summaryrefslogtreecommitdiffstats
path: root/sandbox
Commit message (Collapse)AuthorAgeFilesLines
* sandbox: Fix RedirectToUserSpacePolicyWrapperleecam2014-09-033-24/+112
| | | | | | | | | | | | | This fixes RedirectToUserSpacePolicyWrapper to handle ERRNO ErrorCodes that are behind Conditional ErrorCodes. BUG=408953 TEST=sandbox_linux_unittests: Added new test Review URL: https://codereview.chromium.org/524603002 Cr-Commit-Position: refs/heads/master@{#293116}
* Revert of Revert of clang/win: Fix a few warnings in targets not in ↵thakis2014-08-311-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | chromium_builder_tests. (patchset #1 id:1 of https://codereview.chromium.org/519463003/) Reason for revert: Reland, wasn't the reason for the redness. (https://codereview.chromium.org/521703002/ was.) Original issue's description: > Revert of clang/win: Fix a few warnings in targets not in chromium_builder_tests. (patchset #3 id:40001 of https://codereview.chromium.org/526513002/) > > Reason for revert: > Speculative, might have broken InstallerStateTest.InitializeTwice on XP: http://build.chromium.org/p/chromium.win/builders/XP%20Tests%20%281%29/builds/32481 > > InstallerStateTest.InitializeTwice (run #1): > [ RUN ] InstallerStateTest.InitializeTwice > c:\b\build\slave\win_builder\build\src\chrome\installer\util\installer_state_unittest.cc(639): error: Value of: wcsstr(installer_state.target_path().value().c_str(), BrowserDistribution::GetSpecificDistribution( BrowserDistribution::CHROME_BINARIES)-> GetInstallSubDir().c_str()) > Actual: false > Expected: true > [ FAILED ] InstallerStateTest.InitializeTwice (219 ms) > > InstallerStateTest.InitializeTwice (run #2): > [ RUN ] InstallerStateTest.InitializeTwice > c:\b\build\slave\win_builder\build\src\chrome\installer\util\installer_state_unittest.cc(639): error: Value of: wcsstr(installer_state.target_path().value().c_str(), BrowserDistribution::GetSpecificDistribution( BrowserDistribution::CHROME_BINARIES)-> GetInstallSubDir().c_str()) > Actual: false > Expected: true > [ FAILED ] InstallerStateTest.InitializeTwice (109 ms) > > > Original issue's description: > > clang/win: Fix a few warnings in targets not in chromium_builder_tests. > > > > Also don't use "default" as a variable name, as it's a keyword. > > Also fix a bug where a wstring was passed to %ls. > > > > No real behavior change. > > > > BUG=82385 > > R=hans@chromium.org > > TBR=cpu, vitalybuka > > > > Committed to pending queue: https://chromium.googlesource.com/chromium/src/+/d7efa09 > > TBR=hans@chromium.org,robertshield@chromium.org,gab@chromium.org,vitalybuka@chromium.org,cpu@chromium.org > NOTREECHECKS=true > NOTRY=true > BUG=82385 > > Committed: https://chromium.googlesource.com/chromium/src/+/dae20dafcfd3ccc439ccc5b3d5e21ad611d13c23 TBR=hans@chromium.org,robertshield@chromium.org,gab@chromium.org,vitalybuka@chromium.org,cpu@chromium.org NOTREECHECKS=true NOTRY=true BUG=82385 Review URL: https://codereview.chromium.org/525013003 Cr-Commit-Position: refs/heads/master@{#292818}
* Revert of clang/win: Fix a few warnings in targets not in ↵thakis2014-08-301-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | chromium_builder_tests. (patchset #3 id:40001 of https://codereview.chromium.org/526513002/) Reason for revert: Speculative, might have broken InstallerStateTest.InitializeTwice on XP: http://build.chromium.org/p/chromium.win/builders/XP%20Tests%20%281%29/builds/32481 InstallerStateTest.InitializeTwice (run #1): [ RUN ] InstallerStateTest.InitializeTwice c:\b\build\slave\win_builder\build\src\chrome\installer\util\installer_state_unittest.cc(639): error: Value of: wcsstr(installer_state.target_path().value().c_str(), BrowserDistribution::GetSpecificDistribution( BrowserDistribution::CHROME_BINARIES)-> GetInstallSubDir().c_str()) Actual: false Expected: true [ FAILED ] InstallerStateTest.InitializeTwice (219 ms) InstallerStateTest.InitializeTwice (run #2): [ RUN ] InstallerStateTest.InitializeTwice c:\b\build\slave\win_builder\build\src\chrome\installer\util\installer_state_unittest.cc(639): error: Value of: wcsstr(installer_state.target_path().value().c_str(), BrowserDistribution::GetSpecificDistribution( BrowserDistribution::CHROME_BINARIES)-> GetInstallSubDir().c_str()) Actual: false Expected: true [ FAILED ] InstallerStateTest.InitializeTwice (109 ms) Original issue's description: > clang/win: Fix a few warnings in targets not in chromium_builder_tests. > > Also don't use "default" as a variable name, as it's a keyword. > Also fix a bug where a wstring was passed to %ls. > > No real behavior change. > > BUG=82385 > R=hans@chromium.org > TBR=cpu, vitalybuka > > Committed to pending queue: https://chromium.googlesource.com/chromium/src/+/d7efa09 TBR=hans@chromium.org,robertshield@chromium.org,gab@chromium.org,vitalybuka@chromium.org,cpu@chromium.org NOTREECHECKS=true NOTRY=true BUG=82385 Review URL: https://codereview.chromium.org/519463003 Cr-Commit-Position: refs/heads/master@{#292809}
* clang/win: Fix a few warnings in targets not in chromium_builder_tests.Nico Weber2014-08-291-1/+1
| | | | | | | | | | | | | | | Also don't use "default" as a variable name, as it's a keyword. Also fix a bug where a wstring was passed to %ls. No real behavior change. BUG=82385 R=hans@chromium.org TBR=cpu, vitalybuka Review URL: https://codereview.chromium.org/526513002 Cr-Commit-Position: refs/heads/master@{#292699}
* sandbox: style cleanupMatthew Dempsky2014-08-293-52/+96
| | | | | | | | | | Based on readability review by Dean Berris at Google. R=jln@chromium.org Review URL: https://codereview.chromium.org/511993005 Cr-Commit-Position: refs/heads/master@{#292698}
* Remove cevans@ from OWNERS files.thestig2014-08-291-1/+0
| | | | | | | | NOTRY=true Review URL: https://codereview.chromium.org/524433003 Cr-Commit-Position: refs/heads/master@{#292549}
* Revert of Disable slow ↵rsesek2014-08-271-2/+1
| | | | | | | | | | | | | | | | | | | | | | | BootstrapSandboxTest.DistributedNotifications_SandboxDeny (patchset #1 of https://codereview.chromium.org/498193002/) Reason for revert: Unclear reason for disable. Original issue's description: > Disable slow BootstrapSandboxTest.DistributedNotifications_SandboxDeny > > TBR=rsesek > > BUG=407066 > > Committed to pending queue: https://chromium.googlesource.com/chromium/src/+/f1bd7f7 TBR=phajdan.jr@chromium.org BUG=407066 Review URL: https://codereview.chromium.org/508873003 Cr-Commit-Position: refs/heads/master@{#292207}
* Remove implicit conversions from scoped_refptr to T* in sandbox/dcheng2014-08-272-2/+2
| | | | | | | | | | | This patch was generated by running the rewrite_scoped_refptr clang tool on a Linux build. BUG=110610 Review URL: https://codereview.chromium.org/467183005 Cr-Commit-Position: refs/heads/master@{#292052}
* Disable SandboxBPF.StartSingleThreadedAsMultiThreaded under TSan v2.jyasskin2014-08-271-0/+3
| | | | | | | | | | BUG=407357 TBR=rsesek@chromium.org NOTRY=true Review URL: https://codereview.chromium.org/510473003 Cr-Commit-Position: refs/heads/master@{#292042}
* sandbox: Add support for the new seccomp() system call in kernel 3.17.rsesek2014-08-254-28/+160
| | | | | | | | | | | | | This new system call can be used to set NO_NEW_PRIVS, install a BPF filter, and synchronize that filter across the thread group. BUG=308763,308775 TEST=New SandboxBPF.Tsync test where kernel support is available. R=jln@chromium.org Review URL: https://codereview.chromium.org/494743003 Cr-Commit-Position: refs/heads/master@{#291739}
* Disable slow BootstrapSandboxTest.DistributedNotifications_SandboxDenyPaweł Hajdan, Jr2014-08-251-1/+2
| | | | | | | | | | TBR=rsesek BUG=407066 Review URL: https://codereview.chromium.org/498193002 Cr-Commit-Position: refs/heads/master@{#291669}
* sandbox: Add Arm64 support for seccomp-BPFleecam2014-08-2414-70/+1342
