We added the following: static_assert(offsetof(MemoryChunk, was_used_for_allocation_) == MemoryChunkLayout::kWasUsedForAllocationOffset); static_assert(sizeof(MemoryChunk) == MemoryChunkLayout::kWasUsedForAllocationOffset + sizeof(size_t));
The second assert fails, as if MemoryChunk had an extra field or was padded out to an 8-byte boundary. Probably the latter? On Wednesday, November 30, 2022 at 8:34:36 AM UTC-5 ClearScript Developers wrote: > Hi Dominik, > > Thanks for your reply. Here's our GN config: > > enable_precompiled_headers = false > fatal_linker_warnings = false > is_component_build = false > is_debug = false > target_cpu = "x86" > use_custom_libcxx = false > use_thin_lto = false > v8_enable_pointer_compression = false > v8_enable_31bit_smis_on_64bit_arch = false > v8_monolithic = true > v8_target_cpu = "x86" > v8_use_external_startup_data = false > > Thanks! > On Wednesday, November 30, 2022 at 3:34:19 AM UTC-5 [email protected] > wrote: > >> Hi, >> >> We recently looked into such an issue. This was caused by std::bitset >> (used by MemoryChunk::active_system_pages_) having different sizes in clang >> vs msvc. Are you using msvc by chance? >> >> Cheers, >> Dominik >> >> On Tuesday, November 29, 2022 at 5:35:44 PM UTC+1 [email protected] >> wrote: >> >>> Folks, >>> >>> We're hitting the following in Windows x86 builds of 10.8: >>> >>> ./../..\src/heap/spaces.h(344,35): note: expression evaluates to '152 <= >>> 148' >>> static_assert(sizeof(MemoryChunk) <= MemoryChunk::kHeaderSize); >>> >>> Anyone else? >>> >> -- -- v8-dev mailing list [email protected] http://groups.google.com/group/v8-dev --- You received this message because you are subscribed to the Google Groups "v8-dev" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/v8-dev/36cf50f9-fbf7-4076-9ee0-b4cf5fa2aa44n%40googlegroups.com.
