[PATCH] fast vdso implementation for CLOCK_THREAD_CPUTIME_ID
The extract cpu time instruction (ectg) instruction allows the user
process to get the current thread cputime without calling into the
kernel. The code that uses the instruction needs to switch to the
access registers mode to get access to the per-cpu info page that
contains the two base values that are needed to calculate the current
cputime from the CPU timer with the ectg instruction.
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
diff --git a/arch/s390/kernel/vdso.c b/arch/s390/kernel/vdso.c
index 10a6cce..25a6a82 100644
--- a/arch/s390/kernel/vdso.c
+++ b/arch/s390/kernel/vdso.c
@@ -31,9 +31,6 @@
#include <asm/sections.h>
#include <asm/vdso.h>
-/* Max supported size for symbol names */
-#define MAX_SYMNAME 64
-
#if defined(CONFIG_32BIT) || defined(CONFIG_COMPAT)
extern char vdso32_start, vdso32_end;
static void *vdso32_kbase = &vdso32_start;
@@ -71,6 +68,119 @@
struct vdso_data *vdso_data = &vdso_data_store.data;
/*
+ * Setup vdso data page.
+ */
+static void vdso_init_data(struct vdso_data *vd)
+{
+ unsigned int facility_list;
+
+ facility_list = stfl();
+ vd->ectg_available = switch_amode && (facility_list & 1);
+}
+
+#ifdef CONFIG_64BIT
+/*
+ * Setup per cpu vdso data page.
+ */
+static void vdso_init_per_cpu_data(int cpu, struct vdso_per_cpu_data *vpcd)
+{
+}
+
+/*
+ * Allocate/free per cpu vdso data.
+ */
+#ifdef CONFIG_64BIT
+#define SEGMENT_ORDER 2
+#else
+#define SEGMENT_ORDER 1
+#endif
+
+int vdso_alloc_per_cpu(int cpu, struct _lowcore *lowcore)
+{
+ unsigned long segment_table, page_table, page_frame;
+ u32 *psal, *aste;
+ int i;
+
+ lowcore->vdso_per_cpu_data = __LC_PASTE;
+
+ if (!switch_amode || !vdso_enabled)
+ return 0;
+
+ segment_table = __get_free_pages(GFP_KERNEL, SEGMENT_ORDER);
+ page_table = get_zeroed_page(GFP_KERNEL | GFP_DMA);
+ page_frame = get_zeroed_page(GFP_KERNEL);
+ if (!segment_table || !page_table || !page_frame)
+ goto out;
+
+ clear_table((unsigned long *) segment_table, _SEGMENT_ENTRY_EMPTY,
+ PAGE_SIZE << SEGMENT_ORDER);
+ clear_table((unsigned long *) page_table, _PAGE_TYPE_EMPTY,
+ 256*sizeof(unsigned long));
+
+ *(unsigned long *) segment_table = _SEGMENT_ENTRY + page_table;
+ *(unsigned long *) page_table = _PAGE_RO + page_frame;
+
+ psal = (u32 *) (page_table + 256*sizeof(unsigned long));
+ aste = psal + 32;
+
+ for (i = 4; i < 32; i += 4)
+ psal[i] = 0x80000000;
+
+ lowcore->paste[4] = (u32)(addr_t) psal;
+ psal[0] = 0x20000000;
+ psal[2] = (u32)(addr_t) aste;
+ *(unsigned long *) (aste + 2) = segment_table +
+ _ASCE_TABLE_LENGTH + _ASCE_USER_BITS + _ASCE_TYPE_SEGMENT;
+ aste[4] = (u32)(addr_t) psal;
+ lowcore->vdso_per_cpu_data = page_frame;
+
+ vdso_init_per_cpu_data(cpu, (struct vdso_per_cpu_data *) page_frame);
+ return 0;
+
+out:
+ free_page(page_frame);
+ free_page(page_table);
+ free_pages(segment_table, SEGMENT_ORDER);
+ return -ENOMEM;
+}
+
+#ifdef CONFIG_HOTPLUG_CPU
+void vdso_free_per_cpu(int cpu, struct _lowcore *lowcore)
+{
+ unsigned long segment_table, page_table, page_frame;
+ u32 *psal, *aste;
+
+ if (!switch_amode || !vdso_enabled)
+ return;
+
+ psal = (u32 *)(addr_t) lowcore->paste[4];
+ aste = (u32 *)(addr_t) psal[2];
+ segment_table = *(unsigned long *)(aste + 2) & PAGE_MASK;
+ page_table = *(unsigned long *) segment_table;
+ page_frame = *(unsigned long *) page_table;
+
+ free_page(page_frame);
+ free_page(page_table);
+ free_pages(segment_table, SEGMENT_ORDER);
+}
+#endif /* CONFIG_HOTPLUG_CPU */
+
+static void __vdso_init_cr5(void *dummy)
+{
+ unsigned long cr5;
+
+ cr5 = offsetof(struct _lowcore, paste);
+ __ctl_load(cr5, 5, 5);
+}
+
+static void vdso_init_cr5(void)
+{
+ if (switch_amode && vdso_enabled)
+ on_each_cpu(__vdso_init_cr5, NULL, 1);
+}
+#endif /* CONFIG_64BIT */
+
+/*
* This is called from binfmt_elf, we create the special vma for the
* vDSO and insert it into the mm struct tree
*/
@@ -172,6 +282,9 @@
{
int i;
+ if (!vdso_enabled)
+ return 0;
+ vdso_init_data(vdso_data);
#if defined(CONFIG_32BIT) || defined(CONFIG_COMPAT)
/* Calculate the size of the 32 bit vDSO */
vdso32_pages = ((&vdso32_end - &vdso32_start
@@ -208,6 +321,10 @@
}
vdso64_pagelist[vdso64_pages - 1] = virt_to_page(vdso_data);
vdso64_pagelist[vdso64_pages] = NULL;
+#ifndef CONFIG_SMP
+ BUG_ON(vdso_alloc_per_cpu(0, S390_lowcore));
+#endif
+ vdso_init_cr5();
#endif /* CONFIG_64BIT */
get_page(virt_to_page(vdso_data));