On 6/23/21 6:57 AM, Paul Gofman wrote:
Fixes the crash in test.
Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=51152 Signed-off-by: Paul Gofman pgofman@codeweavers.com
dlls/ntdll/tests/exception.c | 14 ++++++++++++++ 1 file changed, 14 insertions(+)
diff --git a/dlls/ntdll/tests/exception.c b/dlls/ntdll/tests/exception.c index 328882c1a23..7b83c06a8db 100644 --- a/dlls/ntdll/tests/exception.c +++ b/dlls/ntdll/tests/exception.c @@ -2908,13 +2908,27 @@ static const struct exception 0x31, 0xc0, /* xor %eax,%eax */ 0x8e, 0xc0, /* mov %eax,%es */ 0x8e, 0xd8, /* mov %eax,%ds */ +#if 0
/* TODO_WINE: fs/gs base is reset on Linux when setting thesegment value and this is not currently handled.Disable this part to avoid crashing the test. */ 0x8e, 0xe0, /* mov %eax,%fs */ 0x8e, 0xe8, /* mov %eax,%gs */+#else
0x90, 0x90, /* nop */0x90, 0x90, /* nop */+#endif 0xfa, /* cli */ 0x58, /* pop %rax */ +#if 0 0x8e, 0xe8, /* mov %eax,%gs */ 0x58, /* pop %rax */ 0x8e, 0xe0, /* mov %eax,%fs */ +#else
0x58, /* pop %rax */0x90, 0x90, /* nop */0x90, 0x90, /* nop */+#endif 0x58, /* pop %rax */ 0x8e, 0xd8, /* mov %eax,%ds */ 0x58, /* pop %rax */
I don't object to disabling these tests, but I don't think the comment is accurate. We've been able to observe fsbase being reset to zero on one AMD processor, with Linux 5.12, but not another; the behaviour violates the AMD specification, and it's not clear that the kernel is actually getting involved.