[Bug target/108809] New: gcc.target/powerpc/builtins-5-p9-runnable.c fails on power 9 BE

[Bug target/108809] New: gcc.target/powerpc/builtins-5-p9-runnable.c fails on power 9 BE

[seurer at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=108809

            Bug ID: 108809
           Summary: gcc.target/powerpc/builtins-5-p9-runnable.c fails on
                    power 9 BE
           Product: gcc
           Version: 13.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: target
          Assignee: unassigned at gcc dot gnu.org
          Reporter: seurer at gcc dot gnu.org
  Target Milestone: ---

make  -k check-gcc RUNTESTFLAGS="--target_board=unix'{-m32,-m64}'
powerpc.exp=gcc.target/powerpc/builtins-5-p9-runnable.c"
FAIL: gcc.target/powerpc/builtins-5-p9-runnable.c execution test
FAIL: gcc.target/powerpc/builtins-5-p9-runnable.c  -O0  execution test
FAIL: gcc.target/powerpc/builtins-5-p9-runnable.c  -Os  execution test
FAIL: gcc.target/powerpc/builtins-5-p9-runnable.c  -O1  execution test
FAIL: gcc.target/powerpc/builtins-5-p9-runnable.c  -O2  execution test
FAIL: gcc.target/powerpc/builtins-5-p9-runnable.c  -O3  execution test

This fails on power 9 BE (only) and fails for trunk as well as gcc 12, 11, and
10.  

(gdb) run
Starting program: /home/seurer/gcc/git/build/gcc-10/builtins-5-p9-runnable.exe 
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/powerpc64-linux-gnu/libthread_db.so.1".

Program received signal SIGABRT, Aborted.
__pthread_kill_implementation (threadid=<optimized out>, signo=<optimized out>,
no_tid=<optimized out>) at pthread_kill.c:44
44      pthread_kill.c: No such file or directory.
(gdb) where
#0  __pthread_kill_implementation (threadid=<optimized out>, signo=<optimized
out>, no_tid=<optimized out>) at pthread_kill.c:44
#1  0x00007ffff7c90cb4 in __GI_raise (sig=<optimized out>) at
../sysdeps/posix/raise.c:26
#2  0x00007ffff7c749ec in __GI_abort () at abort.c:79
#3  0x00000000100005e8 in main () at
/home/seurer/gcc/git/gcc-10/gcc/testsuite/gcc.target/powerpc/builtins-5-p9-runnable.c:71

If I run it with DEBUG defined I get

seurer@perfzep2:~/gcc/git/build/gcc-10$ ./builtins-5-p9-runnable.exe 
Error: result does not match expected result
vec_xl_len_r(8): vec_uc_expected1[0] to vec_uc_expected1[15]
 8, 7, 6, 5, 4, 3, 2, 1, 0, 0, 0, 0, 0, 0, 0, 0,
vec_xl_len_r(8): vec_uc_result1[0] to vec_uc_result1[15]
 0, 0, 0, 0, 0, 0, 0, 0, 1, 2, 3, 4, 5, 6, 7, 8,

Error: result does not match expected result
vec_xl_len_r(4): vec_uc_expected1[0] to vec_uc_expected1[15]
 4, 3, 2, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
vec_xl_len_r(4): vec_uc_result1[0] to vec_uc_result1[15]
 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 2, 3, 4,

Error: result does not match expected result
vec_xl_len_r(2): vec_uc_expected1[0] to vec_uc_expected1[15]
 2, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
vec_xl_len_r(2) vec_uc_result1[0] to vec_uc_result1[15]
 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 2,

Error: result does not match expected result
vec_xst_len_r(16) vec_uc_expected1[0] to vec_uc_expected1[15]
 16, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1,
vec_xst_len_r(16) result[0] to result[15]
 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16,

Error: result does not match expected result
vec_xst_len_r(2) vec_uc_expected1[0] to vec_uc_expected1[15]
 2, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
vec_xst_len_r(2) result[0] to result[15]
 15, 16, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,

Error: result does not match expected result
vec_xst_len_r(16) vec_uc_expected1[0] to vec_uc_expected1[15]
 16, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1,
vec_xst_len_r(16) result[0] to result[15]
 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16,

Error: result does not match expected result
vec_xst_len_r(14) vec_uc_expected1[0] to vec_uc_expected1[15]
 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0, 0,
vec_xst_len_r(14) result[0] to result[15]
 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 0, 0,

[Bug target/108809] gcc.target/powerpc/builtins-5-p9-runnable.c fails on power 9 BE

[linkw at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=108809

Kewen Lin <linkw at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |linkw at gcc dot gnu.org
   Last reconfirmed|                            |2023-03-09
             Status|UNCONFIRMED                 |NEW
     Ever confirmed|0                           |1

--- Comment #1 from Kewen Lin <linkw at gcc dot gnu.org> ---
It's very likely a test issue, may need to adjust some built-in function for
endianness issue (as they have different element ordering on BE and LE).

