Lines Matching refs:we

38 be shared with other memory accesses). As we increase the number of tracepoints
53 Using the 'asm goto', we can create branches that are either taken or not taken
54 by default, without the need to check memory. Then, at run-time, we can patch
57 For example, if we have a simple branch that is disabled by default:
64 straight-line code path. When the branch is 'flipped', we will patch the
146 to take advantage of this optimization. If there is no architecture support, we
237 If we then include the padding bytes, the jump label code saves, 16 total bytes
239 function is 80 bytes long. Thus, we have saved 20% of the instruction
241 really can be a 2-byte no-op since we can reach the branch with a 2-byte jmp.
242 However, we have not yet implemented optimal no-op sizes (they are currently
283 The percentage of saved branches is .7%, and we've saved 12% on
284 'branch-misses'. This is where we would expect to get the most savings, since
285 this optimization is about reducing the number of branches. In addition, we've