
> >
> > Works OK for us.
> > Thanks
> >
> > --
> >
> > _____________________________________________________________________
> >
> > Xavier BRU BULL ISD/R&D/INTEL office: FREC B1-422
> > tel : +33 (0)4 76 29 77 45 http://www-frec.bull.fr
> > fax : +33 (0)4 76 29 77 70 mailto:Xavier.Bru@bull.net
> > addr: BULL, 1 rue de Provence, BP 208, 38432 Echirolles Cedex, FRANCE
> > _____________________________________________________________________
> >
> > Andrew Morton writes:
> > > William Lee Irwin III <wli@holomorphy.com> wrote:
> > > >
> > > > On Tue, Feb 18, 2003 at 09:46:31AM +0100, Xavier Bru wrote:
> > > > > Thanks for your answers.
> > > > > You are right, we do not need a page structure for mapping /dev/mem in
> > > > > IO space (I am not a mm expert :-).
> > > > > Here after a possible patch that allows the Xserver running on the NUMA
> > > > > platform. (We had the same problem on Azusa).
> > > >
> > > > Looks perfectly reasonable to me. Other discontig/NUMA implementations
> > > > are likely to run into the same issue.
> > > >
> > > > You might want to send this on to lkml and cc: akpm, he's pretty much
> > > > the final destination wrt. VM patches and 2.5.
> > > >
> > >
> > > I'd prefer to do this at the vma level if poss. Given that mmap_mem()
> > > has set VM_RESERVED against the VMA, it should be sufficient to do
> > >
> > > diff -puN mm/memory.c~a mm/memory.c
> > > --- 25/mm/memory.c~a 2003-02-18 02:40:01.000000000 -0800
> > > +++ 25-akpm/mm/memory.c 2003-02-18 02:40:18.000000000 -0800
> > > @@ -211,6 +211,9 @@ int copy_page_range(struct mm_struct *ds
> > > unsigned long cow;
> > > struct pte_chain *pte_chain = NULL;
> > >
> > > + if (vma->vm_flags & (VM_RESERVED|VM_IO))
> > > + return 0;
> > > +
> > > if (is_vm_hugetlb_page(vma))
> > > return copy_hugetlb_page_range(dst, src, vma);
> > >
> > > diff -puN fs/hugetlbfs/inode.c~a fs/hugetlbfs/inode.c
> > > --- 25/fs/hugetlbfs/inode.c~a 2003-02-18 02:42:47.000000000 -0800
> > > +++ 25-akpm/fs/hugetlbfs/inode.c 2003-02-18 02:42:55.000000000 -0800
> > > @@ -63,7 +63,7 @@ static int hugetlbfs_file_mmap(struct fi
> > > down(&inode->i_sem);
> > >
> > > UPDATE_ATIME(inode);
> > > - vma->vm_flags |= VM_HUGETLB | VM_RESERVED;
> > > + vma->vm_flags |= VM_HUGETLB;
> > > vma->vm_ops = &hugetlb_vm_ops;
> > > ret = hugetlb_prefault(mapping, vma);
> > > len = (loff_t)(vma->vm_end - vma->vm_start) +
> > >
> > > _
> > >
> > >
> > > The semantics (and usage!) of the VM_foo flags are rather vague. It needs
> > > definiton, and an audit.
> > >
> >
> >
> > -------------------------------------------------------
> > This sf.net email is sponsored by:ThinkGeek
> > Welcome to geek heaven.
> > http://thinkgeek.com/sf
> > _______________________________________________
> > Discontig-devel mailing list
> > Discontig-devel@lists.sourceforge.net
> > https://lists.sourceforge.net/lists/listinfo/discontig-devel
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2003-03-03 15:37 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2003-02-18 10:43 [Discontig-devel] Re: [Linux-ia64] Re: 2.5.59 & mmap_sem Andrew Morton
2003-02-24 10:14 ` Xavier Bru
2003-03-02 1:32 ` Andrew Morton
2003-03-03 15:37 ` Xavier Bru
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox