diff options
| author | Quincey Koziol <koziol@hdfgroup.org> | 2004-10-04 20:29:31 (GMT) |
|---|---|---|
| committer | Quincey Koziol <koziol@hdfgroup.org> | 2004-10-04 20:29:31 (GMT) |
| commit | 74a448d084ac1e09be38fdf80c7b7298e3957ea4 (patch) | |
| tree | 0f47590955318bf799c5c6d7cefeecc66fc9bebe /src/H5D.c | |
| parent | e2f3ab0ab01044d515eb74b64d24a569d06e3d86 (diff) | |
| download | hdf5-74a448d084ac1e09be38fdf80c7b7298e3957ea4.zip hdf5-74a448d084ac1e09be38fdf80c7b7298e3957ea4.tar.gz hdf5-74a448d084ac1e09be38fdf80c7b7298e3957ea4.tar.bz2 | |
[svn-r9358] Purpose:
Bug fix
Description:
Relax restrictions on parallel I/O to allow compressed, chunked datasets
to be read in parallel (collective access will be degraded to independent
access, but will retrieve the information still).
Platforms tested:
FreeBSD 4.10 (sleipnir) w/parallel
Solaris 2.7 (arabica)
IRIX64 6.5 (modi4)
h5committest
Diffstat (limited to 'src/H5D.c')
| -rw-r--r-- | src/H5D.c | 4 |
1 files changed, 0 insertions, 4 deletions
@@ -2642,10 +2642,6 @@ H5D_open_oid(const H5G_entry_t *ent, hid_t dxpl_id) if(H5P_set(plist, H5D_CRT_DATA_PIPELINE_NAME, &pline) < 0) HGOTO_ERROR(H5E_DATASET, H5E_CANTSET, NULL, "can't set pipeline") - /* If MPI VFD is used, no filter support yet. */ - if(IS_H5FD_MPI(dataset->ent.file) && pline.nused > 0) - HGOTO_ERROR (H5E_DATASET, H5E_UNSUPPORTED, NULL, "Parallel IO does not support filters yet") - /* * Get the raw data layout info. It's actually stored in two locations: * the storage message of the dataset (dataset->storage) and certain |