| | | | | | | | | | | | Adds support for Arm64 to 'sandbox/'. This can be used by future CLs to provide Arm64 policy for Chrome. BUG=355125 TEST=sandbox_linux_unittests Review URL: https://codereview.chromium.org/487143003 Cr-Commit-Position: refs/heads/master@{#291631}
* [MIPS] Fix build issue for MIPSpetarj@mips.com2014-08-221-1/+1
| | | | | | | | | | Use double underscore in the preprocessor symbol. Fixes build breakage caused by r291315. Review URL: https://codereview.chromium.org/495333003 Cr-Commit-Position: refs/heads/master@{#291452} git-svn-id: svn://svn.chromium.org/chrome/trunk/src@291452 0039d316-1c4b-4281-b951-d872f2087c98
* sandbox: Update known syscalls to kernel 3.17leecam@chromium.org2014-08-227-2/+127
| | | | | | | | | | | | | Updated Arm, Mips, x86_32 and x86_64 syscall defines to match the latest kernel 3.17 at the time of this commit. BUG=406123 TEST=sandbox_linux_unittests Review URL: https://codereview.chromium.org/493603003 Cr-Commit-Position: refs/heads/master@{#291315} git-svn-id: svn://svn.chromium.org/chrome/trunk/src@291315 0039d316-1c4b-4281-b951-d872f2087c98
* sandbox: Fix Android clone flagsleecam@chromium.org2014-08-211-2/+4
| | | | | | | | | | | | | | Android's bionic now uses the same clone flags in pthread_create as glibc NOTRY=true BUG=401555 TEST=sandbox_linux_unittests Review URL: https://codereview.chromium.org/484063004 Cr-Commit-Position: refs/heads/master@{#291040} git-svn-id: svn://svn.chromium.org/chrome/trunk/src@291040 0039d316-1c4b-4281-b951-d872f2087c98
* Don't pass const strings to CreateProcess.thakis@chromium.org2014-08-202-11/+19
| | | | | | | | | | | | | | The documentation says that CreateProcess() can modify its second parameter in UNICODE builds. BUG=396705 R=scottmg@chromium.org, vitalybuka@chromium.org TBR=cpu Review URL: https://codereview.chromium.org/487303004 Cr-Commit-Position: refs/heads/master@{#290890} git-svn-id: svn://svn.chromium.org/chrome/trunk/src@290890 0039d316-1c4b-4281-b951-d872f2087c98
* Replace NT prefix in sandbox rules match string to handle correct wildcard ↵forshaw@chromium.org2014-08-163-6/+55
| | | | | | | | | | | | | | | | escaping This patch adds a function to modify file system sandbox rules to replace the \??\ NT prefix with the correct escaped form \/?/?\ for the wildcard matching rules in the broker. This is done generally as it's a common mistake in the sandbox code and so provides some defence in depth. BUG=334882 Review URL: https://codereview.chromium.org/432543005 Cr-Commit-Position: refs/heads/master@{#290131} git-svn-id: svn://svn.chromium.org/chrome/trunk/src@290131 0039d316-1c4b-4281-b951-d872f2087c98
* Resolve registry root paths in HandleCloserjschuh@chromium.org2014-08-151-2/+10
| | | | | | | | | BUG=349162 Review URL: https://codereview.chromium.org/430753002 Cr-Commit-Position: refs/heads/master@{#289764} git-svn-id: svn://svn.chromium.org/chrome/trunk/src@289764 0039d316-1c4b-4281-b951-d872f2087c98
* [MIPS] Add ABI check for syscall numbers definitionsnedeljko.babic@imgtec.com2014-08-081-1/+1
| | | | | | | | | | | | | | In file mips_linux_syscalls.h are definitions of syscall numbers for O32 ABI, so this check is needed in order for Mips architectures with other ABIs to work properly. BUG=400684 TEST=compile sandbox_linux_unittest for MIPS32 and MIPS64 Review URL: https://codereview.chromium.org/446213003 Cr-Commit-Position: refs/heads/master@{#288252} git-svn-id: svn://svn.chromium.org/chrome/trunk/src@288252 0039d316-1c4b-4281-b951-d872f2087c98
* Replace "using namespace" statement with individual "using" statementsmdempsky@chromium.org2014-08-051-1/+1
| | | | | | | | BUG=375497 Review URL: https://codereview.chromium.org/440703002 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@287460 0039d316-1c4b-4281-b951-d872f2087c98
* Revert "Make the sandbox code link with the 10.7 SDK."avi@chromium.org2014-08-013-11/+11
