Christoph Lameter [Fri, 6 Jan 2006 08:11:19 +0000 (00:11 -0800)]
[PATCH] build_zonelists_node(): rename args
Give j and r meaningful names.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Christoph Lameter [Fri, 6 Jan 2006 08:11:18 +0000 (00:11 -0800)]
[PATCH] Fix zone policy determination
The use k in the inner loop means that the highest zone nr is always used
if any zone of a node is populated. This means that the policy zone is not
correctly determined on arches that do no use HIGHMEM like ia64.
Change the loop to decrement k which also simplifies the BUG_ON.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Christoph Lameter [Fri, 6 Jan 2006 08:11:17 +0000 (00:11 -0800)]
[PATCH] mm: move determination of policy_zone into page allocator
Currently the function to build a zonelist for a BIND policy has the side
effect to set the policy_zone. This seems to be a bit strange. policy
zone seems to not be initialized elsewhere and therefore 0. Do we police
ZONE_DMA if no bind policy has been used yet?
This patch moves the determination of the zone to apply policies to into
the page allocator. We determine the zone while building the zonelist for
nodes.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Christoph Lameter [Fri, 6 Jan 2006 08:11:16 +0000 (00:11 -0800)]
[PATCH] mm: simplify build_zonelists_node by removing the case statement.
Simplify build_zonelists_node by removing the case statement.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Con Kolivas [Fri, 6 Jan 2006 08:11:15 +0000 (00:11 -0800)]
[PATCH] mm: add populated_zone() helper
There are numerous places we check whether a zone is populated or not.
Provide a helper function to check for populated zones and convert all
checks for zone->present_pages.
Signed-off-by: Con Kolivas <kernel@kolivas.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Andrew Morton [Fri, 6 Jan 2006 08:11:14 +0000 (00:11 -0800)]
[PATCH] consolidate lru_add_drain() and lru_drain_cache()
Cc: Christoph Lameter <clameter@engr.sgi.com>
Cc: Rajesh Shah <rajesh.shah@intel.com>
Cc: Li Shaohua <shaohua.li@intel.com>
Cc: Srivatsa Vaddagiri <vatsa@in.ibm.com>
Cc: Ashok Raj <ashok.raj@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Andrew Morton [Fri, 6 Jan 2006 08:11:14 +0000 (00:11 -0800)]
[PATCH] vmscan: balancing fix
Revert a patch which went into 2.6.8-rc1. The changelog for that patch was:
The shrink_zone() logic can, under some circumstances, cause far too many
pages to be reclaimed. Say, we're scanning at high priority and suddenly
hit a large number of reclaimable pages on the LRU.
Change things so we bale out when SWAP_CLUSTER_MAX pages have been
reclaimed.
Problem is, this change caused significant imbalance in inter-zone scan
balancing by truncating scans of larger zones.
Suppose, for example, ZONE_HIGHMEM is 10x the size of ZONE_NORMAL. The zone
balancing algorithm would require that if we're scanning 100 pages of
ZONE_HIGHMEM, we should scan 10 pages of ZONE_NORMAL. But this logic will
cause the scanning of ZONE_HIGHMEM to bale out after only 32 pages are
reclaimed. Thus effectively causing smaller zones to be scanned relatively
harder than large ones.
Now I need to remember what the workload was which caused me to write this
patch originally, then fix it up in a different way...
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Nick Piggin [Fri, 6 Jan 2006 08:11:13 +0000 (00:11 -0800)]
[PATCH] mm: pfault optimisation
This atomic operation is superfluous: the pte will be added with the
referenced bit set, and the page will be referenced through this mapping after
the page fault handler returns anyway.
Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Nick Piggin [Fri, 6 Jan 2006 08:11:12 +0000 (00:11 -0800)]
[PATCH] mm: rmap optimisation
Optimise rmap functions by minimising atomic operations when we know there
will be no concurrent modifications.
Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Nick Piggin [Fri, 6 Jan 2006 08:11:11 +0000 (00:11 -0800)]
[PATCH] mm: bad_page optimisation
Cut down size slightly by not passing bad_page the function name (it should be
able to be determined by dump_stack()). And cut down the number of printks in
bad_page.
Also, cut down some branching in the destroy_compound_page path.
Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Nick Piggin [Fri, 6 Jan 2006 08:11:10 +0000 (00:11 -0800)]
[PATCH] mm: dma32 zone statistics
Add dma32 to zone statistics. Also attempt to arrange struct page_state a
bit better (visually).
Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: Andi Kleen <ak@muc.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Andrew Morton [Fri, 6 Jan 2006 08:11:09 +0000 (00:11 -0800)]
[PATCH] kill last zone_reclaim() bits
Remove the last bits of Martin's ill-fated sys_set_zone_reclaim().
Cc: Martin Hicks <mort@wildopensource.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Nikita Danilov [Fri, 6 Jan 2006 08:11:08 +0000 (00:11 -0800)]
[PATCH] find_lock_page(): call __lock_page() directly.
As find_lock_page() already checks with TestSetPageLocked() that page is
locked, there is no need to call lock_page() that will try-lock page again
(chances of page being unlocked in between are small). Call __lock_page()
directly, this saves one atomic operation.
Also, mark truncate-while-slept path as unlikely while we are here.
(akpm: ug. But this is actually a common path for normal old read()s against
a page which is under readahead I/O so ho-hum.)
Signed-off-by: Nikita Danilov <danilov@gmail.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
David Howells [Fri, 6 Jan 2006 08:11:08 +0000 (00:11 -0800)]
[PATCH] FRV: Clean up bootmem allocator's page freeing algorithm
The attached patch cleans up the way the bootmem allocator frees pages.
