<html><head></head><body><div>Hi Peter,</div><div><span><pre><br></pre></span></div><div>On Sat, 2021-11-13 at 16:58 +0000, Peter Blackman wrote:</div><blockquote type="cite" style="margin:0 0 0 .8ex; border-left:2px #729fcf solid;padding-left:1ex"><div>Building C-evo on the Open Build Service, sometimes gives a weird error...<br></div><div><a href="https://build.opensuse.org/package/live_build_log/home:PeterBBB/c-evo/Debian_11/aarch64">https://build.opensuse.org/package/live_build_log/home:PeterBBB/c-evo/Debian_11/aarch64</a> <br></div><div><<a href="https://build.opensuse.org/package/live_build_log/home:PeterBBB/c-evo/Debian_11/aarch64">https://build.opensuse.org/package/live_build_log/home:PeterBBB/c-evo/Debian_11/aarch64</a>><br></div><div><br></div><div><br></div><div>....<br></div><div>[ 124s] [302/432] *installing fp-compiler*-3.2.0-3.2.0+dfsg-12<br></div><div>[ 125s] An unhandled exception occurred at $0000000000470960:<br></div><div>[ 125s] EInvalidOp: *Invalid floating point operation*<br></div><div>[ 125s] $0000000000470960<br></div><div>[ 125s] $0000000000472FC0<br></div><div>....<br></div><div><br></div><div>The problem is intermittent, sometime the aarch64 build succeeds. Builds for Bullseye OK on all other architectures.<br></div></blockquote><div>Many programs build using FPC seem to have build/execution errors with new FPC 3.2.2 release. It seems that 3.2.2 is less stable for some architectures than 3.2.0. However, we are still hunting some minimal programs that show regressions.</div><blockquote type="cite" style="margin:0 0 0 .8ex; border-left:2px #729fcf solid;padding-left:1ex"><div><br></div><div>Floating Point error on an install seems bizarre to me.<br></div></blockquote><div>It may be the result of the above mentioned instability of 3.2.2. However we need probably to understand what leads to that before we can raise a compiler bug.</div><blockquote type="cite" style="margin:0 0 0 .8ex; border-left:2px #729fcf solid;padding-left:1ex"><div><br></div><div>Anyone have any ideas on this? Is it worth raising a bug report?</div></blockquote><div>As I said, it may be a compiler bug, but we will need to have a small program reproducing it if we want upstream to handle it.</div><blockquote type="cite" style="margin:0 0 0 .8ex; border-left:2px #729fcf solid;padding-left:1ex"><div><br></div><div>On the basis of a very low sample, this problem seems occur around 50% of the time.</div></blockquote><div>Until now, mode issues where perfectly reproducible, maybe this is an other one, or maybe related to a non initialized floating point coprocessor register that gets in the way and leads to an invalid floating point operation</div><div style="caret-color: rgb(46, 52, 54); color: rgb(46, 52, 54); font-family: Cantarell; font-size: 14.666666984558105px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; orphans: auto; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; widows: auto; word-spacing: 0px; -webkit-tap-highlight-color: rgba(0, 0, 0, 0.4); -webkit-text-stroke-width: 0px; text-decoration: none;"><span><pre>-- <br></pre><pre>Cheers,
Abou Al Montacir</pre></span></div><blockquote type="cite" style="margin:0 0 0 .8ex; border-left:2px #729fcf solid;padding-left:1ex"></blockquote></body></html>