cvs commit: src/sys/vm vm_contig.c vm_pageout.c vm_pageout.h

2008-09-17 Thread John Baldwin
jhb 2008-09-17 20:30:42 UTC FreeBSD src repository Modified files:(Branch: RELENG_7) sys/vm vm_contig.c vm_pageout.c vm_pageout.h Log: SVN rev 183124 on 2008-09-17 20:30:42Z by jhb MFC: 173901,173918 - Tidy up: Add comments. Eliminate the

cvs commit: src/sys/vm vm_contig.c vm_pageout.c vm_pageout.h

2007-11-25 Thread Alan Cox
alc 2007-11-25 20:37:30 UTC FreeBSD src repository Modified files: sys/vm vm_contig.c vm_pageout.c vm_pageout.h Log: Make contigmalloc(9)'s page laundering more robust. Specifically, use vm_pageout_fallback_object_lock() in vm_contig_launder_page() to better

cvs commit: src/sys/vm vm_contig.c

2007-11-24 Thread Alan Cox
alc 2007-11-25 07:42:34 UTC FreeBSD src repository Modified files: sys/vm vm_contig.c Log: Tidy up: Add comments. Eliminate the pointless malloc_type_allocated(..., 0) calls that occur when contigmalloc() has failed. Eliminate the acquisition and release

cvs commit: src/sys/vm vm_contig.c

2007-06-10 Thread Alan Cox
alc 2007-06-11 03:20:16 UTC FreeBSD src repository Modified files: sys/vm vm_contig.c Log: Conditionally acquire Giant in vm_contig_launder_page(). Revision ChangesPath 1.61 +4 -0 src/sys/vm/vm_contig.c

cvs commit: src/sys/vm vm_contig.c

2007-04-28 Thread Alan Cox
alc 2007-04-28 20:44:57 UTC FreeBSD src repository Modified files:(Branch: RELENG_6) sys/vm vm_contig.c Log: MFC revision 1.58 Correct contigmalloc2()'s implementation of M_ZERO. Revision ChangesPath 1.43.2.6 +1 -1

cvs commit: src/sys/vm vm_contig.c

2007-04-19 Thread Alan Cox
alc 2007-04-19 05:39:54 UTC FreeBSD src repository Modified files: sys/vm vm_contig.c Log: Correct contigmalloc2()'s implementation of M_ZERO. Specifically, contigmalloc2() was always testing the first physical page for PG_ZERO, not the current page of

Re: cvs commit: src/sys/vm vm_contig.c vm_page.c vm_pageq.c vm_zeroidle.c

2007-02-06 Thread Robert Watson
On Mon, 5 Feb 2007, Alan Cox wrote: alc 2007-02-05 06:02:55 UTC FreeBSD src repository Modified files: sys/vm vm_contig.c vm_page.c vm_pageq.c vm_zeroidle.c Log: Change the free page queue lock from a spin mutex to a default (blocking)

Re: cvs commit: src/sys/vm vm_contig.c vm_page.c vm_pageq.c vm_zeroidle.c

2007-02-06 Thread Alan Cox
Robert Watson wrote: On Mon, 5 Feb 2007, Alan Cox wrote: alc 2007-02-05 06:02:55 UTC FreeBSD src repository Modified files: sys/vm vm_contig.c vm_page.c vm_pageq.c vm_zeroidle.c Log: Change the free page queue lock from a spin mutex to a

cvs commit: src/sys/vm vm_contig.c vm_page.c vm_pageq.c vm_zeroidle.c

2007-02-04 Thread Alan Cox
alc 2007-02-05 06:02:55 UTC FreeBSD src repository Modified files: sys/vm vm_contig.c vm_page.c vm_pageq.c vm_zeroidle.c Log: Change the free page queue lock from a spin mutex to a default (blocking) mutex. With the demise of Alpha

cvs commit: src/sys/vm vm_contig.c

2006-11-07 Thread Alan Cox
alc 2006-11-08 06:23:29 UTC FreeBSD src repository Modified files: sys/vm vm_contig.c Log: Ensure that the page's oflags field is initialized by contigmalloc(). Revision ChangesPath 1.56 +1 -0 src/sys/vm/vm_contig.c

cvs commit: src/sys/vm vm_contig.c

2006-10-11 Thread Kip Macy
kmacy 2006-10-12 04:41:39 UTC FreeBSD src repository Modified files: sys/vm vm_contig.c Log: sun4v requires TSBs (translation storage buffers) to be contiguous and be size aligned requiring heavy usage of vm_page_alloc_contig This change makes

cvs commit: src/sys/vm vm_contig.c

2006-09-05 Thread Alan Cox
alc 2006-09-05 23:10:59 UTC FreeBSD src repository Modified files:(Branch: RELENG_6) sys/vm vm_contig.c Log: MFC revision 1.52 Prevent a call to contigmalloc() that asks for more physical memory than the machine has from causing a panic.

