c/7321: ice on vectors with -O3 and -m3dnow

c/7321: ice on vectors with -O3 and -m3dnow

[Andrew Pinski]

>Number:         7321
>Category:       c
>Synopsis:       ice on vectors with -O3 and -m3dnow
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    unassigned
>State:          open
>Class:          ice-on-legal-code
>Submitter-Id:   net
>Arrival-Date:   Mon Jul 15 18:16:02 PDT 2002
>Closed-Date:
>Last-Modified:
>Originator:     Andrew Pinski
>Release:        3.2 20020715 (experimental)
>Organization:
Home
>Environment:
System: Linux lead.geop.uc.edu 2.4.18-0.13 #1 Mon Apr 1 14:59:55 EST 2002 i686 unknown
Architecture: i686

	
host: i686-pc-linux-gnu
build: i686-pc-linux-gnu
target: i686-pc-linux-gnu
configured with: /home/gates/pinskia/src/gnu/gcc/src/configure --target=i686-pc-linux-gnu --prefix=/home/gates/pinskia/linux --enable-threads=posix --enable-shared
>Description:
gcc -O3 -m3dnow vector.c produces the following errorvector.c: In function `tempf':
vector.c:20: error: unable to find a register to spill in class `FLOAT_REGS'
(insn:HI 29 27 30 0 0x4007ae80 (set (reg:SF 8 st(0) [70])
        (minus:SF (subreg:SF (reg/v:DI 30 rmm1 [66]) 0)
            (subreg:SF (reg/v:DI 30 rmm1 [66]) 4))) 556 {*fop_sf_1_nosse} (insn_list 24 (nil))
    (expr_list:REG_EQUIV (mem:SF (reg/v/f:SI 2 ecx [60]) [3 S4 A32])
        (expr_list:REG_DEAD (reg/v:DI 30 rmm1 [66])
            (nil))))
vector.c:20: internal compiler error: Internal compiler error in spill_failure, at reload1.c:1908
Please submit a full bug report,
with preprocessed source if appropriate.
See <URL:http://www.gnu.org/software/gcc/bugs.html> for instructions.

>How-To-Repeat:
gcc -O3 -m3dnow vector.c
vector.c follows:
        floatvect2 temp={x[0],x[1]};
        floatvect2 temp1={y[0],y[1]};
        resfloatvect2 temp2;
        temp2.vector=temp+temp1;
        x[0]=temp2.f[0];
        x[1]=temp2.f[1];

typedef float floatvect2 __attribute__((mode(V2SF)));

typedef union
{
    floatvect2 vector;
    float f[2];
}resfloatvect2;

void tempf(float *a, float *b, float *c)
{
    floatvect2 a1=*(floatvect2*)a;
    floatvect2 b1=*(floatvect2*)b;
    floatvect2 rb1={b[1],b[0]};
    resfloatvect2 r1;
    resfloatvect2 r2;
    r1.vector=a1*b1;
    r2.vector=a1*rb1;
    c[0]=r1.f[0]-r1.f[1];
    c[1]=r2.f[0]+r2.f[1];
}

>Fix:
unknown
>Release-Note:
>Audit-Trail:
>Unformatted: