diff options
author | schmidt <douglascraigschmidt@users.noreply.github.com> | 1999-07-31 22:18:20 +0000 |
---|---|---|
committer | schmidt <douglascraigschmidt@users.noreply.github.com> | 1999-07-31 22:18:20 +0000 |
commit | 10f1729b65638f3f273f523861dd547ec6832e5f (patch) | |
tree | a49c98ad0d34bcb5a87c9f7fb1a256c7bf4df6dd /ace/CDR_Stream.h | |
parent | 3e172150a24e32f471d10eb9a60eebb599de6f81 (diff) | |
download | ATCD-10f1729b65638f3f273f523861dd547ec6832e5f.tar.gz |
ChangeLogTag:Sat Jul 31 12:07:48 1999 Douglas C. Schmidt <schmidt@mambo.cs.wustl.edu>
Diffstat (limited to 'ace/CDR_Stream.h')
-rw-r--r-- | ace/CDR_Stream.h | 20 |
1 files changed, 10 insertions, 10 deletions
diff --git a/ace/CDR_Stream.h b/ace/CDR_Stream.h index 13dad9d7ab5..72a476628c1 100644 --- a/ace/CDR_Stream.h +++ b/ace/CDR_Stream.h @@ -450,9 +450,9 @@ private: // to be packed with the right alignment restrictions. It is mostly // designed for buffers of the basic types. // - // This operation uses memcpy(); as explained above it is expected - // that using assignment is faster that memcpy() for one element, - // but for several elements memcpy() should be more efficient, it + // This operation uses <memcpy>; as explained above it is expected + // that using assignment is faster that <memcpy> for one element, + // but for several elements <memcpy> should be more efficient, it // could be interesting to find the break even point and optimize // for that case, but that would be too platform dependent. @@ -474,7 +474,7 @@ private: char *&buf); // Grow the CDR stream. When it returns <buf> contains a pointer to // memory in the CDR stream, with at least <size> bytes ahead of it - // and aligned to an <align> boundary. It moved the wr_ptr() to <buf + // and aligned to an <align> boundary. It moved the <wr_ptr> to <buf // + size>. int do_byte_swap (void) const; @@ -701,9 +701,9 @@ public: // CDR stream from a socket or file. int grow (size_t newsize); - // Grow the internal buffer, reset rd_ptr() to the first byte in the - // new buffer that is properly aligned, and set wr_ptr() to - // rd_ptr() + newsize + // Grow the internal buffer, reset <rd_ptr> to the first byte in the + // new buffer that is properly aligned, and set <wr_ptr> to <rd_ptr> + // + newsize void reset_byte_order (int byte_order); // After reading and partially parsing the contents the user can @@ -789,9 +789,9 @@ private: // to be packed with the right alignment restrictions. It is mostly // designed for buffers of the basic types. // - // This operation uses memcpy(); as explained above it is expected - // that using assignment is faster that memcpy() for one element, - // but for several elements memcpy() should be more efficient, it + // This operation uses <memcpy>; as explained above it is expected + // that using assignment is faster that <memcpy> for one element, + // but for several elements <memcpy> should be more efficient, it // could be interesting to find the break even point and optimize // for that case, but that would be too platform dependent. |