Move the HSA intercept to the OnLoad function, so that it is available
as soon as the ROCR is loaded.
Layer the HSA API wrappers on top of the basic HSA activity intercept.
Change-Id: Ie636d59755543cda181e76ec29f0b55081136b63
[ROCm/roctracer commit: e282a82e29]
This commit is for code cleanup and for optimizing kernel name search
in the API callback, making sure to get the kernel name accurately
for the hip functions that have any kernel names
Change-Id: Ie9ab917c895748bfb8eee9ddfcbcad81a0b9a9fa
[ROCm/roctracer commit: 054456bcf8]
When ROCP_TRUNCATE_NAMES is not set, getenv returns NULL and std::atoi
crashes. Check that getenv returns a non-NULL string before calling
std::atoi.
Change-Id: Ie479a481f8d23f034b425d14e3cfefb3d62c84e8
[ROCm/roctracer commit: a7cac4b4c9]
Split the public and private HSA profiler/tracer interfaces. Only the
public interface should be exposed in include/roctracer.
Change-Id: I7e4424cd90023693350c31e6b02caca8c984ba84
[ROCm/roctracer commit: 96bcac0f47]
Added a function to truncate the kernel names in case of hip trace.
Change-Id: Ida769679ccc38432538ce74c11c395d6a930018e
[ROCm/roctracer commit: e3db2beca7]
Adding roctracer_timestamp_t to be unified type for every timestamp in the roctracer project
Change-Id: I22f18c303bd2d02d1e6079b76da6b128eb384fdd
[ROCm/roctracer commit: 581c37502f]
The roctracer-tests package contains all the roctracer test binaries
and scripts needed to run the testsuite outside of the build directory.
Change-Id: Id11f862fb4bdb2425d68f455074172c38814ec92
[ROCm/roctracer commit: 3adc56bd9b]
The ROCR now detects already loaded tool libraries and calls OnLoad/
OnUnload in the order specified with HSA_AMD_TOOL_ORDER.
It is no longer necessary to set the HSA_TOOLS_LIB environment variable
to load the roctracer API. The roctracer tool library should be
pre-loaded with LD_PRELOAD.
Change-Id: I6de1b1bd4f93caa08d3554aad2376d242c74fb7e
[ROCm/roctracer commit: d32cf96cd4]
Enabling the new methodology of ROCP_STATS_OPT of getting HIP activities while the application is running
Change-Id: I94b3311b0740db804643dba0e4f77c1f9de0319b
[ROCm/roctracer commit: 7cbee920b7]
In file included from roctracer/src/roctracer/tracker.h:24,
from roctracer/src/roctracer/roctracer.cpp:44:
/opt/rocm/hsa/include/hsa/amd_hsa_signal.h:26:246: note: ‘#pragma message: amd_hsa_signal.h has moved to ...’
26 | ssage("amd_hsa_signal.h has moved to ...")
| ^
Change-Id: I38d151d836688083a4fdb0e86a04fc40923a369f
[ROCm/roctracer commit: 44dbb56fce]
The same information can be generated from the hcc_ops_trace.txt file,
so in a later commit, will add a stage to the tblextr.py script to
generate the .csv files when ROCP_STATS_OPT=1.
Change-Id: I3d1575e096bedf98c66068d9a4ca141421e5bb9d
[ROCm/roctracer commit: 836bab37d3]
Some records may need to point to data with the same lifetime as the
records themselves. One solution is to store the data at the end of
the memory pool buffer. Records in the buffer grow up, and the data
grows down. When the buffer is flushed both records and data are
recycled.
Change-Id: I278fa84478236bf895f7c2d152d47d4256987392
[ROCm/roctracer commit: ef3b2a8aa8]
Only the roctracer API is using exceptions to return error codes to the
client application.
Change-Id: I92e5bd1a044dbde0c80dd9ef87e606550c3ff790
[ROCm/roctracer commit: 1e8e53da1d]
The CallbackTable utility is shared by multiple APIs, so move it out
of the roctracer directory.
Change-Id: Ie2378dca8a4f90bd334dbe2d39ced37c9e790ed0
[ROCm/roctracer commit: c0cbcdff1d]
The roctracer_load, roctracer_unload, and roctrace_flush_buf functions
are not part of the ROCtracer API, and should not be exposed in the API
header file, but keep the functions in the library for backward
compatibility.
Add src/roctracer/backward_compat.cpp to implement retired functions.
Add test/app/backward_compat_test.cpp to test that the retired functions
are still accessible in the latest roctracer library.
Change-Id: I4c94310a7bfccfeae9384dac5db18fc79b4c5b17
[ROCm/roctracer commit: 8ee9d859ed]
Make error codes more informative and have negative values. This is an
ABI break but it does not appear known tools are relying on the exact
error codes.
Use logging for all errors so that roctracer_error_string will be able
to return last error message.
Make internal errors fatal and abort.
Do not use the tracer API exceptions in the tracer tool.
Change-Id: Ie8ed3d50e5ad26625ac9d1263f7e048edb5584c0
[ROCm/roctracer commit: a4db4084db]
Move the tracer_tool from the 'test' directory to the 'src' directory.
Change-Id: I13768b9610cd359f78a66147f0255ab1e4c657e9
[ROCm/roctracer commit: b34e9c2ee4]
Add symbol versioning to the roctracer64 and roctx64 library, and only
expose the OnLoad and OnUnload tracer_tool symbols.
Change-Id: I7f160fc3e568567fd1146ff5b9c0aef3bdcccf53
[ROCm/roctracer commit: 0377c5a191]
Add custom_commands to generate the HSA code objects
Remove the configure time file generation and add custom commands to
generate them at build time.
Change-Id: I167dd9befc6c73f32224935eaab74510922b26f4
[ROCm/roctracer commit: 3773384af8]
Check if a default pool is defined when enabling activities.
Set default pool to undefined if it is deleted.
Disable activities associated with the pool when it is deleted.
Document restrictions on deleting pools.
Change-Id: Ide466b55cab12ca2dd67d9f26b578f421e45a376
[ROCm/roctracer commit: feb652e45d]
Removing unused definitions and compile options
Using cmake variables to set the options needed
Changing the visibility to make it specific for the targets
Change-Id: I80cf0997cd28897d5a06a58c7225ba40dfc51e2d
[ROCm/roctracer commit: 2f5313a0c7]
Each thread has a thread-local record_pair_stack. The stack is
dynamically allocated on first use, but is not detroyed when the
thread exits.
Replaced record_pair_stack pointers with record_pair_stack instances,
the intances are constructed on first odr-use, and destructed when the
thread exits.
Also, converted the cb_journal and act_journal to instances.
Change-Id: I186ac29da477f194880a1ab599f4be5715a23063
[ROCm/roctracer commit: 67481bd295]
The range message stack is mirrored in case ranges are pushed or popped
while tracing is stopped (by the tracer tool?). When a stop event is
reported, the tracer tool emits RangePop events by unwinding the stack,
then when the start event is reported, it emits RangePush events again
by unwinding the stack. The issue is that the RangePush events should
be emitted in reverse order.
For example:
RangePush(M1); RangePush(M2); \
TracerStop; RangePop; RangePop; \
...; \
TracerStart; RangePush(M2); RangePush(M1); \ <- In the wrong order
RangePop; RangePop;
It could be fixed by reversing the stack in RangeStackIterate but is it
worth it? The roctx range markers are supposed to be unintrusive so that
they can be left in the application even when it isn't being traced.
Simplifying the roctx API and reducing its added latency by removing
the range message stack mirroring seems like the better choise.
TODO: A future change should make roctx events immune to tracer start
and tracer stop requests. Or simply remove roctracer_start/stop.
Change-Id: Ie4d76afb5ce8d263848dcf1b599af394db56ddab
[ROCm/roctracer commit: 3d0198c395]
Remove thread_data_init. The C++ standard guarantees that the thread
local variable is initialized before its first odr-use and destructed
when the thread exits. Use a global initializer to set the reference
from the message stack instance in the map.
Remove roctracer_error_string. This does not belong to this library.
ROCTX does not expose errors to the application. The only functions
returning errors are returning -1 (Push/Pop).
Remove memory leaks due to strdup on the ranges messages. The memory
for the messages is guaranteed to be valid for the duration of the
callback, and it is the application's responsibility to strdup the
strings if it needs to extend the message's lifetime.
Add a lock to the RegisterApiCallback implementation. Iterating the
message stack map must be synchronized as a new thread could be adding
a new value to the map.
Change-Id: Iaf5b07ebc9efe4061cb01327d4c7034888727816
[ROCm/roctracer commit: 713db1fce5]
Make CallbackTable::Get return the callback_function/user_arg pair
as an actual return value instead of returning it through arguments
pointers.
Change-Id: Ia2dfcdad8c237a09620518ad67af94add47220da
[ROCm/roctracer commit: 4aeb76f7a8]
At the end of the test, the tracer tool is unloaded and the active
memory pools are flushed. In the flush callback, to get the activity
operation string, the RocpLoader instance is neeeded, and if the
RocpLoader is not already loaded, it attempts to dlopen the rocprofiler
library.
Calling dlopen from a global destructor hangs because the dynamic
loader lock is already owned (e.g. by dlclose).
To temporarily work around the issue, instanciate the RocpLoader when
the activities needing it are enabled.
Change-Id: I712c66d88c43694fe53a95d6a61d7b22abb75262
[ROCm/roctracer commit: 11887f596a]
System clock timestamps should only come from a single source:
util::timestamp_ns(). Externally, this function is exposed as
roctracer_get_timestamp() (used by the tracer tool).
Removed the now unused HSA Runtime Utilities which were never part
of the ROCtracer API.
Change-Id: I044b7f4da60fd8fdb771b0c877622a3143f0e815
[ROCm/roctracer commit: f8462b8637]
Laurent Morichetti
Ammar ELWazir
Sriraksha Nagaraj
Ranjith Ramakrishnan
Tony Tye