Commit 6f3946b4 authored by Steven J. Magnani's avatar Steven J. Magnani Committed by Michal Simek

microblaze: Fix /dev/zero corruption from __clear_user()

A userland read of more than PAGE_SIZE bytes from /dev/zero results in
(a) not all of the bytes returned being zero, and
(b) memory corruption due to zeroing of bytes beyond the user buffer.

This is caused by improper constraints on the assembly __clear_user function.
The constrints don't indicate to the compiler that the pointer argument is
modified. Since the function is inline, this results in double-incrementing
of the pointer when __clear_user() is invoked through a multi-page read() of
/dev/zero.
Signed-off-by: default avatarSteven J. Magnani <steve@digidescorp.com>
Acked-by: default avatarMichal Simek <monstr@monstr.eu>
CC: stable@kernel.org
parent 6f205a4c
...@@ -120,16 +120,16 @@ static inline unsigned long __must_check __clear_user(void __user *to, ...@@ -120,16 +120,16 @@ static inline unsigned long __must_check __clear_user(void __user *to,
{ {
/* normal memset with two words to __ex_table */ /* normal memset with two words to __ex_table */
__asm__ __volatile__ ( \ __asm__ __volatile__ ( \
"1: sb r0, %2, r0;" \ "1: sb r0, %1, r0;" \
" addik %0, %0, -1;" \ " addik %0, %0, -1;" \
" bneid %0, 1b;" \ " bneid %0, 1b;" \
" addik %2, %2, 1;" \ " addik %1, %1, 1;" \
"2: " \ "2: " \
__EX_TABLE_SECTION \ __EX_TABLE_SECTION \
".word 1b,2b;" \ ".word 1b,2b;" \
".previous;" \ ".previous;" \
: "=r"(n) \ : "=r"(n), "=r"(to) \
: "0"(n), "r"(to) : "0"(n), "1"(to)
); );
return n; return n;
} }
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment