On Fri, 2008-05-23 at 21:31 +0100, Alan Cox wrote:
quoted text
> It saves us a lot of memory in several cases where the drivers hang onto
> the firmware so I definitely think we should fix the folks assuming they
> can widdle on the firmware. That doesn't look hard to do and could save
> 50K+ on many systems.

Hm, ok. Do you see any better fix than this for cxusb.c ?

--- cxusb.c~	2008-04-13 13:38:13.000000000 +0100
+++ cxusb.c	2008-05-23 22:01:31.000000000 +0100
@@ -23,6 +23,8 @@
  *
  * see Documentation/dvb/README.dvb-usb for more information
  */
+#include <linux/vmalloc.h>
+
 #include "cxusb.h"
 
 #include "cx22702.h"
@@ -695,12 +697,26 @@ static int bluebird_patch_dvico_firmware
 
 		if (fw->data[idoff] == (USB_VID_DVICO & 0xff) &&
 		    fw->data[idoff + 1] == USB_VID_DVICO >> 8) {
-			fw->data[idoff + 2] =
+			struct firmware new_fw;
+			u8 *new_fw_data = vmalloc(fw->size);
+			int ret;
+
+			if (!new_fw_data)
+				return -ENOMEM;
+
+			memcpy(new_fw_data, fw->data, fw->size);
+			new_fw.size = fw->size;
+			new_fw.data = fw->data;
+
+			new_fw_data[idoff + 2] =
 				le16_to_cpu(udev->descriptor.idProduct) + 1;
-			fw->data[idoff + 3] =
+			new_fw_data[idoff + 3] =
 				le16_to_cpu(udev->descriptor.idProduct) >> 8;
 
-			return usb_cypress_load_firmware(udev, fw, CYPRESS_FX2);
+			ret = usb_cypress_load_firmware(udev, &new_fw,
+							CYPRESS_FX2);
+			vfree(new_fw_data);
+			return ret;
 		}
 	}
 

-- 
dwmw2

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