Sorry for the late reply (just back to work after the Cauldron and LPC conferences).
Thanks a lot for your suggestions. Yes, I agree that the option need a better name -:) and we will figure this out after more study. During this year’s Cauldron and LPC, I got quite some good comments and suggestions in this area, and looks like that 1. This is an area that definitely needs more work; 2. Improvement in this area will be in general needed and beneficial; So, I am planning to spend a little more time to study more and provide a more comprehensive improvement in this area. More discussion is also needed. First, I will try to collect all the opened PR in this area, from my current knowledge, the following PRs belong to this area: 1. https://gcc.gnu.org/bugzilla/show_bug.cgi?id=109071 need more context for -Warray-bounds warnings due to code duplication from jump threading and inlining 2. https://gcc.gnu.org/bugzilla/show_bug.cgi?id=115704 -Wstringop-overread and related warnings should print inline stack Are there any other open PRs belong to this area? Thanks a lot for the help. Qing On Sep 14, 2024, at 14:07, Joern Rennecke <g...@amylaar.uk> wrote: > > > 1. Change the name of the option from: > > > > -fdiagnostic-try-to-explain-harder > > To > > -fdiagnostic-explain-harder > > I can think of a lot of connotations for this name, but alas, they are > unfortunate, off-topic, or both. > > Some more neutral ideas: > > -fdiagnostics-verbose > -fdiagnostic-details > > Or maybe a bit more specific: > > -fdiagnostics-trace-source > -fdiagnostics-show-source-condition > > if part of a group of diagnostic-augmenting options: > > -fdiagnostic-details=condition > > Or if you envisage this facility to be used with some debug info extension so > that gdb can eventually also show where exactly the code comes from: > > -frecord-source-condition > -frecord-code-duplication-condition > > > Although there is also some potential here to misconstrue 'condition' to be > about the style etc of the source, so maybe someone can come up with a better > name still?
