pcre2/backport-Fixed-a-bug-in-character-set-matching-when-JIT-is-en.patch
yangmingtaip cf22193e63 sync community patchs
(cherry picked from commit 157b747b849cd84364a10b496c4338d7faa7f49a)
2022-10-18 21:00:07 +08:00

93 lines
3.1 KiB
Diff

From 3bdc76e4f35dc2f61ac788d829164367cc63fdb6 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Zolt=C3=A1n=20Herczeg?= <hzmester@freemail.hu>
Date: Sat, 19 Sep 2020 03:49:32 +0000
Subject: [PATCH] Fixed a bug in character set matching when JIT is enabled.
Conflict:adapt context and delete Changelog
Reference:https://github.com/PhilipHazel/pcre2/commit/3bdc76e4f35dc2f61ac788d829164367cc63fdb6
---
src/pcre2_jit_compile.c | 36 ++++++++++++++++++++++++++++--------
src/pcre2_jit_test.c | 1 +
2 files changed, 29 insertions(+), 8 deletions(-)
diff --git a/src/pcre2_jit_compile.c b/src/pcre2_jit_compile.c
index fae0f7a..a65eafe 100644
--- a/src/pcre2_jit_compile.c
+++ b/src/pcre2_jit_compile.c
@@ -7670,25 +7670,43 @@ if (needstype || needsscript)
}
cc = ccbegin;
- }
- if (needschar)
- OP1(SLJIT_MOV, TMP1, 0, RETURN_ADDR, 0);
+ if (needstype)
+ {
+ /* TMP2 has already been shifted by 2 */
+ if (!needschar)
+ {
+ OP2(SLJIT_ADD, TMP1, 0, TMP2, 0, TMP2, 0);
+ OP2(SLJIT_ADD, TMP1, 0, TMP1, 0, TMP2, 0);
+
+ OP1(SLJIT_MOV_U8, TMP1, 0, SLJIT_MEM1(TMP1), (sljit_sw)PRIV(ucd_records) + SLJIT_OFFSETOF(ucd_record, chartype));
+ }
+ else
+ {
+ OP2(SLJIT_ADD, TMP1, 0, TMP2, 0, TMP2, 0);
+ OP2(SLJIT_ADD, TMP2, 0, TMP2, 0, TMP1, 0);
- if (needstype)
+ OP1(SLJIT_MOV, TMP1, 0, RETURN_ADDR, 0);
+ OP1(SLJIT_MOV_U8, RETURN_ADDR, 0, SLJIT_MEM1(TMP2), (sljit_sw)PRIV(ucd_records) + SLJIT_OFFSETOF(ucd_record, chartype));
+ typereg = RETURN_ADDR;
+ }
+ }
+ else if (needschar)
+ OP1(SLJIT_MOV, TMP1, 0, RETURN_ADDR, 0);
+ }
+ else if (needstype)
{
+ OP2(SLJIT_SHL, TMP1, 0, TMP2, 0, SLJIT_IMM, 3);
+ OP2(SLJIT_SHL, TMP2, 0, TMP2, 0, SLJIT_IMM, 2);
+
if (!needschar)
{
- OP2(SLJIT_SHL, TMP1, 0, TMP2, 0, SLJIT_IMM, 3);
- OP2(SLJIT_SHL, TMP2, 0, TMP2, 0, SLJIT_IMM, 2);
OP2(SLJIT_ADD, TMP1, 0, TMP1, 0, TMP2, 0);
OP1(SLJIT_MOV_U8, TMP1, 0, SLJIT_MEM1(TMP1), (sljit_sw)PRIV(ucd_records) + SLJIT_OFFSETOF(ucd_record, chartype));
}
else
{
- OP2(SLJIT_SHL, TMP1, 0, TMP2, 0, SLJIT_IMM, 2);
- OP2(SLJIT_SHL, TMP2, 0, TMP2, 0, SLJIT_IMM, 3);
OP2(SLJIT_ADD, TMP2, 0, TMP2, 0, TMP1, 0);
OP1(SLJIT_MOV, TMP1, 0, RETURN_ADDR, 0);
@@ -7696,6 +7714,8 @@ if (needstype || needsscript)
typereg = RETURN_ADDR;
}
}
+ else if (needschar)
+ OP1(SLJIT_MOV, TMP1, 0, RETURN_ADDR, 0);
}
#endif /* SUPPORT_UNICODE */
diff --git a/src/pcre2_jit_test.c b/src/pcre2_jit_test.c
index 2f49867..40a20a1 100644
--- a/src/pcre2_jit_test.c
+++ b/src/pcre2_jit_test.c
@@ -410,6 +410,7 @@ static struct regression_test_case regression_test_cases[] = {
{ MUP, A, 0, 0 | F_PROPERTY, "[\\P{L&}]{2}[^\xc2\x85-\xc2\x89\\p{Ll}\\p{Lu}]{2}", "\xc3\xa9\xe6\x92\xad.a\xe6\x92\xad|\xc2\x8a#" },
{ PCRE2_UCP, 0, 0, 0 | F_PROPERTY, "[a-b\\s]{2,5}[^a]", "AB baaa" },
{ CMUP, 0, 0, 0, "[^S]\\B", "\xe2\x80\x8a" },
+ { MUP, 0, 0, 0 | F_NOMATCH, "[^\\p{Hangul}\\p{Z}]", " " },
/* Possible empty brackets. */
{ MU, A, 0, 0, "(?:|ab||bc|a)+d", "abcxabcabd" },
--
2.27.0