grub2/0290-fbutil-Fix-integer-overflow.patch
Robbie Harwood 445c9bcb9d Font fixes (CVE-2022-2601 batch)
Signed-off-by: Robbie Harwood <rharwood@redhat.com>
2022-11-15 18:09:20 +00:00

84 lines
3.7 KiB
Diff

From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
From: Zhang Boyang <zhangboyang.id@gmail.com>
Date: Tue, 6 Sep 2022 03:03:21 +0800
Subject: [PATCH] fbutil: Fix integer overflow
Expressions like u64 = u32 * u32 are unsafe because their products are
truncated to u32 even if left hand side is u64. This patch fixes all
problems like that one in fbutil.
To get right result not only left hand side have to be u64 but it's also
necessary to cast at least one of the operands of all leaf operators of
right hand side to u64, e.g. u64 = u32 * u32 + u32 * u32 should be
u64 = (u64)u32 * u32 + (u64)u32 * u32.
For 1-bit bitmaps grub_uint64_t have to be used. It's safe because any
combination of values in (grub_uint64_t)u32 * u32 + u32 expression will
not overflow grub_uint64_t.
Other expressions like ptr + u32 * u32 + u32 * u32 are also vulnerable.
They should be ptr + (grub_addr_t)u32 * u32 + (grub_addr_t)u32 * u32.
This patch also adds a comment to grub_video_fb_get_video_ptr() which
says it's arguments must be valid and no sanity check is performed
(like its siblings in grub-core/video/fb/fbutil.c).
Signed-off-by: Zhang Boyang <zhangboyang.id@gmail.com>
Reviewed-by: Daniel Kiper <daniel.kiper@oracle.com>
(cherry picked from commit 50a11a81bc842c58962244a2dc86bbd31a426e12)
(cherry picked from commit 8fa75d647362c938c4cc302cf5945b31fb92c078)
---
grub-core/video/fb/fbutil.c | 4 ++--
include/grub/fbutil.h | 13 +++++++++----
2 files changed, 11 insertions(+), 6 deletions(-)
diff --git a/grub-core/video/fb/fbutil.c b/grub-core/video/fb/fbutil.c
index b98bb51fe8..25ef39f47d 100644
--- a/grub-core/video/fb/fbutil.c
+++ b/grub-core/video/fb/fbutil.c
@@ -67,7 +67,7 @@ get_pixel (struct grub_video_fbblit_info *source,
case 1:
if (source->mode_info->blit_format == GRUB_VIDEO_BLIT_FORMAT_1BIT_PACKED)
{
- int bit_index = y * source->mode_info->width + x;
+ grub_uint64_t bit_index = (grub_uint64_t) y * source->mode_info->width + x;
grub_uint8_t *ptr = source->data + bit_index / 8;
int bit_pos = 7 - bit_index % 8;
color = (*ptr >> bit_pos) & 0x01;
@@ -138,7 +138,7 @@ set_pixel (struct grub_video_fbblit_info *source,
case 1:
if (source->mode_info->blit_format == GRUB_VIDEO_BLIT_FORMAT_1BIT_PACKED)
{
- int bit_index = y * source->mode_info->width + x;
+ grub_uint64_t bit_index = (grub_uint64_t) y * source->mode_info->width + x;
grub_uint8_t *ptr = source->data + bit_index / 8;
int bit_pos = 7 - bit_index % 8;
*ptr = (*ptr & ~(1 << bit_pos)) | ((color & 0x01) << bit_pos);
diff --git a/include/grub/fbutil.h b/include/grub/fbutil.h
index 4205eb917f..78a1ab3b45 100644
--- a/include/grub/fbutil.h
+++ b/include/grub/fbutil.h
@@ -31,14 +31,19 @@ struct grub_video_fbblit_info
grub_uint8_t *data;
};
-/* Don't use for 1-bit bitmaps, addressing needs to be done at the bit level
- and it doesn't make sense, in general, to ask for a pointer
- to a particular pixel's data. */
+/*
+ * Don't use for 1-bit bitmaps, addressing needs to be done at the bit level
+ * and it doesn't make sense, in general, to ask for a pointer
+ * to a particular pixel's data.
+ *
+ * This function assumes that bounds checking has been done in previous phase
+ * and they are opted out in here.
+ */
static inline void *
grub_video_fb_get_video_ptr (struct grub_video_fbblit_info *source,
unsigned int x, unsigned int y)
{
- return source->data + y * source->mode_info->pitch + x * source->mode_info->bytes_per_pixel;
+ return source->data + (grub_addr_t) y * source->mode_info->pitch + (grub_addr_t) x * source->mode_info->bytes_per_pixel;
}
/* Advance pointer by VAL bytes. If there is no unaligned access available,