Skip to contentSkip to conversation

staging: ft1000: Copy from user into correct data

PATCH In review Steven Rostedt 3 participants 3 messages Nov 11, 2010 – Nov 16, 2010 Staging

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 staging: ft1000: Copy from user into correct data

While doing a ktest.pl I used a MIN_CONFIG that had STAGING enabled, and a randconfig with CONFIG_DEBUG_STRICT_USER_COPY_CHECKS enabled caught the following bug:

In file included from /home/rostedt/work/autotest/nobackup/linux-test.git/arch/x86/include/asm/uaccess.h:571:0,

                 from /home/rostedt/work/autotest/nobackup/linux-test.git/include/linux/poll.h:14,
                 from /home/rostedt/work/autotest/nobackup/linux-test.git/drivers/staging/ft1000/ft1000-usb/ft1000_chdev.c:32:
In function 'copy_from_user',
    inlined from 'ft1000_ChIoctl' at /home/rostedt/work/autotest/nobackup/linux-test.git/drivers/staging/ft1000/ft1000-usb/ft1000_chdev.c:702:36:

/home/rostedt/work/autotest/nobackup/linux-test.git/arch/x86/include/asm/uaccess_32.h:212:26: error: call to 'copy_from_user_overflow' declared with attribute error: copy_from_user() buffer size is not provably correct

Looking at the code it was obvious what the problem was. The pointer dpram_data was being allocated but the address was being written to. Looking at the comment above the code shows that it use to write into an element of that pointer where the '&' is appropriate. But now that it writes to the pointer itself, we need to remove the '&' otherwise we write over the pointer and not into the data it points to.

drivers/staging/ft1000/ft1000-usb/ft1000_chdev.c +1 -1
index 87a6487..8e8197d 100644
@@ -699,7 +699,7 @@ static long ft1000_ChIoctl (struct file *File, unsigned int Command,
 			break;
 
                 //if ( copy_from_user(&(dpram_command.dpram_blk), (PIOCTL_DPRAM_BLK)Argument, msgsz+2) ) {
-                if ( copy_from_user(&dpram_data, argp, msgsz+2) ) {
+                if ( copy_from_user(dpram_data, argp, msgsz+2) ) {
                     DEBUG("FT1000:ft1000_ChIoctl: copy fault occurred\n");
                     result = -EFAULT;
                 }
#1 lore raw

Hi Steven,

Quoted 12 lines · Steven Rostedt

While doing a ktest.pl I used a MIN_CONFIG that had STAGING enabled, and a randconfig with CONFIG_DEBUG_STRICT_USER_COPY_CHECKS enabled caught the following bug:

In file included from /home/rostedt/work/autotest/nobackup/linux-test.git/arch/x86/include/asm/uaccess.h:571:0,
                from /home/rostedt/work/autotest/nobackup/linux-test.git/include/linux/poll.h:14,
                from /home/rostedt/work/autotest/nobackup/linux-test.git/drivers/staging/ft1000/ft1000-usb/ft1000_chdev.c:32:
In function 'copy_from_user',
   inlined from 'ft1000_ChIoctl' at /home/rostedt/work/autotest/nobackup/linux-test.git/drivers/staging/ft1000/ft1000-usb/ft1000_chdev.c:702:36:
/home/rostedt/work/autotest/nobackup/linux-test.git/arch/x86/include/asm/uaccess_32.h:212:26: error: call to 'copy_from_user_overflow' declared with attribute error: copy_from_user() buffer size is not provably correct

Quoted 6 lines

Looking at the code it was obvious what the problem was. The pointer dpram_data was being allocated but the address was being written to. Looking at the comment above the code shows that it use to write into an element of that pointer where the '&' is appropriate. But now that it writes to the pointer itself, we need to remove the '&' otherwise we write over the pointer and not into the data it points to.

Good catch but anyway this interface is removed. Already send patches to Greg-KH but there was some patch problem so they're not applied to next yet. Thanks for effort.

Quoted 11 lines
drivers/staging/ft1000/ft1000-usb/ft1000_chdev.c +0 -0
index 87a6487..8e8197d 100644

drivers/staging/ft1000/ft1000-usb/ft1000_chdev.c static long ft1000_ChIoctl (struct file *File, unsigned int Command, · line 705

Quoted 8 lines
drivers/staging/ft1000/ft1000-usb/ft1000_chdev.c +1 -1
@@ -699,7 +699,7 @@ static long ft1000_ChIoctl (struct file *File, unsigned int Command,
                        break;
 
                 //if ( copy_from_user(&(dpram_command.dpram_blk), (PIOCTL_DPRAM_BLK)Argument, msgsz+2) ) {
-                if ( copy_from_user(&dpram_data, argp, msgsz+2) ) {
+                if ( copy_from_user(dpram_data, argp, msgsz+2) ) {
                     DEBUG("FT1000:ft1000_ChIoctl: copy fault occurred\n");
                     result = -EFAULT;
                 }

thanks,

marek

Signature · 10 lines
as simple and primitive as possible
-------------------------------------------------
Marek Belisko - OPEN-NANDRA
Freelance Developer

Ruska Nova Ves 219 | Presov, 08005 Slovak Republic
Tel: +421 915 052 184
skype: marekwhite
icq: 290551086
web: http://open-nandra.com
#2 lore raw
Quoted 15 lines · Belisko Marek

Hi Steven,

Quoted 12 lines · Steven Rostedt

While doing a ktest.pl I used a MIN_CONFIG that had STAGING enabled, and a randconfig with CONFIG_DEBUG_STRICT_USER_COPY_CHECKS enabled caught the following bug:

In file included from /home/rostedt/work/autotest/nobackup/linux-test.git/arch/x86/include/asm/uaccess.h:571:0,
                from /home/rostedt/work/autotest/nobackup/linux-test.git/include/linux/poll.h:14,
                from /home/rostedt/work/autotest/nobackup/linux-test.git/drivers/staging/ft1000/ft1000-usb/ft1000_chdev.c:32:
In function 'copy_from_user',
   inlined from 'ft1000_ChIoctl' at /home/rostedt/work/autotest/nobackup/linux-test.git/drivers/staging/ft1000/ft1000-usb/ft1000_chdev.c:702:36:
/home/rostedt/work/autotest/nobackup/linux-test.git/arch/x86/include/asm/uaccess_32.h:212:26: error: call to 'copy_from_user_overflow' declared with attribute error: copy_from_user() buffer size is not provably correct

Quoted 8 lines
Quoted 6 lines

Looking at the code it was obvious what the problem was. The pointer dpram_data was being allocated but the address was being written to. Looking at the comment above the code shows that it use to write into an element of that pointer where the '&' is appropriate. But now that it writes to the pointer itself, we need to remove the '&' otherwise we write over the pointer and not into the data it points to.

Good catch but anyway this interface is removed. Already send patches to Greg-KH but there was some patch problem so they're not applied to next yet.

But this patch is correct, and I don't have your other patches in my inbox, so I'm going to apply this one.

Please rebase your fixes on the next linux-next and all should be fine.

thanks,

greg k-h