Yann Collet [Wed, 25 Jan 2017 20:31:07 +0000 (12:31 -0800)]
fixed zstdmt cli freeze issue with large nb of threads
fileio.c was continually pushing more content without giving a chance to flush compressed one.
It would block the job queue when input data was accumulated too fast (requiring to define many threads).
Fixed : fileio flushes whatever it can after each input attempt.
Yann Collet [Wed, 25 Jan 2017 01:02:26 +0000 (17:02 -0800)]
zstdmt cli and API allow selection of section sizes
By default, section sizes are 4x window size.
This new setting allow manual selection of section sizes.
The larger they are, the (slightly) better the compression ratio,
but also the higher the memory allocation cost,
and eventually the lesser the nb of possible threads,
since each section is compressed by a single thread.
It also introduces a prototype to set generic parameters,
ZSTDMT_setMTCtxParameter()
The idea is that it's possible to add enums
to extend the list of parameters that can be set this way.
This is more long-term oriented than a fixed-size struct.
Consider it as a test.
Yann Collet [Mon, 23 Jan 2017 07:49:52 +0000 (23:49 -0800)]
ZSTDMT_compressStream() becomes blocking when required to ensure forward progresses
In some (rare) cases, job list could be blocked by a first job still being processed,
while all following ones are completed, waiting to be flushed.
In such case, the current job-table implementation is unable to accept new job.
As a consequence, a call to ZSTDMT_compressStream() can be useless (nothing read, nothing flushed),
with the risk to trigger a busy-wait on the caller side
(needlessly loop over ZSTDMT_compressStream() ).
In such a case, ZSTDMT_compressStream() will block until the first job is completed and ready to flush.
It ensures some forward progress by guaranteeing it will flush at least a part of the completed job.
Energy-wasting busy-wait is avoided.
Yann Collet [Mon, 23 Jan 2017 00:40:06 +0000 (16:40 -0800)]
ZSTDMT_initCStream_usingDict() can outlive dict
Like ZSTD_initCStream_usingDict(),
ZSTDMT_initCStream_usingDict() now keep a copy of dict internally.
This way, dict can be released :
it does not longer have to outlive all future compression sessions.
Yann Collet [Thu, 19 Jan 2017 23:32:07 +0000 (15:32 -0800)]
Added ZSTDMT_initCStream_advanced() variant
Correctly compress with custom params and dictionary
Added relevant fuzzer test in zstreamtest
Also :
new macro ZSTDMT_SECTION_LOGSIZE_MIN, which sets a minimum size for a full job
(note : a flush() command can still generate a partial job anytime)
Yann Collet [Thu, 19 Jan 2017 20:12:50 +0000 (12:12 -0800)]
added streaming fuzzer tests for MT API
Also : fixed corner case, where nb of jobs completed becomes > jobQueueSize
which is possible when many flushes are issued
while there is not enough dst buffer to flush completed ones.
Yann Collet [Thu, 19 Jan 2017 18:32:55 +0000 (10:32 -0800)]
fixed Multi-threaded compression
MT compression generates a single frame.
Multi-threading operates by breaking the frames into independent sections.
But from a decoder perspective, there is no difference :
it's just a suite of blocks.
Problem is, decoder preserves repCodes from previous block to start decoding next block.
This is also valid between sections, since they are no different than changing block.
Previous version would incorrectly initialize repcodes to their default value at the beginning of each section.
When using them, there was a mismatch between encoder (default values) and decoder (values from previous block).
This change ensures that repcodes won't be used at the beginning of a new section.
It works by setting them to 0.
This only works with regular (single segment) variants : extDict variants will fail !
Fortunately, sections beyond the 1st one belong to this category.
To be checked : btopt strategy.
This change was only validated from fast to btlazy2 strategies.
Yann Collet [Wed, 18 Jan 2017 01:46:33 +0000 (17:46 -0800)]
ZSTDMT_free() scrubs potentially unfinished jobs to release their resources
In some complex scenarios (free() without finishing compression),
it is possible that some resources are still into jobs
and not collected back into pools.
In which case, previous version of free() would miss them.
This would be equivalent to a leak.
New version ensures that it even foes after such resource.
It requires job consumers to properly mark resources as released,
by replacing entries by NULL after releasing back to the pool.
Obviously, it's not recommended to free() zstdmt context mid-term,
still that's now a supported scenario.
The same methodology is also used to ensure proper resource collection
after an error is detected.
Still to do :
- detect compression errors (not just allocation ones)
- properly manage resource when init() is called without finishing previous compression.
Yann Collet [Wed, 18 Jan 2017 00:15:18 +0000 (16:15 -0800)]
ZSTDMT_{flush,end}Stream() now block on next job completion when nothing to flush
The main issue was to avoid a caller to continually loop on {flush,end}Stream()
when there was nothing ready to be flushed but still some compression work ongoing in a worker thread.
The continuous loop would have resulted in wasted energy.
The new version makes call to {flush,end}Stream blocking when there is nothing ready to be flushed.
Of course, if all worker threads have exhausted job, it will return zero (all flush completed).
Note : There are still some remaining issues to report error codes
and properly collect back resources into pools when an error is triggered.
Gregory Szorc [Sun, 15 Jan 2017 01:44:54 +0000 (17:44 -0800)]
Set dictionary ID in ZSTD_initCStream_usingCDict()
When porting python-zstandard to use ZSTD_initCStream_usingCDict()
so compression dictionaries could be reused, an automated test
failed due to compressed content changing.
I tracked this down to ZSTD_initCStream_usingCDict() not
setting the dictID field of the ZSTD_CCtx attached to the
ZSTD_CStream instance.
I'm not 100% convinced this is the correct or full solution,
as I'm still seeing one automated test failing with this change.
Yann Collet [Thu, 12 Jan 2017 02:06:35 +0000 (03:06 +0100)]
zstdmt : fix : resources properly collected even when early fail
In previous version, main function would return early when detecting a job error.
Late threads resources were therefore not collected back into pools.
New version just register the error, but continue the collecting process.
All buffers and context should be released back to pool before leaving main function.
Yann Collet [Thu, 12 Jan 2017 01:01:28 +0000 (02:01 +0100)]
zstdmt : correctly check for cctx and buffer allocation
Result from getBuffer and getCCtx could be NULL when allocation fails.
Now correctly checks : job creation stop and last job reports an allocation error.
releaseBuffer and releaseCCtx are now also compatible with NULL input.
Identified a new potential issue :
when early job fails, later jobs are not collected for resource retrieval.
Yann Collet [Wed, 11 Jan 2017 17:21:25 +0000 (18:21 +0100)]
ZSTDMT_compress() creates a single frame
The new strategy involves cutting frame at block level.
The result is a single frame, preserving ZSTD_getDecompressedSize()
As a consequence, bench can now make a full round-trip,
since the result is compatible with ZSTD_decompress().
This strategy will not make it possible to decode the frame with multiple threads
since the exact cut between independent blocks is not known.
MT decoding needs further discussions.
Yann Collet [Wed, 11 Jan 2017 14:44:26 +0000 (15:44 +0100)]
improved ZSTD_createCCtxPool() cancellation
use ZSTD_freeCCtxPool() to release the partially created pool.
avoids to duplicate logic.
Also : identified a new difficult corner case :
when freeing the Pool, all CCtx should be previously released back to the pool.
Otherwise, it means some CCtx are still in use.
There is currently no clear policy on what to do in such a case.
Note : it's supposed to never happen.
Since pool creation/usage is static, it has no external user,
which limits risks.