From: elasota <1137273+elasota@users.noreply.github.com>
Date: Sun, 19 Nov 2023 20:33:37 +0000 (-0500)
Subject: Specify offset 0 as invalid
X-Git-Tag: v1.5.6^2~50^2
X-Git-Url: http://git.ipfire.org/cgi-bin/gitweb.cgi?a=commitdiff_plain;h=f06b18b3ff009ef7dc90294fca674658ddf139bf;p=thirdparty%2Fzstd.git

Specify offset 0 as invalid
---

diff --git a/doc/decompressor_accepted_invalid_data.md b/doc/decompressor_accepted_invalid_data.md
new file mode 100644
index 000000000..f08f963d9
--- /dev/null
+++ b/doc/decompressor_accepted_invalid_data.md
@@ -0,0 +1,14 @@
+Decompressor Accepted Invalid Data
+==================================
+
+This document describes the behavior of the reference decompressor in cases
+where it accepts an invalid frame instead of reporting an error.
+
+Zero offsets converted to 1
+---------------------------
+If a sequence is decoded with `literals_length = 0` and `offset_value = 3`
+while `Repeated_Offset_1 = 1`, the computed offset will be `0`, which is
+invalid.
+
+The reference decompressor will process this case as if the computed
+offset was `1`, including inserting `1` into the repeated offset list.
\ No newline at end of file
diff --git a/doc/zstd_compression_format.md b/doc/zstd_compression_format.md
index 4a8d338b7..a8d4a0f35 100644
--- a/doc/zstd_compression_format.md
+++ b/doc/zstd_compression_format.md
@@ -929,7 +929,10 @@ There is an exception though, when current sequence's `literals_length = 0`.
 In this case, repeated offsets are shifted by one,
 so an `offset_value` of 1 means `Repeated_Offset2`,
 an `offset_value` of 2 means `Repeated_Offset3`,
-and an `offset_value` of 3 means `Repeated_Offset1 - 1_byte`.
+and an `offset_value` of 3 means `Repeated_Offset1 - 1`.
+
+In the final case, if `Repeated_Offset1 - 1` evaluates to 0, then the
+data is considered corrupted.
 
 For the first block, the starting offset history is populated with following values :
 `Repeated_Offset1`=1, `Repeated_Offset2`=4, `Repeated_Offset3`=8,