On Tue, Jun 21, 2016 at 9:51 PM, H.J. Lu <hjl.to...@gmail.com> wrote:
>> I have attached my version of the patch. It handles all your
>> testcases, plus &foo+1 case. Bootstrap is still running.
>>
>> Does the patch work for you?
>
> It works.
Attached version of the patch was committed to mainline SVN. Regarding
the testcases - I have made them to compile on non-ia32 only ATM.
Let's change them when ia32 support is added (it is a trivial change).
2016-06-23 Uros Bizjak <ubiz...@gmail.com>
H.J. Lu <hongjiu...@intel.com>
PR target/67400
* config/i386/i386-protos.h (ix86_force_load_from_GOT_p): New.
* config/i386/i386.c (ix86_force_load_from_GOT_p): New function.
(ix86_legitimate_constant_p): Do not allow UNSPEC_GOTPCREL if
ix86_force_load_from_GOT_p returns true.
(ix86_legitimate_address_p): Allow UNSPEC_GOTPCREL if
ix86_force_load_from_GOT_p returns true.
(ix86_print_operand_address_as): Support UNSPEC_GOTPCREL if
ix86_force_load_from_GOT_p returns true.
(ix86_expand_move): Load the external function address via the
GOT slot if ix86_force_load_from_GOT_p returns true.
* config/i386/predicates.md (x86_64_immediate_operand): Return
false for SYMBOL_REFs where ix86_force_load_from_GOT_p returns true.
(x86_64_zext_immediate_operand): Ditto.
testsuite/ChangeLog:
2016-06-23 H.J. Lu <hongjiu...@intel.com>
PR target/67400
* gcc.target/i386/pr67400-1.c: New test.
* gcc.target/i386/pr67400-2.c: Likewise.
* gcc.target/i386/pr67400-3.c: Likewise.
* gcc.target/i386/pr67400-4.c: Likewise.
* gcc.target/i386/pr67400-5.c: Likewise.
* gcc.target/i386/pr67400-6.c: Likewise.
* gcc.target/i386/pr67400-7.c: Likewise.
Bootstrapped and regression tested on x86_64-linux-gnu {,-m32}.
Committed to mainline SVN.
Uros.
Index: config/i386/i386-protos.h
===================================================================
--- config/i386/i386-protos.h (revision 237716)
+++ config/i386/i386-protos.h (working copy)
@@ -70,6 +70,7 @@ extern bool ix86_expand_set_or_movmem (rtx, rtx, r
extern bool constant_address_p (rtx);
extern bool legitimate_pic_operand_p (rtx);
extern bool legitimate_pic_address_disp_p (rtx);
+extern bool ix86_force_load_from_GOT_p (rtx);
extern void print_reg (rtx, int, FILE*);
extern void ix86_print_operand (FILE *, rtx, int);
Index: config/i386/i386.c
===================================================================
--- config/i386/i386.c (revision 237716)
+++ config/i386/i386.c (working copy)
@@ -15120,6 +15120,19 @@ darwin_local_data_pic (rtx disp)
&& XINT (disp, 1) == UNSPEC_MACHOPIC_OFFSET);
}
+/* True if operand X should be loaded from GOT. */
+
+bool
+ix86_force_load_from_GOT_p (rtx x)
+{
+ return (TARGET_64BIT && !TARGET_PECOFF && !TARGET_MACHO
+ && !flag_plt && !flag_pic
+ && ix86_cmodel != CM_LARGE
+ && GET_CODE (x) == SYMBOL_REF
+ && SYMBOL_REF_FUNCTION_P (x)
+ && !SYMBOL_REF_LOCAL_P (x));
+}
+
/* Determine if a given RTX is a valid constant. We already know this
satisfies CONSTANT_P. */
@@ -15188,6 +15201,12 @@ ix86_legitimate_constant_p (machine_mode mode, rtx
if (MACHO_DYNAMIC_NO_PIC_P)
return machopic_symbol_defined_p (x);
#endif
+
+ /* External function address should be loaded
+ via the GOT slot to avoid PLT. */
+ if (ix86_force_load_from_GOT_p (x))
+ return false;
+
break;
CASE_CONST_SCALAR_INT:
@@ -15596,6 +15615,9 @@ ix86_legitimate_address_p (machine_mode, rtx addr,
return false;
case UNSPEC_GOTPCREL:
+ if (ix86_force_load_from_GOT_p (XVECEXP (XEXP (disp, 0), 0, 0)))
+ goto is_legitimate_pic;
+ /* FALLTHRU */
case UNSPEC_PCREL:
gcc_assert (flag_pic);
goto is_legitimate_pic;
@@ -18169,6 +18191,12 @@ ix86_print_operand_address_as (FILE *file, rtx add
fputs ("ds:", file);
fprintf (file, HOST_WIDE_INT_PRINT_DEC, INTVAL (disp));
}
+ /* Load the external function address via the GOT slot to avoid PLT. */
+ else if (GET_CODE (disp) == CONST
+ && GET_CODE (XEXP (disp, 0)) == UNSPEC
+ && XINT (XEXP (disp, 0), 1) == UNSPEC_GOTPCREL
+ && ix86_force_load_from_GOT_p (XVECEXP (XEXP (disp, 0), 0, 0)))
+ output_pic_addr_const (file, disp, 0);
else if (flag_pic)
output_pic_addr_const (file, disp, 0);
else
@@ -19417,6 +19445,15 @@ ix86_expand_move (machine_mode mode, rtx operands[
if (model)
op1 = legitimize_tls_address (op1, model, true);
+ else if (ix86_force_load_from_GOT_p (op1))
+ {
+ /* Load the external function address via GOT slot to avoid PLT. */
+ op1 = gen_rtx_UNSPEC (Pmode, gen_rtvec (1, op1),
+ UNSPEC_GOTPCREL);
+ op1 = gen_rtx_CONST (Pmode, op1);
+ op1 = gen_const_mem (Pmode, op1);
+ set_mem_alias_set (op1, ix86_GOT_alias_set ());
+ }
else
{
tmp = legitimize_pe_coff_symbol (op1, addend != NULL_RTX);
Index: config/i386/predicates.md
===================================================================
--- config/i386/predicates.md (revision 237716)
+++ config/i386/predicates.md (working copy)
@@ -160,13 +160,18 @@
return trunc_int_for_mode (val, SImode) == val;
}
case SYMBOL_REF:
+ /* TLS symbols are not constant. */
+ if (SYMBOL_REF_TLS_MODEL (op))
+ return false;
+
+ /* Load the external function address via the GOT slot. */
+ if (ix86_force_load_from_GOT_p (op))
+ return false;
+
/* For certain code models, the symbolic references are known to fit.
in CM_SMALL_PIC model we know it fits if it is local to the shared
library. Don't count TLS SYMBOL_REFs here, since they should fit
only if inside of UNSPEC handled below. */
- /* TLS symbols are not constant. */
- if (SYMBOL_REF_TLS_MODEL (op))
- return false;
return (ix86_cmodel == CM_SMALL || ix86_cmodel == CM_KERNEL
|| (ix86_cmodel == CM_MEDIUM && !SYMBOL_REF_FAR_ADDR_P (op)));
@@ -207,6 +212,11 @@
/* TLS symbols are not constant. */
if (SYMBOL_REF_TLS_MODEL (op1))
return false;
+
+ /* Load the external function address via the GOT slot. */
+ if (ix86_force_load_from_GOT_p (op1))
+ return false;
+
/* For CM_SMALL assume that latest object is 16MB before
end of 31bits boundary. We may also accept pretty
large negative constants knowing that all objects are
@@ -273,10 +283,15 @@
return !(INTVAL (op) & ~(HOST_WIDE_INT) 0xffffffff);
case SYMBOL_REF:
- /* For certain code models, the symbolic references are known to fit. */
/* TLS symbols are not constant. */
if (SYMBOL_REF_TLS_MODEL (op))
return false;
+
+ /* Load the external function address via the GOT slot. */
+ if (ix86_force_load_from_GOT_p (op))
+ return false;
+
+ /* For certain code models, the symbolic references are known to fit. */
return (ix86_cmodel == CM_SMALL
|| (ix86_cmodel == CM_MEDIUM
&& !SYMBOL_REF_FAR_ADDR_P (op)));
@@ -301,6 +316,11 @@
/* TLS symbols are not constant. */
if (SYMBOL_REF_TLS_MODEL (op1))
return false;
+
+ /* Load the external function address via the GOT slot. */
+ if (ix86_force_load_from_GOT_p (op1))
+ return false;
+
/* For small code model we may accept pretty large positive
offsets, since one bit is available for free. Negative
offsets are limited by the size of NULL pointer area
Index: testsuite/gcc.target/i386/pr67400-1.c
===================================================================
--- testsuite/gcc.target/i386/pr67400-1.c (nonexistent)
+++ testsuite/gcc.target/i386/pr67400-1.c (working copy)
@@ -0,0 +1,13 @@
+/* { dg-do compile { target { *-*-linux* && { ! ia32 } } } } */
+/* { dg-options "-O2 -fno-pic -fno-plt" } */
+
+extern void bar (void);
+
+void *
+foo (void)
+{
+ return &bar;
+}
+
+/* { dg-final { scan-assembler "mov\(l|q\)\[ \t\]*bar@GOTPCREL" } } */
+/* { dg-final { scan-assembler-not "\(mov|lea\)\(l|q\)\[ \t\]*\(\\\$|\)bar," }
} */
Index: testsuite/gcc.target/i386/pr67400-2.c
===================================================================
--- testsuite/gcc.target/i386/pr67400-2.c (nonexistent)
+++ testsuite/gcc.target/i386/pr67400-2.c (working copy)
@@ -0,0 +1,14 @@
+/* { dg-do compile { target { *-*-linux* && { ! ia32 } } } } */
+/* { dg-options "-O2 -fno-pic -fno-plt" } */
+
+extern void bar (void);
+extern void *p;
+
+void
+foo (void)
+{
+ p = &bar;
+}
+
+/* { dg-final { scan-assembler "mov\(l|q\)\[ \t\]*bar@GOTPCREL" } } */
+/* { dg-final { scan-assembler-not "mov\(l|q\)\[ \t\]*\\\$bar," } } */
Index: testsuite/gcc.target/i386/pr67400-3.c
===================================================================
--- testsuite/gcc.target/i386/pr67400-3.c (nonexistent)
+++ testsuite/gcc.target/i386/pr67400-3.c (working copy)
@@ -0,0 +1,16 @@
+/* { dg-do compile { target { *-*-linux* && { ! ia32 } } } } */
+/* { dg-options "-O2 -fno-pic -fno-plt" } */
+
+static void
+bar (void)
+{
+}
+
+void *
+foo (void)
+{
+ return &bar;
+}
+
+/* { dg-final { scan-assembler "mov\(l|q\)\[ \t\]*\\\$bar," } } */
+/* { dg-final { scan-assembler-not "mov\(l|q\)\[ \t\]*bar@GOTPCREL" } } */
Index: testsuite/gcc.target/i386/pr67400-4.c
===================================================================
--- testsuite/gcc.target/i386/pr67400-4.c (nonexistent)
+++ testsuite/gcc.target/i386/pr67400-4.c (working copy)
@@ -0,0 +1,13 @@
+/* { dg-do compile { target { *-*-linux* && { ! ia32 } } } } */
+/* { dg-options "-O2 -fno-pic -fno-plt" } */
+
+extern void bar (void) __attribute__ ((visibility ("hidden")));
+
+void *
+foo (void)
+{
+ return &bar;
+}
+
+/* { dg-final { scan-assembler "mov\(l|q\)\[ \t\]*\\\$bar," } } */
+/* { dg-final { scan-assembler-not "mov\(l|q\)\[ \t\]*bar@GOTPCREL" } } */
Index: testsuite/gcc.target/i386/pr67400-5.c
===================================================================
--- testsuite/gcc.target/i386/pr67400-5.c (nonexistent)
+++ testsuite/gcc.target/i386/pr67400-5.c (working copy)
@@ -0,0 +1,11 @@
+/* { dg-do compile { target { *-*-linux* && { ! ia32 } } } } */
+/* { dg-options "-O2 -fno-pic -fno-plt" } */
+
+extern void foo (void);
+extern void bar (int, int, int, int, int, int, void *);
+
+void
+x (void)
+{
+ bar (1, 2, 3, 4, 5, 6, foo);
+}
Index: testsuite/gcc.target/i386/pr67400-6.c
===================================================================
--- testsuite/gcc.target/i386/pr67400-6.c (nonexistent)
+++ testsuite/gcc.target/i386/pr67400-6.c (working copy)
@@ -0,0 +1,13 @@
+/* { dg-do compile { target { *-*-linux* && { ! ia32 } } } } */
+/* { dg-options "-O2 -fno-pic -fno-plt" } */
+
+extern int bar (void);
+
+int
+check (void *p)
+{
+ return p != &bar;
+}
+
+/* { dg-final { scan-assembler "cmp\(l|q\)\[ \t\]*.*bar@GOTPCREL" } } */
+/* { dg-final { scan-assembler-not "mov\(l|q\)\[ \t\]*\\\$bar," } } */
Index: testsuite/gcc.target/i386/pr67400-7.c
===================================================================
--- testsuite/gcc.target/i386/pr67400-7.c (nonexistent)
+++ testsuite/gcc.target/i386/pr67400-7.c (working copy)
@@ -0,0 +1,13 @@
+/* { dg-do compile { target { *-*-linux* && { ! ia32 } } } } */
+/* { dg-options "-O2 -fno-pic -fno-plt" } */
+
+extern void bar (void);
+
+void *
+foo (void)
+{
+ return &bar+1;
+}
+
+/* { dg-final { scan-assembler "mov\(l|q\)\[ \t\]*bar@GOTPCREL" } } */
+/* { dg-final { scan-assembler-not "\(mov|lea\)\(l|q\)\[ \t\]*\(\\\$|\)bar," }
} */
