| To: | glame <glchen@ict.ac.cn> |
|---|---|
| Subject: | Re: why gcc 2.95.3 generate different result with option -mips2 and -mips3 for the same code |
| From: | Thiemo Seufer <ica2_ts@csv.ica.uni-stuttgart.de> |
| Date: | Tue, 14 Sep 2004 14:06:40 +0200 |
| Cc: | linux-mips <linux-mips@linux-mips.org> |
| In-reply-to: | <20040914092941Z8224931-1530+9945@linux-mips.org> |
| Original-recipient: | rfc822;linux-mips@linux-mips.org |
| References: | <20040914092941Z8224931-1530+9945@linux-mips.org> |
| Sender: | linux-mips-bounce@linux-mips.org |
| User-agent: | Mutt/1.5.6i |
glame wrote: > hi, > when i disassemble the code, the result is as the following > > -ffffffff80010b8c: 01001021 move $v0,$t0 (-mips2) > +ffffffff80010b8c: 0100102d move $v0,$t0 (-mips3) > > why? Because you told it to do so. :-) The 32bit move is "addu v0, $0, t0", the 64bit move "daddu ...". Thiemo |
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | why gcc 2.95.3 generate different result with option -mips2 and -mips3 for the same code, glame |
|---|---|
| Next by Date: | Re: gcc 2.95 patch for IP28, Florian Lohoff |
| Previous by Thread: | why gcc 2.95.3 generate different result with option -mips2 and -mips3 for the same code, glame |
| Next by Thread: | MIPS Magnum ARCsystem, Dennis Grevenstein |
| Indexes: | [Date] [Thread] [Top] [All Lists] |