Re: lib/49745: tests/include/sys/t_bitops: improvement

[Tetsuya Isaki <isaki%pastel-flower.jp@localhost>]

The following reply was made to PR lib/49745; it has been noted by GNATS.

From: Tetsuya Isaki <isaki%pastel-flower.jp@localhost>
To: gnats-bugs%NetBSD.org@localhost
Cc: lib-bug-people%netbsd.org@localhost,
	gnats-admin%netbsd.org@localhost,
	netbsd-bugs%netbsd.org@localhost, christos%zoulas.com@localhost (Christos Zoulas)
Subject: Re: lib/49745: tests/include/sys/t_bitops: improvement
Date: Wed, 18 Mar 2015 23:38:27 +0900

 At Sun, 15 Mar 2015 10:32:48 -0400,
 Christos Zoulas wrote:
 > |  >  | >Description:
 > |  >  | The current ilog2()'s implementation consists of 32bit variable
 > |  >  | part, 64bit variable part and constant part.  But 'ilog2_basic'
 > |  >  | test checks only 64bit variable.  32bit variable and constant
 > |  >  | were not checked.
 > |  :
 > |  >  I would prefer if you put the constants in an array and loop instead
 > |  >  of inlining it.
 > |  
 > |  It can not be changed to loop because it tests
 > |  __builtin_constant_p() part of ilog2().  If all these inline tests
 > |  passed, ATF_TC_BODY(ilog2_const, tc) is compiled into
 > 
 > Ah, ok. Please put a comment to that effect.
 
 Ok, I'll add a comment as following:
 
 --- t_bitops.c.ORG	2015-03-18 23:31:26.000000000 +0900
 +++ t_bitops.c	2015-03-18 23:32:55.000000000 +0900
 @@ -216,6 +216,10 @@
  
  ATF_TC_BODY(ilog2_const, tc)
  {
 +	/*
 +	 * These inlines test __builtin_constant_p() part of ilog2()
 +	 * at compile time, so don't change it to loop.  PR lib/49745
 +	 */
  	ATF_REQUIRE(ilog2(0x0000000000000001ULL) == 0);
  	ATF_REQUIRE(ilog2(0x0000000000000002ULL) == 1);
  	ATF_REQUIRE(ilog2(0x0000000000000004ULL) == 2);
 
 ---
 Tetsuya Isaki <isaki%pastel-flower.jp@localhost / isaki%NetBSD.org@localhost>