Re: [PATCH 4.19 24/25] uprobes: ensure that uprobe->offset and ->ref_ctr_offset are properly aligned

[Oleg Nesterov <oleg@xxxxxxxxxx>]

On 06/09, Greg Kroah-Hartman wrote:
>
> On Wed, Jun 10, 2020 at 12:25:56AM +0530, Naresh Kamboju wrote:
> > > @@ -911,6 +907,15 @@ static int __uprobe_register(struct inod
> > >         if (offset > i_size_read(inode))
> > >                 return -EINVAL;
> > >
> > > +       /*
> > > +        * This ensures that copy_from_page(), copy_to_page() and
> > > +        * __update_ref_ctr() can't cross page boundary.
> > > +        */
> > > +       if (!IS_ALIGNED(offset, UPROBE_SWBP_INSN_SIZE))
> > > +               return -EINVAL;
> > > +       if (!IS_ALIGNED(ref_ctr_offset, sizeof(short)))
> >
> > stable-rc 4.19 build failure for x86_64, i386 and arm.
> > make -sk KBUILD_BUILD_USER=TuxBuild -C/linux -j16 ARCH=x86 HOSTCC=gcc
> > CC="sccache gcc" O=build
> >
> > 75 #
> > 76 In file included from ../kernel/events/uprobes.c:25:
> > 77 ../kernel/events/uprobes.c: In function ‘__uprobe_register’:
> > 78 ../kernel/events/uprobes.c:916:18: error: ‘ref_ctr_offset’
> > undeclared (first use in this function); did you mean
> > ‘per_cpu_offset’?
> > 79  916 | if (!IS_ALIGNED(ref_ctr_offset, sizeof(short)))
> > 80  | ^~~~~~~~~~~~~~
> > 81 ../include/linux/kernel.h:62:30: note: in definition of macro ‘IS_ALIGNED’
> > 82  62 | #define IS_ALIGNED(x, a) (((x) & ((typeof(x))(a) - 1)) == 0)
> > 83  | ^
> > 84 ../kernel/events/uprobes.c:916:18: note: each undeclared identifier
> > is reported only once for each function it appears in
> > 85  916 | if (!IS_ALIGNED(ref_ctr_offset, sizeof(short)))
> > 86  | ^~~~~~~~~~~~~~
> > 87 ../include/linux/kernel.h:62:30: note: in definition of macro ‘IS_ALIGNED’
> > 88  62 | #define IS_ALIGNED(x, a) (((x) & ((typeof(x))(a) - 1)) == 0)
> > 89  | ^
> > 90 make[3]: *** [../scripts/Makefile.build:304: kernel/events/uprobes.o] Error 1
> >
> > --
> > Linaro LKFT
> > https://lkft.linaro.org
>
> Good catch, my builders just caught it too :(
>
> 4.19, 4.14, 4.9, and 4.4 are all broken, I have a fix will test it and
> push out -rc2 for all of those with it in a bit, thanks.

Yes, SDT markers were added by 1cc33161a83d20b5462b1e93f95d3ce6388079ee in v4.20.

See the patch for v4.4 below. It changes uprobe_register(), not __uprobe_register()
to check IS_ALIGNED(offset, UPROBE_SWBP_INSN_SIZE) only.

Greg, please let me know if you want me to send the patches for 4.9/4.14/4.19.

Oleg.

--- a/kernel/events/uprobes.c
+++ b/kernel/events/uprobes.c
@@ -601,10 +601,6 @@ static int prepare_uprobe(struct uprobe *uprobe, struct file *file,
 	if (ret)
 		goto out;
 
-	/* uprobe_write_opcode() assumes we don't cross page boundary */
-	BUG_ON((uprobe->offset & ~PAGE_MASK) +
-			UPROBE_SWBP_INSN_SIZE > PAGE_SIZE);
-
 	smp_wmb(); /* pairs with rmb() in find_active_uprobe() */
 	set_bit(UPROBE_COPY_INSN, &uprobe->flags);
 
@@ -883,6 +879,13 @@ int uprobe_register(struct inode *inode, loff_t offset, struct uprobe_consumer *
 	if (offset > i_size_read(inode))
 		return -EINVAL;
 
+	/*
+	 * This ensures that copy_from_page() an copy_to_page()
+	 * can't cross page boundary.
+	 */
+	if (!IS_ALIGNED(offset, UPROBE_SWBP_INSN_SIZE))
+		return -EINVAL;
+
  retry:
 	uprobe = alloc_uprobe(inode, offset);
 	if (!uprobe)
@@ -1691,6 +1694,9 @@ static int is_trap_at_addr(struct mm_struct *mm, unsigned long vaddr)
 	uprobe_opcode_t opcode;
 	int result;
 
+	if (WARN_ON_ONCE(!IS_ALIGNED(vaddr, UPROBE_SWBP_INSN_SIZE)))
+		return -EINVAL;
+
 	pagefault_disable();
 	result = __copy_from_user_inatomic(&opcode, (void __user*)vaddr,
 							sizeof(opcode));