Re: [sqlite] Bug in GCC - suggestions sought for a backup tool chain

2011-08-23 Thread Jan Hudec
On Mon, Aug 22, 2011 at 20:22:30 -0400, Richard Hipp wrote: > (1) Compilers sometimes make mistakes. So it is important that you test > your object code - not just your source code. That means running your test > cases using exactly the same *.o files that you use for delivery. "Fly what > you

Re: [sqlite] Bug in GCC - suggestions sought for a backup tool chain

2011-08-23 Thread Andreas Kupries
On 8/22/2011 5:22 PM, Richard Hipp wrote: > Consider this line of code in the "build.c" source file of SQLite: > As it happens, the GCC bug is harmless in this case. SQLite never invokes > the sqlite3SrcListShiftJoinType() function with a non-NULL SrcList pointer > that has a NULL p->a value.

Re: [sqlite] Bug in GCC - suggestions sought for a backup tool chain

2011-08-23 Thread Richard Hipp
On Tue, Aug 23, 2011 at 12:16 PM, David Garfield < garfi...@irving.iisd.sra.com> wrote: > 2) Member "a" was an array, not a pointer. > You are correct. I had in mind that p->a was a pointer. So this is my mistake and not a bug in GCC after all. My apologies to the GCC team for the false

Re: [sqlite] Bug in GCC - suggestions sought for a backup tool chain

2011-08-23 Thread David Garfield
Well, I just looked in the 3.7.2 amalgamation version I use. 1) sqlite3SrcListShiftJoinType was declared with SQLITE_PRIVATE. Even if the static isn't here, if it is in the prototype, it can apply. Since this is an amalgamation-only feature, this might not be the issue. (3.7.7.1 src

Re: [sqlite] Bug in GCC - suggestions sought for a backup tool chain

2011-08-22 Thread Roger Binns
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 08/23/2011 05:52 AM, Richard Hipp wrote: > (1) Compilers sometimes make mistakes. So it is important that you test > your object code - not just your source code. That means running your test > cases using exactly the same *.o files that you use

Re: [sqlite] Bug in GCC - suggestions sought for a backup tool chain

2011-08-22 Thread Richard Hipp
On Mon, Aug 22, 2011 at 11:15 PM, Simon Slavin wrote: > > On 23 Aug 2011, at 1:22am, Richard Hipp wrote: > > > It appears that GCC 4.1.0 is not generating any code for the second test > in > > the conditional. In other words, GCC 4.1.0 is compiling that statement > as > >

Re: [sqlite] Bug in GCC - suggestions sought for a backup tool chain

2011-08-22 Thread Simon Slavin
On 23 Aug 2011, at 1:22am, Richard Hipp wrote: > It appears that GCC 4.1.0 is not generating any code for the second test in > the conditional. In other words, GCC 4.1.0 is compiling that statement as > if it omitted the "&& p->a" term How interesting. Can't solve your problem but pure

[sqlite] Bug in GCC - suggestions sought for a backup tool chain

2011-08-22 Thread Richard Hipp
Consider this line of code in the "build.c" source file of SQLite: http://www.sqlite.org/src/artifact/77be7c217430?ln=3372 It appears that GCC 4.1.0 is not generating any code for the second test in the conditional. In other words, GCC 4.1.0 is compiling that statement as if it omitted