Date: Sat, 7 Apr 2012 13:51:24 +0530 From: "Jayachandran C." <c.jayachandran@gmail.com> To: Stanislav Sedov <stas@freebsd.org> Cc: mips@freebsd.org Subject: Re: There is a known problem with MIPS tinderbox. Message-ID: <CA%2B7sy7DC3XLfOAiXoxRocvx57NV5ebV=wt=YJhescinZQ2%2BgyA@mail.gmail.com> In-Reply-To: <20120407005801.204e52db.stas@FreeBSD.org> References: <CACVs6=_kNJQZLouh6b%2BW9Ktj-CUFyeg7MVhas1qzoLGymJL7Kg@mail.gmail.com> <20120405142423.90d48b27.stas@FreeBSD.org> <CA%2B7sy7BwFQxUaHJsom36xGJXO9rd71%2BuERFcxcR4dUGkbwgNjA@mail.gmail.com> <036C5A57-2B63-40C1-A0DA-0678B7BF29F8@freebsd.org> <CA%2B7sy7BLpd-c54OU%2BuyJmETgcjvtDUGeHDB3_xoA5gyNiBPsTA@mail.gmail.com> <20120406135447.6ca86bd9.stas@FreeBSD.org> <CA%2B7sy7CFiffrYP0jwjB=_-iFAptCCVsn8MQuRzLG47Uv%2BV%2B7Eg@mail.gmail.com> <20120407005801.204e52db.stas@FreeBSD.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sat, Apr 7, 2012 at 1:28 PM, Stanislav Sedov <stas@freebsd.org> wrote: > On Sat, 7 Apr 2012 11:40:43 +0530 > "Jayachandran C." <c.jayachandran@gmail.com> mentioned: > >> I would still suggest the asn1 change if we are doing a temporary fix, b= ecause: >> - It does not add any new symbol versioning, it just converts the >> 'global: *' to an explicit list of symbols. >> - there are other libraries lined to asn1 which may have the same >> issue as libkafs5 (I haven't really looked at this. > > On one hand, it seems that the issue is triggered only when you > link versioned library against non-versioned, and all other > kerberos libraries that use libasn1 are versioned. =A0On the other, > you're right and if we don't fix libasn1 anything else linking > agains it (e.g. something from ports) might trigger the bug. > >> The only downside I see is that libasn1 has a lot of symbols. > > Yep, that was my concern. :-( =A0But I think you're right and we > should go with the edited libasn1 map. =A0I noticed that you > added some internal symbols to the map, though (.e.g. _init > and _fini). =A0What do you think if we take them out? You are right, we should probably take out _init and _fini, > There're some other symbols as well which no libraries seem > to use. I haven't really gone thru the symbols, this is a dump of 'nm -D' of libasn1 with a few obvious ones taken out. I do not have any background on this component, that is why I did not want to take a stab at cutting down the symbols or testing the resultant binaries. If the workaround works fine in your testing, you are welcome to check in after removing symbols you feel can be taken out. Thanks, JC.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CA%2B7sy7DC3XLfOAiXoxRocvx57NV5ebV=wt=YJhescinZQ2%2BgyA>