summaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
authorMuQun Yang <ymuqun@hdfgroup.org>2005-08-31 16:05:40 (GMT)
committerMuQun Yang <ymuqun@hdfgroup.org>2005-08-31 16:05:40 (GMT)
commitab4127629f7dfd5728691a28928185c8de5fa90c (patch)
tree3c0f8fb362bdf992f8aa64110e422dc439ea1190
parentf1179bc82cbfbc7b8cc55257fa7dcb062af5add7 (diff)
downloadhdf5-ab4127629f7dfd5728691a28928185c8de5fa90c.zip
hdf5-ab4127629f7dfd5728691a28928185c8de5fa90c.tar.gz
hdf5-ab4127629f7dfd5728691a28928185c8de5fa90c.tar.bz2
[svn-r11322]
Purpose: Add the known problem for IRIX6.5 C version >7.4. Description: Somehow SGI's MPI implementation doesn't free MPI_TYPE_entries, So for large application to use MPI derived datatype, environment variable MPI_TYPE_MAX needs to be reset. Solution: Platforms tested: Misc. update:
-rw-r--r--release_docs/RELEASE.txt5
1 files changed, 5 insertions, 0 deletions
diff --git a/release_docs/RELEASE.txt b/release_docs/RELEASE.txt
index 7e2e57b..42906c1 100644
--- a/release_docs/RELEASE.txt
+++ b/release_docs/RELEASE.txt
@@ -799,6 +799,11 @@ Linux 2.4 IA64 Intel y n y n y y y
Known Problems
==============
+* On IRIX6.5, when C compiler version >7.4, the complicate MPI derived data type
+ code will work. However, the user should be aware to enlarge MPI_TYPE_MAX environment
+ variable to some certian value in order to use collective irregular selection code.
+ For example, the current parallel HDF5 test needs to enlarge MPI_TYPE_MAX to
+ 200,000 to make the test pass.
* The dataset created or rewritten with the v1.6.3 library or after can't
be read with the v1.6.2 library or before when Fletcher32 EDC(filter) is
enabled. There was a bug in the calculating code of the Fletcher32