[clang] [llvm] Remove mmx 3dnow (PR #96246)

2024-07-11 Thread James Y Knight via cfe-commits
https://github.com/jyknight edited https://github.com/llvm/llvm-project/pull/96246 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [llvm] Remove mmx 3dnow (PR #96246)

2024-06-25 Thread Simon Pilgrim via cfe-commits
RKSimon wrote: > > So can we get all the mmx patches done in time for the 19.0 branch? > > Maybe? Depends on how the discussion on the other PR goes -- whether that > proposal is acceptable, or if a different transition strategy is desired. Apart from #96540 how many more patches do you envisi

[clang] [llvm] Remove mmx 3dnow (PR #96246)

2024-06-24 Thread Craig Topper via cfe-commits
@@ -6,16 +6,11 @@ ; RUN: llc < %s -mtriple=i686-- -mcpu=slm | FileCheck %s -check-prefix=X86-PRFCHWSSE ; RUN: llc < %s -mtriple=i686-- -mcpu=btver2 | FileCheck %s -check-prefix=X86-PRFCHWSSE ; RUN: llc < %s -mtriple=i686-- -mcpu=btver2 -mattr=-prfchw | FileCheck %s -check-pr

[clang] [llvm] Remove mmx 3dnow (PR #96246)

2024-06-24 Thread Craig Topper via cfe-commits
@@ -6,16 +6,11 @@ ; RUN: llc < %s -mtriple=i686-- -mcpu=slm | FileCheck %s -check-prefix=X86-PRFCHWSSE ; RUN: llc < %s -mtriple=i686-- -mcpu=btver2 | FileCheck %s -check-prefix=X86-PRFCHWSSE ; RUN: llc < %s -mtriple=i686-- -mcpu=btver2 -mattr=-prfchw | FileCheck %s -check-pr

[clang] [llvm] Remove mmx 3dnow (PR #96246)

2024-06-24 Thread Craig Topper via cfe-commits
topperc wrote: This LGTM. The prefetch mess was one thing I knew was a little complicated, but it looks like you've taken care of that. https://github.com/llvm/llvm-project/pull/96246 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lis

[clang] [llvm] Remove mmx 3dnow (PR #96246)

2024-06-24 Thread Craig Topper via cfe-commits
topperc wrote: Was `3dnow` never part of `sys:getHostCPUFeatures` for -march=native? I expected you would need to remove it from there, but it looks like it isn't there https://github.com/llvm/llvm-project/pull/96246 ___ cfe-commits mailing list cfe-

[clang] [llvm] Remove mmx 3dnow (PR #96246)

2024-06-24 Thread James Y Knight via cfe-commits
jyknight wrote: > I'd prefer we rip off ALL the 3dnow/mmx bandaid in one big series of patches > - and not split across the 19.X/20.X releases. The 3dnow part is a lot easier -- because it's not trying to be compatible, just dropping support entirely. And I don't think anyone will care, since

[clang] [llvm] Remove mmx 3dnow (PR #96246)

2024-06-24 Thread James Y Knight via cfe-commits
@@ -1825,32 +1818,32 @@ def : ProcModel; -def : Proc<"k6-2", [FeatureX87, FeatureCX8, Feature3DNow], +def : Proc<"k6-2", [FeatureX87, FeatureCX8, FeatureMMX], jyknight wrote: Done. https://github.com/llvm/llvm-project/pull/96246 _

[clang] [llvm] Remove mmx 3dnow (PR #96246)

2024-06-21 Thread James Y Knight via cfe-commits
@@ -7,151 +7,16 @@ *===---=== */ +// 3dNow intrinsics are no longer supported, and this header remains only as a +// stub for users who were including it to get to _m_prefetch or +// _m_prefetchw. Such uses

[clang] [llvm] Remove mmx 3dnow (PR #96246)

2024-06-21 Thread Simon Pilgrim via cfe-commits
@@ -1825,32 +1818,32 @@ def : ProcModel; -def : Proc<"k6-2", [FeatureX87, FeatureCX8, Feature3DNow], +def : Proc<"k6-2", [FeatureX87, FeatureCX8, FeatureMMX], RKSimon wrote: I think the best we can do is add FeaturePRFCHW as well? https://github.com/llvm/llvm-pr

[clang] [llvm] Remove mmx 3dnow (PR #96246)

2024-06-21 Thread James Y Knight via cfe-commits
@@ -1825,32 +1818,32 @@ def : ProcModel; -def : Proc<"k6-2", [FeatureX87, FeatureCX8, Feature3DNow], +def : Proc<"k6-2", [FeatureX87, FeatureCX8], jyknight wrote: Oops! I had fixed that but in a not-yet-committed diff in my checkout. (Tests fail, otherwise!) ht

[clang] [llvm] Remove mmx 3dnow (PR #96246)

2024-06-21 Thread via cfe-commits
github-actions[bot] wrote: :warning: C/C++ code formatter, clang-format found issues in your code. :warning: You can test this locally with the following command: ``bash git-clang-format --diff af6acd7442646fde56de919964bd52d7bb7922b2 8a0252f670f7339a8c8b63db9bbb9a8c9fd0e8d6 --e

[clang] [llvm] Remove mmx 3dnow (PR #96246)

2024-06-21 Thread Simon Pilgrim via cfe-commits
@@ -1825,32 +1818,32 @@ def : ProcModel; -def : Proc<"k6-2", [FeatureX87, FeatureCX8, Feature3DNow], +def : Proc<"k6-2", [FeatureX87, FeatureCX8], RKSimon wrote: These still need FeatureMMX and SSEPrefetch (or a similar flag) https://github.com/llvm/llvm-project

[clang] [llvm] Remove mmx 3dnow (PR #96246)

2024-06-21 Thread Simon Pilgrim via cfe-commits
@@ -7,151 +7,16 @@ *===---=== */ +// 3dNow intrinsics are no longer supported, and this header remains only as a +// stub for users who were including it to get to _m_prefetch or +// _m_prefetchw. Such uses

[clang] [llvm] Remove mmx 3dnow (PR #96246)

2024-06-21 Thread Simon Pilgrim via cfe-commits
RKSimon wrote: I'd prefer we rip off ALL the 3dnow/mmx bandaid in one big series of patches - and not split across the 19.X/20.X releases. So can we get all the mmx patches done in time for the 19.0 branch? Also, this "might" have bigger impact than the KNL changes - should we announce the p

[clang] [llvm] Remove mmx 3dnow (PR #96246)

2024-06-21 Thread Phoebe Wang via cfe-commits
https://github.com/phoebewang commented: This is much like what we have removed for KNL intrinsics/instructions., so general LGTM. But I'd like Simon to sign off given I'm not familar with 3DNOW instructions. https://github.com/llvm/llvm-project/pull/96246 _

[clang] [llvm] Remove mmx 3dnow (PR #96246)

2024-06-20 Thread via cfe-commits
llvmbot wrote: @llvm/pr-subscribers-clang-codegen @llvm/pr-subscribers-clang-driver Author: James Y Knight (jyknight) Changes Remove support for 3DNow!, both intrinsics and builtins. This set of instructions was only supported by AMD chips starting in the K6-2 (introduced 1998), and befor