Re: [patch 3/6] mm: fix fault vs invalidate race for linear mappings

2007-03-06 Thread Nick Piggin
On Tue, Mar 06, 2007 at 11:08:41PM -0800, Andrew Morton wrote: > On Wed, 7 Mar 2007 07:57:27 +0100 Nick Piggin <[EMAIL PROTECTED]> wrote: > > > > > > > Why was truncate_inode_pages_range() altered to unmap the page if it got > > > mapped again? > > > > > > Oh. Because the unmap_mapping_range()

Re: [patch 3/6] mm: fix fault vs invalidate race for linear mappings

2007-03-06 Thread Andrew Morton
On Wed, 7 Mar 2007 07:57:27 +0100 Nick Piggin <[EMAIL PROTECTED]> wrote: > > > > Why was truncate_inode_pages_range() altered to unmap the page if it got > > mapped again? > > > > Oh. Because the unmap_mapping_range() call got removed from vmtruncate(). > > Why? (Please send suitable updates

Re: [patch 3/6] mm: fix fault vs invalidate race for linear mappings

2007-03-06 Thread Nick Piggin
On Tue, Mar 06, 2007 at 10:36:41PM -0800, Andrew Morton wrote: > On Wed, 21 Feb 2007 05:50:05 +0100 (CET) Nick Piggin <[EMAIL PROTECTED]> > wrote: > > > Fix the race between invalidate_inode_pages and do_no_page. > > > > Andrea Arcangeli identified a subtle race between invalidation of > >

Re: [patch 3/6] mm: fix fault vs invalidate race for linear mappings

2007-03-06 Thread Andrew Morton
On Wed, 21 Feb 2007 05:50:05 +0100 (CET) Nick Piggin <[EMAIL PROTECTED]> wrote: > Fix the race between invalidate_inode_pages and do_no_page. > > Andrea Arcangeli identified a subtle race between invalidation of > pages from pagecache with userspace mappings, and do_no_page. > > The issue is

Re: [patch 3/6] mm: fix fault vs invalidate race for linear mappings

2007-03-06 Thread Andrew Morton
On Wed, 21 Feb 2007 05:50:05 +0100 (CET) Nick Piggin [EMAIL PROTECTED] wrote: Fix the race between invalidate_inode_pages and do_no_page. Andrea Arcangeli identified a subtle race between invalidation of pages from pagecache with userspace mappings, and do_no_page. The issue is that

Re: [patch 3/6] mm: fix fault vs invalidate race for linear mappings

2007-03-06 Thread Nick Piggin
On Tue, Mar 06, 2007 at 10:36:41PM -0800, Andrew Morton wrote: On Wed, 21 Feb 2007 05:50:05 +0100 (CET) Nick Piggin [EMAIL PROTECTED] wrote: Fix the race between invalidate_inode_pages and do_no_page. Andrea Arcangeli identified a subtle race between invalidation of pages from

Re: [patch 3/6] mm: fix fault vs invalidate race for linear mappings

2007-03-06 Thread Andrew Morton
On Wed, 7 Mar 2007 07:57:27 +0100 Nick Piggin [EMAIL PROTECTED] wrote: Why was truncate_inode_pages_range() altered to unmap the page if it got mapped again? Oh. Because the unmap_mapping_range() call got removed from vmtruncate(). Why? (Please send suitable updates to the

Re: [patch 3/6] mm: fix fault vs invalidate race for linear mappings

2007-03-06 Thread Nick Piggin
On Tue, Mar 06, 2007 at 11:08:41PM -0800, Andrew Morton wrote: On Wed, 7 Mar 2007 07:57:27 +0100 Nick Piggin [EMAIL PROTECTED] wrote: Why was truncate_inode_pages_range() altered to unmap the page if it got mapped again? Oh. Because the unmap_mapping_range() call got removed

[patch 3/6] mm: fix fault vs invalidate race for linear mappings

2007-02-20 Thread Nick Piggin
Fix the race between invalidate_inode_pages and do_no_page. Andrea Arcangeli identified a subtle race between invalidation of pages from pagecache with userspace mappings, and do_no_page. The issue is that invalidation has to shoot down all mappings to the page, before it can be discarded from

[patch 3/6] mm: fix fault vs invalidate race for linear mappings

2007-02-20 Thread Nick Piggin
Fix the race between invalidate_inode_pages and do_no_page. Andrea Arcangeli identified a subtle race between invalidation of pages from pagecache with userspace mappings, and do_no_page. The issue is that invalidation has to shoot down all mappings to the page, before it can be discarded from