[6.1] Track PF uksmd patches

arch/alpha/kernel/syscalls/syscall.tbl

@@ -490,3 +490,4 @@
 558	common	process_mrelease		sys_process_mrelease
 559	common  futex_waitv                     sys_futex_waitv
 560	common	set_mempolicy_home_node		sys_ni_syscall
+561	common	pmadv_ksm			sys_pmadv_ksm

arch/arm/tools/syscall.tbl

@@ -464,3 +464,4 @@
 448	common	process_mrelease		sys_process_mrelease
 449	common	futex_waitv			sys_futex_waitv
 450	common	set_mempolicy_home_node		sys_set_mempolicy_home_node
+451	common	pmadv_ksm			sys_pmadv_ksm

arch/arm64/include/asm/unistd.h

@@ -39,7 +39,7 @@
 #define __ARM_NR_compat_set_tls		(__ARM_NR_COMPAT_BASE + 5)
 #define __ARM_NR_COMPAT_END		(__ARM_NR_COMPAT_BASE + 0x800)
 
-#define __NR_compat_syscalls		451
+#define __NR_compat_syscalls		452
 #endif
 
 #define __ARCH_WANT_SYS_CLONE

arch/arm64/include/asm/unistd32.h

@@ -907,6 +907,8 @@ __SYSCALL(__NR_process_mrelease, sys_process_mrelease)
 __SYSCALL(__NR_futex_waitv, sys_futex_waitv)
 #define __NR_set_mempolicy_home_node 450
 __SYSCALL(__NR_set_mempolicy_home_node, sys_set_mempolicy_home_node)
+#define __NR_pmadv_ksm 451
+__SYSCALL(__NR_pmadv_ksm, sys_pmadv_ksm)
 
 /*
  * Please add new compat syscalls above this comment and update

arch/ia64/kernel/syscalls/syscall.tbl

@@ -371,3 +371,4 @@
 448	common	process_mrelease		sys_process_mrelease
 449	common  futex_waitv                     sys_futex_waitv
 450	common	set_mempolicy_home_node		sys_set_mempolicy_home_node
+451	common	pmadv_ksm			sys_pmadv_ksm

arch/m68k/kernel/syscalls/syscall.tbl

@@ -450,3 +450,4 @@
 448	common	process_mrelease		sys_process_mrelease
 449	common  futex_waitv                     sys_futex_waitv
 450	common	set_mempolicy_home_node		sys_set_mempolicy_home_node
+451	common	pmadv_ksm			sys_pmadv_ksm

arch/microblaze/kernel/syscalls/syscall.tbl

@@ -456,3 +456,4 @@
 448	common	process_mrelease		sys_process_mrelease
 449	common  futex_waitv                     sys_futex_waitv
 450	common	set_mempolicy_home_node		sys_set_mempolicy_home_node
+451	common	pmadv_ksm			sys_pmadv_ksm

arch/mips/kernel/syscalls/syscall_n32.tbl

@@ -389,3 +389,4 @@
 448	n32	process_mrelease		sys_process_mrelease
 449	n32	futex_waitv			sys_futex_waitv
 450	n32	set_mempolicy_home_node		sys_set_mempolicy_home_node
+451	n32	pmadv_ksm			sys_pmadv_ksm

arch/mips/kernel/syscalls/syscall_n64.tbl

@@ -365,3 +365,4 @@
 448	n64	process_mrelease		sys_process_mrelease
 449	n64	futex_waitv			sys_futex_waitv
 450	common	set_mempolicy_home_node		sys_set_mempolicy_home_node
+451	n64	pmadv_ksm			sys_pmadv_ksm

arch/mips/kernel/syscalls/syscall_o32.tbl

@@ -438,3 +438,4 @@
 448	o32	process_mrelease		sys_process_mrelease
 449	o32	futex_waitv			sys_futex_waitv
 450	o32	set_mempolicy_home_node		sys_set_mempolicy_home_node
+451	o32	pmadv_ksm			sys_pmadv_ksm

arch/parisc/kernel/syscalls/syscall.tbl

@@ -448,3 +448,4 @@
 448	common	process_mrelease		sys_process_mrelease
 449	common	futex_waitv			sys_futex_waitv
 450	common	set_mempolicy_home_node		sys_set_mempolicy_home_node
+451	common	pmadv_ksm			sys_pmadv_ksm

