From: Kees Cook <keescook@chromium.org>
To: linux-kernel@vger.kernel.org
Cc: Kees Cook <keescook@chromium.org>,
Thadeu Lima de Souza Cascardo <cascardo@canonical.com>,
Max Filippov <jcmvbkbc@gmail.com>,
Michael Ellerman <mpe@ellerman.id.au>,
Christian Brauner <christian@brauner.io>,
Andy Lutomirski <luto@amacapital.net>,
Will Drewry <wad@chromium.org>,
linux-kselftest@vger.kernel.org, linux-mips@vger.kernel.org,
linux-xtensa@linux-xtensa.org,
linux-arm-kernel@lists.infradead.org,
linuxppc-dev@lists.ozlabs.org
Subject: [PATCH 11/15] selftests/seccomp: Remove SYSCALL_NUM_RET_SHARE_REG in favor of SYSCALL_RET_SET
Date: Sat, 12 Sep 2020 04:08:16 -0700 [thread overview]
Message-ID: <20200912110820.597135-12-keescook@chromium.org> (raw)
In-Reply-To: <20200912110820.597135-1-keescook@chromium.org>
Instead of special-casing the specific case of shared registers, create
a default SYSCALL_RET_SET() macro (mirroring SYSCALL_NUM_SET()), that
writes to the SYSCALL_RET register. For architectures that can't set the
return value (for whatever reason), they can define SYSCALL_RET_SET()
without an associated SYSCALL_RET() macro. This also paves the way for
architectures that need to do special things to set the return value
(e.g. powerpc).
Signed-off-by: Kees Cook <keescook@chromium.org>
---
tools/testing/selftests/seccomp/seccomp_bpf.c | 33 +++++++++++++------
1 file changed, 23 insertions(+), 10 deletions(-)
diff --git a/tools/testing/selftests/seccomp/seccomp_bpf.c b/tools/testing/selftests/seccomp/seccomp_bpf.c
index 2790d9cd50f4..623953a53032 100644
--- a/tools/testing/selftests/seccomp/seccomp_bpf.c
+++ b/tools/testing/selftests/seccomp/seccomp_bpf.c
@@ -1753,8 +1753,8 @@ TEST_F(TRACE_poke, getpid_runs_normally)
#elif defined(__s390__)
# define ARCH_REGS s390_regs
# define SYSCALL_NUM(_regs) (_regs).gprs[2]
-# define SYSCALL_RET(_regs) (_regs).gprs[2]
-# define SYSCALL_NUM_RET_SHARE_REG
+# define SYSCALL_RET_SET(_regs, _val) \
+ TH_LOG("Can't modify syscall return on this architecture")
#elif defined(__mips__)
# include <asm/unistd_nr_n32.h>
# include <asm/unistd_nr_n64.h>
@@ -1776,8 +1776,8 @@ TEST_F(TRACE_poke, getpid_runs_normally)
else \
(_regs).regs[2] = _nr; \
} while (0)
-# define SYSCALL_RET(_regs) (_regs).regs[2]
-# define SYSCALL_NUM_RET_SHARE_REG
+# define SYSCALL_RET_SET(_regs, _val) \
+ TH_LOG("Can't modify syscall return on this architecture")
#elif defined(__xtensa__)
# define ARCH_REGS struct user_pt_regs
# define SYSCALL_NUM(_regs) (_regs).syscall
@@ -1804,9 +1804,26 @@ TEST_F(TRACE_poke, getpid_runs_normally)
SYSCALL_NUM(_regs) = (_nr); \
} while (0)
#endif
+/*
+ * Most architectures can change the syscall return value by just
+ * writing to the SYSCALL_RET register. This is the default if not
+ * defined above. If an architecture cannot set the return value
+ * (for example when the syscall and return value register is
+ * shared), report it with TH_LOG() in an arch-specific definition
+ * of SYSCALL_RET_SET() above, and leave SYSCALL_RET undefined.