| | | | | | | | | | | | | | This reverts commit r287000. This turns out to not help, and the proper fix is insanely complicated. Also fix some broken declarations. BUG=399414 TBR=rsesek@chromium.org NOTRY=true Review URL: https://codereview.chromium.org/439493002 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@287034 0039d316-1c4b-4281-b951-d872f2087c98
* Make the sandbox code link with the 10.7 SDK.avi@chromium.org2014-08-012-9/+9
| | | | | | | | | | BUG=399414 TEST=Mac 64 bot goes green NOTRY=true Review URL: https://codereview.chromium.org/436713003 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@287000 0039d316-1c4b-4281-b951-d872f2087c98
* Annotate IA-32's SyscallAsm implementation with CFI directivesmdempsky@chromium.org2014-08-012-11/+8
| | | | | | | | | | | | | | | This is necessary so that glibc's backtrace() routine can properly unwind the stack when called within a Trap handler triggered by Syscall::Call(). Otherwise the .eh_frame data will mislead backtrace() into thinking %ebp is still valid, when we've actually clobbered it with a syscall parameter. This allows us to also enable the InvalidSyscall BPF tests on IA-32. BUG=399396 Review URL: https://codereview.chromium.org/430353002 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@286959 0039d316-1c4b-4281-b951-d872f2087c98
* Convert content's BPF policies to use BPF DSLmdempsky@chromium.org2014-08-017-206/+168
| | | | | | | | BUG=375497 Review URL: https://codereview.chromium.org/410063002 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@286929 0039d316-1c4b-4281-b951-d872f2087c98
* Linux sandbox: more verbose test failuresjln@chromium.org2014-07-311-2/+4
| | | | | | | | | | | Restore printing the stderr of the dead process in death tests even if the exit code is unexpected. R=mdempsky@chromium.org Review URL: https://codereview.chromium.org/438703002 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@286891 0039d316-1c4b-4281-b951-d872f2087c98
* Cleanup SandboxBPF policies now that we have InvalidSyscall()mdempsky@chromium.org2014-07-316-1/+42
| | | | | | | | | | In particular, restore previous behavior where most policies would crash on invalid system call numbers with SIGSYS rather than returning ENOSYS. Review URL: https://codereview.chromium.org/294643005 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@286890 0039d316-1c4b-4281-b951-d872f2087c98
* Sandbox: Fix the BrokerProcess RecvMsgDescriptorLeak testleecam@chromium.org2014-07-311-0/+5
| | | | | | | | | | | | | RecvMsgDescriptorLeak was failing on Android due to LOG() openning a socket on first use. This extra file descriptor screwed with the test. So we LOG before the test. BUG=none TEST=sandbox_linux_unittests Review URL: https://codereview.chromium.org/434653002 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@286881 0039d316-1c4b-4281-b951-d872f2087c98
* Fix line endings on BUILD.gnwfh@chromium.org2014-07-311-285/+285
| | | | git-svn-id: svn://svn.chromium.org/chrome/trunk/src@286813 0039d316-1c4b-4281-b951-d872f2087c98
* Android sandbox: EPERM fork() properlyjln@chromium.org2014-07-312-1/+9
| | | | | | | | | | | | | | | On Android, there seems to be fallback code to the fork() system call if clone() is EPERM-ed. Instead of crashing on the fork system call, we now simply EPERM it. Moreover, we allow prctl(PR_GET_NAME,...), which is used by Android and should not affect security. BUG=398611 Review URL: https://codereview.chromium.org/429223003 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@286781 0039d316-1c4b-4281-b951-d872f2087c98
* Android sandbox: fix most BaselinePolicy testsjln@chromium.org2014-07-304-14/+67
