public inbox for gentoo-commits@lists.gentoo.org
 help / color / mirror / Atom feed
From: "Sam James" <sam@gentoo.org>
To: gentoo-commits@lists.gentoo.org
Subject: [gentoo-commits] repo/gentoo:master commit in: dev-libs/openssl/, dev-libs/openssl/files/
Date: Tue, 29 Oct 2024 02:10:03 +0000 (UTC)	[thread overview]
Message-ID: <1730167762.536e382d480933cfc6721f129368a8468ebd2321.sam@gentoo> (raw)

commit:     536e382d480933cfc6721f129368a8468ebd2321
Author:     Sam James <sam <AT> gentoo <DOT> org>
AuthorDate: Tue Oct 29 01:53:30 2024 +0000
Commit:     Sam James <sam <AT> gentoo <DOT> org>
CommitDate: Tue Oct 29 02:09:22 2024 +0000
URL:        https://gitweb.gentoo.org/repo/gentoo.git/commit/?id=536e382d

dev-libs/openssl: fix CVE-2024-9143 for 3.1.7-r1

Bug: https://bugs.gentoo.org/941643
Signed-off-by: Sam James <sam <AT> gentoo.org>

 .../files/openssl-3.1.7-CVE-2024-9143.patch        | 192 ++++++++++++++
 dev-libs/openssl/openssl-3.1.7-r1.ebuild           | 288 +++++++++++++++++++++
 2 files changed, 480 insertions(+)

diff --git a/dev-libs/openssl/files/openssl-3.1.7-CVE-2024-9143.patch b/dev-libs/openssl/files/openssl-3.1.7-CVE-2024-9143.patch
new file mode 100644
index 000000000000..4f33ef000dca
--- /dev/null
+++ b/dev-libs/openssl/files/openssl-3.1.7-CVE-2024-9143.patch
@@ -0,0 +1,192 @@
+https://bugs.gentoo.org/941643
+https://github.com/openssl/openssl/commit/fdf6723362ca51bd883295efe206cb5b1cfa5154
+
+From fdf6723362ca51bd883295efe206cb5b1cfa5154 Mon Sep 17 00:00:00 2001
+From: Viktor Dukhovni <viktor@openssl.org>
+Date: Thu, 19 Sep 2024 01:02:40 +1000
+Subject: [PATCH] Harden BN_GF2m_poly2arr against misuse.
+
+The BN_GF2m_poly2arr() function converts characteristic-2 field
+(GF_{2^m}) Galois polynomials from a representation as a BIGNUM bitmask,
+to a compact array with just the exponents of the non-zero terms.
+
+These polynomials are then used in BN_GF2m_mod_arr() to perform modular
+reduction.  A precondition of calling BN_GF2m_mod_arr() is that the
+polynomial must have a non-zero constant term (i.e. the array has `0` as
+its final element).
+
+Internally, callers of BN_GF2m_poly2arr() did not verify that
+precondition, and binary EC curve parameters with an invalid polynomial
+could lead to out of bounds memory reads and writes in BN_GF2m_mod_arr().
+
+The precondition is always true for polynomials that arise from the
+standard form of EC parameters for characteristic-two fields (X9.62).
+See the "Finite Field Identification" section of:
+
+    https://www.itu.int/ITU-T/formal-language/itu-t/x/x894/2018-cor1/ANSI-X9-62.html
+
+The OpenSSL GF(2^m) code supports only the trinomial and pentanomial
+basis X9.62 forms.
+
+This commit updates BN_GF2m_poly2arr() to return `0` (failure) when
+the constant term is zero (i.e. the input bitmask BIGNUM is not odd).
+
+Additionally, the return value is made unambiguous when there is not
+enough space to also pad the array with a final `-1` sentinel value.
+The return value is now always the number of elements (including the
+final `-1`) that would be filled when the output array is sufficiently
+large.  Previously the same count was returned both when the array has
+just enough room for the final `-1` and when it had only enough space
+for non-sentinel values.
+
+Finally, BN_GF2m_poly2arr() is updated to reject polynomials whose
+degree exceeds `OPENSSL_ECC_MAX_FIELD_BITS`, this guards against
+CPU exhausition attacks via excessively large inputs.