+ */
+#if !defined(SYSCALL_RET) && !defined(SYSCALL_RET_SET)
+# error "One of SYSCALL_RET or SYSCALL_RET_SET is needed for this arch"
+#endif
+#ifndef SYSCALL_RET_SET
+# define SYSCALL_RET_SET(_regs, _val) \
+ do { \
+ SYSCALL_RET(_regs) = (_val); \
+ } while (0)
+#endif
/* When the syscall return can't be changed, stub out the tests for it. */
-#ifdef SYSCALL_NUM_RET_SHARE_REG
+#ifndef SYSCALL_RET
# define EXPECT_SYSCALL_RETURN(val, action) EXPECT_EQ(-1, action)
#else
# define EXPECT_SYSCALL_RETURN(val, action) \
@@ -1870,11 +1887,7 @@ void change_syscall(struct __test_metadata *_metadata,
/* If syscall is skipped, change return value. */
if (syscall == -1)
-#ifdef SYSCALL_NUM_RET_SHARE_REG
- TH_LOG("Can't modify syscall return on this architecture");
-#else
- SYSCALL_RET(regs) = result;
-#endif
+ SYSCALL_RET_SET(regs, result);
/* Flush any register changes made. */
if (memcmp(&orig, ®s, sizeof(orig)) != 0)
--
2.25.1
next prev parent reply other threads:[~2020-09-12 11:10 UTC|newest]
Thread overview: 39+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-09-12 11:08 [PATCH 00/15] selftests/seccomp: Refactor change_syscall() Kees Cook
2020-09-12 11:08 ` [PATCH 01/15] selftests/seccomp: Refactor arch register macros to avoid xtensa special case Kees Cook
2020-09-15 15:51 ` Christian Brauner
2020-09-12 11:08 ` [PATCH 02/15] selftests/seccomp: Provide generic syscall setting macro Kees Cook
2020-09-15 15:53 ` Christian Brauner
2020-09-12 11:08 ` [PATCH 03/15] selftests/seccomp: mips: Define SYSCALL_NUM_SET macro Kees Cook
2020-09-15 15:55 ` Christian Brauner
2020-09-18 22:00 ` Kees Cook
2020-09-12 11:08 ` [PATCH 04/15] selftests/seccomp: arm: " Kees Cook
2020-09-15 15:56 ` Christian Brauner
2020-09-12 11:08 ` [PATCH 05/15] selftests/seccomp: arm64: " Kees Cook
2020-09-15 15:58 ` Christian Brauner
2020-09-12 11:08 ` [PATCH 06/15] selftests/seccomp: mips: Remove O32-specific macro Kees Cook
2020-09-15 16:00 ` Christian Brauner
2020-09-12 11:08 ` [PATCH 07/15] selftests/seccomp: Remove syscall setting #ifdefs Kees Cook
2020-09-15 16:01 ` Christian Brauner
2020-09-12 11:08 ` [PATCH 08/15] selftests/seccomp: Convert HAVE_GETREG into ARCH_GETREG/ARCH_SETREG Kees Cook
2020-09-15 16:03 ` Christian Brauner
2020-09-12 11:08 ` [PATCH 09/15] selftests/seccomp: Convert REGSET calls " Kees Cook
2020-09-15 16:05 ` Christian Brauner
2020-09-12 11:08 ` [PATCH 10/15] selftests/seccomp: Avoid redundant register flushes Kees Cook
2020-09-15 16:08 ` Christian Brauner
2020-09-12 11:08 ` Kees Cook [this message]
2020-09-15 16:11 ` [PATCH 11/15] selftests/seccomp: Remove SYSCALL_NUM_RET_SHARE_REG in favor of SYSCALL_RET_SET Christian Brauner
2020-09-12 11:08 ` [PATCH 12/15] selftests/seccomp: powerpc: Fix seccomp return value testing Kees Cook
2020-09-14 3:38 ` Michael Ellerman
2020-09-12 11:08 ` [PATCH 13/15] selftests/seccomp: powerpc: Set syscall return during ptrace syscall exit Kees Cook
2020-09-14 5:47 ` Michael Ellerman
2020-09-14 20:20 ` Kees Cook
2020-09-12 11:08 ` [PATCH 14/15] selftests/clone3: Avoid OS-defined clone_args Kees Cook
2020-09-15 16:25 ` Christian Brauner
2020-09-18 22:00 ` Kees Cook
2020-09-12 11:08 ` [PATCH 15/15] selftests/seccomp: Use __NR_mknodat instead of __NR_mknod Kees Cook
2020-09-15 16:16 ` Christian Brauner
2020-09-14 12:15 ` [PATCH 00/15] selftests/seccomp: Refactor change_syscall() Michael Ellerman
2020-09-14 20:32 ` Kees Cook
2020-09-15 11:12 ` Max Filippov
2020-09-15 12:52 ` Michael Ellerman
2020-09-15 8:45 ` Max Filippov
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=20200912110820.597135-12-keescook@chromium.org \
--to=keescook@chromium.org \
--cc=cascardo@canonical.com \
--cc=christian@brauner.io \
--cc=jcmvbkbc@gmail.com \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-kselftest@vger.kernel.org \
--cc=linux-mips@vger.kernel.org \
--cc=linux-xtensa@linux-xtensa.org \
--cc=linuxppc-dev@lists.ozlabs.org \
--cc=luto@amacapital.net \
--cc=mpe@ellerman.id.au \
--cc=wad@chromium.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;
as well as URLs for NNTP newsgroup(s).