A new function, __free_pages_bootmem(), is provided in mm/page_alloc.c that is
called from mm/bootmem.c to turn pages over to the main allocator. All the
bits of code to initialise pages (clearing PG_reserved and setting the page
count) are moved to here. The checks on page validity are removed, on the
assumption that the struct page arrays will have been prepared correctly.
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Ravikiran G Thirumalai [Fri, 6 Jan 2006 08:11:01 +0000 (00:11 -0800)]
[PATCH] Cleanup bootmem allocator and fix alloc_bootmem_low
Patch cleans up the alloc_bootmem fix for swiotlb. Patch removes
alloc_bootmem_*_limit api and fixes alloc_boot_*low api to do the right
thing -- allocate from low32 memory.
Signed-off-by: Ravikiran Thirumalai <kiran@scalex86.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Nick Piggin [Fri, 6 Jan 2006 08:11:01 +0000 (00:11 -0800)]
[PATCH] mm: page_alloc cleanups
Small cleanups that does not change generated code with the gcc's I've tested
with.
Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Nick Piggin [Fri, 6 Jan 2006 08:11:00 +0000 (00:11 -0800)]
[PATCH] mm: page_state fixes
read_page_state and __get_page_state only traverse online CPUs, which will
cause results to fluctuate when CPUs are plugged in or out.
Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Nick Piggin [Fri, 6 Jan 2006 08:10:59 +0000 (00:10 -0800)]
[PATCH] mm: remove pcp low
struct per_cpu_pages.low is useless. Remove it.
Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Nick Piggin [Fri, 6 Jan 2006 08:10:58 +0000 (00:10 -0800)]
[PATCH] mm: remove bad_range
bad_range is supposed to be a temporary check. It would be a pity to throw it
out. Make it depend on CONFIG_DEBUG_VM instead.
CONFIG_HOLES_IN_ZONE systems were relying on this to check pfn_valid in the
page allocator. Add that to page_is_buddy instead.
Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Nick Piggin [Fri, 6 Jan 2006 08:10:57 +0000 (00:10 -0800)]
[PATCH] mm: microopt conditions
Micro optimise some conditionals where we don't need lazy evaluation.
Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Nick Piggin [Fri, 6 Jan 2006 08:10:57 +0000 (00:10 -0800)]
[PATCH] mm: set_page_refs opt
Inline set_page_refs.
Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Nick Piggin [Fri, 6 Jan 2006 08:10:56 +0000 (00:10 -0800)]
[PATCH] mm: pagealloc opt
Slightly optimise some page allocation and freeing functions by taking
advantage of knowing whether or not interrupts are disabled.
Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Hugh Dickins [Fri, 6 Jan 2006 08:10:55 +0000 (00:10 -0800)]
[PATCH] mm: free_pages_and_swap_cache opt
Minor optimization (though it doesn't help in the PREEMPT case, severely
constrained by small ZAP_BLOCK_SIZE). free_pages_and_swap_cache works in
chunks of 16, calling release_pages which works in chunks of PAGEVEC_SIZE.
But PAGEVEC_SIZE was dropped from 16 to 14 in 2.6.10, so we're now doing more
spin_lock_irq'ing than necessary: use PAGEVEC_SIZE throughout.
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Nick Piggin <nickpiggin@yahoo.com.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Andy Whitcroft [Fri, 6 Jan 2006 08:10:54 +0000 (00:10 -0800)]
[PATCH] sparsemem: provide pfn_to_nid
Before SPARSEMEM is initialised we cannot provide an efficient pfn_to_nid()
implmentation; before initialisation is complete we use early_pfn_to_nid()
to provide location information. Until recently there was no non-init user
of this functionality. Provide a post init pfn_to_nid() implementation.
Note that this implmentation assumes that the pfn passed has been validated
with pfn_valid(). The current single user of this function already has
this check.
Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Andy Whitcroft [Fri, 6 Jan 2006 08:10:53 +0000 (00:10 -0800)]
[PATCH] flatmem split out memory model
There are three places we define pfn_to_nid(). Two in linux/mmzone.h and one
in asm/mmzone.h. These in essence represent the three memory models. The
definition in linux/mmzone.h under !NEED_MULTIPLE_NODES is both the FLATMEM
definition and the optimisation for single NUMA nodes; the one under SPARSEMEM
is the NUMA sparsemem one; the one in asm/mmzone.h under DISCONTIGMEM is the
discontigmem one. This is not in the least bit obvious, particularly the
connection between the non-NUMA optimisations and the memory models.
Two patches:
flatmem-split-out-memory-model: simplifies the selection of pfn_to_nid()
implementations. The selection is based primarily off the memory model
selected. Optimisations for non-NUMA are applied where needed.
sparse-provide-pfn_to_nid: implement pfn_to_nid() for SPARSEMEM
This patch:
pfn_to_nid is memory model specific
The pfn_to_nid() call is memory model specific. It represents the locality
identifier for the memory passed. Classically this would be a NUMA node,
but not a chunk of memory under DISCONTIGMEM.
The SPARSEMEM and FLATMEM memory model non-NUMA versions of pfn_to_nid()
are folded together under NEED_MULTIPLE_NODES, while DISCONTIGMEM has its
own optimisation. This is all very confusing.
This patch splits out each implementation of pfn_to_nid() so that we can
see them and the optimisations to each.
Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Russell King [Fri, 6 Jan 2006 08:10:52 +0000 (00:10 -0800)]
[PATCH] Shut up warnings in ipc/shm.c
Fix two warnings in ipc/shm.c
ipc/shm.c:122: warning: statement with no effect
ipc/shm.c:560: warning: statement with no effect
by converting the macros to empty inline functions. For safety, let's do
all three. This also has the advantage that typechecking gets performed
even without CONFIG_SHMEM enabled.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Cc: Manfred Spraul <manfred@colorfullife.com>
Cc: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Mike Kravetz [Fri, 6 Jan 2006 08:10:51 +0000 (00:10 -0800)]
[PATCH] mm: remove arch independent NODES_SPAN_OTHER_NODES
The NODES_SPAN_OTHER_NODES config option was created so that DISCONTIGMEM
could handle pSeries numa layouts. However, support for DISCONTIGMEM has
been replaced by SPARSEMEM on powerpc. As a result, this config option and
supporting code is no longer needed.
I have already sent a patch to Paul that removes the option from powerpc
specific code. This removes the arch independent piece. Doesn't really
matter which is applied first.
Signed-off-by: Mike Kravetz <kravetz@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Andy Whitcroft [Fri, 6 Jan 2006 08:10:50 +0000 (00:10 -0800)]
[PATCH] mm: pfn_to_pgdat not used in common code
pfn_to_pgdat() isn't used in common code. Remove definition.
Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Andy Whitcroft [Fri, 6 Jan 2006 08:10:50 +0000 (00:10 -0800)]
[PATCH] mm: kvaddr_to_nid not used in common code
kvaddr_to_nid() isn't used in common code nor in i386 code. Remove these
definitions.
Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Christoph Lameter [Fri, 6 Jan 2006 08:10:49 +0000 (00:10 -0800)]
[PATCH] hugepages: fold find_or_alloc_pages into huge_no_page()
The number of parameters for find_or_alloc_page increases significantly after
policy support is added to huge pages. Simplify the code by folding
find_or_alloc_huge_page() into hugetlb_no_page().
Adam Litke objected to this piece in an earlier patch but I think this is a
good simplification. Diffstat shows that we can get rid of almost half of the
lines of find_or_alloc_page(). If we can find no consensus then lets simply
drop this patch.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Cc: Andi Kleen <ak@muc.de>
Acked-by: William Lee Irwin III <wli@holomorphy.com>
Cc: Adam Litke <agl@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Christoph Lameter [Fri, 6 Jan 2006 08:10:47 +0000 (00:10 -0800)]
[PATCH] Remove old node based policy interface from mempolicy.c
mempolicy.c contains provisional interface for huge page allocation based on
node numbers. This is in use in SLES9 but was never used (AFAIK) in upstream
versions of Linux.
Huge page allocations now use zonelists to figure out where to allocate pages.
The use of zonelists allows us to find the closest hugepage which was the
consideration of the NUMA distance for huge page allocations.
Remove the obsolete functions.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Cc: Andi Kleen <ak@muc.de>
Acked-by: William Lee Irwin III <wli@holomorphy.com>
Cc: Adam Litke <agl@us.ibm.com>
Acked-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Christoph Lameter [Fri, 6 Jan 2006 08:10:46 +0000 (00:10 -0800)]
[PATCH] Add NUMA policy support for huge pages.
The huge_zonelist() function in the memory policy layer provides an list of
zones ordered by NUMA distance. The hugetlb layer will walk that list looking
for a zone that has available huge pages but is also in the nodeset of the
current cpuset.
This patch does not contain the folding of find_or_alloc_huge_page() that was
controversial in the earlier discussion.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Cc: Andi Kleen <ak@muc.de>
Acked-by: William Lee Irwin III <wli@holomorphy.com>
Cc: Adam Litke <agl@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Christoph Lameter [Fri, 6 Jan 2006 08:10:45 +0000 (00:10 -0800)]
[PATCH] mm: dequeue a huge page near to this node
This was discussed at
http://marc.theaimsgroup.com/?l=linux-kernel&m=
113166526217117&w=2
This patch changes the dequeueing to select a huge page near the node
executing instead of always beginning to check for free nodes from node 0.
This will result in a placement of the huge pages near the executing
processor improving performance.
The existing implementation can place the huge pages far away from the
executing processor causing significant degradation of performance. The
search starting from zero also means that the lower zones quickly run out
of memory. Selecting a huge page near the process distributed the huge
pages better.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Cc: William Lee Irwin III <wli@holomorphy.com>
Cc: Adam Litke <agl@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
David Gibson [Fri, 6 Jan 2006 08:10:44 +0000 (00:10 -0800)]
[PATCH] Hugetlb: Copy on Write support
Implement copy-on-write support for hugetlb mappings so MAP_PRIVATE can be
supported. This helps us to safely use hugetlb pages in many more
applications. The patch makes the following changes. If needed, I also have
it broken out according to the following paragraphs.
1. Add a pair of functions to set/clear write access on huge ptes. The
writable check in make_huge_pte is moved out to the caller for use by COW
later.
2. Hugetlb copy-on-write requires special case handling in the following
situations:
- copy_hugetlb_page_range() - Copied pages must be write protected so
a COW fault will be triggered (if necessary) if those pages are written
to.
- find_or_alloc_huge_page() - Only MAP_SHARED pages are added to the
page cache. MAP_PRIVATE pages still need to be locked however.
3. Provide hugetlb_cow() and calls from hugetlb_fault() and
hugetlb_no_page() which handles the COW fault by making the actual copy.
4. Remove the check in hugetlbfs_file_map() so that MAP_PRIVATE mmaps
will be allowed. Make MAP_HUGETLB exempt from the depricated VM_RESERVED
mapping check.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Adam Litke <agl@us.ibm.com>
Cc: William Lee Irwin III <wli@holomorphy.com>
Cc: "Seth, Rohit" <rohit.seth@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Adam Litke [Fri, 6 Jan 2006 08:10:43 +0000 (00:10 -0800)]
[PATCH] Hugetlb: Reorganize hugetlb_fault to prepare for COW
This patch splits the "no_page()" type activity into its own function,
hugetlb_no_page(). hugetlb_fault() becomes the entry point for hugetlb faults
and delegates to the appropriate handler depending on the type of fault.
