Jump to content

My Browser Builds (Part 3)

roytam1
 Share

Recommended Posts

roytam1

roytam1

Posted
  • Author
Posted
1 hour ago, Egorkaru said:

Why are images on the site https://lurkmore.to not loaded in the Pale Moon 26.5.0 browser build for Windows 2000 SP4? When accessing the server https://lurkmore.so, the error "SSL_ERROR_NO_CYPHER_OVERLAP" appears. The same error appears in the RetroZilla Browser, RetroZilla Suite and K-Meleon 1.5.4 for Windows 9x. There is no such problem with Wikipedia.

https://www.ssllabs.com/ssltest/analyze.html?d=lurkmore.so&s=172.67.156.114

Cipher Suites

# TLS 1.3 (server has no preference)

TLS_AES_128_GCM_SHA256 (0x1301)   ECDH x25519 (eq. 3072 bits RSA)   FS 128

TLS_AES_256_GCM_SHA384 (0x1302)   ECDH x25519 (eq. 3072 bits RSA)   FS 256

TLS_CHACHA20_POLY1305_SHA256 (0x1303)   ECDH x25519 (eq. 3072 bits RSA)   FS 256

# TLS 1.2 (suites in server-preferred order)

TLS_ECDHE_ECDSA_WITH_AES_128_GCM_SHA256 (0xc02b)   ECDH x25519 (eq. 3072 bits RSA)   FS 128

OLD_TLS_ECDHE_ECDSA_WITH_CHACHA20_POLY1305_SHA256 (0xcc14)   ECDH x25519 (eq. 3072 bits RSA)   FS 256 P

TLS_ECDHE_ECDSA_WITH_CHACHA20_POLY1305_SHA256 (0xcca9)   ECDH x25519 (eq. 3072 bits RSA)   FS 256 P

TLS_ECDHE_ECDSA_WITH_AES_128_CBC_SHA (0xc009)   ECDH x25519 (eq. 3072 bits RSA)   FS   WEAK 128

TLS_ECDHE_ECDSA_WITH_AES_256_GCM_SHA384 (0xc02c)   ECDH x25519 (eq. 3072 bits RSA)   FS 256

TLS_ECDHE_ECDSA_WITH_AES_256_CBC_SHA (0xc00a)   ECDH x25519 (eq. 3072 bits RSA)   FS   WEAK 256

TLS_ECDHE_ECDSA_WITH_AES_128_CBC_SHA256 (0xc023)   ECDH x25519 (eq. 3072 bits RSA)   FS   WEAK 128

TLS_ECDHE_ECDSA_WITH_AES_256_CBC_SHA384 (0xc024)   ECDH x25519 (eq. 3072 bits RSA)   FS   WEAK 256

# TLS 1.1 (suites in server-preferred order)

TLS_ECDHE_ECDSA_WITH_AES_128_CBC_SHA (0xc009)   ECDH x25519 (eq. 3072 bits RSA)   FS   WEAK 128

TLS_ECDHE_ECDSA_WITH_AES_256_CBC_SHA (0xc00a)   ECDH x25519 (eq. 3072 bits RSA)   FS   WEAK 256

# TLS 1.0 (suites in server-preferred order)

TLS_ECDHE_ECDSA_WITH_AES_128_CBC_SHA (0xc009)   ECDH x25519 (eq. 3072 bits RSA)   FS   WEAK 128

TLS_ECDHE_ECDSA_WITH_AES_256_CBC_SHA (0xc00a)   ECDH x25519 (eq. 3072 bits RSA)   FS   WEAK 256

 

there are some cipher suites that old platform NSS supports, you may check if you disabled them via about:config. supported suites with my old platform NSS belows:

Cipher Suites (in order of preference)

TLS_ECDHE_ECDSA_WITH_AES_128_GCM_SHA256 (0xc02b)   Forward Secrecy 128

TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256 (0xc02f)   Forward Secrecy 128

TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384 (0xc030)   Forward Secrecy 256

TLS_ECDHE_ECDSA_WITH_CHACHA20_POLY1305_SHA256 (0xcca9)   Forward Secrecy 256

TLS_ECDHE_RSA_WITH_CHACHA20_POLY1305_SHA256 (0xcca8)   Forward Secrecy 256

TLS_ECDHE_ECDSA_WITH_AES_256_CBC_SHA (0xc00a)  WEAK 256

TLS_ECDHE_ECDSA_WITH_AES_128_CBC_SHA (0xc009)  WEAK 128

TLS_ECDHE_RSA_WITH_AES_128_CBC_SHA (0xc013)  WEAK 128

TLS_ECDHE_RSA_WITH_AES_128_CBC_SHA256 (0xc027)  WEAK 128

TLS_ECDHE_RSA_WITH_AES_256_CBC_SHA (0xc014)  WEAK 256

TLS_DHE_RSA_WITH_AES_128_CBC_SHA (0x33)  WEAK 128

TLS_DHE_RSA_WITH_AES_256_CBC_SHA (0x39)  WEAK 256

TLS_DHE_RSA_WITH_3DES_EDE_CBC_SHA (0x16)  WEAK 112

TLS_ECDH_ECDSA_WITH_AES_128_CBC_SHA (0xc004)  WEAK 128

TLS_ECDH_RSA_WITH_AES_128_CBC_SHA (0xc00e)  WEAK 128

TLS_ECDH_ECDSA_WITH_AES_256_CBC_SHA (0xc005)  WEAK 256

TLS_ECDH_RSA_WITH_AES_256_CBC_SHA (0xc00f)  WEAK 256

TLS_RSA_WITH_AES_128_CBC_SHA (0x2f)  WEAK 128

TLS_RSA_WITH_AES_256_CBC_SHA (0x35)  WEAK 256

1 hour ago, Egorkaru said:

add support for TLS 1.3

not quite possible to add TLS 1.3 to old platform NSS since old compilers (MSVC6) can't accept new C/C++ syntax appeared in newer NSS changes.

but patches are welcomed. if anyone ported them and want me to merge it, once it is tested with old MSVC6, changes will be merged.

2
Link to comment
Share on other sites

ED_Sln

ED_Sln

Posted
Posted (edited)
3 hours ago, Egorkaru said:

Why are images on the site https://lurkmore.to not loaded in the Pale Moon 26.5.0 browser build for Windows 2000 SP4?

Все открывается. Возможно сайт опять заблокировал РКН, но у меня он открывается без проблем.

Everything opens. Perhaps the site again blocked RKN, but it opens without problems.

spacer.png

Edited by ED_Sln
Link to comment
Share on other sites
ClassicNick

ClassicNick

Posted
Posted
15 hours ago, roytam1 said:

not quite possible to add TLS 1.3 to old platform NSS since old compilers (MSVC6) can't accept new C/C++ syntax appeared in newer NSS changes.

but patches are welcomed. if anyone ported them and want me to merge it, once it is tested with old MSVC6, changes will be merged.

I tried building palemoon26-master on Visual C++ 2010 with the Windows 7 SDK, and I get errors relating to "subdata : undeclared identifier". Challenge: Try building NSS 3.48.5.0 on Visual C++ 2010 or earlier, and post your error messages. I'm reluctant to try building New Moon 27, but is there a way to build NSS 3.48.5.0 without building New Moon 27?

Link to comment
Share on other sites
roytam1

roytam1

Posted
  • Author
Posted (edited)
3 hours ago, ClassicNick said:

I tried building palemoon26-master on Visual C++ 2010 with the Windows 7 SDK, and I get errors relating to "subdata : undeclared identifier". Challenge: Try building NSS 3.48.5.0 on Visual C++ 2010 or earlier, and post your error messages. I'm reluctant to try building New Moon 27, but is there a way to build NSS 3.48.5.0 without building New Moon 27?

NM26-master branch targets VS2012/2013, and VC2010 is not supported.

oldvc branch may support VC2010.

for building NSS alone, you may check mozilla's doc out at: https://developer.mozilla.org/en-US/docs/Mozilla/Projects/NSS/Building

Edited by roytam1
Link to comment
Share on other sites
VistaLover

VistaLover

Posted
Posted
9 hours ago, roytam1 said:

new UXP based builds are not available tomorrow.

And, as far as upstream are concerned, the same stands true for the remainder of this month and the first week of November: 

https://forum.palemoon.org/viewtopic.php?p=220767#p220767
http://developer.palemoon.org/docs/release-engineering/ 

Pale Moon Version - CLOSED TREE - Intended Release Date - Type of Release

29.5.0	            2021-11-02	  2021-11-09	          Major/Feature Release

:P

1
Link to comment
Share on other sites
roytam1

roytam1

Posted
  • Author
Posted

New NewMoon 27 Build!

32bit https://o.rthost.win/palemoon/palemoon-27.10.0.win32-git-20211016-f355dbe3d-xpmod.7z
32bit SSE https://o.rthost.win/palemoon/palemoon-27.10.0.win32-git-20211016-f355dbe3d-xpmod-sse.7z
32bit noSSE https://o.rthost.win/palemoon/palemoon-27.10.0.win32-git-20211016-f355dbe3d-xpmod-ia32.7z

64bit https://o.rthost.win/palemoon/palemoon-27.10.0.win64-git-20211016-f355dbe3d-xpmod.7z

source repo: https://github.com/roytam1/palemoon27

repo changes since my last build:
- import changes from `dev' branch of rmottola/Arctic-Fox:
 - Bug 774364 - Part 4: Add setRNGState testing function. r=sstangl (648a491f0)
 - Bug 1206591 - IonMonkey: MIPS32: Move LIRGeneratorMIPS::visitRandom to architecture-specific. r=arai (edcb0fa60)
 - Bug 1205917 - IonMonkey: MIPS32: Use rshiftPtrArithmetic in clampIntToUint8. r=arai (10e234d60)
 - Bug 1205921 - IonMonkey: MIPS32: Fix scratch register in MacroAssembler::memIntToValue. r=evilpie (5682d93f1)
 - Bug 1204604: Check for int32 value, not int32 type, in SIMD.T.{replaceLane,shuffle,swizzle}; r=lth (18dca9450)
 - Bug 1180763 - Part 1: Add the size of the tenured object to the tenure promotions log entry; r=sfink (456827a5e)
 - Bug 1180763 - Part 0: Document the size property of entries in the tenure promotions log; r=sfink (96244a8f9)
 - Bug 1180763 - Part 2: Test the size property of entries in the tenure promotions log; r=sfink (602225219)
 - Bug 1208867 - Add xxx64 operations to MacroAssembler-none. r=sstangl (c3a8207bf)
 - Bug 1203964 - Part 1: Move MacroAssembler::rshiftPtr into generic macro assembler. r=h4writer (2c6486a37)
 - Bug 1203964 - Part 2: Move MacroAssembler::rshiftPtrArithmetic into generic macro assembler. r=jandem (ce3e31ade)
 - Bug 1203964 - Part 3: Move MacroAssembler::rshift64 into generic macro assembler. r=hev (8f395fad1)
 - Bug 1203964 - Part 4: Move MacroAssembler::lshiftPtr into generic macro assembler. r=nbp2 (99a45b756)
 - Bug 1203964 - Part 5: Move MacroAssembler::lshift64 into generic macro assembler. r=sstangl (3bcd9eb5f)
 - Bug 1136806 - Fix assertion failure with Debugger and oomAfterAllocations; r=shu (efcab9ab7) (3d7e107cc)
- import changes from `dev' branch of rmottola/Arctic-Fox:
 - Bug 1149185 - override GetMessageLoop() r=nical (42bec9405)
 - Add a "blacklisted" property to d3d11 telemetry. (bug 1187453, r=mattwoodrow) (83bcb0e8e)
 - Don't create a D3D11 compositor device on content processes. (bug 1183910 part 8, r=bas,mattwoodrow) (c03d496a6)
 - Bug 1168889 - ContentClientSingleBuffered needs to upload during the transaction. r=nical (35171fe14)
 - Bug 1176077 - Only recompute PaintedLayer item visibility when display list has changed. r=mattwoodrow (0ea691594)
 - Bug 1179298. ClientTiledPaintedLayer should allow empty transactions to succeed if it doesn't need to draw anything. r=mattwoodrow (4aaad97b2)
 - Bug 1169200 - Send OpRemoveTextureAsync before actor destroyed r=nical (86007156c)
 - Bug 1158838 - Add some gfxCriticalError logging. r=milan (32bf80848)
 - Bug 1129871 - Avoid copying TileClient when not necessary. r=BenWa (1ac30bcc2)
 - Bug 1186911 - Fix progressive paint when using tiled-drawtarget. r=nical (87649fb8d)
 - Bug 1118876 - Add edge padding with DrawTargetTiled. r=jrmuizel (64b12cf6a)
 - Bug 1157677 - Fix uninitialized members warning in ClientTiledLayerBuffer. r=nical (d11d0da30)
 - Bug 1151071 - Make sure low-resolution tiles are transparent. r=mattwoodrow (164446378)
 - Bug 1170390 - Use Optimal2DFormatForContent over OptimalFormatForContent to avoid enum conversions. r=jrmuizel (db7ee2c31)
 - Bug 1180326 - Part 1: Abstract parts of the client side tiling that we need. r=jrmuizel (0ad7651fe)
 - Bug 1179987 - Remove painted region from SurfaceDescriptorTiles since it isn't used any more. r=nical (653e9389e)
 - Bug 1180326 - Part 2: Add support for variable tile sizes. r=jrmuizel (081349a5e)
 - Bug 1180326 - Part 3: Move repeated code into GetBackBuffer. r=jrmuizel (2e9091ae1)
 - Bug 1180326 - Part 4: Add new content client. r=jrmuizel (7362d805b)
 - Bug 1180326 - Part 5: Support HWC with tiling when we only have 1 tile. r=jrmuizel (09ce93b7e)
 - Bug 1152461 - Don't use tiled layers for overflow:hidden scroll frames. r=tn (804a683a1)
 - Bug 961887 - Add some reftest fuzz. (dcd707004)
 - Bug 1186061 patch 1 - Add feature to reftest harness to allow skipping the flush. r=mattwoodrow (8cefe9640)
 - Bug 1176969 - Disable OMT animation for any frame in a preserve-3d scene rather than only frames whose parent and child are in a preserve-3d scene. r=mattwoodrow (693d0a518)
 - Bug 1176969 followup - Annotate new reftests as intermittently fuzzy on Mac. (0269d5d8f)
 - Bug 1176969 followup - Increase error amounts for intermittent fuzzy-if(cocoaWidget). (116039ed0)
 - Bug 1176969 followup - Mark test as random for now until I have a chance to rewrite it. (8ba561f83)
 - Bug 1176969 followup - Reorganize the timing of the reftest to make it more reliable. (f3ffc93cd)
 - Bug 1186061 patch 2 - Disable compositor thread animation of transforms when backface-visibility is hidden. r=mattwoodrow (e67bed9fb)
 - Bug 1169666 - Mark reftests failing due to scrollbar transparency random on GTK3. r=karlt (0c6080d55)
 - Bug 1180326 - Part 6: Use SingleTiledContentClient for non-scrollable layers on b2g and OSX. r=jrmuizel (079951f0a)
 - Bug 1189261 - Mark tile invalid region in coordinates relative to the tile. r=jrmuizel (7e696c911)
 - Bug 1187619 - Only optmimize FrameLayerBuilder visibility calculations if correct. r=mattwoodrow (05228ad31)
 - Bug 1190950: Check mappings and surfaces for UpdateFromSurface more robustly. r=mattwoodrow (c00adba10)
 - Remove the backend flag to TextureClient::CreateForDrawing. (bug 1183910 part 9, r=mattwoodrow) (5df1d964c)
 - Bug 1135935 - Part 1: Don't implement ISurfaceAllocator for the texture recycler. r=sotaro (bcba93d3d)
 - Bug 1135935 - Part 2: Merge TextureClientRecycleAllocator into base class. r=sotaro (8f4e7b6b1)
 - Bug 1135935 - Part 3: Allow overriding of allocations for texture client recycling. r=sotaro (35d7499e9)
 - Bug 1135935 - Part 4: Move IDirect3DTexture9 allocations and ownership into the TextureClient. r=jrmuizel (e62cf9acf)
 - Bug 1135935 - Part 5: Add D3D9 texture recycler. r=jrmuizel (8ab85b064)
 - Bug 1135935 - Part 6: Wait for the compositor to stop using textures before recycling them. r=jrmuizel (ae0c45cfd) (96836f9b3)
- import changes from `dev' branch of rmottola/Arctic-Fox:
 - Bug 1156600 - asmjscache: fix assert when reading metadata file for cache write fails (r=janv) (2b9895e97)
 - Bug 1173756 - Part 1: Update asmjscache to use PBackground; r=luke (adb326d80)
 - Bug 1173756 - Part 2: Merge MainProcessRunnable and ParentProcessRunnable; r=luke (b42ffa5f3)
 - Bug 1173756 - Part 3: Merge File and ChildRunnable; r=luke (961701adf)
 - Bug 1194555 - Part 0: Cleanup GetReportsState constructor. r=njn (4ae530854)
 - Bug 1145050 - Fix uninitialized class members in Http2Session. r=hurley (a12a9d372)
 - Bug 1132357 - remove h2-draft* support. r=mcmanus (24b8f347f)
 - Bug 986302 - Add memory reporting for HPACK tables r=hurley r=njn (0d2ebae56)
 - Bug 1123386 - Part 1: Update the clang build script for the most recent clang; r=rail (394f24496)
 - Bug 1123386 - Part 3: Mark DoNothingCallback::Callback as override in order to fix a build issue in warnings as errors mode (1d236acbf)
 - Bug 1123386 - Part 2: Temporarily relax the libstdc++ symbol version requirements for Linux static analysis builds; r=glandium (1415a4d90)
 - Bug 828844 - Add a "system-heap-allocated" memory report on Linux. r=glandium. (eead43b4e)
 - Bug 1194061 - Implement "system-heap-allocated" reporter for Windows. r=dmajor. (228385cb4) (d0c58681f)
- import changes from `dev' branch of rmottola/Arctic-Fox:
 - Bug 1194555 - Part 1: Remove RunReportersForThisProcess. r=njn (5c745b3ce)
 - Bug 1145007 (part 1) - Test |residentUnique| in test_memoryReporters.xul. r=erahm. (bc2fa3d30)
 - Bug 1145007 (part 2) - Add a "resident-peak" distinguished amount and memory reporter on Unix. r=erahm. (007256c58)
 - Bug 1194555 - Part 2: Remove |explicit| attribute from nsIMemoryReporterManager. r=njn (a0db6068e)
 - Bug 1203786 - test_memoryReporters.xul should ignore namespaces. r=njn (04e8842aa)
 - Bug 1194555 - Part 3: Remove |getReportsForThisProcess| from the nsIMemoryReporterManager interface. r=njn (cc3282c7f)
 - Bug 1194555 - Part 4: Rename GetReportsState PendingProcessesState. r=njn (3dab1b600)
 - Bug 1194555 - Part 5: Add option for async memory reporters. r=njn (f05ae83da)
 - Bug 1194555 - Part 6: Run reporters asynchronously. r=njn,jld,ted (d513fcd77) (f355dbe3d)

1
Link to comment
Share on other sites
roytam1

roytam1

Posted
  • Author
Posted

New regular/weekly KM-Goanna release:
https://o.rthost.win/kmeleon/KM76.4.5-Goanna-20211016.7z

Changelog:

Out-of-tree changes:
* update Goanna3 to git fae40c10f...f355dbe3d:
- import changes from `dev' branch of rmottola/Arctic-Fox:
 - Bug 774364 - Part 4: Add setRNGState testing function. r=sstangl (648a491f0)
 - Bug 1206591 - IonMonkey: MIPS32: Move LIRGeneratorMIPS::visitRandom to architecture-specific. r=arai (edcb0fa60)
 - Bug 1205917 - IonMonkey: MIPS32: Use rshiftPtrArithmetic in clampIntToUint8. r=arai (10e234d60)
 - Bug 1205921 - IonMonkey: MIPS32: Fix scratch register in MacroAssembler::memIntToValue. r=evilpie (5682d93f1)
 - Bug 1204604: Check for int32 value, not int32 type, in SIMD.T.{replaceLane,shuffle,swizzle}; r=lth (18dca9450)
 - Bug 1180763 - Part 1: Add the size of the tenured object to the tenure promotions log entry; r=sfink (456827a5e)
 - Bug 1180763 - Part 0: Document the size property of entries in the tenure promotions log; r=sfink (96244a8f9)
 - Bug 1180763 - Part 2: Test the size property of entries in the tenure promotions log; r=sfink (602225219)
 - Bug 1208867 - Add xxx64 operations to MacroAssembler-none. r=sstangl (c3a8207bf)
 - Bug 1203964 - Part 1: Move MacroAssembler::rshiftPtr into generic macro assembler. r=h4writer (2c6486a37)
 - Bug 1203964 - Part 2: Move MacroAssembler::rshiftPtrArithmetic into generic macro assembler. r=jandem (ce3e31ade)
 - Bug 1203964 - Part 3: Move MacroAssembler::rshift64 into generic macro assembler. r=hev (8f395fad1)
 - Bug 1203964 - Part 4: Move MacroAssembler::lshiftPtr into generic macro assembler. r=nbp2 (99a45b756)
 - Bug 1203964 - Part 5: Move MacroAssembler::lshift64 into generic macro assembler. r=sstangl (3bcd9eb5f)
 - Bug 1136806 - Fix assertion failure with Debugger and oomAfterAllocations; r=shu (efcab9ab7) (3d7e107cc)
- import changes from `dev' branch of rmottola/Arctic-Fox:
 - Bug 1149185 - override GetMessageLoop() r=nical (42bec9405)
 - Add a "blacklisted" property to d3d11 telemetry. (bug 1187453, r=mattwoodrow) (83bcb0e8e)
 - Don't create a D3D11 compositor device on content processes. (bug 1183910 part 8, r=bas,mattwoodrow) (c03d496a6)
 - Bug 1168889 - ContentClientSingleBuffered needs to upload during the transaction. r=nical (35171fe14)
 - Bug 1176077 - Only recompute PaintedLayer item visibility when display list has changed. r=mattwoodrow (0ea691594)
 - Bug 1179298. ClientTiledPaintedLayer should allow empty transactions to succeed if it doesn't need to draw anything. r=mattwoodrow (4aaad97b2)
 - Bug 1169200 - Send OpRemoveTextureAsync before actor destroyed r=nical (86007156c)
 - Bug 1158838 - Add some gfxCriticalError logging. r=milan (32bf80848)
 - Bug 1129871 - Avoid copying TileClient when not necessary. r=BenWa (1ac30bcc2)
 - Bug 1186911 - Fix progressive paint when using tiled-drawtarget. r=nical (87649fb8d)
 - Bug 1118876 - Add edge padding with DrawTargetTiled. r=jrmuizel (64b12cf6a)
 - Bug 1157677 - Fix uninitialized members warning in ClientTiledLayerBuffer. r=nical (d11d0da30)
 - Bug 1151071 - Make sure low-resolution tiles are transparent. r=mattwoodrow (164446378)
 - Bug 1170390 - Use Optimal2DFormatForContent over OptimalFormatForContent to avoid enum conversions. r=jrmuizel (db7ee2c31)
 - Bug 1180326 - Part 1: Abstract parts of the client side tiling that we need. r=jrmuizel (0ad7651fe)
 - Bug 1179987 - Remove painted region from SurfaceDescriptorTiles since it isn't used any more. r=nical (653e9389e)
 - Bug 1180326 - Part 2: Add support for variable tile sizes. r=jrmuizel (081349a5e)
 - Bug 1180326 - Part 3: Move repeated code into GetBackBuffer. r=jrmuizel (2e9091ae1)
 - Bug 1180326 - Part 4: Add new content client. r=jrmuizel (7362d805b)
 - Bug 1180326 - Part 5: Support HWC with tiling when we only have 1 tile. r=jrmuizel (09ce93b7e)
 - Bug 1152461 - Don't use tiled layers for overflow:hidden scroll frames. r=tn (804a683a1)
 - Bug 961887 - Add some reftest fuzz. (dcd707004)
 - Bug 1186061 patch 1 - Add feature to reftest harness to allow skipping the flush. r=mattwoodrow (8cefe9640)
 - Bug 1176969 - Disable OMT animation for any frame in a preserve-3d scene rather than only frames whose parent and child are in a preserve-3d scene. r=mattwoodrow (693d0a518)
 - Bug 1176969 followup - Annotate new reftests as intermittently fuzzy on Mac. (0269d5d8f)
 - Bug 1176969 followup - Increase error amounts for intermittent fuzzy-if(cocoaWidget). (116039ed0)
 - Bug 1176969 followup - Mark test as random for now until I have a chance to rewrite it. (8ba561f83)
 - Bug 1176969 followup - Reorganize the timing of the reftest to make it more reliable. (f3ffc93cd)
 - Bug 1186061 patch 2 - Disable compositor thread animation of transforms when backface-visibility is hidden. r=mattwoodrow (e67bed9fb)
 - Bug 1169666 - Mark reftests failing due to scrollbar transparency random on GTK3. r=karlt (0c6080d55)
 - Bug 1180326 - Part 6: Use SingleTiledContentClient for non-scrollable layers on b2g and OSX. r=jrmuizel (079951f0a)
 - Bug 1189261 - Mark tile invalid region in coordinates relative to the tile. r=jrmuizel (7e696c911)
 - Bug 1187619 - Only optmimize FrameLayerBuilder visibility calculations if correct. r=mattwoodrow (05228ad31)
 - Bug 1190950: Check mappings and surfaces for UpdateFromSurface more robustly. r=mattwoodrow (c00adba10)
 - Remove the backend flag to TextureClient::CreateForDrawing. (bug 1183910 part 9, r=mattwoodrow) (5df1d964c)
 - Bug 1135935 - Part 1: Don't implement ISurfaceAllocator for the texture recycler. r=sotaro (bcba93d3d)
 - Bug 1135935 - Part 2: Merge TextureClientRecycleAllocator into base class. r=sotaro (8f4e7b6b1)
 - Bug 1135935 - Part 3: Allow overriding of allocations for texture client recycling. r=sotaro (35d7499e9)
 - Bug 1135935 - Part 4: Move IDirect3DTexture9 allocations and ownership into the TextureClient. r=jrmuizel (e62cf9acf)
 - Bug 1135935 - Part 5: Add D3D9 texture recycler. r=jrmuizel (8ab85b064)
 - Bug 1135935 - Part 6: Wait for the compositor to stop using textures before recycling them. r=jrmuizel (ae0c45cfd) (96836f9b3)
- import changes from `dev' branch of rmottola/Arctic-Fox:
 - Bug 1156600 - asmjscache: fix assert when reading metadata file for cache write fails (r=janv) (2b9895e97)
 - Bug 1173756 - Part 1: Update asmjscache to use PBackground; r=luke (adb326d80)
 - Bug 1173756 - Part 2: Merge MainProcessRunnable and ParentProcessRunnable; r=luke (b42ffa5f3)
 - Bug 1173756 - Part 3: Merge File and ChildRunnable; r=luke (961701adf)
 - Bug 1194555 - Part 0: Cleanup GetReportsState constructor. r=njn (4ae530854)
 - Bug 1145050 - Fix uninitialized class members in Http2Session. r=hurley (a12a9d372)
 - Bug 1132357 - remove h2-draft* support. r=mcmanus (24b8f347f)
 - Bug 986302 - Add memory reporting for HPACK tables r=hurley r=njn (0d2ebae56)
 - Bug 1123386 - Part 1: Update the clang build script for the most recent clang; r=rail (394f24496)
 - Bug 1123386 - Part 3: Mark DoNothingCallback::Callback as override in order to fix a build issue in warnings as errors mode (1d236acbf)
 - Bug 1123386 - Part 2: Temporarily relax the libstdc++ symbol version requirements for Linux static analysis builds; r=glandium (1415a4d90)
 - Bug 828844 - Add a "system-heap-allocated" memory report on Linux. r=glandium. (eead43b4e)
 - Bug 1194061 - Implement "system-heap-allocated" reporter for Windows. r=dmajor. (228385cb4) (d0c58681f)
- import changes from `dev' branch of rmottola/Arctic-Fox:
 - Bug 1194555 - Part 1: Remove RunReportersForThisProcess. r=njn (5c745b3ce)
 - Bug 1145007 (part 1) - Test |residentUnique| in test_memoryReporters.xul. r=erahm. (bc2fa3d30)
 - Bug 1145007 (part 2) - Add a "resident-peak" distinguished amount and memory reporter on Unix. r=erahm. (007256c58)
 - Bug 1194555 - Part 2: Remove |explicit| attribute from nsIMemoryReporterManager. r=njn (a0db6068e)
 - Bug 1203786 - test_memoryReporters.xul should ignore namespaces. r=njn (04e8842aa)
 - Bug 1194555 - Part 3: Remove |getReportsForThisProcess| from the nsIMemoryReporterManager interface. r=njn (cc3282c7f)
 - Bug 1194555 - Part 4: Rename GetReportsState PendingProcessesState. r=njn (3dab1b600)
 - Bug 1194555 - Part 5: Add option for async memory reporters. r=njn (f05ae83da)
 - Bug 1194555 - Part 6: Run reporters asynchronously. r=njn,jld,ted (d513fcd77) (f355dbe3d)

* Notice: the changelog above may not always applicable to XULRunner code which K-Meleon uses.

A goanna3 source tree that has kmeleon adaption patch applied is available here: https://github.com/roytam1/palemoon27/tree/kmeleon76

Link to comment
Share on other sites
mixit

mixit

Posted
Posted (edited)

Firefox's libcubeb has been updated and the "Youtube video freeze" fix is now officially part of mozilla-central: https://hg.mozilla.org/mozilla-central/rev/c7768090c505715c543b2a71e42b07fcb905d7a9.

Thanks again @roytam1 for submitting it! Let it not be said that MSFN hasn't contributed to upstream! :D

Edited by mixit
5
Link to comment
Share on other sites
UCyborg

UCyborg

Posted
Posted
8 hours ago, mixit said:

Thanks again @roytam1 for submitting it! Let it not be said that MSFN hasn't contributed to upstream! :D

You'd have to get out of your way and block MMDeviceEnumerator from being invoked in order for it to fall back to WinMM!:buehehe: If you wanted to try WinMM on Windows OS they target, that is.

1
Link to comment
Share on other sites
mixit

mixit

Posted
Posted
1 hour ago, UCyborg said:

You'd have to get out of your way and block MMDeviceEnumerator from being invoked in order for it to fall back to WinMM!:buehehe: If you wanted to try WinMM on Windows OS they target, that is.

You can actually change the audio backend in current Firefox versions by setting the media.cubeb.backend (string) pref. If you set it to winmm and restart the browser, about:support should show it under Audio Backend. At least it does on Windows 7, haven't tried on later versions. But since they rewrote the drivers after XP, the fix would be superfluous even when using WinMM. But it's still nice that it'll now be part of modern Firefox, considering the "MSFN hackers" meme, etc. :P :lol:

2
Link to comment
Share on other sites
UCyborg

UCyborg

Posted
Posted
1 hour ago, mixit said:

You can actually change the audio backend in current Firefox versions by setting the media.cubeb.backend (string) pref.

One of those that must be created manually, eh? Must be why I missed it since I didn't look at the code much. I just tested by preventing access to relevant registry key under HKCR\Classes\CLSID.

Link to comment
Share on other sites
Guest
This topic is now closed to further replies.
 Share
Go to topic listing

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...