Re: BUG: KASAN: global-out-of-bounds in ata_exec_internal_sg+0x50f/0xc70

[Kees Cook <keescook@chromium.org>]

On Tue, Jul 16, 2019 at 11:28:29AM -0700, Nick Desaulniers wrote:
> On Wed, Jul 10, 2019 at 10:44 AM Jeffrin Thalakkottoor
> <jeffrin@rajagiritech.edu.in> wrote:
> >
> > hello all ,
> >
> > i encountered a KASAN bug related .    here are some related information...
> >
> >
> > -------------------x-----------------------------x------------------
> > [   30.037312] BUG: KASAN: global-out-of-bounds in
> > ata_exec_internal_sg+0x50f/0xc70
> > [   30.037447] Read of size 16 at addr ffffffff91f41f80 by task scsi_eh_1/149
> >
> >
> > [   30.039935] The buggy address belongs to the variable:
> > [   30.040059]  cdb.48319+0x0/0x40
> > (gdb) l *ata_exec_internal_sg+0x50f
> > 0xffffffff81c7b59f is in ata_exec_internal_sg (./include/linux/string.h:359).
> 
> So looks like ata_exec_internal_sg() is panic'ing when...
> 
> > 354 if (q_size < size)
> > 355 __read_overflow2();
> > 356 }
> > 357 if (p_size < size || q_size < size)
> > 358 fortify_panic(__func__);
> > 359 return __builtin_memcpy(p, q, size);

^^^ here, so within memcpy(), but after the "easy" sanity checks.

The only place where I see ata_exec_internal_sg() calling memcpy() is
here:

        /* prepare & issue qc */
        qc->tf = *tf;
        if (cdb)
                memcpy(qc->cdb, cdb, ATAPI_CDB_LEN);

the "16" is consistent with the report:

include/linux/ata.h:    ATAPI_CDB_LEN           = 16,

which matches the claim about the cdb variable from KASAN. And it's a
read, so "cdb" is wrong. Do you have a longer back trace? What called
ata_exec_internal_sg()?

ata_exec_internal() is the only caller of ata_exec_internal_sg(). Nearly
all callers of ata_exec_internal() pass a NULL cdb. Those that don't
are:

atapi_eh_tur()
	u8 cdb[ATAPI_CDB_LEN] = ...
atapi_eh_request_sense()
	u8 cdb[ATAPI_CDB_LEN] = ...

These two are on the static and correctly sized.

eject_tray()
        static const char cdb[ATAPI_CDB_LEN] = ...
zpodd_get_mech_type()
	static const char cdb[] = ...

These are both in rodata, and only the first is correctly sized. I
assume the following will fix it:


diff --git a/drivers/ata/libata-zpodd.c b/drivers/ata/libata-zpodd.c
index 173e6f2dd9af..eefda51f97d3 100644
--- a/drivers/ata/libata-zpodd.c
+++ b/drivers/ata/libata-zpodd.c
@@ -56,7 +56,7 @@ static enum odd_mech_type zpodd_get_mech_type(struct ata_device *dev)
 	unsigned int ret;
 	struct rm_feature_desc *desc;
 	struct ata_taskfile tf;
-	static const char cdb[] = {  GPCMD_GET_CONFIGURATION,
+	static const char cdb[ATAPI_CDB_LEN] = {  GPCMD_GET_CONFIGURATION,
 			2,      /* only 1 feature descriptor requested */
 			0, 3,   /* 3, removable medium feature */
 			0, 0, 0,/* reserved */



-- 
Kees Cook