Thread (8 messages) flat view 8 messages, 2 authors, 2021-07-22
STALE1859d

Revision v2 of 2 in this series.

Revisions (2)
  1. v1 [diff vs current]
  2. v2 current

[PATCH v2 1/4] tools headers UAPI: add cache aligning related macro

From: Yunsheng Lin <hidden>
Date: 2021-07-20 02:38:39
Also in: lkml
Subsystem: the rest · Maintainer: Linus Torvalds

____cacheline_aligned_in_smp macro is needed to avoid
cache bouncing in SMP system, which is used in ptr_ring
lib.

So add the related macro in order to bulid ptr_ring from
user space.

As SMP_CACHE_BYTES is 64 bytes for arm64 and most of x86
system, so use 64 bytes as the default SMP_CACHE_BYTES if
SMP_CACHE_BYTES is not defined.

Signed-off-by: Yunsheng Lin <redacted>
---
 tools/include/linux/cache.h | 25 +++++++++++++++++++++++++
 1 file changed, 25 insertions(+)
 create mode 100644 tools/include/linux/cache.h
diff --git a/tools/include/linux/cache.h b/tools/include/linux/cache.h
new file mode 100644
index 0000000..df04307
--- /dev/null
+++ b/tools/include/linux/cache.h
@@ -0,0 +1,25 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef __TOOLS_LINUX__CACHE_H
+#define __TOOLS_LINUX__CACHE_H
+
+#ifndef CONFIG_SMP
+#define CONFIG_SMP	1
+#endif
+
+#ifndef SMP_CACHE_BYTES
+#define SMP_CACHE_BYTES	64
+#endif
+
+#ifndef ____cacheline_aligned
+#define ____cacheline_aligned __attribute__((__aligned__(SMP_CACHE_BYTES)))
+#endif
+
+#ifndef ____cacheline_aligned_in_smp
+#ifdef CONFIG_SMP
+#define ____cacheline_aligned_in_smp ____cacheline_aligned
+#else
+#define ____cacheline_aligned_in_smp
+#endif /* CONFIG_SMP */
+#endif
+
+#endif /* __LINUX_CACHE_H */
-- 
2.7.4
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help