http://blog.gmane.org/gmane.linux.systemtap hourly 1 1901-01-01T00:00+00:00 http://gmane.org/img/gmane-25t.png http://gmane.org http://comments.gmane.org/gmane.linux.systemtap/19375 <pre>A new automated systemtap source snapshot is available ftp://sourceware.org/pub/systemtap/snapshots/systemtap-20120526.tar.bz2 1973196 bytes, 143cad5 tag See also ftp://sourceware.org/pub/systemtap/snapshots/ </pre> fche< at >sourceware.org 2012-05-26T14:27:15 http://comments.gmane.org/gmane.linux.systemtap/19373 <pre>http://sourceware.org/bugzilla/show_bug.cgi?id=14168 Bug #: 14168 Summary: sanitize environment better for invoking kernel-module-builder make Product: systemtap Version: unspecified Status: NEW Severity: normal Priority: P2 Component: translator AssignedTo: systemtap&lt; at &gt;sourceware.org ReportedBy: fche&lt; at &gt;redhat.com Classification: Unclassified Sometimes stap is invoked with a $PATH etc. that is not suitable for building kernel modules, for example if an oddball compiler happens to be first. Encourage buildrun.cxx to use the "system" compiler, assumed in /usr/bin. Unfortunately, the kernel kbuild mechanism doesn't offer hints as to where the compiler may have been when the original kernel was built. </pre> fche at redhat dot com 2012-05-25T14:49:34 http://comments.gmane.org/gmane.linux.systemtap/19371 <pre>http://sourceware.org/bugzilla/show_bug.cgi?id=14165 Bug #: 14165 Summary: netfilter.stp -- extract IPV6 protocol info Product: systemtap Version: unspecified Status: NEW Severity: normal Priority: P2 Component: tapsets AssignedTo: systemtap&lt; at &gt;sourceware.org ReportedBy: smakarov&lt; at &gt;redhat.com Classification: Unclassified netfilter.stp needs to fully expose netfilter's IPv6 support. Currently we do not extract the protocol header, nor any of the information which requires extracting it. We need to wrap ipv6_skip_exthdr() for the extraction. </pre> smakarov at redhat dot com 2012-05-24T19:17:24 http://comments.gmane.org/gmane.linux.systemtap/19369 <pre>http://sourceware.org/bugzilla/show_bug.cgi?id=14164 Bug #: 14164 Summary: netfilter.stp -- expose arp and bridge protocol info Product: systemtap Version: unspecified Status: NEW Severity: normal Priority: P2 Component: tapsets AssignedTo: systemtap&lt; at &gt;sourceware.org ReportedBy: smakarov&lt; at &gt;redhat.com Classification: Unclassified Add support to the netfilter.stp tapset to use netfilter support for probing ARP and "bridge" protocols (e.g. probe netfilter.arp.in). Need to find out exactly which protocols correspond to "bridge". (follows on from PR13667) </pre> smakarov at redhat dot com 2012-05-24T19:04:58 http://comments.gmane.org/gmane.linux.systemtap/19367 <pre>http://sourceware.org/bugzilla/show_bug.cgi?id=14146 Bug #: 14146 Summary: tapset::* man pages should be generated from actual tapset files Product: systemtap Version: unspecified Status: NEW Severity: normal Priority: P2 Component: documentation AssignedTo: systemtap&lt; at &gt;sourceware.org ReportedBy: smakarov&lt; at &gt;redhat.com Classification: Unclassified Currently, the man/tapset::*.3stap manual pages are hand-written. This is unnecessary and error-prone duplication, as they could just as easily be auto-generated from the comments in the corresponding .stp file (as is already done for the tapset reference manual). </pre> smakarov at redhat dot com 2012-05-23T15:18:28 http://comments.gmane.org/gmane.linux.systemtap/19365 <pre>http://sourceware.org/bugzilla/show_bug.cgi?id=14137 Bug #: 14137 Summary: buildok/netfilter02.stp not ok Product: systemtap Version: unspecified Status: NEW Severity: normal Priority: P2 Component: translator AssignedTo: systemtap&lt; at &gt;sourceware.org ReportedBy: mjw&lt; at &gt;redhat.com CC: cmeek&lt; at &gt;redhat.com, smakarov&lt; at &gt;redhat.com Classification: Unclassified Host: Linux toonder.wildebeest.org 3.3.6-3.fc16.x86_64 #1 SMP Wed May 16 21:43:01 UTC 2012 x86_64 x86_64 x86_64 GNU/Linux Snapshot: version 1.8/0.153 commit release-1.7-249-g57db0e6 GCC: 4.6.3 [gcc (GCC) 4.6.3 20120306 (Red Hat 4.6.3-2)] Distro: Fedora release 16 (Verne) Running /home/mark/src/systemtap/testsuite/buildok/netfilter02.stp starting /home/mark/src/systemtap/testsuite/buildok/netfilter02.stp spawn1 stap -p4 /home/mark/src/systemtap/testsuite/buildok/netfilter02.stp spawn stap -p4 /home/mark/src/systemtap/testsuite/buildok/netfilter02.stp WARNING: side-effect-free probe 'probe_2009': keyword at /home/mark/src/systemtap/testsuite/buildok/netfilter02.stp:4:1 source: probe netfilter.hook("NF_INET_PRE_ROUTING").pf("NFPROTO_IPV4").priority("1") { } ^ WARNING: side-effect-free probe 'probe_2010': keyword at :5:1 WARNING: side-effect-free probe 'probe_2009': keyword at /home/mark/src/systemtap/testsuite/buildok/netfilter02.stp:4:1 source: probe netfilter.hook("NF_INET_PRE_ROUTING").pf("NFPROTO_IPV4").priority("1") { } ^ WARNING: side-effect-free probe 'probe_2010': keyword at :5:1 source: probe netfilter.hook("NF_INET_PRE_ROUTING").pf("NFPROTO_IPV4") { } ^ source: probe netfilter.hook("NF_INET_PRE_ROUTING").pf("NFPROTO_IPV4") { } ^ /tmp/stapTNRaVj/stap_ed0c182d855bb272214e55124b74424b_883_src.c:231:21: error: redefinition of 'enter_netfilter_probe_probe_2009' /tmp/stapTNRaVj/stap_ed0c182d855bb272214e55124b74424b_883_src.c:106:21: note: previous definition of 'enter_netfilter_probe_probe_2009' was here /tmp/stapTNRaVj/stap_ed0c182d855bb272214e55124b74424b_883_src.c:231:21: error: redefinition of 'enter_netfilter_probe_probe_2009' /tmp/stapTNRaVj/stap_ed0c182d855bb272214e55124b74424b_883_src.c:106:21: note: previous definition of 'enter_netfilter_probe_probe_2009' was here /tmp/stapTNRaVj/stap_ed0c182d855bb272214e55124b74424b_883_src.c:349:27: error: redefinition of 'netfilter_opts_probe_2009' /tmp/stapTNRaVj/stap_ed0c182d855bb272214e55124b74424b_883_src.c:224:27: note: previous definition of 'netfilter_opts_probe_2009' was here /tmp/stapTNRaVj/stap_ed0c182d855bb272214e55124b74424b_883_src.c:349:27: error: redefinition of 'netfilter_opts_probe_2009' /tmp/stapTNRaVj/stap_ed0c182d855bb272214e55124b74424b_883_src.c:224:27: note: previous definition of 'netfilter_opts_probe_2009' was here make[4]: *** [/tmp/stapTNRaVj/stap_ed0c182d855bb272214e55124b74424b_883_src.o] Error 1 make[4]: *** [/tmp/stapTNRaVj/stap_ed0c182d855bb272214e55124b74424b_883_src.o] Error 1 make[3]: *** [_module_/tmp/stapTNRaVj] Error 2 make[3]: *** [_module_/tmp/stapTNRaVj] Error 2 WARNING: make exited with status: 2 Pass 4: compilation failed. Try again with another '--vp 0001' option. WARNING: make exited with status: 2 Pass 4: compilation failed. Try again with another '--vp 0001' option. wait results: 30850 exp24 0 1 FAIL: buildok/netfilter02.stp </pre> mjw at redhat dot com 2012-05-23T11:30:27 http://comments.gmane.org/gmane.linux.systemtap/19363 <pre>Here's what I'm getting with the latest run of the inode-uprobes code: === systemtap Summary === # of expected passes3087 # of unexpected failures136 # of unexpected successes7 # of expected failures246 # of unknown successes1 # of known failures42 # of untested testcases83 # of unsupported tests2 Not too shabby. A good number of the failures, 73, comes from the fact that the inode-uprobes code doesn't have return probes yet and that the systemtap code doesn't support .absolute probes with inode-uprobes since there is no (semi-easy) way to turn a raw address into file inode+offset. Note that it is possible that the failures related to process.return and process.absolute probes are masking other problems. - Tests failing because of the lack of process.return probes (and the number of failures in parens): systemtap.base/at_var.exp (1) systemtap.base/bz10078.exp (2) systemtap.base/bz6850.exp (2) systemtap.base/global_var.exp (6) systemtap.base/library.exp (8) systemtap.base/process_by_cmd.exp (1) systemtap.base/uprobes.exp (4) systemtap.context/fib.exp (2) systemtap.context/uprobe_uaddr.exp (6) systemtap.unprivileged/unprivileged_myproc.exp (18) systemtap.unprivileged/unprivileged_probes.exp (18) - Tests failing because of the lack of process.absolute probes (and the number of failures in parens): systemtap.pass1-4/buildok.exp (1) systemtap.unprivileged/unprivileged_myproc.exp (2) systemtap.unprivileged/unprivileged_probes.exp (2) - Here's a list of failing testcases that looks like legitimate bugs in the new code: systemtap.base/itrace.exp systemtap.base/plt.exp systemtap.clone/dtrace_vfork_exec.exp systemtap.context/uprobe_stmt_num.exp systemtap.exelib/pthreadprobes.exp I've attached the systemtap.sum. </pre> David Smith 2012-05-21T18:58:56 http://comments.gmane.org/gmane.linux.systemtap/19361 <pre>A new automated systemtap source snapshot is available ftp://sourceware.org/pub/systemtap/snapshots/systemtap-20120519.tar.bz2 1966866 bytes, 4e8c027 tag See also ftp://sourceware.org/pub/systemtap/snapshots/ </pre> fche< at >sourceware.org 2012-05-19T14:27:23 http://comments.gmane.org/gmane.linux.systemtap/19351 <pre>http://sourceware.org/bugzilla/show_bug.cgi?id=14110 Bug #: 14110 Summary: Use XDG dirs instead of $HOME Product: systemtap Version: unspecified Status: NEW Severity: normal Priority: P2 Component: runtime AssignedTo: systemtap&lt; at &gt;sourceware.org ReportedBy: william.jon.mccann&lt; at &gt;gmail.com Classification: Unclassified Seems to currently use ~/.systemtap. It would be better to use the locations defined in the XDG Base Directory specification. https://live.gnome.org/GnomeGoals/XDGConfigFolders </pre> william.jon.mccann at gmail dot com 2012-05-15T00:01:38 http://comments.gmane.org/gmane.linux.systemtap/19346 <pre>http://sourceware.org/bugzilla/show_bug.cgi?id=14107 Bug #: 14107 Summary: Bad user unwinding from kernel fatal signal handler for some x86_64 kernels Product: systemtap Version: unspecified Status: NEW Severity: normal Priority: P2 Component: runtime AssignedTo: systemtap&lt; at &gt;sourceware.org ReportedBy: mjw&lt; at &gt;redhat.com CC: atomlin&lt; at &gt;redhat.com, bmr&lt; at &gt;redhat.com Classification: Unclassified The following program: int func (void) { int *foo = (void *) 0x1234; *foo = 0x12345; return 0; } int main (void) { return func (); } compiled with gcc -o bad_code bad_code.c and the following stap script: probe kernel.function("show_signal_msg") { /*(PF_USER | PR_WRITE) */ if (execname() == "bad_code") { if ($error_code &amp; 0x6) { printf ("\nUser mode process %s [pid: %d] received a SIGSEGV - error_code: 0x%x\n", execname(), pid(), $error_code) print_ubacktrace() } } } ran with: stap -d ./bad_code --ldd show_signal_msg.stp -c ./bad_code produces the following (correct) user backtrace on 3.3.5-2.fc16.x86_64: User mode process bad_code [pid: 18431] received a SIGSEGV - error_code: 0x6 0x400484 : func+0x10/0x1d [/usr/local/build/systemtap-obj/bad_code] 0x40049a : main+0x9/0xf [/usr/local/build/systemtap-obj/bad_code] 0x7fd419d1069d : __libc_start_main+0xed/0x1c0 [/lib64/libc-2.14.90.so] 0x4003b9 : _start+0x29/0x2c [/usr/local/build/systemtap-obj/bad_code] But on some other x86_64 kernels it produces: WARNING: _stp_read_address failed to access memory location User mode process bad_code [pid: 12152] received a SIGSEGV - error_code: 0x6 0x400484 : func+0x10/0x1d [/home/mark/build/systemtap-obj/bad_code] Warning: child process exited with signal 11 (Segmentation fault) WARNING: Number of errors: 0, skipped probes: 1 WARNING: /usr/local/install/systemtap/bin/staprun exited with status: 1 </pre> mjw at redhat dot com 2012-05-14T15:39:55 http://comments.gmane.org/gmane.linux.systemtap/19344 <pre>replace DEBUG_TASK_FINDER_VMA with a dbug_unwind-like macro: dbug_task_vma --- runtime/debug.h | 10 ++++++++++ runtime/task_finder.c | 8 ++------ runtime/task_finder2.c | 8 ++------ runtime/uprobes-common.c | 20 +++++--------------- runtime/vma.c | 40 ++++++++++------------------------------ 5 files changed, 29 insertions(+), 57 deletions(-) diff --git a/runtime/debug.h b/runtime/debug.h index 5c9cca7..cce06d1 100644 --- a/runtime/debug.h +++ b/runtime/debug.h &lt; at &gt;&lt; at &gt; -57,6 +57,16 &lt; at &gt;&lt; at &gt; #define dbug_unwind(level, args...) ; #endif +#if defined(DEBUG_TASK_FINDER_VMA) +#define dbug_task_vma(level, args...) do { \ + if ((level) &lt;= DEBUG_TASK_FINDER_VMA) \ + _stp_dbug(__FUNCTION__, __LINE__, args); \ + } while (0) +#else +#define dbug_task_vma(level, args...) ; +#endif + + #ifdef DEBUG_SYMBOLS #define dbug_sym(level, args...) do {\ if ((level) &lt;= DEBUG_SYMBOLS)\ diff --git a/runtime/task_finder.c b/runtime/task_finder.c index e436dec..5a00618 100644 --- a/runtime/task_finder.c +++ b/runtime/task_finder.c &lt; at &gt;&lt; at &gt; -668,8 +668,7 &lt; at &gt;&lt; at &gt; __stp_call_mmap_callbacks(struct stap_task_finder_target *tgt, if (tgt == NULL || tsk == NULL) return; -#ifdef DEBUG_TASK_FINDER_VMA -_stp_dbug(__FUNCTION__, __LINE__, +dbug_task_vma(1, "pid %d, a/l/o/p/path 0x%lx 0x%lx 0x%lx %c%c%c%c %s\n", tsk-&gt;pid, addr, length, offset, vm_flags &amp; VM_READ ? 'r' : '-', &lt; at &gt;&lt; at &gt; -677,7 +676,6 &lt; at &gt;&lt; at &gt; __stp_call_mmap_callbacks(struct stap_task_finder_target *tgt, vm_flags &amp; VM_EXEC ? 'x' : '-', vm_flags &amp; VM_MAYSHARE ? 's' : 'p', path); -#endif list_for_each(cb_node, &amp;tgt-&gt;callback_list_head) { struct stap_task_finder_target *cb_tgt; &lt; at &gt;&lt; at &gt; -1543,8 +1541,7 &lt; at &gt;&lt; at &gt; __stp_utrace_task_finder_target_syscall_exit(enum utrace_resume_action action, __stp_tf_handler_start(); rv = syscall_get_return_value(tsk, regs); -#ifdef DEBUG_TASK_FINDER_VMA -_stp_dbug(__FUNCTION__, __LINE__, +dbug_task_vma(1, "tsk %d found %s(0x%lx), returned 0x%lx\n", tsk-&gt;pid, ((entry-&gt;syscall_no == MMAP_SYSCALL_NO(tsk)) ? "mmap" &lt; at &gt;&lt; at &gt; -1555,7 +1552,6 &lt; at &gt;&lt; at &gt; __stp_utrace_task_finder_target_syscall_exit(enum utrace_resume_action action, ? "munmap" : "UNKNOWN")))), entry-&gt;arg0, rv); -#endif if (entry-&gt;syscall_no == MUNMAP_SYSCALL_NO(tsk)) { // Call the callbacks diff --git a/runtime/task_finder2.c b/runtime/task_finder2.c index 58486aa..a91a22b 100644 --- a/runtime/task_finder2.c +++ b/runtime/task_finder2.c &lt; at &gt;&lt; at &gt; -624,8 +624,7 &lt; at &gt;&lt; at &gt; __stp_call_mmap_callbacks(struct stap_task_finder_target *tgt, if (tgt == NULL || tsk == NULL) return; -#ifdef DEBUG_TASK_FINDER_VMA -_stp_dbug(__FUNCTION__, __LINE__, +dbug_task_vma(1, "pid %d, a/l/o/p/path 0x%lx 0x%lx 0x%lx %c%c%c%c %s\n", tsk-&gt;pid, addr, length, offset, vm_flags &amp; VM_READ ? 'r' : '-', &lt; at &gt;&lt; at &gt; -633,7 +632,6 &lt; at &gt;&lt; at &gt; __stp_call_mmap_callbacks(struct stap_task_finder_target *tgt, vm_flags &amp; VM_EXEC ? 'x' : '-', vm_flags &amp; VM_MAYSHARE ? 's' : 'p', path); -#endif list_for_each(cb_node, &amp;tgt-&gt;callback_list_head) { struct stap_task_finder_target *cb_tgt; &lt; at &gt;&lt; at &gt; -1458,8 +1456,7 &lt; at &gt;&lt; at &gt; __stp_utrace_task_finder_target_syscall_exit(u32 action, __stp_tf_handler_start(); rv = syscall_get_return_value(tsk, regs); -#ifdef DEBUG_TASK_FINDER_VMA -_stp_dbug(__FUNCTION__, __LINE__, +dbug_task_vma(1, "tsk %d found %s(0x%lx), returned 0x%lx\n", tsk-&gt;pid, ((entry-&gt;syscall_no == MMAP_SYSCALL_NO(tsk)) ? "mmap" &lt; at &gt;&lt; at &gt; -1470,7 +1467,6 &lt; at &gt;&lt; at &gt; __stp_utrace_task_finder_target_syscall_exit(u32 action, ? "munmap" : "UNKNOWN")))), entry-&gt;arg0, rv); -#endif if (in_atomic() || irqs_disabled()) { struct task_work *work; diff --git a/runtime/uprobes-common.c b/runtime/uprobes-common.c index f557f71..64b62cc 100644 --- a/runtime/uprobes-common.c +++ b/runtime/uprobes-common.c &lt; at &gt;&lt; at &gt; -253,9 +253,7 &lt; at &gt;&lt; at &gt; static int stap_uprobe_change_minus (struct task_struct *tsk, unsigned long relo static int stap_uprobe_process_found (struct stap_task_finder_target *tgt, struct task_struct *tsk, int register_p, int process_p) { const struct stap_uprobe_tf *stf = container_of(tgt, struct stap_uprobe_tf, finder); if (! process_p) return 0; /* ignore threads */ - #ifdef DEBUG_TASK_FINDER_VMA - _stp_dbug (__FUNCTION__,__LINE__, "%cproc pid %d stf %p %p path %s\n", register_p?'+':'-', tsk-&gt;tgid, tgt, stf, stf-&gt;pathname); - #endif + dbug_task_vma(1, "%cproc pid %d stf %p %p path %s\n", register_p?'+':'-', tsk-&gt;tgid, tgt, stf, stf-&gt;pathname); /* ET_EXEC events are like shlib events, but with 0 relocation bases */ if (register_p) { int rc = stap_uprobe_change_plus (tsk, 0, TASK_SIZE, stf, 0, 0); &lt; at &gt;&lt; at &gt; -288,12 +286,10 &lt; at &gt;&lt; at &gt; stap_uprobe_mmap_found (struct stap_task_finder_target *tgt, /* Check non-writable, executable sections for probes. */ if ((vm_flags &amp; VM_EXEC) &amp;&amp; !(vm_flags &amp; VM_WRITE)) { - #ifdef DEBUG_TASK_FINDER_VMA - _stp_dbug (__FUNCTION__,__LINE__, + dbug_task_vma (1, "+mmap X pid %d path %s addr %p length %u offset %p stf %p %p path %s\n", tsk-&gt;tgid, path, (void *) addr, (unsigned)length, (void*) offset, tgt, stf, stf-&gt;pathname); - #endif rc = stap_uprobe_change_plus (tsk, addr, length, stf, offset, vm_flags); } &lt; at &gt;&lt; at &gt; -303,12 +299,10 &lt; at &gt;&lt; at &gt; stap_uprobe_mmap_found (struct stap_task_finder_target *tgt, * no error (rc == 0), we need to look for semaphores too. */ if ((rc == 0) &amp;&amp; (vm_flags &amp; VM_WRITE)) { - #ifdef DEBUG_TASK_FINDER_VMA - _stp_dbug (__FUNCTION__,__LINE__, + dbug_task_vma (1, "+mmap W pid %d path %s addr %p length %u offset %p stf %p %p path %s\n", tsk-&gt;tgid, path, (void *) addr, (unsigned)length, (void*) offset, tgt, stf, stf-&gt;pathname); - #endif rc = stap_uprobe_change_semaphore_plus (tsk, addr, length, stf); } &lt; at &gt;&lt; at &gt; -318,9 +312,7 &lt; at &gt;&lt; at &gt; stap_uprobe_mmap_found (struct stap_task_finder_target *tgt, /* The task_finder_munmap_callback */ static int stap_uprobe_munmap_found (struct stap_task_finder_target *tgt, struct task_struct *tsk, unsigned long addr, unsigned long length) { const struct stap_uprobe_tf *stf = container_of(tgt, struct stap_uprobe_tf, finder); - #ifdef DEBUG_TASK_FINDER_VMA - _stp_dbug (__FUNCTION__,__LINE__, "-mmap pid %d addr %p length %lu stf %p %p path %s\n", tsk-&gt;tgid, (void *) addr, length, tgt, stf, stf-&gt;pathname); - #endif + dbug_task_vma (1, "-mmap pid %d addr %p length %lu stf %p %p path %s\n", tsk-&gt;tgid, (void *) addr, length, tgt, stf, stf-&gt;pathname); return stap_uprobe_change_minus (tsk, addr, length, stf); } &lt; at &gt;&lt; at &gt; -330,9 +322,7 &lt; at &gt;&lt; at &gt; static int stap_uprobe_munmap_found (struct stap_task_finder_target *tgt, struct static int stap_uprobe_process_munmap (struct stap_task_finder_target *tgt, struct task_struct *tsk, int register_p, int process_p) { const struct stap_uprobe_tf *stf = container_of(tgt, struct stap_uprobe_tf, finder); if (! process_p) return 0; /* ignore threads */ - #ifdef DEBUG_TASK_FINDER_VMA - _stp_dbug (__FUNCTION__,__LINE__, "%cproc pid %d stf %p %p path %s\n", register_p?'+':'-', tsk-&gt;tgid, tgt, stf, stf-&gt;pathname); - #endif + dbug_task_vma (1, "%cproc pid %d stf %p %p path %s\n", register_p?'+':'-', tsk-&gt;tgid, tgt, stf, stf-&gt;pathname); /* Covering 0-&gt;TASK_SIZE means "unmap everything" */ if (!register_p) return stap_uprobe_change_minus (tsk, 0, TASK_SIZE, stf); diff --git a/runtime/vma.c b/runtime/vma.c index 8f758d0..b1091f2 100644 --- a/runtime/vma.c +++ b/runtime/vma.c &lt; at &gt;&lt; at &gt; -34,9 +34,7 &lt; at &gt;&lt; at &gt; static void _stp_vma_match_vdso(struct task_struct *tsk) unsigned long vdso_addr = tsk-&gt;mm-&gt;context.vdso_base; #endif -#ifdef DEBUG_TASK_FINDER_VMA - _dbug("tsk: %d vdso: 0x%lx\n", tsk-&gt;pid, vdso_addr); -#endif + dbug_task_vma(1,"tsk: %d vdso: 0x%lx\n", tsk-&gt;pid, vdso_addr); for (i = 0; i &lt; _stp_num_modules &amp;&amp; found == NULL; i++) { struct _stp_module *m = _stp_modules[i]; &lt; at &gt;&lt; at &gt; -47,10 +45,8 &lt; at &gt;&lt; at &gt; static void _stp_vma_match_vdso(struct task_struct *tsk) unsigned long notes_addr; int all_ok = 1; notes_addr = vdso_addr + m-&gt;build_id_offset; -#ifdef DEBUG_TASK_FINDER_VMA - _dbug("notes_addr %s: 0x%lx + 0x%lx = 0x%lx (len: %x)\n", m-&gt;name, + dbug_task_vma(1,"notes_addr %s: 0x%lx + 0x%lx = 0x%lx (len: %x)\n", m-&gt;name, vdso_addr, m-&gt;build_id_offset, notes_addr, m-&gt;build_id_len); -#endif for (j = 0; j &lt; m-&gt;build_id_len; j++) { int rc; &lt; at &gt;&lt; at &gt; -80,10 +76,8 &lt; at &gt;&lt; at &gt; static void _stp_vma_match_vdso(struct task_struct *tsk) #endif if (rc || b != m-&gt;build_id_bits[j]) { -#ifdef DEBUG_TASK_FINDER_VMA - _dbug("darn, not equal (rc=%d) at %d (0x%x != 0x%x)\n", + dbug_task_vma(1,"darn, not equal (rc=%d) at %d (0x%x != 0x%x)\n", rc, j, b, m-&gt;build_id_bits[j]); -#endif all_ok = 0; break; } &lt; at &gt;&lt; at &gt; -97,9 +91,7 &lt; at &gt;&lt; at &gt; static void _stp_vma_match_vdso(struct task_struct *tsk) stap_add_vma_map_info(tsk, vdso_addr, vdso_addr + found-&gt;sections[0].size, "vdso", found); -#ifdef