sys-kernel/coreos-sources: bump to 4.4.1

This commit is contained in:
Alex Crawford 2016-02-09 16:50:08 -08:00
parent 1f80337458
commit f11fdc61bb
23 changed files with 43 additions and 120 deletions

View File

@ -37,6 +37,4 @@ UNIPATCH_LIST="
${PATCH_DIR}/0019-overlayfs-use-a-minimal-buffer-in-ovl_copy_xattr.patch \ ${PATCH_DIR}/0019-overlayfs-use-a-minimal-buffer-in-ovl_copy_xattr.patch \
${PATCH_DIR}/0020-kbuild-derive-relative-path-for-KBUILD_SRC-from-CURD.patch \ ${PATCH_DIR}/0020-kbuild-derive-relative-path-for-KBUILD_SRC-from-CURD.patch \
${PATCH_DIR}/0021-Don-t-verify-write-permissions-on-lower-inodes-on-ov.patch \ ${PATCH_DIR}/0021-Don-t-verify-write-permissions-on-lower-inodes-on-ov.patch \
${PATCH_DIR}/0022-KEYS-Fix-keyring-ref-leak-in-join_session_keyring.patch \
" "

View File

@ -1,7 +1,7 @@
From ed3da1ded7b7581a9a1dc2b48f8ddc7975f3ea67 Mon Sep 17 00:00:00 2001 From 2e10f053682b2614c8689ab7cd792030adb37c3d Mon Sep 17 00:00:00 2001
From: Matthew Garrett <matthew.garrett@nebula.com> From: Matthew Garrett <matthew.garrett@nebula.com>
Date: Fri, 9 Aug 2013 17:58:15 -0400 Date: Fri, 9 Aug 2013 17:58:15 -0400
Subject: [PATCH 01/22] Add secure_modules() call Subject: [PATCH 01/21] Add secure_modules() call
Provide a single call to allow kernel code to determine whether the system Provide a single call to allow kernel code to determine whether the system
has been configured to either disable module loading entirely or to load has been configured to either disable module loading entirely or to load

View File

@ -1,7 +1,7 @@
From e797ce01ad3c0faa578734900a7c03ee04c06c08 Mon Sep 17 00:00:00 2001 From 8161285fced6623edd4c66f9c2d3ece69014a392 Mon Sep 17 00:00:00 2001
From: Matthew Garrett <matthew.garrett@nebula.com> From: Matthew Garrett <matthew.garrett@nebula.com>
Date: Thu, 8 Mar 2012 10:10:38 -0500 Date: Thu, 8 Mar 2012 10:10:38 -0500
Subject: [PATCH 02/22] PCI: Lock down BAR access when module security is Subject: [PATCH 02/21] PCI: Lock down BAR access when module security is
enabled enabled
Any hardware that can potentially generate DMA has to be locked down from Any hardware that can potentially generate DMA has to be locked down from

View File

@ -1,7 +1,7 @@
From e1e4b600d77353180227e93c3dda49ebde147578 Mon Sep 17 00:00:00 2001 From f55838325eadbb393aaf61a61a177fd7ad2f0280 Mon Sep 17 00:00:00 2001
From: Matthew Garrett <matthew.garrett@nebula.com> From: Matthew Garrett <matthew.garrett@nebula.com>
Date: Thu, 8 Mar 2012 10:35:59 -0500 Date: Thu, 8 Mar 2012 10:35:59 -0500
Subject: [PATCH 03/22] x86: Lock down IO port access when module security is Subject: [PATCH 03/21] x86: Lock down IO port access when module security is
enabled enabled
IO port access would permit users to gain access to PCI configuration IO port access would permit users to gain access to PCI configuration

View File

@ -1,7 +1,7 @@
From 15647227ed911e525339ece57b4af9d369390bb0 Mon Sep 17 00:00:00 2001 From 957b35947b86b16d1baadce8ec63db80bfb6466a Mon Sep 17 00:00:00 2001
From: Matthew Garrett <matthew.garrett@nebula.com> From: Matthew Garrett <matthew.garrett@nebula.com>
Date: Fri, 9 Mar 2012 08:39:37 -0500 Date: Fri, 9 Mar 2012 08:39:37 -0500
Subject: [PATCH 04/22] ACPI: Limit access to custom_method Subject: [PATCH 04/21] ACPI: Limit access to custom_method
custom_method effectively allows arbitrary access to system memory, making custom_method effectively allows arbitrary access to system memory, making
it possible for an attacker to circumvent restrictions on module loading. it possible for an attacker to circumvent restrictions on module loading.

View File

@ -1,7 +1,7 @@
From 5b0f82c10dd93fd281e5f31c01deea1f3e2af1d1 Mon Sep 17 00:00:00 2001 From 86c4a0683e7310bad411a1834ce2b949d5bd4534 Mon Sep 17 00:00:00 2001
From: Matthew Garrett <matthew.garrett@nebula.com> From: Matthew Garrett <matthew.garrett@nebula.com>
Date: Fri, 9 Mar 2012 08:46:50 -0500 Date: Fri, 9 Mar 2012 08:46:50 -0500
Subject: [PATCH 05/22] asus-wmi: Restrict debugfs interface when module Subject: [PATCH 05/21] asus-wmi: Restrict debugfs interface when module
loading is restricted loading is restricted
We have no way of validating what all of the Asus WMI methods do on a We have no way of validating what all of the Asus WMI methods do on a

View File

@ -1,7 +1,7 @@
From 37f5217e456a13bb92814e515616b0524fbf0a89 Mon Sep 17 00:00:00 2001 From 03bc662b54a1a5978a2c840eba182b28e65f0c81 Mon Sep 17 00:00:00 2001
From: Matthew Garrett <matthew.garrett@nebula.com> From: Matthew Garrett <matthew.garrett@nebula.com>
Date: Fri, 9 Mar 2012 09:28:15 -0500 Date: Fri, 9 Mar 2012 09:28:15 -0500
Subject: [PATCH 06/22] Restrict /dev/mem and /dev/kmem when module loading is Subject: [PATCH 06/21] Restrict /dev/mem and /dev/kmem when module loading is
restricted restricted
Allowing users to write to address space makes it possible for the kernel Allowing users to write to address space makes it possible for the kernel

View File

@ -1,7 +1,7 @@
From f41415ab2cf92434113fbc97fc856ddd6e8a88da Mon Sep 17 00:00:00 2001 From 16d485311fc3079de4f5b986f2fc2f7d70274f8d Mon Sep 17 00:00:00 2001
From: Josh Boyer <jwboyer@redhat.com> From: Josh Boyer <jwboyer@redhat.com>
Date: Mon, 25 Jun 2012 19:57:30 -0400 Date: Mon, 25 Jun 2012 19:57:30 -0400
Subject: [PATCH 07/22] acpi: Ignore acpi_rsdp kernel parameter when module Subject: [PATCH 07/21] acpi: Ignore acpi_rsdp kernel parameter when module
loading is restricted loading is restricted
This option allows userspace to pass the RSDP address to the kernel, which This option allows userspace to pass the RSDP address to the kernel, which

View File

@ -1,7 +1,7 @@
From e227953c81434fb5156dd2504aeee7960c37a0ad Mon Sep 17 00:00:00 2001 From 7d0d3cb705bb1ae5a739d0087e62844d3bec5e6f Mon Sep 17 00:00:00 2001
From: Matthew Garrett <mjg59@coreos.com> From: Matthew Garrett <mjg59@coreos.com>
Date: Thu, 19 Nov 2015 18:55:53 -0800 Date: Thu, 19 Nov 2015 18:55:53 -0800
Subject: [PATCH 08/22] kexec: Disable at runtime if the kernel enforces module Subject: [PATCH 08/21] kexec: Disable at runtime if the kernel enforces module
loading restrictions loading restrictions
kexec permits the loading and execution of arbitrary code in ring 0, which kexec permits the loading and execution of arbitrary code in ring 0, which

View File

@ -1,7 +1,7 @@
From 1636adeff714c17d2c9a872e6be9b025df85ef64 Mon Sep 17 00:00:00 2001 From c682c72e808feb7c4dcb42ecaae7016c13ce5610 Mon Sep 17 00:00:00 2001
From: Matthew Garrett <matthew.garrett@nebula.com> From: Matthew Garrett <matthew.garrett@nebula.com>
Date: Fri, 8 Feb 2013 11:12:13 -0800 Date: Fri, 8 Feb 2013 11:12:13 -0800
Subject: [PATCH 09/22] x86: Restrict MSR access when module loading is Subject: [PATCH 09/21] x86: Restrict MSR access when module loading is
restricted restricted
Writing to MSRs should not be allowed if module loading is restricted, Writing to MSRs should not be allowed if module loading is restricted,

