https://sourceware.org/bugzilla/show_bug.cgi?id=30967
--- Comment #9 from Aleksei Vetrov <vvvvvv at google dot com> --- (In reply to Frank Ch. Eigler from comment #7) > So my preferred workaround: I like this workaround and it works in our use case. > appears to be based on the assumption that truncated bitfields will not > collide. Has this assumption been tested? No such assumption can be made, there is no guarantees from DWARF standard and it is possible to manually generate DWARF with discriminators exceeding even "unsigned int": just use any library and shift all values by 32 bits. It will be well within standard, however compilers are not expected to behave like this. In an "liblog.so" attachment truncated values don't collide. I'm more interested in this: what is worst that can happen on discriminator collision? We are not using discriminator at all in ABI monitoring, so I'm not familiar with its use cases. -- You are receiving this mail because: You are on the CC list for the bug.
