- 09 Jan, 2006 40 commits
-
-
David Howells authored
Fix a number of miscellanous items: (1) Declare lock sections in the linker script. (2) Recurse in the correct manner in the arch makefile. (3) asm/bug.h requires asm/linkage.h to be included first. One C file puts asm/bug.h first. (4) Add an empty RTC header file to avoid missing header file errors. (5) sg_dma_address() should use the dma_address member of a scatter list. (6) Add trivial pci_unmap support. (7) Add pgprot_noncached() (8) Discard u_quad_t. (9) Use ~0UL rather than ULONG_MAX in unistd.h in case the latter isn't declared. (10) Add an empty VGA header file to avoid missing header file errors. (11) Add an XOR header file to use the generic XOR stuff. Signed-off-by:David Howells <dhowells@redhat.com> Signed-off-by:
Andrew Morton <akpm@osdl.org> Signed-off-by:
Linus Torvalds <torvalds@osdl.org>
-
David Howells authored
Make the get_user macro cast the source pointer to an appropriate type for the specified size. Signed-off-by:
-
David Howells authored
Force the 8230 serial driver to be built in if the on-CPU UARTs are to be used. It can't be used as a module because the arch setup needs to call into it. Signed-off-by:
-
David Howells authored
Fix PCMCIA configuration for FRV by including the stock PCMCIA configuration description file. Signed-off-by:
-
David Howells authored
Implement pci_iomap() for FRV. Signed-off-by:
-
David Howells authored
Add stubs for FRV module support. Signed-off-by:
-
David Howells authored
Supply various I/O access primitives that are missing for the FRV arch: (*) mmiowb() (*) read*_relaxed() (*) ioport_*map() (*) ioread*(), iowrite*(), ioread*_rep() and iowrite*_rep() (*) pci_io*map() (*) check_signature() The patch also makes __is_PCI_addr() more efficient. Signed-off-by:
-
David Howells authored
Fix the exception table handling so that modules exceptions are dealt with. Signed-off-by:
-
David Howells authored
Export a number of features required to build all the modules. It also implements the following simple features: (*) csum_partial_copy_from_user() for MMU as well as no-MMU. (*) __ucmpdi2(). so that they can be exported too. Signed-off-by:
-
David Howells authored
Drop support for debugging features that aren't supported on FRV: (*) EARLY_PRINTK The on-chip UARTs are set up early enough that this isn't required, and VGA support isn't available. There's also a gdbstub available. (*) DEBUG_PAGEALLOC This can't be easily be done since we use huge static mappings to cover the kernel, not pages. Signed-off-by:
-
David Howells authored
Drop support for 8-bit and 16-bit xchg and cmpxchg emulation and implements 32-bit xchg with the SWAP/SWAPI instruction. Signed-off-by:
-
David Howells authored
Suppress configuration of certain features for the FRV arch as they can't be built for FRV at the moment: (*) RTC (*) HISAX_* (*) PARPORT_PC (*) VGA_CONSOLE (*) BINFMT_ELF Signed-off-by:
-
Lennert Buytenhek authored
The Kconfig symbol for pnx0105 was recently renamed to ARCH_PNX010X. Signed-off-by:
Lennert Buytenhek <buytenh@wantstofly.org> Cc: dmitry pervushin <dpervushin@ru.mvista.com> Cc: <dsaxena@plexity.net> Cc: Jeff Garzik <jgarzik@pobox.com> Signed-off-by:
-
Lennert Buytenhek authored
PNX010X support for CS89x0 should be conditional on NET_PCI, as it is an 'on board controller' and NET_PCI includes that category of NICs. Since ARCH_PNX0105 was recently changed to ARCH_PNX010X, incorporate that change as well while we're at it. Signed-off-by:
-
Lennert Buytenhek authored
Implement readwords/writewords that use readword/writeword, and switch the rest of the driver over to use these. Signed-off-by:
-
Lennert Buytenhek authored
Implement suitable versions of the readword/writeword macros for ixdp2x01 and pnx0501. Handle the 32-bit spacing of the registers in these functions instead of in the header file. Signed-off-by:
-
Lennert Buytenhek authored
Make readreg/writereg use readword/writeword. Signed-off-by:
-
Lennert Buytenhek authored
Reverse the order of readreg/writereg and readword/writeword in the file, so that we can make readreg/writereg use readword/writeword. Signed-off-by:
-
Lennert Buytenhek authored
Switch all occurences of inw/outw in the driver over to readword/writeword. Signed-off-by:
-
Lennert Buytenhek authored
readword() and writeword() take a 'struct net_device *' and deref its ->base_addr member. Make them take the base_addr directly instead, so that we can switch the other occurences of inw/outw in the file over to readword/writeword as well. Signed-off-by:
-
Kirill Korotaev authored
When oom_killer kills current there's no need to call schedule_timeout_interruptible() since task must die ASAP. Signed-Off-By:
Pavel Emelianov <xemul@sw.ru> Signed-Off-By:
Kirill Korotaev <dev@openvz.org> Signed-off-by:
-
Christoph Lameter authored
Group page migration functions in mempolicy.c Add a forward declaration for migrate_page_add (like gather_stats()) and use our new found mobility to group all page migration related function around do_migrate_pages(). Signed-off-by:
Christoph Lameter <clameter@sgi.com> Cc: Andi Kleen <ak@muc.de> Signed-off-by:
-
Christoph Lameter authored
Since the numa_maps functionality is now in mempolicy.c we no longer need to export get_vma_policy(). Signed-off-by:
-
Christoph Lameter authored
migrate_page_add cannot be called with a spinlock held (calls isolate_lru_page which calles schedule_on_each_cpu). Drop ptl lock in check_pte_range before calling migrate_page_add(). Signed-off-by:
-
Christoph Lameter authored
First discussed at http://marc.theaimsgroup.com/?t=113149255100001&r=1&w=2 - Use the check_range() in mempolicy.c to gather statistics. - Improve the numa_maps code in general and fix some comments. Signed-off-by:
-
Christoph Lameter authored
This was was first posted at http://marc.theaimsgroup.com/?l=linux-mm&m=113149240227584&w=2 (Part of this functionality is also contained in the direct migration pathset. The functionality here is more generic and independent of that patchset.) - Add internal flags MPOL_MF_INVERT to control check_range() behavior. - Replace the pagelist passed through by check_range by a general private pointer that may be used for other purposes. (The following patches will use that to merge numa_maps into mempolicy.c and to better group the page migration code in the policy layer) - Improve some comments. Signed-off-by:
-
Dave Jones authored
We seem to be hitting this assertion failure too often for it to be hardware bugs. Cc: Hugh Dickins <hugh@veritas.com> Cc: Nick Piggin <nickpiggin@yahoo.com.au> Signed-off-by:
-
Tobias Klauser authored
Clean up a local variable with the same name as a variable in a larger block. Also move a variable into the block where it's actually used. Spotted by http://linuxicc.sourceforge.net/Signed-off-by:
Tobias Klauser <tklauser@nuerscht.ch> Signed-off-by:
-
Avishay Traeger authored
Fix set_page_count() macro to handle complex arguments. Signed-off-by:
Avishay Traeger <atraeger@cs.sunysb.edu> Signed-off-by:
-
Christoph Lameter authored
See http://marc.theaimsgroup.com/?l=linux-kernel&m=113167000201265&w=2 http://marc.theaimsgroup.com/?l=linux-mm&m=113167267527312&w=2 Make hugepages obey cpusets. Signed-off-by:
William Irwin <wli@holomorphy.com> Signed-off-by:
-
Paul Jackson authored
Add a boolean "memory_migrate" to each cpuset, represented by a file containing "0" or "1" in each directory below /dev/cpuset. It defaults to false (file contains "0"). It can be set true by writing "1" to the file. If true, then anytime that a task is attached to the cpuset so marked, the pages of that task will be moved to that cpuset, preserving, to the extent practical, the cpuset-relative placement of the pages. Also anytime that a cpuset so marked has its memory placement changed (by writing to its "mems" file), the tasks in that cpuset will have their pages moved to the cpusets new nodes, preserving, to the extent practical, the cpuset-relative placement of the moved pages. Signed-off-by:
Paul Jackson <pj@sgi.com> Cc: Christoph Lameter <christoph@lameter.com> Signed-off-by:
-
Christoph Lameter authored
Use -Exxx instead of numeric return codes and cleanup the code in migrate_pages() using -Exx error codes. Consolidate successful migration handling Signed-off-by:
-
Christoph Lameter authored
Extend the parameters of migrate_pages() to allow the caller control over the fate of successfully migrated or impossible to migrate pages. Swap migration and direct migration will have the same interface after this patch so that patches can be independently applied to the policy layer and the core migration code. Signed-off-by:
-
Christoph Lameter authored
Drop unused pages immediately If a page is encountered that is only referenced by the migration code then there is no reason to swap or migrate the page. Release the page by calling move_to_lru(). Signed-off-by:
-
Christoph Lameter authored
Add gfp_mask to add_to_swap add_to_swap does allocations with GFP_ATOMIC in order not to interfere with swapping. During migration we may have use add_to_swap extensively which may lead to out of memory errors. This patch makes add_to_swap take a parameter that specifies the gfp mask. The page migration code can then make add_to_swap use GFP_KERNEL. Signed-off-by:
Hirokazu Takahashi <taka@valinux.co.jp> Signed-off-by:
Dave Hansen <haveblue@us.ibm.com> Signed-off-by:
-
Christoph Lameter authored
Move move_to_lru, putback_lru_pages and isolate_lru in section surrounded by CONFIG_MIGRATION saving some codesize for single processor kernels. Signed-off-by:
-
Christoph Lameter authored
sys_migrate_pages implementation using swap based page migration This is the original API proposed by Ray Bryant in his posts during the first half of 2005 on linux-mm@kvack.org and linux-kernel@vger.kernel.org. The intent of sys_migrate is to migrate memory of a process. A process may have migrated to another node. Memory was allocated optimally for the prior context. sys_migrate_pages allows to shift the memory to the new node. sys_migrate_pages is also useful if the processes available memory nodes have changed through cpuset operations to manually move the processes memory. Paul Jackson is working on an automated mechanism that will allow an automatic migration if the cpuset of a process is changed. However, a user may decide to manually control the migration. This implementation is put into the policy layer since it uses concepts and functions that are also needed for mbind and friends. The patch also provides a do_migrate_pages function that may be useful for cpusets to automatically move memory. sys_migrate_pages does not modify policies in contrast to Ray's implementation. The current code here is based on the swap based page migration capability and thus is not able to preserve the physical layout relative to it containing nodeset (which may be a cpuset). When direct page migration becomes available then the implementation needs to be changed to do a isomorphic move of pages between different nodesets. The current implementation simply evicts all pages in source nodeset that are not in the target nodeset. Patch supports ia64, i386 and x86_64. Signed-off-by:
-
Christoph Lameter authored
Add page migration support via swap to the NUMA policy layer This patch adds page migration support to the NUMA policy layer. An additional flag MPOL_MF_MOVE is introduced for mbind. If MPOL_MF_MOVE is specified then pages that do not conform to the memory policy will be evicted from memory. When they get pages back in new pages will be allocated following the numa policy. Signed-off-by:
-
Christoph Lameter authored
Include page migration if the system is NUMA or having a memory model that allows distinct areas of memory (SPARSEMEM, DISCONTIGMEM). And: - Only include lru_add_drain_per_cpu if building for an SMP system. Signed-off-by:
-
Christoph Lameter authored
This adds the basic page migration function with a minimal implementation that only allows the eviction of pages to swap space. Page eviction and migration may be useful to migrate pages, to suspend programs or for remapping single pages (useful for faulty pages or pages with soft ECC failures) The process is as follows: The function wanting to migrate pages must first build a list of pages to be migrated or evicted and take them off the lru lists via isolate_lru_page(). isolate_lru_page determines that a page is freeable based on the LRU bit set. Then the actual migration or swapout can happen by calling migrate_pages(). migrate_pages does its best to migrate or swapout the pages and does multiple passes over the list. Some pages may only be swappable if they are not dirty. migrate_pages may start writing out dirty pages in the initial passes over the pages. However, migrate_pages may not be able to migrate or evict all pages for a variety of reasons. The remaining pages may be returned to the LRU lists using putback_lru_pages(). Changelog V4->V5: - Use the lru caches to return pages to the LRU Changelog V3->V4: - Restructure code so that applying patches to support full migration does require minimal changes. Rename swapout_pages() to migrate_pages(). Changelog V2->V3: - Extract common code from shrink_list() and swapout_pages() Signed-off-by:
Mike Kravetz <kravetz@us.ibm.com> Signed-off-by:
-
