bootstrap broken on powerpc-linux and powerpc64-linux

Joern RENNECKE joern.rennecke@st.com
Wed Dec 14 15:47:00 GMT 2005


Alan Modra wrote:

> (gdb) p debug_rtx (i2)
>(insn:HI 301 299 603 29 /src/gcc-current/libgfortran/generated/product_c8.c:133 (set (mem:DF (plus:SI (reg/v/f:SI 28 28 [orig:163 dest ] [163])
>                (const_int 8 [0x8])) [17 S8 A64])
>        (reg:DF 45 13 [266])) 328 {*movdf_hardfloat32} (insn_list:REG_DEP_TRUE 300 (nil))
>    (expr_list:REG_DEAD (reg:DF 45 13 [266])
>        (nil)))
>$5 = void
>
>  
>
The match is not attempted with my cross compiler because insn 296 is in 
the way:

 (gdb) call debug_rtx_list(yi,-5)
(insn:HI 299 300 301 29 
/src/gcc-current/libgfortran/generated/product_c8.c:133 (set (mem:DF 
(reg/v/f:SI 28 28 [orig:163 dest ] [163]) [17 S8 A64])
        (reg:DF 32 0 [265])) 257 {*movdf_hardfloat32} 
(insn_list:REG_DEP_TRUE 298 (nil))
    (expr_list:REG_DEAD (reg:DF 32 0 [265])
        (nil)))

(insn:HI 301 299 296 29 
/src/gcc-current/libgfortran/generated/product_c8.c:133 (set (mem:DF 
(plus:SI (reg/v/f:SI 28 28 [orig:163 dest ] [163])
                (const_int 8 [0x8])) [17 S8 A64])
        (reg:DF 45 13 [266])) 257 {*movdf_hardfloat32} 
(insn_list:REG_DEP_TRUE 300 (nil))
    (expr_list:REG_DEAD (reg:DF 45 13 [266])
        (nil)))

(insn 296 301 603 29 
/src/gcc-current/libgfortran/generated/product_c8.c:133 (use 
(symbol_ref/f:SI ("*.LC2") [flags 0x2] <complex_cst 0xb5900f30>)) -1 (nil)
    (nil))

(jump_insn:HI 603 296 604 29 (set (pc)
        (label_ref 342)) 413 {jump} (insn_list:REG_DEP_ANTI 296 
(insn_list:REG_DEP_TRUE 298 (insn_list:REG_DEP_TRUE 300 
(insn_list:REG_DEP_ANTI 299 (insn_list:REG_DEP_ANTI 301 (nil))))))
    (nil))

(barrier:HI 604 603 304)

Since your insn numbers are otherwise the same, I suspect that insn 296 
was also emitted in your compiler, but later deleted or moved.  
Apparently, the code
that does not work correctly either or your or my system.  Could you 
find out when and where insn 296 is moved / deleted on your system?



More information about the Gcc-patches mailing list