diff options
| author | Neil Fortner <nfortne2@hdfgroup.org> | 2009-02-12 20:46:32 (GMT) |
|---|---|---|
| committer | Neil Fortner <nfortne2@hdfgroup.org> | 2009-02-12 20:46:32 (GMT) |
| commit | 0bdedf0a39e859956b6810e848fa67d16b4e10a4 (patch) | |
| tree | e558e84649e86dbf4422206089baf08ccef659f7 /src/H5D.c | |
| parent | 0eb811d83156bc15162607c967d98e8ca093a8df (diff) | |
| download | hdf5-0bdedf0a39e859956b6810e848fa67d16b4e10a4.zip hdf5-0bdedf0a39e859956b6810e848fa67d16b4e10a4.tar.gz hdf5-0bdedf0a39e859956b6810e848fa67d16b4e10a4.tar.bz2 | |
[svn-r16480] Purpose: Improve chunk cache
Description:
The meaning of the "nbytes" field in H5D_rdcc_t was not clear, and some places
assumed it was the maximum size of the chunk cache, while some assumed it was
the current size of the chunk cache. The end result was that only 1 chunk could
be held in cache at a time. This field has been replaced by "nbytes_max" and
"nbytes_used". Performance of cached I/O should improve greatly.
Tested: jam, smirom (h5committest)
Diffstat (limited to 'src/H5D.c')
| -rw-r--r-- | src/H5D.c | 2 |
1 files changed, 1 insertions, 1 deletions
@@ -712,7 +712,7 @@ H5Dget_access_plist(hid_t dset_id) if (dset->shared->layout.type == H5D_CHUNKED) { if (H5P_set(new_plist, H5D_ACS_DATA_CACHE_NUM_SLOTS_NAME, &(dset->shared->cache.chunk.nslots)) < 0) HGOTO_ERROR(H5E_PLIST, H5E_CANTSET, FAIL, "can't set data cache number of slots") - if (H5P_set(new_plist, H5D_ACS_DATA_CACHE_BYTE_SIZE_NAME, &(dset->shared->cache.chunk.nbytes)) < 0) + if (H5P_set(new_plist, H5D_ACS_DATA_CACHE_BYTE_SIZE_NAME, &(dset->shared->cache.chunk.nbytes_max)) < 0) HGOTO_ERROR(H5E_PLIST, H5E_CANTSET, FAIL, "can't set data cache byte size") if (H5P_set(new_plist, H5D_ACS_PREEMPT_READ_CHUNKS_NAME, &(dset->shared->cache.chunk.w0)) < 0) HGOTO_ERROR(H5E_PLIST, H5E_CANTSET, FAIL, "can't set preempt read chunks") |