lkml.org 
[lkml]   [2015]   [Apr]   [21]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
From
Subject[PATCH 2/5] perf/x86/intel/pt: Untangle pt_buffer_reset_markers()
Date
Currently, pt_buffer_reset_markers() is a difficult to read knot of
arithmetics with a redundant check for multiple-entry topa capability,
a commented out wakeup marker placement and a logical error wrt to
stop marker placement. The latter happens when write head is not page
aligned and results in stop marker being placed one page earlier than
it actually should.

All these problems only affect PT implementations that support
multiple-entry topa tables (read: proper sg). For single-entry topa
implementations, there is no functional impact.

This patch deals with all of the above. Tested on both single-entry
and multiple-entry topa PT implementations.

Signed-off-by: Alexander Shishkin <alexander.shishkin@linux.intel.com>
---
arch/x86/kernel/cpu/perf_event_intel_pt.c | 34 ++++++++++++++++++++-----------
1 file changed, 22 insertions(+), 12 deletions(-)

diff --git a/arch/x86/kernel/cpu/perf_event_intel_pt.c b/arch/x86/kernel/cpu/perf_event_intel_pt.c
index ac08b67c07..8a4595650d 100644
--- a/arch/x86/kernel/cpu/perf_event_intel_pt.c
+++ b/arch/x86/kernel/cpu/perf_event_intel_pt.c
@@ -615,7 +615,8 @@ static int pt_buffer_reset_markers(struct pt_buffer *buf,
struct perf_output_handle *handle)

{
- unsigned long idx, npages, end;
+ unsigned long head = local64_read(&buf->head);
+ unsigned long idx, npages, wakeup;

if (buf->snapshot)
return 0;
@@ -634,17 +635,26 @@ static int pt_buffer_reset_markers(struct pt_buffer *buf,
buf->topa_index[buf->stop_pos]->stop = 0;
buf->topa_index[buf->intr_pos]->intr = 0;

- if (pt_cap_get(PT_CAP_topa_multiple_entries)) {
- npages = (handle->size + 1) >> PAGE_SHIFT;
- end = (local64_read(&buf->head) >> PAGE_SHIFT) + npages;
- /*if (end > handle->wakeup >> PAGE_SHIFT)
- end = handle->wakeup >> PAGE_SHIFT;*/
- idx = end & (buf->nr_pages - 1);
- buf->stop_pos = idx;
- idx = (local64_read(&buf->head) >> PAGE_SHIFT) + npages - 1;
- idx &= buf->nr_pages - 1;
- buf->intr_pos = idx;
- }
+ /* how many pages till the STOP marker */
+ npages = handle->size >> PAGE_SHIFT;
+
+ /* if it's on a page boundary, fill up one more page */
+ if (!offset_in_page(head + handle->size + 1))
+ npages++;
+
+ idx = (head >> PAGE_SHIFT) + npages;
+ idx &= buf->nr_pages - 1;
+ buf->stop_pos = idx;
+
+ wakeup = handle->wakeup >> PAGE_SHIFT;
+
+ /* in the worst case, wake up the consumer one page before hard stop */
+ idx = (head >> PAGE_SHIFT) + npages - 1;
+ if (idx > wakeup)
+ idx = wakeup;
+
+ idx &= buf->nr_pages - 1;
+ buf->intr_pos = idx;

buf->topa_index[buf->stop_pos]->stop = 1;
buf->topa_index[buf->intr_pos]->intr = 1;
--
2.1.4


\
 
 \ /
  Last update: 2015-04-21 16:01    [W:0.069 / U:0.836 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site