Re: [PATCH v5 5/7] powerpc/mm: Write to PTCR only if ultravisor disabled
From: Michael Ellerman <mpe@ellerman.id.au>
Date: 2019-08-14 12:07:09
Claudio Carvalho [off-list ref] writes:
quoted hunk ↗ jump to hunk
In ultravisor enabled systems, PTCR becomes ultravisor privileged only for writing and an attempt to write to it will cause a Hypervisor Emulation Assitance interrupt. This patch adds the try_set_ptcr(val) macro as an accessor to mtspr(SPRN_PTCR, val), which will be executed only if ultravisor disabled. Signed-off-by: Claudio Carvalho <redacted> --- arch/powerpc/include/asm/reg.h | 13 +++++++++++++ arch/powerpc/mm/book3s64/hash_utils.c | 4 ++-- arch/powerpc/mm/book3s64/pgtable.c | 2 +- arch/powerpc/mm/book3s64/radix_pgtable.c | 6 +++--- 4 files changed, 19 insertions(+), 6 deletions(-)diff --git a/arch/powerpc/include/asm/reg.h b/arch/powerpc/include/asm/reg.h index 10caa145f98b..14139b1ebdb8 100644 --- a/arch/powerpc/include/asm/reg.h +++ b/arch/powerpc/include/asm/reg.h@@ -15,6 +15,7 @@ #include <asm/cputable.h> #include <asm/asm-const.h> #include <asm/feature-fixups.h> +#include <asm/firmware.h>
reg.h is already too big and unwieldy. Can you put this in ultravisor.h and include that in the appropriate places.
quoted hunk ↗ jump to hunk
@@ -1452,6 +1453,18 @@ static inline void update_power8_hid0(unsigned long hid0) */ asm volatile("sync; mtspr %0,%1; isync":: "i"(SPRN_HID0), "r"(hid0)); } + +/* + * In ultravisor enabled systems, PTCR becomes ultravisor privileged only for + * writing and an attempt to write to it will cause a Hypervisor Emulation + * Assistance interrupt. + */ +#define try_set_ptcr(val) \ + do { \ + if (!firmware_has_feature(FW_FEATURE_ULTRAVISOR)) \ + mtspr(SPRN_PTCR, val); \ + } while (0)
This should be a static inline please, not a macro. Sorry, I don't like the name, we're not trying to set it, we know when to set it and when not to. It is awkward to come up with a good name because we don't have a term for "hypervisor that's not running under an ultravisor". Maybe set_ptcr_when_no_uv() Which is kinda messy, someone feel free to come up with something better. I also see some more accesses to the PTCR in arch/powerpc/platforms/powernv/idle.c which you haven't patched? cheers