On Thu, Feb 23, 2023 at 12:04:47AM +0100, Mark Wielaard wrote: > Hi, Hi Mark,
> > On Wed, Feb 22, 2023 at 11:12:07AM -0600, Guillermo E. Martinez wrote: > > The conclusion was basically not use section flags to identify which > > section should be stripped out or not, so, it requires other mechanisms > > for explicitly specifying which sections should be removed, as eu-strip > > does using arguments. > > > > https://sourceware.org/bugzilla/show_bug.cgi?id=29737 > > OK, thanks. > > > Worth it mention here, that my last test shows that .ctf section is not > > stripped by ue-strip because it doesn't have a "debug" section name. > > Are you sure that is what happens? It might depend on whether or not > you give eu-strip -g or not. > Oh, sorry. Yes, you are right!. I looked what was wrong with my test environment and I didn't realize the use of `-g' option in my find-debuginfo script. > With -g only debug symbols and .debug sections are removed, but it > keeps any other unused/unallocated symbol/section. Without -g I would > expect eu-strip to remove the .ctf section (unless it is an allocated > section or referenced from an allocated section/symbol table). > I see, thanks for explanation!. I'll send immediately PATCH-v3 rebased from master. Kind regards, guillermo
