um: Add stub side of SECCOMP/futex based process handling

This adds the stub side for the new seccomp process management code. In
this case we do register save/restore through the signal handler
mcontext.

Add special code for handling TLS, which for x86_64 means setting the
FS_BASE/GS_BASE registers while for i386 it means calling the
set_thread_area syscall.

Co-authored-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: Benjamin Berg <benjamin@sipsolutions.net>
Signed-off-by: Benjamin Berg <benjamin.berg@intel.com>
Link: https://patch.msgid.link/20250602130052.545733-3-benjamin@sipsolutions.net
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

arch/um/include/shared/common-offsets.h

@@ -14,3 +14,5 @@ DEFINE(UM_THREAD_SIZE, THREAD_SIZE);
 
 DEFINE(UM_NSEC_PER_SEC, NSEC_PER_SEC);
 DEFINE(UM_NSEC_PER_USEC, NSEC_PER_USEC);
+
+DEFINE(UM_KERN_GDT_ENTRY_TLS_ENTRIES, GDT_ENTRY_TLS_ENTRIES);

arch/um/include/shared/skas/stub-data.h

@@ -11,6 +11,10 @@
 #include <linux/compiler_types.h>
 #include <as-layout.h>
 #include <sysdep/tls.h>
+#include <sysdep/stub-data.h>
+
+#define FUTEX_IN_CHILD 0
+#define FUTEX_IN_KERN 1
 
 struct stub_init_data {
 	unsigned long stub_start;
@@ -52,6 +56,16 @@ struct stub_data {
 	/* 128 leaves enough room for additional fields in the struct */
 	struct stub_syscall syscall_data[(UM_KERN_PAGE_SIZE - 128) / sizeof(struct stub_syscall)] __aligned(16);
 
+	/* data shared with signal handler (only used in seccomp mode) */
+	short restart_wait;
+	unsigned int futex;
+	int signal;
+	unsigned short si_offset;
+	unsigned short mctx_offset;
+
+	/* seccomp architecture specific state restore */
+	struct stub_data_arch arch_data;
+
 	/* Stack for our signal handlers and for calling into . */
 	unsigned char sigstack[UM_KERN_PAGE_SIZE] __aligned(UM_KERN_PAGE_SIZE);
 };

arch/um/kernel/skas/stub.c

@@ -5,6 +5,9 @@
 
 #include <sysdep/stub.h>
 
+#include <linux/futex.h>
+#include <errno.h>
+
 static __always_inline int syscall_handler(struct stub_data *d)
 {
 	int i;
@@ -57,3 +60,49 @@ stub_syscall_handler(void)
 
 	trap_myself();
 }
+
+void __section(".__syscall_stub")
+stub_signal_interrupt(int sig, siginfo_t *info, void *p)
+{
+	struct stub_data *d = get_stub_data();
+	ucontext_t *uc = p;
+	long res;
+
+	d->signal = sig;
+	d->si_offset = (unsigned long)info - (unsigned long)&d->sigstack[0];
+	d->mctx_offset = (unsigned long)&uc->uc_mcontext - (unsigned long)&d->sigstack[0];
+
+restart_wait:
+	d->futex = FUTEX_IN_KERN;
+	do {
+		res = stub_syscall3(__NR_futex, (unsigned long)&d->futex,
+				    FUTEX_WAKE, 1);
+	} while (res == -EINTR);
+	do {
+		res = stub_syscall4(__NR_futex, (unsigned long)&d->futex,
+				    FUTEX_WAIT, FUTEX_IN_KERN, 0);
+	} while (res == -EINTR || d->futex == FUTEX_IN_KERN);
+
+	if (res < 0 && res != -EAGAIN)
+		stub_syscall1(__NR_exit_group, 1);
+
+	/* Try running queued syscalls. */
+	if (syscall_handler(d) < 0 || d->restart_wait) {
+		/* Report SIGSYS if we restart. */
+		d->signal = SIGSYS;
+		d->restart_wait = 0;
+		goto restart_wait;
+	}
+
+	/* Restore arch dependent state that is not part of the mcontext */
+	stub_seccomp_restore_state(&d->arch_data);
+
+	/* Return so that the host modified mcontext is restored. */
+}
+
+void __section(".__syscall_stub")
+stub_signal_restorer(void)
+{
+	/* We must not have anything on the stack when doing rt_sigreturn */
+	stub_syscall0(__NR_rt_sigreturn);
+}

arch/x86/um/shared/sysdep/stub-data.h

@@ -0,0 +1,23 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef __ARCH_STUB_DATA_H
+#define __ARCH_STUB_DATA_H
+
+#ifdef __i386__
+#include <generated/asm-offsets.h>
+#include <asm/ldt.h>
+
+struct stub_data_arch {
+	int sync;
+	struct user_desc tls[UM_KERN_GDT_ENTRY_TLS_ENTRIES];
+};
+#else
+#define STUB_SYNC_FS_BASE (1 << 0)
+#define STUB_SYNC_GS_BASE (1 << 1)
+struct stub_data_arch {
+	int sync;
+	unsigned long fs_base;
+	unsigned long gs_base;
+};
+#endif
+
+#endif /* __ARCH_STUB_DATA_H */

arch/x86/um/shared/sysdep/stub.h

@@ -13,3 +13,5 @@
 
 extern void stub_segv_handler(int, siginfo_t *, void *);
 extern void stub_syscall_handler(void);
+extern void stub_signal_interrupt(int, siginfo_t *, void *);
+extern void stub_signal_restorer(void);

arch/x86/um/shared/sysdep/stub_32.h

@@ -131,4 +131,17 @@ static __always_inline void *get_stub_data(void)
 		"call *%%eax ;"						\
 		:: "i" ((1 + STUB_DATA_PAGES) * UM_KERN_PAGE_SIZE),	\
 		   "i" (&fn))
+
+static __always_inline void
+stub_seccomp_restore_state(struct stub_data_arch *arch)
+{
+	for (int i = 0; i < sizeof(arch->tls) / sizeof(arch->tls[0]); i++) {
+		if (arch->sync & (1 << i))
+			stub_syscall1(__NR_set_thread_area,
+				      (unsigned long) &arch->tls[i]);
+	}
+
+	arch->sync = 0;
+}
+
 #endif

arch/x86/um/shared/sysdep/stub_64.h

@@ -10,6 +10,7 @@
 #include <sysdep/ptrace_user.h>
 #include <generated/asm-offsets.h>
 #include <linux/stddef.h>
+#include <asm/prctl.h>
 
 #define STUB_MMAP_NR __NR_mmap
 #define MMAP_OFFSET(o) (o)
@@ -134,4 +135,20 @@ static __always_inline void *get_stub_data(void)
 		"call *%%rax ;"						\
 		:: "i" ((1 + STUB_DATA_PAGES) * UM_KERN_PAGE_SIZE),	\
 		   "i" (&fn))
+
+static __always_inline void
+stub_seccomp_restore_state(struct stub_data_arch *arch)
+{
+	/*
+	 * We could use _writefsbase_u64/_writegsbase_u64 if the host reports
+	 * support in the hwcaps (HWCAP2_FSGSBASE).
+	 */
+	if (arch->sync & STUB_SYNC_FS_BASE)
+		stub_syscall2(__NR_arch_prctl, ARCH_SET_FS, arch->fs_base);
+	if (arch->sync & STUB_SYNC_GS_BASE)
+		stub_syscall2(__NR_arch_prctl, ARCH_SET_GS, arch->gs_base);
+
+	arch->sync = 0;
+}
+
 #endif