View Single Post
Old 05-23-2012, 10:31 AM   #109
geekmaster
Carpe diem, c'est la vie.
geekmaster ought to be getting tired of karma fortunes by now.geekmaster ought to be getting tired of karma fortunes by now.geekmaster ought to be getting tired of karma fortunes by now.geekmaster ought to be getting tired of karma fortunes by now.geekmaster ought to be getting tired of karma fortunes by now.geekmaster ought to be getting tired of karma fortunes by now.geekmaster ought to be getting tired of karma fortunes by now.geekmaster ought to be getting tired of karma fortunes by now.geekmaster ought to be getting tired of karma fortunes by now.geekmaster ought to be getting tired of karma fortunes by now.geekmaster ought to be getting tired of karma fortunes by now.
 
geekmaster's Avatar
 
Posts: 6,433
Karma: 10773670
Join Date: Nov 2011
Location: Multiverse 6627A
Device: K1 to PW3
Quote:
Originally Posted by knc1 View Post
Yup - the diagnostic path was destroyed by my pathname typo.

The "file" command above was done on the /lib/libpthread.so file, not (as I later edited it) the /usr/lib/libpthread.so file.

The "too small" /usr/lib/libpthread.so file is a loader script (which file properly identifies as ASCII TEXT).


Which is a lab126 "hack" in the worst meaning of the term, see my signature.
I reported this library loader script problem in earlier posts (for both the libm and libpthread scripts). This is exactly the same problem (and same fix) that I used when "-lm" caused this problem before, extended to support the pthread library in addition to the math library already supported. I also provided a replacement script with the extra crap stripped out, which I bind mount to prevent changing the original contents on the root partition.

For some strange reason, this only affects SOME versions of 5.1.0 firmware. How many versions of 5.1.0 are there, and how can we tell them apart?

Last edited by geekmaster; 05-23-2012 at 10:37 AM.
geekmaster is offline   Reply With Quote