Age | Commit message (Collapse) | Author |
|
current pipe.S seems broken for error case. For example, "la t0,
__syscall_error" should be "la t9, __syscall_error"? Jumping to
__syscall_error without reverting sp is correct?
I updated this patch against recent snapshot and tested: it looks OK.
* Get rid of saving/restoring a0. No need to adjust sp.
* Fix code in error path.
|
|
|
|
of POSIX_MADV_*
|
|
|
|
|
|
most of global data relocations are back
|
|
|
|
that needs to overwrite this
|
|
|
|
|
|
Patch tested.
|
|
to be weaks
|
|
|
|
|
|
|
|
|
|
questionable ifdefs
|
|
|
|
|
|
|
|
|
|
covered by common/longjmp.c
|
|
libpthread as well
|
|
|
|
is a useless attempt
|
|
|
|
specific header file to make porting/updates a lot easier
|
|
|
|
|
|
|
|
|
|
|
|
missing headers, other jump relocs removed
|
|
|
|
|
|
|
|
needed to shrink the pads as well to maintain ABI compat
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|