From 32f70839e9d14837a7f46233db21b0757a2e9803 Mon Sep 17 00:00:00 2001
From: Mauricio Faria de Oliveira <mauricfo@linux.vnet.ibm.com>
Date: Wed, 30 Nov 2022 08:44:31 +0000
Subject: [PATCH] Skip optimizations for big-endian PowerPC.
MIME-Version: 1.0
Content-Type: text/plain; charset=utf8
Content-Transfer-Encoding: 8bit

Bug-Debian: https://bugs.debian.org/766695
Forwarded: http://sourceforge.net/p/math-atlas/patches/65/
Reviewed-by: Sébastien Villemot <sebastien@debian.org>
Last-Update: 2014-10-28

Some of the existing optimized files/cases for PowerPC
contain assembly instructions with implicit big-endian
behavior - thus incorrect for the little-endian mode -
incurring tests failures during the build (i.e., FTBFS).
This is being worked on; this is the workaround for now.

Note: this is an alternative for Michael Normand's patch
'atlas.3.10.2-ppc64le_do_not_use_files_with_lvx.patch',
in order not to touch file lists/indexes, but only C
source code, so the C preprocessor can avoid code per-arch.

For more details, see:
https://bugzilla.redhat.com/show_bug.cgi?id=1080073#c40

Gbp-Pq: Name ppc64el-ifdef-files-with-lvx.patch
---
 tune/blas/gemm/CASES/ATL_cmm4x4x128_av.c | 7 +++++++
 tune/blas/gemm/CASES/ATL_dmm4x4x80_ppc.c | 6 ++++++
 tune/blas/gemm/CASES/ATL_dmm8x4x2_vsx.c  | 6 ++++++
 tune/blas/gemm/CASES/ATL_smm4x4x128_av.c | 5 +++++
 tune/blas/gemm/CASES/ATL_smm8x2x4_av.c   | 6 ++++++
 5 files changed, 30 insertions(+)

diff --git a/tune/blas/gemm/CASES/ATL_cmm4x4x128_av.c b/tune/blas/gemm/CASES/ATL_cmm4x4x128_av.c
index 59747c0..980d852 100644
--- a/tune/blas/gemm/CASES/ATL_cmm4x4x128_av.c
+++ b/tune/blas/gemm/CASES/ATL_cmm4x4x128_av.c
@@ -7,6 +7,11 @@
  * meaning that this kernel will do an extra store of vector C to memory
  * for load by the scalar FPU.
  */
+
+#if defined(__powerpc64__) && (__BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__)
+/* FIXME: port to little-endian powerpc64. */
+#else
+
 #include "atlas_asm.h"
 
 #if !defined(ATL_AS_OSX_PPC) && !defined(ATL_GAS_LINUX_PPC)
@@ -4257,3 +4262,5 @@ MPEELEDU:
         ld      r3, FST(r1)
         addi    r1, r1, FSIZE
         blr
+
+#endif	/* little-endian powerpc64 */
diff --git a/tune/blas/gemm/CASES/ATL_dmm4x4x80_ppc.c b/tune/blas/gemm/CASES/ATL_dmm4x4x80_ppc.c
index ec2f78b..b5a4ae0 100644
--- a/tune/blas/gemm/CASES/ATL_dmm4x4x80_ppc.c
+++ b/tune/blas/gemm/CASES/ATL_dmm4x4x80_ppc.c
@@ -28,6 +28,10 @@
  *
  */
 
+#if defined(__powerpc64__) && (__BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__)
+/* FIXME: port to little-endian powerpc64. */
+#else
+
 #include "atlas_asm.h"
 /*
  * NOTE: this kernel written by R. Clint Whaley, but it uses two key ideas
@@ -7502,3 +7506,5 @@ DONE:
         addi    r1, r1, FSIZE
 #endif
         blr
+
+#endif  /* little-endian powerpc64 */
diff --git a/tune/blas/gemm/CASES/ATL_dmm8x4x2_vsx.c b/tune/blas/gemm/CASES/ATL_dmm8x4x2_vsx.c
index 2761a39..1feb0ca 100644
--- a/tune/blas/gemm/CASES/ATL_dmm8x4x2_vsx.c
+++ b/tune/blas/gemm/CASES/ATL_dmm8x4x2_vsx.c
@@ -27,6 +27,10 @@
  *
  */
 
+#if defined(__powerpc64__) && (__BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__)
+/* FIXME: port to little-endian powerpc64. */
+#else
+
 #include <altivec.h>
 #include "atlas_misc.h"
 #include "atlas_prefetch.h"               /* ATL_pfl1R, ATL_pfl1W */
@@ -659,3 +663,5 @@ void ATL_USERMM
         }
     }
 }
+
+#endif  /* little-endian powerpc64 */
diff --git a/tune/blas/gemm/CASES/ATL_smm4x4x128_av.c b/tune/blas/gemm/CASES/ATL_smm4x4x128_av.c
index b2bf952..0eb8866 100644
--- a/tune/blas/gemm/CASES/ATL_smm4x4x128_av.c
+++ b/tune/blas/gemm/CASES/ATL_smm4x4x128_av.c
@@ -28,6 +28,9 @@
  *
  */
 
+#if defined(__powerpc64__) && (__BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__)
+/* FIXME: port to little-endian powerpc64. */
+#else
 
 #include "atlas_asm.h"
 
@@ -8340,3 +8343,5 @@ MPEELEDU:
         ld      r3, FST(r1)
         addi    r1, r1, FSIZE
         blr
+
+#endif  /* little-endian powerpc64 */
diff --git a/tune/blas/gemm/CASES/ATL_smm8x2x4_av.c b/tune/blas/gemm/CASES/ATL_smm8x2x4_av.c
index 0e0cb5b..a959929 100644
--- a/tune/blas/gemm/CASES/ATL_smm8x2x4_av.c
+++ b/tune/blas/gemm/CASES/ATL_smm8x2x4_av.c
@@ -27,6 +27,10 @@
  *
  */
 
+#if defined(__powerpc64__) && (__BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__)
+/* FIXME: port to little-endian powerpc64. */
+#else
+
 #include <altivec.h>
 #include "atlas_misc.h"
 #include "atlas_prefetch.h"               /* ATL_pfl1R, ATL_pfl1W */
@@ -326,3 +330,5 @@ void ATL_USERMM
         }
     }
 }
+
+#endif  /* little-endian powerpc64 */
-- 
2.30.2