On Tue, Apr 06, 2021 at 11:52:32AM +0300, Bayduraev, Alexey V wrote:
>
> Introduce decompressor into trace reader object so that decompression
> could be executed on per data file basis separately for every data
> file located in data directory.
>
> Load data directory files and provide basic raw dump and aggregated
> analysis support of data directories in report mode, still with no
> memory consumption optimizations.
>
> Design and implementation are based on the prototype [1], [2].
>
> [1] git clone https://git.kernel.org/pub/scm/linux/kernel/git/jolsa/perf.git
> -b perf/record_threads
> [2] https://lore.kernel.org/lkml/20180913125450.21342-1-jo...@kernel.org/
>
> Suggested-by: Jiri Olsa <jo...@kernel.org>
> Signed-off-by: Alexey Bayduraev <alexey.v.baydur...@linux.intel.com>
hi,
this and following patch seem to squash several changes from
perf/record_threads:
21289ea28f2f perf session: Add reader return codes
44d19279d0e9 perf session: Add eof flag to reader state
490cece1ff49 perf session: Add reader__read_event function
ff9d91751367 perf session: Move head/file_offset computation into
reader__mmap function
75a6af52ecae perf session: Move unmap into reader__mmap function
e1fc76225518 perf session: Add reader__mmap function
f7f0473104cd perf session: Add reader__init function
262c08cb7332 perf session: Move head in reader_state
fd91847190f0 perf session: Move data_size in reader_state
dcb49769bdde perf session: Move file_offset in reader_state
69c7f9557e83 perf session: Move file_pos in reader_state
c5c5c03f09f9 perf session: Add mmap_cur to reader_state
faf7b7176f2a perf session: Move mmap_idx in reader_state
42504d97a18f perf session: Move mmap_size in reader_state
7bf7ccdabc84 perf session: Move mmaps in reader_state
fc1eb45de565 perf session: Add path to reader object
making this a really big change of important code, which can't
be easily reviewed.. that's why I split it in the first place ;-)
I think we need to bring this code first with incremental changes
jirka
