提交歷史

作者 SHA1 備註 提交日期
  florian 9f587625e7 + handle not and neg in isFoldableArithOp as well 10 年之前
  florian de17e9fa1c + cpu capability CPUX86_HAS_CMOV 10 年之前
  florian 06fd4223e9 + optimization setting level4 10 年之前
  florian 35a04ad095 * modified patch of Paul W to replace leave by mov/pop, resolves #26455 10 年之前
  sergei 2ee0c8de45 * i386: For integer comparisons with zero, emit "test $-1,%reg" instead of "test %reg,%reg". It is more spilling-friendly, because it transforms into "test $-1,spilltemp" and does not require a register. 11 年之前
  florian cbefdcf887 * patch by Martin Friebe to avoid wrong debugger output, resolves #25584 11 年之前
  nickysn 1de847c754 * fixed error in comment 11 年之前
  florian 4d5119bf1c * fixes several issues which cause warnings by the dfa code when using it to detect uninitialized variables 11 年之前
  Jonas Maebe 0302453912 * disabled the lea->add transformation added in r25907 because it breaks 11 年之前
  florian 1843ed59b5 * set p to a valid instruction 11 年之前
  florian a16ceb6cdf * replace lea by add if possible 11 年之前
  florian 80f8f66c6e * set file positions correctly in PeepHoleOptPass1 11 年之前
  florian 4148637c8c * fold mov/lea/mov as well if possible 11 年之前
  florian 524cde8140 + MatchReference 11 年之前
  florian 440cb33e80 * continue to use lea to adjust the stack pointer, it is the recommended way 11 年之前
  florian 5eb578a44c * generate sub instead of add from lea if appropriate 11 年之前
  florian d575a435c3 * some debugging infrastructure 11 年之前
  florian bfbb0c5b9d * optimize mov/lea 11 年之前
  nickysn f288859fba * i386 optimization - remove the test/or %y, %y instruction in the sequence: 11 年之前
  florian 507edb16de * use add/sub instead of inc/dec on modern CPUs when optimizing for speed 12 年之前
  florian 2ea6979277 * handle Movzl/Op/Mov optimization correctly, resolves #23431 12 年之前
  florian df01717c38 * mov x, %treg; mov %treg, y might be only optimized if y does not use treg 13 年之前
  Jonas Maebe dd03bc3e66 * i386 version of fix from r21113 13 年之前
  florian a6b2fcfe22 * fixes breakage of r17766, should resolve #19598 14 年之前
  florian b00a610aec * optimize andb/w const,reg andw/l const,reg sequences as well 14 年之前
  Jonas Maebe d10f46ec59 * skip instructions containing a reference with a segment prefix (fixes 14 年之前
  Jonas Maebe 8f22a5d224 * perform simple imul->lea transformations also for Pentium 3-class target 14 年之前
  Jonas Maebe ffc9e07275 * only don't perform the complex mul->lea optimizations for the Pentium3 15 年之前
  Jonas Maebe 7d459cf12a * the compiler now explicitly keeps track of the minimally guaranteed 16 年之前
  Jonas Maebe 7bfe1f7865 * fixed indentation 17 年之前