From ac645c6469dcc5a4bdbce8ed98e40182e8e54140 Mon Sep 17 00:00:00 2001 From: Vineet Gupta Date: Tue, 21 May 2013 15:25:11 +0530 Subject: ARC: copy_(to|from)_user() to honor usermode-access permissions commit a950549c675f2c8c504469dec7d780da8a6433dc upstream. This manifested as grep failing psuedo-randomly: -------------->8--------------------- [ARCLinux]$ ip address show lo | grep inet [ARCLinux]$ ip address show lo | grep inet [ARCLinux]$ ip address show lo | grep inet [ARCLinux]$ [ARCLinux]$ ip address show lo | grep inet inet 127.0.0.1/8 scope host lo -------------->8--------------------- ARC700 MMU provides fully orthogonal permission bits per page: Ur, Uw, Ux, Kr, Kw, Kx The user mode page permission templates used to have all Kernel mode access bits enabled. This caused a tricky race condition observed with uClibc buffered file read and UNIX pipes. 1. Read access to an anon mapped page in libc .bss: write-protected zero_page mapped: TLB Entry installed with Ur + K[rwx] 2. grep calls libc:getc() -> buffered read layer calls read(2) with the internal read buffer in same .bss page. The read() call is on STDIN which has been redirected to a pipe. read(2) => sys_read() => pipe_read() => copy_to_user() 3. Since page has Kernel-write permission (despite being user-mode write-protected), copy_to_user() suceeds w/o taking a MMU TLB-Miss Exception (page-fault for ARC). core-MM is unaware that kernel erroneously wrote to the reserved read-only zero-page (BUG #1) 4. Control returns to userspace which now does a write to same .bss page Since Linux MM is not aware that page has been modified by kernel, it simply reassigns a new writable zero-init page to mapping, loosing the prior write by kernel - effectively zero'ing out the libc read buffer under the hood - hence grep doesn't see right data (BUG #2) The fix is to make all kernel-mode access permissions mirror the user-mode ones. Note that the kernel still has full access to pages, when accessed directly (w/o MMU) - this fix ensures that kernel-mode access in copy_to_from() path uses the same faulting access model as for pure user accesses to keep MM fully aware of page state. The issue is peudo-random because it only shows up if the TLB entry installed in #1 is present at the time of #3. If it is evicted out, due to TLB pressure or some-such, then copy_to_user() does take a TLB Miss Exception, with a routine write-to-anon COW processing installing a fresh page for kernel writes and also usable as it is in userspace. Further the issue was dormant for so long as it depends on where the libc internal read buffer (in .bss) is mapped at runtime. If it happens to reside in file-backed data mapping of libc (in the page-aligned slack space trailing the file backed data), loader zero padding the slack space, does the early cow page replacement, setting things up at the very beginning itself. With gcc 4.8 based builds, the libc buffer got pushed out to a real anon mapping which triggers the issue. Reported-by: Anton Kolesov Signed-off-by: Vineet Gupta Signed-off-by: Greg Kroah-Hartman --- arch/arc/include/asm/pgtable.h | 26 +++++++++++++++----------- 1 file changed, 15 insertions(+), 11 deletions(-) (limited to 'arch/arc/include/asm/pgtable.h') diff --git a/arch/arc/include/asm/pgtable.h b/arch/arc/include/asm/pgtable.h index b7e36684c091..8ca472c2f5ca 100644 --- a/arch/arc/include/asm/pgtable.h +++ b/arch/arc/include/asm/pgtable.h @@ -57,9 +57,9 @@ #define _PAGE_ACCESSED (1<<1) /* Page is accessed (S) */ #define _PAGE_CACHEABLE (1<<2) /* Page is cached (H) */ -#define _PAGE_EXECUTE (1<<3) /* Page has user execute perm (H) */ -#define _PAGE_WRITE (1<<4) /* Page has user write perm (H) */ -#define _PAGE_READ (1<<5) /* Page has user read perm (H) */ +#define _PAGE_U_EXECUTE (1<<3) /* Page has user execute perm (H) */ +#define _PAGE_U_WRITE (1<<4) /* Page has user write perm (H) */ +#define _PAGE_U_READ (1<<5) /* Page has user read perm (H) */ #define _PAGE_K_EXECUTE (1<<6) /* Page has kernel execute perm (H) */ #define _PAGE_K_WRITE (1<<7) /* Page has kernel write perm (H) */ #define _PAGE_K_READ (1<<8) /* Page has kernel perm (H) */ @@ -72,9 +72,9 @@ /* PD1 */ #define _PAGE_CACHEABLE (1<<0) /* Page is cached (H) */ -#define _PAGE_EXECUTE (1<<1) /* Page has user execute perm (H) */ -#define _PAGE_WRITE (1<<2) /* Page has user write perm (H) */ -#define _PAGE_READ (1<<3) /* Page has user read perm (H) */ +#define _PAGE_U_EXECUTE (1<<1) /* Page has user execute perm (H) */ +#define _PAGE_U_WRITE (1<<2) /* Page has user write perm (H) */ +#define _PAGE_U_READ (1<<3) /* Page has user read perm (H) */ #define _PAGE_K_EXECUTE (1<<4) /* Page has kernel execute perm (H) */ #define _PAGE_K_WRITE (1<<5) /* Page has kernel write perm (H) */ #define _PAGE_K_READ (1<<6) /* Page has kernel perm (H) */ @@ -93,7 +93,8 @@ #endif /* Kernel allowed all permissions for all pages */ -#define _K_PAGE_PERMS (_PAGE_K_EXECUTE | _PAGE_K_WRITE | _PAGE_K_READ) +#define _K_PAGE_PERMS (_PAGE_K_EXECUTE | _PAGE_K_WRITE | _PAGE_K_READ | \ + _PAGE_GLOBAL | _PAGE_PRESENT) #ifdef CONFIG_ARC_CACHE_PAGES #define _PAGE_DEF_CACHEABLE _PAGE_CACHEABLE @@ -106,7 +107,11 @@ * -by default cached, unless config otherwise * -present in memory */ -#define ___DEF (_PAGE_PRESENT | _K_PAGE_PERMS | _PAGE_DEF_CACHEABLE) +#define ___DEF (_PAGE_PRESENT | _PAGE_DEF_CACHEABLE) + +#define _PAGE_READ (_PAGE_U_READ | _PAGE_K_READ) +#define _PAGE_WRITE (_PAGE_U_WRITE | _PAGE_K_WRITE) +#define _PAGE_EXECUTE (_PAGE_U_EXECUTE | _PAGE_K_EXECUTE) /* Set of bits not changed in pte_modify */ #define _PAGE_CHG_MASK (PAGE_MASK | _PAGE_ACCESSED | _PAGE_MODIFIED) @@ -125,11 +130,10 @@ * kernel vaddr space - visible in all addr spaces, but kernel mode only * Thus Global, all-kernel-access, no-user-access, cached */ -#define PAGE_KERNEL __pgprot(___DEF | _PAGE_GLOBAL) +#define PAGE_KERNEL __pgprot(_K_PAGE_PERMS | _PAGE_DEF_CACHEABLE) /* ioremap */ -#define PAGE_KERNEL_NO_CACHE __pgprot(_PAGE_PRESENT | _K_PAGE_PERMS | \ - _PAGE_GLOBAL) +#define PAGE_KERNEL_NO_CACHE __pgprot(_K_PAGE_PERMS) /************************************************************************** * Mapping of vm_flags (Generic VM) to PTE flags (arch specific) -- cgit v1.2.3