| Message ID | 20190220053040.10831-1-ira.weiny@intel.com (mailing list archive) |
|---|---|
| Headers | show |
| Series | Add FOLL_LONGTERM to GUP fast and use it | expand |
On Tue, Feb 19, 2019 at 09:30:33PM -0800, ira.weiny@intel.com wrote: > From: Ira Weiny <ira.weiny@intel.com> > > Resending these as I had only 1 minor comment which I believe we have covered > in this series. I was anticipating these going through the mm tree as they > depend on a cleanup patch there and the IB changes are very minor. But they > could just as well go through the IB tree. > > NOTE: This series depends on my clean up patch to remove the write parameter > from gup_fast_permitted()[1] > > HFI1, qib, and mthca, use get_user_pages_fast() due to it performance > advantages. These pages can be held for a significant time. But > get_user_pages_fast() does not protect against mapping of FS DAX pages. This I don't get - if you do lock down long term mappings performance of the actual get_user_pages call shouldn't matter to start with. What do I miss?
On Wed, Feb 20, 2019 at 07:19:30AM -0800, Christoph Hellwig wrote: > On Tue, Feb 19, 2019 at 09:30:33PM -0800, ira.weiny@intel.com wrote: > > From: Ira Weiny <ira.weiny@intel.com> > > > > Resending these as I had only 1 minor comment which I believe we have covered > > in this series. I was anticipating these going through the mm tree as they > > depend on a cleanup patch there and the IB changes are very minor. But they > > could just as well go through the IB tree. > > > > NOTE: This series depends on my clean up patch to remove the write parameter > > from gup_fast_permitted()[1] > > > > HFI1, qib, and mthca, use get_user_pages_fast() due to it performance > > advantages. These pages can be held for a significant time. But > > get_user_pages_fast() does not protect against mapping of FS DAX pages. > > This I don't get - if you do lock down long term mappings performance > of the actual get_user_pages call shouldn't matter to start with. > > What do I miss? A couple of points. First "longterm" is a relative thing and at this point is probably a misnomer. This is really flagging a pin which is going to be given to hardware and can't move. I've thought of a couple of alternative names but I think we have to settle on if we are going to use FL_LAYOUT or something else to solve the "longterm" problem. Then I think we can change the flag to a better name. Second, It depends on how often you are registering memory. I have spoken with some RDMA users who consider MR in the performance path... For the overall application performance. I don't have the numbers as the tests for HFI1 were done a long time ago. But there was a significant advantage. Some of which is probably due to the fact that you don't have to hold mmap_sem. Finally, architecturally I think it would be good for everyone to use *_fast. There are patches submitted to the RDMA list which would allow the use of *_fast (they reworking the use of mmap_sem) and as soon as they are accepted I'll submit a patch to convert the RDMA core as well. Also to this point others are looking to use *_fast.[2] As an asside, Jasons pointed out in my previous submission that *_fast and *_unlocked look very much the same. I agree and I think further cleanup will be coming. But I'm focused on getting the final solution for DAX at the moment. Ira
On Tue, Feb 19, 2019 at 09:30:33PM -0800, 'Ira Weiny' wrote: > From: Ira Weiny <ira.weiny@intel.com> > > Resending these as I had only 1 minor comment which I believe we have covered > in this series. I was anticipating these going through the mm tree as they > depend on a cleanup patch there and the IB changes are very minor. But they > could just as well go through the IB tree. > > NOTE: This series depends on my clean up patch to remove the write parameter > from gup_fast_permitted()[1] > > HFI1, qib, and mthca, use get_user_pages_fast() due to it performance > advantages. These pages can be held for a significant time. But > get_user_pages_fast() does not protect against mapping of FS DAX pages. > > Introduce FOLL_LONGTERM and use this flag in get_user_pages_fast() which > retains the performance while also adding the FS DAX checks. XDP has also > shown interest in using this functionality.[2] > > In addition we change get_user_pages() to use the new FOLL_LONGTERM flag and > remove the specialized get_user_pages_longterm call. > > [1] https://lkml.org/lkml/2019/2/11/237 > [2] https://lkml.org/lkml/2019/2/11/1789 Is there anything I need to do on this series or does anyone have any objections to it going into 5.1? And if so who's tree is it going to go through? Thanks, Ira > > Ira Weiny (7): > mm/gup: Replace get_user_pages_longterm() with FOLL_LONGTERM > mm/gup: Change write parameter to flags in fast walk > mm/gup: Change GUP fast to use flags rather than a write 'bool' > mm/gup: Add FOLL_LONGTERM capability to GUP fast > IB/hfi1: Use the new FOLL_LONGTERM flag to get_user_pages_fast() > IB/qib: Use the new FOLL_LONGTERM flag to get_user_pages_fast() > IB/mthca: Use the new FOLL_LONGTERM flag to get_user_pages_fast() > > arch/mips/mm/gup.c | 11 +- > arch/powerpc/kvm/book3s_64_mmu_hv.c | 4 +- > arch/powerpc/kvm/e500_mmu.c | 2 +- > arch/powerpc/mm/mmu_context_iommu.c | 4 +- > arch/s390/kvm/interrupt.c | 2 +- > arch/s390/mm/gup.c | 12 +- > arch/sh/mm/gup.c | 11 +- > arch/sparc/mm/gup.c | 9 +- > arch/x86/kvm/paging_tmpl.h | 2 +- > arch/x86/kvm/svm.c | 2 +- > drivers/fpga/dfl-afu-dma-region.c | 2 +- > drivers/gpu/drm/via/via_dmablit.c | 3 +- > drivers/infiniband/core/umem.c | 5 +- > drivers/infiniband/hw/hfi1/user_pages.c | 5 +- > drivers/infiniband/hw/mthca/mthca_memfree.c | 3 +- > drivers/infiniband/hw/qib/qib_user_pages.c | 8 +- > drivers/infiniband/hw/qib/qib_user_sdma.c | 2 +- > drivers/infiniband/hw/usnic/usnic_uiom.c | 9 +- > drivers/media/v4l2-core/videobuf-dma-sg.c | 6 +- > drivers/misc/genwqe/card_utils.c | 2 +- > drivers/misc/vmw_vmci/vmci_host.c | 2 +- > drivers/misc/vmw_vmci/vmci_queue_pair.c | 6 +- > drivers/platform/goldfish/goldfish_pipe.c | 3 +- > drivers/rapidio/devices/rio_mport_cdev.c | 4 +- > drivers/sbus/char/oradax.c | 2 +- > drivers/scsi/st.c | 3 +- > drivers/staging/gasket/gasket_page_table.c | 4 +- > drivers/tee/tee_shm.c | 2 +- > drivers/vfio/vfio_iommu_spapr_tce.c | 3 +- > drivers/vfio/vfio_iommu_type1.c | 3 +- > drivers/vhost/vhost.c | 2 +- > drivers/video/fbdev/pvr2fb.c | 2 +- > drivers/virt/fsl_hypervisor.c | 2 +- > drivers/xen/gntdev.c | 2 +- > fs/orangefs/orangefs-bufmap.c | 2 +- > include/linux/mm.h | 17 +- > kernel/futex.c | 2 +- > lib/iov_iter.c | 7 +- > mm/gup.c | 220 ++++++++++++-------- > mm/gup_benchmark.c | 5 +- > mm/util.c | 8 +- > net/ceph/pagevec.c | 2 +- > net/rds/info.c | 2 +- > net/rds/rdma.c | 3 +- > 44 files changed, 232 insertions(+), 180 deletions(-) > > -- > 2.20.1 >
From: Ira Weiny <ira.weiny@intel.com> Resending these as I had only 1 minor comment which I believe we have covered in this series. I was anticipating these going through the mm tree as they depend on a cleanup patch there and the IB changes are very minor. But they could just as well go through the IB tree. NOTE: This series depends on my clean up patch to remove the write parameter from gup_fast_permitted()[1] HFI1, qib, and mthca, use get_user_pages_fast() due to it performance advantages. These pages can be held for a significant time. But get_user_pages_fast() does not protect against mapping of FS DAX pages. Introduce FOLL_LONGTERM and use this flag in get_user_pages_fast() which retains the performance while also adding the FS DAX checks. XDP has also shown interest in using this functionality.[2] In addition we change get_user_pages() to use the new FOLL_LONGTERM flag and remove the specialized get_user_pages_longterm call. [1] https://lkml.org/lkml/2019/2/11/237 [2] https://lkml.org/lkml/2019/2/11/1789 Ira Weiny (7): mm/gup: Replace get_user_pages_longterm() with FOLL_LONGTERM mm/gup: Change write parameter to flags in fast walk mm/gup: Change GUP fast to use flags rather than a write 'bool' mm/gup: Add FOLL_LONGTERM capability to GUP fast IB/hfi1: Use the new FOLL_LONGTERM flag to get_user_pages_fast() IB/qib: Use the new FOLL_LONGTERM flag to get_user_pages_fast() IB/mthca: Use the new FOLL_LONGTERM flag to get_user_pages_fast() arch/mips/mm/gup.c | 11 +- arch/powerpc/kvm/book3s_64_mmu_hv.c | 4 +- arch/powerpc/kvm/e500_mmu.c | 2 +- arch/powerpc/mm/mmu_context_iommu.c | 4 +- arch/s390/kvm/interrupt.c | 2 +- arch/s390/mm/gup.c | 12 +- arch/sh/mm/gup.c | 11 +- arch/sparc/mm/gup.c | 9 +- arch/x86/kvm/paging_tmpl.h | 2 +- arch/x86/kvm/svm.c | 2 +- drivers/fpga/dfl-afu-dma-region.c | 2 +- drivers/gpu/drm/via/via_dmablit.c | 3 +- drivers/infiniband/core/umem.c | 5 +- drivers/infiniband/hw/hfi1/user_pages.c | 5 +- drivers/infiniband/hw/mthca/mthca_memfree.c | 3 +- drivers/infiniband/hw/qib/qib_user_pages.c | 8 +- drivers/infiniband/hw/qib/qib_user_sdma.c | 2 +- drivers/infiniband/hw/usnic/usnic_uiom.c | 9 +- drivers/media/v4l2-core/videobuf-dma-sg.c | 6 +- drivers/misc/genwqe/card_utils.c | 2 +- drivers/misc/vmw_vmci/vmci_host.c | 2 +- drivers/misc/vmw_vmci/vmci_queue_pair.c | 6 +- drivers/platform/goldfish/goldfish_pipe.c | 3 +- drivers/rapidio/devices/rio_mport_cdev.c | 4 +- drivers/sbus/char/oradax.c | 2 +- drivers/scsi/st.c | 3 +- drivers/staging/gasket/gasket_page_table.c | 4 +- drivers/tee/tee_shm.c | 2 +- drivers/vfio/vfio_iommu_spapr_tce.c | 3 +- drivers/vfio/vfio_iommu_type1.c | 3 +- drivers/vhost/vhost.c | 2 +- drivers/video/fbdev/pvr2fb.c | 2 +- drivers/virt/fsl_hypervisor.c | 2 +- drivers/xen/gntdev.c | 2 +- fs/orangefs/orangefs-bufmap.c | 2 +- include/linux/mm.h | 17 +- kernel/futex.c | 2 +- lib/iov_iter.c | 7 +- mm/gup.c | 220 ++++++++++++-------- mm/gup_benchmark.c | 5 +- mm/util.c | 8 +- net/ceph/pagevec.c | 2 +- net/rds/info.c | 2 +- net/rds/rdma.c | 3 +- 44 files changed, 232 insertions(+), 180 deletions(-)