From: "Yoann Congal" <yoann.congal@smile.fr>
To: "Darsh Kelaiya -X (dkelaiya - E INFOCHIPS PRIVATE LIMITED at
Cisco)" <dkelaiya@cisco.com>,
<openembedded-core@lists.openembedded.org>
Subject: Re: [OE-core][wrynose][PATCH v2 3/5] curl: fix CVE-2026-80255
Date: Tue, 17 Nov 2026 10:54:08 +0100 [thread overview]
Message-ID: <DMXE85B25SK0.3VK8FVR2RZ6AI@smile.fr> (raw)
In-Reply-To: <20261023093331.1886272-4-dkelaiya@cisco.com>
On Fri Oct 23, 2026 at 11:33 AM CEST, Darsh Kelaiya -X (dkelaiya - E INFOCHIPS PRIVATE LIMITED at Cisco) wrote:
> From: Darsh Kelaiya <dkelaiya@cisco.com>
>
> This patch applies the upstream fix as referenced in [2],
> using the commit shown in [1].
>
> Apply prerequisite [3] to reject TABs in attribute values, then
> prerequisite [4] for per-attribute name/value validation. Carry them
> as separate dependent_p1 and dependent_p2 patches before the main
> TAB-handling fix. Preserve test 1685 and the invalid_octets()
> adaptation required by the older parser.
>
> Keep test 2885's Secure<TAB>HttpOnly regression case alongside the
> upstream TAB-before-Secure case and normal Secure control.
>
> [1] https://github.com/curl/curl/commit/4f6aa41a0145e930e766775dbe860883d350aa0a
> [2] https://curl.se/docs/CVE-2026-80255.html
> [3] https://github.com/curl/curl/commit/8e8bdd36048c146fe843d8948a912fbd398d1c5d
> [4] https://github.com/curl/curl/commit/67300814294ab0aa311b8568b460619ecc522949
>
> Signed-off-by: Darsh Kelaiya <dkelaiya@cisco.com>
> ---
> Changes in v2:
> - Reject malformed attribute names/values before dispatch so a cookie
> cannot lose its Secure flag.
> - Carry the TAB-value fix and adapted validation in separate
> dependent_p1 and dependent_p2 patches before the main fix.
> - Extend test 2885 with the malformed attribute case, retaining the
> TAB-before-Secure and normal Secure controls.
> ---
> .../curl/CVE-2026-80255-dependent_p1.patch | 139 ++++++++++++++++++
> .../curl/CVE-2026-80255-dependent_p2.patch | 81 ++++++++++
> .../curl/curl/CVE-2026-80255.patch | 115 +++++++++++++++
> meta/recipes-support/curl/curl_8.19.0.bb | 3 +
> 4 files changed, 338 insertions(+)
> create mode 100644 meta/recipes-support/curl/curl/CVE-2026-80255-dependent_p1.patch
> create mode 100644 meta/recipes-support/curl/curl/CVE-2026-80255-dependent_p2.patch
> create mode 100644 meta/recipes-support/curl/curl/CVE-2026-80255.patch
>
> diff --git a/meta/recipes-support/curl/curl/CVE-2026-80255-dependent_p1.patch b/meta/recipes-support/curl/curl/CVE-2026-80255-dependent_p1.patch
> new file mode 100644
> index 0000000000..ca7737e5dd
> --- /dev/null
> +++ b/meta/recipes-support/curl/curl/CVE-2026-80255-dependent_p1.patch
> @@ -0,0 +1,139 @@
> +From 8e8bdd36048c146fe843d8948a912fbd398d1c5d Mon Sep 17 00:00:00 2001
> +From: Daniel Stenberg <daniel@haxx.se>
> +Date: Wed, 1 Apr 2026 08:09:47 +0200
> +Subject: [PATCH] cookie: fix rejection when tabs in value
> +
> +A previous refactor changed the TAB check so that the octet could be
> +accepted in the 'path', which would cause an invalid line in the saved
> +cookie file so not possible to read the cookie back. Not terrible
> +because the path cannot contain a raw tab anyway so it would never match
> +anyway.