+
+The above issues do not arise in processing X.509 certificates.  These
+generally have EC keys from "named curves", and RFC5840 (Section 2.1.1)
+disallows explicit EC parameters.  The TLS code in OpenSSL enforces this
+constraint only after the certificate is decoded, but, even if explicit
+parameters are specified, they are in X9.62 form, which cannot represent
+problem values as noted above.
+
+Initially reported as oss-fuzz issue 71623.
+
+A closely related issue was earlier reported in
+<https://github.com/openssl/openssl/issues/19826>.
+
+Severity: Low, CVE-2024-9143
+
+Reviewed-by: Matt Caswell <matt@openssl.org>
+Reviewed-by: Bernd Edlinger <bernd.edlinger@hotmail.de>
+Reviewed-by: Paul Dale <ppzgs1@gmail.com>
+Reviewed-by: Tomas Mraz <tomas@openssl.org>
+(Merged from https://github.com/openssl/openssl/pull/25639)
+
+(cherry picked from commit 8e008cb8b23ec7dc75c45a66eeed09c815b11cd2)
+--- a/crypto/bn/bn_gf2m.c
++++ b/crypto/bn/bn_gf2m.c
+@@ -15,6 +15,7 @@
+ #include "bn_local.h"
+ 
+ #ifndef OPENSSL_NO_EC2M
++# include <openssl/ec.h>
+ 
+ /*
+  * Maximum number of iterations before BN_GF2m_mod_solve_quad_arr should
+@@ -1140,16 +1141,26 @@ int BN_GF2m_mod_solve_quad(BIGNUM *r, const BIGNUM *a, const BIGNUM *p,
+ /*
+  * Convert the bit-string representation of a polynomial ( \sum_{i=0}^n a_i *
+  * x^i) into an array of integers corresponding to the bits with non-zero
+- * coefficient.  Array is terminated with -1. Up to max elements of the array
+- * will be filled.  Return value is total number of array elements that would
+- * be filled if array was large enough.
++ * coefficient.  The array is intended to be suitable for use with
++ * `BN_GF2m_mod_arr()`, and so the constant term of the polynomial must not be
++ * zero.  This translates to a requirement that the input BIGNUM `a` is odd.
++ *
++ * Given sufficient room, the array is terminated with -1.  Up to max elements
++ * of the array will be filled.
++ *
++ * The return value is total number of array elements that would be filled if
++ * array was large enough, including the terminating `-1`.  It is `0` when `a`
++ * is not odd or the constant term is zero contrary to requirement.
++ *
++ * The return value is also `0` when the leading exponent exceeds
++ * `OPENSSL_ECC_MAX_FIELD_BITS`, this guards against CPU exhaustion attacks,
+  */
+ int BN_GF2m_poly2arr(const BIGNUM *a, int p[], int max)
+ {
+     int i, j, k = 0;
+     BN_ULONG mask;
+ 
+-    if (BN_is_zero(a))
++    if (!BN_is_odd(a))
+         return 0;
+ 
+     for (i = a->top - 1; i >= 0; i--) {
+@@ -1167,12 +1178,13 @@ int BN_GF2m_poly2arr(const BIGNUM *a, int p[], int max)
+         }
+     }
+ 
+-    if (k < max) {
++    if (k > 0 && p[0] > OPENSSL_ECC_MAX_FIELD_BITS)
++        return 0;
++
++    if (k < max)
+         p[k] = -1;
+-        k++;
+-    }
+ 
+-    return k;
++    return k + 1;
+ }
+ 
+ /*
+--- a/test/ec_internal_test.c
++++ b/test/ec_internal_test.c
+@@ -155,6 +155,56 @@ static int field_tests_ecp_mont(void)
+ }
+ 
+ #ifndef OPENSSL_NO_EC2M
++/* Test that decoding of invalid GF2m field parameters fails. */
++static int ec2m_field_sanity(void)
++{
++    int ret = 0;
++    BN_CTX *ctx = BN_CTX_new();
++    BIGNUM *p, *a, *b;
++    EC_GROUP *group1 = NULL, *group2 = NULL, *group3 = NULL;
++
++    TEST_info("Testing GF2m hardening\n");
++
++    BN_CTX_start(ctx);
++    p = BN_CTX_get(ctx);
++    a = BN_CTX_get(ctx);
++    if (!TEST_ptr(b = BN_CTX_get(ctx))
++        || !TEST_true(BN_one(a))
++        || !TEST_true(BN_one(b)))
++        goto out;
++
++    /* Even pentanomial value should be rejected */
++    if (!TEST_true(BN_set_word(p, 0xf2)))
++        goto out;
++    if (!TEST_ptr_null(group1 = EC_GROUP_new_curve_GF2m(p, a, b, ctx)))
++        TEST_error("Zero constant term accepted in GF2m polynomial");
++
++    /* Odd hexanomial should also be rejected */
++    if (!TEST_true(BN_set_word(p, 0xf3)))
++        goto out;
++    if (!TEST_ptr_null(group2 = EC_GROUP_new_curve_GF2m(p, a, b, ctx)))
++        TEST_error("Hexanomial accepted as GF2m polynomial");
++
++    /* Excessive polynomial degree should also be rejected */
++    if (!TEST_true(BN_set_word(p, 0x71))
++        || !TEST_true(BN_set_bit(p, OPENSSL_ECC_MAX_FIELD_BITS + 1)))
++        goto out;
++    if (!TEST_ptr_null(group3 = EC_GROUP_new_curve_GF2m(p, a, b, ctx)))
++        TEST_error("GF2m polynomial degree > %d accepted",
++                   OPENSSL_ECC_MAX_FIELD_BITS);
++
++    ret = group1 == NULL && group2 == NULL && group3 == NULL;
++
++ out:
++    EC_GROUP_free(group1);
++    EC_GROUP_free(group2);
++    EC_GROUP_free(group3);
++    BN_CTX_end(ctx);
++    BN_CTX_free(ctx);
++
++    return ret;
++}
++
+ /* test EC_GF2m_simple_method directly */
+ static int field_tests_ec2_simple(void)
+ {
+@@ -443,6 +493,7 @@ int setup_tests(void)
+     ADD_TEST(field_tests_ecp_simple);
+     ADD_TEST(field_tests_ecp_mont);
+ #ifndef OPENSSL_NO_EC2M
++    ADD_TEST(ec2m_field_sanity);
+     ADD_TEST(field_tests_ec2_simple);
+ #endif
+     ADD_ALL_TESTS(field_tests_default, crv_len);

diff --git a/dev-libs/openssl/openssl-3.1.7-r1.ebuild b/dev-libs/openssl/openssl-3.1.7-r1.ebuild
new file mode 100644
index 000000000000..5ca73111c8f3
--- /dev/null
+++ b/dev-libs/openssl/openssl-3.1.7-r1.ebuild
@@ -0,0 +1,288 @@
+# Copyright 1999-2024 Gentoo Authors
+# Distributed under the terms of the GNU General Public License v2
+
+EAPI=8
+
+VERIFY_SIG_OPENPGP_KEY_PATH=/usr/share/openpgp-keys/openssl.org.asc
+inherit edo flag-o-matic linux-info toolchain-funcs
+inherit multilib multilib-minimal multiprocessing preserve-libs verify-sig
+
+DESCRIPTION="Robust, full-featured Open Source Toolkit for the Transport Layer Security (TLS)"
+HOMEPAGE="https://openssl-library.org/"
+
+MY_P=${P/_/-}
+
+if [[ ${PV} == 9999 ]] ; then
+	EGIT_REPO_URI="https://github.com/openssl/openssl.git"
+
+	inherit git-r3
+else
+	SRC_URI="
+		https://github.com/openssl/openssl/releases/download/${P}/${P}.tar.gz
+		verify-sig? (
+			https://github.com/openssl/openssl/releases/download/${P}/${P}.tar.gz.asc
+		)
+	"
+	KEYWORDS="~alpha ~amd64 ~arm ~arm64 ~hppa ~loong ~m68k ~mips ~ppc ~ppc64 ~riscv ~s390 ~sparc ~x86 ~arm64-macos ~ppc-macos ~x64-macos ~x64-solaris"
+fi
+
+S="${WORKDIR}"/${MY_P}
+
+LICENSE="Apache-2.0"
+SLOT="0/$(ver_cut 1)" # .so version of libssl/libcrypto
+IUSE="+asm cpu_flags_x86_sse2 fips ktls rfc3779 sctp static-libs test tls-compression vanilla verify-sig weak-ssl-ciphers"
+RESTRICT="!test? ( test )"
+
+COMMON_DEPEND="
+	!<net-misc/openssh-9.2_p1-r3
+	tls-compression? ( >=sys-libs/zlib-1.2.8-r1[static-libs(+)?,${MULTILIB_USEDEP}] )
+"
+BDEPEND="
+	>=dev-lang/perl-5
+	sctp? ( >=net-misc/lksctp-tools-1.0.12 )
+	test? (
+		sys-apps/diffutils
+		app-alternatives/bc
+		sys-process/procps
+	)
+	verify-sig? ( >=sec-keys/openpgp-keys-openssl-20240920 )
+"
+
+DEPEND="${COMMON_DEPEND}"
+RDEPEND="${COMMON_DEPEND}"
+PDEPEND="app-misc/ca-certificates"
+
+MULTILIB_WRAPPED_HEADERS=(
+	/usr/include/openssl/configuration.h
+)
+
+PATCHES=(
+	"${FILESDIR}"/${P}-CVE-2024-9143.patch
+)
+
+pkg_setup() {
+	if use ktls ; then
+		if kernel_is -lt 4 18 ; then
+			ewarn "Kernel implementation of TLS (USE=ktls) requires kernel >=4.18!"
