@@ -41,6 +41,7 @@
#include <asm/system_misc.h>
#include <asm/mach/time.h>
#include <asm/tls.h>
+#include <asm/vdso.h>
#ifdef CONFIG_CC_STACKPROTECTOR
#include <linux/stackprotector.h>
@@ -476,7 +477,7 @@ const char *arch_vma_name(struct vm_area_struct *vma)
}
/* If possible, provide a placement hint at a random offset from the
- * stack for the signal page.
+ * stack for the sigpage and vdso pages.
*/
static unsigned long sigpage_addr(const struct mm_struct *mm,
unsigned int npages)
@@ -520,6 +521,7 @@ int arch_setup_additional_pages(struct linux_binprm *bprm, int uses_interp)
{
struct mm_struct *mm = current->mm;
struct vm_area_struct *vma;
+ unsigned long npages;
unsigned long addr;
unsigned long hint;
int ret = 0;
@@ -529,9 +531,12 @@ int arch_setup_additional_pages(struct linux_binprm *bprm, int uses_interp)
if (!signal_page)
return -ENOMEM;
+ npages = 1; /* for sigpage */
+ npages += vdso_total_pages;
+
down_write(&mm->mmap_sem);
- hint = sigpage_addr(mm, 1);
- addr = get_unmapped_area(NULL, hint, PAGE_SIZE, 0, 0);
+ hint = sigpage_addr(mm, npages);
+ addr = get_unmapped_area(NULL, hint, npages << PAGE_SHIFT, 0, 0);
if (IS_ERR_VALUE(addr)) {
ret = addr;
goto up_fail;
@@ -548,6 +553,12 @@ int arch_setup_additional_pages(struct linux_binprm *bprm, int uses_interp)
mm->context.sigpage = addr;
+ /* Unlike the sigpage, failure to install the vdso is unlikely
+ * to be fatal to the process, so no error check needed
+ * here.
+ */
+ arm_install_vdso(mm, addr + PAGE_SIZE);
+
up_fail:
up_write(&mm->mmap_sem);
return ret;
new file mode 100644
@@ -0,0 +1,211 @@
+/*
+ * Adapted from arm64 version.
+ *
+ * Copyright (C) 2012 ARM Limited
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include <linux/err.h>
+#include <linux/kernel.h>
+#include <linux/mm.h>
+#include <linux/slab.h>
+#include <linux/timekeeper_internal.h>
+#include <linux/vmalloc.h>
+
+#include <asm/barrier.h>
+#include <asm/cacheflush.h>
+#include <asm/page.h>
+#include <asm/vdso.h>
+#include <asm/vdso_datapage.h>
+
+#include <asm/arch_timer.h>
+
+static struct page **vdso_text_pagelist;
+
+/* Total number of pages needed for the data and text portions of the VDSO. */
+unsigned int vdso_total_pages __read_mostly;
+
+/*
+ * The vDSO data page.
+ */
+static union vdso_data_store vdso_data_store __page_aligned_data;
+static struct vdso_data *vdso_data = &vdso_data_store.data;
+
+static struct page *vdso_data_page;
+static struct vm_special_mapping vdso_data_mapping = {
+ .name = "[vvar]",
+ .pages = &vdso_data_page,
+};
+
+static struct vm_special_mapping vdso_text_mapping = {
+ .name = "[vdso]",
+};
+
+static int __init vdso_init(void)
+{
+ unsigned int text_pages;
+ int i;
+
+ if (memcmp(&vdso_start, "\177ELF", 4)) {
+ pr_err("vDSO is not a valid ELF object!\n");
+ return -ENOEXEC;
+ }
+
+ text_pages = (&vdso_end - &vdso_start) >> PAGE_SHIFT;
+ pr_debug("vdso: %i text pages at base %p\n", text_pages, &vdso_start);
+
+ /* Allocate the vDSO text pagelist */
+ vdso_text_pagelist = kcalloc(text_pages, sizeof(struct page *),
+ GFP_KERNEL);
+ if (vdso_text_pagelist == NULL)
+ return -ENOMEM;
+
+ /* Grab the vDSO data page. */
+ vdso_data_page = virt_to_page(vdso_data);
+
+ /* Grab the vDSO text pages. */
+ for (i = 0; i < text_pages; i++) {
+ struct page *page;
+
+ page = virt_to_page(&vdso_start + i * PAGE_SIZE);
+ vdso_text_pagelist[i] = page;
+ }
+
+ vdso_text_mapping.pages = vdso_text_pagelist;
+
+ vdso_total_pages = 1; /* for the data/vvar page */
+ vdso_total_pages += text_pages;
+
+ return 0;
+}
+arch_initcall(vdso_init);
+
+static int install_vvar(struct mm_struct *mm, unsigned long addr)
+{
+ struct vm_area_struct *vma;
+
+ vma = _install_special_mapping(mm, addr, PAGE_SIZE,
+ VM_READ | VM_MAYREAD,
+ &vdso_data_mapping);
+
+ return IS_ERR(vma) ? PTR_ERR(vma) : 0;
+}
+
+/* assumes mmap_sem is write-locked */
+void arm_install_vdso(struct mm_struct *mm, unsigned long addr)
+{
+ struct vm_area_struct *vma;
+ unsigned long len;
+
+ mm->context.vdso = 0;
+
+ if (vdso_text_pagelist == NULL)
+ return;
+
+ if (install_vvar(mm, addr))
+ return;
+
+ /* Account for vvar page. */
+ addr += PAGE_SIZE;
+ len = (vdso_total_pages - 1) << PAGE_SHIFT;
+
+ vma = _install_special_mapping(mm, addr, len,
+ VM_READ | VM_EXEC | VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC,
+ &vdso_text_mapping);
+
+ if (!IS_ERR(vma))
+ mm->context.vdso = addr;
+}
+
+static void vdso_write_begin(struct vdso_data *vdata)
+{
+ ++vdso_data->seq_count;
+ smp_wmb(); /* Pairs with smp_rmb in vdso_read_retry */
+}
+
+static void vdso_write_end(struct vdso_data *vdata)
+{
+ smp_wmb(); /* Pairs with smp_rmb in vdso_read_begin */
+ ++vdso_data->seq_count;
+}
+
+static bool vdso_can_use_arch_timer(const struct timekeeper *tk)
+{
+#ifdef CONFIG_ARM_ARCH_TIMER
+ u32 cntkctl;
+
+ if (strcmp(tk->tkr.clock->name, "arch_sys_counter") != 0)
+ return false;
+
+ cntkctl = arch_timer_get_cntkctl();
+
+ return cntkctl & ARCH_TIMER_USR_VCT_ACCESS_EN;
+#else
+ return false;
+#endif
+}
+
+/**
+ * update_vsyscall - update the vdso data page
+ *
+ * Increment the sequence counter, making it odd, indicating to
+ * userspace that an update is in progress. Update the fields used
+ * for coarse clocks and, if the architected system timer is in use,
+ * the fields used for high precision clocks. Increment the sequence
+ * counter again, making it even, indicating to userspace that the
+ * update is finished.
+ *
+ * Userspace is expected to sample seq_count before reading any other
+ * fields from the data page. If seq_count is odd, userspace is
+ * expected to wait until it becomes even. After copying data from
+ * the page, userspace must sample seq_count again; if it has changed
+ * from its previous value, userspace must retry the whole sequence.
+ *
+ * Calls to update_vsyscall are serialized by the timekeeping core.
+ */
+void update_vsyscall(struct timekeeper *tk)
+{
+ struct timespec xtime_coarse;
+ struct timespec64 *wtm = &tk->wall_to_monotonic;
+ bool use_syscall = !vdso_can_use_arch_timer(tk);
+
+ vdso_write_begin(vdso_data);
+
+ xtime_coarse = __current_kernel_time();
+ vdso_data->use_syscall = use_syscall;
+ vdso_data->xtime_coarse_sec = xtime_coarse.tv_sec;
+ vdso_data->xtime_coarse_nsec = xtime_coarse.tv_nsec;
+ vdso_data->wtm_clock_sec = wtm->tv_sec;
+ vdso_data->wtm_clock_nsec = wtm->tv_nsec;
+
+ if (!use_syscall) {
+ vdso_data->cs_cycle_last = tk->tkr.cycle_last;
+ vdso_data->xtime_clock_sec = tk->xtime_sec;
+ vdso_data->xtime_clock_snsec = tk->tkr.xtime_nsec;
+ vdso_data->cs_mult = tk->tkr.mult;
+ vdso_data->cs_shift = tk->tkr.shift;
+ vdso_data->cs_mask = tk->tkr.mask;
+ }
+
+ vdso_write_end(vdso_data);
+
+ flush_dcache_page(virt_to_page(vdso_data));
+}
+
+void update_vsyscall_tz(void)
+{
+ vdso_data->tz_minuteswest = sys_tz.tz_minuteswest;
+ vdso_data->tz_dsttime = sys_tz.tz_dsttime;
+ flush_dcache_page(virt_to_page(vdso_data));
+}
Initialize the vdso page list at boot, install the vdso mapping at exec time, and update the data page during timer ticks. This code is not built if CONFIG_VDSO is not enabled. Account for the vdso length when randomizing the offset from the stack. The vdso is placed immediately following the sigpage with a separate install_special_mapping call in arm_install_vdso. Signed-off-by: Nathan Lynch <nathan_lynch@mentor.com> --- arch/arm/kernel/process.c | 17 +++- arch/arm/kernel/vdso.c | 211 ++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 225 insertions(+), 3 deletions(-) create mode 100644 arch/arm/kernel/vdso.c