summaryrefslogtreecommitdiffstats
path: root/runtime
Commit message (Collapse)AuthorAgeFilesLines
...
| | * | | | | | add runtime/autoconf magic for itrace on pre-find_task_by_vpid kernelsFrank Ch. Eigler2009-03-191-0/+6
| | | | | | | |
| | * | | | | | Fix for spinlock bad magic error with itrace probe pointMaynard Johnson2009-03-181-6/+2
| | | |_|_|_|/ | | |/| | | |
| | * | | | | Don't mutex_unlock in __stp_tf_get_vma_map_entry_internal.Mark Wielaard2009-03-181-1/+0
| | | |_|_|/ | | |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | All callers of __stp_tf_get_vma_map_entry_internal lock and unlock the mutex correctly themselves in all cases. This unlock would trigger a double unlock. * runtime/task_finder_vma.c (__stp_tf_get_vma_map_entry_internal): Don't mutex_unlock(&__stp_tf_vma_mutex).
| | * | | | Get the canonical path of the main file for comparison at runtime.Mark Wielaard2009-03-172-1/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When given directly by the user through -d or in case of the kernel name and path might differ. path should be used for matching. * runtime/sym.h (_stp_module): Add path field. * runtime/task_finder.c (__stp_tf_vm_cb): Use module path to compare vm_path. * translate.cxx (dump_unwindsyms): Output canonical path.
| | * | | | PR9951: Prevent GCC warnings in deref()Josh Stone2009-03-161-6/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | In some configurations, GCC was warning about a possible use of _v in the deref macros. I could not reproduce the error, but the only case where _v is not written is if lookup_bad_addr rejects the address, in which case we will hit DEREF_FAULT and _v won't be used. Now we're priming _v=0 anyway, so GCC has no right to complain...
| | * | | | Move vma module tracking from pr6866 branch to master.Mark Wielaard2009-03-152-7/+45
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * tapsets.cxx (utrace_derived_probe_group::emit_module_decls): Always emit vm callback probe for __stp_tf_vm_cb. * runtime/task_finder.c (__stp_tf_vm_cb): Always expose, move _stp_dbug statements under ifdef DEBUG_TASK_FINDER_VMA. Find and record corresponding module when vm_path not NULL. * runtime/task_finder_vma.c (struct __stp_tf_vma_entry): Add _stp_module. (stap_add_vma_map_info): Add _stp_module argument and assign. (__stp_tf_get_vma_entry_addr): New static function to get the __stp_tf_vma_entry given an address.
| | * | | | Move lookup_bad_addr call in STAPCONF_PROBE_KERNELJosh Stone2009-03-131-8/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | With most of the implementations, kread/kwrite call deref/store_deref, and so it makes sense to have lookup_bad_addr in the latter as an underlying address check. However, in the STAPCONF_PROBE_KERNEL case that uses probe_kernel_read and probe_kernel_write, the roles are reversed, so lookup_bad_addr needs to be in kread/kwrite. Also note that __deref_bad and __store_deref_bad should only be used in cases that can be determined at compile time. These turn into invalid symbols which prevent the module from loading. (They might be better replaced with compile-time assertions.)
| | * | | | PR9947: move runtime cleanup out of the work queueJosh Stone2009-03-134-8/+35
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The kernel lockdep checking found a possible deadlock if a forced rmmod tried to destroy _stp_work_queue at the same time that the work queue was unregistering tracepoints. An unlikely scenario, but still possible. Now the work queue will just issue a STP_REQUEST_EXIT down to usermode, and usermode will echo back an STP_EXIT that triggers the actual probe cleanup. This way the unregistrations are happening in exactly the same context as the registrations were.
| * | | | | Don't reject opcode 0x9b -- fwait, which is also the first byte ofJim Keniston2009-03-114-16/+12
| |/ / / / | | | | | | | | | | | | | | | | | | | | | | | | | pseudo-instructions like fclex, finit, fstsw, and fstcw. Do reject opcode 0x82 in 64-bit mode.
* | | | | Start at supporting the original transport (STP_TRANSPORT_VERSION=1).David Smith2009-05-125-114/+100
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * runtime/print.c: Only use print_new.c. * runtime/print_new.c (stp_print_flush): Add STP_TRANSPORT_VERSION 1 support. * runtime/transport/transport.c: Removed inclusion of procfs.c. * runtime/transport/procfs.c (_stp_proc_read): Adapt to new interface. (_stp_proc_write): Ditto. * runtime/transport/relayfs.c: Ditto.
* | | | | Sets uid/gid on "dropped" file.David Smith2009-05-081-0/+2
| | | | | | | | | | | | | | | | | | | | | | | | | * runtime/transport/relay_v2.c (_stp_transport_data_fs_init): Sets uid/gid on "dropped" file.
* | | | | Cleaned up runtime/transport/relay_v2.c.David Smith2009-05-081-181/+15
| | | | | | | | | | | | | | | | | | | | * runtime/transport/relay_v2.c: Cleanup.
* | | | | Working transport version 2 (RHEL5 and recent fedora kernels).David Smith2009-05-081-414/+376
| | | | | | | | | | | | | | | | | | | | | | | | | * runtime/transport/relay_v2.c: Morphed utt.c into relay_v2.c. Actually works.
* | | | | Hid details of internal ring_buffer.c structure.David Smith2009-05-073-52/+79
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * runtime/print_new.c (stp_print_flush): Calls _stp_data_entry_data() to get data pointer. * runtime/transport/transport.h: Removed _stp_entry definition. Added _stp_data_entry_data() declaration. * runtime/transport/ring_buffer.c: Uses new _stp_data_entry structure. (_stp_data_entry_data): Added.
* | | | | Added comments.David Smith2009-05-061-0/+37
| | | | | | | | | | | | | | | | | | | | * runtime/transport/transport.h: Added comments.
* | | | | Start of support for older transport and cleanup.David Smith2009-05-064-51/+573
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * runtime/transport/control.c: Added inclusion of linux/delay.h to get declaration of msleep(). * runtime/transport/relay_v2.c: New skeleton file for transport version 2. * runtime/transport/ring_buffer.c: Cleanup. * runtime/transport/transport.c: Ditto.
* | | | | Renamed STP_OLD_TRANSPORT to STP_TRANSPORT_VERSION.David Smith2009-04-028-15/+26
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 2009-04-02 David Smith <dsmith@redhat.com> * runtime.h: Defines STP_TRANSPORT_VERSION instead of STP_OLD_TRANSPORT. * staprun/staprun.h (STP_OLD_TRANSPORT): Ditto. * print.c: Changed STP_OLD_TRANSPORT to STP_TRANSPORT_VERSION. * staprun/mainloop.c (stp_main_loop): Ditto. * transport/transport.c: Ditto. * transport/transport.h: Ditto. * transport/transport_msgs.h: Ditto. * transport/utt.h: Ditto.
* | | | | 2009-03-31 David Smith <dsmith@redhat.com>David Smith2009-03-313-105/+64
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * print_new.c (stp_print_flush): Pushed MAX_RESERVE logic down to _stp_data_write_reserve(). Now just keeps calling _stp_data_write_reserve() until it has written the entire print buffer. * transport/ring_buffer.c (_stp_data_write_reserve): Breaks large reserve requests down into smaller ones for better buffer use. Returns the number of bytes reserved. * transport/transport.h: Updated _stp_data_write_reserve() prototype.
* | | | | 2009-03-30 David Smith <dsmith@redhat.com>David Smith2009-03-301-89/+79
| | | | | | | | | | | | | | | | | | | | * print_new.c: Whitespace/indent changes only.
* | | | | Working bulkmode support.David Smith2009-03-132-12/+93
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 2009-03-13 David Smith <dsmith@redhat.com> * print_new.c (stp_print_flush): Added bulkmode support for new transport. * transport/ring_buffer.c (__stp_find_next_entry): Fixed syntax error in bulkmode code. (_stp_transport_data_fs_init): Changed 'for_each_possible_cpu()' to 'for_each_online_cpu()' so that non-online cpu's won't have a trace file created.
* | | | | More bulkmode support.David Smith2009-03-131-44/+23
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 2009-03-13 David Smith <dsmith@redhat.com> * transport/ring_buffer.c (_stp_data_open_trace): Passes the inode's private data into the file's private data. (_stp_trace_seq_to_user): Deleted unused function. (trace_seq_reset): Ditto. (__stp_find_next_entry): Renamed and added bulkmode support. (_stp_find_next_entry): Updated call. (_stp_data_read_trace): Updated call.
* | | | | More bulkmode support. Only allows for one reader of all trace files.David Smith2009-03-131-7/+29
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 2009-03-13 David Smith <dsmith@redhat.com> * transport/ring_buffer.c (__stp_free_ring_buffer): Frees _stp_trace_reader_cpumask. (__stp_alloc_ring_buffer): Allocates and clears _stp_trace_reader_cpumask. (_stp_data_open_trace): Instead of using an atomic variable, uses a cpumask variable to allow for only one reader of trace files. (_stp_data_release_trace): Clears cpumask when trace files are closed.
* | | | | Start of STP_BULKMODE support.David Smith2009-03-121-12/+38
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 2009-03-12 David Smith <dsmith@redhat.com> * transport/ring_buffer.c (_stp_transport_data_fs_init): Start of STP_BULKMODE support - creates one trace file per cpu. (_stp_transport_data_fs_close): Cleans up one trace file per cpu.
* | | | | Merge branch 'master' of ssh://sources.redhat.com/git/systemtap into pr7043David Smith2009-03-128-118/+153
|\| | | |
| * | | | Adapt to linux-next commit changing __alloc_percpu API.Frank Ch. Eigler2009-03-082-22/+9
| | | | | | | | | | | | | | | | | | | | | | | | | After linux-next commit f2a8205c, it takes two parameters again, so we autoconf for it rather than use KERNEL_VERSION ifdefs.
| * | | | stap autoconf test for kernel stack trace supportTim Moore2009-03-044-6/+27
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * buildrun.cxx (compile_pass): Add autoconf line for stack trace test, which defines STAPCONF_KERNEL_STACKTRACE. * runtime/autoconf-save-stack-trace.c: New file. * runtime/stack.c : Use STAPCONF_KERNEL_STACKTRACE instead of tests for kernel configuration and versions. * runtime/stack-i386.c : ditto * runtime/stack-x86_64.c : ditto
| * | | | This patch updates the itrace code to support the new utrace interface.Maynard Johnson2009-03-031-29/+117
| | | | | | | | | | | | | | | | | | | | | | | | | It also adds a private copy of access_process_vm to runtime/itrace.c since that function is not consistently exported by all distros.
| * | | | PR9875: Remove sduprobesWenji Huang2009-03-031-61/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Impact: cleanup Since sdt.h is in git tree, sduprobes is no longer needed. * Makefile.am: Remove sduprobes. * Makefile.in: Regenerated. * doc/Makefile.in: Ditto. * runtime/sduprobes.c: Removed. Signed-off-by: Wenji Huang <wenji.huang@oracle.com>
* | | | | Updated buffer size.David Smith2009-03-121-1/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 2009-03-12 David Smith <dsmith@redhat.com> * transport/ring_buffer.c (__stp_alloc_ring_buffer): Updated buffer size.
* | | | | Small update.David Smith2009-03-112-3/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 2009-03-11 David Smith <dsmith@redhat.com> * print_new.c (stp_print_flush): Updated MAX_RESERVE_SIZE. * transport/ring_buffer.c (__stp_alloc_ring_buffer): Reserves more space.
* | | | | Handles polling and breaks down large buffers.David Smith2009-03-112-8/+60
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 2009-03-11 David Smith <dsmith@redhat.com> * print_new.c (stp_print_flush): Breaks up the buffer into smaller pieces (since the ring_buffer likes lots of small events, not one large one). * transport/ring_buffer.c (__stp_alloc_ring_buffer): Reserves space for struct _stp_entry. (_stp_data_poll_trace): New function. (_stp_data_write_commit): Wakes up any tasks waiting on data.
* | | | | Cleanup.David Smith2009-03-091-162/+15
| | | | |
* | | | | First working version that actually produces output.David Smith2009-03-094-72/+262
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 2009-03-09 David Smith <dsmith@redhat.com> * print_new.c (stp_print_flush): Calls _stp_data_write_reserve() and _stp_data_write_commit(). * transport/ring_buffer.c (__stp_alloc_ring_buffer): Sets up a default buffer size of STP_BUFFER_SIZE; (trace_seq_reset): New function. (peek_next_entry): New function. (__find_next_entry): New function. (find_next_entry_inc): New function. (_stp_data_read_trace): Uses find_next_entry_inc() to get the next entry, then calls _stp_entry_to_user() to copy it to the user's buffer, then calls ring_buffer_consume() to consume it. (_stp_data_write_reserve): New function. (_stp_data_write_commit): New function. * transport/transport.c (_stp_transport_close): Calls functions that were ifdef'ed out. * transport/transport.h (struct _stp_entry): Added _stp_entry definition and _stp_data_write_reserve()/_stp_data_write_commit() prototypes.
* | | | | Start of new ring_buffer transport.David Smith2009-03-044-10/+363
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 2009-03-04 David Smith <dsmith@redhat.com> * runtime.h: Added _stp_exit() prototype. * transport/ring_buffer.c: New file. * transport/transport.c: Removed unneeded utt_trace lines. Includes transport/ring_buffer.c. (_stp_transport_fs_init): Calls _stp_transport_data_fs_init(). (_stp_transport_fs_close): Calls _stp_transport_data_fs_close(). * transport/transport.h: Added prototypes.
* | | | | More cleanup.David Smith2009-02-258-40/+68
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 2009-02-25 David Smith <dsmith@redhat.com> * debug.h: Removed unused variable '_stp_transport_state'. * print_new.c (stp_print_flush): Ifdef'ed out call to utt_reserve(). * runtime.h: Added _stp_warn() prototype. * transport/control.c: Includes control.h, mempool.c, and symbols.c. Renamed '_stp_attached' to '_stp_ctl_attached'. * transport/control.h: Removed _stp_pool_q declaration. * transport/debugfs.c (_stp_register_ctl_channel_fs): Uses _stp_get_module_dir(). * transport/transport.c: Cleanup. * transport/transport.h: Ditto.
* | | | | Major hacking to minimal config.David Smith2009-02-2410-40/+175
|/ / / / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 2009-02-24 David Smith <dsmith@redhat.com> * print.h: New file. * print.c: Includes new print.h file. * print_new.c (stp_print_flush): Commented out some code that needs replacing. * sym.c: Includes sym.h. * transport/debugfs.c: Includes transport.h. * transport/relayfs.c: Updated _stp_get_root_dir() call. * transport/transport.c: Ifdef'ed out most setup. * transport/transport.h: Ifdef'ed out all but basics. * transport/utt.c (utt_create_tree): Updated _stp_get_root_dir() call. * transport/utt.h: Commented out 'rchan' in struct utt_trace. (utt_reserve): Ifdef'ed out guts.
* | | | Add selective use of _stp_time_init/kill (PR9822)Josh Stone2009-02-232-7/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Our gettimeofday runtime has frequent wakeups to stay in sync with kernel time, but this is wasted effort if gettimeofday is not used in the script. This patch moves the calls to _stp_time_init and _stp_time_kill into begin and end/error probes, which only get pulled in if one of the gettimeofday variants is called.
* | | | Better <linux/debugfs.h> header file inclusion.David Smith2009-02-202-1/+2
| |_|/ |/| | | | | | | | | | | | | | | | | 2009-02-20 David Smith <dsmith@redhat.com> * debugfs.c: Added inclusion of linux/debugfs.h. * relayfs.h: Removed inclusion of linux/debugfs.h.
* | | Kill all ChangeLogsJosh Stone2009-02-193-4291/+0
| | | | | | | | | | | | | | | | | | Mark Wielaard, a dear friend of the departed, will be performing a song and dance at the funeral services to commemorate the joyous times that they had together.
* | | PR9866 band-aid: deactivate new CONFIG_STACKTRACE code for older kernelsFrank Ch. Eigler2009-02-184-5/+12
| | |
* | | Remove context argument from _stp_stack_print_fallback (PR 9860).William Cohen2009-02-182-4/+9
| | |
* | | Removed unused functions and variables.David Smith2009-02-1816-374/+43
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 2009-02-18 David Smith <dsmith@redhat.com> * io.c (_stp_log): Removed unused function. * map.c (_stp_cmp): Replace _stp_log() with dbug(). * mempool.c (_stp_mempool_resize): Removed unused function. * print.c (next_fmt): Removed unused function. * procfs.c: Removed unused variable '_stp_num_procfs_files'. * regs.c (_stp_ret_addr): Removed unused function. * string.c (_stp_text_str): Removed unused variable 'len'. * string.h: Removed unused variable '_stdout_' and function declaration for deleted function '_stp_vsprintf'. * sym.c: Removed unused variables. * unwind.c (_stp_create_unwind_hdr): Removed unused function. 2009-02-18 David Smith <dsmith@redhat.com> * control.c: Removed unused variable '_stp_current_buffers'. * procfs.c (_stp_set_buffers): Removed unused function. (_stp_register_ctl_channel_fs): Removed unused variables and label. * symbols.c (u32_swap): Removed unused function. (generic_swap): Ditto. (_stp_sort): Ditto. (_stp_section_is_interesting): Ditto. * transport.c (_stp_transport_init): Removed unused variable 'ret'.
* | | Move sdt.h to includes/sys and use in tests.Mark Wielaard2009-02-172-180/+4
| | |
* | | Reduced control channel code duplication.David Smith2009-02-176-270/+132
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 2009-02-17 David Smith <dsmith@redhat.com> * control.c: Contains generic control channel functions. * procfs.c: Specific procfs control channel functions. All generic control channel functions moved to control.c. * debugfs.c: New file containing debugfs specific control channel functions. * control.h: New file. * transport.c: Updated file inclusion.
* | | sdt.h (STAP_PROBE): Take provider and probe as arguments.Mark Wielaard2009-02-162-1/+5
| | |
* | | Fix )( typo in sdt.h STAP_PROBE9 parm9.Mark Wielaard2009-02-162-1/+5
| | |
* | | Handle c++ static user probes via .probe, c via .label.Stan Cox2009-02-152-99/+51
| | |
* | | Fixed stack.c compile problems on systems with older kernels (like RHEL4).David Smith2009-02-132-0/+5
| | | | | | | | | | | | | | | | | | | | | 2009-02-13 David Smith <dsmith@redhat.com> * stack.c: Fixed compile problems on systems with older kernels (like RHEL4).
* | | Fixed transport/procfs.c for systems that use the old transport (RHEL4).David Smith2009-02-134-3/+19
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 2009-02-13 David Smith <dsmith@redhat.com> * procfs.c: Added macros to guard against multiple inclusion. 2009-02-13 David Smith <dsmith@redhat.com> * procfs.c: Added inclusion of ../procfs.c for _stp_mkdir_proc_module().
* | | Moved inclusion of runtime/procfs.c to a better place.David Smith2009-02-122-1/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 2009-02-12 David Smith <dsmith@redhat.com> * tapsets.cxx (procfs_derived_probe_group::emit_module_decls): Moved inclusion of procfs.c here from runtime/transport.c 2009-02-12 David Smith <dsmith@redhat.com> * transport.c: Moved inclusion of procfs.c to procfs_derived_probe_group::emit_module_decls() in tapsets.cxx.
generated by cgit (git 2.34.1) at 2025-08-21 05:57:54 +0000