This is the mail archive of the libc-hacker@sources.redhat.com mailing list for the glibc project.
Note that libc-hacker is a closed list. You may look at the archives of this list, but subscription and posting are not open.
Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
---|---|---|
Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |
On Sun, Sep 09, 2001 at 07:44:26PM -0400, Roland McGrath wrote: > > So isn't the librt.so linker script the right solution then? I mean the > > fact that -lrt uses -lpthread for its job is its implementation detail. > > For that very reason I don't think the linker script is right. The > librt.so binary could very well be replaced with a fully ABI-compatible one > that does not use libpthread in its implementation. > > Ideally librt.so.1's dependency on libpthread.so.0 would make this work out > right. But I guess it doesn't because the program's own breadth-first > dependency order will put libc before libpthread. Can we add DF_1_INITFIRST to librt.so.1? What will ld.so do for DSOs with DF_1_INITFIRST on the DT_NEEDED list? > > It seems to me that for this sort of thing to really work properly, > all of libc's definitions that might be overridden by libpthread must be weak. I don't think it is a long time solution since ld.so will treat weak as strong one day. H.J.
Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
---|---|---|
Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |