Tried to use coccinelle on the linux kernel for the first time. It
seems to get things mostly right, but not quite. Here is the semantic
patch:
@@ expression E; @@
-down_read(&E->mmap_sem)
+mm_read_lock(E)
@@ expression E; @@
-down_read_trylock(&E->mmap_sem)
+mm_read_trylock(E)
@@ expression E; @@
-up_read(&E->mmap_sem)
+mm_read_unlock(E)
@@ expression E; @@
-down_write(&E->mmap_sem)
+mm_write_lock(E)
@@ expression E; @@
-down_write_trylock(&E->mmap_sem)
+mm_write_trylock(E)
@@ expression E; @@
-up_write(&E->mmap_sem)
+mm_write_unlock(E)
I run into problems whenever handling system calls. Looks like the
SYSCALL_DEFINE5() macro and friends are causing problems. Here is my
attempt to handle things.
#define SYSCALL_DEFINE1(func, t1, a1) \
asmlinkage unsigned long func(t1 a1)
#define SYSCALL_DEFINE2(func, t1, a1, t2, a2) \
asmlinkage unsigned long func(t1 a1, t2 a2)
#define SYSCALL_DEFINE3(func, t1, a1, t2, a2, t3, a3) \
asmlinkage unsigned long func(t1 a1, t2 a2, t3 a3)
#define SYSCALL_DEFINE4(func, t1, a1, t2, a2, t3, a3, t4, a4) \
asmlinkage unsigned long func(t1 a1, t2 a2, t3 a3, t4 a4)
#define SYSCALL_DEFINE5(func, t1, a1, t2, a2, t3, a3, t4, a4, t5, a5) \
asmlinkage unsigned long func(t1 a1, t2 a2, t3 a3, t4 a4, t5 a5)
#define SYSCALL_DEFINE6(func, t1, a1, t2, a2, t3, a3, t4, a4, t5, a5, t6, a6) \
asmlinkage unsigned long func(t1 a1, t2 a2, t3 a3, t4 a4, t5 a5, t6 a6)
Should that be included into standard.h?
Jörn
--
It is a cliché that most clichés are true, but then, like most clichés,
that cliché is untrue.
-- Stephen Fry
_______________________________________________
Cocci mailing list
[email protected]
https://systeme.lip6.fr/mailman/listinfo/cocci
