[PATCH v2 2/3] AArch64: Remove eMAG memset ifunc

[Wilco Dijkstra]

v2: Split patch

As a cleanup remove the eMAG ifunc for memset.

Passes regress, OK for commit?

---

diff --git a/sysdeps/aarch64/multiarch/Makefile b/sysdeps/aarch64/multiarch/Makefile
index bc497782555896bb030139eae3b219577728eb43..988f7cec2562127cdae463b449245aab2a0e00dc 100644
--- a/sysdeps/aarch64/multiarch/Makefile
+++ b/sysdeps/aarch64/multiarch/Makefile
@@ -8,7 +8,6 @@ sysdep_routines += \
   memcpy_sve \
   memmove_mops \
   memset_a64fx \
-  memset_emag \
   memset_generic \
   memset_kunpeng \
   memset_mops \
diff --git a/sysdeps/aarch64/multiarch/ifunc-impl-list.c b/sysdeps/aarch64/multiarch/ifunc-impl-list.c
index b4416b044394f8811fc6f0c416b6580d23e4b0d2..ea5f5853c312e61fcf12458654a76740114c8a4d 100644
--- a/sysdeps/aarch64/multiarch/ifunc-impl-list.c
+++ b/sysdeps/aarch64/multiarch/ifunc-impl-list.c
@@ -50,7 +50,6 @@ __libc_ifunc_impl_list (const char *name, struct libc_ifunc_impl *array,
   IFUNC_IMPL (i, name, memset,
 	      IFUNC_IMPL_ADD (array, i, memset, (zva_size == 64), __memset_zva64)
 	      IFUNC_IMPL_ADD (array, i, memset, (zva_size == 64), __memset_oryon1)
-	      IFUNC_IMPL_ADD (array, i, memset, 1, __memset_emag)
 	      IFUNC_IMPL_ADD (array, i, memset, 1, __memset_kunpeng)
 	      IFUNC_IMPL_ADD (array, i, memset, sve && zva_size == 256, __memset_a64fx)
 	      IFUNC_IMPL_ADD (array, i, memset, sve && zva_size == 64, __memset_sve_zva64)
diff --git a/sysdeps/aarch64/multiarch/memset.c b/sysdeps/aarch64/multiarch/memset.c
index 76c6532b811e3bfab31bb60b4270305c8b60a811..438845eec9e2d273272a270a0f87833e26ea8c84 100644
--- a/sysdeps/aarch64/multiarch/memset.c
+++ b/sysdeps/aarch64/multiarch/memset.c
@@ -30,7 +30,6 @@
 extern __typeof (__redirect_memset) __libc_memset;
 
 extern __typeof (__redirect_memset) __memset_zva64 attribute_hidden;
-extern __typeof (__redirect_memset) __memset_emag attribute_hidden;
 extern __typeof (__redirect_memset) __memset_kunpeng attribute_hidden;
 extern __typeof (__redirect_memset) __memset_a64fx attribute_hidden;
 extern __typeof (__redirect_memset) __memset_generic attribute_hidden;
@@ -61,9 +60,6 @@ select_memset_ifunc (void)
   if (IS_KUNPENG920 (midr))
     return __memset_kunpeng;
 
-  if (IS_EMAG (midr))
-    return __memset_emag;
-
   if (zva_size == 64)
     return __memset_zva64;
 
diff --git a/sysdeps/aarch64/multiarch/memset_emag.S b/sysdeps/aarch64/multiarch/memset_emag.S
deleted file mode 100644
index 8f68de6f72d0f75acc068351c8689b004bdbab0d..0000000000000000000000000000000000000000
--- a/sysdeps/aarch64/multiarch/memset_emag.S
+++ /dev/null
@@ -1,115 +0,0 @@
-/* Optimized memset for AmpereComputing emag processor.
-   Copyright (C) 2018-2026 Free Software Foundation, Inc.
-
-   This file is part of the GNU C Library.
-
-   The GNU C Library is free software; you can redistribute it and/or
-   modify it under the terms of the GNU Lesser General Public
-   License as published by the Free Software Foundation; either
-   version 2.1 of the License, or (at your option) any later version.
-
-   The GNU C Library is distributed in the hope that it will be useful,
-   but WITHOUT ANY WARRANTY; without even the implied warranty of
-   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
-   Lesser General Public License for more details.
-
-   You should have received a copy of the GNU Lesser General Public
-   License along with the GNU C Library.  If not, see
-   <https://www.gnu.org/licenses/>.  */
-
-#include <sysdep.h>
-
-/* Assumptions:
- *
- * ARMv8-a, AArch64, unaligned accesses
- *
- */
-
-#define dstin	x0
-#define val	x1
-#define valw	w1
-#define count	x2
-#define dst	x3
-#define dstend	x4
-
-ENTRY (__memset_emag)
-
-	bfi	valw, valw, 8, 8
-	bfi	valw, valw, 16, 16
-	bfi	val, val, 32, 32
-
-	add	dstend, dstin, count
-
-	cmp	count, 96
-	b.hi	L(set_long)
-	cmp	count, 16
-	b.hs	L(set_medium)
-
-	/* Set 0..15 bytes.  */
-	tbz	count, 3, 1f
-	str	val, [dstin]
-	str	val, [dstend, -8]
-	ret
-
-	.p2align 3
-1:	tbz	count, 2, 2f
-	str	valw, [dstin]
-	str	valw, [dstend, -4]
-	ret
-2:	cbz	count, 3f
-	strb	valw, [dstin]
-	tbz	count, 1, 3f
-	strh	valw, [dstend, -2]
-3:	ret
-
-	.p2align 3
-	/* Set 16..96 bytes.  */
-L(set_medium):
-	stp	val, val, [dstin]
-	tbnz	count, 6, L(set96)
-	stp	val, val, [dstend, -16]
-	tbz	count, 5, 1f
-	stp	val, val, [dstin, 16]
-	stp	val, val, [dstend, -32]
-1:	ret
-
-	.p2align 4
-	/* Set 64..96 bytes.  Write 64 bytes from the start and
-	   32 bytes from the end.  */
-L(set96):
-	stp	val, val, [dstin, 16]
-	stp	val, val, [dstin, 32]
-	stp	val, val, [dstin, 48]
-	stp	val, val, [dstend, -32]
-	stp	val, val, [dstend, -16]
-	ret
-
-	.p2align 4
-L(set_long):
-	stp	val, val, [dstin]
-	bic	dst, dstin, 15
-	/* Small-size or non-zero memset does not use DC ZVA. */
-	sub	count, dstend, dst
-
-	/*
-	 * Adjust count and bias for loop. By subtracting extra 1 from count,
-	 * it is easy to use tbz instruction to check whether loop tailing
-	 * count is less than 33 bytes, so as to bypass 2 unnecessary stps.
-	 */
-	sub	count, count, 64+16+1
-
-1:	stp	val, val, [dst, 16]
-	stp	val, val, [dst, 32]
-	stp	val, val, [dst, 48]
-	stp	val, val, [dst, 64]!
-	subs	count, count, 64
-	b.hs	1b
-
-	tbz	count, 5, 1f	/* Remaining count is less than 33 bytes? */
-	stp	val, val, [dst, 16]
-	stp	val, val, [dst, 32]
-1:	stp	val, val, [dstend, -32]
-	stp	val, val, [dstend, -16]
-	ret
-
-END (__memset_emag)