43e58e33c8
Bug 760331: Coalesce data for inline style across nodes. r=bz
...
This patch enables sharing of an nsAttrValue's MiscContainer between nodes for style rules. MiscContainers of type eCSSStyleRule are now refcounted (with some clever struct packing to ensure that the amount of memory allocated for MiscContainer remains unchanged on 32 and 64 bit). This infrastructure can be used to share most MiscContainer types in the future if we find advantages to sharing other types than just eCSSStyleRuley. A cache mapping strings to MiscContainers has been added to nsHTMLCSSStyleSheet. MiscContainers can be shared between nsAttrValues when one nsAttrValue is SetTo another nsAttrValue or when there is a cache hit in this cache. This patch also adds the ability to tell a style rule that it belongs to an nsHTMLCSSStyleSheet, with appropriate accessor functions to separate that from the existing case of belonging to an nsCSSStyleSheet.
The primary use case is to reduce memory use for pages that have lots of inline style attributes with the same value. This can happen easily with large pages that are automatically generated. An (admittedly pathological) testcase in Bug 686975 sees over 250 MB of memory savings with this change. Reusing the same MiscContainer for multiple nodes saves the overhead of maintaining separate copies of the string containing the serialized value of the style attribute and of creating separate style rules for each node. Eliminating duplicate style rules enables further savings in layout through style context sharing. The testcase sees the amount of memory used by style contexts go from over 250 MB to 10 KB.
Because the cache is based on the text value of the style attribute, it will not handle attributes that have different text values but are parsed into identical style rules. We also do not attempt to share MiscContainers when the node's base URI differs from the document URI. The effect of these limitations is expected to be low.
2012-09-30 09:40:24 -07:00
13c3b3bdbc
Add vendor prefix to :dir selector. Bug 562169, r=dbaron
2012-09-29 13:41:10 -07:00
e368dc9c85
Bug 579517 - Part 1: Automated conversion of NSPR numeric types to stdint types in Gecko; r=bsmedberg
...
This patch was generated by a script. Here's the source of the script for
future reference:
function convert() {
echo "Converting $1 to $2..."
find . ! -wholename "*nsprpub*" \
! -wholename "*security/nss*" \
! -wholename "*/.hg*" \
! -wholename "obj-ff-dbg*" \
! -name nsXPCOMCID.h \
! -name prtypes.h \
-type f \
\( -iname "*.cpp" \
-o -iname "*.h" \
-o -iname "*.c" \
-o -iname "*.cc" \
-o -iname "*.idl" \
-o -iname "*.ipdl" \
-o -iname "*.ipdlh" \
-o -iname "*.mm" \) | \
xargs -n 1 sed -i -e "s/\b$1\b/$2/g"
}
convert PRInt8 int8_t
convert PRUint8 uint8_t
convert PRInt16 int16_t
convert PRUint16 uint16_t
convert PRInt32 int32_t
convert PRUint32 uint32_t
convert PRInt64 int64_t
convert PRUint64 uint64_t
convert PRIntn int
convert PRUintn unsigned
convert PRSize size_t
convert PROffset32 int32_t
convert PROffset64 int64_t
convert PRPtrdiff ptrdiff_t
convert PRFloat64 double
2012-08-22 11:56:38 -04:00
4e199a1679
Bug 780618 - Move all error codes to nsError.h; r=ehsan
2012-07-27 17:03:27 +03:00
7b0bd7bf8b
Bug 562169. Add support for :dir(ltr/rtl) CSS selector, layout/style part, r=bz
2012-08-09 20:37:58 -07:00
3cfaa1db23
Back out bug 562169 because of Dromaeo DOM regressions
...
Backed out changeset 841cd6e1e585 (bug 562169)
Backed out changeset e226e413dd27 (bug 562169)
Backed out changeset 9c24b7287c5b (bug 562169)
2012-08-08 08:14:48 -07:00
aed3f621eb
Add support for :dir(ltr/rtl) CSS selector, layout/style part. Bug 562159, r=bz
2012-08-07 21:44:24 -07:00
1528758ac5
Bug 649740 - Implement @supports. r=dbaron,bz
2012-08-02 10:32:12 +10:00
d0ad5a7d0c
Bug 777292 part 2 - Change all nsnull to nullptr
2012-07-30 17:20:58 +03:00
65eef91085
Bug 626472 part 1 - Define nsnull as nullptr where available; r=ehsan
2012-07-20 14:16:17 +03:00
fcf99ba406
Bug 758885. Don't apply the dynamic :hover reresolution skipping optimization to selectors which can match on mutable state other than :hover. r=dbaron
2012-05-29 00:17:45 -04:00
82ff7027aa
Bug 716478 - update licence to MPL 2.
2012-05-21 12:12:37 +01:00
0a84303b99
Bug 722853 - Derive private browsing status from docshell when assigning element states. r=dbaron
2012-03-30 02:16:22 -04:00
cd6e787733
Backout bug 722853 - changeset f40a17213344; r=orange
2012-03-26 18:41:19 +02:00
5b05bfee0b
Bug 722853 - Derive private browsing status from docshell when assigning element states. r=dbaron
...
--HG--
extra : rebase_source : d59985b415ccafc4a9c8e8dbea9b69fa61602b8b
2012-02-21 09:57:09 -05:00
64e0bf573c
Bug 735481. Skip ancestor filtering on IDs and classes in quirks mode, because that matching needs to be case-insensitive. r=dbaron
2012-03-14 00:44:13 -05:00
0b6e42d918
Bug 732667. Only do checks for :hover selectors when hover state changes on nodes which have relevant hover rules. r=dbaron
2012-03-12 22:54:15 -05:00
4e80e64553
Bug 705877 part 4. Use the TreeMatchContext's ancestor filter, if any, in EnumerateAllRules. r=dbaron
2012-03-12 22:54:15 -05:00
3e12270964
Bug 705877 part 3. Hang an optional Bloom filter and some methods for managing it off the TreeMatchContext. r=dbaron
2012-03-12 22:54:15 -05:00
0f31ab2288
Bug 705877 part 2. Store the hashes of the atoms occurring as IDs, classes, and tag names in selectors that would need to match one of our ancestors. r=dbaron
...
We store a maximum of 4 hashes to save space. Interesting questions are whether 4 is the right number
and whether we should, prioritize ids and classes over tags. In practice, this seems to be good enough
at least performance-wise, though it's possible we could use less memory and still do OK on the performance.
2012-03-12 22:54:14 -05:00
ef5d0d917f
Back out 83a33626d965:6e9330f5a9bd (bug 732667 and bug 705877) for Windows build bustage
2012-03-12 22:12:41 -07:00
4de57bccc9
Bug 732667. Only do checks for :hover selectors when hover state changes on nodes which have relevant hover rules. r=dbaron
2012-03-12 22:54:15 -05:00
c3ed68f368
Bug 705877 part 4. Use the TreeMatchContext's ancestor filter, if any, in EnumerateAllRules. r=dbaron
2012-03-12 22:54:15 -05:00
f593cd5e83
Bug 705877 part 3. Hang an optional Bloom filter and some methods for managing it off the TreeMatchContext. r=dbaron
2012-03-12 22:54:15 -05:00
e9f9272eeb
Bug 705877 part 2. Store the hashes of the atoms occurring as IDs, classes, and tag names in selectors that would need to match one of our ancestors. r=dbaron
...
We store a maximum of 4 hashes to save space. Interesting questions are whether 4 is the right number
and whether we should, prioritize ids and classes over tags. In practice, this seems to be good enough
at least performance-wise, though it's possible we could use less memory and still do OK on the performance.
2012-03-12 22:54:14 -05:00
68a79f10ae
Bug 731615 - Eliminate duplicate headers in layout. rs=smontagu
2012-02-29 10:57:47 -05:00
f4fb4a1991
Bug 729142 - Convert layout/style to MOZ_STATIC_ASSERT. r=dbaron
2012-02-23 08:19:00 -08:00
580e841242
Bug 708874 - API for locking pseudo-class state of an element. r=bz
2012-02-08 10:42:38 -08:00
7559a70a04
Bug 715453 - Remove computedSize from nsMallocSizeOfFun. r=jlebar,bhackett.
...
--HG--
extra : rebase_source : a65039a407daab45360a5b375b53cbf1bc05b7f6
2012-01-25 00:52:51 -08:00
0089711fa7
Bug 707865 - Convert nsTArray::SizeOf() to nsTArray::SizeOfExcludingThis(). r=jlebar.
...
--HG--
extra : rebase_source : d802d58bc7dedda2490878793923adc0ab55f779
2011-12-15 14:59:53 -08:00
293c6cddef
Bug 700914. Initialize the rulehash's hashtables lazily. r=dbaron
2011-12-09 05:29:01 -05:00
8ea852b335
Bug 705987 - Use mallocSizeOf in the layout memory reporters. r=khuey,bzbarsky
2011-12-08 21:01:52 -08:00
5da1a0cc40
Bug 689443. Switch PerWeightData back to using linked lists instead of arrays for the actual per-weight rule data, so as to reduce fragmentation. r=dbaron
...
This incidentally fixes a bug where we were copying the per-weight data very inefficiently.
2011-12-08 01:32:11 -05:00
8582f44c33
Bug 704911. Don't restyle based on state selectors that match our node but don't depend on the state that's changing. r=dbaron
2011-12-05 23:58:15 -05:00
5db488079f
Backout 15cf68a3c027, 2f403e4c42c8, 61534874aed7, 1cb3705557d9 & 03970d1fc919 (bug 661746) for Win PGO build failures; a=bz
2011-11-14 18:10:22 +00:00
1475029a2d
Bug 661746 - Part 5: Avoid creating atoms during selector matching. r=bzbarsky
2011-11-14 16:30:28 +13:00
3cf667fd11
Bug 661746 - Part 4: use const nsIDocument in selector matching. r=bzbarsky
2011-11-14 16:30:22 +13:00
1e9e43f12d
Bug 661746 - Part 3: Use const selectors during selector matching. r=bzbarsky
2011-11-14 16:30:16 +13:00
87c926f7b6
Bug 661746 - Part 2: Make selector matching work with const elements. r=bzbarsky
2011-11-14 16:30:08 +13:00
dd2d24bcbf
Bug 661746 - Part 1: Delay setting slow selector flags during selector matching to avoid mutating Elements. r=bzbarsky
2011-11-14 16:29:56 +13:00
52ccc4cb61
Bug 660959 - Update link state less lazily r=bz
2011-11-14 16:24:41 +13:00
8ded9164c4
Bug 641731: Ignore :visited status in SVG image documents. r=dbaron
2011-10-29 02:43:43 -07:00
2c2da3099f
Bug 682420 - Rename nsINode::GetOwnerDoc to nsINode::OwnerDoc, part 1, r=jst
...
--HG--
extra : rebase_source : baf1a25cdea68d499a7673fdf96e27b5a12dc83c
2011-10-18 13:53:36 +03:00
92064e6d3f
Bug 690892 - Replace PR_TRUE/PR_FALSE with true/false on mozilla-central; rs=dbaron
...
Landing on a CLOSED TREE
2011-10-17 10:59:28 -04:00
b5e57820bf
Bug 693469 - Implement mozilla::ArrayLength and mozilla::ArrayEnd, and replace uses of NS_ARRAY_LENGTH whenever possible. (Exceptions: assigning to static initializers, use in static assertions, as template parameters, etc. These will go away when the relevant compilers have C++11 constexpr support.) r=cjones
...
--HG--
extra : rebase_source : b4bae9a0f85abf2feb828609b50e756916b99a6f
2011-10-10 22:50:08 -07:00
d0532c3e95
Back out everything since 5435ee09cf7b. Tinderbox compilers hate me. r=epic-fail
2011-10-12 12:21:53 -07:00
0f30b277bf
Bug 639469 - Implement mozilla::ArrayLength and mozilla::ArrayEnd, and replace uses of NS_ARRAY_LENGTH whenever possible. r=cjones
...
--HG--
extra : rebase_source : 860a8271bf7e51cde358f2a4185c410de1fc2960
2011-10-10 22:50:08 -07:00
46d661d94c
Bug 681755. Switch some of the style system's arrays in hashtables to be auto arrays. r=dbaron
2011-10-05 22:06:37 -04:00
9e403ddbf6
Back out 81ca82f27831 (bug 681755) due to incorrect leak reporting.
...
(I accidentally pushed the wrong version of this patch.)
2011-10-05 22:02:44 -04:00
eb19b99546
Bug 681755. Switch some of the style system's arrays in hashtables to be auto arrays. r=dbaron
2011-09-15 15:21:52 -07:00
Kyle Huey
Simon Montagu
Ehsan Akhgari
Aryeh Gregor
Matt Brubeck
Cameron McCormack
Boris Zbarsky
Gervase Markham
Josh Matthews
Ms2ger
Phil Ringnalda
Nathan Froyd
Zack Weinberg
Heather Arthur
Nicholas Nethercote
Ed Morley
David Zbarsky
Daniel Holbert
Olli Pettay
Jeff Walden
Justin Lebar