- 26 Aug, 2009 9 commits
-
-
Venkatesh Pallipadi authored
Lookup the reserved memtype during vm_insert_pfn and use that memtype for the new mapping. This takes care or handling of vm_insert_pfn() interface in track_pfn_vma*/untrack_pfn_vma. Signed-off-by:
Venkatesh Pallipadi <venkatesh.pallipadi@intel.com> Signed-off-by:
Suresh Siddha <suresh.b.siddha@intel.com> Signed-off-by:
H. Peter Anvin <hpa@zytor.com>
-
Venkatesh Pallipadi authored
Add a new routine lookup_memtype() to get the current memtype based on the PAT reserves and frees. Signed-off-by:
-
Venkatesh Pallipadi authored
Change reserve_ram_pages_type and free_ram_pages_type to use 2 page flags to track UC_MINUS, WC, WB and default types. Previous RAM tracking just tracked WB or NonWB, which was not complete and did not allow tracking of RAM fully and there was no way to get the actual type reserved by looking at the page flags. We use the memtype_lock spinlock for atomicity in dealing with memtype tracking in struct page. Signed-off-by:
-
Venkatesh Pallipadi authored
Only IA64 was using PG_uncached as of now. We now intend to use this bit in x86 as well, to keep track of memory type of those addresses that have page struct for them. So, generalize the use of that bit across ia64 and x86. Signed-off-by:
-
Venkatesh Pallipadi authored
PAT memtype tracking uses a linear link list to keep track of IO (non-RAM) regions and their memtypes. The code used a last_accessed pointer as a cache to speedup the lookup. As per discussions with H. Peter Anvin a while back, having a rbtree here will avoid bad performances in pathological cases where we may end up with huge linked list. This may not add any noticable performance speedup in normal case as the number of entires in PAT memtype list tend to be ~20-30 range. The patch removes the "cached_entry" logic as with rbtree we have more generic way of speeding up the lookup. With this patch, we use rbtree to do the quick lookup. We still use linked list as the memtype range tracked can be of different sizes and can overlap in different ways. We also keep track of usage counts with linked list. Example: Multiple ioremaps with different sizes uncached-minus @ 0xfffff00000-0xfffff04000 uncached-minus @ 0xfffff02000-0xfffff03000 And one userlevel mmap and the thread forks a new process uncached-minus @ 0xbf453000-0xbf454000 uncached-minus @ 0xbf453000-0xbf454000 Signed-off-by:
-
Venkatesh Pallipadi authored
io_mapping_* interfaces were added, mainly for graphics drivers. Make this interface go through the PAT reserve/free, instead of hardcoding WC mapping. This makes sure that there are no aliases due to unconditional WC setting. Signed-off-by:
-
Venkatesh Pallipadi authored
Add new routines to request memtype for IO regions. This will currently be a backend for io_mapping_* routines. But, it can also be made available to drivers directly in future, in case it is needed. reserve interface reserves the memory, makes sure we have a compatible memory type available and keeps the identity map in sync when needed. Signed-off-by:
-
Venkatesh Pallipadi authored
ioremap has this hard-coded check for new type and requested type. That check differs from other PAT users like /dev/mem mmap, remap_pfn_range in only one condition where requested type is UC_MINUS and new type is WC. Under that condition, ioremap fails. But other PAT interfaces succeed with a WC mapping. Change to make ioremap be in sync with other PAT APIs and use the same macro as others. Also changes the error print to KERN_ERR instead of pr_debug. Signed-off-by:
-
Venkatesh Pallipadi authored
Make reserve_memtype internally take care of pat disabled case and fallback to default return values. Remove the specific pat_disabled checks in track_* routines. Change kernel_map_sync_memtype to sync identity map even when pat_disabled. This change ensures that, even for pat_disabled case, we take care of keeping identity map in sync. Before this patch, in pat disabled case, ioremap() keeps the identity maps in sync and other APIs like pci and /dev/mem mmap don't, which is not a very consistent behavior. Signed-off-by:
-
- 22 Aug, 2009 1 commit
-
-
H. Peter Anvin authored
mtr_aps_delayed_init was declared u32 and made global, but it only ever takes boolean values and is only ever used in arch/x86/kernel/cpu/mtrr/main.c. Declare it "static bool" and remove external references. Signed-off-by:
-
- 21 Aug, 2009 2 commits
-
-
Suresh Siddha authored
SDM Vol 3a section titled "MTRR considerations in MP systems" specifies the need for synchronizing the logical cpu's while initializing/updating MTRR. Currently Linux kernel does the synchronization of all cpu's only when a single MTRR register is programmed/updated. During an AP online (during boot/cpu-online/resume) where we initialize all the MTRR/PAT registers, we don't follow this synchronization algorithm. This can lead to scenarios where during a dynamic cpu online, that logical cpu is initializing MTRR/PAT with cache disabled (cr0.cd=1) etc while other logical HT sibling continue to run (also with cache disabled because of cr0.cd=1 on its sibling). Starting from Westmere, VMX transitions with cr0.cd=1 don't work properly (because of some VMX performance optimizations) and the above scenario (with one logical cpu doing VMX activity and another logical cpu coming online) can result in system crash. Fix the MTRR initialization by doing rendezvous of all the cpus. During boot and resume, we delay the MTRR/PAT init for APs till all the logical cpu's come online and the rendezvous process at the end of AP's bringup, will initialize the MTRR/PAT for all AP's. For dynamic single cpu online, we synchronize all the logical cpus and do the MTRR/PAT init on the AP that is coming online. Signed-off-by:
-
Suresh Siddha authored
Because of deadlock possiblities smp_call_function() is not allowed to be called with interrupts disabled. Add an exception for the cpu not yet online, as no one else can send smp call function interrupt to this cpu that is not yet online and as such deadlock condition is not possible. Signed-off-by:
Nick Piggin <npiggin@suse.de> Signed-off-by:
-
- 09 Aug, 2009 1 commit
-
-
Marcin Slusarz authored
Signed-off-by:
Marcin Slusarz <marcin.slusarz@gmail.com> Cc: "H. Peter Anvin" <hpa@zytor.com> LKML-Reference: <1249847649-11631-6-git-send-email-marcin.slusarz@gmail.com> Signed-off-by:
Ingo Molnar <mingo@elte.hu>
-
- 11 Jul, 2009 2 commits
-
-
Alan Cox authored
No code changes except printk levels (although some of the K6 mtrr code might be clearer if there were a few as would splitting out some of the intel cache code). Signed-off-by:
Alan Cox <alan@linux.intel.com> LKML-Reference: <new-submission> Signed-off-by:
-
Huang Weiyi authored
Remove duplicated #include in: arch/x86/kernel/dumpstack.c Signed-off-by:
Huang Weiyi <weiyi.huang@gmail.com> Signed-off-by:
-
- 10 Jul, 2009 2 commits
-
-
Cyrill Gorcunov authored
We already use a lot of cpu_has_ helpers. Lets do here the same for consistency. Signed-off-by:
Cyrill Gorcunov <gorcunov@openvz.org> Cc: Yinghai Lu <yinghai@kernel.org> LKML-Reference: <20090705160154.GB4791@lenovo> Signed-off-by:
-
Cyrill Gorcunov authored
Signed-off-by:
-
- 05 Jul, 2009 1 commit
-
-
Ingo Molnar authored
Yinghai noticed that i defined BIOS_BUG_MSG but added no usage for it. The usage is to clean up this turd in generic.c: printk(KERN_WARNING "WARNING: BIOS bug: VAR MTRR %d " "contains strange UC entry under 1M, check " "with your system vendor!\n", i); Breaking printk lines in the middle looks ugly, is hard to read and breaks 'git grep'. Use the BIOS_BUG_MSG instead. Also complete the moving of structure definitions and variables to the top of the file. Reported-by:
Yinghai Lu <yinghai@kernel.org> LKML-Reference: <20090703164225.GA21447@elte.hu> Signed-off-by:
-
- 04 Jul, 2009 10 commits
-
-
Jaswinder Singh Rajput authored
Fix following trivial style problems: ERROR: trailing whitespace X 25 WARNING: Use #include <linux/uaccess.h> instead of <asm/uaccess.h> WARNING: Use #include <linux/kvm_para.h> instead of <asm/kvm_para.h> ERROR: do not initialise externals to 0 or NULL X 2 ERROR: "foo * bar" should be "foo *bar" X 5 ERROR: do not use assignment in if condition X 2 WARNING: line over 80 characters X 8 ERROR: return is not a function, parentheses are not required WARNING: braces {} are not necessary for any arm of this statement ERROR: space required before the open parenthesis '(' X 2 ERROR: open brace '{' following function declarations go on the next line ERROR: space required after that ',' (ctx:VxV) X 8 ERROR: space required before the open parenthesis '(' X 3 ERROR: else should follow close brace '}' WARNING: space prohibited between function name and open parenthesis '(' WARNING: EXPORT_SYMBOL(foo); should immediately follow its function/variable X 2 Also use pr_debug and pr_warning where possible. total: 50 errors, 14 warnings arch/x86/kernel/cpu/mtrr/main.o: text data bss dec hex filename 3668 116 4156 7940 1f04 main.o.before 3668 116 4156 7940 1f04 main.o.after md5: e01af2fd28deef77c8d01e71acfbd365 main.o.before.asm e01af2fd28deef77c8d01e71acfbd365 main.o.after.asm Suggested-by:Alan Cox <alan@lxorguk.ukuu.org.uk> Signed-off-by:
Jaswinder Singh Rajput <jaswinderrajput@gmail.com> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Yinghai Lu <yinghai@kernel.org> LKML-Reference: <20090703164225.GA21447@elte.hu> Cc: Avi Kivity <avi@redhat.com> # Avi, please have a look at the kvm_para.h bit [ More cleanups ] Signed-off-by:
-
Jaswinder Singh Rajput authored
Fix: WARNING: Use #include <linux/io.h> instead of <asm/io.h> WARNING: line over 80 characters X 4 arch/x86/kernel/cpu/mtrr/state.o: text data bss dec hex filename 864 0 0 864 360 state.o.before 864 0 0 864 360 state.o.after md5: c5c4364b9aeac74d70111e1e49667a2c state.o.before.asm c5c4364b9aeac74d70111e1e49667a2c state.o.after.asm Suggested-by: -
Jaswinder Singh Rajput authored
Fix: ERROR: do not use C99 // comments ERROR: "foo * bar" should be "foo *bar" X 2 Suggested-by:
-
Jaswinder Singh Rajput authored
Fix: WARNING: Use #include <linux/uaccess.h> instead of <asm/uaccess.h> ERROR: trailing whitespace X 7 ERROR: trailing statements should be on next line X 3 WARNING: line over 80 characters X 5 ERROR: space required before the open parenthesis '(' arch/x86/kernel/cpu/mtrr/if.o: text data bss dec hex filename 2239 4 0 2243 8c3 if.o.before 2239 4 0 2243 8c3 if.o.after md5: 78d1f2aa4843ec6509c18e2dee54bc7f if.o.before.asm 78d1f2aa4843ec6509c18e2dee54bc7f if.o.after.asm Suggested-by: -
Jaswinder Singh Rajput authored
Fix following trivial style problems: ERROR: trailing whitespace X 4 WARNING: Use #include <linux/io.h> instead of <asm/io.h> WARNING: braces {} are not necessary for single statement blocks X 3 ERROR: "foo * bar" should be "foo *bar" WARNING: line over 80 characters X 6 ERROR: "foo * bar" should be "foo *bar" ERROR: spaces required around that '=' (ctx:VxO) ERROR: space required before that '-' (ctx:OxV) WARNING: suspect code indent for conditional statements (8, 12) ERROR: spaces required around that '=' (ctx:VxV) ERROR: do not initialise statics to 0 or NULL ERROR: space prohibited after that open parenthesis '(' X 2 ERROR: space prohibited before that close parenthesis ')' X 2 ERROR: trailing statements should be on next line ERROR: return is not a function, parentheses are not required Also use pr_debug and pr_warning where possible. arch/x86/kernel/cpu/mtrr/generic.o: text data bss dec hex filename 5652 77 4224 9953 26e1 generic.o.before 5652 77 4220 9949 26dd generic.o.after The md5 changed: b34d6c045f06daa4ed092b90cc760e8f generic.o.before.asm a490c6251cfd8442fbffecc0e09a573d generic.o.after.asm Because mtrr_state moved from data to bss, changing its offsets - and also because __LINE__ numbers changed. Suggested-by: -
