37924 2010-04-21 07:43:55 -0700 Web Inspector: slow scroll speed of Timeline in case of many timeline marks 2010-04-22 04:22:16 -0700 1 1 1 Unclassified WebKit Web Inspector (Deprecated) 528+ (Nightly build) All All RESOLVED FIXED P2 Normal --- 0 loislo loislo commit-queue eric pfeldman timothy yurys oldest_to_newest 215315 0 loislo 2010-04-21 07:43:55 -0700 if we have significant amount of timeline marks then we have significant scrolling speed degradation. (2k marks) 215328 1 53959 loislo 2010-04-21 08:12:56 -0700 Created attachment 53959 [patch] initial version. 215532 2 53959 pfeldman 2010-04-21 15:14:26 -0700 Comment on attachment 53959 [patch] initial version. Note that this does not address gmail's slowness. There are 1800 marks generated upon initial gmail load, i.e. we create 1800 elements. This obviously will not scale no matter what we do to scrolling. I was thinking of fixing it by introducing a canvas element instead of separate divs per record. Feel free to land your patch, but I don't think we should close the bug yet. 215605 3 eric 2010-04-21 18:14:28 -0700 Attachment 53959 was posted by a committer and has review+, assigning to Ilya Tikhonovsky for commit. 215692 4 loislo 2010-04-21 22:14:06 -0700 (In reply to comment #2) > (From update of attachment 53959 [details]) > Note that this does not address gmail's slowness. There are 1800 marks > generated upon initial gmail load, i.e. we create 1800 elements. This obviously > will not scale no matter what we do to scrolling. I was thinking of fixing it > by introducing a canvas element instead of separate divs per record. Feel free > to land your patch, but I don't think we should close the bug yet. Everything was ok but when I created 2k marks by markTimeline I got scrolling speed slowed down as you described. The reason of that was the fact that each of these marks had additional DOM elements in timeline pane grid and timeline overview grid. As result for that use case we have not more then 100 visible rows in timeline panel and 2k x 2 additional DOM elements recreated and reattached for the each scroll event. I've checked this use case by 10k events and now the scrolling speed is fine. 215822 5 53959 commit-queue 2010-04-22 04:22:11 -0700 Comment on attachment 53959 [patch] initial version. Clearing flags on attachment: 53959 Committed r58087: <http://trac.webkit.org/changeset/58087> 215823 6 commit-queue 2010-04-22 04:22:16 -0700 All reviewed patches have been landed. Closing bug. 53959 2010-04-21 08:12:56 -0700 2010-04-22 04:22:10 -0700 [patch] initial version. patch text/plain 5174 loislo ZGlmZiAtLWdpdCBhL1dlYkNvcmUvQ2hhbmdlTG9nIGIvV2ViQ29yZS9DaGFuZ2VMb2cKaW5kZXgg NjQxNGI1Yy4uZDY4YmMyOSAxMDA2NDQKLS0tIGEvV2ViQ29yZS9DaGFuZ2VMb2cKKysrIGIvV2Vi Q29yZS9DaGFuZ2VMb2cKQEAgLTEsMyArMSwxOSBAQAorMjAxMC0wNC0yMSAgSWx5YSBUaWtob25v dnNreSAgPGxvaXNsb0BjaHJvbWl1bS5vcmc+CisKKyAgICAgICAgUmV2aWV3ZWQgYnkgTk9CT0RZ IChPT1BTISkuCisKKyAgICAgICAgV2ViIEluc3BlY3RvcjogVGltZWxpbmUgc2Nyb2xsaW5nIHNw ZWVkIGlzIHNsb3cgaWYgaXQgaGFzIG1vcmUgdGhhbiAxayB0aW1lbGluZSBtYXJrcy4KKyAgICAg ICAgaHR0cHM6Ly9idWdzLndlYmtpdC5vcmcvc2hvd19idWcuY2dpP2lkPTM3OTI0CisKKyAgICAg ICAgKiBpbnNwZWN0b3IvZnJvbnQtZW5kL1RpbWVsaW5lR3JpZC5qczoKKyAgICAgICAgKFdlYklu c3BlY3Rvci5UaW1lbGluZUdyaWQucHJvdG90eXBlLnVwZGF0ZURpdmlkZXJzKToKKyAgICAgICAg KFdlYkluc3BlY3Rvci5UaW1lbGluZUdyaWQucHJvdG90eXBlLmFkZEV2ZW50RGl2aWRlcnMpOgor ICAgICAgICAqIGluc3BlY3Rvci9mcm9udC1lbmQvVGltZWxpbmVPdmVydmlld1BhbmUuanM6Cisg ICAgICAgIChXZWJJbnNwZWN0b3IuVGltZWxpbmVPdmVydmlld1BhbmUucHJvdG90eXBlLnVwZGF0 ZUV2ZW50RGl2aWRlcnMpOgorICAgICAgICAqIGluc3BlY3Rvci9mcm9udC1lbmQvVGltZWxpbmVQ YW5lbC5qczoKKyAgICAgICAgKFdlYkluc3BlY3Rvci5UaW1lbGluZVBhbmVsLnByb3RvdHlwZS5f dXBkYXRlRXZlbnREaXZpZGVycyk6CisgICAgICAgIChXZWJJbnNwZWN0b3IuVGltZWxpbmVQYW5l bC5wcm90b3R5cGUuX3JlZnJlc2gpOgorCiAyMDEwLTA0LTIxICBCZW5qYW1pbiBQb3VsYWluICA8 aWtpcG91QGdtYWlsLmNvbT4KIAogICAgICAgICBSZXZpZXdlZCBieSBTaW1vbiBGcmFzZXIuCmRp ZmYgLS1naXQgYS9XZWJDb3JlL2luc3BlY3Rvci9mcm9udC1lbmQvVGltZWxpbmVHcmlkLmpzIGIv V2ViQ29yZS9pbnNwZWN0b3IvZnJvbnQtZW5kL1RpbWVsaW5lR3JpZC5qcwppbmRleCBmYjkzYjhm Li4zMjg2ZTY1IDEwMDY0NAotLS0gYS9XZWJDb3JlL2luc3BlY3Rvci9mcm9udC1lbmQvVGltZWxp bmVHcmlkLmpzCisrKyBiL1dlYkNvcmUvaW5zcGVjdG9yL2Zyb250LWVuZC9UaW1lbGluZUdyaWQu anMKQEAgLTY2LDcgKzY2LDYgQEAgV2ViSW5zcGVjdG9yLlRpbWVsaW5lR3JpZC5wcm90b3R5cGUg PSB7CiAgICAgICAgICAgICBwYWRkaW5nTGVmdCA9IDA7CiAgICAgICAgIHRoaXMuX2N1cnJlbnRE aXZpZGVyU2xpY2UgPSBzbGljZTsKIAotICAgICAgICB0aGlzLl9ldmVudERpdmlkZXJzRWxlbWVu dC5yZW1vdmVDaGlsZHJlbigpOwogICAgICAgICAvLyBSZXVzZSBkaXZpZGVyIGVsZW1lbnRzIGFu ZCBsYWJlbHMuCiAgICAgICAgIHZhciBkaXZpZGVyID0gdGhpcy5fZGl2aWRlcnNFbGVtZW50LmZp cnN0Q2hpbGQ7CiAgICAgICAgIHZhciBkaXZpZGVyTGFiZWxCYXIgPSB0aGlzLl9kaXZpZGVyc0xh YmVsQmFyRWxlbWVudC5maXJzdENoaWxkOwpAQCAtMTM1LDYgKzEzNCwxNSBAQCBXZWJJbnNwZWN0 b3IuVGltZWxpbmVHcmlkLnByb3RvdHlwZSA9IHsKICAgICAgICAgdGhpcy5fZXZlbnREaXZpZGVy c0VsZW1lbnQuYXBwZW5kQ2hpbGQoZGl2aWRlcik7CiAgICAgfSwKIAorICAgIGFkZEV2ZW50RGl2 aWRlcnM6IGZ1bmN0aW9uKGRpdmlkZXJzKQorICAgIHsKKyAgICAgICAgdGhpcy5lbGVtZW50LnJl bW92ZUNoaWxkKHRoaXMuX2V2ZW50RGl2aWRlcnNFbGVtZW50KTsKKyAgICAgICAgZm9yICh2YXIg aSA9IDA7IGkgPCBkaXZpZGVycy5sZW5ndGg7ICsraSkKKyAgICAgICAgICAgIGlmIChkaXZpZGVy c1tpXSkKKyAgICAgICAgICAgICAgICB0aGlzLl9ldmVudERpdmlkZXJzRWxlbWVudC5hcHBlbmRD aGlsZChkaXZpZGVyc1tpXSk7CisgICAgICAgIHRoaXMuZWxlbWVudC5hcHBlbmRDaGlsZCh0aGlz Ll9ldmVudERpdmlkZXJzRWxlbWVudCk7CisgICAgfSwKKwogICAgIHJlbW92ZUV2ZW50RGl2aWRl cnM6IGZ1bmN0aW9uKCkKICAgICB7CiAgICAgICAgIHRoaXMuX2V2ZW50RGl2aWRlcnNFbGVtZW50 LnJlbW92ZUNoaWxkcmVuKCk7CmRpZmYgLS1naXQgYS9XZWJDb3JlL2luc3BlY3Rvci9mcm9udC1l bmQvVGltZWxpbmVPdmVydmlld1BhbmUuanMgYi9XZWJDb3JlL2luc3BlY3Rvci9mcm9udC1lbmQv VGltZWxpbmVPdmVydmlld1BhbmUuanMKaW5kZXggNzE5MWVmOC4uYjE0MDQ0ZiAxMDA2NDQKLS0t IGEvV2ViQ29yZS9pbnNwZWN0b3IvZnJvbnQtZW5kL1RpbWVsaW5lT3ZlcnZpZXdQYW5lLmpzCisr KyBiL1dlYkNvcmUvaW5zcGVjdG9yL2Zyb250LWVuZC9UaW1lbGluZU92ZXJ2aWV3UGFuZS5qcwpA QCAtMTY2LDEzICsxNjYsMTggQEAgV2ViSW5zcGVjdG9yLlRpbWVsaW5lT3ZlcnZpZXdQYW5lLnBy b3RvdHlwZSA9IHsKICAgICB1cGRhdGVFdmVudERpdmlkZXJzOiBmdW5jdGlvbihyZWNvcmRzLCBk aXZpZGVyQ29uc3RydWN0b3IpCiAgICAgewogICAgICAgICB0aGlzLl9vdmVydmlld0dyaWQucmVt b3ZlRXZlbnREaXZpZGVycygpOworICAgICAgICB2YXIgZGl2aWRlcnMgPSBbXTsKICAgICAgICAg Zm9yICh2YXIgaSA9IDA7IGkgPCByZWNvcmRzLmxlbmd0aDsgKytpKSB7CiAgICAgICAgICAgICB2 YXIgcmVjb3JkID0gcmVjb3Jkc1tpXTsKICAgICAgICAgICAgIHZhciBwb3NpdGlvbnMgPSB0aGlz Ll9vdmVydmlld0NhbGN1bGF0b3IuY29tcHV0ZUJhckdyYXBoUGVyY2VudGFnZXMocmVjb3JkKTsK KyAgICAgICAgICAgIHZhciBkaXZpZGVyUG9zaXRpb24gPSBNYXRoLnJvdW5kKHBvc2l0aW9ucy5z dGFydCAqIDEwKTsKKyAgICAgICAgICAgIGlmIChkaXZpZGVyc1tkaXZpZGVyUG9zaXRpb25dKQor ICAgICAgICAgICAgICAgIGNvbnRpbnVlOwogICAgICAgICAgICAgdmFyIGRpdmlkZXIgPSBkaXZp ZGVyQ29uc3RydWN0b3IocmVjb3JkKTsKICAgICAgICAgICAgIGRpdmlkZXIuc3R5bGUubGVmdCA9 IHBvc2l0aW9ucy5zdGFydCArICIlIjsKLSAgICAgICAgICAgIHRoaXMuX292ZXJ2aWV3R3JpZC5h ZGRFdmVudERpdmlkZXIoZGl2aWRlcik7CisgICAgICAgICAgICBkaXZpZGVyc1tkaXZpZGVyUG9z aXRpb25dID0gZGl2aWRlcjsKICAgICAgICAgfQorICAgICAgICB0aGlzLl9vdmVydmlld0dyaWQu YWRkRXZlbnREaXZpZGVycyhkaXZpZGVycyk7CiAgICAgfSwKIAogICAgIHNldFNpZGViYXJXaWR0 aDogZnVuY3Rpb24od2lkdGgpCmRpZmYgLS1naXQgYS9XZWJDb3JlL2luc3BlY3Rvci9mcm9udC1l bmQvVGltZWxpbmVQYW5lbC5qcyBiL1dlYkNvcmUvaW5zcGVjdG9yL2Zyb250LWVuZC9UaW1lbGlu ZVBhbmVsLmpzCmluZGV4IDdiM2UzYjAuLjg0M2Q0MTAgMTAwNjQ0Ci0tLSBhL1dlYkNvcmUvaW5z cGVjdG9yL2Zyb250LWVuZC9UaW1lbGluZVBhbmVsLmpzCisrKyBiL1dlYkNvcmUvaW5zcGVjdG9y L2Zyb250LWVuZC9UaW1lbGluZVBhbmVsLmpzCkBAIC0xODUsMTcgKzE4NSwxOCBAQCBXZWJJbnNw ZWN0b3IuVGltZWxpbmVQYW5lbC5wcm90b3R5cGUgPSB7CiAgICAgewogICAgICAgICB0aGlzLl90 aW1lbGluZUdyaWQucmVtb3ZlRXZlbnREaXZpZGVycygpOwogICAgICAgICB2YXIgY2xpZW50V2lk dGggPSB0aGlzLl9ncmFwaFJvd3NFbGVtZW50Lm9mZnNldFdpZHRoIC0gdGhpcy5fZXhwYW5kT2Zm c2V0OworICAgICAgICB2YXIgZGl2aWRlcnMgPSBbXTsKICAgICAgICAgZm9yICh2YXIgaSA9IDA7 IGkgPCB0aGlzLl9tYXJrVGltZWxpbmVSZWNvcmRzLmxlbmd0aDsgKytpKSB7CiAgICAgICAgICAg ICB2YXIgcmVjb3JkID0gdGhpcy5fbWFya1RpbWVsaW5lUmVjb3Jkc1tpXTsKICAgICAgICAgICAg IHZhciBwb3NpdGlvbnMgPSB0aGlzLl9jYWxjdWxhdG9yLmNvbXB1dGVCYXJHcmFwaFdpbmRvd1Bv c2l0aW9uKHJlY29yZCwgY2xpZW50V2lkdGgpOwotICAgICAgICAgICAgaWYgKHBvc2l0aW9ucy5s ZWZ0IDwgMCB8fCBwb3NpdGlvbnMubGVmdCA+PSBjbGllbnRXaWR0aCkKKyAgICAgICAgICAgIHZh ciBkaXZpZGVyUG9zaXRpb24gPSBNYXRoLnJvdW5kKHBvc2l0aW9ucy5sZWZ0KTsKKyAgICAgICAg ICAgIGlmIChkaXZpZGVyUG9zaXRpb24gPCAwIHx8IGRpdmlkZXJQb3NpdGlvbiA+PSBjbGllbnRX aWR0aCB8fCBkaXZpZGVyc1tkaXZpZGVyUG9zaXRpb25dKQogICAgICAgICAgICAgICAgIGNvbnRp bnVlOwotCiAgICAgICAgICAgICB2YXIgZGl2aWRlciA9IHRoaXMuX2NyZWF0ZUV2ZW50RGl2aWRl cihyZWNvcmQpOwotICAgICAgICAgICAgZGl2aWRlci5zdHlsZS5sZWZ0ID0gKHBvc2l0aW9ucy5s ZWZ0ICsgdGhpcy5fZXhwYW5kT2Zmc2V0KSArICJweCI7Ci0KLSAgICAgICAgICAgIHRoaXMuX3Rp bWVsaW5lR3JpZC5hZGRFdmVudERpdmlkZXIoZGl2aWRlcik7CisgICAgICAgICAgICBkaXZpZGVy LnN0eWxlLmxlZnQgPSAoZGl2aWRlclBvc2l0aW9uICsgdGhpcy5fZXhwYW5kT2Zmc2V0KSArICJw eCI7CisgICAgICAgICAgICBkaXZpZGVyc1tkaXZpZGVyUG9zaXRpb25dID0gZGl2aWRlcjsKICAg ICAgICAgfQorICAgICAgICB0aGlzLl90aW1lbGluZUdyaWQuYWRkRXZlbnREaXZpZGVycyhkaXZp ZGVycyk7CiAgICAgICAgIHRoaXMuX292ZXJ2aWV3UGFuZS51cGRhdGVFdmVudERpdmlkZXJzKHRo aXMuX21hcmtUaW1lbGluZVJlY29yZHMsIHRoaXMuX2NyZWF0ZUV2ZW50RGl2aWRlci5iaW5kKHRo aXMpKTsKICAgICB9LAogCkBAIC00MTcsNyArNDE4LDggQEAgV2ViSW5zcGVjdG9yLlRpbWVsaW5l UGFuZWwucHJvdG90eXBlID0gewogICAgICAgICB0aGlzLl9vdmVydmlld1BhbmUudXBkYXRlKHRo aXMuX3Jvb3RSZWNvcmQuY2hpbGRyZW4sIHRoaXMuX2NhbGN1bGF0b3IuX3Nob3dTaG9ydEV2ZW50 cyk7CiAgICAgICAgIHRoaXMuX3JlZnJlc2hSZWNvcmRzKCF0aGlzLl9ib3VuZGFyaWVzQXJlVmFs aWQpOwogICAgICAgICB0aGlzLl91cGRhdGVSZWNvcmRzQ291bnRlcigpOwotICAgICAgICB0aGlz Ll91cGRhdGVFdmVudERpdmlkZXJzKCk7CisgICAgICAgIGlmKCF0aGlzLl9ib3VuZGFyaWVzQXJl VmFsaWQpCisgICAgICAgICAgICB0aGlzLl91cGRhdGVFdmVudERpdmlkZXJzKCk7CiAgICAgICAg IHRoaXMuX2JvdW5kYXJpZXNBcmVWYWxpZCA9IHRydWU7CiAgICAgfSwKIAo=