On 22/08/2017 11:57, Jozef Lawrynowicz wrote:
On 02/08/2017 17:45, Joseph Myers wrote:
On Wed, 2 Aug 2017, Jeff Law wrote:
I think Joseph's suggestion for looking at partial float handling may be
useful, though ia64's RFmode may be more interesting as it's not a
multiple of 8 in bitsize. IF/KF modes on the ppc port have similar
properties.
The key issue those floating-point modes engage is the meaning of
precision. IFmode and KFmode have precision set to 106 and 113 to
distinguish them. But that's precision in the sense of number of
mantissa
bits; as normally understood in GCC, precision should be the number of
significant bits, so 128 for both those modes (but having multiple
different binary floating-point modes with the same precision would
require changing how we deal with laying out long double, so the target
specifies a mode for floating-point types rather than leaving it to be
deduced from values such as LONG_DOUBLE_TYPE_SIZE).
Thanks for the advice, I'm looking into how the ppc KFmode behaves in
this situation now.
I also looked through the front end code a bit more, and the behaviour
of stor-layout.c:layout_type for RECORD_TYPE looks likes a smoking gun
to me.
For BOOLEAN_TYPE, INTEGER_TYPE, ENUMERAL_TYPE, REAL_TYPE,
FIXED_POINT_TYPE etc. layout_type sets:
TYPE_SIZE (type) = bitsize_int (GET_MODE_BITSIZE (TYPE_MODE (type)));
But for the individual field types in RECORD_TYPE, UNION_TYPE or
QUAL_UNION_TYPE, this same setting of TYPE_SIZE and friends is not
performed.
So a field in a RECORD_TYPE might be an INTEGER_TYPE, but TYPE_SIZE for
this INTEGER_TYPE will not be set as it would have been had the type not
been a field in a RECORD_TYPE.
So the fix to me looks to be to ensure that the code for the base types
(BOOLEAN_TYPE, INTEGER_TYPE etc.) from layout_type is also executed for
each type that happens to be a field in a RECORD/UNION/QUAL_UNION_TYPE.
Actually, the issue turned out to be that TYPE_SIZE is specifically set
in the first place.
When the internal data structures to handle __intN types are initialised
in tree.c, the compiler is also setting TYPE_SIZE.
For the other "standard" types, layout_type sets TYPE_SIZE. So rather
than specifically setting TYPE_SIZE in tree.c, I've removed this code so
TYPE_SIZE will get set like it does for every other type.
If the attached patch is acceptable, I would appreciate if someone could
commit it for me, as I do not have write access.
Thanks,
Jozef
From 5437c7ffa48f974c6960a1e308c4cdf0ea0a2648 Mon Sep 17 00:00:00 2001
From: Jozef Lawrynowicz <joze...@somniumtech.com>
Date: Thu, 24 Aug 2017 11:40:04 +0000
Subject: [PATCH] MSP430: Dont specifically set TYPE_SIZE for __intN types
2017-08-XX Jozef Lawrynowicz <joze...@somniumtech.com>
PR target/78849
* gcc/tree.c (build_common_tree_nodes): Dont set TYPE_SIZE for __intN
types.
gcc/testsuite
2017-08-XX Jozef Lawrynowicz <joze...@somniumtech.com>
PR target/78849
* gcc.target/msp430/msp430.exp: Remove -pedantic-errors from
DEFAULT_CFLAGS.
* gcc.target/msp430/pr78849.c: New test.
---
gcc/testsuite/gcc.target/msp430/msp430.exp | 13 +++++---
gcc/testsuite/gcc.target/msp430/pr78849.c | 50 ++++++++++++++++++++++++++++++
gcc/tree.c | 2 --
3 files changed, 59 insertions(+), 6 deletions(-)
create mode 100644 gcc/testsuite/gcc.target/msp430/pr78849.c
diff --git a/gcc/testsuite/gcc.target/msp430/msp430.exp
b/gcc/testsuite/gcc.target/msp430/msp430.exp
index e54d531..3be8711 100644
--- a/gcc/testsuite/gcc.target/msp430/msp430.exp
+++ b/gcc/testsuite/gcc.target/msp430/msp430.exp
@@ -24,10 +24,15 @@ if { ![istarget msp430-*-*] } then {
# Load support procs.
load_lib gcc-dg.exp
-# If a testcase doesn't have special options, use these.
+# The '-pedantic-errors' option in the global variable DEFAULT_CFLAGS that is
+# set by other drivers causes an error when the __int20 type is used, so remove
+# this option from DEFAULT_CFLAGS for the msp430 tests.
global DEFAULT_CFLAGS
-if ![info exists DEFAULT_CFLAGS] then {
- set DEFAULT_CFLAGS ""
+if [info exists DEFAULT_CFLAGS] then {
+ set MSP430_DEFAULT_CFLAGS \
+ [ string map { "-pedantic-errors" "" } $DEFAULT_CFLAGS ]
+} else {
+ set MSP430_DEFAULT_CFLAGS ""
}
# Initialize `dg'.
@@ -35,7 +40,7 @@ dg-init
# Main loop.
dg-runtest [lsort [glob -nocomplain $srcdir/$subdir/*.\[cCS\]]] \
- "" $DEFAULT_CFLAGS
+ "" $MSP430_DEFAULT_CFLAGS
# All done.
dg-finish
diff --git a/gcc/testsuite/gcc.target/msp430/pr78849.c
b/gcc/testsuite/gcc.target/msp430/pr78849.c
new file mode 100644
index 0000000..f70f0bb
--- /dev/null
+++ b/gcc/testsuite/gcc.target/msp430/pr78849.c
@@ -0,0 +1,50 @@
+/* { dg-do compile } */
+/* { dg-final { scan-assembler ".size.*instance.*52" } } */
+
+struct t_inner
+{
+ __int20 a;
+ char val1;
+ __int20 b[3];
+ char val2;
+};
+
+struct t_full
+{
+ __int20 array[2];
+ char val1;
+ struct t_inner bb[2];
+ char val2;
+};
+
+struct t_full instance =
+{
+ {
+ 4231,
+ 3212,
+ },
+ 5,
+ {
+ {
+ 87680,
+ 20,
+ {
+ 2534,
+ 3,
+ 41,
+ },
+ 55,
+ },
+ {
+ 567,
+ 4,
+ {
+ 43522,
+ 5433,
+ 454,
+ },
+ 88,
+ },
+ },
+ 8,
+};
diff --git a/gcc/tree.c b/gcc/tree.c
index 4f56892..1c085ba 100644
--- a/gcc/tree.c
+++ b/gcc/tree.c
@@ -9526,8 +9526,6 @@ build_common_tree_nodes (bool signed_char)
{
int_n_trees[i].signed_type = make_signed_type (int_n_data[i].bitsize);
int_n_trees[i].unsigned_type = make_unsigned_type
(int_n_data[i].bitsize);
- TYPE_SIZE (int_n_trees[i].signed_type) = bitsize_int
(int_n_data[i].bitsize);
- TYPE_SIZE (int_n_trees[i].unsigned_type) = bitsize_int
(int_n_data[i].bitsize);
if (int_n_data[i].bitsize > LONG_LONG_TYPE_SIZE
&& int_n_enabled_p[i])
--
1.8.3.1
