diff options
| author | Bernhard Schommer <bernhardschommer@gmail.com> | 2016-08-05 14:05:34 +0200 |
|---|---|---|
| committer | Bernhard Schommer <bernhardschommer@gmail.com> | 2016-08-05 14:05:34 +0200 |
| commit | 028aaefc44b8ed8bafd8b8896fedb53f6e68df3c (patch) | |
| tree | d7f9325da52050a64e5c9ced1017a4f57c674ff3 /runtime/arm/i64_shr.S | |
| parent | 4ac759d0bceef49d16197e3bb8c9767ece693c5e (diff) | |
| download | compcert-kvx-028aaefc44b8ed8bafd8b8896fedb53f6e68df3c.tar.gz compcert-kvx-028aaefc44b8ed8bafd8b8896fedb53f6e68df3c.zip | |
Implement support for big endian arm targets.
Adds support for the big endian arm targets by making the target
endianess flag configurable, adding support for the big endian
calling conventions, rewriting memory access patterns and adding
big endian versions of the runtime functions.
Bug 19418
Diffstat (limited to 'runtime/arm/i64_shr.S')
| -rw-r--r-- | runtime/arm/i64_shr.S | 22 |
1 files changed, 11 insertions, 11 deletions
diff --git a/runtime/arm/i64_shr.S b/runtime/arm/i64_shr.S index a5418f4a..43325092 100644 --- a/runtime/arm/i64_shr.S +++ b/runtime/arm/i64_shr.S @@ -17,7 +17,7 @@ @ * Neither the name of the <organization> nor the @ names of its contributors may be used to endorse or promote products @ derived from this software without specific prior written permission. -@ +@ @ THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS @ "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT @ LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR @@ -36,20 +36,20 @@ #include "sysdeps.h" -@@@ Shift right unsigned +@@@ Shift right unsigned @ Note on ARM shifts: the shift amount is taken modulo 256. @ If shift amount mod 256 >= 32, the shift produces 0. @ Algorithm: @ RL = (XL >> N) | (XH << (32-N) | (XH >> (N-32)) -@ RH = XH >> N +@ RH = XH >> N @ If N = 0: @ RL = XL | 0 | 0 @ RH = XH @ If 1 <= N <= 31: 1 <= 32-N <= 31 and 255 <= N-32 mod 256 <= 255 @ RL = (XL >> N) | (XH >> (32-N) | 0 -@ RH = XH >> N +@ RH = XH >> N @ If N = 32: @ RL = 0 | XH | 0 @ RH = 0 @@ -60,12 +60,12 @@ FUNCTION(__i64_shr) AND r2, r2, #63 @ normalize amount to 0...63 RSB r3, r2, #32 @ r3 = 32 - amount - LSR r0, r0, r2 - LSL r3, r1, r3 - ORR r0, r0, r3 - SUB r3, r2, #32 @ r3 = amount - 32 - LSR r3, r1, r3 - ORR r0, r0, r3 - LSR r1, r1, r2 + LSR Reg0LO, Reg0LO, r2 + LSL r3, Reg0HI, r3 + ORR Reg0LO, Reg0LO, r3 + SUB r3, r2, #32 @ r3 = amount - 32 + LSR r3, Reg0HI, r3 + ORR Reg0LO, Reg0LO, r3 + LSR Reg0HI, Reg0HI, r2 bx lr ENDFUNCTION(__i64_shr) |