[Bug target/108809] gcc.target/powerpc/builtins-5-p9-runnable.c fails on power 9 BE

[guojiufu at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=108809

Jiu Fu Guo <guojiufu at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |guojiufu at gcc dot gnu.org

--- Comment #2 from Jiu Fu Guo <guojiufu at gcc dot gnu.org> ---
(In reply to Kewen Lin from comment #1)
> It's very likely a test issue, may need to adjust some built-in function for
> endianness issue (as they have different element ordering on BE and LE).

Yeap it should be the test case issue on the difference between BE and LE:
For a buff `\001\002\003\004\005\006\a\b\t\n\v\f\r\016\017\020\021\022`,
after loading to a vector right justified, the result is different between BE
and LE.
e.g. for a vector (from 128bit view: 0x0000000000000000102030405060708), 
from v16_int8 view on BE, it is v16_int8 = {0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0,
0x0, 0x1, 0x2, 0x3, 0x4, 0x5, 0x6, 0x7, 0x8}; but from the v16_int8 view on LE,
it is v16_int8 = {0x8, 0x7, 0x6, 0x5, 0x4, 0x3, 0x2, 0x1, 0x0, 0x0, 0x0, 0x0,
0x0, 0x0, 0x0, 0x0}.

So, we would just need to update the expected result for BE for this test case.

[Bug target/108809] gcc.target/powerpc/builtins-5-p9-runnable.c fails on power 9 BE

[guojiufu at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=108809

--- Comment #3 from Jiu Fu Guo <guojiufu at gcc dot gnu.org> ---
A similar different view between BE and LE on the vector for vec_xst_len_r.
For: 
store_data_uc = (vector unsigned char){ 1, 2, 3, 4, 5, 6, 7, 8,
                                             9, 10, 11, 12, 13, 14, 15, 16 };
vec_xst_len_r(store_data_uc, address, size);

On BE, from 128bit view (corresponding to store_data_uc) is
`0x102030405060708090a0b0c0d0e0f10`.
But on LE, from 128bit view, it is `0x100f0e0d0c0b0a090807060504030201`.
After right justified (left clean to 0s), the effective bytes to be stored to
buff on LE are those with smaller values (1,2,...);  
On BE, the bytes to be stored are those with bigger values (16,15,...)

BTW: the generated insn sequence aligns with the example implementation in
PVIPR.

[Bug testsuite/108809] gcc.target/powerpc/builtins-5-p9-runnable.c fails on power 9 BE

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=108809

--- Comment #4 from CVS Commits <cvs-commit at gcc dot gnu.org> ---
The master branch has been updated by Jiu Fu Guo <guojiufu@gcc.gnu.org>:

https://gcc.gnu.org/g:a1f25e04b8d10bbe5dcaf98bb7e17cdaec9f169d

commit r13-7202-ga1f25e04b8d10bbe5dcaf98bb7e17cdaec9f169d
Author: Jiufu Guo <guojiufu@linux.ibm.com>
Date:   Fri Apr 14 10:50:28 2023 +0800

    testsuite: update builtins-5-p9-runnable.c for BE

    Hi,

    As PR108809 mentioned, vec_xl_len_r and vec_xst_len_r are tested
    in gcc.target/powerpc/builtins-5-p9-runnable.c.
    The vector operand of these two bifs are different from the view
    of v16_int8 between BE and LE, even it is same from the view of
    128bits(uint128/V1TI).

    The test case gcc.target/powerpc/builtins-5-p9-runnable.c was
    written for LE environment, this patch updates it for BE.

    Tested on ppc64 BE and LE.
    Is this ok for trunk?

    BR,
    Jeff (Jiufu)

    gcc/testsuite/ChangeLog:

            PR testsuite/108809
            * gcc.target/powerpc/builtins-5-p9-runnable.c: Update for BE.

[Bug testsuite/108809] gcc.target/powerpc/builtins-5-p9-runnable.c fails on power 9 BE

[guojiufu at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=108809

Jiu Fu Guo <guojiufu at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         Resolution|---                         |FIXED
             Status|NEW                         |RESOLVED

--- Comment #5 from Jiu Fu Guo <guojiufu at gcc dot gnu.org> ---
Fixed in the trunk.