Skip to contentSkip to conversation

python3-cryptography: set CVE_PRODUCT

PATCH Superseded [email protected] 3 participants 7 messages Jul 10 – Jul 13

v1 of 2 Superseded Latest is v2 →

lore
Share & tools
lore
#0 lore raw

PATCH 1/4 python3-cryptography: set CVE_PRODUCT

From: "mark.yang" <[email protected]>

NVD lists it as cryptography.io:cryptography,
and CNA lists it as pyca:cryptography, so CVE_PRODUCT is set to cryptography.

---
 meta/recipes-devtools/python/python3-cryptography.bb | 2 ++
 1 file changed, 2 insertions(+)
meta/recipes-devtools/python/python3-cryptography.bb +2 -0
index 4730c34997..6138028d55 100644
@@ -71,4 +71,6 @@ do_install_ptest() {
     cp -r ${S}/pyproject.toml ${D}${PTEST_PATH}/
 }
 
+CVE_PRODUCT = "cryptography"
+
 BBCLASSEXTEND = "native nativesdk"
#1 lore raw

PATCH 2/4 python3-ply: set CVE_PRODUCT and CVE_STATUS for CVE-2025-56005 as disputed

From: "mark.yang" <[email protected]>

For python3-ply, the CVE_PRODUCT should be set to ply, not python:ply.
NVD registers ply as dabeaz:ply, so the default python:ply vendor prefix never matches and no CVEs are reported.

CVE-2025-56005 is disputed.
See https://nvd.nist.gov/vuln/detail/CVE-2025-56005

---
 meta/recipes-devtools/python/python3-ply_3.11.bb | 4 ++++
 1 file changed, 4 insertions(+)
meta/recipes-devtools/python/python3-ply_3.11.bb +4 -0
index 2c5fa3f215..ce45bbc1bd 100644
@@ -14,4 +14,8 @@ RDEPENDS:${PN}:class-target += "\
     python3-shell \
 "
 
+CVE_PRODUCT = "ply"
+# https://nvd.nist.gov/vuln/detail/CVE-2025-56005 Disputed
+CVE_STATUS[CVE-2025-56005] = "disputed: PoC fails to demonstrate code execution; picklefile is a developer-controlled parser-table cache parameter"
+
 BBCLASSEXTEND = "native nativesdk"
#2 lore raw

Re: [OE-core] [PATCH 2/4] python3-ply: set CVE_PRODUCT and CVE_STATUS for CVE-2025-56005 as disputed

Quoted 18 lines · mark.yang via lists.openembedded.org

From: "mark.yang" <[email protected]>

For python3-ply, the CVE_PRODUCT should be set to ply, not python:ply.
NVD registers ply as dabeaz:ply, so the default python:ply vendor prefix never matches and no CVEs are reported.

CVE-2025-56005 is disputed.
See https://nvd.nist.gov/vuln/detail/CVE-2025-56005

---
 meta/recipes-devtools/python/python3-ply_3.11.bb | 4 ++++
 1 file changed, 4 insertions(+)
meta/recipes-devtools/python/python3-ply_3.11.bb +0 -0
index 2c5fa3f215..ce45bbc1bd 100644

meta/recipes-devtools/python/python3-ply_3.11.bb RDEPENDS:${PN}:class-target += "\ · line 20

Quoted 7 lines
meta/recipes-devtools/python/python3-ply_3.11.bb +4 -0
@@ -14,4 +14,8 @@ RDEPENDS:${PN}:class-target += "\
     python3-shell \
 "
 
+CVE_PRODUCT = "ply"
+# https://nvd.nist.gov/vuln/detail/CVE-2025-56005 Disputed
+CVE_STATUS[CVE-2025-56005] = "disputed: PoC fails to demonstrate code execution; picklefile is a developer-controlled parser-table cache parameter"
+

Hi,

The NVD entry [1] says:

A third-party states that this vulnerability should be rejected because the proof of concept does not demonstrate arbitrary code execution and fails to complete successfully.

Following the links in the NVD entry I see what I assume is the argument to dispute this [2] and an argument to dispute the dispute [3]. It's not very convincing to me. I think we need a more clear reason why we want to ignore this CVE.

[1]: https://nvd.nist.gov/vuln/detail/CVE-2025-56005
[2]: https://github.com/tom025/ply_exploit_rejection
[3]: https://www.openwall.com/lists/oss-security/2026/01/29/1

Best regards,

Signature · 1 line
Paul Barker
#3 lore raw

PATCH 3/4 python3-pyasn1: set CVE_PRODUCT

In reply to [email protected]

From: "mark.yang" <[email protected]>

Set the product to pyasn1.

The default python:pyasn1 does not match the NVD/CNA entries which use pyasn1 as vendor, so CVEs like CVE-2026-30922 are never reported.

---
 meta/recipes-devtools/python/python3-pyasn1_0.6.3.bb | 2 ++
 1 file changed, 2 insertions(+)
meta/recipes-devtools/python/python3-pyasn1_0.6.3.bb +2 -0
index fb572d9da4..ee3cfc9813 100644
@@ -14,4 +14,6 @@ RDEPENDS:${PN}:class-target += " \
     python3-shell \
 "
 
+CVE_PRODUCT = "pyasn1"
+
 BBCLASSEXTEND = "native nativesdk"
#4 lore raw

Re: [OE-core] [PATCH 3/4] python3-pyasn1: set CVE_PRODUCT

Quoted 2 lines · mark.yang via lists.openembedded.org

The default python:pyasn1 does not match the NVD/CNA entries which use pyasn1 as vendor, so CVEs like CVE-2026-30922 are never reported.

Where we know the exact CPE that is used, we should use it. Can you set CVE_PRODUCT to pyasn1:pyasn1 instead.

Ross

#5 lore raw

PATCH 4/4 libx11-compose-data: set CVE_PRODUCT to empty value

In reply to [email protected]

From: "mark.yang" <[email protected]>

This recipe only builds and installs the compose data files (nls/) from the libX11 sources and contains no compiled libX11 code, so libx11 CVEs do not apply to it. When the x11 DISTRO_FEATURE is enabled this recipe is skipped and libx11 itself is built and scanned instead.

Referred glibc-locale case:
https://github.com/openembedded/openembedded-core/commit/1f9a963b9ff7ebe052ba54b9fcbdf7d09478dd17

---
 .../recipes-graphics/xorg-lib/libx11-compose-data_1.8.12.bb | 6 ++++++
 1 file changed, 6 insertions(+)
meta/recipes-graphics/xorg-lib/libx11-compose-data_1.8.12.bb +6 -0
index 562a8cbf16..4bf9f07aba 100644
@@ -29,3 +29,9 @@ do_install() {
 PACKAGES = "${PN}"
 
 FILES:${PN} = "${datadir}/X11/locale ${libdir}/X11/locale"
+
+# This recipe only builds and installs the compose data files (nls/) from
+# the libX11 sources and contains no compiled libX11 code, so libx11 CVEs
+# do not apply to it. When the x11 DISTRO_FEATURE is enabled this recipe
+# is skipped and libx11 itself is built and scanned instead.
+CVE_PRODUCT = ""
#6 lore raw
In reply to [email protected]
Quoted 2 lines · mark.yang via lists.openembedded.org

NVD lists it as cryptography.io:cryptography,
and CNA lists it as pyca:cryptography, so CVE_PRODUCT is set to cryptography.

“cryptography” is vague and will no doubt cause false-positives in the future, and CVE_PRODUCT is actually a list, so setting it to "cryptography.io:cryptography pyca:cryptography” will match correctly and precisely.

Ross