linux-kernel - [PATCH 3/5] [RFC] staging: rts5208: make len a u16 in rtsx_write_cfg

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-Id: <20200120074344.504-4-dja@axtens.net>
Date:   Mon, 20 Jan 2020 18:43:42 +1100
From:   Daniel Axtens <dja@...ens.net>
To:     kernel-hardening@...ts.openwall.com, linux-mm@...ck.org,
        keescook@...omium.org
Cc:     linux-kernel@...r.kernel.org, akpm@...ux-foundation.org,
        Daniel Axtens <dja@...ens.net>
Subject: [PATCH 3/5] [RFC] staging: rts5208: make len a u16 in rtsx_write_cfg_seq

A warning occurs when vzalloc is annotated in a subsequent patch to tell
the compiler that its parameter is an allocation size:

drivers/staging/rts5208/rtsx_chip.c: In function ‘rtsx_write_cfg_seq’:
drivers/staging/rts5208/rtsx_chip.c:1453:7: warning: argument 1 value ‘18446744073709551615’ exceeds maximum object size 9223372036854775807 [-Walloc-size-larger-than=]
  data = vzalloc(array_size(dw_len, 4));
  ~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

This occurs because len and dw_len are signed integers and the parameter to
array_size is a size_t. If dw_len is a negative integer, it will become a
very large positive number when cast to size_t. This could cause an
overflow, so array_size(), will return SIZE_MAX _at compile time_. gcc then
notices that this value is too large for an allocation and throws a
warning.

rtsx_write_cfg_seq is only called from write_cfg_byte in rtsx_scsi.c.
There, len is a u16. So make len a u16 in rtsx_write_cfg_seq too. This
means dw_len can never be negative, avoiding the potential overflow and the
warning.

This should not cause a functional change, but was compile tested only.

Signed-off-by: Daniel Axtens <dja@...ens.net>
---
 drivers/staging/rts5208/rtsx_chip.c | 2 +-
 drivers/staging/rts5208/rtsx_chip.h | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/staging/rts5208/rtsx_chip.c b/drivers/staging/rts5208/rtsx_chip.c
index 17c4131f5f62..4a8cbf7362f7 100644
--- a/drivers/staging/rts5208/rtsx_chip.c
+++ b/drivers/staging/rts5208/rtsx_chip.c
@@ -1432,7 +1432,7 @@ int rtsx_read_cfg_dw(struct rtsx_chip *chip, u8 func_no, u16 addr, u32 *val)
 }
 
 int rtsx_write_cfg_seq(struct rtsx_chip *chip, u8 func, u16 addr, u8 *buf,
-		       int len)
+		       u16 len)
 {
 	u32 *data, *mask;
 	u16 offset = addr % 4;
diff --git a/drivers/staging/rts5208/rtsx_chip.h b/drivers/staging/rts5208/rtsx_chip.h
index bac65784d4a1..9b0024557b7e 100644
--- a/drivers/staging/rts5208/rtsx_chip.h
+++ b/drivers/staging/rts5208/rtsx_chip.h
@@ -963,7 +963,7 @@ int rtsx_write_cfg_dw(struct rtsx_chip *chip,
 		      u8 func_no, u16 addr, u32 mask, u32 val);
 int rtsx_read_cfg_dw(struct rtsx_chip *chip, u8 func_no, u16 addr, u32 *val);
 int rtsx_write_cfg_seq(struct rtsx_chip *chip,
-		       u8 func, u16 addr, u8 *buf, int len);
+		       u8 func, u16 addr, u8 *buf, u16 len);
 int rtsx_read_cfg_seq(struct rtsx_chip *chip,
 		      u8 func, u16 addr, u8 *buf, int len);
 int rtsx_write_phy_register(struct rtsx_chip *chip, u8 addr, u16 val);
-- 
2.20.1