Statistics
| Branch: | Revision:

root / tcg @ 5a37532d

Name Size
  arm
  hppa
  i386
  ia64
  mips
  ppc
  ppc64
  s390
  sparc
  tci
LICENSE 146 Bytes
README 15.9 kB
TODO 394 Bytes
optimize.c 34.3 kB
tcg-op.h 94.3 kB
tcg-opc.h 9.6 kB
tcg-runtime.h 825 Bytes
tcg.c 82.4 kB
tcg.h 23.6 kB

Latest revisions

# Date Author Comment
39dc85b9 04/15/2013 09:09 pm Richard Henderson

tcg-ppc64: Handle deposit of zero

The TCG optimizer does great work when inserting constants, being able
to fold the open-coded deposit expansion to just an AND or an OR. Avoid
a bit the regression caused by having the deposit opcode by expanding
deposit of zero as an AND....
027ffea9 04/15/2013 09:09 pm Richard Henderson

tcg-ppc64: Implement movcond

Reviewed-by: Aurelien Jarno <>
Signed-off-by: Richard Henderson <>
1e6e9aca 04/15/2013 09:09 pm Richard Henderson

tcg-ppc64: Use getauxval for ISA detection

Glibc 2.16 includes an easy way to get feature bits previously
buried in /proc or the program startup auxiliary vector. Use it.

Reviewed-by: Aurelien Jarno <>
Signed-off-by: Richard Henderson <>
6c858762 04/15/2013 09:09 pm Richard Henderson

tcg-ppc64: Implement add2/sub2_i64

Reviewed-by: Aurelien Jarno <>
Signed-off-by: Richard Henderson <>
6645c147 04/15/2013 09:09 pm Richard Henderson

tcg-ppc64: Implement mulu2/muls2_i64

Reviewed-by: Aurelien Jarno <>
Signed-off-by: Richard Henderson <>
991041a4 04/15/2013 09:09 pm Richard Henderson

tcg-ppc64: Cleanup i32 constants to tcg_out_cmp

Nothing else in the call chain ensures that these
constants don't have garbage in the high bits.

Reviewed-by: Aurelien Jarno <>
Signed-off-by: Richard Henderson <>
6995a4a0 04/15/2013 09:09 pm Richard Henderson

tcg-ppc64: Use MFOCRF instead of MFCR

It takes half the cycles to read one CR register instead of all 8.
This is a backward compatible addition to the ISA, so chips prior
to Power 2.00 spec will simply continue to read the entire CR register.

Reviewed-by: Aurelien Jarno <>...
70fac59a 04/15/2013 09:09 pm Richard Henderson

tcg-ppc64: Use ISEL for setcond

There are a few simple special cases that should be handled first.
Break these out to subroutines to avoid code duplication.

Reviewed-by: Aurelien Jarno <>
Signed-off-by: Richard Henderson <>
ef809300 04/15/2013 09:09 pm Richard Henderson

tcg-ppc64: Use I constraint for mul

The mul_i32 pattern was loading non-16-bit constants into a register,
when we can get the middle-end to do that for us. The mul_i64 pattern
was not considering that MULLI takes 64-bit inputs.

Reviewed-by: Aurelien Jarno <>...
4c314da6 04/15/2013 09:09 pm Richard Henderson

tcg-ppc64: Use TCGType throughout compares

The optimization/bug being fixed is that tcg_out_cmp was not applying the
right type to loading a constant, in the case it can't be implemented
directly. Rather than recomputing the TCGType enum from the arch64 bool,...

View revisions

Also available in: Atom