Page MenuHomePhabricator

koviankevin (GuanHong Liu)
User

Projects

User does not belong to any projects.

User Details

User Since
Feb 12 2014, 9:44 PM (344 w, 6 d)

Recent Activity

Jan 22 2015

koviankevin added a comment to D2804: [COMPILER-RT] Implement __fixtfsi, __fixunstfsi.

OK, but the following two cases in the test will fail if there's no saturation logic in fixunstfsi. I think it's better to remove them, thanks.

Jan 22 2015, 7:35 PM

Jan 21 2015

koviankevin added a comment to D2804: [COMPILER-RT] Implement __fixtfsi, __fixunstfsi.

Yes, it's not resonable to implement saturation logic for fixtfsi but not for fixuntfsi.

Jan 21 2015, 11:53 PM

Nov 23 2014

koviankevin added a comment to D2804: [COMPILER-RT] Implement __fixtfsi, __fixunstfsi.

Saturation logic for _fixunstfsi is also needed but it has been removed in previous revision. (i.e, range check part in fixuntfsi.cpp)

Nov 23 2014, 11:00 PM

Sep 18 2014

koviankevin added a comment to D2804: [COMPILER-RT] Implement __fixtfsi, __fixunstfsi.

__fixunstfsi should have range check, too.

Sep 18 2014, 7:42 PM

Sep 17 2014

koviankevin added a comment to D2804: [COMPILER-RT] Implement __fixtfsi, __fixunstfsi.

Since these functions are implemented for using compiler-rt instead of libgcc, and __fixtfsi returns INT_MAX/INT_MIN when the input is out of range in libgcc, I think range check is still needed to avoid inconsistent results.

Sep 17 2014, 12:05 AM

Sep 15 2014

koviankevin added a comment to D2805: [COMPILER-RT] Implement __floatsitf, __floatunstfsi.

This patch blocks quad math stuff on aarch64 from using compiler-rt instead of gnu libgcc
Can this patch be commited, now? Thanks

Sep 15 2014, 1:03 AM
koviankevin added a comment to D2804: [COMPILER-RT] Implement __fixtfsi, __fixunstfsi.

This patch blocks quad math stuff on aarch64 from using compiler-rt instead of gnu libgcc
Can this patch be commited, now? Thanks

Sep 15 2014, 1:03 AM

Jun 16 2014

koviankevin added a comment to D2800: [COMPILER-RT] Implement __divtf3.

Sorry for leaving this hanging so long.
I just refered the double-pecision version(i.e, divdf3) to write this function
Since it used 32-bit instead of the highest 8-bit in divdf3, I used 64-bit constant here.

Jun 16 2014, 1:07 AM
koviankevin retitled D4154: [COMPILER-RT] add myself in CREDITS.txt from to [COMPILER-RT] add myself in CREDITS.txt.
Jun 16 2014, 12:42 AM

May 28 2014

koviankevin added inline comments to D2800: [COMPILER-RT] Implement __divtf3.
May 28 2014, 7:52 PM

May 27 2014

koviankevin added a comment to D3886: [COMPILER-RT] refactoring mulsf3.c and muldf3.c.

NO, I don't have access. Could you help me to commit these patches? Thanks

May 27 2014, 7:01 PM

May 25 2014

koviankevin updated the diff for D2805: [COMPILER-RT] Implement __floatsitf, __floatunstfsi.

Adjusted as discussed

May 25 2014, 8:00 PM
koviankevin updated the diff for D2804: [COMPILER-RT] Implement __fixtfsi, __fixunstfsi.

fix typo

May 25 2014, 7:57 PM

May 23 2014

koviankevin updated the diff for D2804: [COMPILER-RT] Implement __fixtfsi, __fixunstfsi.

use UINT64_C and drop le64

May 23 2014, 1:15 AM
koviankevin updated the diff for D2805: [COMPILER-RT] Implement __floatsitf, __floatunstfsi.

use UINT64_C and drop le64

May 23 2014, 1:15 AM
koviankevin updated the diff for D2803: [COMPILER-RT] Implement __trunctfdf2, __trunctfsf2.

use UINT64_C and drop le64
truncdfsf2.c is refactored in D3888

May 23 2014, 1:13 AM
koviankevin retitled D3888: [COMPILER-RT] refactoring truncdfsf2.c from to [COMPILER-RT] refactoring truncdfsf2.c.
May 23 2014, 1:12 AM
koviankevin retitled D2802: [COMPILER-RT] Implement __extendsftf2, __extenddftf2 from [COMPILER_RT] Implement __extendsftf2, __extenddftf2 to [COMPILER-RT] Implement __extendsftf2, __extenddftf2.

use UINT64_C and drop le64
extendsfdf2.c is refactored in D3887

May 23 2014, 1:11 AM
koviankevin retitled D3887: [COMPILER-RT] refactoring extendsfdf2.c from to [COMPILER-RT] refactoring extendsfdf2.c.
May 23 2014, 1:09 AM
koviankevin updated the diff for D2800: [COMPILER-RT] Implement __divtf3.

use UINT64_C and drop le64

May 23 2014, 1:08 AM
koviankevin updated the diff for D2799: [COMPILER-RT] Implement __multf3.

use UINT64_C and drop le64
muldf3.c and mulsf3.c are refactored in D3886

May 23 2014, 1:06 AM
koviankevin retitled D3886: [COMPILER-RT] refactoring mulsf3.c and muldf3.c from to [COMPILER-RT] refactoring mulsf3.c and muldf3.c.
May 23 2014, 1:05 AM
koviankevin updated the diff for D2798: [COMPILER-RT] Implement __addtf3 and __subtf3.

use UINT64_C and drop le64
adddf3.c and addsf3.c are refactored in D3885

May 23 2014, 1:03 AM
koviankevin retitled D3885: [COMPILER-RT] refactoring addsf3.c and adddf3.c from to [COMPILER-RT] refactoring addsf3.c and adddf3.c.
May 23 2014, 1:01 AM