> +
> +Add test 1685 to verify
> +
> +Reported-by: Izan on hackerone
> +
> +Closes #21185
> +
> +CVE: CVE-2026-80255
> +Upstream-Status: Backport [https://github.com/curl/curl/commit/8e8bdd36048c146fe843d8948a912fbd398d1c5d]
> +
> +Backport Changes:
> +- Registered test 1685 in the older curl 8.19.0 test-list row.
> +
> +(cherry picked from commit 8e8bdd36048c146fe843d8948a912fbd398d1c5d)
> +Signed-off-by: Darsh Kelaiya <dkelaiya@cisco.com>
> +---
> + lib/cookie.c | 14 ++++++-------
> + tests/data/Makefile.am | 2 +-
> + tests/data/test1685 | 57 ++++++++++++++++++++++++++++++++++++++++++++++++++
> + 3 files changed, 65 insertions(+), 8 deletions(-)
> +
> +diff --git a/lib/cookie.c b/lib/cookie.c
> +index 4d53cc4..cd95130 100644
> +--- a/lib/cookie.c
> ++++ b/lib/cookie.c
> +@@ -461,6 +461,13 @@ parse_cookie_header(struct Curl_easy *data,
> + sep = TRUE; /* a '=' was used */
> + if(!curlx_str_cspn(&ptr, &val, ";\r\n"))
> + curlx_str_trimblanks(&val);
> ++
> ++ /* Reject cookies with a TAB inside the value */
> ++ if(curlx_strlen(&val) &&
> ++ memchr(curlx_str(&val), '\t', curlx_strlen(&val))) {
> ++ infof(data, "cookie contains TAB, dropping");
> ++ return CURLE_OK;
> ++ }
> + }
> + else
> + curlx_str_init(&val);
> +@@ -489,13 +496,6 @@ parse_cookie_header(struct Curl_easy *data,
> + return CURLE_OK;
> + }
> +
> +- /* Reject cookies with a TAB inside the value */
> +- if(curlx_strlen(&val) &&
> +- memchr(curlx_str(&val), '\t', curlx_strlen(&val))) {
> +- infof(data, "cookie contains TAB, dropping");
> +- return CURLE_OK;
> +- }
> +-
> + /* Check if we have a reserved prefix set. */
> + if(!strncmp("__Secure-", curlx_str(&name), 9))
> + co->prefix_secure = TRUE;
> +diff --git a/tests/data/Makefile.am b/tests/data/Makefile.am
> +index ea700cd..5bb4f7e 100644
> +--- a/tests/data/Makefile.am
> ++++ b/tests/data/Makefile.am
> +@@ -226,7 +226,7 @@ test1660 test1661 test1662 test1663 test1664 test1665 \
> + \
> + test1670 test1671 \
> + \
> +-test1680 test1681 test1682 test1683 \
> ++test1680 test1681 test1682 test1683 test1685 \
> + \
> + test1700 test1701 test1702 test1703 test1704 test1705 test1706 test1707 \
> + test1708 test1709 test1710 test1711 test1712 \
> +diff --git a/tests/data/test1685 b/tests/data/test1685
> +new file mode 100644
> +index 0000000..b98dc20
> +--- /dev/null
> ++++ b/tests/data/test1685
> +@@ -0,0 +1,57 @@
> ++<?xml version="1.0" encoding="US-ASCII"?>
> ++<testcase>
> ++<info>
> ++<keywords>
> ++HTTP
> ++cookies
> ++</keywords>
> ++</info>
> ++
> ++# Server-side
> ++<reply>
> ++<data crlf="headers">
> ++HTTP/1.0 200 OK swsclose
> ++Date: Tue, 09 Nov 2010 14:49:00 GMT
> ++Content-Type: text/html
> ++Set-Cookie: name=content; path=/we%TABwant
> ++Set-Cookie: accept=this; path=/only/this
> ++
> ++boo
> ++</data>
> ++</reply>
> ++
> ++# Client-side
> ++<client>
> ++<server>
> ++http
> ++</server>
> ++<name>
> ++HTTP, reject cookie with tab in path
> ++</name>
> ++<command>
> ++http://%HOSTIP:%HTTPPORT/ -c %LOGDIR/jar%TESTNUMBER.txt
> ++</command>
> ++
> ++<features>
> ++cookies
> ++</features>
> ++</client>
> ++
> ++# Verify data after the test has been "shot"
> ++<verify>
> ++<protocol crlf="headers">
> ++GET / HTTP/1.1
> ++Host: %HOSTIP:%HTTPPORT
> ++User-Agent: curl/%VERSION
> ++Accept: */*
> ++
> ++</protocol>
> ++<file name="%LOGDIR/jar%TESTNUMBER.txt" mode="text">
> ++# Netscape HTTP Cookie File
> ++# https://curl.se/docs/http-cookies.html
> ++# This file was generated by libcurl! Edit at your own risk.
