Skip to content

Instantly share code, notes, and snippets.

View goodell's full-sized avatar

Dave Goodell goodell

37 followers · 7 following
  • AWS
View GitHub Profile
@goodell
goodell / tmup
Last active December 31, 2015 23:39 — forked from simonjbeaumont/tmup
A version that deals with the empty line at the end so that it doesn't spew the result of a bare "export". It also splits on newlines instead of commas, so embedded commas in the envvar value won't be misinterpreted.
# adapted from https://gist.github.com/simonjbeaumont/4672606
function tmup() {
local IFS
echo -n "Updating to latest tmux environment...";
export IFS=$'\n';
for line in $(tmux showenv -t $(tmux display -p "#S"));
do
if [[ "$line" == -* ]]; then
unset $(echo $line | cut -c2-);
elif [[ -n "$line" ]] ; then
@goodell
goodell / octicons-regular-webfont-d930fd0ced1849be80ea26bc877594354d365893.woff
Created January 30, 2013 04:06
A cached copy of https://a248.e.akamai.net/assets.github.com/assets/octicons/octicons-regular-webfont-d930fd0ced1849be80ea26bc877594354d365893.woff downloaded on 2013-01-29. Cached here in a gist because it seems likely that the CDN link will eventually go stale. Perhaps there is a canonical link for this stuff instead?
wOFF�����`X�����?\�����������������������FFTM��X������fi??GDEF��t������ O�OS/2��?���?���`x?e?cmap��?���V��b?Iֈcvt ��,�������Dgasp��0������??�glyf��8��U��?�?M?head��WX���/���6_??hhea��W?������$hmtx��W?���?��N?ni?loca��XT��F��F8i
?maxp��Z?��� ��� ?name��Z?��H��?<?b?post��\��J�� mEI?Qwebf��`P������~CP?�������?=??����??����?.?x?c`d``?b `b`d`dT?,`�?�U�x?c`f??8?????՘? ?L?t???Ja???~  ?????�?\ , aF$%
?�g?
e�x?c```f?`F???| ?????d30?>`???????:????????� ?d? p???�bdL h
?=��?M?���D���??�x??? `T??7????9'?e?I23?3?IfK2?B 9?CX"?AP??�?R???Z+?.Z?^[QG?뽶?j/?.?B{??+????V?̛????93g? ?_????????????>??!?F??VYBd???'(i???? ?[?P?w'=)K?K??񴂧??T???$??w??p?I~VO??6(KN>2??!?ۿ??(??? 4B???#2nr???`??܁?&G8?n?t?#?V/???4?n թ?F??3??Ny2
xJk?
hD*(?֧?D??T??(?Q??WY??4????4 C"9?8@?~????G?{\??O<iPC5N?S?$???|\?~????a$???z[zz????h?:ՓQd,i!md?;B?jd
???P?̧?t?&ޝ$?"???tk?o?3d?`?_?Ȯ?sd?Y+?1R?aŜ?/_?sz&?]E?Z????_-}p%?????_ad`?D???D?M":Y<|
@goodell
goodell / bar.c
Created January 9, 2013 23:55
my (failed) attempt at creating a small reproducer for macports bug #36654 (https://trac.macports.org/ticket/36654)
__thread int t = 0xdeadbeef;
int foo(void)
{
return t;
}
@goodell
goodell / _git.patch
Created January 3, 2013 20:06
--- ./Completion/Unix/Command/_git 2012-03-06 14:04:43.000000000 -0600
+++ /usr/local/share/zsh/functions/_git 2013-01-03 14:05:17.000000000 -0600
@@ -4706,7 +4706,10 @@
local -a third_party_commands
local command
for command in $_git_third_party_commands; do
- (( $+commands[git-${command%%:*}] )) && third_party_commands+=$command
+ # XXX DJG disabling this filter, since it interferes with my "hub"
+ # subcommand completions. See https://github.com/defunkt/hub/issues/231
+ #(( $+commands[git-${command%%:*}] )) && third_party_commands+=$command
@goodell
goodell / gist:4190680
Created December 2, 2012 19:49
automake code snippet to deal with Fortran 90 modules, extracted from MPICH's (perl-generated) src/binding/f90/Makefile.mk
# variables for custom "silent-rules" for F90 modules
mod_verbose = $(mod_verbose_$(V))
mod_verbose_ = $(mod_verbose_$(AM_DEFAULT_VERBOSITY))
mod_verbose_0 = @echo " MOD " $@;
# we "manually" build the f90 sources and add them with LIBADD
lib_lib@MPILIBNAME@f90_la_SOURCES =
lib_lib@MPILIBNAME@f90_la_LIBADD += src/binding/f90/mpi_constants.lo
# now force libtool FC rules/variables to be generated and cause FC linking to
@goodell
goodell / _git-alias
Created October 31, 2012 05:39
beginning of a possible fix for mislav/hub#231
#compdef git-alias
#description show shell instructions for wrapping git
_arguments \
'-s[output shell script suitable for eval]' \
'1::shell:(zsh bash csh)'
@goodell
goodell / gist:2513853
Created April 27, 2012 22:16
Fatal error in MPI_Finalize: Other MPI error, error stack:
MPI_Finalize(293).................: MPI_Finalize failed
MPI_Finalize(213).................:
MPID_Finalize(117)................:
MPIDI_CH3U_VC_WaitForClose(385)...: an error occurred while the device was waiting for all open connections to close
MPIDI_CH3I_Progress(367)..........:
MPID_nem_mpich2_blocking_recv(904):
MPID_nem_tcp_connpoll(1838).......:
state_listening_handler(1906).....: accept of socket fd failed - Socket operation on non-socket
Fatal error in MPI_Finalize: Other MPI error, error stack:
@goodell
goodell / gist:2503321
Created April 26, 2012 21:17
static int
write_NC(NC *ncp)
{
int status, mpireturn, rank;
void *buf;
MPI_Offset hsz; /* header size with 0-padding if needed */
MPI_Status mpistatus;
assert(!NC_readonly(ncp));
@goodell
goodell / stdinfile1.diff
Created April 19, 2012 23:16
commit a5566493e8acfca25718cbf6539c562c52ddbf3e
Author: Dave Goodell <[email protected]>
Date: Thu Apr 19 18:07:54 2012 -0500
use better algorithms in MPIR_Allreduce_group
This is a copy of the non-SMP-aware code in MPIR_Allreduce_intra that
has been modified to perform rank translation immediately prior to any
communication calls. For cases where incremental group rank translation
is abnormally slow compared to batch translation, this is a non-optimal
@goodell
goodell / gist:2370635
Created April 12, 2012 20:08
commit 7592f7d3de4b8da5a33d387e74d87753608d512c
Author: Dave Goodell <[email protected]>
Date: Thu Apr 12 13:15:03 2012 -0500
tt#1599: eliminate MPIDI_CH3_PktGeneric_t
Without this change we are vulnerable to a nasty bit of compiler
behavior. MPIDI_CH3_PktGeneric_t sometimes included padding between the
"kind" and "pktptrs" fields. Then whole-structure assignments using the
generic packet (e.g., ch3_istartmsg.c:96) of buffers that are actually