The patch titled Subject: selftests: mm: hugepage-vmemmap fails on 64K page size systems. has been added to the -mm mm-hotfixes-unstable branch. Its filename is selftests-mm-hugepage-vmemmap-fails-on-64k-page-size-systems.patch
This patch will shortly appear at https://git.kernel.org/pub/scm/linux/kernel/git/akpm/25-new.git/tree/patches...
This patch will later appear in the mm-hotfixes-unstable branch at git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm
Before you just go and hit "reply", please: a) Consider who else should be cc'ed b) Prefer to cc a suitable mailing list as well c) Ideally: find the original patch on the mailing list and do a reply-to-all to that, adding suitable additional cc's
*** Remember to use Documentation/process/submit-checklist.rst when testing your code ***
The -mm tree is included into linux-next via the mm-everything branch at git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm and is updated there every 2-3 working days
------------------------------------------------------ From: Donet Tom donettom@linux.vnet.ibm.com Subject: selftests: mm: hugepage-vmemmap fails on 64K page size systems. Date: Wed, 10 Jan 2024 14:03:35 +0530
The kernel sefltest mm/hugepage-vmemmap fails on architectures which has different page size other than 4K. In hugepage-vmemmap page size used is 4k so the pfn calculation will go wrong on systems which has different page size .The length of MAP_HUGETLB memory must be hugepage aligned but in hugepage-vmemmap map length is 2M so this will not get aligned if the system has differnet hugepage size.
Added psize() to get the page size and default_huge_page_size() to get the default hugepage size at run time, hugepage-vmemmap test pass on powerpc with 64K page size and x86 with 4K page size.
Result on powerpc without patch (page size 64K) *# ./hugepage-vmemmap Returned address is 0x7effff000000 whose pfn is 0 Head page flags (100000000) is invalid check_page_flags: Invalid argument *#
Result on powerpc with patch (page size 64K) *# ./hugepage-vmemmap Returned address is 0x7effff000000 whose pfn is 600 *#
Result on x86 with patch (page size 4K) *# ./hugepage-vmemmap Returned address is 0x7fc7c2c00000 whose pfn is 1dac00 *#
Link: https://lkml.kernel.org/r/3b3a3ae37ba21218481c482a872bbf7526031600.170486575... Fixes: b147c89cd429 ("selftests: vm: add a hugetlb test case") Signed-off-by: Donet Tom donettom@linux.vnet.ibm.com Reported-by: Geetika Moolchandani (geetika@linux.ibm.com) Tested-by: Geetika Moolchandani (geetika@linux.ibm.com) Cc: Muchun Song muchun.song@linux.dev Cc: stable@vger.kernel.org Signed-off-by: Andrew Morton akpm@linux-foundation.org ---
tools/testing/selftests/mm/hugepage-vmemmap.c | 29 +++++++++------- 1 file changed, 18 insertions(+), 11 deletions(-)
--- a/tools/testing/selftests/mm/hugepage-vmemmap.c~selftests-mm-hugepage-vmemmap-fails-on-64k-page-size-systems +++ a/tools/testing/selftests/mm/hugepage-vmemmap.c @@ -10,10 +10,7 @@ #include <unistd.h> #include <sys/mman.h> #include <fcntl.h> - -#define MAP_LENGTH (2UL * 1024 * 1024) - -#define PAGE_SIZE 4096 +#include "vm_util.h"
#define PAGE_COMPOUND_HEAD (1UL << 15) #define PAGE_COMPOUND_TAIL (1UL << 16) @@ -39,6 +36,9 @@ #define MAP_FLAGS (MAP_PRIVATE | MAP_ANONYMOUS | MAP_HUGETLB) #endif
+static size_t pagesize; +static size_t maplength; + static void write_bytes(char *addr, size_t length) { unsigned long i; @@ -56,7 +56,7 @@ static unsigned long virt_to_pfn(void *a if (fd < 0) return -1UL;
- lseek(fd, (unsigned long)addr / PAGE_SIZE * sizeof(pagemap), SEEK_SET); + lseek(fd, (unsigned long)addr / pagesize * sizeof(pagemap), SEEK_SET); read(fd, &pagemap, sizeof(pagemap)); close(fd);
@@ -86,7 +86,7 @@ static int check_page_flags(unsigned lon * this also verifies kernel has correctly set the fake page_head to tail * while hugetlb_free_vmemmap is enabled. */ - for (i = 1; i < MAP_LENGTH / PAGE_SIZE; i++) { + for (i = 1; i < maplength / pagesize; i++) { read(fd, &pageflags, sizeof(pageflags)); if ((pageflags & TAIL_PAGE_FLAGS) != TAIL_PAGE_FLAGS || (pageflags & HEAD_PAGE_FLAGS) == HEAD_PAGE_FLAGS) { @@ -106,18 +106,25 @@ int main(int argc, char **argv) void *addr; unsigned long pfn;
- addr = mmap(MAP_ADDR, MAP_LENGTH, PROT_READ | PROT_WRITE, MAP_FLAGS, -1, 0); + pagesize = psize(); + maplength = default_huge_page_size(); + if (!maplength) { + printf("Unable to determine huge page size\n"); + exit(1); + } + + addr = mmap(MAP_ADDR, maplength, PROT_READ | PROT_WRITE, MAP_FLAGS, -1, 0); if (addr == MAP_FAILED) { perror("mmap"); exit(1); }
/* Trigger allocation of HugeTLB page. */ - write_bytes(addr, MAP_LENGTH); + write_bytes(addr, maplength);
pfn = virt_to_pfn(addr); if (pfn == -1UL) { - munmap(addr, MAP_LENGTH); + munmap(addr, maplength); perror("virt_to_pfn"); exit(1); } @@ -125,13 +132,13 @@ int main(int argc, char **argv) printf("Returned address is %p whose pfn is %lx\n", addr, pfn);
if (check_page_flags(pfn) < 0) { - munmap(addr, MAP_LENGTH); + munmap(addr, maplength); perror("check_page_flags"); exit(1); }
/* munmap() length of MAP_HUGETLB memory must be hugepage aligned */ - if (munmap(addr, MAP_LENGTH)) { + if (munmap(addr, maplength)) { perror("munmap"); exit(1); } _
Patches currently in -mm which might be from donettom@linux.vnet.ibm.com are
selftests-mm-hugepage-vmemmap-fails-on-64k-page-size-systems.patch