Messages in this thread Patch in this message | | | Date | Sat, 13 Mar 2004 21:25:10 -0800 | From | "Barry K. Nathan" <> | Subject | [PATCH] (2.6.x) toshiba_acpi needs copy_from_user (fixes oops) |
| |
On kernels with the 4G/4G patch (like some of the recent kernels in Fedora Core 2 development), writing stuff to the /proc/acpi/toshiba/* files causes an oops. As it turns out, this is because the driver is accessing userspace data without first doing copy_from_user(). IOW, this is a bug in toshiba_acpi, not a bug in the 4G/4G patch.
Here's a patch to fix this bug. I've tested it on 2.6.4 + some patches from the FC kernels (including the 4G/4G patch) and it fixes my oopses. I have also tested it against vanilla 2.6.4 and I haven't encountered any regressions.
If there are any problems with this patch, let me know.
-Barry K. Nathan <barryn@pobox.com>
diff -ruN linux-2.6.4/drivers/acpi/toshiba_acpi.c linux-2.6.4-bkn1/drivers/acpi/toshiba_acpi.c --- linux-2.6.4/drivers/acpi/toshiba_acpi.c 2004-03-12 21:31:59.000000000 -0800 +++ linux-2.6.4-bkn1/drivers/acpi/toshiba_acpi.c 2004-03-12 22:27:07.000000000 -0800 @@ -41,6 +41,7 @@ #include <linux/init.h> #include <linux/types.h> #include <linux/proc_fs.h> +#include <asm/uaccess.h> #include <acpi/acpi_drivers.h> @@ -269,10 +270,18 @@ } static int -dispatch_write(struct file* file, const char* buffer, unsigned long count, - ProcItem* item) +dispatch_write(struct file* file, const char __user *buffer, + unsigned long count, ProcItem* item) { - return item->write_func(buffer, count); + char str[48] = {'\0'}; + + if (count > sizeof(str) - 1) + return count; + + if (copy_from_user(str, buffer, count)) + return -EFAULT; + + return item->write_func(str, count); } static char* - 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/
| |