| ca50b874039e326858e38bc67457369532a9c4dd |
|
12-Jun-2017 |
Yuqian Li <liyuqian@google.com> |
Combine duplicate code in preparing clipRgn for fill_path
Bug: skia:
Change-Id: I52ed98491457aa426e2bb74a29131f4a20330017
Reviewed-on: https://skia-review.googlesource.com/19341
Commit-Queue: Yuqian Li <liyuqian@google.com>
Reviewed-by: Cary Clark <caryclark@google.com>
/external/skia/src/core/SkScan_AAAPath.cpp
|
| 1c8408847cc626a34fc885b1c6313533d030e192 |
|
26-May-2017 |
Yuqian Li <liyuqian@google.com> |
Explicitly check edge count >= 2 in non-debug build
We also preserve the check in debug mode that the
count should never be 1 when the path is convex.
Bug: skia:6684
Change-Id: I4d4c9ad9f9d704e94bbe51f10a96f8b3066afaa1
Reviewed-on: https://skia-review.googlesource.com/17983
Commit-Queue: Yuqian Li <liyuqian@google.com>
Reviewed-by: Mike Reed <reed@google.com>
/external/skia/src/core/SkScan_AAAPath.cpp
|
| d3b65972aad96453ff4510caa3e25a2b847c6d1e |
|
22-Mar-2017 |
Brian Salomon <bsalomon@google.com> |
Mark overridden destructors with 'override' and remove 'virtual'
This silences a new warning in clang 5.0
Change-Id: Ieb5b75a6ffed60107c3fd16075d2ecfd515b55e8
Reviewed-on: https://skia-review.googlesource.com/10006
Commit-Queue: Brian Salomon <bsalomon@google.com>
Reviewed-by: Mike Klein <mtklein@chromium.org>
/external/skia/src/core/SkScan_AAAPath.cpp
|
| a33b43d796f03e2f4b5abd8060272f4f775d7390 |
|
17-Mar-2017 |
Yuqian Li <liyuqian@google.com> |
Combine some duplicate code in AA using template
BUG=skia:
Change-Id: Id048425e8199167feba0268f89f526bbf5b7242b
Reviewed-on: https://skia-review.googlesource.com/9832
Reviewed-by: Cary Clark <caryclark@google.com>
Commit-Queue: Yuqian Li <liyuqian@google.com>
/external/skia/src/core/SkScan_AAAPath.cpp
|
| 8de05ffcb26c757a6169e217ef46105286811dd9 |
|
06-Mar-2017 |
Yuqian Li <liyuqian@google.com> |
Remove SK_SUPPORT_LEGACY_AAA flag as chromium now turns it off
BUG=skia:
Change-Id: I3b385b8aeab08f402f4471637bf641cd456c8c56
Reviewed-on: https://skia-review.googlesource.com/9327
Reviewed-by: Cary Clark <caryclark@google.com>
Commit-Queue: Yuqian Li <liyuqian@google.com>
/external/skia/src/core/SkScan_AAAPath.cpp
|
| 6987b00fae08ef0731042bb341b4e002a2da72c2 |
|
19-Jan-2017 |
Yuqian Li <liyuqian@google.com> |
Fix performance issue and compute intersections more often.
(This CL also turned on Analytic AA for concave paths by removing SK_SUPPORT_LEGACY_AAA flag.)
Performance:
The SK_ALWAYS_INLINE was restored because it could bring 30%-50% speedup
in certain convex cases (e.g., fill_big_triangle). We also have to
reduce the number of branchings in the concave code path to enable such
speedup. (Although the speedup is for convex cases. The assembly code is
so strange...)
Intersection:
Previously, the criterion is too loose and that caused some bad pixels
(mostly unnoticeable by human eyes without magnifying). For example,
pixel (198, 222) of
https://gold.skia.org/detail?test=parsedpaths&digest=979e81de6f7b3f9e7e8dc810e31cad8d
BUG=skia:
Change-Id: I5e8191865c3df625f895cd4588c67c283fcbeaec
Reviewed-on: https://skia-review.googlesource.com/7318
Reviewed-by: Cary Clark <caryclark@google.com>
Reviewed-by: Mike Reed <reed@google.com>
Commit-Queue: Yuqian Li <liyuqian@google.com>
/external/skia/src/core/SkScan_AAAPath.cpp
|
| 98cf99b3ae4cc956158d1fd843fc8a84059ae132 |
|
17-Jan-2017 |
Yuqian Li <liyuqian@google.com> |
Turn on concave analytic AA
This will allow Skia trybots to exercise analytic AA. But there's still
a guard flag in Chromium that prevents layout tests failure.
Additionally, we
1. fixed nagative shift problem
2. relax the ASSERT when slope is too large: If slope is large, the accuracy of the slope is limited due
to conversion to SkFDot6 and division. Hence we have to relax the constraint.
3. handle the special case where dx != 0 while slope == 0 because of very large dy and limited precision.
BUG=skia:
Change-Id: Ice70164f3f396f0db3896bedc7b96fbd613078dc
Reviewed-on: https://skia-review.googlesource.com/7120
Reviewed-by: Mike Reed <reed@google.com>
Commit-Queue: Yuqian Li <liyuqian@google.com>
/external/skia/src/core/SkScan_AAAPath.cpp
|
| 550148ba572acc75e73d3c7d600ea7fa43db55f2 |
|
13-Jan-2017 |
Yuqian Li <liyuqian@google.com> |
Reland https://skia-review.googlesource.com/c/6091/
The only difference is that we now also put the guard flag
SK_SUPPORT_LEGACY_AAA in SkUserConfig.h. Previously, SkAnalyticEdge.cpp doesn't
get that flag from SkScan.h and that caused many problems.
BUG=skia:
TBR=reed@google.com,caryclark@google.com
Change-Id: I134bb76cebd6fffa712f438076668765321bba3b
Reviewed-on: https://skia-review.googlesource.com/6992
Reviewed-by: Yuqian Li <liyuqian@google.com>
Commit-Queue: Yuqian Li <liyuqian@google.com>
/external/skia/src/core/SkScan_AAAPath.cpp
|
| aeef5610d7a3fe4a53fc27b20486da79bf7a949a |
|
13-Jan-2017 |
Yuqian Li <liyuqian@google.com> |
Revert "Reland https://skia-review.googlesource.com/c/6091/"
This reverts commit b46fff60bc82fe6f0c64b2241d854a121f7cb5f9.
Reason for revert: possible chromium cc unit tests failure
Change-Id: Ie174c55e4d0fc3ae45854b5897ba26b7ad5a9c13
Reviewed-on: https://skia-review.googlesource.com/6981
Commit-Queue: Yuqian Li <liyuqian@google.com>
Reviewed-by: Yuqian Li <liyuqian@google.com>
/external/skia/src/core/SkScan_AAAPath.cpp
|
| b46fff60bc82fe6f0c64b2241d854a121f7cb5f9 |
|
12-Jan-2017 |
Yuqian Li <liyuqian@google.com> |
Reland https://skia-review.googlesource.com/c/6091/
The only difference is that we now put the guard flag SK_SUPPORT_LEGACY_AAA in
SkUserConfig.h instead of SkScan.h. Previously, SkAnalyticEdge.cpp doesn't get
that flag from SkScan.h and that caused many problems.
BUG=skia:
TBR=reed@google.com,caryclark@google.com
Change-Id: I7b89d3cb64ad71715101d2a5e8e77be3a8a6fa16
Reviewed-on: https://skia-review.googlesource.com/6972
Reviewed-by: Yuqian Li <liyuqian@google.com>
Commit-Queue: Yuqian Li <liyuqian@google.com>
/external/skia/src/core/SkScan_AAAPath.cpp
|
| 90ee03b0d1778f37448dfd82d1f5cfbbec9f798f |
|
12-Jan-2017 |
Yuqian Li <liyuqian@google.com> |
Revert "Implement Analytic AA for General Paths (with Guard against Chrome)"
This reverts commit 89a0e72287e991cfa2f860f92fad545ca59defe1.
Reason for revert: <INSERT REASONING HERE>
Original change's description:
> Implement Analytic AA for General Paths (with Guard against Chrome)
>
> I've set up a SK_SUPPORT_LEGACY_AAA flag to guard against Chromium layout tests. I also set that flag in this CL so theoretically this CL won't trigger any GM changes. I'll use this to verify my guard, and remove that flag and actually enables concave AAA in a future CL.
>
> When enabled, for most simple concave paths (e.g., rectangle stroke, rrect stroke, sawtooth, stars...), the Analytic AA achieves 1.3x-2x speedup, and they look much prettier. And they probably are the majority in our use cases by number. But they probably are not the majority by time cost; a single complicated path may cost 10x-100x more time to render than a rectangle stroke... For those complicated paths, we fall back to supersampling by default as we're likely to be 1.1-1.2x slower and the quality improvement is not visually significant. However, one can use gSkForceAnalyticAA to disable that fallback.
>
> BUG=skia:
>
> Change-Id: If9549a3acc4a187cfaf7eb51890c148da3083d31
> Reviewed-on: https://skia-review.googlesource.com/6091
> Reviewed-by: Mike Reed <reed@google.com>
> Commit-Queue: Yuqian Li <liyuqian@google.com>
>
TBR=caryclark@google.com,liyuqian@google.com,reed@google.com
BUG=skia:
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
Change-Id: I13c05aaa1bcb14956bd0fe01bb404e41be75af22
Reviewed-on: https://skia-review.googlesource.com/6961
Commit-Queue: Yuqian Li <liyuqian@google.com>
Reviewed-by: Yuqian Li <liyuqian@google.com>
/external/skia/src/core/SkScan_AAAPath.cpp
|
| 89a0e72287e991cfa2f860f92fad545ca59defe1 |
|
11-Jan-2017 |
Yuqian Li <liyuqian@google.com> |
Implement Analytic AA for General Paths (with Guard against Chrome)
I've set up a SK_SUPPORT_LEGACY_AAA flag to guard against Chromium layout tests. I also set that flag in this CL so theoretically this CL won't trigger any GM changes. I'll use this to verify my guard, and remove that flag and actually enables concave AAA in a future CL.
When enabled, for most simple concave paths (e.g., rectangle stroke, rrect stroke, sawtooth, stars...), the Analytic AA achieves 1.3x-2x speedup, and they look much prettier. And they probably are the majority in our use cases by number. But they probably are not the majority by time cost; a single complicated path may cost 10x-100x more time to render than a rectangle stroke... For those complicated paths, we fall back to supersampling by default as we're likely to be 1.1-1.2x slower and the quality improvement is not visually significant. However, one can use gSkForceAnalyticAA to disable that fallback.
BUG=skia:
Change-Id: If9549a3acc4a187cfaf7eb51890c148da3083d31
Reviewed-on: https://skia-review.googlesource.com/6091
Reviewed-by: Mike Reed <reed@google.com>
Commit-Queue: Yuqian Li <liyuqian@google.com>
/external/skia/src/core/SkScan_AAAPath.cpp
|
| 95e3c058ef633782f7549e9e1c2727d60dbc8ee5 |
|
11-Jan-2017 |
Hal Canary <halcanary@google.com> |
SkTypes.h : move SkAutoMalloc into SkAutoMalloc.h
* SkAutoFree moved to SkTemplates.h (now implmented with unique_ptr).
* SkAutoMalloc and SkAutoSMalloc moved to SkAutoMalloc.h
* "SkAutoFree X(sk_malloc_throw(N));" --> "SkAutoMalloc X(N);"
Revert "Revert 'SkTypes.h : move SkAutoMalloc into SkAutoMalloc.h'"
This reverts commit c456b73fef9589bbdc5eb83eaa83e53c357bb3da.
Change-Id: Ie2c1a17c20134b8ceab85a68b3ae3e61c24fbaab
Reviewed-on: https://skia-review.googlesource.com/6886
Reviewed-by: Hal Canary <halcanary@google.com>
Commit-Queue: Hal Canary <halcanary@google.com>
/external/skia/src/core/SkScan_AAAPath.cpp
|
| c456b73fef9589bbdc5eb83eaa83e53c357bb3da |
|
11-Jan-2017 |
Kevin Lubick <kjlubick@google.com> |
Revert "SkTypes.h : move SkAutoMalloc into SkAutoMalloc.h"
This reverts commit a5494f117086d712855e4b6289c58c92d1549bcf.
Reason for revert: Broke Google3
Original change's description:
> SkTypes.h : move SkAutoMalloc into SkAutoMalloc.h
>
> * SkAutoFree moved to SkTemplates.h (now implmented with unique_ptr).
>
> * SkAutoMalloc and SkAutoSMalloc moved to SkAutoMalloc.h
>
> * "SkAutoFree X(sk_malloc_throw(N));" --> "SkAutoMalloc X(N);"
>
> Change-Id: Idacd86ca09e22bf092422228599ae0d9bedded88
> Reviewed-on: https://skia-review.googlesource.com/4543
> Reviewed-by: Ben Wagner <bungeman@google.com>
> Reviewed-by: Mike Reed <reed@google.com>
> Commit-Queue: Hal Canary <halcanary@google.com>
>
TBR=halcanary@google.com,bungeman@google.com,reed@google.com,reviews@skia.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
Change-Id: Ie8bd176121c3ee83c110d66c0d0ac65e09bfc9c5
Reviewed-on: https://skia-review.googlesource.com/6884
Commit-Queue: Kevin Lubick <kjlubick@google.com>
Reviewed-by: Kevin Lubick <kjlubick@google.com>
/external/skia/src/core/SkScan_AAAPath.cpp
|
| a5494f117086d712855e4b6289c58c92d1549bcf |
|
10-Jan-2017 |
Hal Canary <halcanary@google.com> |
SkTypes.h : move SkAutoMalloc into SkAutoMalloc.h
* SkAutoFree moved to SkTemplates.h (now implmented with unique_ptr).
* SkAutoMalloc and SkAutoSMalloc moved to SkAutoMalloc.h
* "SkAutoFree X(sk_malloc_throw(N));" --> "SkAutoMalloc X(N);"
Change-Id: Idacd86ca09e22bf092422228599ae0d9bedded88
Reviewed-on: https://skia-review.googlesource.com/4543
Reviewed-by: Ben Wagner <bungeman@google.com>
Reviewed-by: Mike Reed <reed@google.com>
Commit-Queue: Hal Canary <halcanary@google.com>
/external/skia/src/core/SkScan_AAAPath.cpp
|
| dba05384e1a21015b7df72d1006ee7054b1906f7 |
|
02-Dec-2016 |
Yuqian Li <liyuqian@google.com> |
Remove SK_ANALYTIC_AA_GUARD flag
BUG=skia:
Change-Id: I4e10ba7afc76ed41c6c41275f9b6a5dde3a2a4be
Reviewed-on: https://skia-review.googlesource.com/5502
Reviewed-by: Florin Malita <fmalita@chromium.org>
Commit-Queue: Yuqian Li <liyuqian@google.com>
/external/skia/src/core/SkScan_AAAPath.cpp
|
| db13a093bec6ac1c06bb7c2ec1bde99c42130280 |
|
29-Nov-2016 |
Yuqian Li <liyuqian@google.com> |
Use AdditiveBlitter for partial rows
Previously, we forgot to use AdditiveBlitter in two places where partial
rows are blitterred. That causes SkAAClip to complain as in skia:6003.
BUG=skia:6003
Change-Id: I4f4a896072448bdb3f287a2eb61cb64b1256ea78
Reviewed-on: https://skia-review.googlesource.com/5273
Reviewed-by: Cary Clark <caryclark@google.com>
Commit-Queue: Yuqian Li <liyuqian@google.com>
/external/skia/src/core/SkScan_AAAPath.cpp
|
| 4eb35bb7d72d0e87f1546e822909929e1eef12bb |
|
17-Nov-2016 |
Yuqian Li <liyuqian@google.com> |
Use SkFixedMul instead of SkFixedMul_lowprec
It seems that SkFixedMul_lowprec doesn't have too much performance gain (maybe
1%). But its precision loss is siginificant. No serious issues have been found
in convex cases, but things get much trickier for concave shapes. I'm removing
it now to improve the stability and reliability of our algorithm, which may
potentially benefit our Chrome landing. (Even if we do not remove
SkFixedMul_lowprec now, we eventually will remove it when we land concave AAA
code.)
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=4926
Change-Id: Iae5739a3780bb77ce6237888eee51a502fea5cf2
Reviewed-on: https://skia-review.googlesource.com/4926
Reviewed-by: Cary Clark <caryclark@google.com>
Commit-Queue: Yuqian Li <liyuqian@google.com>
/external/skia/src/core/SkScan_AAAPath.cpp
|
| d930752957aaa26077870ebc7fb1c05307f4c27e |
|
16-Nov-2016 |
Yuqian Li <liyuqian@google.com> |
Turn off all SkDebugfs in SkScan_AAAPath.cpp
TBR=mtklein@google.com
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=4912
Change-Id: Ia69f2932936e34bc487c9c468bba176de0b2b0df
Reviewed-on: https://skia-review.googlesource.com/4912
Reviewed-by: Yuqian Li <liyuqian@google.com>
Reviewed-by: Mike Klein <mtklein@chromium.org>
Commit-Queue: Mike Klein <mtklein@chromium.org>
/external/skia/src/core/SkScan_AAAPath.cpp
|
| 8de17f75982efe643c0f34afa685b99c05a5f9d7 |
|
16-Nov-2016 |
Yuqian Li <liyuqian@google.com> |
No SkDebugf for approximate intersection
TBR=mtklein@google.com
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=4911
Change-Id: I23bc298186f93c1a97a15d00b34f65cd72f964e7
Reviewed-on: https://skia-review.googlesource.com/4911
Reviewed-by: Yuqian Li <liyuqian@google.com>
Reviewed-by: Mike Klein <mtklein@chromium.org>
Commit-Queue: Yuqian Li <liyuqian@google.com>
Commit-Queue: Mike Klein <mtklein@chromium.org>
/external/skia/src/core/SkScan_AAAPath.cpp
|
| 20079a94e82fd7345dfb119a8777e5ba482a041c |
|
16-Nov-2016 |
Yuqian Li <liyuqian@google.com> |
Fix mask overflow caused by edge drift
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=4797
Change-Id: Ica1568b67c1e1ce4aae2bdaba2c5b1f2155d1382
Reviewed-on: https://skia-review.googlesource.com/4797
Reviewed-by: Cary Clark <caryclark@google.com>
/external/skia/src/core/SkScan_AAAPath.cpp
|
| 721625b25e3e99d23b7765c75ba0b2ae5a351f7e |
|
16-Nov-2016 |
Yuqian Li <liyuqian@google.com> |
Drop forceRLE to simplify aaa_walk_convex_edges
The virtual flush function doesn't seem to affect the performance much.
Maybe there's a 1% drop in performance in nanobench against
fill_big_triangle and fill_big_circle, but that's too small a change for
nanobench to reliabily diffrentiate.
The smooth jump (ignore fractional y if edges don't change their
directions significantly) no longer needs to be guarded against SkAAClip
because our recent CL (https://skia-review.googlesource.com/c/4636/)
make left/rightBound much tighter.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=4899
Change-Id: If323013b810cc1ff5f6dbb868a8981354ee6f9b5
Reviewed-on: https://skia-review.googlesource.com/4899
Reviewed-by: Mike Reed <reed@google.com>
Reviewed-by: Stan Iliev <stani@google.com>
Commit-Queue: Yuqian Li <liyuqian@google.com>
Commit-Queue: Stan Iliev <stani@google.com>
/external/skia/src/core/SkScan_AAAPath.cpp
|
| e4b8b5283f45cff6a490ee73677f75b641c22bd1 |
|
16-Nov-2016 |
Yuqian Li <liyuqian@google.com> |
Simplify the signature of sk/aaa_fill_path
Previously, the clipRect is either equal to nullptr or clipRgn's bound
(after necessary supersampling shift). Hence we drop one of them to make
the signature simpler.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=4874
Change-Id: Ied8d5313809d6cf90374365b01f2b8d52f2236e2
Reviewed-on: https://skia-review.googlesource.com/4874
Reviewed-by: Mike Reed <reed@google.com>
Commit-Queue: Yuqian Li <liyuqian@google.com>
/external/skia/src/core/SkScan_AAAPath.cpp
|
| 06196411cb3e06bd041b7ec12e69d0a7a790d5b7 |
|
14-Nov-2016 |
Yuqian Li <liyuqian@google.com> |
Catch width overflow in ASAN using int64
This is related with our previous CL:
https://skia-review.googlesource.com/c/4628/
That previous CL fix won't pass the ASAN test (../../../include/core/SkRect.h:72:39: runtime error: signed integer overflow: 1 - -2147483648 cannot be represented in type 'int') so we use int64 in this CL.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=4785
Change-Id: I39f7329c427d49f4be07588f37120722ac9892c2
Reviewed-on: https://skia-review.googlesource.com/4785
Reviewed-by: Mike Reed <reed@google.com>
/external/skia/src/core/SkScan_AAAPath.cpp
|
| bfebe22ed54d1e3a00888292f10ed8b9714135d3 |
|
14-Nov-2016 |
liyuqian <liyuqian@google.com> |
Set analytic AA as default.
We'll use https://codereview.chromium.org/2471123002/ to keep Chrome from using
analytic AA until we fix all the Chrome tests.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2471133002
Committed: https://skia.googlesource.com/skia/+/dca4f6530013cf43a2557ccb07f5cb4fd916b8e8
Review-Url: https://codereview.chromium.org/2471133002
/external/skia/src/core/SkScan_AAAPath.cpp
|
| 06dd5c77a6ca6c484c683f1006a7bb6742f8917e |
|
11-Nov-2016 |
liyuqian <liyuqian@google.com> |
Revert of Set analytic AA as default. (patchset #8 id:140001 of https://codereview.chromium.org/2471133002/ )
Reason for revert:
SkASSERT triggered in some perf tests. Roll back.
Original issue's description:
> Set analytic AA as default.
>
> We'll use https://codereview.chromium.org/2471123002/ to keep Chrome from using
> analytic AA until we fix all the Chrome tests.
>
> BUG=skia:
> GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2471133002
>
> Committed: https://skia.googlesource.com/skia/+/dca4f6530013cf43a2557ccb07f5cb4fd916b8e8
TBR=fmalita@chromium.org,msarett@google.com,mtklein@google.com,reed@google.com,mtklein@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:
Review-Url: https://codereview.chromium.org/2499723002
/external/skia/src/core/SkScan_AAAPath.cpp
|
| dca4f6530013cf43a2557ccb07f5cb4fd916b8e8 |
|
11-Nov-2016 |
liyuqian <liyuqian@google.com> |
Set analytic AA as default.
We'll use https://codereview.chromium.org/2471123002/ to keep Chrome from using
analytic AA until we fix all the Chrome tests.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2471133002
Review-Url: https://codereview.chromium.org/2471133002
/external/skia/src/core/SkScan_AAAPath.cpp
|
| 041da389dd923689041ba5e45cda34901ba4fe8a |
|
11-Nov-2016 |
liyuqian <liyuqian@google.com> |
Tighten SkAAClip Bounds by Path Bounds
This bug chromium:662780 exists after our original fix (https://codereview.chromium.org/2477393002/) because this path (added in unit test) is calling blitAntiRect rather than blitAntiH when the path is drifted across the boundary. (The quadratic edge drifts across the boundary after an update and sets a dX=0 line segment which triggers blitAntiRect.)
Note that I didn't assert for the dLeft = dRite = 0 case because the left/right there won't drift after the SkTMin/SkTMax in line 964/966.
Theoretically we can revert the relaxation in https://codereview.chromium.org/2477393002/ (that's only a relaxation for analytic AA, not supersampled AA). However, consider that the initial landing of analytic AA is so painful, I decide to revert that relaxation only after our successful landing...
BUG=chromium:662780, chromium:662862
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2482193004
Review-Url: https://codereview.chromium.org/2482193004
/external/skia/src/core/SkScan_AAAPath.cpp
|
| c4f66af20e3a58a09554eb3db2bf45c6291fe3a8 |
|
11-Nov-2016 |
Yuqian Li <liyuqian@google.com> |
Catch width overflow
BUG=chromium:662730
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=4628
Change-Id: Iaf3a30d39fda3166a6f8fc62a30580629418dc88
Reviewed-on: https://skia-review.googlesource.com/4628
Reviewed-by: Cary Clark <caryclark@google.com>
Commit-Queue: Yuqian Li <liyuqian@google.com>
/external/skia/src/core/SkScan_AAAPath.cpp
|
| c78eff97549e8e346394d3e228395ceb8a467b35 |
|
09-Nov-2016 |
liyuqian <liyuqian@google.com> |
Do not call blitV with 0 height
This is causing SkAAClipBlitter to crash.
BUG=chromium:662952
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2485243002
Review-Url: https://codereview.chromium.org/2485243002
/external/skia/src/core/SkScan_AAAPath.cpp
|
| 3ce89dad5bc324dad9c4b77393e16f3bcb7396a7 |
|
09-Nov-2016 |
liyuqian <liyuqian@google.com> |
Fix the meaning of stop_y
stop_y means that we should stop exactly at stop_y, so the last row should be
[stop_y - 1, stop_y], not [stop_y, stop_y + 1].
Somehow this misunderstanding didn't trigger any issue until Chrome exercises
SkAAClip with some websites (e.g.,
http://www.lemonde.fr/elections-americaines/article/2016/11/07/deux-programmes-deux-visions-de-l-amerique_5026444_829254.html).
When we blitter the extra row [stop_y, stop_y + 1], the SkAAClip will return
nullptr by findRow. Later when that nullptr row is used to findX, the crash
happened.
BUG=chromium:662925, chromium:662776
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2481703004
Review-Url: https://codereview.chromium.org/2481703004
/external/skia/src/core/SkScan_AAAPath.cpp
|
| f6bddfd33daeade5418a2ccb36b7945fac703f40 |
|
08-Nov-2016 |
liyuqian <liyuqian@google.com> |
Do not call blitAntiRect for empty rect
BUG=chromium:662800
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2490673002
Review-Url: https://codereview.chromium.org/2490673002
/external/skia/src/core/SkScan_AAAPath.cpp
|
| 525e13c1d318b067e679d7ca540fc1436c38385b |
|
05-Nov-2016 |
liyuqian <liyuqian@google.com> |
Fix the sum of alpha = 254 instead of 255 by rounding
This patch is a subpatch from https://codereview.chromium.org/2471133002/
I created this CL mainly for Florin to land before he rebaseline the Chrome
layout tests. (Landing the whole patch of 2471133002 would be nice but most
people already left for the weekend so I'd rather not take the risk to break
the Skia during the weekend.)
BUG=skia:
TBR=reed@google.com, mtklein@google.com, fmalita@chromium.org
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2483523002
Review-Url: https://codereview.chromium.org/2483523002
/external/skia/src/core/SkScan_AAAPath.cpp
|
| 5de9afc8526fe266cee4f32e1cdf652622fa1e3d |
|
31-Oct-2016 |
liyuqian <liyuqian@google.com> |
Disable some SkDebugfs for dbg bots debugging
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2464893002
Review-Url: https://codereview.chromium.org/2464893002
/external/skia/src/core/SkScan_AAAPath.cpp
|
| a3316adf63ca168c9125df965cfa85bdd7b50a1c |
|
29-Oct-2016 |
liyuqian <liyuqian@google.com> |
Prevent overflow by falling back to non-AA
This piece of code is directly copied from our old supersampling AA
code. I didn not preserve this code because it's not triggered by any Skia
tests. However, this will be critical to Chromium's tests as some
websites will generate huge paths!
I'm not so sure whether the long_running_idle_gmail_background_tbmv2 test
failed because of this. But I'm sure that www.nationalgeographic.com from
page_cycler_v2.typical_25 failed because of this.
BUG=chromium:660394
TBR=reed@google.com,caryclark@google.com
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2461133002
Review-Url: https://codereview.chromium.org/2461133002
/external/skia/src/core/SkScan_AAAPath.cpp
|
| 552889ea44baaa50368fc4aec5c304c83698040f |
|
27-Oct-2016 |
liyuqian <liyuqian@google.com> |
Do not skip fractional y for SkAAClip
This should fix the following bugs:
https://bugs.chromium.org/p/chromium/issues/detail?id=659883
https://bugs.chromium.org/p/chromium/issues/detail?id=659972
BUG=chromium:659883,chromium:659972
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2452393002
Review-Url: https://codereview.chromium.org/2452393002
/external/skia/src/core/SkScan_AAAPath.cpp
|
| 7a86423c42433ab8a7df2b06bd954b2427217bb9 |
|
21-Oct-2016 |
liyuqian <liyuqian@google.com> |
Fall back from analytic to sampling earlier
This magically fixes some strange js unit tests in Chrome. Those tests
only fail in virtual/gpu which somehow interacts with my cpu rasterizer?
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2439973004
Review-Url: https://chromiumcodereview.appspot.com/2439973004
/external/skia/src/core/SkScan_AAAPath.cpp
|
| 6a7287c14ba39784d66bb299a1340c0d7ca7b683 |
|
21-Oct-2016 |
liyuqian <liyuqian@google.com> |
Another forgotten forceRLE
BUG=skia:
TBR=reed@google.com,caryclark@google.com
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2438163003
Review-Url: https://chromiumcodereview.appspot.com/2438163003
/external/skia/src/core/SkScan_AAAPath.cpp
|
| c83ada98e2e60be75ae03e228c4d35eb45357add |
|
20-Oct-2016 |
liyuqian <liyuqian@google.com> |
Quickly fix the forgotten forceRLE
BUG=skia:
TBR=reed@google.com,caryclark@google.com
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2436593004
Review-Url: https://chromiumcodereview.appspot.com/2436593004
/external/skia/src/core/SkScan_AAAPath.cpp
|
| 2add0ffdaab4c33ca6702c81533c88c7b5ca7294 |
|
20-Oct-2016 |
liyuqian <liyuqian@google.com> |
Use Analytic AA in SkAAClip.
This will fix some Chrome ref tests.
The main issue here is that the SkAAClip requires: (1) we blit in order, and (2) we must forceRLE.
Note that this still depends on the global gSkUseAnalyticAA to turn on the Analytic AA.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2430343003
Review-Url: https://chromiumcodereview.appspot.com/2430343003
/external/skia/src/core/SkScan_AAAPath.cpp
|
| ebcb8aadae70b25d4ac023b4f987a469b6d91f41 |
|
11-Oct-2016 |
liyuqian <liyuqian@google.com> |
Add AA Analytic Compile Flag (and small fixes)
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2410303002
Review-Url: https://codereview.chromium.org/2410303002
/external/skia/src/core/SkScan_AAAPath.cpp
|
| 511f2d79bf6ea0ee858a3468d3aa8dcbfca58648 |
|
04-Oct-2016 |
Mike Klein <mtklein@chromium.org> |
Make global use-analytic-AA bit threadsafe.
I also had to cut it down to just a global atomic bool... as a field in a global singleton accessed through instance(), it's very hard to make threadsafe.
CQ_INCLUDE_TRYBOTS=master.client.skia:Test-Ubuntu-Clang-GCE-CPU-AVX2-x86_64-Release-TSAN-Trybot
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2937
Change-Id: If80be987906dd521fbe644d1d0d577009f06d0e3
Reviewed-on: https://skia-review.googlesource.com/2937
Reviewed-by: Yuqian Li <liyuqian@google.com>
Commit-Queue: Yuqian Li <liyuqian@google.com>
Commit-Queue: Mike Klein <mtklein@chromium.org>
/external/skia/src/core/SkScan_AAAPath.cpp
|
| 38911a7cb53474575e1cd1cb545902b50ee00889 |
|
04-Oct-2016 |
liyuqian <liyuqian@google.com> |
Resubmit issue 2221103002 to fix the iOS build by declaring the flag in
SkCommonFlags.h
TBR=reed@google.com,caryclark@google.com
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2393643002
Review-Url: https://codereview.chromium.org/2393643002
/external/skia/src/core/SkScan_AAAPath.cpp
|
| d36baa7a4a5ae3cc94cc4a45379f55658f80c0a6 |
|
04-Oct-2016 |
stephana <stephana@google.com> |
Revert of Analytic AntiAlias for Convex Shapes (patchset #14 id:260001 of https://codereview.chromium.org/2221103002/ )
Reason for revert:
Breaks iOS build.
Original issue's description:
> Implement AnalyticAA for convex shapes.
>
> Design doc: go/analyticAA
>
> A performance test can be found here: https://docs.google.com/a/google.com/spreadsheets/d/1n9LSjFzrQzx0hovFddWey0GSMXNRjl1oFuSypMlHWZk/edit?usp=sharing
>
> Our best case is filling big triangles, which according to our experiment has ~2.9x speedup. Our worst case is filling small ovals/circles, which has a ~1.06x slowdown.
>
> To see how our new algorithm changes the DM images, see: https://x20web.corp.google.com/~liyuqian/dmdiff/index.html
> The most significant changes are in convexpaths and analytic_antialias_convex
>
> BUG=skia:
> GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2221103002
>
> Committed: https://skia.googlesource.com/skia/+/7795822807478143120c33228b68d2ab3918af2c
TBR=reed@google.com,caryclark@google.com,liyuqian@google.com
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:
Review-Url: https://codereview.chromium.org/2388213003
/external/skia/src/core/SkScan_AAAPath.cpp
|
| 7795822807478143120c33228b68d2ab3918af2c |
|
04-Oct-2016 |
liyuqian <liyuqian@google.com> |
Implement AnalyticAA for convex shapes.
Design doc: go/analyticAA
A performance test can be found here: https://docs.google.com/a/google.com/spreadsheets/d/1n9LSjFzrQzx0hovFddWey0GSMXNRjl1oFuSypMlHWZk/edit?usp=sharing
Our best case is filling big triangles, which according to our experiment has ~2.9x speedup. Our worst case is filling small ovals/circles, which has a ~1.06x slowdown.
To see how our new algorithm changes the DM images, see: https://x20web.corp.google.com/~liyuqian/dmdiff/index.html
The most significant changes are in convexpaths and analytic_antialias_convex
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2221103002
Review-Url: https://codereview.chromium.org/2221103002
/external/skia/src/core/SkScan_AAAPath.cpp
|