| | | | | | | | | | | | | | | | | | | | On Android, the default signal handler is used and on SIGSEGV the process will be terminated by the signal. On Linux, the SIGSEGV signal handler terminates the process with _exit(1). Unit tests that tested for SIGSEGV implicitly relied on the signal handler terminating the process with _exit(1). We create a new type of "death check" that specifically checks that the process was terminated due to a SIGSEGV. However, on Linux this is still not very reliable. BUG=398611 R=mdempsky@chromium.org Review URL: https://codereview.chromium.org/419683013 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@286576 0039d316-1c4b-4281-b951-d872f2087c98
* Android seccomp-bpf: fix a few unittestsjln@chromium.org2014-07-303-22/+26
| | | | | | | | | | | | | | | | | | This CL fixes the following tests: - SandboxBPF.UseOpenBroker - SandboxBPF.ErrnoTest - SandboxBPF.SimpleCondTest The cause of failure was assuming that dup(), open() and access() are always implemented with specific system calls (https://codereview.chromium.org/226923003). This CL also introduces two trivial sanity checks tests of our testing framework. BUG=398611 Review URL: https://codereview.chromium.org/424253002 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@286444 0039d316-1c4b-4281-b951-d872f2087c98
* [MIPS] Add seccomp bpf supportnedeljko.babic@imgtec.com2014-07-2520-75/+1909
| | | | | | | | | | | | | | | | | Add support for seccomp bpf sandboxing on MIPS architecture. Enable testing of seccomp bpf sandbox. Support for seccomp bpf for MIPS was added in Linux kernel version 3.15. This patchset was reviewed in https://chromiumcodereview.appspot.com/260793003/ and re-created to workaround a commit-queue issue. BUG=369594 TEST=sandbox_linux_unittests Review URL: https://codereview.chromium.org/409403003 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@285538 0039d316-1c4b-4281-b951-d872f2087c98
* Make HandleCloserAgent walk in 4x multiples on x64jschuh@chromium.org2014-07-242-2/+2
| | | | | | | | | | | | | Unfortunately, there's no non-flakey way to test this, because we can't control handle allocation. BUG=397130 R=wfh@chromium.org NOTRY=true Review URL: https://codereview.chromium.org/419533003 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@285397 0039d316-1c4b-4281-b951-d872f2087c98
* Patch NtOpenKeyEx on Windows Server 2008jschuh@chromium.org2014-07-241-1/+4
| | | | | | | | | | R=wfh@chromium.org BUG=51909 TEST=Verify that chrome still runs on Windows Server 2008 Review URL: https://codereview.chromium.org/417923006 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@285352 0039d316-1c4b-4281-b951-d872f2087c98
* GN: Add //ui/base/ime, fix up printing on macjamesr@chromium.org2014-07-231-0/+2
| | | | | | | | | | | | | | | | | This adds the //ui/base/ime target to the GN build, which is used everywhere except for ios. In the gyp build this isn't a real target, it's a set of properties that are included into the ui_base target which is the best you can do in GYP since you can't link a static library into a component without dropping some symbols. In GN, we can define this as a source_set with its own dependencies and settings. This also fixes up some miscellaneous mac issues. R=brettw@chromium.org,ben@chromium.org Review URL: https://codereview.chromium.org/389083002 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@284981 0039d316-1c4b-4281-b951-d872f2087c98
* Fix the //sandbox/mac build when using a modern (10.7+) SDK.rsesek@chromium.org2014-07-214-40/+61
| | | | | | | | | | | | This doesn't use the dynamic library loader if the SDK has libxpc.dylib, and instead just relies on libSystem to load it. BUG=395364 R=mark@chromium.org Review URL: https://codereview.chromium.org/403193002 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@284427 0039d316-1c4b-4281-b951-d872f2087c98
* Run XPCMessageServerTest on 10.8+ rather than 10.7+.rsesek@chromium.org2014-07-191-1/+1
| | | | | | | | | BUG=382931 TBR=mark@chromium.org Review URL: https://codereview.chromium.org/399643008 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@284303 0039d316-1c4b-4281-b951-d872f2087c98
* Add an XPC implementation of sandbox::MessageServer.rsesek@chromium.org2014-07-1910-1/+446