arch/powerpc/kernel/syscalls/syscall.tbl

@@ -537,3 +537,4 @@
 448	common	process_mrelease		sys_process_mrelease
 449	common  futex_waitv                     sys_futex_waitv
 450 	nospu	set_mempolicy_home_node		sys_set_mempolicy_home_node
+451	common	pmadv_ksm			sys_pmadv_ksm

arch/s390/kernel/syscalls/syscall.tbl

@@ -453,3 +453,4 @@
 448  common	process_mrelease	sys_process_mrelease		sys_process_mrelease
 449  common	futex_waitv		sys_futex_waitv			sys_futex_waitv
 450  common	set_mempolicy_home_node	sys_set_mempolicy_home_node	sys_set_mempolicy_home_node
+451  common	pmadv_ksm		sys_pmadv_ksm			sys_pmadv_ksm

arch/sh/kernel/syscalls/syscall.tbl

@@ -453,3 +453,4 @@
 448	common	process_mrelease		sys_process_mrelease
 449	common  futex_waitv                     sys_futex_waitv
 450	common	set_mempolicy_home_node		sys_set_mempolicy_home_node
+451	common	pmadv_ksm			sys_pmadv_ksm

arch/sparc/kernel/syscalls/syscall.tbl

@@ -496,3 +496,4 @@
 448	common	process_mrelease		sys_process_mrelease
 449	common  futex_waitv                     sys_futex_waitv
 450	common	set_mempolicy_home_node		sys_set_mempolicy_home_node
+451	common	pmadv_ksm			sys_pmadv_ksm

arch/x86/entry/syscalls/syscall_32.tbl

@@ -455,3 +455,4 @@
 448	i386	process_mrelease	sys_process_mrelease
 449	i386	futex_waitv		sys_futex_waitv
 450	i386	set_mempolicy_home_node		sys_set_mempolicy_home_node
+451	i386	pmadv_ksm		sys_pmadv_ksm

arch/x86/entry/syscalls/syscall_64.tbl

@@ -372,6 +372,7 @@
 448	common	process_mrelease	sys_process_mrelease
 449	common	futex_waitv		sys_futex_waitv
 450	common	set_mempolicy_home_node	sys_set_mempolicy_home_node
+451	common	pmadv_ksm		sys_pmadv_ksm
 
 #
 # Due to a historical design error, certain syscalls are numbered differently

arch/xtensa/kernel/syscalls/syscall.tbl

@@ -421,3 +421,4 @@
 448	common	process_mrelease		sys_process_mrelease
 449	common  futex_waitv                     sys_futex_waitv
 450	common	set_mempolicy_home_node		sys_set_mempolicy_home_node
+451	common	pmadv_ksm			sys_pmadv_ksm

include/linux/ksm.h

@@ -16,6 +16,10 @@
 #include <linux/sched/coredump.h>
 
 #ifdef CONFIG_KSM
+int ksm_madvise_merge(struct mm_struct *mm, struct vm_area_struct *vma,
+		unsigned long *vm_flags);
+int ksm_madvise_unmerge(struct vm_area_struct *vma, unsigned long start,
+		unsigned long end, unsigned long *vm_flags);
 int ksm_madvise(struct vm_area_struct *vma, unsigned long start,
 		unsigned long end, int advice, unsigned long *vm_flags);
 int __ksm_enter(struct mm_struct *mm);

include/linux/syscalls.h

@@ -917,6 +917,7 @@ asmlinkage long sys_madvise(unsigned long start, size_t len, int behavior);
 asmlinkage long sys_process_madvise(int pidfd, const struct iovec __user *vec,
 			size_t vlen, int behavior, unsigned int flags);
 asmlinkage long sys_process_mrelease(int pidfd, unsigned int flags);
+asmlinkage long sys_pmadv_ksm(int pidfd, int behavior, unsigned int flags);
 asmlinkage long sys_remap_file_pages(unsigned long start, unsigned long size,
 			unsigned long prot, unsigned long pgoff,
 			unsigned long flags);

include/uapi/asm-generic/unistd.h

@@ -886,8 +886,11 @@ __SYSCALL(__NR_futex_waitv, sys_futex_waitv)
 #define __NR_set_mempolicy_home_node 450
 __SYSCALL(__NR_set_mempolicy_home_node, sys_set_mempolicy_home_node)
 
