On Thu, 26 Jul 2018 21:14:14 +0200, Carlos O'Donell wrote:
Bug 1430223 - In some conditions, tcmalloc memalign will segfault
https://bugzilla.redhat.com/show_bug.cgi?id=1430223
It looks as a tcmalloc bug which could be fixed; it also has been probably
fixed in the meantime as stated there.
I think a key point here is to reduce the number of allocators being
used by the distribution so we can keep the quality high and help
our users when they have problems.
So why glibc greated an N+1 allocator (by DJ Delorie) instead of just
importing/using tcmalloc (which is license-compatible with glibc)?
On Thu, 26 Jul 2018 22:34:09 +0200, Florian Weimer wrote:
=
gperftools: tcmalloc breaks ABI on x86_64
You (Florian Weimer) have not provided any countercase (*) why small
allocations under alignment size (16 bytes) still should be aligned to 16 bytes.
This is why tcmalloc upstream closed it:
https://github.com/gperftools/gperftools/issues/724
(*)
https://github.com/gperftools/gperftools/issues/724#issuecomment-147369562
I do not see how that can be useful in practice, do you?
= tcsh:
interposed malloc is not ABI-compliant due to lack of alignment
This was a bug in tcsh custom allocator. (That is unrelated to tcmalloc.)
Looking at its source it looks to me mozjemalloc still in use by Fedora
Firefox still has only 8-byte alignment.
Jan