linux-sparse.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Luc Van Oostenryck <luc.vanoostenryck@gmail.com>
To: linux-sparse@vger.kernel.org
Cc: Luc Van Oostenryck <luc.vanoostenryck@gmail.com>
Subject: [PATCH 3/3] fix linearize (*fun)()
Date: Thu, 21 Dec 2017 01:06:42 +0100	[thread overview]
Message-ID: <20171221000642.56954-4-luc.vanoostenryck@gmail.com> (raw)
In-Reply-To: <20171221000642.56954-1-luc.vanoostenryck@gmail.com>

A function call via a function pointer can be written like:
	fp(), (fp)() or (*fp)()
In the latter case the dereference is unneeded but legal and
idiomatic.

However, the linearization doesn't handle this unneeded deref
and leads to the generation of a load of the pointer:

	int foo(int a, int (*fun)(int))
	{
		(*fun)(a);
	}

gives something like:
	foo:
		load        %r2 <- 0[%arg2]
		call.32     %r3 <- %r2, %arg1
		ret.32      %r3

This happens because, at linearization, the deref is dropped
but only if the sub-expression is a symbol and the test for
node is not done.

Fix this by using is_func_type() to test the type of all call
expressions.

Signed-off-by: Luc Van Oostenryck <luc.vanoostenryck@gmail.com>
---
 linearize.c                             | 10 +++-------
 validation/linear/call-basic.c          |  1 -
 validation/linear/call-casted-pointer.c |  1 -
 validation/linear/call-indirect.c       |  1 -
 4 files changed, 3 insertions(+), 10 deletions(-)

diff --git a/linearize.c b/linearize.c
index 7ad348c00..2e146de7a 100644
--- a/linearize.c
+++ b/linearize.c
@@ -1291,13 +1291,9 @@ static pseudo_t linearize_call_expression(struct entrypoint *ep, struct expressi
 		add_symbol(&insn->fntypes, arg->ctype);
 	} END_FOR_EACH_PTR(arg);
 
-	if (fn->type == EXPR_PREOP) {
-		if (fn->unop->type == EXPR_SYMBOL) {
-			struct symbol *sym = fn->unop->symbol;
-			if (sym->ctype.base_type->type == SYM_FN)
-				fn = fn->unop;
-		}
-	}
+	if (fn->type == EXPR_PREOP && fn->op == '*' && is_func_type(fn->ctype))
+		fn = fn->unop;
+
 	if (fn->type == EXPR_SYMBOL) {
 		call = symbol_pseudo(ep, fn->symbol);
 	} else {
diff --git a/validation/linear/call-basic.c b/validation/linear/call-basic.c
index 235b443d2..60517e2e3 100644
--- a/validation/linear/call-basic.c
+++ b/validation/linear/call-basic.c
@@ -23,7 +23,6 @@ int builtin(int a)
 /*
  * check-name: basic function calls
  * check-command: test-linearize -Wno-decl $file
- * check-known-to-fail
  *
  * check-output-start
 symbol:
diff --git a/validation/linear/call-casted-pointer.c b/validation/linear/call-casted-pointer.c
index e66f69990..610d67482 100644
--- a/validation/linear/call-casted-pointer.c
+++ b/validation/linear/call-casted-pointer.c
@@ -23,7 +23,6 @@ int quz(void *a, void *fun)
 /*
  * check-name: call via casted function pointer
  * check-command: test-linearize -Wno-decl $file
- * check-known-to-fail
  *
  * check-output-ignore
  * check-output-excludes: load
diff --git a/validation/linear/call-indirect.c b/validation/linear/call-indirect.c
index 8ab78676b..f5f2adaf1 100644
--- a/validation/linear/call-indirect.c
+++ b/validation/linear/call-indirect.c
@@ -6,7 +6,6 @@ int g1(int (*fun)(void)) { return (*fun)(); }	// C99,C11 6.5.3.2p4
 /*
  * check-name: indirect calls
  * check-command: test-linearize -Wno-decl $file
- * check-known-to-fail
  *
  * check-output-ignore
  * check-output-excludes: load
-- 
2.15.0


      parent reply	other threads:[~2017-12-21  0:08 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-12-21  0:06 [PATCH 0/3] fix function calls via pointers Luc Van Oostenryck
2017-12-21  0:06 ` [PATCH 1/3] add testcases for the linearization of calls Luc Van Oostenryck
2017-12-21  0:06 ` [PATCH 2/3] simplify linearize_call_expression() Luc Van Oostenryck
2017-12-21  0:06 ` Luc Van Oostenryck [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20171221000642.56954-4-luc.vanoostenryck@gmail.com \
    --to=luc.vanoostenryck@gmail.com \
    --cc=linux-sparse@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).