lkml.org 
[lkml]   [2019]   [Jul]   [12]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
From
Date
SubjectRe: [PATCH] waitqueue: fix clang -Wuninitialized warnings
On Fri, Jul 12, 2019 at 9:54 AM Nathan Chancellor
<natechancellor@gmail.com> wrote:
>
> On Fri, Jul 12, 2019 at 09:45:06AM +0200, Arnd Bergmann wrote:
> > On Fri, Jul 12, 2019 at 2:49 AM Andrew Morton <akpm@linux-foundation.org> wrote:
> > > On Wed, 3 Jul 2019 10:10:55 +0200 Arnd Bergmann <arnd@arndb.de> wrote:
> >
> > > <scratches head>
> > >
> > > Surely clang is being extraordinarily dumb here?
> > >
> > > DECLARE_WAIT_QUEUE_HEAD_ONSTACK() is effectively doing
> > >
> > > struct wait_queue_head name = ({ __init_waitqueue_head(&name) ; name; })
> > >
> > > which is perfectly legitimate! clang has no business assuming that
> > > __init_waitqueue_head() will do any reads from the pointer which it was
> > > passed, nor can clang assume that __init_waitqueue_head() leaves any of
> > > *name uninitialized.
> > >
> > > Does it also warn if code does this?
> > >
> > > struct wait_queue_head name;
> > > __init_waitqueue_head(&name);
> > > name = name;
> > >
> > > which is equivalent, isn't it?
> >
> > No, it does not warn for this.
> >
> > I've tried a few more variants here: https://godbolt.org/z/ykSX0r
> >
> > What I think is going on here is a result of clang and gcc fundamentally
> > treating -Wuninitialized warnings differently. gcc tries to make the warnings
> > as helpful as possible, but given the NP-complete nature of this problem
> > it won't always get it right, and it traditionally allowed this syntax as a
> > workaround.
> >
> > int f(void)
> > {
> > int i = i; // tell gcc not to warn
> > return i;
> > }
> >
> > clang apparently implements the warnings in a way that is as
> > completely predictable (and won't warn in cases that it
> > doesn't completely understand), but decided as a result that the
> > gcc 'int i = i' syntax is bogus and it always warns about a variable
> > used in its own declaration that is later referenced, without looking
> > at whether the declaration does initialize it or not.
> >
> > > The proposed solution is, effectively, to open-code
> > > __init_waitqueue_head() at each DECLARE_WAIT_QUEUE_HEAD_ONSTACK()
> > > callsite. That's pretty unpleasant and calls for an explanatory
> > > comment at the __WAIT_QUEUE_HEAD_INIT_ONSTACK() definition site as well
> > > as a cautionary comment at the __init_waitqueue_head() definition so we
> > > can keep the two versions in sync as code evolves.
> >
> > Yes, makes sense.
> >
> > > Hopefully clang will soon be hit with the cluebat (yes?) and this
> > > change becomes obsolete in the quite short term. Surely 6-12 months
> > > from now nobody will be using the uncluebatted version of clang on
> > > contemporary kernel sources so we get to remove this nastiness again.
> > > Which makes me wonder whether we should merge it at all.
> >
> > Would it make you feel better to keep the current code but have an alternative
> > version guarded with e.g. "#if defined(__clang__ && (__clang_major__ <= 9)"?
> >
> > While it is probably a good idea to fix clang here, this is one of the last
> > issues that causes a significant difference between gcc and clang in build
> > testing with kernelci:
> > https://kernelci.org/build/next/branch/master/kernel/next-20190709/
> > I'm trying to get all the warnings fixed there so we can spot build-time
> > regressions more easily.
> >
> > Arnd
>
> I'm just spitballing here since I am about to go to sleep but could we
> do something like you did for bee20031772a ("disable -Wattribute-alias
> warning for SYSCALL_DEFINEx()") and disable the warning in
> DECLARE_WAIT_QUEUE_HEAD_ONSTACK only since we know it is not going to
> be a problem? That way, if/when Clang is fixed, we can just have the
> warning be disabled for older versions?

I managed to get that to work, but there are two problems:

- the __diag_ignore() infrastructure was never added for clang, so
I ended up copying a lot from gcc. There is probably a nicer way
to do this, but that would require a larger rework
- adding __diag_pop() between two variable declarations is seen as
a statement that causes a warning with both gcc and clang,
so I had to turn that warning off as well for all compilers, and at that
point it gets rather ugly in the macro.

Arnd

diff --git a/include/linux/compiler-clang.h b/include/linux/compiler-clang.h
index 333a6695a918..0d30c0489ad7 100644
--- a/include/linux/compiler-clang.h
+++ b/include/linux/compiler-clang.h
@@ -42,3 +42,31 @@
* compilers, like ICC.
*/
#define barrier() __asm__ __volatile__("" : : : "memory")
+
+/*
+ * Turn individual warnings and errors on and off locally, depending
+ * on version.
+ */
+#define __diag_clang(version, severity, s) \
+ __diag_clang_ ## version(__diag_clang_ ## severity s)
+
+/* Severity used in pragma directives */
+#define __diag_clang_ignore ignored
+#define __diag_clang_warn warning
+#define __diag_clang_error error
+
+#define __diag_str1(s) #s
+#define __diag_str(s) __diag_str1(s)
+#define __diag(s) _Pragma(__diag_str(clang diagnostic s))
+
+#if __clang_major__ >= 8
+#define __diag_clang_8(s) __diag(s)
+#else
+#define __diag_clang_8(s)
+#endif
+
+#if __clang_major__ >= 9
+#define __diag_clang_9(s) __diag(s)
+#else
+#define __diag_clang_9(s)
+#endif
diff --git a/include/linux/compiler-gcc.h b/include/linux/compiler-gcc.h
index e8579412ad21..c5f8d9ae0530 100644
--- a/include/linux/compiler-gcc.h
+++ b/include/linux/compiler-gcc.h
@@ -165,8 +165,16 @@
#define __diag_str(s) __diag_str1(s)
#define __diag(s) _Pragma(__diag_str(GCC diagnostic s))

+#if GCC_VERSION >= 40006
+#define __diag_GCC_4_6(s) __diag(s)
+#else
+#define __diag_GCC_4_6(s)
+#endif
+
#if GCC_VERSION >= 80000
#define __diag_GCC_8(s) __diag(s)
#else
#define __diag_GCC_8(s)
#endif
+
+#define __diag_clang(s...)
diff --git a/include/linux/wait.h b/include/linux/wait.h
index ddb959641709..0e33fe589f49 100644
--- a/include/linux/wait.h
+++ b/include/linux/wait.h
@@ -71,7 +71,12 @@ extern void __init_waitqueue_head(struct
wait_queue_head *wq_head, const char *n
# define __WAIT_QUEUE_HEAD_INIT_ONSTACK(name) \
({ init_waitqueue_head(&name); name; })
# define DECLARE_WAIT_QUEUE_HEAD_ONSTACK(name) \
- struct wait_queue_head name = __WAIT_QUEUE_HEAD_INIT_ONSTACK(name)
+ __diag_push(); \
+ __diag_ignore(clang, 8, "-Wuninitialized",
"https://godbolt.org/z/ykSX0r"); \
+ __diag_ignore(clang, 8, "-Wdeclaration-after-statement", "for
__diag_pop"); \
+ __diag_ignore(GCC, 4_6, "-Wdeclaration-after-statement", "for
__diag_pop"); \
+ struct wait_queue_head name = __WAIT_QUEUE_HEAD_INIT_ONSTACK(name); \
+ __diag_pop()
#else
# define DECLARE_WAIT_QUEUE_HEAD_ONSTACK(name) DECLARE_WAIT_QUEUE_HEAD(name)
#endif
\
 
 \ /
  Last update: 2019-07-12 16:53    [W:0.033 / U:8.176 seconds]
©2003-2018 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site