ARM-specific optimisations for inflate.#256
Closed
Conversation
added 2 commits
April 26, 2017 17:19
Change-Id: Id4cda552b39bfb39ab35ec499dbe122b43b6d1a1
In inflate_fast() the output pointer always has plenty of room to write. This means that so long as the target is capable, wide un-aligned loads and stores can be used to transfer several bytes at once. When the reference distance is too short simply unroll the data a little to increase the distance. Change-Id: I59854eb25d2b1e43561c8a2afaf9175bf10cf674
Author
|
@ProgramMax, more of the PNG optimisation here, FYI. Corresponding Chromium patch (now with green bots!) is here. |
|
Thank you for pinging me. :) |
nigeltao
reviewed
Sep 22, 2017
| exploited. | ||
| */ | ||
| static inline unsigned char FAR *chunkcopy_safe(unsigned char FAR *out, | ||
| const unsigned char FAR * Z_RESTRICT from, |
There was a problem hiding this comment.
I'm not sure if you can use Z_RESTRICT here. Maybe that's true if you came in via inflate.c, but maybe not if you came in via infback.c.
There's a longer discussion of that at https://chromium-review.googlesource.com/c/chromium/src/+/641575/4/third_party/zlib/contrib/arm/chunkcopy.h#230
Author
There was a problem hiding this comment.
My inclination is to give infback.c and inflate.c different implementations; but it could still be argued that the assumption is too dangerous for something, somewhere out there, written in the last twenty-something years.
Author
|
This is past-life work, now, and I'm not sure how I'm supposed to reconcile that now that I need to fix it. So I won't. |
|
I can fix it and add it to the Adler-32 + CRC32 merge request in: #251 |
jow-
pushed a commit
to lede-project/source
that referenced
this pull request
Jan 2, 2018
This adds two optimizations for ARM: NEON optimized Adler(-)32 checksum algorithm (ARMv7 and newer NEON CPUs) ARM(v7+) specific optimization for inflate I've also connected inflate optimization to the build using the following source as template. mirror/chromium@0397489#diff-a62ad2db6c83dbc205d34bb9a8884f16 Additional info: https://codereview.chromium.org/2676493007/ https://codereview.chromium.org/2722063002/ Sources: madler/zlib#251 (only the first commit) madler/zlib#256 Signed-off-by: Daniel Engberg <daniel.engberg.lists@pyret.net>
SpiralP
pushed a commit
to SpiralP/lede-source
that referenced
this pull request
Jan 2, 2018
This adds two optimizations for ARM: NEON optimized Adler(-)32 checksum algorithm (ARMv7 and newer NEON CPUs) ARM(v7+) specific optimization for inflate I've also connected inflate optimization to the build using the following source as template. mirror/chromium@0397489#diff-a62ad2db6c83dbc205d34bb9a8884f16 Additional info: https://codereview.chromium.org/2676493007/ https://codereview.chromium.org/2722063002/ Sources: madler/zlib#251 (only the first commit) madler/zlib#256 Signed-off-by: Daniel Engberg <daniel.engberg.lists@pyret.net>
jollaman999
pushed a commit
to jollaman999/openwrt
that referenced
this pull request
Jan 13, 2018
This adds two optimizations for ARM: NEON optimized Adler(-)32 checksum algorithm (ARMv7 and newer NEON CPUs) ARM(v7+) specific optimization for inflate I've also connected inflate optimization to the build using the following source as template. mirror/chromium@0397489#diff-a62ad2db6c83dbc205d34bb9a8884f16 Additional info: https://codereview.chromium.org/2676493007/ https://codereview.chromium.org/2722063002/ Sources: madler/zlib#251 (only the first commit) madler/zlib#256 Signed-off-by: Daniel Engberg <daniel.engberg.lists@pyret.net>
This pull request was closed.
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
3 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.
In
inflate_fast()the output pointer always has plenty of room to write. This means that so long as the target is capable, wide un-aligned loads and stores can be used to transfer several bytes at once. When the reference distance is too short simply unroll the data a little to increase the distance.For PNG decode this comes out at about 33% faster overall across a wide set of files. Small PNGs tend to benefit the least because they don't ever enter into
inflate_fast()where the most straightforward assumptions can be made.