cvs commit: src/sys/vm vm_contig.c vm_page.h

2006-09-03 Thread Alan Cox
alc 2006-09-03 22:24:09 UTC FreeBSD src repository Modified files: sys/vm vm_contig.c vm_page.h Log: Make vm_page_release_contig() static. Revision ChangesPath 1.53 +1 -1 src/sys/vm/vm_contig.c 1.144 +0 -1 src/sys/vm/vm_page.h

cvs commit: src/sys/vm vm_contig.c

2006-08-25 Thread Alan Cox
alc 2006-08-26 02:43:23 UTC FreeBSD src repository Modified files: sys/vm vm_contig.c Log: Prevent a call to contigmalloc() that asks for more physical memory than the machine has from causing a panic. Submitted by: Michael Plass PR: 101668 MFC

cvs commit: src/sys/vm vm_contig.c

2006-04-25 Thread Scott Long
scottl 2006-04-25 15:04:30 UTC FreeBSD src repository Modified files:(Branch: RELENG_6) sys/vm vm_contig.c Log: Don't scan pages that are above the passed in high value. This is a partial MFC of rev 1.48; the rest of that change is incomplete and not

cvs commit: src/sys/vm vm_contig.c

2006-04-25 Thread Scott Long
scottl 2006-04-25 15:29:50 UTC FreeBSD src repository Modified files:(Branch: RELENG_6_1) sys/vm vm_contig.c Log: MFC rev 1.43.2.4: avoid high ordered pages that are outside of the requested range. Approved by: re Revision ChangesPath

cvs commit: src/sys/vm vm_contig.c

2006-03-07 Thread Tor Egge
tegge 2006-03-08 00:51:01 UTC FreeBSD src repository Modified files: sys/vm vm_contig.c Log: Ignore dirty pages owned by dead objects. Revision ChangesPath 1.51 +4 -0 src/sys/vm/vm_contig.c ___

cvs commit: src/sys/vm vm_contig.c

2006-03-02 Thread Tor Egge
tegge 2006-03-02 21:38:38 UTC FreeBSD src repository Modified files: sys/vm vm_contig.c Log: Hold extra reference to vm object while cleaning pages. Revision ChangesPath 1.49 +2 -0 src/sys/vm/vm_contig.c

cvs commit: src/sys/vm vm_contig.c

2006-01-29 Thread Scott Long
scottl 2006-01-29 08:24:54 UTC FreeBSD src repository Modified files: sys/vm vm_contig.c Log: The change a few years ago of having contigmalloc start its scan at the top of physical RAM instead of the bottom was a sound idea, but the implementation left a lot

Re: cvs commit: src/sys/vm vm_contig.c

2006-01-29 Thread Nate Lawson
Scott Long wrote: scottl 2006-01-29 08:24:54 UTC FreeBSD src repository Modified files: sys/vm vm_contig.c Log: The change a few years ago of having contigmalloc start its scan at the top of physical RAM instead of the bottom was a sound idea, but the

Re: cvs commit: src/sys/vm vm_contig.c

2006-01-29 Thread Scott Long
Nate Lawson wrote: Scott Long wrote: scottl 2006-01-29 08:24:54 UTC FreeBSD src repository Modified files: sys/vm vm_contig.c Log: The change a few years ago of having contigmalloc start its scan at the top of physical RAM instead of the bottom was a sound

cvs commit: src/sys/vm vm_contig.c

2006-01-25 Thread Alan Cox
alc 2006-01-26 05:51:26 UTC FreeBSD src repository Modified files: sys/vm vm_contig.c Log: Plug a leak in the newer contigmalloc() implementation. Specifically, if a multipage allocation was aborted midway, the pages that were already allocated were not