Op 09-10-2024 om 20:08 schreef Mike Rapoport:
From: "Mike Rapoport (Microsoft)" <r...@kernel.org>

Using large pages to map text areas reduces iTLB pressure and improves
performance.

Extend execmem_alloc() with an ability to use huge pages with ROX
permissions as a cache for smaller allocations.

To populate the cache, a writable large page is allocated from vmalloc with
VM_ALLOW_HUGE_VMAP, filled with invalid instructions and then remapped as
ROX.

Portions of that large page are handed out to execmem_alloc() callers
without any changes to the permissions.

When the memory is freed with execmem_free() it is invalidated again so
that it won't contain stale instructions.

The cache is enabled when an architecture sets EXECMEM_ROX_CACHE flag in
definition of an execmem_range.

Signed-off-by: Mike Rapoport (Microsoft) <r...@kernel.org>
---
  include/linux/execmem.h |   2 +
  mm/execmem.c            | 317 +++++++++++++++++++++++++++++++++++++++-
  mm/internal.h           |   1 +
  mm/vmalloc.c            |   5 +
  4 files changed, 320 insertions(+), 5 deletions(-)
[...]
+static void execmem_cache_clean(struct work_struct *work)
+{
+       struct maple_tree *free_areas = &execmem_cache.free_areas;
+       struct mutex *mutex = &execmem_cache.mutex;
+       MA_STATE(mas, free_areas, 0, ULONG_MAX);
+       void *area;
+
+       mutex_lock(mutex);
+       mas_for_each(&mas, area, ULONG_MAX) {
+               size_t size;
+
No need to check for !area, because it is already guaranteed by the while loop condition (mas_for_each)
+               if (!area)
+                       continue;
+
+               size = mas_range_len(&mas);
+
+               if (IS_ALIGNED(size, PMD_SIZE) &&
+                   IS_ALIGNED(mas.index, PMD_SIZE)) {
+                       struct vm_struct *vm = find_vm_area(area);
+
+                       execmem_set_direct_map_valid(vm, true);
+                       mas_store_gfp(&mas, NULL, GFP_KERNEL);
+                       vfree(area);
+               }
+       }
+       mutex_unlock(mutex);
+}


Reply via email to