Jaswinder Singh Rajput authored
Fix trivial style problems: WARNING: Use #include <linux/io.h> instead of <asm/io.h> WARNING: line over 80 characters ERROR: do not initialise statics to 0 or NULL ERROR: space prohibited after that open parenthesis '(' X 2 ERROR: space prohibited before that close parenthesis ')' X 2 ERROR: trailing whitespace X 2 ERROR: trailing statements should be on next line ERROR: do not use C99 // comments X 2 arch/x86/kernel/cpu/mtrr/cyrix.o: text data bss dec hex filename 1637 32 8 1677 68d cyrix.o.before 1637 32 8 1677 68d cyrix.o.after md5: 6f52abd06905be3f4cabb5239f9b0ff0 cyrix.o.before.asm 6f52abd06905be3f4cabb5239f9b0ff0 cyrix.o.after.asm Suggested-by: -
Jaswinder Singh Rajput authored
Fix trivial style problems: WARNING: Use #include <linux/uaccess.h> instead of <asm/uaccess.h> WARNING: Use #include <linux/kvm_para.h> instead of <asm/kvm_para.h> Also, nr_mtrr_spare_reg should be unsigned long. arch/x86/kernel/cpu/mtrr/cleanup.o: text data bss dec hex filename 6241 8992 2056 17289 4389 cleanup.o.before 6241 8992 2056 17289 4389 cleanup.o.after The md5 has changed: 1a7a27513aef1825236daf29110fe657 cleanup.o.before.asm bcea358efa2532b6020e338e158447af cleanup.o.after.asm Because a WARN_ON()'s __LINE__ value changed by 3 lines. Suggested-by:
-
Jaswinder Singh Rajput authored
Remove dead code and fix trivial style problems: ERROR: trailing whitespace X 2 WARNING: line over 80 characters X 3 ROR: trailing whitespace ERROR: do not use C99 // comments X 2 arch/x86/kernel/cpu/mtrr/centaur.o: text data bss dec hex filename 605 32 68 705 2c1 centaur.o.before 605 32 68 705 2c1 centaur.o.after md5: a4865ea98ce3c163bb1d376a3949b3e3 centaur.o.before.asm a4865ea98ce3c163bb1d376a3949b3e3 centaur.o.after.asm Suggested-by: -
Jaswinder Singh Rajput authored
Fix trivial style problems : ERROR: trailing whitespace WARNING: line over 80 characters ERROR: do not use C99 // comments arch/x86/kernel/cpu/mtrr/amd.o: text data bss dec hex filename 501 32 0 533 215 amd.o.before 501 32 0 533 215 amd.o.after md5: 62f795eb840ee2d17b03df89e789e76c amd.o.before.asm 62f795eb840ee2d17b03df89e789e76c amd.o.after.asm Suggested-by: -
Ingo Molnar authored
Merge reason: We were on an older pre-rc1 base, move to almost-rc2. Signed-off-by:
-
- 02 Jul, 2009 12 commits
-
-
git://git.infradead.org/users/eparis/notifyLinus Torvalds authored
* 'for-linus' of git://git.infradead.org/users/eparis/notify: fs/notify/inotify: decrement user inotify count on close
-
git://git.kernel.org/pub/scm/linux/kernel/git/mason/btrfs-unstableLinus Torvalds authored
* git://git.kernel.org/pub/scm/linux/kernel/git/mason/btrfs-unstable: Btrfs: fix error message formatting Btrfs: fix use after free in btrfs_start_workers fail path Btrfs: honor nodatacow/sum mount options for new files Btrfs: update backrefs while dropping snapshot Btrfs: account for space we may use in fallocate Btrfs: fix the file clone ioctl for preallocated extents Btrfs: don't log the inode in file_write while growing the file
-
git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi-rc-fixes-2.6Linus Torvalds authored
* git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi-rc-fixes-2.6: [SCSI] cxgb3i: fix connection error when vlan is enabled [SCSI] FC transport: Locking fix for common-code FC pass-through patch [SCSI] zalon: fix oops on attach failure [SCSI] fnic: use DMA_BIT_MASK(nn) instead of deprecated DMA_nnBIT_MASK [SCSI] fnic: remove redundant BUG_ONs and fix checks on unsigned [SCSI] ibmvscsi: Fix module load hang
-
git://git.infradead.org/iommu-2.6Linus Torvalds authored
* git://git.infradead.org/iommu-2.6: (38 commits) intel-iommu: Don't keep freeing page zero in dma_pte_free_pagetable() intel-iommu: Introduce first_pte_in_page() to simplify PTE-setting loops intel-iommu: Use cmpxchg64_local() for setting PTEs intel-iommu: Warn about unmatched unmap requests intel-iommu: Kill superfluous mapping_lock intel-iommu: Ensure that PTE writes are 64-bit atomic, even on i386 intel-iommu: Make iommu=pt work on i386 too intel-iommu: Performance improvement for dma_pte_free_pagetable() intel-iommu: Don't free too much in dma_pte_free_pagetable() intel-iommu: dump mappings but don't die on pte already set intel-iommu: Combine domain_pfn_mapping() and domain_sg_mapping() intel-iommu: Introduce domain_sg_mapping() to speed up intel_map_sg() intel-iommu: Simplify __intel_alloc_iova() intel-iommu: Performance improvement for domain_pfn_mapping() intel-iommu: Performance improvement for dma_pte_clear_range() intel-iommu: Clean up iommu_domain_identity_map() intel-iommu: Remove last use of PHYSICAL_PAGE_MASK, for reserving PCI BARs intel-iommu: Make iommu_flush_iotlb_psi() take pfn as argument intel-iommu: Change aligned_size() to aligned_nrpages() intel-iommu: Clean up intel_map_sg(), remove domain_page_mapping() ...
-
Yinghai Lu authored
fix hang with HIGHMEM_64G and 32bit resource. According to hpa and Linus, use (resource_size_t)-1 to fend off big ranges. Analyzed by hpa Reported-and-tested-by:
Mikael Pettersson <mikpe@it.uu.se> Signed-off-by:
Linus Torvalds <torvalds@linux-foundation.org>
-
Linus Torvalds authored
These macros had two bugs: - the type of the mask was not correctly expanded to the full size of the argument being expanded, resulting in possible loss of high bits when mixing types. - the alignment argument was evaluated twice, despite the macro looking like a fancy function (but it really does need to be a macro, since it works on arbitrary integer types) Noticed by Peter Anvin, and with a fix that is a modification of his suggestion (bug noticed by Yinghai Lu). Cc: Peter Anvin <hpa@zytor.com> Cc: Yinghai Lu <yinghai@kernel.org> Signed-off-by: -
Hu Tao authored
Make an error msg look nicer by inserting a space between number and word. Signed-off-by:
Hu Tao <hu.taoo@gmail.com> Signed-off-by:
Jiri Kosina <jkosina@suse.cz> Signed-off-by:
Chris Mason <chris.mason@oracle.com>
-
Jiri Slaby authored
worker memory is already freed on one fail path in btrfs_start_workers, but is still dereferenced. Switch the dereference and kfree. Signed-off-by:
Jiri Slaby <jirislaby@gmail.com> Signed-off-by:
-
Chris Mason authored
The btrfs attr patches unconditionally inherited the inode flags field without honoring nodatacow and nodatasum. This fix makes sure we properly record the nodatacow/sum mount options in new inodes. Signed-off-by: -
Yan Zheng authored
The new backref format has restriction on type of backref item. If a tree block isn't referenced by its owner tree, full backrefs must be used for the pointers in it. When a tree block loses its owner tree's reference, backrefs for the pointers in it should be updated to full backrefs. Current btrfs_drop_snapshot misses the code that updates backrefs, so it's unsafe for general use. This patch adds backrefs update code to btrfs_drop_snapshot. It isn't a problem in the restricted form btrfs_drop_snapshot is used today, but for general snapshot deletion this update is required. Signed-off-by:
Yan Zheng <zheng.yan@oracle.com> Signed-off-by:
-
Josef Bacik authored
Using Eric Sandeen's xfstest for fallocate, you can easily trigger a ENOSPC panic on btrfs. This is because we do not account for data we may use when doing the fallocate. This patch fixes the problem by properly reserving space, and then just freeing it when we are done. The reservation stuff was made with delalloc in mind, so its a little crude for this case, but it keeps the box from panicing. Signed-off-by:
Josef Bacik <jbacik@redhat.com> Signed-off-by:
-
Chris Mason authored
-
