diff options
| author | Donald Washburn <dwashbur@us.ibm.com> | 2017-06-08 10:39:35 -0500 |
|---|---|---|
| committer | Joshua Hunsberger <jahunsbe@us.ibm.com> | 2017-10-23 17:50:28 -0500 |
| commit | 176045b1a9d7280c345b97a59a35744e51f29fac (patch) | |
| tree | adea54e2ba0259017730edab103a84ef88bd6af3 /import/hwpf/fapi2 | |
| parent | 28b656a07f0b8c698135a0bc77e42edc21be3a98 (diff) | |
| download | talos-hcode-176045b1a9d7280c345b97a59a35744e51f29fac.tar.gz talos-hcode-176045b1a9d7280c345b97a59a35744e51f29fac.zip | |
Enable and fix error log variable_buffer support.
Errorlog support for the fapi2::variable_buffer type was not enabled in
hostboot. Tests showed that when enabled, variable_buffer data was not
being propagated properly to the error log. The issue was found to be
that the pointer to the variable_buffer's internal data was not being
properly passed to an ffdc_t object. Also, transferring the size of
the variable_buffer data was not being correctly communicated to an
ffdc_t object because a specialization of the getErrorInfoFfdcSize
template function is needed. Becuase the specialization of the
getErrorInfoFfdcSize function with an ffdc_t parameter did not exist,
the code base used the primary function template for the
getErrorInfoFfdcSize function which just returns the size of an ffdc_t
object passed to it instead of the size of the contained data within
the ffdc_t.
Changes:
* Added specialization of getErrorInfoFfdcSize for fapi2::ffdc_t.
* Enabled variable_buffer support in parseErrorInfo.mk.
* Added const overload of the pointer() method for the fapi2::buffer
and fapi2::variable_buffer classes. This to allow these methods
to be used in the set_BUFFER methods that take a const reference
to objects of these classes.
* Modified parseErrorInfo.pl to generate code to use the above
mentioned pointer methods. The adjusted generated code fixes
the problem of assigning an incorrect buffer pointer.
Change-Id: I96dc89fbb68ee6a153ca43191181c56804b84ae8
RTC: 175239
Reviewed-on: http://ralgit01.raleigh.ibm.com/gerrit1/41541
Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com>
Tested-by: PPE CI <ppe-ci+hostboot@us.ibm.com>
Tested-by: Hostboot CI <hostboot-ci+hostboot@us.ibm.com>
Reviewed-by: William G. Hoffa <wghoffa@us.ibm.com>
Reviewed-by: Richard J. Knight <rjknight@us.ibm.com>
Reviewed-by: Prachi Gupta <pragupta@us.ibm.com>
Diffstat (limited to 'import/hwpf/fapi2')
| -rw-r--r-- | import/hwpf/fapi2/include/buffer.H | 9 |
1 files changed, 9 insertions, 0 deletions
diff --git a/import/hwpf/fapi2/include/buffer.H b/import/hwpf/fapi2/include/buffer.H index 43019030..0015103b 100644 --- a/import/hwpf/fapi2/include/buffer.H +++ b/import/hwpf/fapi2/include/buffer.H @@ -401,6 +401,15 @@ class buffer return &iv_data; } + /// + /// @brief Get a pointer to the buffer bits + /// @return Pointer to the buffer itself + /// + inline const T* pointer(void) const + { + return &iv_data; + } + // Note: Many (all?) of these are not needed and the compiler complains // as the cast to T yields a better operator. There are here mainly for // documenation purposes. |

