search for: rtld_now

..., both can be installed diff --git a/plugins/vddk/vddk.c b/plugins/vddk/vddk.c index c49eebcd..b988946b 100644 --- a/plugins/vddk/vddk.c +++ b/plugins/vddk/vddk.c @@ -267,7 +267,11 @@ load_library (void) exit (EXIT_FAILURE); } +#ifdef HAVE_DLMOPEN + dl = dlmopen (LM_ID_NEWLM, path, RTLD_NOW); +#else dl = dlopen (path, RTLD_NOW); +#endif if (dl != NULL) break; if (i == 0) { -- 2.25.0

I couldn't get this to work in the end. This is the latest non-working version. This email documents what doesn't work for the permanent record. The central problem is that VDDK InitEx() appears to dlopen() various of its own plugins. Although I wasn't able to capture exactly what dlopen() command it is running, the plugins cannot be loaded because they rely on the recompiled

...runs programs. > Is it using LD_PRELOAD trickery, or replacing libc, or using seccomp? If I understand the question correctly - it's about how collectd loads its plugins? If so it uses: static int plugin_load_file(const char *file, _Bool global) { void (*reg_handle)(void); int flags = RTLD_NOW; if (global) flags |= RTLD_GLOBAL; void *dlh = *dlopen*(file, flags); //... reg_handle = (void (*)(void))*dlsym*(dlh, "module_register"); //... *(*reg_handle)();* } Does this give any clues? Best Regards, Peter On Wed, Nov 7, 2018 at 12:56 PM Richard W.M. Jones &l...

http://dovecot.procontrol.fi/test/ Again major maildir syncing updates: - uidlist file is accessed and rewritten a lot less often now than before - mail duplicates are detected and handled properly by renaming them to new base name - message flag updates and expunging before did two full syncs for cur/ directory. now it does only one. - some fixes to syncing in read-only / out-of-quota

...newest library in case multiple exist. */ - "libvixDiskLib.so.6", - "libvixDiskLib.so.5", - }; - size_t i; - CLEANUP_FREE char *orig_error = NULL; - - /* Load the library. */ - for (i = 0; i < sizeof sonames / sizeof sonames[0]; ++i) { - dl = dlopen (sonames[i], RTLD_NOW); - if (dl != NULL) - break; - if (i == 0) { - orig_error = dlerror (); - if (orig_error) - orig_error = strdup (orig_error); - } - } - if (dl == NULL) { - nbdkit_error ("%s\n\n" - "If '%s' is located on a non-standard path...

...in gdb, it >> reports >> >> [Thread debugging using libthread_db enabled] >> Cannot find new threads: generic error >> >> at >> >> Plugin::load() >> { >> #ifdef ENABLE_PLUGINS >>  this->handle_ = dlopen(this->filename_.c_str(), RTLD_NOW); >> >> in binutil/src/gold/plugin.cc >> >> The gdb version is 7.0.1. I tried it with other multi-threaded >> programs without any problem. Have I missed something? Or I shouldn't >> debug my pass this way at all? >> > > Probably the best approach...

...return odelay; } $NetBSD: patch-ac,v 1.1.1.1 2001/03/15 20:35:58 wiz Exp $ --- ../libao-0.6.0.orig/src/audio_out.c Fri Feb 23 17:31:46 2001 +++ src/audio_out.c Thu Mar 15 10:21:55 2001 @@ -63,7 +63,7 @@ driver_tree_t *dt; void *handle; - handle = dlopen(plugin_file, RTLD_NOW); + handle = dlopen(plugin_file, DL_LAZY); if (handle) { dt = (driver_tree_t *)malloc(sizeof(driver_tree_t)); if (!dt) return NULL; $NetBSD: patch-ad,v 1.1 2001/03/16 10:10:35 wiz Exp $ --- src/ao_wav.c.orig Sat Feb 24 02:31:46 2001 +++ src/ao_wav.c @@ -2...

Thomas Lumley writes: > > > I've been trying to get Matt Calder's S compiler to work in R using R > COMPILE and R SHLIB so it will be fairly platform-independent. > > I'm sure someone has claimed in the past that it is possible/easy to > dyn.load() two dynamic libraries and have one call functions in the other. > I can't get it to work (Red Hat

...LOBAL 0 # endif #endif /* We may have to define LTDL_LAZY_OR_NOW in the command line if we find out it does not work in some platform. */ #ifndef LTDL_LAZY_OR_NOW # ifdef RTLD_LAZY # define LTDL_LAZY_OR_NOW RTLD_LAZY # else # ifdef DL_LAZY # define LTDL_LAZY_OR_NOW DL_LAZY # else # ifdef RTLD_NOW # define LTDL_LAZY_OR_NOW RTLD_NOW # else # ifdef DL_NOW # define LTDL_LAZY_OR_NOW DL_NOW # else # define LTDL_LAZY_OR_NOW 0 # endif # endif # endif # endif #endif fnord() { int i=42;} main() { void *self, *ptr1, *ptr2; self=dlopen(0,LTDL_GLOBAL|LTDL_LAZY_OR_NOW); if(s...

...--git a/plugins/vddk/vddk.c b/plugins/vddk/vddk.c index db61c1d8..c49eebcd 100644 --- a/plugins/vddk/vddk.c +++ b/plugins/vddk/vddk.c @@ -256,7 +256,18 @@ load_library (void) /* Load the library. */ for (i = 0; i < sizeof sonames / sizeof sonames[0]; ++i) { - dl = dlopen (sonames[i], RTLD_NOW); + CLEANUP_FREE char *path; + + /* Set the full path so that dlopen will preferentially load the + * system libraries from the same directory. + */ + if (asprintf (&path, "%s/lib%d/%s", + libdir ? : VDDK_LIBDIR, 8*SIZEOF_LONG, sonames[i]) == -1) { +...

Hi, I have downloaded the following files libao-0.6.0.tar.gz libogg-1.0beta4.tar.gz libvorbis-1.0beta4.tar.gz vorbis-tools-1.0beta4.tar.gz as said from the download section. But I'm at a loss to what do after that. I don't have much experience/knowledge in compiling etc, but I still want to try out ogg. (I downloaded the source coz I have an OpenBSD machine, and there was no

Differences from v4: Patch 1 is simplified: I realized that since we already use -rdynamic for nbdkit (after all, we WANT our dlopen()d plugins to be able to call our nbdkit_* exports), it is only a matter of adding dlopen to the set of symbols that we export. With that done, there is no separate shared library needed; our dlopen shim is now part of nbdkit proper, and we don't have to tweak

..."libvixDiskLib.so.6", + "lib64/libvixDiskLib.so.5", "libvixDiskLib.so.5", }; size_t i; @@ -256,9 +263,28 @@ load_library (void) /* Load the library. */ for (i = 0; i < sizeof sonames / sizeof sonames[0]; ++i) { - dl = dlopen (sonames[i], RTLD_NOW); - if (dl != NULL) + CLEANUP_FREE char *path; + + /* Set the full path so that dlopen will preferentially load the + * system libraries from the same directory. + */ + if (asprintf (&path, "%s/%s", libdir, sonames[i]) == -1) { + nbdkit_error ("asprintf:...

I'm a lot happier with this version: no mucking with dlmopen(). It does add a bit more coupling between nbdkit proper and the vddk plugin (namely, nbdkit now exports a new function that the vddk plugin relies on), but patch 2 adds testsuite coverage of the new function so we shouldn't regress. Patch 1 and 2 are new, patch 3 is unchanged from when Rich posted it in v2, and patch 4 is

...quot;lib64/libvixDiskLib.so.5", + "lib32/libvixDiskLib.so.5", "libvixDiskLib.so.5", }; size_t i; @@ -348,9 +374,25 @@ load_library (void) /* Load the library. */ for (i = 0; i < sizeof sonames / sizeof sonames[0]; ++i) { - dl = dlopen (sonames[i], RTLD_NOW); - if (dl != NULL) + CLEANUP_FREE char *path; + + /* Set the full path so that dlopen will preferentially load the + * system libraries from the same directory. + */ + if (asprintf (&path, "%s/%s", libdir, sonames[i]) == -1) { + nbdkit_error ("asprintf:...

..., and now I want to debug it using gdb. I use ld as my target file, but when I run the program in gdb, it reports [Thread debugging using libthread_db enabled] Cannot find new threads: generic error at Plugin::load() { #ifdef ENABLE_PLUGINS this->handle_ = dlopen(this->filename_.c_str(), RTLD_NOW); in binutil/src/gold/plugin.cc The gdb version is 7.0.1. I tried it with other multi-threaded programs without any problem. Have I missed something? Or I shouldn't debug my pass this way at all? btw, "llvm-gcc -use-gold-plugin" works fine when I compile my program. Thanks ver...

..._PRELOAD trickery, or replacing libc, or using seccomp? > > If I understand the question correctly - it's about how collectd loads its > plugins? If so it uses: > > static int plugin_load_file(const char *file, _Bool global) { > void (*reg_handle)(void); > int flags = RTLD_NOW; > if (global) > flags |= RTLD_GLOBAL; > void *dlh = *dlopen*(file, flags); > //... > reg_handle = (void (*)(void))*dlsym*(dlh, "module_register"); > //... > *(*reg_handle)();* > } > > > Does this give any clues? Can you strace the c...

...: `RTLD_DEFAULT' undeclared (first use this > function) > DynamicLinker.cpp:40: error: (Each undeclared identifier is reported only once > for each function it appears in.) > ----------------- > > well, the only value supported in cygwin's dlfcn.h are RTLD_LAZY, > RTLD_NOW and RTLD_GLOBAL. Does the cygwin version of dlopen return the main process if 0 is passed in as the first parameter? The wording in the linux man page looks like this: If filename is a NULL pointer, then the returned handle is for the main program. If so, this patch should fix the...

...;: DynamicLinker.cpp:40: error: `RTLD_DEFAULT' undeclared (first use this function) DynamicLinker.cpp:40: error: (Each undeclared identifier is reported only once for each function it appears in.) ----------------- well, the only value supported in cygwin's dlfcn.h are RTLD_LAZY, RTLD_NOW and RTLD_GLOBAL. So, just for checking compilation I've changed the line return dlsym (RTLD_DEFAULT, symbolName); in DynamicLinker.cpp to the following: return dlsym (/*RTLD_DEFAULT*/(void*)RTLD_LAZY, symbolName); than compilation itself finished, just linkage failed with one screen ou...

Hi I met problem developing plug-in under FreeBSD 6.1. First I tried to use internal plugins like quota and acl but I got: IMAP(xxx): module /usr/local/lib/dovecot/imap/lib01_acl_plugin.so: dlsym(acl_plugin_init) failed: Shared object "nss_dns.so.1" not found, required by "imap" This library doesn't exist on FreeBSD 6.1 Same story when u mention any plugin in