Commit graph

14 commits

Author SHA1 Message Date
Diego Frias
a6ebd100ec AK/LEB128: Speed up reading unsigned LEB128 values
Unroll the first byte as a fast path, and remove a branch. This speeds
up the instantiation of spidermonkey by 10ms.
2024-07-27 08:19:51 +02:00
Diego
596dd5252d AK: Read signed LEB128 integers without 64-bit assumptions
This fixes some errors where too many bytes were allowed to be read for
signed integers of a smaller size (e.g. i32). The new parser doesn't
make 64-bit assumptions and now matches the generality of its unsigned
counterpart.
2024-06-18 16:58:33 +02:00
Tim Schumacher
43f98ac6e1 Everywhere: Remove the AK:: qualifier from Stream usages 2023-02-13 00:50:07 +00:00
Tim Schumacher
47531a42a9 AK: Make LEB128 decoding work with read_value 2023-02-04 18:41:27 -07:00
Tim Schumacher
787f4d639a AK: Port LEB128 to the new AK::Stream 2023-02-04 18:41:27 -07:00
Tim Schumacher
31f59855b4 AK: Don't restore the stream offset when failing to read an LEB128
Pretty much no other read function does this, and getting rid of the
typename template parameter for the stream makes the transition to the
new AK::Stream a bit easier.
2023-02-04 18:41:27 -07:00
Tim Schumacher
ae64b68717 AK: Deprecate the old AK::Stream
This also removes a few cases where the respective header wasn't
actually required to be included.
2023-01-29 19:16:44 -07:00
Andreas Kling
ae3ffdd521 AK: Make it possible to not using AK classes into the global namespace
This patch adds the `USING_AK_GLOBALLY` macro which is enabled by
default, but can be overridden by build flags.

This is a step towards integrating Jakt and AK types.
2022-11-26 15:51:34 +01:00
kleines Filmröllchen
41b2d37e8a AK: Always check shift amount in LEB128 read functions
Even shifting 0 by more than the value size is UB.
2022-07-09 22:04:31 +00:00
Idan Horowitz
086969277e Everywhere: Run clang-format 2022-04-01 21:24:45 +01:00
Ali Mohammad Pur
60d43d6969 AK: Don't perform the shift when it's too large when decoding LEB128
Prior to this, we calculated whether the shift was too large for the
result, and then did the shift regardless.
Found by OSS-Fuzz: https://oss-fuzz.com/testcase-detail/6046441716973568
2021-08-31 16:37:49 +02:00
Andrew Kaster
0af192ff8d AK: Handle LEB128 encoded values that are too large for the result type
Previously, we would go crazy and shift things way out of bounds.
Add tests to verify that the decoding algorithm is safe around the
limits of the result type.
2021-05-31 14:25:27 +04:30
Ali Mohammad Pur
d288f6654e AK: Make LEB128 capable of reading into any type
And not just ssize_t/size_t.
This is useful in cases where the output size is supposed to be larger
than size_t.
2021-05-04 22:33:35 +02:00
Ali Mohammad Pur
48260b5054 AK: Move the LEB128 logic to AK and make it usable with InputStream 2021-05-04 22:33:35 +02:00