openharmony/third_party_spirv-tools
1
0
Fork 0
mirror of https://gitee.com/openharmony/third_party_spirv-tools synced 2024-11-27 17:40:28 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

Change HexFloat to work with gcc8. (#2109)

Browse Source 
When we want to set a the value of a HexFloat to inf or nan, we
construct the specific bit pattern in an appropriately sized integer.
That integer is copied to a FloatProxy object through a memcpy.  GCC8
complains about the memcpy because it is overwriting a private member of
the class.

The original solution worked well because the template to the HexFloat
could be anything.  However, we only used some instantiation of FloatProxy,
which has a construction from that takes its uint_type, so I decided to use
that constructor instead of the memcpy.  This puts an extra requirement
on the templace for HexFloat, but it will be fine for us.

Part of #1541.
This commit is contained in:
Steven Perron 2018-11-26 15:47:48 -05:00 committed by GitHub
parent d543f7dfed
commit 72d4e5414b
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 2 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
source/util/hex_float.h
View File

@ -672,7 +672,7 @@ class HexFloat {
// If we are Nan or Inf we should pass that through.
if (is_inf) {
other.set_value(BitwiseCast<typename other_T::underlying_type>(
other.set_value(typename other_T::underlying_type(
static_cast<typename other_T::uint_type>(
(negate ? other_T::sign_mask : 0) | other_T::exponent_mask)));
return;
@ -687,7 +687,7 @@ class HexFloat {
// We are some sort of Nan. We try to keep the bit-pattern of the Nan
// as close as possible. If we had to shift off bits so we are 0, then we
// just set the last bit.
other.set_value(BitwiseCast<typename other_T::underlying_type>(
other.set_value(typename other_T::underlying_type(
static_cast<typename other_T::uint_type>(
(negate ? other_T::sign_mask : 0) | other_T::exponent_mask |
(shifted_significand == 0 ? 0x1 : shifted_significand))));