drm/mgag200: fix memory leak

Commit Message

Sudip Mukherjee Sept. 14, 2015, 10:29 a.m. UTC

On Mon, Sep 14, 2015 at 08:05:37PM +1000, Dave Airlie wrote:
> (this time with correct email address).
> 
> On 14 September 2015 at 20:04, Dave Airlie <[e-mail]> wrote:
> >>
> >>
> >> There's a new regression: v4.3-rc1 crashes on bootup on non-supported hardware, if
> >> CONFIG_DRM_MGAG200=y (built into the kernel).
> >
> > Archit, I'm guessing this is some fallout from the fbdev changes.
> >
> > There is no reason we should need CONFIG_FB_LITTLE_ENDIAN I don't think.
I am assuming v4.2 has worked for Ingo. So in that case I don't see any
change in fbdev/core/fbmem.c between 4.2 and 4.3-rc1. But in drm (I
almost know nothing about drm) commit e829d7ef9f17 changes few codes
related to the card revision. Can this help?

regards
sudip

Patch

diff --git a/drivers/gpu/drm/mgag200/mgag200_mode.c b/drivers/gpu/drm/mgag200/mgag200_mode.c
index c99d3fe..5cdfa53 100644
--- a/drivers/gpu/drm/mgag200/mgag200_mode.c
+++ b/drivers/gpu/drm/mgag200/mgag200_mode.c
@@ -1602,7 +1602,7 @@  static int mga_vga_mode_valid(struct drm_connector *connector,
 			if (mga_vga_calculate_mode_bandwidth(mode, bpp)
 				> (24400 * 1024))
 				return MODE_BANDWIDTH;
-		} else if (mdev->unique_rev_id == 0x02) {
+		} else if (mdev->unique_rev_id >= 0x02) {
 			if (mode->hdisplay > 1920)
 				return MODE_VIRTUAL_X;
 			if (mode->vdisplay > 1200)