DEBUG_TASK_FINDER_VMA - _dbug("found vdso: %s\n", found-&gt;path); -#endif + dbug_task_vma(1,"found vdso: %s\n", found-&gt;path); } } #endif /* STAPCONF_MM_CONTEXT_VDSO */ &lt; at &gt;&lt; at &gt; -112,11 +104,9 &lt; at &gt;&lt; at &gt; static int _stp_vma_exec_cb(struct stap_task_finder_target *tgt, int register_p, int process_p) { -#ifdef DEBUG_TASK_FINDER_VMA - _stp_dbug(__FUNCTION__, __LINE__, + dbug_task_vma(1, "tsk %d:%d , register_p: %d, process_p: %d\n", tsk-&gt;pid, tsk-&gt;tgid, register_p, process_p); -#endif if (process_p) { if (register_p) &lt; at &gt;&lt; at &gt; -144,11 +134,9 &lt; at &gt;&lt; at &gt; static int _stp_vma_mmap_cb(struct stap_task_finder_target *tgt, if (path == NULL || *path == '\0') /* unknown? */ path = (char *)name; /* we'll copy this soon, in ..._add_vma_... */ -#ifdef DEBUG_TASK_FINDER_VMA -_stp_dbug(__FUNCTION__, __LINE__, +dbug_task_vma(1, "mmap_cb: tsk %d:%d path %s, addr 0x%08lx, length 0x%08lx, offset 0x%lx, flags 0x%lx\n", tsk-&gt;pid, tsk-&gt;tgid, path, addr, length, offset, vm_flags); -#endif // We are only interested in the first load of the whole module that // is executable. We register whether or not we know the module, // so we can later lookup the name given an address for this task. &lt; at &gt;&lt; at &gt; -159,11 +147,9 &lt; at &gt;&lt; at &gt; static int _stp_vma_mmap_cb(struct stap_task_finder_target *tgt, { unsigned long vm_start = 0; unsigned long vm_end = 0; -#ifdef DEBUG_TASK_FINDER_VMA - _stp_dbug(__FUNCTION__, __LINE__, + dbug_task_vma(1, "vm_cb: matched path %s to module (sec: %s)\n", path, _stp_modules[i]-&gt;sections[0].name); -#endif module = _stp_modules[i]; /* Make sure we really don't know about this module yet. If we do know, we might want to extend &lt; at &gt;&lt; at &gt; -200,12 +186,10 &lt; at &gt;&lt; at &gt; static int _stp_vma_mmap_cb(struct stap_task_finder_target *tgt, { res = stap_add_vma_map_info(tsk-&gt;group_leader, addr, addr + length, path, NULL); -#ifdef DEBUG_TASK_FINDER_VMA - _stp_dbug(__FUNCTION__, __LINE__, + dbug_task_vma(1, "registered '%s' for %d (res:%d) [%lx-%lx]\n", path, tsk-&gt;group_leader-&gt;pid, res, addr, addr + length); -#endif } } else if (path != NULL) { &lt; at &gt;&lt; at &gt; -216,12 +200,10 &lt; at &gt;&lt; at &gt; static int _stp_vma_mmap_cb(struct stap_task_finder_target *tgt, // precisely to module names and symbols. res = stap_extend_vma_map_info(tsk-&gt;group_leader, addr, addr + length); -#ifdef DEBUG_TASK_FINDER_VMA -_stp_dbug(__FUNCTION__, __LINE__, +dbug_task_vma(1, "extended '%s' for %d (res:%d) [%lx-%lx]\n", path, tsk-&gt;group_leader-&gt;pid, res, addr, addr + length); -#endif } return 0; } &lt; at &gt;&lt; at &gt; -265,10 +247,8 &lt; at &gt;&lt; at &gt; static int _stp_vma_init(void) _stp_error("Couldn't initialize vma map: %d\n", rc); return rc; } -#ifdef DEBUG_TASK_FINDER_VMA -_stp_dbug(__FUNCTION__, __LINE__, +dbug_task_vma(1, "registering vmcb (_stap_target: %d)\n", _stp_target); -#endif rc = stap_register_task_finder_target (&amp; vmcb); if (rc != 0) _stp_error("Couldn't register task finder target: %d\n", rc); </pre> Negreanu Marius Adrian 2012-05-12T14:31:58 http://comments.gmane.org/gmane.linux.systemtap/19343 <pre>A new automated systemtap source snapshot is available ftp://sourceware.org/pub/systemtap/snapshots/systemtap-20120512.tar.bz2 1967475 bytes, d6d8634 tag See also ftp://sourceware.org/pub/systemtap/snapshots/ </pre> fche< at >sourceware.org 2012-05-12T14:27:20 http://comments.gmane.org/gmane.linux.systemtap/19340 <pre> Attached patch fixes build of systemtap with automake 1.12 </pre> Nitin A Kamble 2012-05-08T16:45:28 http://comments.gmane.org/gmane.linux.systemtap/19338 <pre>http://sourceware.org/bugzilla/show_bug.cgi?id=14079 Bug #: 14079 Summary: caller() pass-4 error if no uretprobes in script Product: systemtap Version: unspecified Status: NEW Severity: normal Priority: P2 Component: tapsets AssignedTo: systemtap&lt; at &gt;sourceware.org ReportedBy: fche&lt; at &gt;redhat.com Classification: Unclassified stap -e 'probe module("tun").function("tun_net_xmit&lt; at &gt;drivers/net/tun.c").call { printf("tun_net_xmit %x from %s\n", $skb-&gt;ip_summed, caller()); }' results in a stap -p4 