Re: [patches] Re: RISC-V glibc port, v4

[Joseph Myers <joseph at codesourcery dot com>]

On Tue, 16 Jan 2018, Darius Rad wrote:

> On 01/15/2018 03:47 PM, Joseph Myers wrote:
> > 
> > If you copy libgcc_s.so.1 and libstdc++.so.6 into the build directory 
> > before "make check", I expect you should get much better results - and 
> > more meaningful results for identifying other problems.  I think it's a 
> > lot more than 35 tests affected (note all the "libgcc_s.so.1 must be 
> > installed for pthread_cancel to work" in output.log, for example).
> > 
> 
> After doing so, the results are much improved.
> 
> Summary of test results:
>      61 FAIL
>    5511 PASS
>       4 UNRESOLVED
>      21 UNSUPPORTED
>      18 XFAIL
> 
> I placed the output here [1].
> 
> [1] http://bluespec.com/tmp/glibc-check.tar.xz

Thanks.  It seems you still need to sort out (a) some NPTL issues, (b) 
probably problems with context functions, (c) a backtrace implementation 
(maybe a #include of the x86_64 one which is a generic 
backtrace-via-unwind-info one), (d) some TLS issues which could well be 
issues with tls-macros.h rather than the actual TLS implementation, and 
(e) an issue with NaN signs in strfromf (and maybe (f) a few other 
miscellaneous issues).  In addition to the issues from review of this 
version of the port.

(When the next revision of the port is posted, please include such test 
results for all four ABI variants - RV32I and RV64I, hard and soft float.)

-- 
Joseph S. Myers
joseph@codesourcery.com