search for: cpages

Hi all, I'm upgrading my old F9 mail server to Centos 7 and all is going well. However, I've got a problem with recently created users. I've rsync'd /home and /var/spool/mail after moving /etc/passwd and /etc/group I used the default locations on the old F9 server, i.e. mail delivered to /var/spool/mail/%u by EXIM and Dovecot using that as the inbox. Dovecot then used ~/mail/

...grated). * * Once the caller is done cleaning up things and updating its page table (if it * chose to do so, this is not an obligation) it finally calls @@ -2657,10 +2656,11 @@ int migrate_vma_setup(struct migrate_vma *args) args->npages = 0; migrate_vma_collect(args); - if (args->cpages) - migrate_vma_prepare(args); - if (args->cpages) - migrate_vma_unmap(args); + if (!args->cpages) + return 0; + + migrate_vma_prepare(args); + migrate_vma_unmap(args); /* * At this point pages are locked and unmapped, and thus they have

On Fri, Aug 16, 2019 at 05:11:07PM +0000, Jason Gunthorpe wrote: > - if (args->cpages) > - migrate_vma_prepare(args); > - if (args->cpages) > - migrate_vma_unmap(args); > + if (!args->cpages) > + return 0; > + > + migrate_vma_prepare(args); > + migrate_vma_unmap(args); I don't think this is ok. Both migrate_vma_prepare and migrate_vma_unmap can...

These changes are based on Jason's latest hmm branch. Patch 1 was previously posted here [1] but was dropped from the orginal series. Hopefully, the tests will reduce concerns about edge conditions. I'm sure more tests could be usefully added but I thought this was a good starting point. [1] https://lore.kernel.org/linux-mm/20190726005650.2566-6-rcampbell at nvidia.com/ Ralph Campbell

...fault too. > */ > - ret = migrate_vma(&nouveau_dmem_fault_migrate_ops, vmf->vma, > - vmf->address, vmf->address + PAGE_SIZE, > - src, dst, &fault); > - if (ret) > + if (migrate_vma_setup(&args) < 0) > return VM_FAULT_SIGBUS; > + if (!args.cpages) > + return 0; > + > + nouveau_dmem_fault_alloc_and_copy(args.vma, src, dst, args.start, > + args.end, &fault); > + migrate_vma_pages(&args); > + nouveau_dmem_fault_finalize_and_map(&fault); > > + migrate_vma_finalize(&args); > if (dst[0] == MIGRA...

...likely that more surrounding page will CPU fault too. */ - ret = migrate_vma(&nouveau_dmem_fault_migrate_ops, vmf->vma, - vmf->address, vmf->address + PAGE_SIZE, - src, dst, &fault); - if (ret) + if (migrate_vma_setup(&args) < 0) return VM_FAULT_SIGBUS; + if (!args.cpages) + return 0; + + nouveau_dmem_fault_alloc_and_copy(args.vma, src, dst, args.start, + args.end, &fault); + migrate_vma_pages(&args); + nouveau_dmem_fault_finalize_and_map(&fault); + migrate_vma_finalize(&args); if (dst[0] == MIGRATE_PFN_ERROR) return VM_FAULT_SIGBUS; @@ -...

This series adds new functions to the mmu interval notifier API to allow device drivers with MMUs to dynamically mirror a process' page tables based on device faults and invalidation callbacks. The Nouveau driver is updated to use the extended API and a set of stand alone self tests is added to help validate and maintain correctness. The patches are based on linux-5.5.0-rc6 and are for

...since v2: - don't unmap pages when returning 0 from nouveau_dmem_migrate_to_ram - minor style fixes - add a new patch to remove CONFIG_MIGRATE_VMA_HELPER Changes since v1: - fix a few whitespace issues - drop the patch to remove MIGRATE_PFN_WRITE for now - various spelling fixes - clear cpages and npages in migrate_vma_setup - fix the nouveau_dmem_fault_copy_one return value - minor improvements to some nouveau internal calling conventions

...likely that more surrounding page will CPU fault too. */ - ret = migrate_vma(&nouveau_dmem_fault_migrate_ops, vmf->vma, - vmf->address, vmf->address + PAGE_SIZE, - src, dst, &fault); - if (ret) + if (migrate_vma_setup(&args) < 0) return VM_FAULT_SIGBUS; + if (!args.cpages) + return 0; + + nouveau_dmem_fault_alloc_and_copy(args.vma, src, dst, args.start, + args.end, &fault); + migrate_vma_pages(&args); + nouveau_dmem_fault_finalize_and_map(&fault); + migrate_vma_finalize(&args); if (dst[0] == MIGRATE_PFN_ERROR) return VM_FAULT_SIGBUS; @@ -...

