[mpich-commits] [mpich] MPICH primary repository branch, master, updated. v3.1.1-24-gbfdad82
Service Account
noreply at mpich.org
Fri Jun 20 15:49:39 CDT 2014
This is an automated email from the git hooks/post-receive script. It was
generated because a ref change was pushed to the repository containing
the project "MPICH primary repository".
The branch, master has been updated
via bfdad82283520038a8abad059ef637442321b93b (commit)
via 2b63c241a709c63308b9f0e88802a349665cdc74 (commit)
from f8c8d5aa6e3081716c3ee38f83f528bc7a061c49 (commit)
Those revisions listed above that are new to this repository have
not appeared on any other notification email; so we list those
revisions in full, below.
- Log -----------------------------------------------------------------
http://git.mpich.org/mpich.git/commitdiff/bfdad82283520038a8abad059ef637442321b93b
commit bfdad82283520038a8abad059ef637442321b93b
Author: Sangmin Seo <sseo at anl.gov>
Date: Thu Jun 19 16:38:26 2014 -0500
Removed a unnecessary MPIU_Assert.
MPIU_Assert(rreq->dev.iov_offset >= 0) generates warning about
comparison which is always true. This happens because the type of
iov_offset is size_t.
Signed-off-by: Rob Latham <robl at mcs.anl.gov>
diff --git a/src/mpid/ch3/channels/nemesis/netmod/tcp/socksm.c b/src/mpid/ch3/channels/nemesis/netmod/tcp/socksm.c
index b34af09..7e8ce8f 100644
--- a/src/mpid/ch3/channels/nemesis/netmod/tcp/socksm.c
+++ b/src/mpid/ch3/channels/nemesis/netmod/tcp/socksm.c
@@ -1579,7 +1579,6 @@ static int MPID_nem_tcp_recv_handler(sockconn_t *const sc)
int (*reqFn)(MPIDI_VC_t *, MPID_Request *, int *);
MPIU_Assert(rreq->dev.iov_count > 0);
- MPIU_Assert(rreq->dev.iov_offset >= 0);
MPIU_Assert(rreq->dev.iov_count + rreq->dev.iov_offset <= MPID_IOV_LIMIT);
bytes_recvd = MPL_large_readv(sc_fd, iov, rreq->dev.iov_count);
http://git.mpich.org/mpich.git/commitdiff/2b63c241a709c63308b9f0e88802a349665cdc74
commit 2b63c241a709c63308b9f0e88802a349665cdc74
Author: Sangmin Seo <sseo at anl.gov>
Date: Thu Jun 19 16:16:29 2014 -0500
Replaced -Wno-type-limits with -Wtype-limits.
-Wno-type-limits was removed because it masks important failures
(see ticket #2094). However, since Intel compiler currently does
not include -Wtype-limits in -Wextra, -Wtype-limits was added to
handle warnings with the Intel compiler.
Signed-off-by: Rob Latham <robl at mcs.anl.gov>
diff --git a/confdb/aclocal_cc.m4 b/confdb/aclocal_cc.m4
index f03e0c2..8fc999b 100644
--- a/confdb/aclocal_cc.m4
+++ b/confdb/aclocal_cc.m4
@@ -475,13 +475,6 @@ if test "$enable_strict_done" != "yes" ; then
# msg_sz_t) variables.
# -Wno-format-zero-length -- this warning is irritating and useless, since
# a zero-length format string is very well defined
- # -Wno-type-limits -- There are places where we compare an unsigned to
- # a constant that happens to be zero e.g., if x is unsigned and
- # MIN_VAL is zero, we'd like to do "MPIU_Assert(x >= MIN_VAL);".
- # Note this option is not supported by gcc 4.2. This needs to be added
- # after most other warning flags, so that we catch a gcc bug on 32-bit
- # that doesn't give a warning that this is unsupported, unless another
- # warning is triggered, and then if gives an error.
# These were removed to reduce warnings:
# -Wcast-qual -- Sometimes we need to cast "volatile char*" to
# "char*", e.g., for memcpy.
@@ -513,6 +506,18 @@ if test "$enable_strict_done" != "yes" ; then
# important check, but is temporarily disabled, since it is
# throwing too many (correct) warnings currently, causing us
# to miss other warnings.
+ #
+ # This was removed because it masks important failures (see ticket #2094).
+ # However, since Intel compiler currently does not include -Wtype-limits
+ # in -Wextra, -Wtype-limits was added to handle warnings with the Intel
+ # compiler.
+ # -Wno-type-limits -- There are places where we compare an unsigned to
+ # a constant that happens to be zero e.g., if x is unsigned and
+ # MIN_VAL is zero, we'd like to do "MPIU_Assert(x >= MIN_VAL);".
+ # Note this option is not supported by gcc 4.2. This needs to be added
+ # after most other warning flags, so that we catch a gcc bug on 32-bit
+ # that doesn't give a warning that this is unsupported, unless another
+ # warning is triggered, and then if gives an error.
# the embedded newlines in this string are safe because we evaluate each
# argument in the for-loop below and append them to the CFLAGS with a space
# as the separator instead
@@ -543,7 +548,7 @@ if test "$enable_strict_done" != "yes" ; then
-Wno-pointer-sign
-Wvariadic-macros
-Wno-format-zero-length
- -Wno-type-limits
+ -Wtype-limits
-Werror-implicit-function-declaration
"
-----------------------------------------------------------------------
Summary of changes:
confdb/aclocal_cc.m4 | 21 +++++++++++++--------
src/mpid/ch3/channels/nemesis/netmod/tcp/socksm.c | 1 -
2 files changed, 13 insertions(+), 9 deletions(-)
hooks/post-receive
--
MPICH primary repository
More information about the commits
mailing list