Right now we still have only hugetlb_no_page() but a later patch introduces a
COW fault.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Adam Litke <agl@us.ibm.com>
Cc: William Lee Irwin III <wli@holomorphy.com>
Cc: "Seth, Rohit" <rohit.seth@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Adam Litke [Fri, 6 Jan 2006 08:10:42 +0000 (00:10 -0800)]
[PATCH] Hugetlb: Rename find_lock_page to find_or_alloc_huge_page
find_lock_huge_page() isn't a great name, since it does extra things not
analagous to find_lock_page(). Rename it find_or_alloc_huge_page() which is
closer to the mark.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Adam Litke <agl@us.ibm.com>
Cc: William Lee Irwin III <wli@holomorphy.com>
Cc: "Seth, Rohit" <rohit.seth@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Adam Litke [Fri, 6 Jan 2006 08:10:40 +0000 (00:10 -0800)]
[PATCH] Hugetlb: Remove duplicate i_size check
cleanup
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Adam Litke <agl@us.ibm.com>
Cc: William Lee Irwin III <wli@holomorphy.com>
Cc: "Seth, Rohit" <rohit.seth@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Badari Pulavarty [Fri, 6 Jan 2006 08:10:38 +0000 (00:10 -0800)]
[PATCH] madvise(MADV_REMOVE): remove pages from tmpfs shm backing store
Here is the patch to implement madvise(MADV_REMOVE) - which frees up a
given range of pages & its associated backing store. Current
implementation supports only shmfs/tmpfs and other filesystems return
-ENOSYS.
"Some app allocates large tmpfs files, then when some task quits and some
client disconnect, some memory can be released. However the only way to
release tmpfs-swap is to MADV_REMOVE". - Andrea Arcangeli
Databases want to use this feature to drop a section of their bufferpool
(shared memory segments) - without writing back to disk/swap space.
This feature is also useful for supporting hot-plug memory on UML.
Concerns raised by Andrew Morton:
- "We have no plan for holepunching! If we _do_ have such a plan (or
might in the future) then what would the API look like? I think
sys_holepunch(fd, start, len), so we should start out with that."
- Using madvise is very weird, because people will ask "why do I need to
mmap my file before I can stick a hole in it?"
- None of the other madvise operations call into the filesystem in this
manner. A broad question is: is this capability an MM operation or a
filesytem operation? truncate, for example, is a filesystem operation
which sometimes has MM side-effects. madvise is an mm operation and with
this patch, it gains FS side-effects, only they're really, really
significant ones."
Comments:
- Andrea suggested the fs operation too but then it's more efficient to
have it as a mm operation with fs side effects, because they don't
immediatly know fd and physical offset of the range. It's possible to
fixup in userland and to use the fs operation but it's more expensive,
the vmas are already in the kernel and we can use them.
Short term plan & Future Direction:
- We seem to need this interface only for shmfs/tmpfs files in the short
term. We have to add hooks into the filesystem for correctness and
completeness. This is what this patch does.
- In the future, plan is to support both fs and mmap apis also. This
also involves (other) filesystem specific functions to be implemented.
- Current patch doesn't support VM_NONLINEAR - which can be addressed in
the future.
Signed-off-by: Badari Pulavarty <pbadari@us.ibm.com>
Cc: Hugh Dickins <hugh@veritas.com>
Cc: Andrea Arcangeli <andrea@suse.de>
Cc: Michael Kerrisk <mtk-manpages@gmx.net>
Cc: Ulrich Drepper <drepper@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Hans Reiser [Fri, 6 Jan 2006 08:10:36 +0000 (00:10 -0800)]
[PATCH] reiser4: vfs: add truncate_inode_pages_range()
This patch makes truncate_inode_pages_range from truncate_inode_pages.
truncate_inode_pages became a one-liner call to truncate_inode_pages_range.
Reiser4 needs truncate_inode_pages_ranges because it tries to keep
correspondence between existences of metadata pointing to data pages and pages
to which those metadata point to. So, when metadata of certain part of file
is removed from filesystem tree, only pages of corresponding range are to be
truncated.
(Needed by the madvise(MADV_REMOVE) patch)
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Andy Whitcroft [Fri, 6 Jan 2006 08:10:35 +0000 (00:10 -0800)]
[PATCH] memhotplug: register_memory should be global
register_memory is global and declared so in linux/memory.h. Update the
HOTPLUG specific definition to match. This fixes a compile warning when
HOTPLUG is enabled.
Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Andy Whitcroft [Fri, 6 Jan 2006 08:10:35 +0000 (00:10 -0800)]
[PATCH] memhotplug: register_ and unregister_memory_notifier should be global
Both register_memory_notifer and unregister_memory_notifier are global and
declared so in linux/memory.h. Update the HOTPLUG specific definitions to
match. This fixes a compile warning when HOTPLUG is enabled.
Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Andy Whitcroft [Fri, 6 Jan 2006 08:10:33 +0000 (00:10 -0800)]
[PATCH] memhotplug: __add_section remove unused pgdat definition
__add_section defines an unused pointer to the zones pgdat. Remove this
definition. This fixes a compile warning.
Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Paul Jackson [Fri, 6 Jan 2006 08:10:32 +0000 (00:10 -0800)]
[PATCH] mm: fix __alloc_pages cpuset ALLOC_* flags
Two changes to the setting of the ALLOC_CPUSET flag in
mm/page_alloc.c:__alloc_pages()
- A bug fix - the "ignoring mins" case should not be honoring ALLOC_CPUSET.
This case of all cases, since it is handling a request that will free up
more memory than is asked for (exiting tasks, e.g.) should be allowed to
escape cpuset constraints when memory is tight.
- A logic change to make it simpler. Honor cpusets even on GFP_ATOMIC
(!wait) requests. With this, cpuset confinement applies to all requests
except ALLOC_NO_WATERMARKS, so that in a subsequent cleanup patch, I can
remove the ALLOC_CPUSET flag entirely. Since I don't know any real reason
this logic has to be either way, I am choosing the path of the simplest
code.
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Andrew Morton [Fri, 6 Jan 2006 08:09:50 +0000 (00:09 -0800)]
[PATCH] swsusp: resume_store() retval fix
- This function returns -EINVAL all the time. Fix.
- Decruftify it a bit too.
- Writing to it doesn't seem to do what it's suppoed to do.
Cc: Pavel Machek <pavel@ucw.cz>
Cc: "Rafael J. Wysocki" <rjw@sisk.pl>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Andrew Morton [Fri, 6 Jan 2006 08:09:50 +0000 (00:09 -0800)]
[PATCH] alpha: dma_map_page() fix
Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Cc: Richard Henderson <rth@twiddle.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
NeilBrown [Fri, 6 Jan 2006 08:09:49 +0000 (00:09 -0800)]
[PATCH] knfsd: fix hash function for IP addresses on 64bit little-endian machines.
The hash.h hash_long function, when used on a 64 bit machine, ignores many
of the middle-order bits. (The prime chosen it too bit-sparse).
IP addresses for clients of an NFS server are very likely to differ only in
the low-order bits. As addresses are stored in network-byte-order, these
bits become middle-order bits in a little-endian 64bit 'long', and so do
not contribute to the hash. Thus you can have the situation where all
clients appear on one hash chain.
So, until hash_long is fixed (or maybe forever), us a hash function that
works well on IP addresses - xor the bytes together.
Thanks to "Iozone" <capps@iozone.org> for identifying this problem.
Cc: "Iozone" <capps@iozone.org>
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Herbert Xu [Fri, 6 Jan 2006 08:09:47 +0000 (00:09 -0800)]
[PATCH] nbd: fix TX/RX race condition
Janos Haar of First NetCenter Bt. reported numerous crashes involving the
NBD driver. With his help, this was tracked down to bogus bio vectors
which in turn was the result of a race condition between the
receive/transmit routines in the NBD driver.
The bug manifests itself like this:
CPU0 CPU1
do_nbd_request
add req to queuelist
nbd_send_request
send req head
for each bio
kmap
send
nbd_read_stat
nbd_find_request
nbd_end_request
kunmap
When CPU1 finishes nbd_end_request, the request and all its associated
bio's are freed. So when CPU0 calls kunmap whose argument is derived from
the last bio, it may crash.
Under normal circumstances, the race occurs only on the last bio. However,
if an error is encountered on the remote NBD server (such as an incorrect
magic number in the request), or if there were a bug in the server, it is
possible for the nbd_end_request to occur any time after the request's
addition to the queuelist.
The following patch fixes this problem by making sure that requests are not
added to the queuelist until after they have been completed transmission.
In order for the receiving side to be ready for responses involving
requests still being transmitted, the patch introduces the concept of the
active request.
When a response matches the current active request, its processing is
delayed until after the tranmission has come to a stop.
This has been tested by Janos and it has been successful in curing this
race condition.
From: Herbert Xu <herbert@gondor.apana.org.au>
Here is an updated patch which removes the active_req wait in
nbd_clear_queue and the associated memory barrier.
I've also clarified this in the comment.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Cc: <djani22@dynamicweb.hu>
Cc: Paul Clements <Paul.Clements@SteelEye.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Joshua Kwan [Fri, 6 Jan 2006 08:09:45 +0000 (00:09 -0800)]
[PATCH] hfsplus oops fix
nls_utf8 is available, and the check in hfsplus_fill_super checks the wrong
pointer for NULLness (it checks the saved nls, not the new one that it
needs to use.)
Signed-off-by: Joshua Kwan <joshk@triplehelix.org>
Cc: Roman Zippel <zippel@linux-m68k.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Chuck Ebbert [Fri, 6 Jan 2006 04:11:29 +0000 (23:11 -0500)]
[PATCH] i386: PTRACE_POKEUSR: allow changing RF bit in EFLAGS register.
Setting RF (resume flag) allows a debugger to resume execution after a
code breakpoint without tripping the breakpoint again. It is reset by
the CPU after execution of one instruction.
Requested by Stephane Eranian:
"I am trying to the user HW debug registers on i386 and I am running
into a problem with ptrace() not allowing access to EFLAGS_RF for
POKEUSER (see FLAG_MASK). [ ... ] It avoids the need to remove the
breakpoint, single step, and reinstall. The equivalent functionality
exists on IA-64 and is allowed by ptrace()"
Cc: Stephane Eranian <eranian@hpl.hp.com>
Signed-off-by: Chuck Ebbert <76306.1226@compuserve.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Linus Torvalds [Fri, 6 Jan 2006 04:43:11 +0000 (20:43 -0800)]
Merge http://oss.oracle.com/git/ocfs2
Jody McIntyre [Fri, 6 Jan 2006 04:04:08 +0000 (23:04 -0500)]
Update MAINTAINERS - Jody is no longer at Steamballoon.
Jody McIntyre [Fri, 6 Jan 2006 03:22:50 +0000 (22:22 -0500)]
Merge ... /linux/kernel/git/torvalds/linux-2.6.git
Kris Katterjohn [Fri, 6 Jan 2006 00:35:42 +0000 (16:35 -0800)]
[NET]: Change 1500 to ETH_DATA_LEN in some files
These patches add the header linux/if_ether.h and change 1500 to
ETH_DATA_LEN in some files.
