Hi,
On Tue, 2025-01-14 at 20:30 +0900, Hajime Tazaki wrote:
> This patchset is another spin of nommu mode addition to UML. It doesn't
> change a lot since the last version (v5), but contain clean ups. It would
> be nice to hear about your opinions on that.
>
> There are still several limitations/issues which we already found;
> here is the list of those issues.
>
> - memory mapped by loadable modules are not distinguished from
> userspace memory.
Maybe I am missing it, but I do not yet see proper FP register
handling. This will be needed for task/thread switches and also signal
emission/sigreturn. I am attaching the test program that I used to
verify the correct behaviour when dealing with the recent changes to FP
register handling in UML.
Benjamin
>
> -- Hajime
>
> v6:
> - rebase to the latest uml/next tree
> - more clean up on mmu/nommu for signal handling [10/13]
> - rename functions of mcontext routines [06,10/13]
> - added Acked-by tag for binfmt_elf_fdpic [02/13]
>
> v5:
> - clean up stack manipulation code [05,06,07,10/13]
> -
> https://lore.kernel.org/linux-um/cover.1733998168.git.thehaj...@gmail.com/
>
> v4:
> - add arch/um/nommu, arch/x86/um/nommu to contain !MMU specific codes
> - remove zpoline patch
> - drop binfmt_elf_fdpic patch
> - reduce ifndef CONFIG_MMU if possible
> - split to elf header cleanup patch [01/13]
> - fix kernel test robot warnings [06/13]
> - fix coding styles [07/13]
> - move task_top_of_stack definition [05/13]
> -
> https://lore.kernel.org/linux-um/cover.1733652929.git.thehaj...@gmail.com/
>
> v3:
> -
> https://lore.kernel.org/linux-um/cover.1733199769.git.thehaj...@gmail.com/
> - add seccomp-based syscall hook in addition to zpoline [06/13]
> - remove RFC, add a line to MAINTAINERS file
> - fix kernel test robot warnings [02/13,08/13,10/13]
> - add base-commit tag to cover letter
> - pull the latest uml/next
> - clean up SIGSEGV handling [10/13]
> - detect fsgsbase availability with elf aux vector [08/13]
> - simplify vdso code with macros [09/13]
>
> RFC v2:
> -
> https://lore.kernel.org/linux-um/cover.1731290567.git.thehaj...@gmail.com/
> - base branch is now uml/linux.git instead of torvalds/linux.git.
> - reorganize the patch series to clean up
> - fixed various coding styles issues
> - clean up exec code path [07/13]
> - fixed the crash/SIGSEGV case on userspace programs [10/13]
> - add seccomp filter to limit syscall caller address [06/13]
> - detect fsgsbase availability with sigsetjmp/siglongjmp [08/13]
> - removes unrelated changes
> - removes unneeded ifndef CONFIG_MMU
> - convert UML_CONFIG_MMU to CONFIG_MMU as using uml/linux.git
> - proposed a patch of maple-tree issue (resolving a limitation in RFC
> v1)
>
> https://lore.kernel.org/linux-mm/20241108222834.3625217-1-thehaj...@gmail.com/
>
> RFC:
> -
> https://lore.kernel.org/linux-um/cover.1729770373.git.thehaj...@gmail.com/
>
> Hajime Tazaki (13):
> x86/um: clean up elf specific definitions
> x86/um: nommu: elf loader for fdpic
> um: decouple MMU specific code from the common part
> um: nommu: memory handling
> x86/um: nommu: syscall handling
> um: nommu: seccomp syscalls hook
> x86/um: nommu: process/thread handling
> um: nommu: configure fs register on host syscall invocation
> x86/um/vdso: nommu: vdso memory update
> x86/um: nommu: signal handling
> um: change machine name for uname output
> um: nommu: add documentation of nommu UML
> um: nommu: plug nommu code into build system
>
> Documentation/virt/uml/nommu-uml.rst | 177 ++++++++++++++++++++++
> MAINTAINERS | 1 +
> arch/um/Kconfig | 14 +-
> arch/um/Makefile | 10 ++
> arch/um/configs/x86_64_nommu_defconfig | 64 ++++++++
> arch/um/include/asm/Kbuild | 1 +
> arch/um/include/asm/futex.h | 4 +
> arch/um/include/asm/mmu.h | 8 +
> arch/um/include/asm/mmu_context.h | 2 +
> arch/um/include/asm/ptrace-generic.h | 6 +
> arch/um/include/asm/uaccess.h | 7 +-
> arch/um/include/shared/kern_util.h | 12 ++
> arch/um/include/shared/os.h | 16 ++
> arch/um/kernel/Makefile | 5 +-
> arch/um/kernel/mem-pgtable.c | 55 +++++++
> arch/um/kernel/mem.c | 39 +----
> arch/um/kernel/process.c | 25 ++++
> arch/um/kernel/skas/process.c | 27 ----
> arch/um/kernel/um_arch.c | 3 +
> arch/um/nommu/Makefile | 3 +
> arch/um/nommu/os-Linux/Makefile | 7 +
> arch/um/nommu/os-Linux/signal.c | 28 ++++
> arch/um/nommu/trap.c | 188
> ++++++++++++++++++++++++
> arch/um/os-Linux/Makefile | 8 +-
> arch/um/os-Linux/internal.h | 5 +
> arch/um/os-Linux/mem.c | 4 +
> arch/um/os-Linux/process.c | 149 ++++++++++++++++++-
> arch/um/os-Linux/seccomp.c | 87 +++++++++++
> arch/um/os-Linux/signal.c | 31 +++-
> arch/um/os-Linux/skas/process.c | 132 -----------------
> arch/um/os-Linux/start_up.c | 20 +++
> arch/um/os-Linux/util.c | 3 +-
> arch/x86/um/Makefile | 7 +-
> arch/x86/um/asm/elf.h | 8 +-
> arch/x86/um/asm/module.h | 24 ---
> arch/x86/um/nommu/Makefile | 8 +
> arch/x86/um/nommu/do_syscall_64.c | 74 ++++++++++
> arch/x86/um/nommu/entry_64.S | 113 ++++++++++++++
> arch/x86/um/nommu/os-Linux/Makefile | 6 +
> arch/x86/um/nommu/os-Linux/mcontext.c | 24 +++
> arch/x86/um/nommu/syscalls.h | 16 ++
> arch/x86/um/nommu/syscalls_64.c | 115 +++++++++++++++
> arch/x86/um/shared/sysdep/mcontext.h | 4 +
> arch/x86/um/shared/sysdep/syscalls_64.h | 6 +
> arch/x86/um/signal.c | 7 +
> arch/x86/um/vdso/vma.c | 17 ++-
> fs/Kconfig.binfmt | 2 +-
> 47 files changed, 1328 insertions(+), 244 deletions(-)
> create mode 100644 Documentation/virt/uml/nommu-uml.rst
> create mode 100644 arch/um/configs/x86_64_nommu_defconfig
> create mode 100644 arch/um/kernel/mem-pgtable.c
> create mode 100644 arch/um/nommu/Makefile
> create mode 100644 arch/um/nommu/os-Linux/Makefile
> create mode 100644 arch/um/nommu/os-Linux/signal.c
> create mode 100644 arch/um/nommu/trap.c
> create mode 100644 arch/um/os-Linux/seccomp.c
> delete mode 100644 arch/x86/um/asm/module.h
> create mode 100644 arch/x86/um/nommu/Makefile
> create mode 100644 arch/x86/um/nommu/do_syscall_64.c
> create mode 100644 arch/x86/um/nommu/entry_64.S
> create mode 100644 arch/x86/um/nommu/os-Linux/Makefile
> create mode 100644 arch/x86/um/nommu/os-Linux/mcontext.c
> create mode 100644 arch/x86/um/nommu/syscalls.h
> create mode 100644 arch/x86/um/nommu/syscalls_64.c
>
>
> base-commit: 2d2b61ae38bd91217ea7cc5bc700a2b9e75b3937
/*
* gcc test-signal-restore.c -o test-signal-restore-amd64
* gcc -m32 -march=i686 -lm test-signal-restore.c -o test-signal-restore-i386
*/
/* Is there a better way to *not* include bits/sigcontext.h? */
#include <features.h>
#undef __USE_MISC
#include <asm/sigcontext.h>
#include <elf.h>
#include <math.h>
#include <stdio.h>
#include <signal.h>
#include <stdlib.h>
#include <sys/mman.h>
#include <sys/wait.h>
#include <errno.h>
#include <unistd.h>
#include <sys/ptrace.h>
#include <sys/user.h>
#include <sys/uio.h>
#include <asm/unistd.h>
#define ST0_EXP_ADD 10
void *scratch_page;
void sighandler(int sig, siginfo_t *info, void *p)
{
ucontext_t *uc = p;
printf("sighandler: extended_size: %d, xstate_size: %d\n",
((struct _fpstate *)uc->uc_mcontext.__fpregs)->sw_reserved.extended_size,
((struct _fpstate *)uc->uc_mcontext.__fpregs)->sw_reserved.xstate_size);
uc->uc_mcontext.__fpregs->_st[0].__exponent += ST0_EXP_ADD;
}
int test_fp()
{
double num = 0.5;
long ret;
printf("pre-signal: %g\n", num);
/*
* This does kill(getpid(), SIGUSR1); with "num" being passed in AND
* out of the floating point stack. We can therefore modify num by
* changing st[0] when handling the signal.
*/
#ifdef __i386__
asm volatile (
"int $0x80;"
: "=t" (num), "=a" (ret)
: "0" (num), "1" (__NR_kill), "b" (getpid()), "c" (SIGUSR1) : );
#else
asm volatile (
"syscall;"
: "=t" (num), "=a" (ret)
: "0" (num), "1" (__NR_kill), "D" (getpid()), "S" (SIGUSR1) : "r11", "rcx");
#endif
printf("post-signal: %g\n", num);
if (num != pow(2, ST0_EXP_ADD - 1)) {
printf("floating point register was not manipulated\n");
return 1;
}
return 0;
}
enum source {
S_FPREGS = 0,
S_FPXREGS = 1,
S_GETREGS_FPREGS = 2,
S_GETREGS_XFPREGS = 3,
S_GETREGS_XSTATE = 4,
};
int test_fp_ptrace(enum source source)
{
int pid, status, ret;
pid = fork();
if (pid < 0)
return 127;
if (pid == 0) {
/* child */
ptrace(PTRACE_TRACEME, 0, 0, 0);
kill(getpid(), SIGSTOP);
if (test_fp())
exit(1);
exit(0);
}
/* Wait for child to stop itself */
do {
ret = waitpid(pid, &status, 0);
} while (ret < 0 && errno == EINTR);
if (!WIFSTOPPED(status))
return 127;
/* Continue until SIGUSR1 to self */
ptrace(PTRACE_CONT, pid, NULL, 0);
do {
ret = waitpid(pid, &status, 0);
} while (ret < 0 && errno == EINTR);
if (!WIFSTOPPED(status))
return 127;
if (source == S_FPXREGS || source == S_GETREGS_XFPREGS) {
#ifdef __i386__
struct user_fpxregs_struct *fpstate;
struct iovec iov = {
.iov_len = sizeof(*fpstate),
};
int ret;
fpstate = scratch_page + 4096 - iov.iov_len;
iov.iov_base = fpstate;
if (source == S_GETREGS_XFPREGS)
ret = ptrace(PTRACE_GETREGSET, pid, NT_PRXFPREG, &iov);
else
ret = ptrace(PTRACE_GETFPXREGS, pid, NULL, fpstate);
if (ret) {
kill(pid, SIGKILL);
if (errno == EINVAL) {
printf("Getting FPX regs not supported\n");
return 0;
} else {
printf("Error getting FPX regs: %d\n", errno);
return 127;
}
}
((struct _fpxreg*)&fpstate->st_space[0])->exponent += ST0_EXP_ADD;
if (source == S_GETREGS_XFPREGS)
ret = ptrace(PTRACE_SETREGSET, pid, NT_PRXFPREG, &iov);
else
ret = ptrace(PTRACE_SETFPXREGS, pid, NULL, fpstate);
if (ret)
return -127;
#else
printf("No FPXREGS on x86_64\n");
kill(pid, SIGKILL);
return 127;
#endif
} else if (source == S_FPREGS || source == S_GETREGS_FPREGS) {
struct _fpstate *fpstate;
struct iovec iov = {
.iov_len = sizeof(*fpstate),
};
fpstate = scratch_page; // + 4096 - sizeof(*fpstate);
iov.iov_base = fpstate;
if (source == S_GETREGS_FPREGS)
ret = ptrace(PTRACE_GETREGSET, pid, NT_PRFPREG, &iov);
else
ret = ptrace(PTRACE_GETFPREGS, pid, NULL, fpstate);
if (ret) {
kill(pid, SIGKILL);
if (errno == EINVAL) {
printf("Getting FP regs not supported\n");
return 0;
} else {
printf("Error getting FPX regs: %d\n", errno);
return 127;
}
}
#ifdef __i386__
((struct _fpreg*) &fpstate->_st[0])->exponent += ST0_EXP_ADD;
#else
((struct _fpxreg*) &fpstate->st_space[0])->exponent += ST0_EXP_ADD;
#endif
if (source == S_GETREGS_FPREGS)
ret = ptrace(PTRACE_SETREGSET, pid, NT_PRFPREG, &iov);
else
