gcc-2.95 19990602 (prerelease) on alpha-dec-osf4.0d

[Jeffrey A Law]

  In message < 11EB515FBDB3D111BF270000F84A9ED603D24745@shrexc1.shr.dec.com >you 
write:
  > I am seeing this bug on both alphaev56-dec-osf4.0d and alphaev6-dec-osf4.0e
  > .
  > I am using the 6/8/99 snapshot:
  > 
  > stage1/xgcc -Bstage1/
  > -B/cadunx/hansolo/people/jfitz/fakey/alphaev6-dec-osf4.0e/bin/ -c  -DIN_GCC
  > -DHAIFA    -O2 -g -O2  -DHAVE_CONFIG_H    -I.
  > -I/cadunx/hansolo/people/jfitz/gnu/egcs-19990608/gcc
  > -I/cadunx/hansolo/people/jfitz/gnu/egcs-19990608/gcc/config
  > -I/cadunx/hansolo/people/jfitz/gnu/egcs-19990608/gcc/../include
  > /cadunx/hansolo/people/jfitz/gnu/egcs-19990608/gcc/gcov.c
  > /cadunx/hansolo/people/jfitz/gnu/egcs-19990608/gcc/gcov.c: In function
  > `output_data':
  > /cadunx/hansolo/people/jfitz/gnu/egcs-19990608/gcc/gcov.c:1403: fixed or
  > forbidden register 32 ($f0) was spilled for class FLOAT_REGS.
  > /cadunx/hansolo/people/jfitz/gnu/egcs-19990608/gcc/gcov.c:1403: This may be
  > due to a compiler bug or to impossible asm
  > /cadunx/hansolo/people/jfitz/gnu/egcs-19990608/gcc/gcov.c:1403: statements
  > or clauses.
  > /cadunx/hansolo/people/jfitz/gnu/egcs-19990608/gcc/gcov.c:1403: This is the
  > instruction:
  > (insn 2638 2634 18 (set (reg:DF 398)
  >         (mem/u:DF (reg:DI 384) 0)) 279 {movsf-2} (insn_list 2632 (nil))
  >     (expr_list:REG_EQUIV (const_double:DF (mem/u:DF (symbol_ref/u:DI
  > ("*$LC28")) 0) 4636737291354636288 [0x4059000000000000] 0 [0x0] [100])
  >         (nil)))
  > make[2]: *** [gcov.o] Error 1
  > make[2]: Leaving directory `/cadunx/hansolo/people/jfitz/gnu/objdir/gcc'
  > make[1]: *** [bootstrap] Error 2
  > make[1]: Leaving directory `/cadunx/hansolo/people/jfitz/gnu/objdir/gcc'
  > make: *** [bootstrap] Error 2
  > 754.623u 166.621s 29:40.90 51.7% 1+35k 21145+50214io 2537pf+0w
  > watman:/cadunx/hansolo/people/jfitz/gnu/objdir>
This should be fixed now.
jeff