From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 9781 invoked by alias); 28 Oct 2003 16:49:49 -0000 Mailing-List: contact gdb-patches-help@sources.redhat.com; run by ezmlm Precedence: bulk List-Subscribe: List-Archive: List-Post: List-Help: , Sender: gdb-patches-owner@sources.redhat.com Received: (qmail 9767 invoked from network); 28 Oct 2003 16:49:47 -0000 Received: from unknown (HELO nevyn.them.org) (66.93.172.17) by sources.redhat.com with SMTP; 28 Oct 2003 16:49:47 -0000 Received: from drow by nevyn.them.org with local (Exim 4.24 #1 (Debian)) id 1AEX2O-0007uE-4r; Tue, 28 Oct 2003 11:49:44 -0500 Date: Tue, 28 Oct 2003 16:49:00 -0000 From: Daniel Jacobowitz To: Kris Warkentin Cc: Elena Zannoni , Michael Elizabeth Chastain , gdb-patches@sources.redhat.com Subject: Re: [patch] New testcase for PR:1291 Message-ID: <20031028164944.GA29487@nevyn.them.org> Mail-Followup-To: Kris Warkentin , Elena Zannoni , Michael Elizabeth Chastain , gdb-patches@sources.redhat.com References: <16286.37627.229387.242404@localhost.redhat.com> <0c6901c39d71$d41d2150$0202040a@catdog> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <0c6901c39d71$d41d2150$0202040a@catdog> User-Agent: Mutt/1.5.1i X-SW-Source: 2003-10/txt/msg00812.txt.bz2 On Tue, Oct 28, 2003 at 11:37:50AM -0500, Kris Warkentin wrote: > You'll have to explain that to me a bit better. KFAIL? Does that mean I > leave the test as is and mark it as an expected failure somehow? Or do I > modify the test and make a note? It's just like xfail. Grep for kfail in the testsuite for examples; kfail takes as argument a PR number (file it in GNATS). > > cheers, > > Kris > > ----- Original Message ----- > From: "Elena Zannoni" > To: "Kris Warkentin" > Cc: "Michael Elizabeth Chastain" ; ; > > Sent: Tuesday, October 28, 2003 11:02 AM > Subject: Re: [patch] New testcase for PR:1291 > > > > Kris Warkentin writes: > > > > > A) move it and make it generic. > > > > > B) leave it and write it in assembly. > > > > > > > > I like (B) better. The backtrace code is all about specific > > > > assembly instructions. So I'd like it better in the style of the > i386 > > > > prologue tests, with explicit assembly. > > > > > > > > Michael C > > > > > > Okay so this bit of assembly adequately demonstrates the problem. > There is > > > a bit of an issue here though that perhaps some sh4 expert can clear up > for > > > me. If I run this and go "until sub2", it stops on sub2 and works > fine. > > > If, however, I go "until sub1", followed by "until sub2", it stops a > couple > > > instructions before sub2 and I have to 'si' into it. I can modify the > test > > > case to not bother checking sub1 (since we know it's okay anyway) but > I'm > > > still puzzled as to why this is having the problem. > > > > > > cheers, > > > > > > Kris > > > > > > > Ok. Not sure why that happens. It defeinitely seems like a bug. File > > a pr, and KFAIL it in the testsuite, with the pr number. > > > > elena > > > > > > > > > > void sub1 (void); > > > void sub2 (void); > > > > > > main() > > > { > > > sub1(); > > > sub2(); > > > } > > > > > > asm(".text\n" > > > " .align 5\n" > > > "sub1:\n" > > > " mov.l r14,@-r15\n" > > > " add #-128,r15\n" > > > " add #-128,r15\n" > > > " mov r15,r14\n" > > > " mov.w .STACK1,r7\n" > > > " add r7,r14\n" > > > " mov r14,r15\n" > > > " mov.l @r15+,r14\n" > > > " rts\n" > > > " nop\n" > > > " .align 1\n" > > > ".STACK1:\n" > > > " .short 256\n"); > > > > > > asm(".text\n" > > > " .align 5\n" > > > "sub2:\n" > > > " mov.l r14,@-r15\n" > > > " mov.w .STACK2,r3\n" > > > " sub r3,r15\n" > > > " mov r15,r14\n" > > > " mov.w .STACK2,r7\n" > > > " add r7,r14\n" > > > " mov r14,r15\n" > > > " mov.l @r15+,r14\n" > > > " rts\n" > > > " nop\n" > > > " .align 1\n" > > > ".STACK2:\n" > > > " .short 260\n"); > > > > > > > > > > > -- Daniel Jacobowitz MontaVista Software Debian GNU/Linux Developer