+#define __NR_pmadv_ksm 451
+__SYSCALL(__NR_pmadv_ksm, sys_pmadv_ksm)
+
 #undef __NR_syscalls
-#define __NR_syscalls 451
+#define __NR_syscalls 452
 
 /*
  * 32 bit systems traditionally used different

kernel/sys_ni.c

@@ -292,6 +292,7 @@ COND_SYSCALL(mincore);
 COND_SYSCALL(madvise);
 COND_SYSCALL(process_madvise);
 COND_SYSCALL(process_mrelease);
+COND_SYSCALL(pmadv_ksm);
 COND_SYSCALL(remap_file_pages);
 COND_SYSCALL(mbind);
 COND_SYSCALL(get_mempolicy);

mm/ksm.c

@@ -2438,54 +2438,78 @@ static int ksm_scan_thread(void *nothing)
 	return 0;
 }
 
-int ksm_madvise(struct vm_area_struct *vma, unsigned long start,
-		unsigned long end, int advice, unsigned long *vm_flags)
+int ksm_madvise_merge(struct mm_struct *mm, struct vm_area_struct *vma,
+		unsigned long *vm_flags)
 {
-	struct mm_struct *mm = vma->vm_mm;
 	int err;
 
-	switch (advice) {
-	case MADV_MERGEABLE:
-		/*
-		 * Be somewhat over-protective for now!
-		 */
-		if (*vm_flags & (VM_MERGEABLE | VM_SHARED  | VM_MAYSHARE   |
-				 VM_PFNMAP    | VM_IO      | VM_DONTEXPAND |
-				 VM_HUGETLB | VM_MIXEDMAP))
-			return 0;		/* just ignore the advice */
+	/*
+	 * Be somewhat over-protective for now!
+	 */
+	if (*vm_flags & (VM_MERGEABLE | VM_SHARED  | VM_MAYSHARE   |
+			 VM_PFNMAP    | VM_IO      | VM_DONTEXPAND |
+			 VM_HUGETLB | VM_MIXEDMAP))
+		return 0;		/* just ignore the advice */
 
-		if (vma_is_dax(vma))
-			return 0;
+	if (vma_is_dax(vma))
+		return 0;
 
 #ifdef VM_SAO
 		if (*vm_flags & VM_SAO)
 			return 0;
 #endif
 #ifdef VM_SPARC_ADI
-		if (*vm_flags & VM_SPARC_ADI)
-			return 0;
+	if (*vm_flags & VM_SPARC_ADI)
+		return 0;
 #endif
 
-		if (!test_bit(MMF_VM_MERGEABLE, &mm->flags)) {
-			err = __ksm_enter(mm);
-			if (err)
-				return err;
-		}
+	if (!test_bit(MMF_VM_MERGEABLE, &mm->flags)) {
+		err = __ksm_enter(mm);
+		if (err)
+			return err;
+	}
 
-		*vm_flags |= VM_MERGEABLE;
-		break;
+	*vm_flags |= VM_MERGEABLE;
 
-	case MADV_UNMERGEABLE:
-		if (!(*vm_flags & VM_MERGEABLE))
-			return 0;		/* just ignore the advice */
+	return 0;
+}
 
-		if (vma->anon_vma) {
-			err = unmerge_ksm_pages(vma, start, end);
-			if (err)
-				return err;
-		}
+int ksm_madvise_unmerge(struct vm_area_struct *vma, unsigned long start,
+		unsigned long end, unsigned long *vm_flags)
+{
+	int err;
+
+	if (!(*vm_flags & VM_MERGEABLE))
+		return 0;		/* just ignore the advice */
+
+	if (vma->anon_vma) {
+		err = unmerge_ksm_pages(vma, start, end);
+		if (err)
+			return err;
+	}
 
-		*vm_flags &= ~VM_MERGEABLE;
+	*vm_flags &= ~VM_MERGEABLE;
+
+	return 0;
+}
+
+int ksm_madvise(struct vm_area_struct *vma, unsigned long start,
+		unsigned long end, int advice, unsigned long *vm_flags)
+{
+	struct mm_struct *mm = vma->vm_mm;
+	int err;
+
+	switch (advice) {
+	case MADV_MERGEABLE:
+		err = ksm_madvise_merge(mm, vma, vm_flags);
+		if (err)
+			return err;
+		break;
+
+	case MADV_UNMERGEABLE:
+		err = ksm_madvise_unmerge(vma, start, end, vm_flags);
+		if (err)
+			return err;
 		break;
 	}
 

mm/madvise.c

@@ -1512,3 +1512,116 @@ SYSCALL_DEFINE5(process_madvise, int, pidfd, const struct iovec __user *, vec,
 out:
 	return ret;
 }
+
+SYSCALL_DEFINE3(pmadv_ksm, int, pidfd, int, behaviour, unsigned int, flags)
+{
+#ifdef CONFIG_KSM
+	ssize_t ret;
+	struct pid *pid;
+	struct task_struct *task;
+	struct mm_struct *mm;
+	unsigned int f_flags;
+	struct vm_area_struct *vma;
+	struct vma_iterator vmi;
+
+	if (flags != 0) {
+		ret = -EINVAL;
+		goto out;
+	}
+
+	switch (behaviour) {
+		case MADV_MERGEABLE:
+		case MADV_UNMERGEABLE:
+			break;
+		default:
+			ret = -EINVAL;
+			goto out;
+			break;
+	}
+
+	pid = pidfd_get_pid(pidfd, &f_flags);
+	if (IS_ERR(pid)) {
+		ret = PTR_ERR(pid);
+		goto out;
+	}
+
+	task = get_pid_task(pid, PIDTYPE_PID);
+	if (!task) {
+		ret = -ESRCH;
+		goto put_pid;
+	}
+
+	/* Require PTRACE_MODE_READ to avoid leaking ASLR metadata. */
+	mm = mm_access(task, PTRACE_MODE_READ_FSCREDS);
+	if (IS_ERR_OR_NULL(mm)) {
+		ret = IS_ERR(mm) ? PTR_ERR(mm) : -ESRCH;
+		goto release_task;
+	}
+
+	/* Require CAP_SYS_NICE for influencing process performance. */
+	if (!capable(CAP_SYS_NICE)) {
+		ret = -EPERM;
+		goto release_mm;
+	}
+
+	if (mmap_write_lock_killable(mm)) {
+		ret = -EINTR;
+		goto release_mm;
+	}
+
+	vma_iter_init(&vmi, mm, 0);
+	for_each_vma(vmi, vma) {
+		switch (behaviour) {
+			case MADV_MERGEABLE:
+				ret = ksm_madvise_merge(vma->vm_mm, vma, &vma->vm_flags);
+				break;
+			case MADV_UNMERGEABLE:
+				ret = ksm_madvise_unmerge(vma, vma->vm_start, vma->vm_end, &vma->vm_flags);
+				break;
+			default:
+				/* look, ma, no brain */
+				break;
+		}
+		if (ret)
+			break;
+	}
+
+	mmap_write_unlock(mm);
+
+release_mm:
+	mmput(mm);
+release_task:
+	put_task_struct(task);
+put_pid:
+	put_pid(pid);
+out:
+	return ret;
+#else /* CONFIG_KSM */
+	return -ENOSYS;
+#endif /* CONFIG_KSM */
+}
+
+#ifdef CONFIG_KSM
+static ssize_t ksm_show(struct kobject *kobj, struct kobj_attribute *attr,
+			 char *buf)
+{
+	return sprintf(buf, "%u\n", __NR_pmadv_ksm);
+}
+static struct kobj_attribute pmadv_ksm_attr = __ATTR_RO(ksm);
+
+static struct attribute *pmadv_sysfs_attrs[] = {
+	&pmadv_ksm_attr.attr,
+	NULL,
+};
+
+static const struct attribute_group pmadv_sysfs_attr_group = {
+	.attrs = pmadv_sysfs_attrs,
+	.name = "pmadv",
+};
+
+static int __init pmadv_sysfs_init(void)
+{
+	return sysfs_create_group(kernel_kobj, &pmadv_sysfs_attr_group);
+}
+subsys_initcall(pmadv_sysfs_init);
+#endif /* CONFIG_KSM */