Add Package: htop
Adds package htop. Changes made based upon e-mails with Mike Frysinger and Gerhard Heift. The autoreconf change is important for anyone using uClibc as there is no widely available backtrace (yet).
---
package/Config.in 2011-03-21 17:29:22.000000000 -0500 renamed from package/Config.in 2011-03-14 10:39:51.000000000 -0500 +3 -0
@@ -476,6 +476,9 @@ menu "System tools" if BR2_PACKAGE_BUSYBOX_SHOW_OTHERS source "package/bootutils/Config.in" +endif +source "package/htop/Config.in" +if BR2_PACKAGE_BUSYBOX_SHOW_OTHERS source "package/module-init-tools/Config.in" source "package/procps/Config.in" source "package/psmisc/Config.in"
package/htop/Config.in 2011-03-21 17:29:22.000000000 -0500 renamed from package/htop/Config.in 1969-12-31 18:00:00.000000000 -0600 +8 -0
@@ -0,0 +1,8 @@ +config BR2_PACKAGE_HTOP + bool "htop" + select BR2_PACKAGE_NCURSES + help + htop is an interactive text-mode process viewer for Linux. + It aims to be a better top. + + http://sourceforge.net/projects/htop/
package/htop/htop.patch 2011-03-21 17:29:22.000000000 -0500 renamed from package/htop/htop.patch 1969-12-31 18:00:00.000000000 -0600 +70 -0
70 lines not on this page · raw message
package/htop/htop.mk 2011-03-22 09:31:54.000000000 -0500 renamed from package/htop/htop.mk 1969-12-31 18:00:00.000000000 -0600 +12 -0
@@ -0,0 +1,12 @@ +############################################################# +# +# htop +# +############################################################# +HTOP_VERSION = 0.9 +HTOP_SOURCE = htop-$(HTOP_VERSION).tar.gz +HTOP_SITE = http://$(BR2_SOURCEFORGE_MIRROR).dl.sourceforge.net/htop/$(HTOP_VERSION) +HTOP_DEPENDENCIES = ncurses +HTOP_AUTORECONF=YES + +$(eval $(call AUTOTARGETS,package,htop))
Quoted 3 lines · Andy Kennedy
Adds package htop. ?Changes made based upon e-mails with Mike Frysinger and Gerhard Heift. ?The autoreconf change is important for anyone using uClibc as there is no widely available backtrace (yet).
this ver looks mostly sane ...
Quoted 1 line
++AC_CHECK_HEADERS([execinfo.h],[:],[:])
no need for munging the other branches
AC_CHECK_HEADERS([execinfo.h])
Quoted 4 lines
++if test "$cross_compiling" = "no"; then
+ AC_CHECK_FILE($PROCDIR/stat,,AC_MSG_ERROR(Cannot find/proc/stat. Make sure you have a Linux-compatible /proc filesystem mounted. See the file README for help.))
+ AC_CHECK_FILE($PROCDIR/meminfo,,AC_MSG_ERROR(Cannot find/proc/meminfo. Make sure you have a Linux-compatible /proc filesystem mounted. See the file README for help.)) ++fi
AC_CHECK_FILE has fallback cases for cross-compiling i thought, so you shouldnt need to check cross_compiling yourself
-mike
Quoted 18 lines · Andy Kennedy
Quoted 1 line
Adds package htop. ?Changes made based upon e-mails with Mike
Frysinger
Quoted 1 line
and Gerhard Heift. ?The autoreconf change is important for anyone
using
Quoted 1 line
uClibc as there is no widely available backtrace (yet).
this ver looks mostly sane ...
Quoted 1 line
++AC_CHECK_HEADERS([execinfo.h],[:],[:])
no need for munging the other branches
AC_CHECK_HEADERS([execinfo.h])Quoted 2 lines
++if test "$cross_compiling" = "no"; then
+ AC_CHECK_FILE($PROCDIR/stat,,AC_MSG_ERROR(Cannot find
/proc/stat. Make sure you have a Linux-compatible /proc filesystem mounted. See the file README for help.))
Quoted 8 lines
Quoted 1 line
+ AC_CHECK_FILE($PROCDIR/meminfo,,AC_MSG_ERROR(Cannot find
/proc/meminfo. Make sure you have a Linux-compatible /proc filesystem mounted. See the file README for help.))Quoted 1 line
++fi
AC_CHECK_FILEhas fallback cases for cross-compiling i thought, so you
shouldnt need to checkcross_compilingyourself
The error was really strange. I submitted this patch to htop and they tested, modified and included it as a fix. The problem was that when one is cross compiling, the test for /proc fails since the build is a cross build -- there were enough smarts to not test when cross building, however, the return value was a failure for the /proc tests. This _should_ not occur so we skip the whole test.
This patch was derived by getting the current branch via svn and running a diff against configure.ac.
Quoted 1 line
-mike
Andy
Quoted 2 lines · ANDY KENNEDY
This patch was derived by getting the current branch via svn and running a diff against configure.ac.
ok, np then. i might mention in the patch header "this is already in upstream svn".
-mike
Quoted 7 lines · ANDY KENNEDY
Quoted 2 lines
This patch was derived by getting the current branch via svn and running a diff against configure.ac.
ok, np then. i might mention in the patch header "this is already in
upstream svn".
Kinda sorta did:
+# This patch removes the hard dependency on backtrace by checking for +# it at compile time and adds a cross-compile /proc check. +# Patch given to me by Hisham <[email protected]> (htop-general)
Hisham is the main developer. You want me to modify this patch again and resubmit? I don't mind -- and it wouldn't take long.
Quoted 1 line
-mike
Andy
Quoted 8 lines · ANDY KENNEDY
Quoted 6 lines · ANDY KENNEDY
Quoted 2 lines
This patch was derived by getting the current branch via svn and running a diff against configure.ac.
ok, np then. ?i might mention in the patch header "this is already in upstream svn".
Kinda sorta did:
yes, but the sorta isnt explicit ;)
Quoted 6 lines
package/htop/htop.patch 2011-03-21 17:29:22.000000000 -0500 renamed from package/htop/htop.patch 1969-12-31 18:00:00.000000000 -0600 +3
-0+# This patch removes the hard dependency on backtrace by checking for +# it at compile time and adds a cross-compile /proc check. +# Patch given to me by Hisham <[email protected]> (htop-general)Hisham is the main developer. ?You want me to modify this patch again and resubmit? ?I don't mind -- and it wouldn't take long.
yes, please add an explicit "This patch is taken from upstream svn." -mike