Hi Jérôme, Ben and Jason, below is a series against the hmm tree which starts revamping the migrate_vma functionality. The prime idea is to export three slightly lower level functions and thus avoid the need for migrate_vma_ops callbacks. Diffstat: 4 files changed, 285 insertions(+), 602 deletions(-) A git tree is also available at: git://git.infradead.org/users/hch/misc.git

...//git.infradead.org/users/hch/misc.git migrate_vma-cleanup.2 Gitweb: http://git.infradead.org/users/hch/misc.git/shortlog/refs/heads/migrate_vma-cleanup.2 Changes since v1: - fix a few whitespace issues - drop the patch to remove MIGRATE_PFN_WRITE for now - various spelling fixes - clear cpages and npages in migrate_vma_setup - fix the nouveau_dmem_fault_copy_one return value - minor improvements to some nouveau internal calling conventions

Hello, there I want to connect to a Windows XP share using the following line in my fstab \\WINDOWS-PC\SHARE /dir/to/mount/share smbfs auto,umask=0002,credentials=/etc/winpassword,iocharset=utf8,gid=0,noauto 0 0 It mounts OK but I can't get the charset shown correctly for non-ASCII chars like ? ? ? et cetera. I've even tried using cpage=850 option when mounting without success.

...ptr_t)buffer->ptr, size); + ret = madvise(map, size, MADV_HUGEPAGE); + ASSERT_EQ(ret, 0); + old_ptr = buffer->ptr; + buffer->ptr = map; + + /* Migrate memory to device. */ + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_MIGRATE, buffer, npages); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, npages); + + /* Check what the device read. */ + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], 0); + + buffer->ptr = old_ptr; + hmm_buffer_free(buffer); +} + +/* + * Migrate private anonymous huge zero page. + */ +TEST_F(hmm, migrate_anon_huge_zero)...

This series adds support for transparent huge page migration to migrate_vma_*() and adds nouveau SVM and HMM selftests as consumers. Earlier versions were posted previously [1] and [2]. The patches apply cleanly to the linux-mm 5.10.0-rc2 tree. There are a lot of other THP patches being posted. I don't think there are any semantic conflicts but there may be some merge conflicts depending on

This series adds support for transparent huge page migration to migrate_vma_*() and adds nouveau SVM and HMM selftests as consumers. An earlier version was posted previously [1]. This version now supports splitting a THP midway in the migration process which led to a number of changes. The patches apply cleanly to the current linux-mm tree. Since there are a couple of patches in linux-mm from Dan

...ing 0 from nouveau_dmem_migrate_to_ram > - minor style fixes > - add a new patch to remove CONFIG_MIGRATE_VMA_HELPER > > Changes since v1: > - fix a few whitespace issues > - drop the patch to remove MIGRATE_PFN_WRITE for now > - various spelling fixes > - clear cpages and npages in migrate_vma_setup > - fix the nouveau_dmem_fault_copy_one return value > - minor improvements to some nouveau internal calling conventions > Some of the patches seem to have been mangled in the mail. I was able to edit them and apply to Jason's tree https://github.c...

...mmu_notifier.h> #include <linux/memory_hotplug.h> -#if IS_ENABLED(CONFIG_HMM_MIRROR) static const struct mmu_notifier_ops hmm_mmu_notifier_ops; static inline struct hmm *mm_get_hmm(struct mm_struct *mm) @@ -1326,4 +1325,3 @@ long hmm_range_dma_unmap(struct hmm_range *range, return cpages; } EXPORT_SYMBOL(hmm_range_dma_unmap); -#endif /* IS_ENABLED(CONFIG_HMM_MIRROR) */ -- 2.20.1

...o system memory and check them. */ + for (i = 0, ptr = buffer->ptr; i < size / (2 * sizeof(*ptr)); ++i) + ASSERT_EQ(ptr[i], i); + + /* Migrate memory to the device again. */ + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_MIGRATE, buffer, npages); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, npages); + + /* Check what the device read. */ + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) ASSERT_EQ(ptr[i], i); hmm_buffer_free(buffer); -- 2.20.1

...c +++ b/mm/hmm.c @@ -26,7 +26,6 @@ #include <linux/mmu_notifier.h> #include <linux/memory_hotplug.h> -#if IS_ENABLED(CONFIG_HMM_MIRROR) static const struct mmu_notifier_ops hmm_mmu_notifier_ops; /** @@ -1289,4 +1288,3 @@ long hmm_range_dma_unmap(struct hmm_range *range, return cpages; } EXPORT_SYMBOL(hmm_range_dma_unmap); -#endif /* IS_ENABLED(CONFIG_HMM_MIRROR) */ -- 2.20.1

These patches apply to linux-5.8.0-rc1. Patches 1-3 should probably go into 5.8, the others can be queued for 5.9. Patches 4-6 improve the HMM self tests. Patch 7-8 prepare nouveau for the meat of this series which adds support and testing for compound page mapping of system memory (patches 9-11) and compound page migration to device private memory (patches 12-16). Since these changes are split