compilation error, due to this bit in the caller embedded-c: #ifdef STAPCONF_UPROBE_GET_PC else if (CONTEXT-&gt;probe_type == _STP_PROBE_HANDLER_URETPROBE) THIS-&gt;__retvalue = (int64_t)(long)_stp_ret_addr_r(CONTEXT-&gt;ips.ri); #endif with apparently no one including enough uprobes.h goop to make this from runtime/common_probe_context.h compilable: /* uretprobe state */ struct uretprobe_instance *ri; This could be fixed a couple of ways, but my favorite would be to use this as an opportunity to do bug #6580, which would reimplement caller() etc. in terms of backtracing functions instead of such low-level embedded-c runes. </pre> fche at redhat dot com 2012-05-08T16:15:38 http://comments.gmane.org/gmane.linux.systemtap/19335 <pre>http://sourceware.org/bugzilla/show_bug.cgi?id=14078 Bug #: 14078 Summary: librpmio.so exports bundled broken glob(3) clone Product: systemtap Version: unspecified Status: NEW Severity: normal Priority: P2 Component: translator AssignedTo: systemtap&lt; at &gt;sourceware.org ReportedBy: mjw&lt; at &gt;redhat.com CC: fche&lt; at &gt;redhat.com Classification: Unclassified Created attachment 6401 --&gt; http://sourceware.org/bugzilla/attachment.cgi?id=6401 RHBZ-819680-librpmio.so-exports-bundled-broken-glob-.patch See https://bugzilla.redhat.com/show_bug.cgi?id=819680 The attached patch makes sure we always explicitly use the glibc glob version. Not yet applied, Frank is looking into another solution using -lc that doesn't require adding our own stap_glob function. Tested on RHEL5, F16 and F17. </pre> mjw at redhat dot com 2012-05-08T14:15:31 http://comments.gmane.org/gmane.linux.systemtap/19332 <pre>http://sourceware.org/bugzilla/show_bug.cgi?id=14073 Bug #: 14073 Summary: setjmp.exp testcase broken Product: systemtap Version: unspecified Status: NEW Severity: normal Priority: P2 Component: testsuite AssignedTo: systemtap&lt; at &gt;sourceware.org ReportedBy: mjw&lt; at &gt;redhat.com Classification: Unclassified The setjmp.exp testcase is broken in several ways: - It uses -D_FORTIFY_SOURCE=2 without -O1 or -O2, which basically does nothing. - When adding -O1 or -O2 (with or without -D_FORTIFY_SOURCE=2) the testcase fails because of backtrace mismatch. - At least on Fedora 17 glibc doesn't have setjmp/longjmp markers, so the whole test doesn't actually test anything </pre> mjw at redhat dot com 2012-05-07T19:23:01 http://comments.gmane.org/gmane.linux.systemtap/19327 <pre>Signed-off-by: Negreanu Marius Adrian &lt;adrian.m.negreanu&lt; at &gt;intel.com&gt; --- runtime/unwind.c | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) diff --git a/runtime/unwind.c b/runtime/unwind.c index a4f35ee..ff36b2b 100644 --- a/runtime/unwind.c +++ b/runtime/unwind.c &lt; at &gt;&lt; at &gt; -382,7 +382,7 &lt; at &gt;&lt; at &gt; static void set_no_state_rule(uleb128_t reg, enum item_location where, static void set_offset_rule(uleb128_t reg, enum item_location where, sleb128_t svalue, struct unwind_state *state) { -dbug_unwind(1, "reg=%lx, where=%d, svalue=%lx\n", reg, where, value); +dbug_unwind(1, "reg=%lx, where=%d, svalue=%lx\n", reg, where, svalue); if (reg &lt; ARRAY_SIZE(REG_STATE.regs)) { REG_STATE.regs[reg].where = where; REG_STATE.regs[reg].state.off = svalue; &lt; at &gt;&lt; at &gt; -485,7 +485,7 &lt; at &gt;&lt; at &gt; static int processCFI(const u8 *start, const u8 *end, unsigned long targetLoc, case DW_CFA_offset_extended_sf: value = get_uleb128(&amp;ptr.p8, end); svalue = get_sleb128(&amp;ptr.p8, end); -dbug_unwind(1, "map DW_CFA_offset_extended_sf value %ld to reg_info idx %ld, with offset: %ld\n", value, DWARF_REG_MAP(value)); +dbug_unwind(1, "map DW_CFA_offset_extended_sf value %ld to reg_info idx %ld, with offset: %ld\n", value, DWARF_REG_MAP(value), svalue); value = DWARF_REG_MAP(value); set_offset_rule(value, Memory, svalue * state-&gt;dataAlign, </pre> Negreanu Marius Adrian 2012-05-07T11:57:16 http://comments.gmane.org/gmane.linux.systemtap/19313 <pre>A new automated systemtap source snapshot is available ftp://sourceware.org/pub/systemtap/snapshots/systemtap-20120505.tar.bz2 1965209 bytes, 3c10a1b tag See also ftp://sourceware.org/pub/systemtap/snapshots/ </pre> fche< at >sourceware.org 