From: Jerome B. <jer...@fr...> - 2001-05-13 22:25:31
|
Hi, I've committed a fix for this bug, now intermediate code is properly generated (as far i can see). There is still a problem in asm code probably due to incorrect register assignement (spill loc). I'll appeciate help for this part. Since there are many changes, let me know if you find a problem or if you are agree with this fix . Regardrs Jérôme. |
From: Sandeep D. <sa...@dd...> - 2001-05-14 04:21:28
|
The fix looks good, it generates a redundant move, but the generated code looks good to me.... do you see anything wrong with it ? Sandeep -----Original Message----- From: sdc...@li... [mailto:sdc...@li...]On Behalf Of Jerome Bessiere Sent: Sunday, May 13, 2001 3:28 PM To: SDCC-devel Subject: [sdcc-devel] Re : Possible Array bug <foo[foo[5]+2]=3D10> Hi, I've committed a fix for this bug, now intermediate code is properly generated (as far i can see). There is still a problem in asm code probab= ly due to incorrect register assignement (spill loc). I'll appeciate help fo= r this part. Since there are many changes, let me know if you find a problem or if you are agree with this fix . Regardrs J=E9r=F4me. _______________________________________________ sdcc-devel mailing list sdc...@li... http://lists.sourceforge.net/lists/listinfo/sdcc-devel |
From: Johan K. <joh...@id...> - 2001-05-14 10:30:17
|
> The fix looks good, it generates a redundant move, but the > generated code looks good to me.... do you see anything wrong with > it ? It looks nice. However the new peephole 223 (which had a typo) is an exact copy of 221b that didn't work either because bindVar() swallowed the trailing ')'. I fixed that in SDCCpeeph.c:bindVar(). Also removed 222 and 223. Johan |