Signed-off-by: Kris Katterjohn <kjak@users.sourceforge.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Andrew Morton [Thu, 5 Jan 2006 22:57:36 +0000 (14:57 -0800)]
[IPVS]: Another file needs linux/interrupt.h
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Linus Torvalds [Thu, 5 Jan 2006 23:55:49 +0000 (15:55 -0800)]
Merge git://git./linux/kernel/git/brodo/pcmcia-2.6
Richard Purdie [Thu, 5 Jan 2006 09:56:03 +0000 (09:56 +0000)]
[PATCH] pcmcia: add some IDs for ide-cs and dtl1_cs
Add some PCMCIA device IDs for the microdrive found in the Sharp Zaurus
and a different revision of the Socket CF+ Bluetooth card.
Signed-off-by: Richard Purdie <rpurdie@rpsys.net>
Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
Daniel Ritz [Fri, 30 Dec 2005 14:12:35 +0000 (15:12 +0100)]
[PATCH] pcmcia: cleanup cs.c, reduce size
kill the socket_shutdown()/shutdown_socket() confusion by making it
one single function. move cs_socket_put() in there. nicer to read and
smaller:
original:
text data bss dec hex filename
25181 1076 32 26289 66b1 drivers/pcmcia/pcmcia_core.ko
patched:
text data bss dec hex filename
24973 1076 32 26081 65e1 drivers/pcmcia/pcmcia_core.ko
Signed-off-by: Daniel Ritz <daniel.ritz@gmx.ch>
Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
Alexey Dobriyan [Fri, 23 Dec 2005 20:51:08 +0000 (23:51 +0300)]
[PATCH] drivers/pcmcia/cistpl.c: fix endian warnings
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
Dominik Brodowski [Sun, 11 Dec 2005 20:18:26 +0000 (21:18 +0100)]
[PATCH] pcmcia: kzalloc conversion
Convert users of kmalloc and memset to kzalloc
Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
Dominik Brodowski [Sun, 11 Dec 2005 19:47:44 +0000 (20:47 +0100)]
[PATCH] pcmcia: no probing of ioports on PARISC
Do not wildly probe the IO ports we're trying to use on PARISC.
Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
Dominik Brodowski [Thu, 8 Dec 2005 22:50:36 +0000 (23:50 +0100)]
[PATCH] pcmcia: export stored values in sysfs
Export the stored values instead of re-reading everything in the socket
information sysfs files, and make them accessible to all users, not only
to root.
Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
Vitaly Bordug [Thu, 8 Dec 2005 15:56:12 +0000 (13:56 -0200)]
[PATCH] 8xx PCMCIA: support for MPC885ADS and MPC866ADS
This adds PCMCIA support for both MPC885ADS and MPC866ADS.
This is established not together with FADS, because 885 does not have
io_block_mapping() for BCSR area.
Also, some cleanups done both for 885ADS and MBX.
Signed-off-by: Vitaly Bordug <vbordug@ru.mvista.com>
Signed-off-by: Marcelo Tosatti <marcelo.tosatti@cyclades.com>
Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
Vitaly Bordug [Thu, 8 Dec 2005 15:53:20 +0000 (13:53 -0200)]
[PATCH] m8xx_pcmcia: support MAP_AUTOSZ required for CF cards
This fixes misconfiguration that could result in odd work of some old CF
cards.
Signed-off-by: Vitaly Bordug <vbordug@ru.mvista.com>
Signed-off-by: Marcelo Tosatti <marcelo.tosatti@cyclades.com>
Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
Dominik Brodowski [Wed, 7 Dec 2005 11:32:20 +0000 (12:32 +0100)]
[PATCH] pcmcia: properly handle static mem, but dynamic io sockets
Some PCMCIA sockets have statically mapped memory windows, but dynamically
mapped IO windows. Using the "nonstatic" socket library is inpractical for
them, as they do neither need a resource database (as we can trust the
kernel resource database on m68k and ppc) nor lots of other features of that
library. Let them get a small "iodyn" socket library (105 lines of code)
instead.
Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
Dominik Brodowski [Thu, 5 Jan 2006 23:27:16 +0000 (00:27 +0100)]
[PATCH] pcmcia: fix sound drivers
Update the PCMCIA sound drivers to handle the recent changes to the PCMCIA
core. A part of this merge was done by Takashi Iwai <tiwai@suse.de>.
Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
Dominik Brodowski [Mon, 14 Nov 2005 20:25:51 +0000 (21:25 +0100)]
[PATCH] pcmcia: unify attach, EVENT_CARD_INSERTION handlers into one probe callback
Unify the EVENT_CARD_INSERTION and "attach" callbacks to one unified
probe() callback. As all in-kernel drivers are changed to this new
callback, there will be no temporary backwards-compatibility. Inside a
probe() function, each driver _must_ set struct pcmcia_device
*p_dev->instance and instance->handle correctly.
With these patches, the basic driver interface for 16-bit PCMCIA drivers
now has the classic four callbacks known also from other buses:
int (*probe) (struct pcmcia_device *dev);
void (*remove) (struct pcmcia_device *dev);
int (*suspend) (struct pcmcia_device *dev);
int (*resume) (struct pcmcia_device *dev);
Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
Dominik Brodowski [Mon, 14 Nov 2005 20:25:35 +0000 (21:25 +0100)]
[PATCH] pcmcia: remove dev_list from drivers
The linked list of devices managed by each PCMCIA driver is, in very most
cases, unused. Therefore, remove it from many drivers.
Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
Dominik Brodowski [Mon, 14 Nov 2005 20:25:23 +0000 (21:25 +0100)]
[PATCH] pcmcia: remove old detach mechanism
Remove the old "detach" mechanism as it is unused now.
Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
Dominik Brodowski [Mon, 14 Nov 2005 20:23:14 +0000 (21:23 +0100)]
[PATCH] pcmcia: unify detach, REMOVAL_EVENT handlers into one remove callback
Unify the "detach" and REMOVAL_EVENT handlers to one "remove" function.
Old functionality is preserved, for the moment.
Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
Dominik Brodowski [Thu, 5 Jan 2006 23:02:03 +0000 (00:02 +0100)]
[PATCH] pcmcia: merge suspend into device model
Merge the suspend and resume methods for 16-bit PCMCIA cards into the
device model -- for both runtime power management and suspend to ram/disk.
Bugfix in ds.c by Richard Purdie
Signed-Off-By: Richard Purdie <rpurdie@rpsys.net>
Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
Dominik Brodowski [Mon, 14 Nov 2005 20:21:18 +0000 (21:21 +0100)]
[PATCH] pcmcia: new suspend core
Move the suspend and resume methods out of the event handler, and into
special functions. Also use these functions for pre- and post-reset, as
almost all drivers already do, and the remaining ones can easily be
converted.
Bugfix to include/pcmcia/ds.c
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
Daniel Ritz [Thu, 3 Nov 2005 20:12:14 +0000 (21:12 +0100)]
[PATCH] yenta: make bridge specific init code configurable
Make the bridge specific initialization code config options depending on
CONFIG_EMBEDDED. Config options for TI/EnE, Toshiba, Ricoh and O2Micro are
available. Disabling all of the specific tweaks cuts off more than half
of yenta_socket.ko.
Signed-off-by: Daniel Ritz <daniel.ritz@gmx.ch>
Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
Andrew Morton [Thu, 17 Nov 2005 05:29:26 +0000 (21:29 -0800)]
[PATCH] pcmcia: validate_mem fix
Also return a value if CONFIG_PCMCIA_PROBE is not set.
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
Dominik Brodowski [Wed, 28 Sep 2005 17:41:56 +0000 (19:41 +0200)]
[PATCH] pcmcia: validate_mem shouldn't be void
Add a return value to pcmcia_validate_mem. Only if we have enough memory
available to map the CIS, we should proceed in trying to determine information
about the device.
Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
Dominik Brodowski [Sat, 12 Nov 2005 22:56:33 +0000 (23:56 +0100)]
[PATCH] pcmcia: remove get_socket callback
The .get_socket callback is never used by the PCMCIA core, therefore remove
it.
Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
Dominik Brodowski [Wed, 28 Sep 2005 17:41:55 +0000 (19:41 +0200)]
[PATCH] pcmcia: remove socket register_callback
Remove the register_callback declaration in struct pccard_operations as it is
unused.
Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
Daniel Ritz [Tue, 20 Sep 2005 21:17:37 +0000 (14:17 -0700)]
[PATCH] yenta: optimize interrupt handler
Don't waste cpu time in yenta interrupt handler when the interrupt was for
another device.
Signed-off-by: Daniel Ritz <daniel.ritz@gmx.ch>
Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
Pavel Machek [Wed, 7 Sep 2005 23:00:26 +0000 (16:00 -0700)]
[PATCH] pcmcia: avoid macro usage in cistpl
Fix macro abuse in pcmcia.
Signed-off-by: Pavel Machek <pavel@suse.cz>
Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
Linus Torvalds [Thu, 5 Jan 2006 22:18:15 +0000 (14:18 -0800)]
Merge branch 'release' of git://git./linux/kernel/git/aegl/linux-2.6
Linus Torvalds [Thu, 5 Jan 2006 22:01:23 +0000 (14:01 -0800)]
Merge /pub/scm/linux/kernel/git/davem/sparc-2.6
Tony Luck [Thu, 5 Jan 2006 21:30:52 +0000 (13:30 -0800)]
[IA64] Fix compile warnings in setup.c
arch/ia64/kernel/setup.c: In function `show_cpuinfo':
arch/ia64/kernel/setup.c:576: warning: long unsigned int format, different type arg (arg 12)
arch/ia64/kernel/setup.c:576: warning: long unsigned int format, different type arg (arg 13)
Introduced by
95235ca2c20ac0b31a8eb39e2d599bcc3e9c9a10
Signed-off-by: Tony Luck <tony.luck@intel.com>
Luis F. Ortiz [Thu, 5 Jan 2006 21:12:41 +0000 (13:12 -0800)]
[ATYFB]: Fix onboard video on SPARC Blade 100 for 2.6.{13,14,15}
I have recently been switching from using 2.4.32 on my trusty
old Sparc Blade 100 to using 2.6.15 . Some of the problems I ran into
were distorted video when the console was active (missing first
character, skipped dots) and when running X windows (colored snow,
stripes, missing pixels). A quick examination of the 2.6 versus 2.4
source for the ATY driver revealed alot of changes.
A closer look at the code/data for the 64GR/XL chip revealed
two minor "typos" that the rewriter(s) of the code made. The first is
a incorrect clock value (230 .vs. 235) and the second is a missing
flag (M64F_SDRAM_MAGIC_PLL). Making both these changes seems to have
fixed my problem. I tend to think the 235 value is the correct one,
as there is a 29.4 Mhz clock crystal close to the video chip and 235.2
(29.4*8) is too close to 235 to make it a coincidence.
The flag for M64F_SDRAM_MAGIC_PLL was dropped during the
changes made by adaplas in file revision 1.72 on the old bitkeeper
repository.
The change relating to the clock rate has been there forever,
at least in the 2.6 tree. I'm not sure where to look for the old 2.5
tree or if anyone cares when it happened.