+		else
+			CONFIG_CHECK="~TLS ~TLS_DEVICE"
+			ERROR_TLS="You will be unable to offload TLS to kernel because CONFIG_TLS is not set!"
+			ERROR_TLS_DEVICE="You will be unable to offload TLS to kernel because CONFIG_TLS_DEVICE is not set!"
+			use test && CONFIG_CHECK+=" ~CRYPTO_USER_API_SKCIPHER"
+
+			linux-info_pkg_setup
+		fi
+	fi
+
+	[[ ${MERGE_TYPE} == binary ]] && return
+
+	# must check in pkg_setup; sysctl doesn't work with userpriv!
+	if use test && use sctp ; then
+		# test_ssl_new will fail with "Ensure SCTP AUTH chunks are enabled in kernel"
+		# if sctp.auth_enable is not enabled.
+		local sctp_auth_status=$(sysctl -n net.sctp.auth_enable 2>/dev/null)
+		if [[ -z "${sctp_auth_status}" ]] || [[ ${sctp_auth_status} != 1 ]] ; then
+			die "FEATURES=test with USE=sctp requires net.sctp.auth_enable=1!"
+		fi
+	fi
+}
+
+src_prepare() {
+	# Make sure we only ever touch Makefile.org and avoid patching a file
+	# that gets blown away anyways by the Configure script in src_configure
+	rm -f Makefile
+
+	if ! use vanilla ; then
+		PATCHES+=(
+			# Add patches which are Gentoo-specific customisations here
+		)
+	fi
+
+	default
+
+	if use test && use sctp && has network-sandbox ${FEATURES} ; then
+		einfo "Disabling test '80-test_ssl_new.t' which is known to fail with FEATURES=network-sandbox ..."
+		rm test/recipes/80-test_ssl_new.t || die
+	fi
+
+	# Test fails depending on kernel configuration, bug #699134
+	rm test/recipes/30-test_afalg.t || die
+}
+
+src_configure() {
+	# Keep this in sync with app-misc/c_rehash
+	SSL_CNF_DIR="/etc/ssl"
+
+	# Quiet out unknown driver argument warnings since openssl
+	# doesn't have well-split CFLAGS and we're making it even worse
+	# and 'make depend' uses -Werror for added fun (bug #417795 again)
+	tc-is-clang && append-flags -Qunused-arguments
+
+	# We really, really need to build OpenSSL w/ strict aliasing disabled.
+	# It's filled with violations and it *will* result in miscompiled
+	# code. This has been in the ebuild for > 10 years but even in 2022,
+	# it's still relevant:
+	# - https://github.com/llvm/llvm-project/issues/55255
+	# - https://github.com/openssl/openssl/issues/12247
+	# - https://github.com/openssl/openssl/issues/18225
+	# - https://github.com/openssl/openssl/issues/18663#issuecomment-1181478057
+	# Don't remove the no strict aliasing bits below!
+	filter-flags -fstrict-aliasing
+	append-flags -fno-strict-aliasing
+	# The OpenSSL developers don't test with LTO right now, it leads to various
+	# warnings/errors (which may or may not be false positives), it's considered
+	# unsupported, and it's not tested in CI: https://github.com/openssl/openssl/issues/18663.
