I'm pleased to announce the new version of libtracefs library has been
released:
libtracefs: 1.1.0
https://git.kernel.org/pub/scm/libs/libtrace/libtracefs.git/snapshot/libtracefs-1.1.0.tar.gz
https://git.kernel.org/pub/scm/libs/libtrace/libtracefs.git/
Changes since 1.0.2:
- Depends on libtraceevent 1.2.0 or later
- New APIs for opening and reading ftrace files
tracefs_instance_file_read_number()
tracefs_instance_file_open()
- New APIs for enable / disable tracing
tracefs_trace_is_on()
tracefs_trace_on()
tracefs_trace_off()
tracefs_trace_on_get_fd()
tracefs_trace_on_fd()
tracefs_trace_off_fd()
- New APIs for trace options
tracefs_option_set()
tracefs_option_clear()
tracefs_option_is_set()
tracefs_options_get_supported()
tracefs_option_is_supported()
tracefs_options_get_enabled()
tracefs_option_is_enabled()
tracefs_option_enable()
tracefs_option_diasble()
tracefs_option_name()
tracefs_option_id()
tarcefs_option_mask_is_set()
- New APIs for getting existing trace instance
tracefs_instance_alloc()
tracefs_instance_get_trace_dir()
- New APIs to set the filtering of functions
tracefs_function_filter()
tracefs_function_notrace()
- New APIs to writing strings into the tracing buffer
tracefs_print_init()
tracefs_printf()
tracefs_vprintf()
tracefs_print_close()
- New APIs to writing binary data into the tracing buffer
tracefs_binary_init()
tracefs_binary_write()
tracefs_binary_close()
- tracefs_iterate_raw_events() now calls the callbacks in
time stamp sorted order (instead of calling all events in
a CPU buffer, then the next CPU buffer). Also fixed record->cpu
to match the CPU that's also passed to the callback, instead
of always being zero.
- Loading of events with tracefs_local_events(),
tracefs_local_event_system() and tracefs_fill_local_events()
now pull in kallsyms, saved_cmdlines, and printk_format mappings
into the tep handler.
- New API to reload the saved_cmdlines
tracefs_load_cmdlines()
- Warnings now can be overwritten with tracefs_warning() and no
longer the too generic name warning()
- All files opened in the library now are done so with O_CLOEXEC
- Cleaner building and installation.
-- Steve
I am pleased to announce the new version of KernelShark has been released:
KernelShark 2.0
https://git.kernel.org/pub/scm/utils/trace-cmd/kernel-shark.git/snapshot/kernel-shark-kernelshark-v2.0.tar.gz
https://git.kernel.org/pub/scm/utils/trace-cmd/kernel-shark.git/
Major changes since 1.3:
- Introduces Data streams:
With the help of Data stream, KernelShark is able to load and
merge multiple trace files (streams). Each stream can have
different plugins or filters, registered for it, which means that
the raw trace data of the streams can have different formats, and
will allow for a great degree of customization of the provided
data visualization.
- New design of the plugin interface - instead of having a single
interface for loading the plugin, we now have 3 different
interfaces:
- The one that exists in version 1 of KernelShark is now renamed
to Data Processing Interface (dpi).
- The first new interface for loading can be used to register
user provided implementation of the Data stream readout and is
called Data Readout Interface (dri). Via this plugin loading
interface the user can open trace data having an arbitrary
format. In order to make this possible the user has to provide
a plugin that contains an implementation of the data readout
methods defined by the Data stream and to register all those
methods.
- The second new plugin loading interface is called Control
interface and can be used to provide the plugin with an access
to the GUI's Main window object. Via this interface the plugin
can became capable to modify the GUI.
- The modification of the C API are used to implement 3 new
plugins:
- KVMCombo plugin: The plugin allows the user to visualize
the execution flow between the host and guest virtual machines.
It exploits the concepts of "Data streams " and "Combo Plots"
That allows to have two normal graphs from two data streams
stacked together (on top of each other). The plugin uses a
"combo" between the task in the host that emulates a virtual
CPU and the corresponding CPU graph from the VM. The plugin
draws additional graphical elements on top of this "combo",
helping the user to intuitively interpret the data and see how
the execution flow goes from host to guest and back.
- LatencyPlot plugin: The plugin allows the user to visualize
the latency between two events under the condition that the
values of given data fields in the two events are identical
(for example having the same PID). The plugin also registers
its own dialog, that allows the user to select the events
(and the matching field) to be visualized.
- EventFieldPlot plugin: The plugin allows the user to visualize
the recorded value of a given data field from a given trace
event. The user can select the event and field to be visualized.
Yordan