Message ID | 1297563761-2786-2-git-send-email-laurent@vivier.eu |
---|---|
State | New |
Headers | show |
On Sun, Feb 13, 2011 at 4:22 AM, Laurent Vivier <laurent@vivier.eu> wrote: > Datatype alignment can be found using following application: > > int main(void) > { > printf("alignof(short) %ld\n", __alignof__(short)); > printf("alignof(int) %ld\n", __alignof__(int)); > printf("alignof(long) %ld\n", __alignof__(long)); > printf("alignof(long long) %ld\n", __alignof__(long long)); > } > > This patch includes following alignments: > > i386 > > alignof(short) 2 > alignof(int) 4 > alignof(long) 4 > alignof(long long) 8 > > x86_64 > > alignof(short) 2 > alignof(int) 4 > alignof(long) 8 > alignof(long long) 8 > > arm > > alignof(short) 2 > alignof(int) 4 > alignof(long) 4 > alignof(long long) 4 > > m68k (680x0) > > alignof(short) 2 > alignof(int) 2 > alignof(long) 2 > alignof(long long) 2 > > mips > > alignof(short) 2 > alignof(int) 4 > alignof(long) 4 > alignof(long long) 8 > > ppc > > alignof(short) 2 > alignof(int) 4 > alignof(long) 4 > alignof(long long) 8 > > for other targets, use by default (2,4,4,8). > > Please, update for your favorite target... For Sparc32 (I think also sparc32plus), the default is OK. For Sparc64, please use 2, 4, 8, 8. I'd guess other 64 bit platforms (Alpha, MIPS64, PPC64 etc) should use the same. Does GCC produce correct code using the attributes on strictly aligned host, when the target is less strictly aligned? Should the alignment of floating point variables be specified as well? The strict alignment required for doubles is 4, but recommended alignment is 8, I'm not sure which one is used for structures containing doubles.
Le dimanche 13 février 2011 à 10:24 +0200, Blue Swirl a écrit : > On Sun, Feb 13, 2011 at 4:22 AM, Laurent Vivier <laurent@vivier.eu> wrote: > > Datatype alignment can be found using following application: > > > > int main(void) > > { > > printf("alignof(short) %ld\n", __alignof__(short)); > > printf("alignof(int) %ld\n", __alignof__(int)); > > printf("alignof(long) %ld\n", __alignof__(long)); > > printf("alignof(long long) %ld\n", __alignof__(long long)); > > } > > > > This patch includes following alignments: > > > > i386 > > > > alignof(short) 2 > > alignof(int) 4 > > alignof(long) 4 > > alignof(long long) 8 > > > > x86_64 > > > > alignof(short) 2 > > alignof(int) 4 > > alignof(long) 8 > > alignof(long long) 8 > > > > arm > > > > alignof(short) 2 > > alignof(int) 4 > > alignof(long) 4 > > alignof(long long) 4 > > > > m68k (680x0) > > > > alignof(short) 2 > > alignof(int) 2 > > alignof(long) 2 > > alignof(long long) 2 > > > > mips > > > > alignof(short) 2 > > alignof(int) 4 > > alignof(long) 4 > > alignof(long long) 8 > > > > ppc > > > > alignof(short) 2 > > alignof(int) 4 > > alignof(long) 4 > > alignof(long long) 8 > > > > for other targets, use by default (2,4,4,8). > > > > Please, update for your favorite target... > > For Sparc32 (I think also sparc32plus), the default is OK. > > For Sparc64, please use 2, 4, 8, 8. I'd guess other 64 bit platforms > (Alpha, MIPS64, PPC64 etc) should use the same. OK, I update my patch. > Does GCC produce correct code using the attributes on strictly aligned > host, when the target is less strictly aligned? It seems it is OK. I did some tests into a mips-linux-user chroot (sparc one is broken ;-) ) : mips is a strictly aligned host, from gcc/config/mips/mips.h: #define STRICT_ALIGNMENT 1 aligments are: alignof(short) 2 alignof(int) 4 alignof(long) 4 alignof(long long) 8 We try to align int on 2. #include <stdio.h> typedef int target_int __attribute__((aligned(2))); struct { char Z; target_int A; } B; int main(void) { B.A = 0xdeadbeaf; printf("%d: %x\n", __alignof__(B.A), B.A); } ./test_align 2: deadbeaf disass: lw $2,%got(B)($28) li $3,-559087616 # 0xffffffffdead0000 ori $3,$3,0xbeaf swl $3,2($2) swr $3,5($2) normal case: lw $3,%got(B)($28) li $2,-559087616 # 0xffffffffdead0000 ori $2,$2,0xbeaf sw $2,4($3) So, gcc seems smart enough to split the memory access in several ones compatible with the strict alignment rules. > Should the alignment of floating point variables be specified as well? At the moment it seems useless. > The strict alignment required for doubles is 4, but recommended > alignment is 8, I'm not sure which one is used for structures > containing doubles. if necessary, some tests will be helpfull. Thank you for your comments. Regards, Laurent
diff --git a/configure b/configure index 25381e5..efc6fa4 100755 --- a/configure +++ b/configure @@ -2919,6 +2919,10 @@ target_nptl="no" interp_prefix1=`echo "$interp_prefix" | sed "s/%M/$target_arch2/g"` echo "CONFIG_QEMU_INTERP_PREFIX=\"$interp_prefix1\"" >> $config_target_mak gdb_xml_files="" +target_short_alignment=2 +target_int_alignment=4 +target_long_alignment=4 +target_long_long_alignment=8 TARGET_ARCH="$target_arch2" TARGET_BASE_ARCH="" @@ -2931,6 +2935,7 @@ case "$target_arch2" in x86_64) TARGET_BASE_ARCH=i386 target_phys_bits=64 + target_long_alignment=8 ;; alpha) target_phys_bits=64 @@ -2942,6 +2947,7 @@ case "$target_arch2" in target_nptl="yes" gdb_xml_files="arm-core.xml arm-vfp.xml arm-vfp3.xml arm-neon.xml" target_phys_bits=32 + target_long_long_alignment=4 ;; cris) target_nptl="yes" @@ -2951,6 +2957,9 @@ case "$target_arch2" in bflt="yes" gdb_xml_files="cf-core.xml cf-fp.xml" target_phys_bits=32 + target_int_alignment=2 + target_long_alignment=2 + target_long_long_alignment=2 ;; microblaze) bflt="yes" @@ -3029,6 +3038,10 @@ case "$target_arch2" in exit 1 ;; esac +echo "TARGET_SHORT_ALIGNMENT=$target_short_alignment" >> $config_target_mak +echo "TARGET_INT_ALIGNMENT=$target_int_alignment" >> $config_target_mak +echo "TARGET_LONG_ALIGNMENT=$target_long_alignment" >> $config_target_mak +echo "TARGET_LONG_LONG_ALIGNMENT=$target_long_long_alignment" >> $config_target_mak echo "TARGET_ARCH=$TARGET_ARCH" >> $config_target_mak target_arch_name="`echo $TARGET_ARCH | tr '[:lower:]' '[:upper:]'`" echo "TARGET_$target_arch_name=y" >> $config_target_mak diff --git a/cpu-defs.h b/cpu-defs.h index 8d4bf86..98c6718 100644 --- a/cpu-defs.h +++ b/cpu-defs.h @@ -37,16 +37,22 @@ #define TARGET_LONG_SIZE (TARGET_LONG_BITS / 8) +typedef int16_t target_short __attribute__ ((aligned(TARGET_SHORT_ALIGNMENT))); +typedef uint16_t target_ushort __attribute__((aligned(TARGET_SHORT_ALIGNMENT))); +typedef int32_t target_int __attribute__((aligned(TARGET_INT_ALIGNMENT))); +typedef uint32_t target_uint __attribute__((aligned(TARGET_INT_ALIGNMENT))); +typedef int64_t target_long_long __attribute__((aligned(TARGET_LONG_LONG_ALIGNMENT))); +typedef uint64_t target_ulong_long __attribute__((aligned(TARGET_LONG_LONG_ALIGNMENT))); /* target_ulong is the type of a virtual address */ #if TARGET_LONG_SIZE == 4 -typedef int32_t target_long; -typedef uint32_t target_ulong; +typedef int32_t target_long __attribute__((aligned(TARGET_LONG_ALIGNMENT))); +typedef uint32_t target_ulong __attribute__((aligned(TARGET_LONG_ALIGNMENT))); #define TARGET_FMT_lx "%08x" #define TARGET_FMT_ld "%d" #define TARGET_FMT_lu "%u" #elif TARGET_LONG_SIZE == 8 -typedef int64_t target_long; -typedef uint64_t target_ulong; +typedef int64_t target_long __attribute__((aligned(TARGET_LONG_ALIGNMENT))); +typedef uint64_t target_ulong __attribute__((aligned(TARGET_LONG_ALIGNMENT))); #define TARGET_FMT_lx "%016" PRIx64 #define TARGET_FMT_ld "%" PRId64 #define TARGET_FMT_lu "%" PRIu64
Datatype alignment can be found using following application: int main(void) { printf("alignof(short) %ld\n", __alignof__(short)); printf("alignof(int) %ld\n", __alignof__(int)); printf("alignof(long) %ld\n", __alignof__(long)); printf("alignof(long long) %ld\n", __alignof__(long long)); } This patch includes following alignments: i386 alignof(short) 2 alignof(int) 4 alignof(long) 4 alignof(long long) 8 x86_64 alignof(short) 2 alignof(int) 4 alignof(long) 8 alignof(long long) 8 arm alignof(short) 2 alignof(int) 4 alignof(long) 4 alignof(long long) 4 m68k (680x0) alignof(short) 2 alignof(int) 2 alignof(long) 2 alignof(long long) 2 mips alignof(short) 2 alignof(int) 4 alignof(long) 4 alignof(long long) 8 ppc alignof(short) 2 alignof(int) 4 alignof(long) 4 alignof(long long) 8 for other targets, use by default (2,4,4,8). Please, update for your favorite target... Signed-off-by: Laurent Vivier <laurent@vivier.eu> --- v2: compute align size for each basic datatype configure | 13 +++++++++++++ cpu-defs.h | 14 ++++++++++---- 2 files changed, 23 insertions(+), 4 deletions(-)