Remove stray code alignment (.align 16 and .align 4 directives)

from i386 and x86_64.

8 files changed, 14 insertions, 11 deletions

diff --git a/libc/string/sparc/_glibc_inc.h b/libc/string/sparc/_glibc_inc.h
index 4eb4d755c..e0aef52c2 100644
--- a/libc/string/sparc/_glibc_inc.h
+++ b/libc/string/sparc/_glibc_inc.h
@@ -6,6 +6,8 @@
 #include <features.h>
 #include <bits/wordsize.h>
 
+/* Is alignment really needed? */
+
 #if __WORDSIZE == 32
 # define ENTRY_ALIGN 4
 #else
diff --git a/libc/string/x86_64/_glibc_inc.h b/libc/string/x86_64/_glibc_inc.h
index 88cef2ea3..415ce90a7 100644
--- a/libc/string/x86_64/_glibc_inc.h
+++ b/libc/string/x86_64/_glibc_inc.h
@@ -6,15 +6,8 @@
 #include <features.h>
 #include <bits/wordsize.h>
 
-#if __WORDSIZE == 32
-# define ENTRY_ALIGN 4
-#else
-# define ENTRY_ALIGN 2
-#endif
-
 #define ENTRY(sym) \
 	.global sym; \
-	.align  ENTRY_ALIGN; \
 	.type   sym,%function; \
 	sym:
 
diff --git a/libc/string/x86_64/memset.S b/libc/string/x86_64/memset.S
index d72d74468..0f054ade6 100644
--- a/libc/string/x86_64/memset.S
+++ b/libc/string/x86_64/memset.S
@@ -71,7 +71,7 @@ ENTRY (memset)
 	jae	11f
 
 	.p2align 4
-3:	/* Copy 64 bytes.  */
+3:	/* Fill 64 bytes.  */
 	mov	%r8,(%rcx)
 	mov	%r8,0x8(%rcx)
 	mov	%r8,0x10(%rcx)
@@ -84,7 +84,7 @@ ENTRY (memset)
 	dec	%rax
 	jne	3b
 
-4:	/* Copy final bytes.  */
+4:	/* Fill final bytes.  */
 	and	$0x3f,%edx
 	mov	%rdx,%rax
 	shr	$0x3,%rax
@@ -107,7 +107,7 @@ ENTRY (memset)
 	jne	8b
 9:
 #if BZERO_P
-	nop
+	nop	/* huh?? */
 #else
 	/* Load result (only if used as memset).  */
 	mov	%rdi,%rax	/* start address of destination is result */
@@ -115,7 +115,7 @@ ENTRY (memset)
 	retq
 
 	.p2align 4
-11:	/* Copy 64 bytes without polluting the cache.  */
+11:	/* Fill 64 bytes without polluting the cache.  */
 	/* We could use	movntdq    %xmm0,(%rcx) here to further
 	   speed up for large cases but let's not use XMM registers.  */
 	movnti	%r8,(%rcx)
diff --git a/libc/string/x86_64/strcat.S b/libc/string/x86_64/strcat.S
index 9b0068981..ddec8ee79 100644
--- a/libc/string/x86_64/strcat.S
+++ b/libc/string/x86_64/strcat.S
@@ -21,6 +21,7 @@
 
 #include "_glibc_inc.h"
 
+/* Seems to be unrolled too much */
 
 	.text
 ENTRY (BP_SYM (strcat))
diff --git a/libc/string/x86_64/strchr.S b/libc/string/x86_64/strchr.S
index 8e59c4c19..c357bfd03 100644
--- a/libc/string/x86_64/strchr.S
+++ b/libc/string/x86_64/strchr.S
@@ -20,6 +20,7 @@
 
 #include "_glibc_inc.h"
 
+/* Seems to be unrolled too much */
 
 	.text
 ENTRY (BP_SYM (strchr))
diff --git a/libc/string/x86_64/strcpy.S b/libc/string/x86_64/strcpy.S
index d9a51b0bb..920d0543a 100644
--- a/libc/string/x86_64/strcpy.S
+++ b/libc/string/x86_64/strcpy.S
@@ -20,6 +20,8 @@
 
 #include "_glibc_inc.h"
 
+/* Seems to be unrolled too much */
+
 #ifndef USE_AS_STPCPY
 # define STRCPY strcpy
 #endif
diff --git a/libc/string/x86_64/strcspn.S b/libc/string/x86_64/strcspn.S
index fed12b5f6..c02fbbb5f 100644
--- a/libc/string/x86_64/strcspn.S
+++ b/libc/string/x86_64/strcspn.S
@@ -25,6 +25,9 @@
 
 #include "_glibc_inc.h"
 
+/* Seems to be unrolled too much */
+/* (and why testb $0xff, %cl, not testb %cl, %cl?) */
+
 /* BEWARE: `#ifdef strcspn' means that strcspn is redefined as `strpbrk' */
 #define STRPBRK_P (defined strcspn)
 
diff --git a/libc/string/x86_64/strlen.S b/libc/string/x86_64/strlen.S
index 0441dc46c..e79537f87 100644
--- a/libc/string/x86_64/strlen.S
+++ b/libc/string/x86_64/strlen.S
@@ -20,6 +20,7 @@
 
 #include "_glibc_inc.h"
 
+/* Seems to be unrolled too much */
 
 	.text
 ENTRY (strlen)