Commit 4193f952 for xz

commit 4193f95253d04af7e43b2c3774a521cf1cba49aa
Author: Lasse Collin <lasse.collin@tukaani.org>
Date:   Tue Nov 17 19:23:30 2026 +0200

    liblzma: Fix lzma_filter_flags_decode() docs about in_pos

    In this case it's better to change the docs than change the code. While
    the function is exported, it's primarily used internally in liblzma
    in lzma_block_header_decode(). There it doesn't matter if *in_pos is
    updated on error.

    Reported-by: GitHub user katie-andtruth

diff --git a/src/liblzma/api/lzma/filter.h b/src/liblzma/api/lzma/filter.h
index e86809c4..58916040 100644
--- a/src/liblzma/api/lzma/filter.h
+++ b/src/liblzma/api/lzma/filter.h
@@ -491,7 +491,9 @@ extern LZMA_API(lzma_ret) lzma_filter_flags_encode(const lzma_filter *filter,
  *                          Set to NULL to use malloc() and free().
  * \param       in          Beginning of the input buffer
  * \param[out]  in_pos      The next byte will be read from in[*in_pos].
- *                          *in_pos is updated only if decoding succeeds.
+ *                          *in_pos may be updated even if an error occurs. In
+ *                          liblzma 5.8.4 and older, the docs incorrectly said
+ *                          that *in_pos is updated only if decoding succeeds.
  * \param       in_size     Size of the input buffer; the first byte that
  *                          won't be read is in[in_size].
  *