Hi all, On Tue, 2 Feb 2021 14:46:24 +0100 Andrey Konovalov wrote: > > On Tue, Feb 2, 2021 at 11:09 AM Anders Roxell wrote: > > > > I've seen this failure on tag next-20210202: > > arch/arm64/kernel/mte.c:121:15: error: 'mte_enable_kernel_sync' > > undeclared here (not in a function); did you mean 'mte_enable_kernel'? > > > > I think it may be a merge conflict that didn't get resolved correctly? > > Yes, that patch was supposed to go on top of another one (which > actually renames mte_enable_kernel to mte_enable_kernel_sync), but the > latter wasn't picked up into mm. > > > The below change fixed the issue: > > > > diff --git a/arch/arm64/kernel/mte.c b/arch/arm64/kernel/mte.c > > index 275b5d0f38b8..8f5bd1293496 100644 > > --- a/arch/arm64/kernel/mte.c > > +++ b/arch/arm64/kernel/mte.c > > @@ -118,7 +118,7 @@ void mte_set_report_once(bool state) > > { > > WRITE_ONCE(report_fault_once, state); > > } > > -EXPORT_SYMBOL(mte_enable_kernel_sync); > > +EXPORT_SYMBOL(mte_enable_kernel); > > EXPORT_SYMBOL(mte_set_report_once); > > > > bool mte_report_once(void) > > The changed export also needs to be moved next to mte_enable_kernel(). > > Thanks! I have applied the following to linux-next today: From: Stephen Rothwell Date: Wed, 3 Feb 2021 07:22:35 +1100 Subject: [PATCH] arm64-kasan-export-mte-symbols-for-kasan-tests-fix Signed-off-by: Stephen Rothwell --- arch/arm64/kernel/mte.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm64/kernel/mte.c b/arch/arm64/kernel/mte.c index 275b5d0f38b8..8c7e32054172 100644 --- a/arch/arm64/kernel/mte.c +++ b/arch/arm64/kernel/mte.c @@ -113,12 +113,12 @@ void mte_enable_kernel(void) sysreg_clear_set(sctlr_el1, SCTLR_ELx_TCF_MASK, SCTLR_ELx_TCF_SYNC); isb(); } +EXPORT_SYMBOL(mte_enable_kernel); void mte_set_report_once(bool state) { WRITE_ONCE(report_fault_once, state); } -EXPORT_SYMBOL(mte_enable_kernel_sync); EXPORT_SYMBOL(mte_set_report_once); bool mte_report_once(void) -- 2.30.0 -- Cheers, Stephen Rothwell