This is the mail archive of the libc-alpha@sourceware.org mailing list for the glibc project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [PATCH] Patch 2 of 2 for ILP32 aarch64


On 09/06/17 09:47, Szabolcs Nagy wrote:
> On 24/02/17 23:46, Steve Ellcey wrote:
>>
>> Here is a new version of patch 2.  I added the one line descriptions
>> that were missing, fixed the configure.ac problems that Andreas pointed
>> out (and ran autoconf to make sure they were tested) and changed the
>> default version in shlib-versions to 2.26.
>>
> 
> i assume this is the latest glibc ilp32 patch.
> 
> without patch 1 (abilist updates) and after i drop the
> sysdeps/unix/sysv/linux/aarch64/mmap.c change and remove
> the malloc/tst-mallocstate test,
> 
> using CC='gcc -mabi=ilp32' (so c++ tests are unsupported), i get
> 

removing abi, c++ and known aarch64 issues

> FAIL: check-local-headers
*** $(common-objpfx)elf/readlib.o: uses /usr/include/gnu/lib-names-ilp32.h
*** /usr/include/gnu/lib-names-ilp32.h: uses /usr/include/gnu/lib-names-ilp32.h:
...
lib-names-ilp32.h was not generated, only lib-names-lp64.h

for some reason default-abi is lp64 in config.make,
even though preconfigure set aarch64_config_abi to ilp32

> FAIL: malloc/tst-dynarray-fail
timeout, cannot reproduce it now so it may be spurious error.

> FAIL: math/test-double-finite-lrint
> FAIL: math/test-double-finite-lround
> FAIL: math/test-double-lrint
> FAIL: math/test-double-lround
> FAIL: math/test-float-finite-lrint
> FAIL: math/test-float-finite-lround
> FAIL: math/test-float-lrint
> FAIL: math/test-float-lround
things like
Failure: lrint (0x1p31): Exception "Invalid operation" not set
Failure: lrint (0x1p32): Exception "Invalid operation" not set

i guess it's x vs w reg in the inline asm code

> FAIL: nptl/tst-initializers1
> FAIL: nptl/tst-initializers1-c11
> FAIL: nptl/tst-initializers1-c89
> FAIL: nptl/tst-initializers1-c99
> FAIL: nptl/tst-initializers1-gnu11
> FAIL: nptl/tst-initializers1-gnu89
> FAIL: nptl/tst-initializers1-gnu99
fails at
if (rwl_writer.__data.__flags
      != PTHREAD_RWLOCK_PREFER_WRITER_NONRECURSIVE_NP)

> FAIL: nptl/tst-rwlock15
timeout

> FAIL: nptl/tst-stack4
segfault

> FAIL: nss/bug17079
info: 10 test items
Timed out: killed the child process

> FAIL: posix/tst-mmap-offset
error: tst-mmap-offset.c:65: ftruncate64 failed
error: 1 test failures

> FAIL: resolv/tst-bug18665
timeout

> FAIL: resolv/tst-resolv-search
timeout

> FAIL: stdlib/tst-makecontext3
making contexts
start f1(a00=00000001,a01=00000004,a02=00000012,a03=00000048,
         a04=00000123,a05=0000048d,a06=00001234,a07=000048d1,
         a08=00012345,a09=00123456,a10=01234567,a11=12345678,
         a12=23456789,a13=3456789a,a14=456789ab,a15=56789abc,
         a16=6789abcd,a17=789abcde,a18=89abcdef,a19=9abcdef0,
         a20=abcdef0f,a21=00000000,a22=00000000,a23=00400b14,
         a24=00000000,a25=00000000,a26=00400aec,a27=00000000,
         a28=00000000,a29=da721e98,a30=c174bda8,a31=004005b0,
         a32=ffffffff) [0]
arg mismatch
program did not reach main again


i dont have debug tools on this machine, but i'd like
to know if others see similar failures on a ilp32 system.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]