View File

@ -1,7 +1,7 @@
From f08b4a4b93bc28efe2d7aab38a6b44592d944dda Mon Sep 17 00:00:00 2001 From abac45cbcaa27170eef195cb48c33a1b37071f2a Mon Sep 17 00:00:00 2001
From: Matthew Garrett <matthew.garrett@nebula.com> From: Matthew Garrett <matthew.garrett@nebula.com>
Date: Fri, 9 Aug 2013 18:36:30 -0400 Date: Fri, 9 Aug 2013 18:36:30 -0400
Subject: [PATCH 10/22] Add option to automatically enforce module signatures Subject: [PATCH 10/21] Add option to automatically enforce module signatures
when in Secure Boot mode when in Secure Boot mode
UEFI Secure Boot provides a mechanism for ensuring that the firmware will UEFI Secure Boot provides a mechanism for ensuring that the firmware will

View File

@ -1,7 +1,7 @@
From 9bfe6c0b8200244a9517979dc06d3d7bcf8fde4a Mon Sep 17 00:00:00 2001 From 76ba8b2fee84c6489316547f19d03a0485f59dc3 Mon Sep 17 00:00:00 2001
From: Josh Boyer <jwboyer@fedoraproject.org> From: Josh Boyer <jwboyer@fedoraproject.org>
Date: Tue, 27 Aug 2013 13:28:43 -0400 Date: Tue, 27 Aug 2013 13:28:43 -0400
Subject: [PATCH 11/22] efi: Make EFI_SECURE_BOOT_SIG_ENFORCE depend on EFI Subject: [PATCH 11/21] efi: Make EFI_SECURE_BOOT_SIG_ENFORCE depend on EFI
The functionality of the config option is dependent upon the platform being The functionality of the config option is dependent upon the platform being
UEFI based. Reflect this in the config deps. UEFI based. Reflect this in the config deps.

View File

@ -1,7 +1,7 @@
From 1b435189fb66e031edc4df509576448a96b4c3ff Mon Sep 17 00:00:00 2001 From 8d2a8d8ce61706a3a778ae9fd79cb5bab91a2817 Mon Sep 17 00:00:00 2001
From: Josh Boyer <jwboyer@fedoraproject.org> From: Josh Boyer <jwboyer@fedoraproject.org>
Date: Tue, 27 Aug 2013 13:33:03 -0400 Date: Tue, 27 Aug 2013 13:33:03 -0400
Subject: [PATCH 12/22] efi: Add EFI_SECURE_BOOT bit Subject: [PATCH 12/21] efi: Add EFI_SECURE_BOOT bit
UEFI machines can be booted in Secure Boot mode. Add a EFI_SECURE_BOOT bit UEFI machines can be booted in Secure Boot mode. Add a EFI_SECURE_BOOT bit
for use with efi_enabled. for use with efi_enabled.

View File

@ -1,7 +1,7 @@
From e62a3871237bb79ef5e51b112eff7d940cf06020 Mon Sep 17 00:00:00 2001 From b671df07aed28fcbc9e470b52b8c1822f78303c0 Mon Sep 17 00:00:00 2001
From: Josh Boyer <jwboyer@fedoraproject.org> From: Josh Boyer <jwboyer@fedoraproject.org>
Date: Fri, 20 Jun 2014 08:53:24 -0400 Date: Fri, 20 Jun 2014 08:53:24 -0400
Subject: [PATCH 13/22] hibernate: Disable in a signed modules environment Subject: [PATCH 13/21] hibernate: Disable in a signed modules environment
There is currently no way to verify the resume image when returning There is currently no way to verify the resume image when returning
from hibernate. This might compromise the signed modules trust model, from hibernate. This might compromise the signed modules trust model,

View File

@ -1,7 +1,7 @@
From 70aadec167cb84865c6e85c1eccc218a024f86ef Mon Sep 17 00:00:00 2001 From 9cb22840851be7a7f842229e6603a6b4b25e824d Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com> From: David Howells <dhowells@redhat.com>
Date: Tue, 16 Jun 2015 14:14:31 +0100 Date: Tue, 16 Jun 2015 14:14:31 +0100
Subject: [PATCH 14/22] Security: Provide copy-up security hooks for unioned Subject: [PATCH 14/21] Security: Provide copy-up security hooks for unioned
files files
Provide two new security hooks for use with security files that are used when Provide two new security hooks for use with security files that are used when

View File

@ -1,7 +1,7 @@
From 2e1d35fb4b10cafc0dac63436f94fda8b4e738ee Mon Sep 17 00:00:00 2001 From 64ef0efdd90f5aae4fae7c76783b09af53d29dfe Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com> From: David Howells <dhowells@redhat.com>
Date: Tue, 16 Jun 2015 14:14:31 +0100 Date: Tue, 16 Jun 2015 14:14:31 +0100
Subject: [PATCH 15/22] Overlayfs: Use copy-up security hooks Subject: [PATCH 15/21] Overlayfs: Use copy-up security hooks
Use the copy-up security hooks previously provided to allow an LSM to adjust Use the copy-up security hooks previously provided to allow an LSM to adjust
the security on a newly created copy and to filter the xattrs copied to that the security on a newly created copy and to filter the xattrs copied to that

View File

@ -1,7 +1,7 @@
From df782b85901bc5a1e1d5c90895b0166cb7ba6260 Mon Sep 17 00:00:00 2001 From 38d19edb9bae02a9e78b26a7b2c4f0980ee13ee3 Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com> From: David Howells <dhowells@redhat.com>
Date: Tue, 16 Jun 2015 14:14:32 +0100 Date: Tue, 16 Jun 2015 14:14:32 +0100
Subject: [PATCH 16/22] SELinux: Stub in copy-up handling Subject: [PATCH 16/21] SELinux: Stub in copy-up handling
Provide stubs for union/overlay copy-up handling. The xattr copy up stub Provide stubs for union/overlay copy-up handling. The xattr copy up stub
discards lower SELinux xattrs rather than letting them be copied up so that discards lower SELinux xattrs rather than letting them be copied up so that

View File

@ -1,7 +1,7 @@
From ce05f979bd98e5f267330f47d9a26bbb138dc54f Mon Sep 17 00:00:00 2001 From 3e6ccc54dd0383a8c57287f9e63f392595e28cb1 Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com> From: David Howells <dhowells@redhat.com>
Date: Tue, 16 Jun 2015 14:14:32 +0100 Date: Tue, 16 Jun 2015 14:14:32 +0100
Subject: [PATCH 17/22] SELinux: Handle opening of a unioned file Subject: [PATCH 17/21] SELinux: Handle opening of a unioned file
Handle the opening of a unioned file by trying to derive the label that would Handle the opening of a unioned file by trying to derive the label that would
be attached to the union-layer inode if it doesn't exist. be attached to the union-layer inode if it doesn't exist.

View File

@ -1,7 +1,7 @@
From f60b70463bb7493f60a27ac2d06058da87b062d9 Mon Sep 17 00:00:00 2001 From 7b0a1257f4b4a35f087db9120b684d3a9c8181e5 Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com> From: David Howells <dhowells@redhat.com>
Date: Tue, 16 Jun 2015 14:14:32 +0100 Date: Tue, 16 Jun 2015 14:14:32 +0100
Subject: [PATCH 18/22] SELinux: Check against union label for file operations Subject: [PATCH 18/21] SELinux: Check against union label for file operations
File operations (eg. read, write) issued against a file that is attached to File operations (eg. read, write) issued against a file that is attached to
the lower layer of a union file needs to be checked against the union-layer the lower layer of a union file needs to be checked against the union-layer

View File

@ -1,7 +1,7 @@
From 116f798bcf3fd2ce4965cb15ec44c8180f0428c1 Mon Sep 17 00:00:00 2001 From 7505098adc7a76c3d001831af40f39c86d624a67 Mon Sep 17 00:00:00 2001
From: Vito Caputo <vito.caputo@coreos.com> From: Vito Caputo <vito.caputo@coreos.com>
Date: Mon, 19 Oct 2015 17:53:12 -0700 Date: Mon, 19 Oct 2015 17:53:12 -0700
Subject: [PATCH 19/22] overlayfs: use a minimal buffer in ovl_copy_xattr Subject: [PATCH 19/21] overlayfs: use a minimal buffer in ovl_copy_xattr
Rather than always allocating the high-order XATTR_SIZE_MAX buffer Rather than always allocating the high-order XATTR_SIZE_MAX buffer
which is costly and prone to failure, only allocate what is needed and which is costly and prone to failure, only allocate what is needed and

