[vector] Do not refer to VLENGTH in package doc

Viswanathan, Sandhya sandhya.viswanathan at intel.com
Wed May 13 00:16:45 UTC 2020


Best Regards,

-----Original Message-----
From: panama-dev <panama-dev-bounces at openjdk.java.net> On Behalf Of Paul Sandoz
Sent: Tuesday, May 12, 2020 12:15 PM
To: panama-dev at openjdk.java.net' <panama-dev at openjdk.java.net>
Subject: [vector] Do not refer to VLENGTH in package doc


Please review this small change [1] to refer to SPECIES.length() in the examples, rather than the term VLENGTH as defined by the Vector class doc.



diff -r c57baebffe2b src/jdk.incubator.vector/share/classes/jdk/incubator/vector/package-info.java
--- a/src/jdk.incubator.vector/share/classes/jdk/incubator/vector/package-info.java     Tue May 12 11:48:13 2020 -0700
+++ b/src/jdk.incubator.vector/share/classes/jdk/incubator/vector/package-info.java     Tue May 12 12:13:15 2020 -0700
@@ -169,7 +169,7 @@
  * The scalar computation after the vector computation is required to
  * process a <em>tail</em> of {@code TLENGTH} array elements, where
- * {@code TLENGTH <= VLENGTH} for the vector species.
+ * {@code TLENGTH < SPECIES.length()} for the vector species.
  * The example above uses vectors hardcoded to a concrete shape
  * (512-bit). Instead, we could use preferred species as shown below, @@ -184,7 +184,7 @@
  * is used in the above code to find the end of the vector loop.
  * A primitive masking expression such as
  * {@code (a.length & ~(SPECIES.length() - 1))} might also be used
- * here, since the species {@code VLENGTH} is known to be 8, which
+ * here, since {@code SPECIES.length()} is known to be 8, which
  * is a power of two.  But this is not always a correct assumption.
  * For example, if the {@code FloatVector.SPECIES_PREFERRED} turns
  * out to have the platform-dependent shape

More information about the panama-dev mailing list