| | | | | | | | | BUG=382931 R=mark@chromium.org Review URL: https://codereview.chromium.org/404893002 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@284288 0039d316-1c4b-4281-b951-d872f2087c98
* Add domain-specific language for BPF policiesmdempsky@chromium.org2014-07-179-0/+885
| | | | | | | | | | | | | This CL adds basic support for equality testing of system call arguments, and conjunctive and disjunctive combinations of tests. Reland of https://codereview.chromium.org/299743002/ BUG=375497 Review URL: https://codereview.chromium.org/396323002 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@283687 0039d316-1c4b-4281-b951-d872f2087c98
* Create DispatchSourceMach to run a MACH_RECV dispatch source.rsesek@chromium.org2014-07-177-33/+258
| | | | | | | | | | | | | This implements RAII semantics so that when the destructor runs, the source is cancelled and event handler blocks cannot call methods on a destructed object. BUG=382931 R=mark@chromium.org Review URL: https://codereview.chromium.org/392273002 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@283664 0039d316-1c4b-4281-b951-d872f2087c98
* Revert of Add domain-specific language for BPF policies ↵tapted@chromium.org2014-07-169-865/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | (https://codereview.chromium.org/299743002/) Reason for revert: Suspected for sandbox_linux_unittests failures on Linux Tests (dbg)(2)(32) at http://build.chromium.org/p/chromium.linux/builders/Linux%20Tests%20%28dbg%29%282%29%2832%29/builds/14560 errors in BPFDSL.MoreBooleanLogic Actual test failure: ../../sandbox/linux/bpf_dsl/bpf_dsl_unittest.cc:117:(11) == ((*__errno_location ())) BPFDSL.MaskTest Actual test failure: ../../sandbox/linux/bpf_dsl/bpf_dsl_unittest.cc:206:(expect_errno) == ((*__errno_location ())) BPFDSL.ElseIfTest Actual test failure: ../../sandbox/linux/bpf_dsl/bpf_dsl_unittest.cc:235:(0) == (setuid(0)) output like Value of: subprocess_exit_status Actual: 1 Expected: kExpectedValue Which is: 42 Original issue's description: > Add domain-specific language for BPF policies > > This CL adds basic support for equality testing of system call > arguments, and conjunctive and disjunctive combinations of tests. > > BUG=375497 > > Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=283350 TBR=mdempsky@chromium.org NOTREECHECKS=true NOTRY=true Review URL: https://codereview.chromium.org/391043003 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@283377 0039d316-1c4b-4281-b951-d872f2087c98
* Do not create a reply IPCMessage for every message received.rsesek@chromium.org2014-07-165-35/+46
| | | | | | | | | | | | | | This is incompatible with the way xpc_dictionary_create_reply() works, which may only be called once per message. This would inhibit forwarding messages, since the receiver of the forwarded XPC message would not be able to reply. Instead, replies now must be created explicitly via MessageServer::CreateReply(). BUG=382931 R=mark@chromium.org Review URL: https://codereview.chromium.org/398563002 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@283376 0039d316-1c4b-4281-b951-d872f2087c98
* Add domain-specific language for BPF policiesmdempsky@chromium.org2014-07-169-0/+865
| | | | | | | | | | | This CL adds basic support for equality testing of system call arguments, and conjunctive and disjunctive combinations of tests. BUG=375497 Review URL: https://codereview.chromium.org/299743002 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@283350 0039d316-1c4b-4281-b951-d872f2087c98
* Fixes for re-enabling more MSVC level 4 warnings: sandbox/ editionpkasting@chromium.org2014-07-109-58/+47
| | | | | | | | | | | | | | | | | This contains fixes for the following sorts of issues: * Assignment inside conditional * Possibly-uninitialized local variable * Signedness mismatch This also contains a small number of other cleanups/simplifications to nearby code. BUG=81439 TEST=none Review URL: https://codereview.chromium.org/382613002 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@282451 0039d316-1c4b-4281-b951-d872f2087c98