On SPARC Blades 100's, which use the ATY MACH64GR video chipset, the
clock crystal frequency is 235.2 Mhz, not 230 Mhz. The chipset also
requires the use of M64F_SDRAM_MAGIC_PLL in order to setup the PLL
properly for the DRAM.
Signed-off-by: Luis F. Ortiz <lfo@Polyad.Org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Yasuyuki Kozakai [Thu, 5 Jan 2006 20:28:57 +0000 (12:28 -0800)]
[NETFILTER]: Use HOPLIMIT metric as TTL of TCP reset sent by REJECT
HOPLIMIT metric is appropriate to TCP reset sent by REJECT target
than hard-coded max TTL. Thanks to David S. Miller for hint.
Signed-off-by: Yasuyuki Kozakai <yasuyuki.kozakai@toshiba.co.jp>
Signed-off-by: David S. Miller <davem@davemloft.net>
Patrick McHardy [Thu, 5 Jan 2006 20:21:52 +0000 (12:21 -0800)]
[NETFILTER]: nf_conntrack_l3proto_ipv4.c needs net/route.h
CC [M] net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.o
net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c: In function 'ipv4_refrag':
net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c:198: error: dereferencing pointer to incomplete type
make[3]: *** [net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.o] Error 1
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Patrick McHardy [Thu, 5 Jan 2006 20:21:34 +0000 (12:21 -0800)]
[NETFILTER]: Export ip6_masked_addrcmp, don't pass IPv6 addresses on stack
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Patrick McHardy [Thu, 5 Jan 2006 20:21:16 +0000 (12:21 -0800)]
[NETFILTER]: make ipv6_find_hdr() find transport protocol header
The original ipv6_find_hdr() finds the specified header in IPv6 packets.
This makes it possible to get transport header so that we can kill similar
loop in ip6_match_packet().
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Patrick McHardy [Thu, 5 Jan 2006 20:20:59 +0000 (12:20 -0800)]
[NETFILTER]: Call POST_ROUTING hook before fragmentation
Call POST_ROUTING hook before fragmentation to get rid of the okfn use
in ip_refrag and save the useless fragmentation/defragmentation step
when NAT is used.
The patch introduces one user-visible change, the POSTROUTING chain
in the mangle table gets entire packets, not fragments, which should
simplify use of the MARK and CLASSIFY targets for queueing as a nice
side-effect.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Patrick McHardy [Thu, 5 Jan 2006 20:20:40 +0000 (12:20 -0800)]
[NETFILTER]: Remove okfn usage in ip_vs_core.c
okfn should only be used from different contexts to avoid deep call chains,
i.e. by nf_queue.
Acked-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Patrick McHardy [Thu, 5 Jan 2006 20:20:02 +0000 (12:20 -0800)]
[NETFILTER]: ctnetlink: Fix dumping of helper name
Properly dump the helper name instead of internal kernel data.
Based on patch by Marcus Sundberg <marcus@ingate.com>.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Patrick McHardy [Thu, 5 Jan 2006 20:19:46 +0000 (12:19 -0800)]
[NETFILTER]: Fix module_param types and permissions
Fix netfilter module_param types and permissions. Also fix an off-by-one in
the ipt_ULOG nlbufsiz < 128k check.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Pablo Neira Ayuso [Thu, 5 Jan 2006 20:19:23 +0000 (12:19 -0800)]
[NETFILTER]: Filter dumped entries based on the layer 3 protocol number
Dump entries of a given Layer 3 protocol number.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Pablo Neira Ayuso [Thu, 5 Jan 2006 20:19:05 +0000 (12:19 -0800)]
[NETFILTER]: Add ctnetlink port for nf_conntrack
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Pablo Neira Ayuso [Thu, 5 Jan 2006 20:18:44 +0000 (12:18 -0800)]
[NETFILTER]: ctnetlink: remove unused variable
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Pablo Neira Ayuso [Thu, 5 Jan 2006 20:18:25 +0000 (12:18 -0800)]
[NETFILTER]: ctnetlink: fix conntrack mark race
Set conntrack mark before it is in hashes.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Pablo Neira Ayuso [Thu, 5 Jan 2006 20:18:08 +0000 (12:18 -0800)]
[NETFILTER]: ctnetlink: ctnetlink_event cleanup
Cleanup: Use 'else if' instead of a ugly 'goto' statement.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Pablo Neira Ayuso [Thu, 5 Jan 2006 20:17:50 +0000 (12:17 -0800)]
[NETFILTER]: ctnetlink: use u_int32_t instead of unsigned int
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Pablo Neira Ayuso [Thu, 5 Jan 2006 20:17:29 +0000 (12:17 -0800)]
[NETFILTER]: ctnetlink: propagate ctnetlink_dump_tuples_proto return value back
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Yasuyuki Kozakai [Thu, 5 Jan 2006 20:17:03 +0000 (12:17 -0800)]
[NETFILTER]: ctnetlink: Add sanity checkings for ICMP
Signed-off-by: Yasuyuki Kozakai <yasuyuki.kozakai@toshiba.co.jp>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Pablo Neira Ayuso [Thu, 5 Jan 2006 20:16:41 +0000 (12:16 -0800)]
[NETFILTER]: ctnetlink: remove bogus checks in ICMP protocol at dumping
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Jesper Juhl [Thu, 5 Jan 2006 20:16:16 +0000 (12:16 -0800)]
[NETFILTER]: Decrease number of pointer derefs in nf_conntrack_core.c
Benefits of the patch:
- Fewer pointer dereferences should make the code slightly faster.
- Size of generated code is smaller
- improved readability
Signed-off-by: Jesper Juhl <jesper.juhl@gmail.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>