Message ID | 20180308073159.1068-1-kraxel@redhat.com |
---|---|
State | New |
Headers | show |
Series | vga: fix region calculation | expand |
+-- On Thu, 8 Mar 2018, Gerd Hoffmann wrote --+
| Typically the scanline length and the line offset are identical.
| But case
But in case ...
| Cc: Ross Lagerwall <ross.lagerwall@citrix.com>
s/Cc:/Reported-by: Ross Lagerwall ...
| Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
| ---
| hw/display/vga.c | 2 ++
| 1 file changed, 2 insertions(+)
|
| diff --git a/hw/display/vga.c b/hw/display/vga.c
| index 28f298b342..72181330b8 100644
| --- a/hw/display/vga.c
| +++ b/hw/display/vga.c
| @@ -1483,6 +1483,8 @@ static void vga_draw_graphic(VGACommonState *s, int full_update)
|
| region_start = (s->start_addr * 4);
| region_end = region_start + (ram_addr_t)s->line_offset * height;
| + region_end += width * s->get_bpp(s) / 8; /* scanline length */
| + region_end -= s->line_offset;
| if (region_end > s->vbe_size) {
| /* wraps around (can happen with cirrus vbe modes) */
| region_start = 0;
It does help to fix a segmentation fault caused by OOB access.
Reviewed-by: Prasad J Pandit <pjp@fedoraproject.org>
Thank you.
--
Prasad J Pandit / Red Hat Product Security Team
47AF CE69 3A90 54AA 9045 1053 DD13 3D32 FE5B 041F
On 03/08/2018 07:31 AM, Gerd Hoffmann wrote: > Typically the scanline length and the line offset are identical. But > case they are not our calculation for region_end is incorrect. Using > line_offset is fine for all scanlines, except the last one where we have > to use the actual scanline length. > > Fixes: CVE-2018-<to-be-filled> > Cc: P J P <ppandit@redhat.com> > Cc: Ross Lagerwall <ross.lagerwall@citrix.com> > Signed-off-by: Gerd Hoffmann <kraxel@redhat.com> > --- Tested-by: Ross Lagerwall <ross.lagerwall@citrix.com> Thanks!
diff --git a/hw/display/vga.c b/hw/display/vga.c index 28f298b342..72181330b8 100644 --- a/hw/display/vga.c +++ b/hw/display/vga.c @@ -1483,6 +1483,8 @@ static void vga_draw_graphic(VGACommonState *s, int full_update) region_start = (s->start_addr * 4); region_end = region_start + (ram_addr_t)s->line_offset * height; + region_end += width * s->get_bpp(s) / 8; /* scanline length */ + region_end -= s->line_offset; if (region_end > s->vbe_size) { /* wraps around (can happen with cirrus vbe modes) */ region_start = 0;
Typically the scanline length and the line offset are identical. But case they are not our calculation for region_end is incorrect. Using line_offset is fine for all scanlines, except the last one where we have to use the actual scanline length. Fixes: CVE-2018-<to-be-filled> Cc: P J P <ppandit@redhat.com> Cc: Ross Lagerwall <ross.lagerwall@citrix.com> Signed-off-by: Gerd Hoffmann <kraxel@redhat.com> --- hw/display/vga.c | 2 ++ 1 file changed, 2 insertions(+)