Commits

Breadcrumbs

History for

ond485e7b

User selector

All users

Datepicker

All time

Commit History

Commits on Jan 31, 2011

• add a note, progress unblocked by PR8575 being fixed.

  Show description for fd2ad87
  lattner

  committed
  fd2ad87

  Copy full SHA for fd2ad87

Commits on Jan 30, 2011

• Teach DAGCombine to fold fold (sra (trunc (sr x, c1)), c2) -> (trunc (sra x, c1+c2) when c1 equals the amount of bits that are truncated off.

  Show description for 9b108a3
  d0k

  committed
  9b108a3

  Copy full SHA for 9b108a3

Commits on Jan 24, 2011

• this isn't a memset, we do convert dest[i] to one though :)

  Show description for 96c0771
  lattner

  committed
  96c0771

  Copy full SHA for 96c0771

• with recent work, we now optimize this into:

  Show description for 7ea7d32
  lattner

  committed
  7ea7d32

  Copy full SHA for 7ea7d32

Commits on Jan 23, 2011

• Add a memset loop that LoopIdiomRecognize doesn't recognize.

  Show description for f330b32
  

  Anders Carlsson

  committed
  f330b32

  Copy full SHA for f330b32

Commits on Jan 18, 2011

• add a note

  Show description for 908d8e8
  lattner

  committed
  908d8e8

  Copy full SHA for 908d8e8

Commits on Jan 16, 2011

• Update README.txt to remove the DAE enhancement.

  Show description for 5822b86
  

  Anders Carlsson

  committed
  5822b86

  Copy full SHA for 5822b86

• add some commentary

  Show description for b0daffc
  lattner

  committed
  b0daffc

  Copy full SHA for b0daffc

• Simplify a README.txt entry significantly to expose the core issue.

  Show description for cad33c6
  chandlerc

  committed
  cad33c6

  Copy full SHA for cad33c6

Commits on Jan 13, 2011

• typo

  Show description for 4c5456a
  lattner

  committed
  4c5456a

  Copy full SHA for 4c5456a

• memcpy + metadata = bliss :)

  Show description for 4cd18f9
  lattner

  committed
  4cd18f9

  Copy full SHA for 4cd18f9

Commits on Jan 11, 2011

• Teach constant folding to perform conversions from constant floating

  Show description for 15ed90c
  chandlerc

  committed
  15ed90c

  Copy full SHA for 15ed90c

• Fix a random missed optimization by making InstCombine more aggressive when determining which bits are demanded by

  Show description for da1c122
  resistor

  committed
  da1c122

  Copy full SHA for da1c122

Commits on Jan 10, 2011

• +0.0 vs -0.0 differences can be handled by looking at the user of the

  Show description for 4a6fb94
  lattner

  committed
  4a6fb94

  Copy full SHA for 4a6fb94

• expand on a note

  Show description for 00a35d0
  lattner

  committed
  00a35d0

  Copy full SHA for 00a35d0

Commits on Jan 9, 2011

• typo

  Show description for 66d7a57
  lattner

  committed
  66d7a57

  Copy full SHA for 66d7a57

• xref a PR #

  Show description for 7cc4e30
  lattner

  committed
  7cc4e30

  Copy full SHA for 7cc4e30

• Add a note about the inability to model FP -> int conversions which

  Show description for fb00e27
  chandlerc

  committed
  fb00e27

  Copy full SHA for fb00e27

• Add a note about a missed FP optimization.

  Show description for 96b1b6c
  chandlerc

  committed
  96b1b6c

  Copy full SHA for 96b1b6c

• Another missed memset in std::vector initialization.

  Show description for 694d753
  chandlerc

  committed
  694d753

  Copy full SHA for 694d753

• Fix a cut-paste-o so that the sample code is correct for my last note.

  Show description for 27a2a13
  chandlerc

  committed
  27a2a13

  Copy full SHA for 27a2a13

• Another missed optimization of trivial vector code.

  Show description for e5ca494
  chandlerc

  committed
  e5ca494

  Copy full SHA for e5ca494

• Add a note about vector's size-constructor producing dead stores.

  Show description for d8723a9
  chandlerc

  committed
  d8723a9

  Copy full SHA for d8723a9

• Add a note about a missed memset optimization from std::fill.

  Show description for 75fbd37
  chandlerc

  committed
  75fbd37

  Copy full SHA for 75fbd37

Commits on Jan 7, 2011

• Revert 122959, it needs more thought. Add it back to README.txt with additional notes.

  Show description for eaff66a
  d0k

  committed
  eaff66a

  Copy full SHA for eaff66a

Commits on Jan 6, 2011

• With Benjamin's recent amazing patches, we should be able to do even better things :)

  Show description for 01cdc20
  lattner

  committed
  01cdc20

  Copy full SHA for 01cdc20

• Add a note from llvmdev, this time with more info.

  Show description for fa36680
  d0k

  committed
  fa36680

  Copy full SHA for fa36680

• EarlyCSE does this now (and GVN always did it).

  Show description for 7d9f51f
  d0k

  committed
  7d9f51f

  Copy full SHA for 7d9f51f

• InstCombine: If we call llvm.objectsize on a malloc call we can replace it with the size passed to malloc.

  Show description for 240d42d
  d0k

  committed
  240d42d

  Copy full SHA for 240d42d

• add a note about object size from drystone, add a poorly optimized loop from 179.art.

  Show description for cce240d
  lattner

  committed
  cce240d

  Copy full SHA for cce240d

• add a trivial instcombine missed in Dhrystone

  Show description for c1853e4
  lattner

  committed
  c1853e4

  Copy full SHA for c1853e4

Commits on Jan 2, 2011

• update a bunch of entries.

  Show description for 527b47d
  lattner

  committed
  527b47d

  Copy full SHA for 527b47d

• Allow loop-idiom to run on multiple BB loops, but still only scan the loop

  Show description for cf078f2
  lattner

  committed
  cf078f2

  Copy full SHA for cf078f2

Commits on Jan 1, 2011

• a missed __builtin_object_size case.

  Show description for 15df044
  lattner

  committed
  15df044

  Copy full SHA for 15df044

• various updates.

  Show description for 9c8fb9e
  lattner

  committed
  9c8fb9e

  Copy full SHA for 9c8fb9e

Pagination

Previous

Next