Re: [PATCH v2 14/14] powerpc: Implement UACCESS validation on PPC32
From: Peter Zijlstra <peterz@infradead.org>
Date: 2023-06-22 11:57:11
Also in:
lkml
On Thu, Jun 22, 2023 at 12:54:36PM +0200, Christophe Leroy wrote:
quoted hunk ↗ jump to hunk
diff --git a/tools/objtool/check.c b/tools/objtool/check.c index f850ab892ad5..8ac5711a055f 100644 --- a/tools/objtool/check.c +++ b/tools/objtool/check.c@@ -218,6 +218,7 @@ static bool __dead_end_function(struct objtool_file *file, struct symbol *func, "kthread_exit", "kunit_try_catch_throw", "lbug_with_loc", + "longjmp", "machine_real_restart", "make_task_dead", "mpt_halt_firmware",@@ -230,7 +231,9 @@ static bool __dead_end_function(struct objtool_file *file, struct symbol *func, "sev_es_terminate", "snp_abort", "start_kernel", + "start_secondary_resume", "stop_this_cpu", + "unrecoverable_exception", "usercopy_abort", "x86_64_start_kernel", "x86_64_start_reservations",
Someone went and changed all that in tip/objtool/core :-) But perhaps, like the uaccess_safe_builtins[] array below, should we start marking sections so we can remember where stuff comes from later?
quoted hunk ↗ jump to hunk
@@ -1335,6 +1338,8 @@ static const char *uaccess_safe_builtin[] = { "rep_stos_alternative", "rep_movs_alternative", "__copy_user_nocache", + "__copy_tofrom_user", + "__arch_clear_user", NULL };
Do we want to rename the 'misc' sectino to 'x86' and start a 'ppc32' section there?