Menu
▾
▴
[sdcc-devel] [ sdcc-Bugs-3502632 ] gcc-torture-execute-20010224-1 fails for z80-related
|
From: SourceForge.net <no...@so...> - 2012-03-12 17:07:11
|
Bugs item #3502632, was opened at 2012-03-12 06:46 Message generated for change (Comment added) made by spth You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=100599&aid=3502632&group_id=599 Please note that this message will contain a full copy of the comment thread, including the initial issue submission, for this request, not just the latest update. Category: z80 port Group: None Status: Open Resolution: None Priority: 6 Private: No Submitted By: Philipp Klaus Krause (spth) Assigned to: Nobody/Anonymous (nobody) Summary: gcc-torture-execute-20010224-1 fails for z80-related Initial Comment: The regression test gcc-torture-execute-20010224-1 fails for the z80, z180, r2k, gbz80 in revision #7441. To reproduce, remove the #ifdef. Philipp ---------------------------------------------------------------------- >Comment By: Philipp Klaus Krause (spth) Date: 2012-03-12 10:07 Message: Ok, it's not j, it's j << 1. Nevertheless, before CSE the sitatuion looks the same for z80 and hc08. But for z80 the error is introduced by CSE (not GCSE). Even though AFAIK there is no port-specific stuff in CSE. Philipp ---------------------------------------------------------------------- Comment By: Philipp Klaus Krause (spth) Date: 2012-03-12 09:34 Message: It is not even %psd[j]. sdcc plainly keeps another copy of j, that is never incremented and uses it in the loop to calculate &psd[j]. Philipp ---------------------------------------------------------------------- Comment By: Philipp Klaus Krause (spth) Date: 2012-03-12 09:27 Message: It seems the problem is that sdcc thinks &psd[j] is an invariant, and calculates it before the loop, even though j changes in the loop. I wonder, why the other ports are not affected. Philipp ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=100599&aid=3502632&group_id=599 |