created | 2019-07-09T07:26:54Z |
---|---|
begin | 2019-06-28T00:00:00Z |
end | 2019-06-29T00:00:00Z |
path | src/sys |
commits | 3 |
date | 2019-06-28T05:51:43Z | |||
---|---|---|---|---|
author | deraadt | |||
files | src/sys/arch/sh/sh/trap.c | log | diff | annotate |
message |
landisk/sh didn't have the code for checking if sp is on MAP_STACK memory, so let's add that where it seems to work gap noticed by miod |
date | 2019-06-28T13:25:09Z | |||
---|---|---|---|---|
author | visa | |||
files | src/sys/kern/vfs_subr.c | log | diff | annotate |
message |
Skip VFS barrier lock during normal operation to reduce overhead. This removes a system-wide serialization point, which might help finding timing-related bugs. OK deraadt@ anton@ |
date | 2019-06-28T21:54:05Z | |||
---|---|---|---|---|
author | bluhm | |||
files | src/sys/arch/amd64/amd64/ucode.c | log | diff | annotate |
src/sys/arch/amd64/include/cpufunc.h | log | diff | annotate | |
src/sys/arch/i386/i386/ucode.c | log | diff | annotate | |
src/sys/arch/i386/include/cpufunc.h | log | diff | annotate | |
message |
Write back and invalidate caches before updating CPU microcode, like Intel does in their patches on githup. Also add a compiler level memory barrier to the wbinvd instruction like Linux does. OK mlarkin@ guenther@ kettenis@ |