* Re: [PATCH v2 2/4] mm: Probe for sub-page faults in fault_in_*()
[not found] <20211201193750.2097885-3-catalin.marinas@arm.com>
@ 2021-12-02 5:10 ` kernel test robot
0 siblings, 0 replies; only message in thread
From: kernel test robot @ 2021-12-02 5:10 UTC (permalink / raw)
To: Catalin Marinas; +Cc: llvm, kbuild-all
Hi Catalin,
I love your patch! Yet something to improve:
[auto build test ERROR on powerpc/next]
[also build test ERROR on arm64/for-next/core linus/master v5.16-rc3]
[cannot apply to hnaz-mm/master kdave/for-next next-20211201]
[If your patch is applied to the wrong git tree, kindly drop us a note.
And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch]
url: https://github.com/0day-ci/linux/commits/Catalin-Marinas/Avoid-live-lock-in-fault-in-uaccess-loops-with-sub-page-faults/20211202-034030
base: https://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux.git next
config: hexagon-randconfig-r045-20211129 (https://download.01.org/0day-ci/archive/20211202/202112021346.8kPeV8ve-lkp@intel.com/config)
compiler: clang version 14.0.0 (https://github.com/llvm/llvm-project 4b553297ef3ee4dc2119d5429adf3072e90fac38)
reproduce (this is a W=1 build):
wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross
chmod +x ~/bin/make.cross
# https://github.com/0day-ci/linux/commit/a25817805b46cbc8d2d7e9d36dfa91e5f8dbf5df
git remote add linux-review https://github.com/0day-ci/linux
git fetch --no-tags linux-review Catalin-Marinas/Avoid-live-lock-in-fault-in-uaccess-loops-with-sub-page-faults/20211202-034030
git checkout a25817805b46cbc8d2d7e9d36dfa91e5f8dbf5df
# save the config file to linux build tree
mkdir build_dir
COMPILER_INSTALL_PATH=$HOME/0day COMPILER=clang make.cross W=1 O=build_dir ARCH=hexagon SHELL=/bin/bash
If you fix the issue, kindly add following tag as appropriate
Reported-by: kernel test robot <lkp@intel.com>
All errors (new ones prefixed by >>):
>> mm/gup.c:1764:48: error: passing 'const char *' to parameter of type 'void *' discards qualifiers [-Werror,-Wincompatible-pointer-types-discards-qualifiers]
(min_size && probe_subpage_safe_writeable(uaddr, min_size)))
^~~~~
include/linux/uaccess.h:306:64: note: passing argument to parameter 'uaddr' here
static inline size_t probe_subpage_safe_writeable(void __user *uaddr,
^
mm/gup.c:1807:42: error: passing 'const char *' to parameter of type 'void *' discards qualifiers [-Werror,-Wincompatible-pointer-types-discards-qualifiers]
(min_size && probe_subpage_readable(start, min_size)))
^~~~~
include/linux/uaccess.h:320:58: note: passing argument to parameter 'uaddr' here
static inline size_t probe_subpage_readable(void __user *uaddr, size_t size)
^
2 errors generated.
vim +1764 mm/gup.c
1700
1701 /*
1702 * fault_in_safe_writeable - fault in an address range for writing
1703 * @uaddr: start of address range
1704 * @size: length of address range
1705 * @min_size: minimum size to be faulted in
1706 *
1707 * Faults in an address range using get_user_pages, i.e., without triggering
1708 * hardware page faults. This is primarily useful when we already know that
1709 * some or all of the pages in the address range aren't in memory.
1710 *
1711 * Other than fault_in_writeable(), this function is non-destructive.
1712 *
1713 * Note that we don't pin or otherwise hold the pages referenced that we fault
1714 * in. There's no guarantee that they'll stay in memory for any duration of
1715 * time.
1716 *
1717 * Returns the number of bytes not faulted in, like copy_to_user() and
1718 * copy_from_user().
1719 */
1720 size_t fault_in_safe_writeable(const char __user *uaddr, size_t size,
1721 size_t min_size)
1722 {
1723 unsigned long start = (unsigned long)untagged_addr(uaddr);
1724 unsigned long end, nstart, nend;
1725 struct mm_struct *mm = current->mm;
1726 struct vm_area_struct *vma = NULL;
1727 int locked = 0;
1728 size_t faulted_in = size;
1729
1730 nstart = start & PAGE_MASK;
1731 end = PAGE_ALIGN(start + size);
1732 if (end < nstart)
1733 end = 0;
1734 for (; nstart != end; nstart = nend) {
1735 unsigned long nr_pages;
1736 long ret;
1737
1738 if (!locked) {
1739 locked = 1;
1740 mmap_read_lock(mm);
1741 vma = find_vma(mm, nstart);
1742 } else if (nstart >= vma->vm_end)
1743 vma = vma->vm_next;
1744 if (!vma || vma->vm_start >= end)
1745 break;
1746 nend = end ? min(end, vma->vm_end) : vma->vm_end;
1747 if (vma->vm_flags & (VM_IO | VM_PFNMAP))
1748 continue;
1749 if (nstart < vma->vm_start)
1750 nstart = vma->vm_start;
1751 nr_pages = (nend - nstart) / PAGE_SIZE;
1752 ret = __get_user_pages_locked(mm, nstart, nr_pages,
1753 NULL, NULL, &locked,
1754 FOLL_TOUCH | FOLL_WRITE);
1755 if (ret <= 0)
1756 break;
1757 nend = nstart + ret * PAGE_SIZE;
1758 }
1759 if (locked)
1760 mmap_read_unlock(mm);
1761 if (nstart != end)
1762 faulted_in = min_t(size_t, nstart - start, size);
1763 if (faulted_in < min_size ||
> 1764 (min_size && probe_subpage_safe_writeable(uaddr, min_size)))
1765 return size;
1766 return size - faulted_in;
1767 }
1768 EXPORT_SYMBOL(fault_in_safe_writeable);
1769
---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all@lists.01.org
^ permalink raw reply [flat|nested] only message in thread