paweld/fpc
1
0
Fork 0
mirror of https://gitlab.com/freepascal.org/fpc/source.git synced 2025-04-14 03:19:27 +02:00
Code Issues Packages Projects Releases Wiki Activity

* emit MUL instead of IMUL in ti8086addnode.second_mul when overflow checking is

Browse Source 
off and the multiplication is 16->16

git-svn-id: trunk@26495 -
This commit is contained in:
nickysn 2014-01-17 23:04:46 +00:00
parent 76c40a49db
commit 3f4111ac37
1 changed files with 7 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
compiler/i8086/n8086add.pas
View File

@ -616,6 +616,13 @@ interface
begin
pass_left_right;
{ MUL is faster than IMUL on the 8086 & 8088 (and equal in speed on 286+),
but it's only safe to use in place of IMUL when overflow checking is off
and we're doing a 16-bit>16-bit multiplication }
if not (cs_check_overflow in current_settings.localswitches) and
(not is_32bitint(resultdef)) then
unsigned:=true;
{The location.register will be filled in later (JM)}
location_reset(location,LOC_REGISTER,def_cgsize(resultdef));
{ Mul supports registers and references, so if not register/reference,