2012-05-05T14:27:17 http://comments.gmane.org/gmane.linux.systemtap/19302 <pre>http://sourceware.org/bugzilla/show_bug.cgi?id=14057 Bug #: 14057 Summary: ERROR: Your privilege credentials (stapsys) are insufficient to run this module (stapdev required). Product: systemtap Version: unspecified Status: NEW Severity: normal Priority: P2 Component: runtime AssignedTo: systemtap&lt; at &gt;sourceware.org ReportedBy: mjw&lt; at &gt;redhat.com CC: brolley&lt; at &gt;redhat.com Classification: Unclassified commit 429a4963c958e082c1199c5c3266c45bd4122907 Author: Dave Brolley &lt;brolley&lt; at &gt;redhat.com&gt; Date: Tue Apr 17 15:16:47 2012 -0400 BZ 813323: stapusr-only users no longer able to run unsigned modules in /lib - Elevate the user's privilege level to stapdev when the module is loaded fr /lib/modules/`uname -r`/systemtap. - Suppress error messages about failure to verify the module's signature when a signature is not required. After this commit I get the following when using stap as root: sudo /home/mark/install/systemtap/bin/stap /home/mark/src/systemtap/testsuite/systemtap.base/backtrace.stp ERROR: Your privilege credentials (stapsys) are insufficient to run this module (stapdev required). WARNING: /home/mark/install/systemtap/bin/staprun exited with status: 1 Pass 5: run failed. Try again with another '--vp 00001' option. This results in lots and lots of sudo make installcheck failures. This is with systemtap from git on Red Hat Enterprise Linux Workstation release 6.2 (Santiago) i686. Reverting the above patch makes everything work again. </pre> mjw at redhat dot com 2012-05-03T09:33:41 http://comments.gmane.org/gmane.linux.systemtap/19285 <pre>Hi, I am following-up remark made in the context of "Making the transport layer more robust" topic. Goal is to reduce number of systemtap wake-ups in the context of embedded low-power use cases like MP3 playback. The last non tunable or small duration source of wake-up is "reader_thread" ppoll timeout value, every 200ms. We compile systemtap with a value of 2s or more for our daily use. So I am proposing to introduce tunables below as an example. Don't know if I should have correlated s and ns, also don't know if this should also be applied to relay_old.c. Please tell me if this is an acceptable change: From 3fac053713be44c00e05423b4f31e2ed8edaa993 Mon Sep 17 00:00:00 2001 From: Frederic Turgis &lt;f-turgis&lt; at &gt;ti.com&gt; Date: Wed, 2 May 2012 01:16:28 +0200 Subject: [PATCH] Make reader_thread poll timeout tunable at compilation --- runtime/staprun/relay.c | 9 ++++++++- 1 files changed, 8 insertions(+), 1 deletions(-) diff --git a/runtime/staprun/relay.c b/runtime/staprun/relay.c index e08ff60..d81a45f 100644 --- a/runtime/staprun/relay.c +++ b/runtime/staprun/relay.c &lt; at &gt;&lt; at &gt; -12,6 +12,13 &lt; at &gt;&lt; at &gt; #include "staprun.h" +#ifndef STP_RELAY_TIMEOUT_S +#define STP_RELAY_TIMEOUT_S 0 +#endif +#ifndef STP_RELAY_TIMEOUT_NS +#define STP_RELAY_TIMEOUT_NS 200000000 +#endif + int out_fd[NR_CPUS]; static pthread_t reader[NR_CPUS]; static int relay_fd[NR_CPUS]; &lt; at &gt;&lt; at &gt; -125,7 +132,7 &lt; at &gt;&lt; at &gt; static void *reader_thread(void *data) char buf[131072]; int rc, cpu = (int)(long)data; struct pollfd pollfd; - struct timespec tim = {.tv_sec=0, .tv_nsec=200000000}, *timeout = &amp;tim; + struct timespec tim = {.tv_sec=STP_RELAY_TIMEOUT_S, .tv_nsec=STP_RELAY_TIMEOUT_NS}, *timeout = &amp;tim; sigset_t sigs; off_t wsize = 0; int fnum = 0; -- Regards FRed OMAP Platform Business Unit - System Platform Engineering - Platform &amp; Product Entitlement Texas Instruments France SA, 821 Avenue Jack Kilby, 06270 Villeneuve Loubet. 036 420 040 R.C.S Antibes. Capital de EUR 753.920 </pre> Turgis, Frederic 2012-05-01T20:09:49 http://comments.gmane.org/gmane.linux.systemtap/19284 <pre>A new automated systemtap source snapshot is available ftp://sourceware.org/pub/systemtap/snapshots/systemtap-20120428.tar.bz2 1966234 bytes, 155e594 tag See also ftp://sourceware.org/pub/systemtap/snapshots/ </pre> fche< at >sourceware.org 2012-04-28T14:27:17 Search the mailing list at Gmane query http://search.gmane.org/?group=$group=gmane.linux.systemtap