sh: fix mmap_sem deadlock in sys_cacheflush

From: Adam Iwaniuk <[email protected]>

We can't be holding the mmap_sem while calling __flush_XXX_region/flush_icache_range because the flush can fault. If we fault on a user address, the page fault handler will try to take mmap_sem again. Since both places acquire the read lock, most of the time it succeeds. However, if another thread tries to acquire the write lock on the mmap_sem (e.g. mmap) in between the call to __flush_XXX_region/flush_icache_range and the fault, the down_read in do_page_fault will deadlock.

---
 arch/sh/kernel/sys_sh.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

index 8c6a350..4aa294c 100644
@@ -74,6 +74,7 @@ asmlinkage int sys_cacheflush(unsigned long addr, unsigned long len, int op)
 		up_read(&current->mm->mmap_sem);
 		return -EFAULT;
 	}
+	up_read(&current->mm->mmap_sem);
 
 	switch (op & CACHEFLUSH_D_PURGE) {
 		case CACHEFLUSH_D_INVAL:

@@ -90,6 +91,5 @@ asmlinkage int sys_cacheflush(unsigned long addr, unsigned long len, int op)
 	if (op & CACHEFLUSH_I)
 		flush_icache_range(addr, addr+len);
 
-	up_read(&current->mm->mmap_sem);
 	return 0;
 }

1.7.9.5