#1132377 libvbz-hdf-plugin: Fix FTBFS on loong64

Package:
src:libvbz-hdf-plugin
Source:
src:libvbz-hdf-plugin
Submitter:
zhangdandan
Date:
2026-06-15 15:59:02 UTC
Severity:
normal
Tags:
#1132377#5
Date:
2026-03-31 07:29:04 UTC
From:
To:
Dear maintainers,

Compiling the libvbz-hdf-plugin failed for loong64 in the Debian Package
Auto-Building environment.
The error log is as follows,
```
[ 33%] Built target hdf_test_utils
In file included from
/build/reproducible-path/libvbz-hdf-plugin-1.0.2/vbz/v0/vbz_streamvbyte.cpp:2:
In static member function ‘static std::vector<U> StreamVByteWorkerV0<T,
UseZigZag>::cast(const gsl::span<V>&) [with U = int; V = const int; T =
int; bool UseZigZag = true]’,
     inlined from ‘static vbz_size_t StreamVByteWorkerV0<T,
UseZigZag>::compress(gsl::span<const char>, gsl::span<char>) [with T =
int; bool UseZigZag = true]’ at
/build/reproducible-path/libvbz-hdf-plugin-1.0.2/vbz/v0/vbz_streamvbyte_impl.h:30:35,
     inlined from ‘vbz_size_t
vbz_delta_zig_zag_streamvbyte_compress_v0(const void*, vbz_size_t,
void*, vbz_size_t, int, bool)’ at
/build/reproducible-path/libvbz-hdf-plugin-1.0.2/vbz/v0/vbz_streamvbyte.cpp:54:73:
/build/reproducible-path/libvbz-hdf-plugin-1.0.2/vbz/v0/vbz_streamvbyte_impl.h:75:23:
error: ‘void* __builtin_memcpy(void*, const void*, long unsigned int)’
offset [0, 3] is out of the bounds [0, 0] [-Werror=array-bounds=]
    75 |             output[i] = input[i];
```

Please consider the patch attached.
I have built libvbz-hdf-plugin normally on locally.

Best regards,
Dandan Zhang
#1132377#10
Date:
2026-03-31 08:03:57 UTC
From:
To:
Control: retitle -1 libvbz-hdf-plugin: Fix FTBFS on loong64
Control: tags -1 +ftbfs

This looks like more than a workaround than an actual fix as it just suppresses a warning.
Is it known why this warning shows on loong64 only?

Adrian
#1132377#19
Date:
2026-03-31 08:03:57 UTC
From:
To:
Control: retitle -1 libvbz-hdf-plugin: Fix FTBFS on loong64
Control: tags -1 +ftbfs

This looks like more than a workaround than an actual fix as it just suppresses a warning.
Is it known why this warning shows on loong64 only?

Adrian
#1132377#24
Date:
2026-06-15 15:57:05 UTC
From:
To:
Hi,

I intend to upload a new upstream version soon.  I wonder whether the
workaround to accept the compiler warning might be needed even for this
version and if so if there might be a better solution than hiding a
potential code problem.

Thank you for the patch anyway
    Andreas.