+	filter-lto
+
+	append-flags $(test-flags-CC -Wa,--noexecstack)
+
+	# bug #895308
+	append-atomic-flags
+	# Configure doesn't respect LIBS
+	export LDLIBS="${LIBS}"
+
+	# bug #197996
+	unset APPS
+	# bug #312551
+	unset SCRIPTS
+	# bug #311473
+	unset CROSS_COMPILE
+
+	tc-export AR CC CXX RANLIB RC
+
+	multilib-minimal_src_configure
+}
+
+multilib_src_configure() {
+	use_ssl() { usex $1 "enable-${2:-$1}" "no-${2:-$1}" " ${*:3}" ; }
+
+	local krb5=$(has_version app-crypt/mit-krb5 && echo "MIT" || echo "Heimdal")
+
+	# See if our toolchain supports __uint128_t.  If so, it's 64bit
+	# friendly and can use the nicely optimized code paths, bug #460790.
+	#local ec_nistp_64_gcc_128
+	#
+	# Disable it for now though (bug #469976)
+	# Do NOT re-enable without substantial discussion first!
+	#
+	#echo "__uint128_t i;" > "${T}"/128.c
+	#if ${CC} ${CFLAGS} -c "${T}"/128.c -o /dev/null >&/dev/null ; then
+	#       ec_nistp_64_gcc_128="enable-ec_nistp_64_gcc_128"
+	#fi
+
+	local sslout=$(bash "${FILESDIR}/gentoo.config-1.0.4")
+	einfo "Using configuration: ${sslout:-(openssl knows best)}"
+
+	# https://github.com/openssl/openssl/blob/master/INSTALL.md#enable-and-disable-features
+	local myeconfargs=(
+		${sslout}
+
+		$(use cpu_flags_x86_sse2 || echo "no-sse2")
+		enable-camellia
+		enable-ec
+		enable-ec2m
+		enable-sm2
+		enable-srp
+		$(use elibc_musl && echo "no-async")
+		enable-idea
+		enable-mdc2
+		enable-rc5
+		$(use fips && echo "enable-fips")
+		$(use_ssl asm)
+		$(use_ssl ktls)
+		$(use_ssl rfc3779)
+		$(use_ssl sctp)
+		$(use test || echo "no-tests")
+		$(use_ssl tls-compression zlib)
+		$(use_ssl weak-ssl-ciphers)
+
+		--prefix="${EPREFIX}"/usr
+		--openssldir="${EPREFIX}"${SSL_CNF_DIR}
+		--libdir=$(get_libdir)
+
+		shared
+		threads
+	)
+
+	edo perl "${S}/Configure" "${myeconfargs[@]}"
+}
+
+multilib_src_compile() {
+	emake build_sw
+
+	if multilib_is_native_abi; then
+		emake build_docs
+	fi
+}
+
+multilib_src_test() {
+	# VFP = show subtests verbosely and show failed tests verbosely
+	# Normal V=1 would show everything verbosely but this slows things down.
+	emake HARNESS_JOBS="$(makeopts_jobs)" -Onone VFP=1 test
+}
+
+multilib_src_install() {
+	# Only -j1 is supported for the install targets:
+	# https://github.com/openssl/openssl/issues/21999#issuecomment-1771150305
+	emake DESTDIR="${D}" -j1 install_sw
+	if use fips; then
+		emake DESTDIR="${D}" -j1 install_fips
+		# Regen this in pkg_preinst, bug 900625
+		rm "${ED}${SSL_CNF_DIR}"/fipsmodule.cnf || die
+	fi
+
+	if multilib_is_native_abi; then
+		emake DESTDIR="${D}" -j1 install_ssldirs
+		emake DESTDIR="${D}" DOCDIR='$(INSTALLTOP)'/share/doc/${PF} -j1 install_docs
+	fi
+
+	# This is crappy in that the static archives are still built even
+	# when USE=static-libs. But this is due to a failing in the openssl
+	# build system: the static archives are built as PIC all the time.
+	# Only way around this would be to manually configure+compile openssl
+	# twice; once with shared lib support enabled and once without.
