crypto/*/Makefile: unify "catch-all" assembler make rules and harmonize

ARM assembler modules.

crypto/aes/Makefile

@@ -68,7 +68,7 @@ aes-parisc.s:	asm/aes-parisc.pl
 	$(PERL) asm/aes-parisc.pl $(PERLASM_SCHEME) $@
 
 # GNU make "catch all"
-aes-%.s:	asm/aes-%.pl;	$(PERL) $< $(CFLAGS) > $@
+aes-%.s:	asm/aes-%.pl;	$(PERL) $< $(PERLASM_SCHEME) $@
 
 files:
 	$(PERL) $(TOP)/util/files.pl Makefile >> $(TOP)/MINFO

crypto/aes/asm/aes-armv4.pl

@@ -22,6 +22,9 @@
 #
 # AES_set_[en|de]crypt_key is added.
 
+while (($output=shift) && ($output!~/^\w[\w\-]*\.\w+$/)) {}
+open STDOUT,">$output";
+
 $s0="r0";
 $s1="r1";
 $s2="r2";
@@ -1029,3 +1032,4 @@ ___
 
 $code =~ s/\bbx\s+lr\b/.word\t0xe12fff1e/gm;	# make it possible to compile with -march=armv4
 print $code;
+close STDOUT;	# enforce flush

crypto/bn/Makefile

@@ -114,7 +114,7 @@ alpha-mont.s:	asm/alpha-mont.pl
 	$(PERL) $< | $(CC) -E - | tee $@ > /dev/null
 
 # GNU make "catch all"
-%-mont.s:	asm/%-mont.pl;	$(PERL) $< $(CFLAGS) > $@
+%-mont.s:	asm/%-mont.pl;	$(PERL) $< $(PERLASM_SCHEME) $@
 
 files:
 	$(PERL) $(TOP)/util/files.pl Makefile >> $(TOP)/MINFO

crypto/modes/Makefile

@@ -55,8 +55,9 @@ ghash-alpha.s:	asm/ghash-alpha.pl
 	$(PERL) $< | $(CC) -E - | tee $@ > /dev/null
 ghash-parisc.s:	asm/ghash-parisc.pl
 	$($PERL) asm/ghash-parisc.pl $(PERLASM_SCHEME) $@
+
 # GNU make "catch all"
-ghash-%.s:	asm/ghash-%.pl;	$(PERL) $< $(PERLASM_SCHEME) $(CFLAGS) > $@
+ghash-%.s:	asm/ghash-%.pl;	$(PERL) $< $(PERLASM_SCHEME) $@
 
 files:
 	$(PERL) $(TOP)/util/files.pl Makefile >> $(TOP)/MINFO

crypto/modes/asm/ghash-armv4.pl

@@ -39,6 +39,9 @@
 # *native* byte order on current platform. See gcm128.c for working
 # example...
 
+while (($output=shift) && ($output!~/^\w[\w\-]*\.\w+$/)) {}
+open STDOUT,">$output";
+
 $Xi="r0";	# argument block
 $Htbl="r1";
 $inp="r2";
@@ -59,9 +62,6 @@ $nhi="r14";
 $rem_4bit=$inp;	# used in gcm_gmult_4bit
 $cnt=$len;
 
-$output=shift;
-open STDOUT,">$output";
-
 sub Zsmash() {
   my $i=12;
   my @args=@_;

crypto/sha/Makefile

@@ -79,9 +79,9 @@ sha256-parisc.s:asm/sha512-parisc.pl;	$(PERL) asm/sha512-parisc.pl $(PERLASM_SCH
 sha512-parisc.s:asm/sha512-parisc.pl;	$(PERL) asm/sha512-parisc.pl $(PERLASM_SCHEME) $@
 
 # GNU make "catch all"
-sha1-%.s:	asm/sha1-%.pl;		$(PERL) $< $@
-sha256-%.s:	asm/sha512-%.pl;	$(PERL) $< $@
-sha512-%.s:	asm/sha512-%.pl;	$(PERL) $< $@
+sha1-%.s:	asm/sha1-%.pl;		$(PERL) $< $(PERLASM_SCHEME) $@
+sha256-%.s:	asm/sha512-%.pl;	$(PERL) $< $(PERLASM_SCHEME) $@
+sha512-%.s:	asm/sha512-%.pl;	$(PERL) $< $(PERLASM_SCHEME) $@
 
 files:
 	$(PERL) $(TOP)/util/files.pl Makefile >> $(TOP)/MINFO

crypto/sha/asm/sha1-armv4-large.pl

@@ -39,7 +39,7 @@
 #	small and always slower.
 # [***]	which is also ~35% better than compiler generated code.
 
-$output=shift;
+while (($output=shift) && ($output!~/^\w[\w\-]*\.\w+$/)) {}
 open STDOUT,">$output";
 
 $ctx="r0";

crypto/sha/asm/sha256-armv4.pl

@@ -13,7 +13,7 @@
 # lute" terms is ~2250 cycles per 64-byte block or ~35 cycles per
 # byte.
 
-$output=shift;
+while (($output=shift) && ($output!~/^\w[\w\-]*\.\w+$/)) {}
 open STDOUT,">$output";
 
 $ctx="r0";	$t0="r0";

crypto/sha/asm/sha512-armv4.pl

@@ -22,7 +22,7 @@ $hi=0;
 $lo=4;
 # ====================================================================
 
-$output=shift;
+while (($output=shift) && ($output!~/^\w[\w\-]*\.\w+$/)) {}
 open STDOUT,">$output";
 
 $ctx="r0";