Open
Conversation
Added time reuse distance and cacheline reuse distance
… event will not generate any overflow
…t64_t *val) to read the counter value
… context PC to construct the calling context
Conflicts: src/tool/hpcrun/sample-sources/perf/linux_perf.c src/tool/hpcrun/sample-sources/watchpoint_support.c
…oinNode + first_use_point [leaf]
Re-added the flag PERF_EVENT_IOC_UPDATE_BREAKPOINT for older kernel
When sampling the load (threshold_above_latency), do not record the sample if it is an L1 hit
…. Otherwise, it may make the value of some event appear smaller.
…ance from "estimation"
Conflicts: src/tool/hpcrun/sample-sources/watchpoint_clients.c
2). add online attribution for memory reuse distance 3). fix a bug about sampling policy setting
1). Added cache miss metric from MEM_TRANS_RETIRED:THRESHOLD_ABOVE_LATENCY 2). Added online reuse distance estimation due to counter multiplexing 3). Disable the dummy counter from watchpoint 4). Start to implement that the number of memeroy accesses is read from overflow counters
Implemented the function linux_perf_read_event_counter()
How to concatenate the use and reuse pair can be determined by the environmental variable HPCRUN_WP_REUSE_CONCATENATE_ORDER
Conflicts: src/tool/hpcrun/sample-sources/perf/linux_perf.c src/tool/hpcrun/sample-sources/perf/perf-util.c src/tool/hpcrun/sample-sources/watchpoint_support.c
Conflicts: src/tool/hpcrun/sample-sources/perf/linux_perf.c src/tool/hpcrun/sample-sources/perf/perf-util.c
Conflicts: src/tool/hpcrun/sample-sources/perf/perf-util.c
Conflicts: src/tool/hpcrun/sample-sources/watchpoint_support.c
Conflicts: src/tool/hpcrun/sample-sources/perf/linux_perf.c src/tool/hpcrun/utilities/tokenize.c
Conflicts: src/tool/hpcrun/sample-sources/perf/linux_perf.c src/tool/hpcrun/sample-sources/perf/perf-util.c src/tool/hpcrun/utilities/tokenize.c
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
When compiling, adding the flag
-DREUSE_HISTOmakes the reuse client into histogram mode; otherwise, it is just profiling mode.When in histogram mode, use the environmental variable HPCRUN_WP_REUSE_BIN_SCHEME to tune the bin.
If it set to TRACE or not set, it will trace all the reuse pair and context.
Otherwise, you can set it to two numbers which determines the bin start and the ratio. For example, 4000,2 indicates the first two bins are [0, 4000) and [4000, 8000)