Swiftshader MSVC Break |
||||
Issue description>ninja -C out\Debug obj/third_party/swiftshader/src/Common/swiftshader_common/CPUID.obj ninja: Entering directory `out\Debug' [ 1/1 - 1 processes @ 33.7/s : 0.030s ] CXX obj/third_party/swiftshader/src/Common/swiftshader_common/CPUID.obj FAILED: obj/third_party/swiftshader/src/Common/swiftshader_common/CPUID.obj ninja -t msvc -e environment.x86 -- "d:\src\third_party\depot_tools\win_toolchain\vs_files\1f52d730755ac72dddaf121b73c9d6fd5c24ddf8\vc\tools\msvc\14.11.25503\bin\hostx64\x86/cl.exe" /nologo /showIncludes @obj/third_party/swiftshader/src/Common/swiftshader_common/CPUID.obj.rsp /c ../../third_party/swiftshader/src/Common/CPUID.cpp /Foobj/third_party/swiftshader/src/Common/swiftshader_common/CPUID.obj /Fd"obj/third_party/swiftshader/src/Common/swiftshader_common_cc.pdb" cl : Command line error D8016 : '/O1' and '/RTC1' command-line options are incompatible ninja: build stopped: subcommand failed.
,
Aug 14 2017
,
Aug 14 2017
https://build.chromium.org/p/chromium.win/builders/WinClang64%20(dbg) is happy, where are you seeing this?
,
Aug 14 2017
It's probably a MSVC 2017 thing
,
Aug 14 2017
Nope, I get this in MSVC 2015 too. There must be another config difference somewhere.
,
Aug 14 2017
The reason why this works for Win Clang (dbg) is because of this setting in its GN args: is_debug = false Turns out our debug bots aren't using a debug config. Oops!
,
Aug 14 2017
That's intentional; it's more important to keep the release config buildable since the release builders use that config. But yes, the breakage is visible on https://build.chromium.org/p/chromium.fyi/console?category=win%20clang (which are also all msvc bots) Looks like other stuff is broken too: FAILED: obj/sandbox/win/sbox_integration_tests/lpc_policy_test.obj ninja -t msvc -e environment.x86 -- C:\b\c\goma_client/gomacc.exe "c:\b\c\win_toolchain\vs_files\f53e4598951162bad6330f7a167486c7ae5db1e5\vc\bin\amd64_x86/cl.exe" /nologo /showIncludes @obj/sandbox/win/sbox_integration_tests/lpc_policy_test.obj.rsp /c ../../sandbox/win/src/lpc_policy_test.cc /Foobj/sandbox/win/sbox_integration_tests/lpc_policy_test.obj /Fd"obj/sandbox/win/sbox_integration_tests_cc.pdb" c:\b\c\builder\crwinclang_shared_\src\sandbox\win\src\lpc_policy_test.cc(35) : error C2220: warning treated as error - no 'object' file generated c:\b\c\builder\crwinclang_shared_\src\sandbox\win\src\lpc_policy_test.cc(35) : warning C4702: unreachable code Consider not using msvc? :-)
,
Aug 14 2017
(to be clear, i'll try to get these bots green, but they'll be broken again until I then fix them again, etc)
,
Aug 14 2017
> Consider not using msvc? :-) If the debug info is on par, sure. Until then, this is what we've got. I've already got a fix on the way for swiftshader btw.
,
Aug 14 2017
,
Aug 14 2017
> If the debug info is on par Are there bug filed for whatever shortcomings you see?
,
Aug 14 2017
The following revision refers to this bug: https://swiftshader.googlesource.com/SwiftShader.git/+/e8d42ae19d9c8433a728fb1b1382a3ff0258926c commit e8d42ae19d9c8433a728fb1b1382a3ff0258926c Author: Robert Liao <robliao@chromium.org> Date: Mon Aug 14 19:24:44 2017 Remove /RTC1 for Windows Debug Builds SwiftShader requests the optimized flags for debug buidls for performance. This leads to an incompatible use of /O1 and /RTC1. BUG= chromium:755195 Change-Id: I53b6f99a67d0b8d5fc5932b677ee278aca300702 Reviewed-on: https://swiftshader-review.googlesource.com/11594 Reviewed-by: Alexis Hétu <sugoi@google.com> Reviewed-by: Nicolas Capens <nicolascapens@google.com> Tested-by: Robert Liao <robliao@chromium.org> [modify] https://crrev.com/e8d42ae19d9c8433a728fb1b1382a3ff0258926c/BUILD.gn
,
Aug 14 2017
Change submitted from https://swiftshader-review.googlesource.com/#/c/SwiftShader/+/11594/
,
Aug 14 2017
> Are there bug filed for whatever shortcomings you see? They're already tracked in http://crbug.com/636111 The main problem here is we don't exercise the debugging system to completeness every day. It depends on the problem we're trying to solve. When something is missing, we won't find out until someone hits that case. Clang has some of the basics down, but as we've discovered it's definitely not up to par for debug information. In fact, in some cases, it's debugger hostile (e.g. the hang on http://crbug.com/753032 , causing us to lose control of the process for minutes at a time).
,
Aug 14 2017
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/fe528e14b412bf3dc1b5879a004d870735311ed0 commit fe528e14b412bf3dc1b5879a004d870735311ed0 Author: Nico Weber <thakis@chromium.org> Date: Mon Aug 14 19:51:49 2017 win: Unbreak msvc unofficial release builds a bit. Bug: 755195 Cq-Include-Trybots: master.tryserver.chromium.win:win10_chromium_x64_rel_ng Change-Id: Ie1b24b0ad37a1c4fe82386a39f132f92dbfa8c2a Reviewed-on: https://chromium-review.googlesource.com/614148 Commit-Queue: Nico Weber <thakis@chromium.org> Reviewed-by: Will Harris <wfh@chromium.org> Cr-Commit-Position: refs/heads/master@{#494142} [modify] https://crrev.com/fe528e14b412bf3dc1b5879a004d870735311ed0/sandbox/win/src/lpc_policy_test.cc
,
Aug 14 2017
'all' builds locally for me with your swiftshader fix and my sandbox one, so things should be fine after swiftshader rolls forward. https://build.chromium.org/p/chromium.fyi/console?category=win%20clang (all msvc bots) looks like it's slowly greening up already.
,
Aug 15 2017
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/693c11eab0ea3cdb8b4f14643769a10a2eba827b commit 693c11eab0ea3cdb8b4f14643769a10a2eba827b Author: Alexis Hetu <sugoi@google.com> Date: Tue Aug 15 05:28:37 2017 Roll SwiftShader 5bf72ee..e8d42ae https://swiftshader.googlesource.com/SwiftShader.git/+log/5bf72ee..e8d42ae BUG= chromium:755195 TBR=kbr@chromium.org TEST=bots CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.win:win_optional_gpu_tests_rel;master.tryserver.chromium.mac:mac_optional_gpu_tests_rel;master.tryserver.chromium.linux:linux_optional_gpu_tests_rel,linux_chromium_cfi_rel_ng;master.tryserver.chromium.android:android_optional_gpu_tests_rel Change-Id: Ifffea9273cf79c8048cbf95003f7eb29b98a17ef Reviewed-on: https://chromium-review.googlesource.com/615027 Reviewed-by: Alexis Hétu <sugoi@chromium.org> Commit-Queue: Alexis Hétu <sugoi@chromium.org> Cr-Commit-Position: refs/heads/master@{#494328} [modify] https://crrev.com/693c11eab0ea3cdb8b4f14643769a10a2eba827b/DEPS
,
Aug 17 2017
|
||||
►
Sign in to add a comment |
||||
Comment 1 by robliao@chromium.org
, Aug 14 2017