* Rename source_prereqs to inputs in the GN build.brettw@chromium.org2014-07-091-1/+1
| | | | | | | | | | This variable changed name. R=jamesr@chromium.org Review URL: https://codereview.chromium.org/375873006 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@282076 0039d316-1c4b-4281-b951-d872f2087c98
* Linux sandbox: add space for 8 parameters to the Syscall() classnedeljko.babic@imgtec.com2014-07-082-23/+80
| | | | | | | | | | | | | | On some architectures (Mips for example) syscalls can take more than six parameters. Add support for 8 native-size parameters in Syscall::Call() BUG=369594 TEST= sandbox_linux_unittests Review URL: https://codereview.chromium.org/357323003 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@281823 0039d316-1c4b-4281-b951-d872f2087c98
* Add mdempsky as OWNER for sandbox/linuxmdempsky@chromium.org2014-07-081-0/+1
| | | | | | | | R=jschuh@chromium.org Review URL: https://codereview.chromium.org/375873004 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@281779 0039d316-1c4b-4281-b951-d872f2087c98
* Revert of Revert of Disable yama on 3.2 kernels with 32bit userland and ↵machenbach@chromium.org2014-07-081-0/+17
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 64bit kernel. (https://codereview.chromium.org/374933002/) Reason for revert: [Sheriff] This was fixed already in 281762. Original issue's description: > Revert of Disable yama on 3.2 kernels with 32bit userland and 64bit kernel. (https://codereview.chromium.org/371113003/) > > Reason for revert: > Broke clang builders: > https://build.chromium.org/p/chromium.linux/builders/Linux%20Clang%20(dbg)/builds/61819/steps/compile/logs/stdio > > Original issue's description: > > Disable two yama tests on 3.2 kernels with 32bit userland and 64bit kernel. > > > > Disabling yama restrictions is broken there. > > > > BUG=391916 > > R=jorgelo@chromium.org > > > > Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=281666 > > > > Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=281757 > > TBR=jorgelo@chromium.org,keescook@chromium.org,thakis@chromium.org > NOTREECHECKS=true > NOTRY=true > BUG=391916 > > Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=281763 TBR=jorgelo@chromium.org,keescook@chromium.org,thakis@chromium.org,eseidel@chromium.org NOTREECHECKS=true NOTRY=true BUG=391916 Review URL: https://codereview.chromium.org/374943002 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@281764 0039d316-1c4b-4281-b951-d872f2087c98
* Revert of Disable yama on 3.2 kernels with 32bit userland and 64bit kernel. ↵eseidel@chromium.org2014-07-081-17/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | (https://codereview.chromium.org/371113003/) Reason for revert: Broke clang builders: https://build.chromium.org/p/chromium.linux/builders/Linux%20Clang%20(dbg)/builds/61819/steps/compile/logs/stdio Original issue's description: > Disable two yama tests on 3.2 kernels with 32bit userland and 64bit kernel. > > Disabling yama restrictions is broken there. > > BUG=391916 > R=jorgelo@chromium.org > > Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=281666 > > Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=281757 TBR=jorgelo@chromium.org,keescook@chromium.org,thakis@chromium.org NOTREECHECKS=true NOTRY=true BUG=391916 Review URL: https://codereview.chromium.org/374933002 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@281763 0039d316-1c4b-4281-b951-d872f2087c98
* build fix (see r281666 and r281672, I screwed up the re-land)thakis@chromium.org2014-07-081-1/+4
| | | | | | | | TBR=maniscalo@chromium.org Review URL: https://codereview.chromium.org/375913002 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@281762 0039d316-1c4b-4281-b951-d872f2087c98
generated by cgit v1.1 at 2025-01-31 05:53:20 +0000