summaryrefslogtreecommitdiffstats
path: root/tapset
Commit message (Collapse)AuthorAgeFilesLines
...
* (kernel_int) added goto success; (PR 4419)dwilder2007-03-012-0/+5
|
* Created an s390x version of syscall.mmap, syscall.mmap2, syscall.mmap.return,dwilder2007-02-223-0/+177
| | | | syscall.mmap2.return.
* 2007-02-09 Frank Ch. Eigler <fche@elastic.org>fche2007-02-092-11/+15
| | | | * conversions.stp (*): Make errmsg usage uniform.
* 2007-02-06 Josh Stone <joshua.i.stone@intel.com>jistone2007-02-0727-804/+691
| | | | | | | | | | | | | | | | | | | | | | | | | * aux_syscalls.stp, inet_sock.stp, ioblock.stp, ioscheduler.stp, nfs.stp, nfs_proc.stp, nfsd.stp, rpc.stp, scsi.stp, signal.stp, socket.stp, task.stp, tcp.stp, vfs.stp: Protect pointer dereferences with kread wherever possible. Some places still have hazards, as marked with FIXMEs. * errno.stp (returnstr): Don't use return in tapset C functions. * aux_syscalls.stp (__uget_timex_m): Ditto. * nfsd.stp (__get_fh): Ditto. * nfs.stp, vfs.stp (<many functions>): Ditto. * string.stp (substr): Ditto. Also make sure start index is valid. * syscalls.stp (syscall.execve): Change __string to kernel_string. LKET/ * nfs.stp, nfs_proc.stp, nfsd.stp, process.stp, tskdispatch.stp: Protect pointer dereferences with kread wherever possible. Some places still have hazards, as marked with FIXMEs. * aio.stp (log_io_getevents): Don't use return in tapset C functions. * timestamp.stp (set_timing_method): Ditto. * utils.stp (filter_by_pid): Ditto.
* * kernel_{string,long,...,char} protected accessor functionsfche2007-02-062-3/+62
| | | | | | | | | | | | | | | | | | | | * test suites for all of 'em too * even documentation * mm, donuts 2007-02-06 Frank Ch. Eigler <fche@elastic.org> * stapfuncs.5.in: Add docs for kernel_{long,int,short,char} and some user_string* variants. 2007-02-06 Frank Ch. Eigler <fche@elastic.org> * conversions.stp (kernel_long/int/short/char): New functions. 2007-02-06 Frank Ch. Eigler <fche@elastic.org> * buildok/conversions.stp: Build-test all conversions.stp functions. * systemtap.stress/conversions.*: New test.
* 2007-01-31 Martin Hunt <hunt@redhat.com>hunt2007-01-312-14/+17
| | | | | | | | * context.stp (print_backtrace): Modify for new _stp_stack_print. (backtrace): Just use _stp_stack_snprint(). (print_stack): Use new _stp_print* functions. (probefunc): Use _stp_symbol_snprint.
* 2007-01-26 Josh Stone <joshua.i.stone@intel.com>jistone2007-01-272-2/+5
| | | | * vfs.stp (ppos_pos): Change deref() to kread() so i686 passes.
* __is_idle should be pure.jistone2007-01-251-1/+1
|
* Fix for missing SOCK_PASSSEC in kernels earlier than 2.6.18.mmason2007-01-252-1/+9
|
* 2007-01-25 Martin Hunt <hunt@localhost.localdomain>hunt2007-01-252-1/+16
| | | | * syscalls2.stp: Add sys32_ustat probe.
* Changed bit checking method in sock_flags_num2str().mmason2007-01-242-8/+17
|
* New sockets tapset.mmason2007-01-232-0/+1081
|
* Added new string functions tokenize() and strtol().mmason2007-01-232-0/+36
|
* In _stp_sockaddr_str() s390x needs sockaddr_ll.sll_addr printed as a long ↵dwilder2007-01-181-1/+1
| | | | int instead of long long int.
* Added several ? to probe definitions for kernel functions that are not ↵dwilder2007-01-181-8/+8
| | | | availableon s390. This is needed to make probe syscall.* work.
* Added s390x support in function errno_str.dwilder2007-01-181-0/+2
|
* 2007-01-17 Martin Hunt <hunt@redhat.com>hunt2007-01-173-0/+23
| | | | | * syscalls.stp: Add syscall.creat. * syscalls2.stp: Add compat_sys_open.
* Fix for ioblocktest.exp test (PR 3867)mmason2007-01-132-2/+6
|
* Added ? to request_key probes in x86_64/syscalls.stp to preventmmason2007-01-132-2/+7
| | | | syscall.* failures on some kernels.
* bugfix for 2.6.19 kernelguanglei2006-12-299-108/+306
|
* fix wrong HOOKID for _addevent.sunrpc.svc.recv.entryguanglei2006-12-232-2/+8
| | | | and _addevent.sunrpc.svc.recv.return
* bugfix of LKET rpc event register codes.guanglei2006-12-203-12/+18
|
* 2006-12-18 Frank Ch. Eigler <fche@elastic.org>fche2006-12-182-1/+11
| | | | | | | | | | | PR 3079 * loc2c-runtime.h (deref, store_deref): Fork x86 and x86-64 variants. Remove dysfunctional 64-bit ops from x86. 2006-12-18 Frank Ch. Eigler <fche@elastic.org> * vfs.stp (ppos_pos): Protect contents with deref(), though this blocks operation on i686 due to bug #3079.
* *** empty log message ***mmason2006-12-131-0/+5
|
* In scheduler.ctxswitch, fixed last line to reference $prev_p->state instead ↵mmason2006-12-131-1/+1
| | | | of $prev_p->pid.
* patch of nfs, nfs_proc, rpc tapsets for 2.6.9 kernelguanglei2006-12-117-95/+229
|
* scheduler.stp, vfs.stp: made idle_balance and buffer_migrate_page optinal.guanglei2006-11-293-3/+9
|
* 2006-11-20 Frank Ch. Eigler <fche@elastic.org>fche2006-11-202-1/+5
| | | | * context.stp (caller_addr): Declare return type.
* bugfix for #3526:guanglei2006-11-194-11/+24
| | | | | | | | | runtime/lket/b2a/lket_b2a.[ch]: bugfix for #3536 lket_trace.stp: add the logging of pid, ppid process.stp: add logging of tid, ppid for lket_internal.process.execve. Change to use for_each_process() in process_snapshot() to get the info of all running processes. register_event.stp: update the registering codes of execve event.
* add addevent.process.exit.entryguanglei2006-11-175-2/+38
|
* use c global variables instead of stap global variables to avoid better ↵guanglei2006-11-158-82/+118
| | | | | | performance(unfinished). visit members in scripts for better safety check
* register_event.stp:guanglei2006-11-112-4/+10
| | | | pass integer values explicitly as 64-bit into _lket_trace() to avoid error when parse va-arg by _stp_printf
* bugfix of NFS event register codes for LKETguanglei2006-11-102-11/+4
| | | | bugfix of lket-b2a
* tapset/LKET/iosyscall.stp:guanglei2006-11-103-819/+870
| | | | | | | | | add some trace aliases tapset/LKET/register_event.stp: register sys events in embedded c codes for stap to better optimize the generated codes Update LKET man page accordingly
* add options for lket-b2a to control the output data fields of lket.out.guanglei2006-11-093-9/+24
| | | | | | | | make fork/execce trace hooks lket internally and turned on by default. add stoptrace_fork and stoptrace_execve flags for lket to control stopping fork/execve in user scripts. update lket-b2a and lket man page.