View File

@ -1,7 +1,7 @@
From 6f682c2c88f74b45c3692a994d90ed51412b932b Mon Sep 17 00:00:00 2001 From b0a4a60266e116f35e31a2054d9769f23dc88a95 Mon Sep 17 00:00:00 2001
From: Vito Caputo <vito.caputo@coreos.com> From: Vito Caputo <vito.caputo@coreos.com>
Date: Wed, 25 Nov 2015 02:59:45 -0800 Date: Wed, 25 Nov 2015 02:59:45 -0800
Subject: [PATCH 20/22] kbuild: derive relative path for KBUILD_SRC from CURDIR Subject: [PATCH 20/21] kbuild: derive relative path for KBUILD_SRC from CURDIR
This enables relocating source and build trees to different roots, This enables relocating source and build trees to different roots,
provided they stay reachable relative to one another. Useful for provided they stay reachable relative to one another. Useful for
@ -12,7 +12,7 @@ by some undesirable path component.
1 file changed, 2 insertions(+), 1 deletion(-) 1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/Makefile b/Makefile diff --git a/Makefile b/Makefile
index 70dea02..987d283 100644 index c6a265b..8125380 100644
--- a/Makefile --- a/Makefile
+++ b/Makefile +++ b/Makefile
@@ -143,7 +143,8 @@ $(filter-out _all sub-make $(CURDIR)/Makefile, $(MAKECMDGOALS)) _all: sub-make @@ -143,7 +143,8 @@ $(filter-out _all sub-make $(CURDIR)/Makefile, $(MAKECMDGOALS)) _all: sub-make

View File

@ -1,7 +1,7 @@
From 06ccab87d8c415e51bcf69e34bb27712bad8398f Mon Sep 17 00:00:00 2001 From 196c562e9a0ef9a1580f35c014ee7f4669cfb5d7 Mon Sep 17 00:00:00 2001
From: Matthew Garrett <mjg59@coreos.com> From: Matthew Garrett <mjg59@coreos.com>
Date: Tue, 22 Dec 2015 07:43:52 +0000 Date: Tue, 22 Dec 2015 07:43:52 +0000
Subject: [PATCH 21/22] Don't verify write permissions on lower inodes on Subject: [PATCH 21/21] Don't verify write permissions on lower inodes on
overlayfs overlayfs
If a user opens a file r/w on overlayfs, and if the underlying inode is If a user opens a file r/w on overlayfs, and if the underlying inode is

View File

@ -1,75 +0,0 @@
From fc94e26e760f2e752aa55f7b2d58fdcbeeef433e Mon Sep 17 00:00:00 2001
From: Yevgeny Pats <yevgeny@perception-point.io>
Date: Mon, 11 Jan 2016 12:05:28 +0000
Subject: [PATCH 22/22] KEYS: Fix keyring ref leak in join_session_keyring()
If a thread is asked to join as a session keyring the keyring that's already
set as its session, we leak a keyring reference.
This can be tested with the following program:
#include <stddef.h>
#include <stdio.h>
#include <sys/types.h>
#include <keyutils.h>
int main(int argc, const char *argv[])
{
int i = 0;
key_serial_t serial;
serial = keyctl(KEYCTL_JOIN_SESSION_KEYRING,
"leaked-keyring");
if (serial < 0) {
perror("keyctl");
return -1;
}
if (keyctl(KEYCTL_SETPERM, serial,
KEY_POS_ALL | KEY_USR_ALL) < 0) {
perror("keyctl");
return -1;
}
for (i = 0; i < 100; i++) {
serial = keyctl(KEYCTL_JOIN_SESSION_KEYRING,
"leaked-keyring");
if (serial < 0) {
perror("keyctl");
return -1;
}
}
return 0;
}
If, after the program has run, there something like the following line in
/proc/keys:
3f3d898f I--Q--- 100 perm 3f3f0000 0 0 keyring leaked-keyring: empty
with a usage count of 100 * the number of times the program has been run,
then the kernel is malfunctioning. If leaked-keyring has zero usages or
has been garbage collected, then the problem is fixed.
Reported-by: Yevgeny Pats <yevgeny@perception-point.io>
Signed-off-by: David Howells <dhowells@redhat.com>
---
security/keys/process_keys.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/security/keys/process_keys.c b/security/keys/process_keys.c
index a3f85d2..e6d50172 100644
--- a/security/keys/process_keys.c
+++ b/security/keys/process_keys.c
@@ -794,6 +794,7 @@ long join_session_keyring(const char *name)
ret = PTR_ERR(keyring);
goto error2;
} else if (keyring == new->session_keyring) {
+ key_put(keyring);
ret = 0;
goto error2;
}
--
2.4.10