Re: [PATCH] mm/hmm.c: Allow VM_MIXEDMAP to work with hmm_range_fault

[Alistair Popple <apopple@nvidia.com>]

On Tuesday, 16 November 2021 3:48:42 PM AEDT Andrew Morton wrote:
> On Mon, 8 Nov 2021 18:58:01 +1100 Alistair Popple <apopple@nvidia.com> wrote:
> 
> > On Friday, 5 November 2021 11:25:57 PM AEDT Jason Gunthorpe wrote:
> > > On Thu, Nov 04, 2021 at 12:20:01PM +1100, Alistair Popple wrote:
> > > > hmm_range_fault() can be used instead of get_user_pages() for devices
> > > > which allow faulting however unlike get_user_pages() it will return an
> > > > error when used on a VM_MIXEDMAP range.
> > > > 
> > > > To make hmm_range_fault() more closely match get_user_pages() remove
> > > > this restriction. This requires dealing with the !ARCH_HAS_PTE_SPECIAL
> > > > case in hmm_vma_handle_pte(). Rather than replicating the logic of
> > > > vm_normal_page() call it directly and do a check for the zero pfn
> > > > similar to what get_user_pages() currently does.
> > > > 
> > > > Also add a test to hmm selftest to verify functionality.
> > > 
> > > Please add a fixes line
> > 
> > This has been the case since hmm_vma_get_pfns() was first introduced. Therefore
> > the fixes line is:
> > 
> > Fixes: da4c3c735ea4 ("mm/hmm/mirror: helper to snapshot CPU page table")
> > 
> > Should I send a v2 adding that tag or does Andrew normally pick these up along
> > with the Reviewed-by?
> 
> I got it.
> 
> It needed a bit of rework due to the newly-added !pte_devmap() change
> in hmm_vma_handle_pte().  Please check carefully?

Thanks. The rework looks good to me - changing pte_special() to
vm_insert_page() does not change the logic of the !pte_devmap() change. In
either case the value of pte_special() or vm_insert_page() is ignored when
pte_devmap()==True as intended by the !pte_devmap() change.

For good measure I rerun the HMM selftest as well.

> From: Alistair Popple <apopple@nvidia.com>
> Subject: mm/hmm.c: Allow VM_MIXEDMAP to work with hmm_range_fault
> 
> hmm_range_fault() can be used instead of get_user_pages() for devices
> which allow faulting however unlike get_user_pages() it will return an
> error when used on a VM_MIXEDMAP range.
> 
> To make hmm_range_fault() more closely match get_user_pages() remove
> this restriction. This requires dealing with the !ARCH_HAS_PTE_SPECIAL
> case in hmm_vma_handle_pte(). Rather than replicating the logic of
> vm_normal_page() call it directly and do a check for the zero pfn
> similar to what get_user_pages() currently does.
> 
> Also add a test to hmm selftest to verify functionality.
> 
> Link: https://lkml.kernel.org/r/20211104012001.2555676-1-apopple@nvidia.com
> Fixes: da4c3c735ea4 ("mm/hmm/mirror: helper to snapshot CPU page table")
> Signed-off-by: Alistair Popple <apopple@nvidia.com>
> Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
> Cc: Jerome Glisse <jglisse@redhat.com>
> Cc: John Hubbard <jhubbard@nvidia.com>
> Cc: Zi Yan <ziy@nvidia.com>
> Cc: Ralph Campbell <rcampbell@nvidia.com>
> Cc: Felix Kuehling <Felix.Kuehling@amd.com>
> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
> ---
> 
>  lib/test_hmm.c                         |   24 +++++++++++++
>  mm/hmm.c                               |    5 +-
>  tools/testing/selftests/vm/hmm-tests.c |   42 +++++++++++++++++++++++
>  3 files changed, 69 insertions(+), 2 deletions(-)
> 
> --- a/lib/test_hmm.c~mm-hmmc-allow-vm_mixedmap-to-work-with-hmm_range_fault
> +++ a/lib/test_hmm.c
> @@ -1086,9 +1086,33 @@ static long dmirror_fops_unlocked_ioctl(
>  	return 0;
>  }
>  
> +static int dmirror_fops_mmap(struct file *file, struct vm_area_struct *vma)
> +{
> +	unsigned long addr;
> +
> +	for (addr = vma->vm_start; addr < vma->vm_end; addr += PAGE_SIZE) {
> +		struct page *page;
> +		int ret;
> +
> +		page = alloc_page(GFP_KERNEL | __GFP_ZERO);
> +		if (!page)
> +			return -ENOMEM;
> +
> +		ret = vm_insert_page(vma, addr, page);
> +		if (ret) {
> +			__free_page(page);
> +			return ret;
> +		}
> +		put_page(page);
> +	}
> +
> +	return 0;
> +}
> +
>  static const struct file_operations dmirror_fops = {
>  	.open		= dmirror_fops_open,
>  	.release	= dmirror_fops_release,
> +	.mmap		= dmirror_fops_mmap,
>  	.unlocked_ioctl = dmirror_fops_unlocked_ioctl,
>  	.llseek		= default_llseek,
>  	.owner		= THIS_MODULE,
> --- a/mm/hmm.c~mm-hmmc-allow-vm_mixedmap-to-work-with-hmm_range_fault
> +++ a/mm/hmm.c
> @@ -300,7 +300,8 @@ static int hmm_vma_handle_pte(struct mm_
>  	 * Since each architecture defines a struct page for the zero page, just
>  	 * fall through and treat it like a normal page.
>  	 */
> -	if (pte_special(pte) && !pte_devmap(pte) &&
> +	if (!vm_normal_page(walk->vma, addr, pte) &&
> +	    !pte_devmap(pte) &&
>  	    !is_zero_pfn(pte_pfn(pte))) {
>  		if (hmm_pte_need_fault(hmm_vma_walk, pfn_req_flags, 0)) {
>  			pte_unmap(ptep);
> @@ -518,7 +519,7 @@ static int hmm_vma_walk_test(unsigned lo
>  	struct hmm_range *range = hmm_vma_walk->range;
>  	struct vm_area_struct *vma = walk->vma;
>  
> -	if (!(vma->vm_flags & (VM_IO | VM_PFNMAP | VM_MIXEDMAP)) &&
> +	if (!(vma->vm_flags & (VM_IO | VM_PFNMAP)) &&
>  	    vma->vm_flags & VM_READ)
>  		return 0;
>  
> --- a/tools/testing/selftests/vm/hmm-tests.c~mm-hmmc-allow-vm_mixedmap-to-work-with-hmm_range_fault
> +++ a/tools/testing/selftests/vm/hmm-tests.c
> @@ -1251,6 +1251,48 @@ TEST_F(hmm, anon_teardown)
>  /*
>   * Test memory snapshot without faulting in pages accessed by the device.
>   */
> +TEST_F(hmm, mixedmap)
> +{
> +	struct hmm_buffer *buffer;
> +	unsigned long npages;
> +	unsigned long size;
> +	unsigned char *m;
> +	int ret;
> +
> +	npages = 1;
> +	size = npages << self->page_shift;
> +
> +	buffer = malloc(sizeof(*buffer));
> +	ASSERT_NE(buffer, NULL);
> +
> +	buffer->fd = -1;
> +	buffer->size = size;
> +	buffer->mirror = malloc(npages);
> +	ASSERT_NE(buffer->mirror, NULL);
> +
> +
> +	/* Reserve a range of addresses. */
> +	buffer->ptr = mmap(NULL, size,
> +			   PROT_READ | PROT_WRITE,
> +			   MAP_PRIVATE,
> +			   self->fd, 0);
> +	ASSERT_NE(buffer->ptr, MAP_FAILED);
> +
> +	/* Simulate a device snapshotting CPU pagetables. */
> +	ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_SNAPSHOT, buffer, npages);
> +	ASSERT_EQ(ret, 0);
> +	ASSERT_EQ(buffer->cpages, npages);
> +
> +	/* Check what the device saw. */
> +	m = buffer->mirror;
> +	ASSERT_EQ(m[0], HMM_DMIRROR_PROT_READ);
> +
> +	hmm_buffer_free(buffer);
> +}
> +
> +/*
> + * Test memory snapshot without faulting in pages accessed by the device.
> + */
>  TEST_F(hmm2, snapshot)
>  {
>  	struct hmm_buffer *buffer;
> _
> 
> 
>