perf_event: Provide vmalloc() based mmap() backing

Some architectures such as Sparc, ARM and MIPS (basically
everything with flush_dcache_page()) need to deal with dcache
aliases by carefully placing pages in both kernel and user maps.

These architectures typically have to use vmalloc_user() for this.

However, on other architectures, vmalloc() is not needed and has
the downsides of being more restricted and slower than regular
allocations.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Acked-by: David Miller <davem@davemloft.net>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Jens Axboe <jens.axboe@oracle.com>
Cc: Paul Mackerras <paulus@samba.org>
LKML-Reference: <1254830228.21044.272.camel@laptop>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

5 files changed, 214 insertions, 62 deletions

diff --git a/arch/sparc/Kconfig b/arch/sparc/Kconfig
index 97fca4695e0b..9b70a2f28dc7 100644
--- a/arch/sparc/Kconfig
+++ b/arch/sparc/Kconfig
@@ -26,6 +26,7 @@ config SPARC
         select RTC_CLASS
         select RTC_DRV_M48T59
         select HAVE_PERF_EVENTS
+        select PERF_USE_VMALLOC
         select HAVE_DMA_ATTRS
         select HAVE_DMA_API_DEBUG
 
@@ -48,6 +49,7 @@ config SPARC64
         select RTC_DRV_SUN4V
         select RTC_DRV_STARFIRE
         select HAVE_PERF_EVENTS
+        select PERF_USE_VMALLOC
 
 config ARCH_DEFCONFIG
         string
diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index 3a9d36d1e92a..2e6d95f97419 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -442,6 +442,7 @@ enum perf_callchain_context {
 #include <linux/hrtimer.h>
 #include <linux/fs.h>
 #include <linux/pid_namespace.h>
+#include <linux/workqueue.h>
 #include <asm/atomic.h>
 
 #define PERF_MAX_STACK_DEPTH            255
@@ -513,6 +514,10 @@ struct file;
 
 struct perf_mmap_data {
         struct rcu_head                 rcu_head;
+#ifdef CONFIG_PERF_USE_VMALLOC
+        struct work_struct              work;
+#endif
+        int                             data_order;
         int                             nr_pages;       /* nr of data pages  */
         int                             writable;       /* are we writable   */
         int                             nr_locked;      /* nr pages mlocked  */
diff --git a/init/Kconfig b/init/Kconfig
index c7bac39d6c61..09c5c6431f42 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -921,6 +921,11 @@ config HAVE_PERF_EVENTS
         help
           See tools/perf/design.txt for details.
 
+config PERF_USE_VMALLOC
+        bool
+        help
+          See tools/perf/design.txt for details
+
 menu "Kernel Performance Events And Counters"
 
 config PERF_EVENTS
@@ -976,6 +981,19 @@ config PERF_COUNTERS
 
           Say N if unsure.
 
+config DEBUG_PERF_USE_VMALLOC
+        default n
+        bool "Debug: use vmalloc to back perf mmap() buffers"
+        depends on PERF_EVENTS && DEBUG_KERNEL
+        select PERF_USE_VMALLOC
+        help
+         Use vmalloc memory to back perf mmap() buffers.
+
+         Mostly useful for debugging the vmalloc code on platforms
+         that don't require it.
+
+         Say N if unsure.
+
 endmenu
 
 config VM_EVENT_COUNTERS
diff --git a/kernel/perf_event.c b/kernel/perf_event.c
index e491fb087939..9d0b5c665883 100644
--- a/kernel/perf_event.c
+++ b/kernel/perf_event.c
@@ -20,6 +20,7 @@
 #include <linux/percpu.h>
 #include <linux/ptrace.h>
 #include <linux/vmstat.h>
+#include <linux/vmalloc.h>
 #include <linux/hardirq.h>
 #include <linux/rculist.h>
 #include <linux/uaccess.h>
@@ -2091,49 +2092,31 @@ unlock:
         rcu_read_unlock();
 }
 
-static int perf_mmap_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
+static unsigned long perf_data_size(struct perf_mmap_data *data)
 {
-        struct perf_event *event = vma->vm_file->private_data;
-        struct perf_mmap_data *data;
-        int ret = VM_FAULT_SIGBUS;
-
-        if (vmf->flags & FAULT_FLAG_MKWRITE) {
-                if (vmf->pgoff == 0)
-                        ret = 0;
-                return ret;
-        }
-
-        rcu_read_lock();
-        data = rcu_dereference(event->data);
-        if (!data)
-                goto unlock;
-
-        if (vmf->pgoff == 0) {
-                vmf->page = virt_to_page(data->user_page);
-        } else {
-                int nr = vmf->pgoff - 1;
-
-                if ((unsigned)nr > data->nr_pages)
-                        goto unlock;
+        return data->nr_pages << (PAGE_SHIFT + data->data_order);
+}
 
-                if (vmf->flags & FAULT_FLAG_WRITE)
-                        goto unlock;
+#ifndef CONFIG_PERF_USE_VMALLOC
 
-                vmf->page = virt_to_page(data->data_pages[nr]);
-        }
+/*
+ * Back perf_mmap() with regular GFP_KERNEL-0 pages.
+ */
 
-        get_page(vmf->page);
-        vmf->page->mapping = vma->vm_file->f_mapping;
-        vmf->page->index   = vmf->pgoff;
+static struct page *
+perf_mmap_to_page(struct perf_mmap_data *data, unsigned long pgoff)
+{
+        if (pgoff > data->nr_pages)
+                return NULL;
 
-        ret = 0;
-unlock:
-        rcu_read_unlock();
+        if (pgoff == 0)
+                return virt_to_page(data->user_page);
 
-        return ret;
+        return virt_to_page(data->data_pages[pgoff - 1]);
 }
 
-static int perf_mmap_data_alloc(struct perf_event *event, int nr_pages)
+static struct perf_mmap_data *
+perf_mmap_data_alloc(struct perf_event *event, int nr_pages)
 {
         struct perf_mmap_data *data;
         unsigned long size;
@@ -2158,19 +2141,10 @@ static int perf_mmap_data_alloc(struct perf_event *event, int nr_pages)
                         goto fail_data_pages;
         }
 
+        data->data_order = 0;
         data->nr_pages = nr_pages;
-        atomic_set(&data->lock, -1);
-
-        if (event->attr.watermark) {
-                data->watermark = min_t(long, PAGE_SIZE * nr_pages,
-                                      event->attr.wakeup_watermark);
-        }
-        if (!data->watermark)
-                data->watermark = max(PAGE_SIZE, PAGE_SIZE * nr_pages / 4);
 
-        rcu_assign_pointer(event->data, data);
-
-        return 0;
+        return data;
 
 fail_data_pages:
         for (i--; i >= 0; i--)
@@ -2182,7 +2156,7 @@ fail_user_page:
         kfree(data);