| Message ID | 20200924081830.GA1568157@mwanda (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | drm/i915: Uninitialized variable in i915_gem_object_map_page() | expand |
Hi Dan, On Thu, 24 Sep 2020 11:18:30 +0300 Dan Carpenter <dan.carpenter@oracle.com> wrote: > > The "i" iterator is never set to zero. This probably doesn't affect > testing because GCC sometimes initializes variables and also we have a > new pluggin to initialize stack variables to zero. > > Fixes: 7edd32a9e614 ("drm/i915: use vmap in i915_gem_object_map") > Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> > --- > Hi Andrew, this should probably go through the -mm tree and get folded > into the original patch. Added to linux-next today.
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_pages.c b/drivers/gpu/drm/i915/gem/i915_gem_pages.c index 90029ea83aed..12bedabc1daa 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_pages.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_pages.c @@ -266,6 +266,7 @@ static void *i915_gem_object_map_page(struct drm_i915_gem_object *obj, return NULL; } + i = 0; for_each_sgt_page(page, iter, obj->mm.pages) pages[i++] = page; vaddr = vmap(pages, n_pages, 0, pgprot); @@ -291,6 +292,7 @@ static void *i915_gem_object_map_pfn(struct drm_i915_gem_object *obj) return NULL; } + i = 0; for_each_sgt_daddr(addr, iter, obj->mm.pages) pfns[i++] = (iomap + addr) >> PAGE_SHIFT; vaddr = vmap_pfn(pfns, n_pfn, pgprot_writecombine(PAGE_KERNEL_IO));
The "i" iterator is never set to zero. This probably doesn't affect testing because GCC sometimes initializes variables and also we have a new pluggin to initialize stack variables to zero. Fixes: 7edd32a9e614 ("drm/i915: use vmap in i915_gem_object_map") Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> --- Hi Andrew, this should probably go through the -mm tree and get folded into the original patch. drivers/gpu/drm/i915/gem/i915_gem_pages.c | 2 ++ 1 file changed, 2 insertions(+)