+	if ! use static-libs ; then
+		rm "${ED}"/usr/$(get_libdir)/lib{crypto,ssl}.a || die
+	fi
+}
+
+multilib_src_install_all() {
+	# openssl installs perl version of c_rehash by default, but
+	# we provide a shell version via app-misc/c_rehash
+	rm "${ED}"/usr/bin/c_rehash || die
+
+	dodoc {AUTHORS,CHANGES,NEWS,README,README-PROVIDERS}.md doc/*.txt doc/${PN}-c-indent.el
+
+	# Create the certs directory
+	keepdir ${SSL_CNF_DIR}/certs
+
+	# bug #254521
+	dodir /etc/sandbox.d
+	echo 'SANDBOX_PREDICT="/dev/crypto"' > "${ED}"/etc/sandbox.d/10openssl
+
+	diropts -m0700
+	keepdir ${SSL_CNF_DIR}/private
+}
+
+pkg_preinst() {
+	if use fips; then
+		# Regen fipsmodule.cnf, bug 900625
+		ebegin "Running openssl fipsinstall"
+		"${ED}/usr/bin/openssl" fipsinstall -quiet \
+			-out "${ED}${SSL_CNF_DIR}/fipsmodule.cnf" \
+			-module "${ED}/usr/$(get_libdir)/ossl-modules/fips.so"
+		eend $?
+	fi
+
+	preserve_old_lib /usr/$(get_libdir)/lib{crypto,ssl}$(get_libname 1) \
+		/usr/$(get_libdir)/lib{crypto,ssl}$(get_libname 1.1)
+}
+
+pkg_postinst() {
+	ebegin "Running 'openssl rehash ${EROOT}${SSL_CNF_DIR}/certs' to rebuild hashes (bug #333069)"
+	openssl rehash "${EROOT}${SSL_CNF_DIR}/certs"
+	eend $?
+
+	preserve_old_lib_notify /usr/$(get_libdir)/lib{crypto,ssl}$(get_libname 1) \
+		/usr/$(get_libdir)/lib{crypto,ssl}$(get_libname 1.1)
+}


             reply	other threads:[~2024-10-29  2:10 UTC|newest]

Thread overview: 36+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-10-29  2:10 Sam James [this message]
  -- strict thread matches above, loose matches on Subject: below --
2024-10-29  2:10 [gentoo-commits] repo/gentoo:master commit in: dev-libs/openssl/, dev-libs/openssl/files/ Sam James
2024-08-07  2:41 Sam James
2024-08-03  5:08 Sam James
2024-07-19 17:03 Jakov Smolić
2024-04-29 17:07 Sam James
2024-04-15  7:16 Sam James
2024-04-15  7:16 Sam James
2024-02-01 16:46 Sam James
2023-07-19 15:06 Sam James
2023-04-20 16:58 Patrick McLean
2023-03-22 23:00 Patrick McLean
2022-12-18  5:00 Andreas K. Hüttel
2022-12-18  2:11 Andreas K. Hüttel
2020-12-22 22:44 Thomas Deutschmann
2020-12-17 16:23 Andreas K. Hüttel
2019-11-25  0:13 Thomas Deutschmann
2019-10-04 15:48 Thomas Deutschmann
2019-10-04 15:48 Thomas Deutschmann
2019-08-23 18:10 Thomas Deutschmann
2019-01-02 21:58 Thomas Deutschmann
2018-12-25 17:24 Lars Wendler
2018-11-20 14:46 Lars Wendler
2018-11-12 18:36 Thomas Deutschmann
2018-06-20 15:38 Lars Wendler
2018-06-12 10:40 Thomas Deutschmann
2017-12-30 19:55 Thomas Deutschmann
2017-12-07 18:53 Thomas Deutschmann
2016-09-22 13:15 Lars Wendler
2016-05-03 14:21 Lars Wendler
2016-02-26 22:46 Doug Goldstein
2016-02-09 13:32 Jason Donenfeld
2016-01-29  6:59 Lars Wendler
2015-12-03 19:58 Mike Frysinger
2015-12-03 18:40 Lars Wendler
2015-12-03 17:10 Mike Frysinger

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=1730167762.536e382d480933cfc6721f129368a8468ebd2321.sam@gentoo \
    --to=sam@gentoo.org \
    --cc=gentoo-commits@lists.gentoo.org \
    --cc=gentoo-dev@lists.gentoo.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