mirror of
https://github.com/RPCS3/llvm-mirror.git
synced 2024-12-23 03:59:12 +00:00
Don't create a backwards compatibility flag for something that was a
regression bug introduced in release 1.2 llvm-svn: 15219
This commit is contained in:
parent
1bf4f4f9e1
commit
bf99424415
@ -279,10 +279,6 @@ private:
|
|||||||
/// 2^32-1 bytes long.
|
/// 2^32-1 bytes long.
|
||||||
bool hasLongBlockHeaders;
|
bool hasLongBlockHeaders;
|
||||||
|
|
||||||
/// LLVM 1.2 and earlier wrote floating point values in a platform specific
|
|
||||||
/// bit ordering. This was fixed in LLVM 1.3
|
|
||||||
bool hasPlatformSpecificFloatingPoint;
|
|
||||||
|
|
||||||
/// LLVM 1.2 and earlier wrote type slot numbers as vbr_uint32. In LLVM 1.3
|
/// LLVM 1.2 and earlier wrote type slot numbers as vbr_uint32. In LLVM 1.3
|
||||||
/// this has been reduced to vbr_uint24. It shouldn't make much difference
|
/// this has been reduced to vbr_uint24. It shouldn't make much difference
|
||||||
/// since we haven't run into a module with > 24 million types, but for safety
|
/// since we haven't run into a module with > 24 million types, but for safety
|
||||||
|
Loading…
Reference in New Issue
Block a user