Bug 165114

Summary: ICE when rebuilding procps-3.2.5-6.3
Product: [Fedora] Fedora Reporter: Karsten Hopp <karsten>
Component: gccAssignee: Jakub Jelinek <jakub>
Status: CLOSED WONTFIX QA Contact:
Severity: medium Docs Contact:
Priority: medium    
Version: rawhideCC: kzak
Target Milestone: ---   
Target Release: ---   
Hardware: All   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2005-08-10 15:41:28 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Attachments:
Description Flags
Preprocessed source none

Description Karsten Hopp 2005-08-04 14:16:12 UTC 
From Bugzilla Helper:
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.8b3) Gecko/20050729 Fedora/1.1-0.2.5.deerpark.alpha2 Firefox/1.0+

Description of problem:
proc/devname.c:52: warning: padding struct size to alignment boundary
proc/devname.c: In function 'link_name':
proc/devname.c:229: internal compiler error: in df_uses_record, at df.c:1061
Please submit a full bug report,

Version-Release number of selected component (if applicable):
gcc-4.0.1-6

How reproducible:
Didn't try

Steps to Reproduce:
1.rebuild procps-3.2.5-6.3
2.
3.
  

Additional info:
Comment 1 Karsten Hopp 2005-08-04 14:19:00 UTC 
Created attachment 117448 [details]
Preprocessed source
Comment 2 Jakub Jelinek 2005-08-09 20:37:59 UTC 
Simplified testcase:
/* { dg-do compile } */
/* { dg-options "-O2 -fstack-protector-all -fweb" } */

struct S
{
  unsigned long long st_dev;
  unsigned long long st_rdev;
};
int st (const char *, struct S *);

int
foo (char *const buf, unsigned maj, unsigned min)
{
  struct S sbuf;
  if (st (buf, &sbuf) < 0)
    return 0;
  if (min !=
      (((unsigned) (sbuf.st_rdev) & 0xffu) |
       (((unsigned) (sbuf.st_rdev) & 0xfff00000u) >> 12u)))
    return 0;
  if (maj != (((unsigned) (sbuf.st_rdev) >> 8u) & 0xfffu))
    return 0;
  return 1;
}

Now, why are you using -fweb?  See
http://gcc.gnu.org/ml/gcc-patches/2004-09/msg01931.html
for details why it is not a good idea to use it.
Comment 3 Karsten Hopp 2005-08-10 10:43:30 UTC 
I'm just the one who found it during a mass rebuild. 
I'll add kzak to CC: for a comment about -fweb
Comment 4 Karel Zak 2005-08-10 14:48:01 UTC 
Thanks for info. I will remove -fweb from procps CFLAGS.
Comment 5 Jakub Jelinek 2005-08-10 15:41:28 UTC 
The problem doesn't seem to be reproducible on GCC HEAD, so I guess I'll just
WONTFIX this.  -fweb simply shouldn't be used, it is certainly not very well
tested.