Re: [Linux-fbdev-devel] [PATCH] viafb: Use sizeof struct rather thanpointer

From: Florian Tobias Schandinat
Date: Sat Nov 21 2009 - 18:59:28 EST


Hi,

Roel Kluin schrieb:
diff --git a/drivers/video/via/viafbdev.c b/drivers/video/via/viafbdev.c
index 56ec696..7b181e7 100644
--- a/drivers/video/via/viafbdev.c
+++ b/drivers/video/via/viafbdev.c
@@ -680,7 +680,7 @@ static int viafb_ioctl(struct fb_info *info, u_int cmd, u_long arg)
if (!viafb_gamma_table)
return -ENOMEM;
if (copy_from_user(viafb_gamma_table, argp,
- sizeof(viafb_gamma_table))) {
+ sizeof(*viafb_gamma_table))) {
kfree(viafb_gamma_table);
return -EFAULT;
}
@@ -694,7 +694,7 @@ static int viafb_ioctl(struct fb_info *info, u_int cmd, u_long arg)
return -ENOMEM;
viafb_get_gamma_table(viafb_gamma_table);
if (copy_to_user(argp, viafb_gamma_table,
- sizeof(viafb_gamma_table))) {
+ sizeof(*viafb_gamma_table))) {
kfree(viafb_gamma_table);
return -EFAULT;
}

I'm sorry but I fear your fix isn't correct. The reason is a few lines above:

u32 *viafb_gamma_table;
...
viafb_gamma_table = kmalloc(256 * sizeof(u32), GFP_KERNEL);

so probably the right solution would look like this:

copy_from_user(viafb_gamma_table, argp, 256 * sizeof(u32))
copy_to_user(argp, viafb_gamma_table, 256 * sizeof(u32))

However viafb has way too many private ioctls. I would be curious whether there exists any program that uses any of them so that they could be tested. As this ioctl didn't work very well I vote to remove it at least temporarily.

Thanks for highlighting this problem,


Florian Tobias Schandinat
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/