Skip to contentSkip to conversation

headers: support pre-3.9 (5-args) hashtable macros

PATCH In review Mathieu Olivari 2 participants 2 messages Mar 27, 2015 – Apr 3, 2015

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 headers: support pre-3.9 (5-args) hashtable macros

Last ath10k makes use of hash tables, which changed in the upstream commit:
* commit 0bbacca7c3911451cea923b0ad6389d58e3d9ce9
* Author: Sasha Levin <[email protected]>
* Date: Thu Feb 7 12:32:18 2013 +1100
*
* hlist: drop the node parameter from iterators

The list portion of this commit has already been backported, but the hashtable part has not. So this change evens it up. It fixes a build issue when selecting ath10k on a kernel before 3.9

---
 backport/backport-include/linux/hashtable.h | 42 +++++++++++++++++++++++++++++
 1 file changed, 42 insertions(+)
 create mode 100644 backport/backport-include/linux/hashtable.h
backport/backport-include/linux/hashtable.h new file +42 -0
new file mode 100644
index 0000000..dc53e74
@@ -0,0 +1,42 @@
+#ifndef __BACKPORT_HASHTABLE_H
+#define __BACKPORT_HASHTABLE_H
+#include_next <linux/hashtable.h>
+#include <linux/version.h>
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(3,9,0)
+/**
+ * backport:
+ *
+ * commit 0bbacca7c3911451cea923b0ad6389d58e3d9ce9
+ * Author: Sasha Levin <[email protected]>
+ * Date:   Thu Feb 7 12:32:18 2013 +1100
+ *
+ *     hlist: drop the node parameter from iterators
+ */
+#include <linux/list.h>
+#include <backport/magic.h>
+
+#undef hash_for_each
+#define hash_for_each(name, bkt, obj, member)				\
+	for ((bkt) = 0, obj = NULL; obj == NULL && (bkt) < HASH_SIZE(name);\
+			(bkt)++)\
+		hlist_for_each_entry(obj, &name[bkt], member)
+
+#undef hash_for_each_safe
+#define hash_for_each_safe(name, bkt, tmp, obj, member)			\
+	for ((bkt) = 0, obj = NULL; obj == NULL && (bkt) < HASH_SIZE(name);\
+			(bkt)++)\
+		hlist_for_each_entry_safe(obj, tmp, &name[bkt], member)
+
+#undef hash_for_each_possible
+#define hash_for_each_possible(name, obj, member, key)			\
+	hlist_for_each_entry(obj, &name[hash_min(key, HASH_BITS(name))], member)
+
+#undef hash_for_each_possible_safe
+#define hash_for_each_possible_safe(name, obj, tmp, member, key)	\
+	hlist_for_each_entry_safe(obj, tmp,\
+		&name[hash_min(key, HASH_BITS(name))], member)
+
+#endif
+
+#endif /* __BACKPORT_HASHTABLE_H */
Signature · 1 line
1.9.1
#1 lore raw
Quoted 8 lines · Mathieu Olivari

Last ath10k makes use of hash tables, which changed in the upstream commit:
* commit 0bbacca7c3911451cea923b0ad6389d58e3d9ce9
* Author: Sasha Levin <[email protected]>
* Date: Thu Feb 7 12:32:18 2013 +1100
*
* hlist: drop the node parameter from iterators

Quoted 5 lines

The list portion of this commit has already been backported, but the hashtable part has not. So this change evens it up. It fixes a build issue when selecting ath10k on a kernel before 3.9

Thank you for the patch it was applied to backports.

Hauke