* signal.stp, memory.stp:guanglei2006-11-073-3/+10
| | | | | make do_mmap2, copy_cow_page and signal.handle.return optional since they are obsolete or failed to probe on 2.6.18.1/ppc64
* 2006-11-02 Frank Ch. Eigler <fche@elastic.org>fche2006-11-022-0/+12
| | | | | | | | | * Makefile.am: Install tapset/README too. * Makefile.in: Regenerated. 2006-11-02 Frank Ch. Eigler <fche@elastic.org> * README: Add a blurb against running these scripts directly.
* update lket.5.in to add more details and re-arrange the sections.guanglei2006-10-312-223/+228
| | | | modify tapsets/LKET/register_events to change the event_desc strings
* 2006-10-29 Thang Nguyen <thang.p.nguyen@intel.com>tpnguyen2006-10-292-1/+9
| | | | | * context.stp: Added proper handling of statement * probe in probefunc() for IA64 (bz# 3423)
* tapset/LKET/register_event.stp: changes of variable name and checking forguanglei2006-10-232-18/+26
| | | | user registered events
* tapset/scsi.stp:guanglei2006-10-163-7/+13
| | | | | | | check whether rq_disk is empty tapset/scheduler.stp revert changes
* * tapset/scsi.stp:guanglei2006-10-135-115/+57
| | | | | | | | | | | | | add more variables * tapset/LKET/register_event.stp: be sync with scsi trace hooks * tapset/LKET/scsi.stp: avoid reference to struct members in embedded c. Use stap variables as more as possible. break scsi_info into four variables. delete cmd_pid from all logging functions
* scheduler.stp:guanglei2006-10-132-13/+19
| | | | skip __switch_to on x86_64
* 2006-10-12 Martin Hunt <hunt@redhat.com>hunt2006-10-122-7/+53
| | | | | | | | | | | | * context.stp (print_backtrace): Pass in new kretprobe instance arg. (backtrace): Ditto. (is_return): Rewrite. (stack_size): New. (stack_used): New. (stack_unused): New. (called_addr): New. (caller): New.
* * ioscheduler.stp:guanglei2006-10-126-112/+119
| | | | | | | | | | | bugfix to avoid refer to NULL pointer * hookid_defs.stp, ioscheduler.stp, register_event.stp: break the old addevent.ioscheduler.elv_next_request into addevent.ioscheduler.elv_next_request.entry and addevent.ioscheduler.elv_next_request.return. Now log_ioscheduler_tracedata_* will use the variables defined in generic ioscheduler tapset
* bugfix of LKET/tapsets for pointer conversion on 32-bit platformguanglei2006-10-1110-83/+98
| | | | make sys_request_key optional on i386
* stapprobes.signal.5.in:guanglei2006-10-107-628/+883
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | document signal syscall hookiddefs.stp: some changes to the hookid: Now the hookid of a return type event(addevent.*.return) should be an even number and its value should be the corresponding entry event hookid +1. The hookid of a non-return type event should be an odd number. Add the hookid of HOOKID_REGEVTDESC register_event.stp: add register_evt_desc() to register the description of an event. Change the register_sys_event() to add the an argument representing event description Allow to use empty format string and empty field name when calling register_event(). Change all calling to register_sys_event() to add an argument which represents the event description. Change register_evt_desc() and register_event() to add an argument representing the event description string, which could help facilitate the error debugging. iosyscall.stp: bugfix to add the variable GroupID and hookID nfs.stp: bugfix to add some omitted probe definitions. nfs_proc.stp: bugfix of the wrong calling to _lket_trace() for nfs_proc.remove.entry.
* 2006-10-04 Frank Ch. Eigler <fche@elastic.org>fche2006-10-042-3/+10
| | | | | * context.stp (probefunc): Return empty string for userspace (< PAGE_OFFSET) pointers.
* 2006-09-26 David Smith <dsmith@redhat.com>dsmith2006-09-262-3/+7
| | | | * test/run.sh: Changed 'stpd' references to 'staprun'.
* 2006-09-26 David Smith <dsmith@redhat.com>dsmith2006-09-262-1/+5
| | | | * tskdispatch.stp: Changed 'stpd' reference to 'staprun'.