Skip to content

S
swiftshader
Switch branch/tag
  1. 30 Jul, 2014 6 commits
  3. 29 Jul, 2014 1 commit
  5. 28 Jul, 2014 2 commits
    • A couple of fixes for using Makefile.standalone on Mac. · 839c4cea
      Jan Voung authored 
      (*) PNaCl toolchain_build builds 64-bit libraries for LLVM on Mac.
    That won't link with subzero code if subzero is built with -m32,
    so add an option to override the -m32.
(*) include locale header
(*) Mark xMacroIntegrityCheck unused to avoid clang compiler warning.
(*) virtual dtor, for inheritable class
(*) Mark compare function const

BUG=none
R=stichnot@chromium.org

Review URL: https://codereview.chromium.org/428733003
    • Subzero: Make Ice::Ostream a typedef for llvm::raw_ostream. · 78282f6c
      Jim Stichnoth authored 
      Previously Ostream was a class that wrapped a raw_ostream pointer,
structured that way in case we wanted to wrap an alternate stream
type.

Also, Ostream used to include a Cfg pointer, but that had to go away
when the Ostream became associated with the GlobalContext which
persists beyond the Cfg lifetime, so the Cfg pointer was removed
leaving only the raw_ostream.

Since llvm::raw_ostream is supposed to be very lightweight, we can
just give up the abstraction and equate it to Ice::Ostream.

BUG= none
R=kschimpf@google.com

Review URL: https://codereview.chromium.org/413393005
  7. 25 Jul, 2014 1 commit
  9. 24 Jul, 2014 4 commits
  11. 23 Jul, 2014 3 commits
  13. 22 Jul, 2014 3 commits
  15. 21 Jul, 2014 1 commit
  17. 18 Jul, 2014 4 commits
  19. 17 Jul, 2014 1 commit
    • Lower the rest of the vector arithmetic operations. · 7fa22d8a
      Matt Wala authored 
      The instructions emitted by the lowering operations require memory
operands to be aligned to 16 bytes. Since there is no support for
aligning memory operands in Subzero, do the arithmetic in registers for
now.

Add vector arithmetic to the arith crosstest. Pass the -mstackrealign
parameter to the crosstest clang so that llc code called back from
Subzero code (helper calls) doesn't assume that the stack is aligned at
the entry to the call.

BUG=none
R=jvoung@chromium.org, stichnot@chromium.org

Review URL: https://codereview.chromium.org/397833002
  21. 16 Jul, 2014 2 commits
    • Lower casting operations that involve vector types. · 83b8036b
      Matt Wala authored 
      Impacted instructions:

bitcast {v4f32, v4i32, v8i16, v16i8} <-> {v4f32, v4i32, v8i16, v16i8}
bitcast v8i1 <-> i8
bitcast v16i1 <-> i16

(There was already code present to handle trivial bitcasts like v16i1 <-> v16i1.)

[sz]ext v4i1 -> v4i32
[sz]ext v8i1 -> v8i16
[sz]ext v16i1 -> v16i8

trunc v4i32 -> v4i1
trunc v8i16 -> v8i1
trunc v16i8 -> v16i1

[su]itofp v4i32 -> v4f32
fpto[su]i v4f32 -> v4i32

Where there is a relatively simple lowering to x86 instructions, it has been used. Otherwise a helper call is used.

Some lowerings require a materialization of a integer vector with 1s in each entry. Since there is no support for vector constant pools, the constant is materialized purely through register operations.

BUG=none
R=jvoung@chromium.org, stichnot@chromium.org

Review URL: https://codereview.chromium.org/383303003
    • Lower bitmanip intrinsics, assuming absence of BMI/SSE4.2 for now. · e4da26f6
      Jan Voung authored 
      We'll need the fallbacks in any case. However, once we've
decided on how to specify the CPU features of the user
machine we can use the nicer LZCNT/TZCNT/POPCNT as well.

Adds cmov, bsf, and bsr instructions.

Calls a popcount helper function for machines without SSE4.2.

Not handling bswap yet (which can also take i16 params).

BUG= https://code.google.com/p/nativeclient/issues/detail?id=3882
R=stichnot@chromium.org, wala@chromium.org

Review URL: https://codereview.chromium.org/390443005
  23. 15 Jul, 2014 2 commits
  25. 14 Jul, 2014 2 commits
  27. 11 Jul, 2014 4 commits
  29. 10 Jul, 2014 1 commit
  31. 09 Jul, 2014 3 commits