> ++
> ++127.0.0.1%TABFALSE%TAB/only/this%TABFALSE%TAB0%TABaccept%TABthis
> ++</file>
> ++</verify>
> ++</testcase>
> diff --git a/meta/recipes-support/curl/curl/CVE-2026-80255-dependent_p2.patch b/meta/recipes-support/curl/curl/CVE-2026-80255-dependent_p2.patch
> new file mode 100644
> index 0000000000..8950b28784
> --- /dev/null
> +++ b/meta/recipes-support/curl/curl/CVE-2026-80255-dependent_p2.patch
> @@ -0,0 +1,81 @@
> +From 67300814294ab0aa311b8568b460619ecc522949 Mon Sep 17 00:00:00 2001
> +From: Daniel Stenberg <daniel@haxx.se>
> +Date: Wed, 3 Jun 2026 08:10:14 +0200
> +Subject: [PATCH] cookie: refactor parse_cookie_header
> +
> +- introduce a few static helper functions
> +- simplify the bad octet checks
> +- simplify the too long cookie/value check
> +
> +Closes #21854
> +
> +CVE: CVE-2026-80255
> +Upstream-Status: Backport [https://github.com/curl/curl/commit/67300814294ab0aa311b8568b460619ecc522949]
> +
> +Backport Changes:
> +- Kept parse_cookie_header() in its curl 8.19.0 form instead of
> + extracting parse_first_pair(), parse_flag(), parse_domain(),
> + parse_maxage() and parse_expires().
> +- Applied the control-octet checks to that existing parser: reject
> + TAB in invalid_octets(), validate every attribute name/value,
> + and remove the duplicate first-pair checks.
> +- Kept the older cookie name/value length limits; omitted upstream's
> + separate simplification of those limits.
> +
> +(cherry picked from commit 67300814294ab0aa311b8568b460619ecc522949)
> +Signed-off-by: Darsh Kelaiya <dkelaiya@cisco.com>
> +---
> + lib/cookie.c | 17 +++++++++--------
> + 1 file changed, 9 insertions(+), 8 deletions(-)
> +
> +diff --git a/lib/cookie.c b/lib/cookie.c
> +index cd95130..b3b1ebd 100644
> +--- a/lib/cookie.c
> ++++ b/lib/cookie.c
> +@@ -354,9 +354,9 @@ static bool bad_domain(const char *domain, size_t len)
> + static bool invalid_octets(const char *ptr, size_t len)
> + {
> + const unsigned char *p = (const unsigned char *)ptr;
> +- /* Reject all bytes \x01 - \x1f (*except* \x09, TAB) + \x7f */
> ++ /* Reject all bytes \x01 - \x1f + \x7f */
> + while(len && *p) {
> +- if(((*p != 9) && (*p < 0x20)) || (*p == 0x7f))
> ++ if((*p < 0x20) || (*p == 0x7f))
> + return TRUE;
> + p++;
> + len--;
> +@@ -457,15 +457,18 @@ parse_cookie_header(struct Curl_easy *data,
> + bool sep = FALSE;
> + curlx_str_trimblanks(&name);
> +
> ++ if(invalid_octets(curlx_str(&name), curlx_strlen(&name))) {
> ++ infof(data, "invalid octets in name, cookie dropped");
> ++ return CURLE_OK;
> ++ }
> ++
> + if(!curlx_str_single(&ptr, '=')) {
> + sep = TRUE; /* a '=' was used */
> + if(!curlx_str_cspn(&ptr, &val, ";\r\n"))
> + curlx_str_trimblanks(&val);
> +
> +- /* Reject cookies with a TAB inside the value */
> +- if(curlx_strlen(&val) &&
> +- memchr(curlx_str(&val), '\t', curlx_strlen(&val))) {
> +- infof(data, "cookie contains TAB, dropping");
> ++ if(invalid_octets(curlx_str(&val), curlx_strlen(&val))) {
> ++ infof(data, "invalid octets in value, cookie dropped");
> + return CURLE_OK;
> + }
> + }
> +@@ -476,8 +479,6 @@ parse_cookie_header(struct Curl_easy *data,
> + /* The first name/value pair is the actual cookie name */
> + if(!sep ||
> + /* Bad name/value pair. */
> +- invalid_octets(curlx_str(&name), curlx_strlen(&name)) ||
> +- invalid_octets(curlx_str(&val), curlx_strlen(&val)) ||
> + !curlx_strlen(&name)) {
> + infof(data, "invalid octets in name/value, cookie dropped");
^ That message does not apply anymore. This
is only verbose/debug code but that makes me
wonder if that doesn't hide more something
more serious...
I know refactors are usually unacceptable from stables. But, in this
case, I wonder if taking the whole commit would make this fix more easy
to review (an upstream-vetted refactor might be preferable over me
checking if a partial backport has all the right pieces)
What do you think?
Regards,
--
Yoann Congal
Smile ECS
next prev parent reply other threads:[~2026-11-17 9:54 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-10-23 9:33 [OE-core][wrynose][PATCH v2 0/5] curl: backport security fixes Darsh Kelaiya -X (dkelaiya - E INFOCHIPS PRIVATE LIMITED at Cisco)
2026-10-23 9:33 ` [OE-core][wrynose][PATCH v2 1/5] curl: fix CVE-2026-8926 Darsh Kelaiya -X (dkelaiya - E INFOCHIPS PRIVATE LIMITED at Cisco)
2026-11-16 14:36 ` Yoann Congal
2026-10-23 9:33 ` [OE-core][wrynose][PATCH v2 2/5] curl: fix CVE-2026-11564 Darsh Kelaiya -X (dkelaiya - E INFOCHIPS PRIVATE LIMITED at Cisco)
2026-11-16 15:51 ` Yoann Congal
2026-11-17 9:50 ` Paul Barker
2026-10-23 9:33 ` [OE-core][wrynose][PATCH v2 3/5] curl: fix CVE-2026-80255 Darsh Kelaiya -X (dkelaiya - E INFOCHIPS PRIVATE LIMITED at Cisco)
2026-11-17 9:54 ` Yoann Congal [this message]
2026-10-23 9:33 ` [OE-core][wrynose][PATCH v2 4/5] curl: fix CVE-2026-82208 Darsh Kelaiya -X (dkelaiya - E INFOCHIPS PRIVATE LIMITED at Cisco)
2026-10-23 9:33 ` [OE-core][wrynose][PATCH v2 5/5] curl: fix CVE-2026-80231 Darsh Kelaiya -X (dkelaiya - E INFOCHIPS PRIVATE LIMITED at Cisco)
2026-11-17 10:31 ` Yoann Congal
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=DMXE85B25SK0.3VK8FVR2RZ6AI@smile.fr \
--to=yoann.congal@smile.fr \
--cc=dkelaiya@cisco.com \
--cc=openembedded-core@lists.openembedded.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox