Thanks. The ir_b2a -st output is not changed at all. Not sure if anything would need to change for handling the remaining pieces, but I don't think so. I wrote this from scratch (so blame me for the bugs :-), and did not attach any gpl license to it, so feel free to incorporate it into both open64.net and pathscale's source.
-----Original Message----- From: "C. Bergström" [mailto:cbergst...@pathscale.com] Sent: Saturday, October 16, 2010 2:13 PM To: Mike Murphy Cc: 'Open64-devel@lists.sourceforge.net' Subject: Re: [Open64-devel] ir_a2b Mike Murphy wrote: > If you've ever looked at the ir_b2a code, you'll see that there was an idea > for ir_a2b back in the beginning, but it was never implemented. I actually > remember Fred asking me about it back at SGI, but I didn't have time and > didn't see an easy way to do it because you'd have to parse the symbol table. > Well, a while ago a colleague mentioned that this would be a nice feature, > and I started thinking about it and realized it should be doable. So in my > spare time I started working on ir_a2b. What I have now works on simple > code, but there is still lots to do. But I don't have time to finish this, > so I'm publishing what I have so far, and hoping that others will take this > and productize it. > > Attached is ir_a2b.cxx, ir_b2a.cxx, and Makefile.gbase, all in ir_tools. The > Makefile.gbase change just adds ir_a2b, the ir_b2a.cxx is the old ir_a2b but > just for ir_b2a, and then the new code is all in ir_a2b.cxx. One other > change was to symtab.cxx to make Init_Constab not be inline (odd that it was > already in symtab.h but was declared inline in symtab.cxx). What is not > handled: more than 1 function, loops, arrays, structs, labels, initos, dst > other than files and includes, plus some other minor things. On the plus > side, the basic code works, and the following test is handled: > int a; > static int b; > > void test (int n) > { > a = b + n; > } > ir_b2a -st test.B test.Ba; ir_a2b test.Ba test.BB; ir_b2a -st test.BB test.Bba > produces the same result except that we don't have the dst-info output. > Hi Mike, It's awesome that you published this.. I've only glanced, but did you write this from scratch and what license is it under? Did you have to change the output of ir_b2a so that it's easier to parse? /* NVISA comment - Having all those additional memory spaces is a bugger isn't it.. ;) */ ./C ----------------------------------------------------------------------------------- This email message is for the sole use of the intended recipient(s) and may contain confidential information. Any unauthorized review, use, disclosure or distribution is prohibited. If you are not the intended recipient, please contact the sender by reply email and destroy all copies of the original message. ----------------------------------------------------------------------------------- ------------------------------------------------------------------------------ Download new Adobe(R) Flash(R) Builder(TM) 4 The new Adobe(R) Flex(R) 4 and Flash(R) Builder(TM) 4 (formerly Flex(R) Builder(TM)) enable the development of rich applications that run across multiple browsers and platforms. Download your free trials today! http://p.sf.net/sfu/adobe-dev2dev _______________________________________________ Open64-devel mailing list Open64-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/open64-devel