ret = ptrace(PTRACE_SETFPREGS, pid, NULL, fpstate);
if (ret)
return 127;
} else if (source == S_GETREGS_XSTATE) {
#ifdef __i386__
struct user_fpxregs_struct *fpstate;
#else
struct user_fpregs_struct *fpstate;
#endif
struct iovec iov = {
.iov_len = 4096,
};
fpstate = scratch_page + 4096 - iov.iov_len;
iov.iov_base = fpstate;
ret = ptrace(PTRACE_GETREGSET, pid, NT_X86_XSTATE, &iov);
if (ret) {
kill(pid, SIGKILL);
if (errno == EINVAL) {
printf("Getting XSTATE not supported\n");
return 0;
} else {
printf("Error getting XSTATE size: %d\n", errno);
return 127;
}
}
printf("host xstate size: %ld\n", iov.iov_len);
/* Second time with the exact length (to test the kernel) */
fpstate = scratch_page + 4096 - iov.iov_len;
iov.iov_base = fpstate;
ret = ptrace(PTRACE_GETREGSET, pid, NT_X86_XSTATE, &iov);
if (ret) {
printf("Error getting XSTATE: %d\n", errno);
return 127;
}
fpstate = scratch_page + 4096 - iov.iov_len;
iov.iov_base = fpstate;
ret = ptrace(PTRACE_GETREGSET, pid, NT_X86_XSTATE, &iov);
if (ret) {
kill(pid, SIGKILL);
printf("Error getting XSTATE (with correct size): %d\n", errno);
return 127;
}
#ifdef __i386__
((struct _fpxreg *)&fpstate->st_space[0])->exponent += ST0_EXP_ADD;
#else
((struct _fpxreg *)&fpstate->st_space[0])->exponent += ST0_EXP_ADD;
#endif
ret = ptrace(PTRACE_SETREGSET, pid, NT_X86_XSTATE, &iov);
if (ret) {
printf("Failed to set XSTATE: %d\n", errno);
return 127;
}
} else {
return 127;
}
/* Run until completion (without handling the signal) */
ptrace(PTRACE_CONT, pid, NULL, 0);
do {
ret = waitpid(pid, &status, 0);
} while (ret < 0 && errno == EINTR);
if (!WIFEXITED(status))
return 127;
return WEXITSTATUS(status);
}
int main()
{
struct sigaction sa = {
.sa_flags = SA_SIGINFO,
.sa_handler = (void (*)(int))sighandler,
};
int ret;
scratch_page = mmap(NULL, 8192, PROT_READ | PROT_WRITE, MAP_PRIVATE | MAP_ANONYMOUS, -1, 0);
munmap(scratch_page + 4096, 4096);
sigaction(SIGUSR1, &sa, NULL);
if (test_fp())
return 1;
sa.sa_handler = SIG_DFL;
sigaction(SIGUSR1, &sa, NULL);
printf("\nmodify using ptrace PTRACE_SETFPREGS instead of sighandler:\n");
ret = test_fp_ptrace(S_FPREGS);
if (ret)
return ret;
#ifdef __i386__
printf("\nmodify using ptrace PTRACE_SETFPXREGS instead of sighandler:\n");
ret = test_fp_ptrace(S_FPXREGS);
if (ret)
return ret;
#endif
printf("\nmodify using ptrace PTRACE_SETREGSET, via NT_PRFPREG instead of sighandler:\n");
ret = test_fp_ptrace(S_GETREGS_FPREGS);
if (ret)
return ret;
#ifdef __i386__
printf("\nmodify using ptrace PTRACE_SETREGSET, via NT_XFPREGS instead of sighandler:\n");
ret = test_fp_ptrace(S_GETREGS_XFPREGS);
if (ret)
return ret;
#endif
printf("\nmodify using ptrace PTRACE_SETREGSET, via NT_X86_XSTATE instead of sighandler:\n");
ret = test_fp_ptrace(S_GETREGS_XSTATE);
if (ret)
return ret;
return 0;
}
