https://sourceware.org/bugzilla/show_bug.cgi?id=27672
Nick Clifton <nickc at redhat dot com> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |nickc at redhat dot com --- Comment #1 from Nick Clifton <nickc at redhat dot com> --- (In reply to Nick Lott from comment #0) Hi Nick, > When listing the symbol table using 'readelf' the size is listed in decimal > for objects smaller than 100,000 bytes. For objects this size or larger the > notation switches to hexadecimal format. > > This is inconsistent with other binutils programs such as 'nm' and 'objdump' > which use consistent notation throughout an output and as specified via the > command options. The problem is that in most cases, large values, ie ones greater than 100,000, often make more sense when viewed as hexadecimal numbers, rather than decimal ones. In the context of displaying the value of dynamic symbols that is. Of course there are times when this is not true, so it seems to me that the best solution would be to allow the user to choose. How about creating (yet another) new command line option --sym-base=[0|8|10|16]' where 0 would mean to continue with the current mixed-base output ? (The 0 value is necessary because there may be users of readelf that rely upon the behaviour). It should apply to all types of symbol, not just dynamic. What do you think ? Cheers Nick -- You are receiving this mail because: You are on the CC list for the bug.