| Message ID | 149713137723.17377.8854203820807564559.stgit@dwillia2-desk3.amr.corp.intel.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
On Sat, Jun 10, 2017 at 02:49:37PM -0700, Dan Williams wrote: > diff --git a/include/linux/huge_mm.h b/include/linux/huge_mm.h > index c4706e2c3358..901ed3767d1b 100644 > --- a/include/linux/huge_mm.h > +++ b/include/linux/huge_mm.h > @@ -1,6 +1,8 @@ > #ifndef _LINUX_HUGE_MM_H > #define _LINUX_HUGE_MM_H > > +#include <linux/fs.h> > + It means <linux/mm.h> now depends on <linux/fs.h>. I don't think it's a good idea.
On Mon, Jun 12, 2017 at 5:07 AM, Kirill A. Shutemov <kirill.shutemov@linux.intel.com> wrote: > On Sat, Jun 10, 2017 at 02:49:37PM -0700, Dan Williams wrote: >> diff --git a/include/linux/huge_mm.h b/include/linux/huge_mm.h >> index c4706e2c3358..901ed3767d1b 100644 >> --- a/include/linux/huge_mm.h >> +++ b/include/linux/huge_mm.h >> @@ -1,6 +1,8 @@ >> #ifndef _LINUX_HUGE_MM_H >> #define _LINUX_HUGE_MM_H >> >> +#include <linux/fs.h> >> + > > It means <linux/mm.h> now depends on <linux/fs.h>. I don't think it's a > good idea. Seems to be ok as far as 0day-kbuild-robot is concerned. The alternative is to move vma_is_dax() out of line. I think transparent_hugepage_enabled() is called frequently enough to make it worth it to keep it inline.
On Mon, Jun 12, 2017 at 07:47:19AM -0700, Dan Williams wrote: > On Mon, Jun 12, 2017 at 5:07 AM, Kirill A. Shutemov > <kirill.shutemov@linux.intel.com> wrote: > > On Sat, Jun 10, 2017 at 02:49:37PM -0700, Dan Williams wrote: > >> diff --git a/include/linux/huge_mm.h b/include/linux/huge_mm.h > >> index c4706e2c3358..901ed3767d1b 100644 > >> --- a/include/linux/huge_mm.h > >> +++ b/include/linux/huge_mm.h > >> @@ -1,6 +1,8 @@ > >> #ifndef _LINUX_HUGE_MM_H > >> #define _LINUX_HUGE_MM_H > >> > >> +#include <linux/fs.h> > >> + > > > > It means <linux/mm.h> now depends on <linux/fs.h>. I don't think it's a > > good idea. > > Seems to be ok as far as 0day-kbuild-robot is concerned. The > alternative is to move vma_is_dax() out of line. I think > transparent_hugepage_enabled() is called frequently enough to make it > worth it to keep it inline. Yea, I played with moving vma_is_dax() to include/linux/mm.h instead, but ran into the issue where IS_DAX() is defined in include/linux/fs.h. So, any way we slice it we end up requiring both MM and FS includes for this to work. Since the way you have it here apparently works and passes zero-day, my vote is to just go with it.
diff --git a/include/linux/dax.h b/include/linux/dax.h index 1f6b6072af64..cbaf3d53d66b 100644 --- a/include/linux/dax.h +++ b/include/linux/dax.h @@ -151,11 +151,6 @@ static inline unsigned int dax_radix_order(void *entry) #endif int dax_pfn_mkwrite(struct vm_fault *vmf); -static inline bool vma_is_dax(struct vm_area_struct *vma) -{ - return vma->vm_file && IS_DAX(vma->vm_file->f_mapping->host); -} - static inline bool dax_mapping(struct address_space *mapping) { return mapping->host && IS_DAX(mapping->host); diff --git a/include/linux/fs.h b/include/linux/fs.h index 803e5a9b2654..5916ab3a12d5 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -18,6 +18,7 @@ #include <linux/bug.h> #include <linux/mutex.h> #include <linux/rwsem.h> +#include <linux/mm_types.h> #include <linux/capability.h> #include <linux/semaphore.h> #include <linux/fiemap.h> @@ -3042,6 +3043,11 @@ static inline bool io_is_direct(struct file *filp) return (filp->f_flags & O_DIRECT) || IS_DAX(filp->f_mapping->host); } +static inline bool vma_is_dax(struct vm_area_struct *vma) +{ + return vma->vm_file && IS_DAX(vma->vm_file->f_mapping->host); +} + static inline int iocb_flags(struct file *file) { int res = 0; diff --git a/include/linux/huge_mm.h b/include/linux/huge_mm.h index c4706e2c3358..901ed3767d1b 100644 --- a/include/linux/huge_mm.h +++ b/include/linux/huge_mm.h @@ -1,6 +1,8 @@ #ifndef _LINUX_HUGE_MM_H #define _LINUX_HUGE_MM_H +#include <linux/fs.h> + extern int do_huge_pmd_anonymous_page(struct vm_fault *vmf); extern int copy_huge_pmd(struct mm_struct *dst_mm, struct mm_struct *src_mm, pmd_t *dst_pmd, pmd_t *src_pmd, unsigned long addr, @@ -92,6 +94,9 @@ static inline bool transparent_hugepage_enabled(struct vm_area_struct *vma) if (transparent_hugepage_flags & (1 << TRANSPARENT_HUGEPAGE_FLAG)) return true; + if (vma_is_dax(vma)) + return true; + if (transparent_hugepage_flags & (1 << TRANSPARENT_HUGEPAGE_REQ_MADV_FLAG)) /* check vma flags */;
The madvise policy for transparent huge pages is meant to avoid unwanted allocations of transparent huge pages. It allows a policy of disabling the extra memory pressure and effort to arrange for a huge page when it is not needed. DAX by definition never incurs this overhead since it is statically allocated. The policy choice makes even less sense for device-dax which tries to guarantee a given tlb-fault size. Specifically, the following setting: echo never > /sys/kernel/mm/transparent_hugepage/enabled ...violates that guarantee and silently disables all device-dax instances with a 2M or 1G alignment. So, let's avoid that non-obvious side effect by force enabling thp for dax mappings in all cases. It is worth noting that the reason this uses vma_is_dax(), and the resulting header include changes, is that previous attempts to add a VM_DAX flag were NAKd. Cc: Jan Kara <jack@suse.cz> Cc: Christoph Hellwig <hch@lst.de> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Ross Zwisler <ross.zwisler@linux.intel.com> Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com> --- include/linux/dax.h | 5 ----- include/linux/fs.h | 6 ++++++ include/linux/huge_mm.h | 5 +++++ 3 files changed, 11 insertions(+), 5 deletions(-)