[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#845749: libwebp FTBFS on armhf: inlining failed in call to always_inline 'vtrnq_s32': target specific option mismatch



Source: libwebp
Version: 0.5.1-3
Severity: serious

https://buildd.debian.org/status/logs.php?pkg=libwebp&arch=armhf

...
libtool: compile:  gcc -DHAVE_CONFIG_H -I. -I../../src/webp -DNDEBUG -Wdate-time -D_FORTIFY_SOURCE=2 -fvisibility=hidden -Wall -Wdeclaration-after-statement -Wextra -Wfloat-conversion -Wformat -Wformat-nonliteral -Wformat -Wformat-security -Wmissing-declarations -Wmissing-prototypes -Wold-style-definition -Wshadow -Wunreachable-code -Wunused-but-set-variable -Wunused -Wvla -g -O2 -fdebug-prefix-map=/«PKGBUILDDIR»=. -fstack-protector-strong -Wformat -Werror=format-security -pthread -c dec_neon.c  -fPIC -DPIC -o .libs/libwebpdspdecode_neon_la-dec_neon.o
In file included from ./neon.h:15:0,
                 from enc_neon.c:20:
./neon.h: In function 'Transpose4x4':
/usr/lib/gcc/arm-linux-gnueabihf/6/include/arm_neon.h:8203:1: error: inlining failed in call to always_inline 'vtrnq_s32': target specific option mismatch
 vtrnq_s32 (int32x4_t __a, int32x4_t __b)
 ^~~~~~~~~
In file included from enc_neon.c:20:0:
./neon.h:71:23: note: called from here
     const int32x4x2_t out23 = vtrnq_s32(vreinterpretq_s32_u64(row23.val[0]),
                       ^~~~~
In file included from ./neon.h:15:0,
                 from enc_neon.c:20:
/usr/lib/gcc/arm-linux-gnueabihf/6/include/arm_neon.h:14375:1: error: inlining failed in call to always_inline 'vreinterpretq_s32_u64': target specific option mismatch
 vreinterpretq_s32_u64 (uint64x2_t __a)
 ^~~~~~~~~~~~~~~~~~~~~
In file included from enc_neon.c:20:0:
./neon.h:71:31: note: called from here
     const int32x4x2_t out23 = vtrnq_s32(vreinterpretq_s32_u64(row23.val[0]),
                               ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
                                         vreinterpretq_s32_u64(row23.val[1]));
                                         ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
...



0.5.1-2 built fine, but back then gcc-5 was the default gcc.


Reply to: