summaryrefslogtreecommitdiffstats
path: root/third_party/mach_override
Commit message (Collapse)AuthorAgeFilesLines
* Roll mach_override to a newer version. badea@adobe.com2012-11-142-80/+35
| | | | | | | | | See https://code.google.com/p/chromium/issues/detail?id=138535#c5 Review URL: https://chromiumcodereview.appspot.com/11360250 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@167693 0039d316-1c4b-4281-b951-d872f2087c98
* Roll mach_override to a newer version. badea@adobe.com2012-11-012-4/+7
| | | | | | | | | See https://code.google.com/p/chromium/issues/detail?id=138535#c5 Review URL: https://chromiumcodereview.appspot.com/11275073 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@165380 0039d316-1c4b-4281-b951-d872f2087c98
* Roll mach_override to newer versions in small increments.badea@adobe.com2012-10-222-29/+76
| | | | | | | | | See https://code.google.com/p/chromium/issues/detail?id=138535#c5 Review URL: https://chromiumcodereview.appspot.com/11186020 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@163308 0039d316-1c4b-4281-b951-d872f2087c98
* Revert 150529 (very likely caused http://crbug.com/141707thakis@chromium.org2012-08-092-131/+103
| | | | | | | | | | | | | - Update third_party/mach_override from https://github.com/rentzsch/mach_star/mach_override to include the fix for building Chromium on 64 bits. BUG=138535 Review URL: https://chromiumcodereview.appspot.com/10834127 TBR=mmaerean@adobe.com Review URL: https://chromiumcodereview.appspot.com/10830249 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@150901 0039d316-1c4b-4281-b951-d872f2087c98
* Update third_party/mach_override from ↵mmaerean@adobe.com2012-08-082-103/+131
| | | | | | | | | | https://github.com/rentzsch/mach_star/mach_override to include the fix for building Chromium on 64 bits. BUG=138535 Review URL: https://chromiumcodereview.appspot.com/10834127 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@150529 0039d316-1c4b-4281-b951-d872f2087c98
* mac make: Fix build.thakis@chromium.org2011-09-161-0/+1
| | | | | | | | | | | | This was broken by http://src.chromium.org/viewvc/chrome?view=rev&revision=101378 BUG=none TEST=mac make build works again TBR=pkasting Review URL: http://codereview.chromium.org/7918016 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@101486 0039d316-1c4b-4281-b951-d872f2087c98
* Allocate mach_override "reentry islands" in high memory using vm_allocate,mark@chromium.org2011-08-222-2/+8
| | | | | | | | | | | rather than putting them in the heap using malloc. Ownership of page protection bits in the heap is dubious. BUG=93736 TEST=Watch the canary crashes tomorrow Review URL: http://codereview.chromium.org/7710011 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@97770 0039d316-1c4b-4281-b951-d872f2087c98
* Roll third_party/mach_override to 87f491f8acef924d2ba90dd55fc23ad64f9d5bbd.rsesek@chromium.org2011-08-202-14/+29
| | | | | | | | | | BUG=none TEST=none Review URL: http://codereview.chromium.org/7693006 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@97582 0039d316-1c4b-4281-b951-d872f2087c98
* Patch mach_override to not use printf() for errors, but fprintf(stderr, ...).rsesek@chromium.org2011-08-182-5/+7
| | | | | | | | | BUG=none TEST=Layout tests go green. Review URL: http://codereview.chromium.org/7655052 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@97379 0039d316-1c4b-4281-b951-d872f2087c98
* [Mac] Implement base::EnableTerminationOnHeapCorruption() by overriding ↵rsesek@chromium.org2011-08-182-1/+3
| | | | | | | | | | | | | | | | | malloc_error_break(). This makes malloc_error_break() fatal for all processes in an attempt to get better stack traces when heap corruption may occur. BUG=90884,91068,93191 TEST=See bug 93191 for repro steps. A crash report gets generated with a hopefully more-useful stack. Originally landed: http://src.chromium.org/viewvc/chrome?view=rev&revision=97315 Reverted: http://src.chromium.org/viewvc/chrome?view=rev&revision=97322 Review URL: http://codereview.chromium.org/7670025 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@97351 0039d316-1c4b-4281-b951-d872f2087c98
* Update mach_override to 51ae3d199463fa84548f466d649f0821d579fdaf to pick upmark@chromium.org2011-07-262-26/+14
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | our recent change (r93690, bug 79642). This also picks up: commit 965084a3fa2d8171508037601e1ea009bc0d8443 Author: rentzsch <jwr.git@redshed.net> Date: Wed Jul 13 11:20:08 2011 -0500 Add support for signal() and longjmp() on x86 and x86_64. (Ramosian Glider) Closes #6. commit 8a6962a636af2ab5e213eb317506cde5f1cdb8b6 Merge: 8f5baad f2061a9 Author: Jonathan 'Wolf' Rentzsch <jwr.git@redshed.net> Date: Sat Jun 25 12:51:34 2011 -0700 Merge pull request #5 from mkrautz/vm-protect-size Change size of region in vm_protect calls to be 8. commit f2061a9c26733a764a37585b7f93f362ca788ce5 Author: Mikkel Krautz <mikkel@krautz.dk> Date: Sat Jun 25 21:13:44 2011 +0200 Change size of region in vm_protect calls to be 8. commit 8f5baad6fe4fa73ecdf76d7d7cacec131723b8f5 Author: Jonathan 'Wolf' Rentzsch <jwr.git@redshed.net> Date: Sat Jun 25 11:52:25 2011 -0700 [FIX] sizeof(long) => sizeof(void*) for vm_protect(). Closes #4. BUG=none TEST=none Review URL: http://codereview.chromium.org/7461053 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@94131 0039d316-1c4b-4281-b951-d872f2087c98
* Lion x86 compatibility for mach_override.mark@chromium.org2011-07-222-13/+31
| | | | | | | | | | | This enables makeIslandExecutable for x86. Main executables linked and run on Mac OS X 10.7 ("Lion") have non-executable heap pages by default. BUG=79642 TEST=Build and run on Lion. Renderers should not crash instantly. Review URL: http://codereview.chromium.org/7492012 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@93690 0039d316-1c4b-4281-b951-d872f2087c98
* Remove the comments setting emands and vim tab width and expansion variables.tony@chromium.org2011-07-111-6/+0
| | | | | | | | | | | | | | | | | These were added in r28089 (http://codereview.chromium.org/256059) but are unnecessary bloat for everyone to carry around, even those that don't use emacs or vim. In an earlier change, I added editor config files in src/tools/emacs/ and src/tools/vim/ so users of the appropriate editor can source those instead. BUG=none TEST=none Review URL: http://codereview.chromium.org/7310019 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@92046 0039d316-1c4b-4281-b951-d872f2087c98
* Globally replace <(library) with static_libraryevan@chromium.org2011-05-191-1/+1
| | | | | | | | | | | We provided <(library) as a variable to support a peculiar build configuration on Linux. We no longer support that build configuration, so we can simplify this code to no longer use a variable. Review URL: http://codereview.chromium.org/7051014 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@86004 0039d316-1c4b-4281-b951-d872f2087c98
* Use new upstream version of mach_override that includes our local changes.thakis@chromium.org2011-04-182-149/+4
| | | | | | | | | BUG=none TEST=none Review URL: http://codereview.chromium.org/6873055 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@82007 0039d316-1c4b-4281-b951-d872f2087c98
* Mac: Let chrome coexist with the Chinese Handwriting IME.thakis@chromium.org2011-04-136-0/+1001
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The renderer sandbox breaks the Chinese Handwriting shortcut while Chrome is running, even when Chrome is not active: In a nutshell, Carbon tries to load the IME in every process, and if that fails, it uninstalls the toggle shorcut. It fails in the renderer process due to the sandbox. See the bug for details. Prewarming (patch set 1) doesn't work because that opens up clipboard access to the renderer – the IME doesn't load correctly if it doesn't have clipboard access. dyld __interpose based interposing (patch set 2) doesn't work because TISCreateInputSourceList and its broken caller are in the same framework (HIToolbox). Hence, use binary rewriting to return an arbitrary array with size > 0 to make HIToolbox believe that the IME loads correctly, so that it doesn't uninstall the global handwriting keyboard shortcut (only on >= 10.6). BUG=31225 TEST=Enable Chinese in the input menu. Make sure Chrome's not running. Note that ctrl-shift-space opens and dismisses the Handwriting IME (you need a multitouch trackpad for that). Open Chrome, note that the IME still works. (If you run Chrome without this patch, the IME will be broken until you do ps aux | grep Chinese and then kill $CHINESE_HANDWRITING_PID and then toggle the IME once through the Menu Extra menu.) It is a good idea to have a Terminal open during testing, because killing the task is the only way to dismiss the IME with the keyboard without this CL. Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=81308 Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=81332 Review URL: http://codereview.chromium.org/6801056 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@81500 0039d316-1c4b-4281-b951-d872f2087c98
* [Reverting 81332. Reason for revert: page_cycler_intl2 showing huge ↵nick@chromium.org2011-04-136-1002/+0
| | | | | | | | | | regression in vm_size_final_r on Mac 10.5 only: http://build.chromium.org/f/chromium/perf/mac-release-10.5/intl2/report.html?history=150&rev=-1&graph=vm_size_final_r] Revert 81332 - Mac: Let chrome coexist with the Chinese Handwriting IME.The renderer sandbox breaks the Chinese Handwriting shortcut while Chromeis running, even when Chrome is not active: In a nutshell, Carbon tries to loadthe IME in every process, and if that fails, it uninstalls the toggle shorcut. Itfails in the renderer process due to the sandbox. See the bug for details.Prewarming (patch set 1) doesn't work because that opens up clipboardaccess to the renderer - the IME doesn't load correctly if it doesn't haveclipboard access.dyld __interpose based interposing (patch set 2) doesn't work becauseTISCreateInputSourceList and its broken caller are in the sameframework (HIToolbox).Hence, use binary rewriting to return an arbitrary array with size > 0 to makeHIToolbox believe that the IME loads correctly, so that it doesn't uninstall theglobal handwriting keyboard shortcut.BUG=31225TEST=Enable Chinese in the input menu. Make sure Chrome's not running.Note that ctrl-shift-space opens and dismisses the Handwriting IME(you need a multitouch trackpad for that). Open Chrome, note that theIME still works.(If you run Chrome without this patch, the IME will be broken until you do ps aux | grep Chineseand then kill $CHINESE_HANDWRITING_PIDand then toggle the IME once through the Menu Extra menu.)It is a good idea to have a Terminal open during testing, because killingthe task is the only way to dismiss the IME with the keyboard without thisCL.Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=81308Review URL: http://codereview.chromium.org/6801056 TBR=thakis@chromium.org Review URL: http://codereview.chromium.org/6836022 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@81468 0039d316-1c4b-4281-b951-d872f2087c98
* Update mach_override's README.chromium with the xor change.thakis@chromium.org2011-04-131-0/+1
| | | | | | | | | | BUG=none TEST=none TBR=mark Review URL: http://codereview.chromium.org/6834009 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@81337 0039d316-1c4b-4281-b951-d872f2087c98
* Mac: Let chrome coexist with the Chinese Handwriting IME.thakis@chromium.org2011-04-126-0/+1001
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The renderer sandbox breaks the Chinese Handwriting shortcut while Chrome is running, even when Chrome is not active: In a nutshell, Carbon tries to load the IME in every process, and if that fails, it uninstalls the toggle shorcut. It fails in the renderer process due to the sandbox. See the bug for details. Prewarming (patch set 1) doesn't work because that opens up clipboard access to the renderer – the IME doesn't load correctly if it doesn't have clipboard access. dyld __interpose based interposing (patch set 2) doesn't work because TISCreateInputSourceList and its broken caller are in the same framework (HIToolbox). Hence, use binary rewriting to return an arbitrary array with size > 0 to make HIToolbox believe that the IME loads correctly, so that it doesn't uninstall the global handwriting keyboard shortcut. BUG=31225 TEST=Enable Chinese in the input menu. Make sure Chrome's not running. Note that ctrl-shift-space opens and dismisses the Handwriting IME (you need a multitouch trackpad for that). Open Chrome, note that the IME still works. (If you run Chrome without this patch, the IME will be broken until you do ps aux | grep Chinese and then kill $CHINESE_HANDWRITING_PID and then toggle the IME once through the Menu Extra menu.) It is a good idea to have a Terminal open during testing, because killing the task is the only way to dismiss the IME with the keyboard without this CL. Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=81308 Review URL: http://codereview.chromium.org/6801056 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@81332 0039d316-1c4b-4281-b951-d872f2087c98
* Revert 81308 (broke 10.5)- Mac: Let chrome coexist with the Chinese ↵thakis@chromium.org2011-04-126-997/+0
| | | | | | | | | Handwriting IME.The renderer sandbox breaks the Chinese Handwriting shortcut while Chromeis running, even when Chrome is not active: In a nutshell, Carbon tries to loadthe IME in every process, and if that fails, it uninstalls the toggle shorcut. Itfails in the renderer process due to the sandbox. See the bug for details.Prewarming (patch set 1) doesn't work because that opens up clipboardaccess to the renderer – the IME doesn't load correctly if it doesn't haveclipboard access.dyld __interpose based interposing (patch set 2) doesn't work becauseTISCreateInputSourceList and its broken caller are in the sameframework (HIToolbox).Hence, use binary rewriting to return an arbitrary array with size > 0 to makeHIToolbox believe that the IME loads correctly, so that it doesn't uninstall theglobal handwriting keyboard shortcut.BUG=31225TEST=Enable Chinese in the input menu. Make sure Chrome's not running.Note that ctrl-shift-space opens and dismisses the Handwriting IME(you need a multitouch trackpad for that). Open Chrome, note that theIME still works.(If you run Chrome without this patch, the IME will be broken until you do ps aux | grep Chineseand then kill $CHINESE_HANDWRITING_PIDand then toggle the IME once through the Menu Extra menu.)It is a good idea to have a Terminal open during testing, because killingthe task is the only way to dismiss the IME with the keyboard without thisCL.Review URL: http://codereview.chromium.org/6801056 TBR=thakis@chromium.org Review URL: http://codereview.chromium.org/6839001 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@81318 0039d316-1c4b-4281-b951-d872f2087c98
* Mac: Let chrome coexist with the Chinese Handwriting IME.thakis@chromium.org2011-04-126-0/+997
The renderer sandbox breaks the Chinese Handwriting shortcut while Chrome is running, even when Chrome is not active: In a nutshell, Carbon tries to load the IME in every process, and if that fails, it uninstalls the toggle shorcut. It fails in the renderer process due to the sandbox. See the bug for details. Prewarming (patch set 1) doesn't work because that opens up clipboard access to the renderer – the IME doesn't load correctly if it doesn't have clipboard access. dyld __interpose based interposing (patch set 2) doesn't work because TISCreateInputSourceList and its broken caller are in the same framework (HIToolbox). Hence, use binary rewriting to return an arbitrary array with size > 0 to make HIToolbox believe that the IME loads correctly, so that it doesn't uninstall the global handwriting keyboard shortcut. BUG=31225 TEST=Enable Chinese in the input menu. Make sure Chrome's not running. Note that ctrl-shift-space opens and dismisses the Handwriting IME (you need a multitouch trackpad for that). Open Chrome, note that the IME still works. (If you run Chrome without this patch, the IME will be broken until you do ps aux | grep Chinese and then kill $CHINESE_HANDWRITING_PID and then toggle the IME once through the Menu Extra menu.) It is a good idea to have a Terminal open during testing, because killing the task is the only way to dismiss the IME with the keyboard without this CL. Review URL: http://codereview.chromium.org/6801056 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@81308 0039d316-1c4b-4281-b951-d872f2087c98
generated by cgit v1.1 at 2024-09-21 15:22:18 +0000