LibGfx: Decode Mellor_ACTITC_10.pdf faster [3/N]#26250
Merged
nico merged 2 commits intoSerenityOS:masterfrom Oct 7, 2025
Merged
LibGfx: Decode Mellor_ACTITC_10.pdf faster [3/N]#26250nico merged 2 commits intoSerenityOS:masterfrom
nico merged 2 commits intoSerenityOS:masterfrom
Conversation
LucasChollet
commented
Oct 6, 2025
| if (!inputs.uses_huffman_encoding || inputs.uses_refinement_or_aggregate_coding) { | ||
| auto bitmap = TRY(read_symbol_bitmap(symbol_width, height_class_height)); | ||
| new_symbols.append(Symbol::create(move(bitmap))); | ||
| new_symbols.append(Symbol::create(bitmap->subbitmap(IntRect(0, 0, bitmap->width(), bitmap->height())))); |
Member
Author
There was a problem hiding this comment.
As the two types are not compatible, it forces a few of these conversions.
These are the only non-mechanical changes in the decoder.
Contributor
There was a problem hiding this comment.
Maybe BilevelImage should have a rect() method, similar to Gfx::Bitmap. That'd make this here slightly less awkward.
(No action needed.)
Member
Author
There was a problem hiding this comment.
Or a as_subbitmap method?
8abcad1 to
8277d84
Compare
nico
approved these changes
Oct 6, 2025
Contributor
nico
left a comment
There was a problem hiding this comment.
Very nice!
The commit message grammar-o is the only real comment.
For non-CCITT jbig2s, the additional rect bounds are a teeny tiny slowdown, but it's probably in the measurement noise. (261.0 ms ± 3.7 ms -> 258.9 ms ± 2.8 ms for big_image.jpg converted to jbig2 and then hyperfine 'Build/lagom/bin/image --no-output big_image.jbig2')
| if (!inputs.uses_huffman_encoding || inputs.uses_refinement_or_aggregate_coding) { | ||
| auto bitmap = TRY(read_symbol_bitmap(symbol_width, height_class_height)); | ||
| new_symbols.append(Symbol::create(move(bitmap))); | ||
| new_symbols.append(Symbol::create(bitmap->subbitmap(IntRect(0, 0, bitmap->width(), bitmap->height())))); |
Contributor
There was a problem hiding this comment.
Maybe BilevelImage should have a rect() method, similar to Gfx::Bitmap. That'd make this here slightly less awkward.
(No action needed.)
This type stores a pointer to an upstream `BilevelImage` and a rectangle, corresponding to the dimensions of the sub-bitmap. It is introduced to avoid the copies that were made in `BilevelImage::subbitmap`. When using this test case: ``` BuildLagom/bin/pdf --render out.png PDF/Mellor_ACTITC_10.pdf --page 4 ``` The number of calls to `subbitmap` is ~25000, and none of the resulting sub-bitmaps are modified. So making shallow copies gives a nice performance improvement: according to valgrind, this makes the time spent in `BilevelImage::subbitmap` drop from 26% to 0.03%.
8277d84 to
f26ef2a
Compare
nico
added a commit
to nico/serenity
that referenced
this pull request
Oct 7, 2025
Symbol was a RefCounted wrapper around BilevelImage. As of SerenityOS#26250, BilevelImage is refcounted itself, and it moved the loader to BilevelSubImage which can cheaply be copied. Just use that directly. No behavior change.
nico
added a commit
to nico/serenity
that referenced
this pull request
Oct 7, 2025
Symbol was a RefCounted wrapper around BilevelImage. As of SerenityOS#26250, BilevelImage is refcounted itself, and it moved the loader to BilevelSubImage which can cheaply be copied. Just use that directly. No behavior change.
nico
added a commit
to nico/serenity
that referenced
this pull request
Oct 7, 2025
Symbol was a RefCounted wrapper around BilevelImage. As of SerenityOS#26250, BilevelImage is refcounted itself, and it moved the loader to BilevelSubImage which can cheaply be copied. Just use that directly. No behavior change.
LucasChollet
pushed a commit
to LucasChollet/serenity
that referenced
this pull request
Oct 8, 2025
Symbol was a RefCounted wrapper around BilevelImage. As of SerenityOS#26250, BilevelImage is refcounted itself, and it moved the loader to BilevelSubImage which can cheaply be copied. Just use that directly. No behavior change.
nico
added a commit
that referenced
this pull request
Oct 8, 2025
Symbol was a RefCounted wrapper around BilevelImage. As of #26250, BilevelImage is refcounted itself, and it moved the loader to BilevelSubImage which can cheaply be copied. Just use that directly. No behavior change.
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.
Related to #26082
Before:
After: