Re: [PATCH v2 2/3] target/ppc: ppc_store_fpscr doesn't update bit 52

[Richard Henderson]

On 11/18/21 2:25 PM, Lucas Mateus Castro (alqotel) wrote:

+/* FPSCR bits that can be set by mtfsf, mtfsfi and mtfsb1 */
+#define FPSCR_MTFS_MASK (~((1ull << 11) | FP_VX | FP_FEX))


If you're going to make the reserved bit 52 read-as-zero-writes-ignored, you should do the 
same for reserved bits 0-31.  Otherwise drop this and let the bits be read-write with no 
effect.



r~

Re: [PATCH v2 2/3] target/ppc: ppc_store_fpscr doesn't update bit 52

[BALATON Zoltan]

On Thu, 18 Nov 2021, Lucas Mateus Castro (alqotel) wrote:

This commit fixes the difference reported in the bug in the reserved
bit 52, it does this by adding this bit to the mask of bits to not be
directly altered in the ppc_store_fpscr function (the hardware used to
compare to QEMU was a Power9).

Although this is a difference reported in the bug, since it's a reserved
bit it may be a "don't care" case, as put in the bug report. Looking at
the ISA it doesn't explicitly mentions this bit can't be set, like it
does for FEX and VX, so I'm unsure if this is necessary.

Resolves: https://gitlab.com/qemu-project/qemu/-/issues/266
Signed-off-by: Lucas Mateus Castro (alqotel) 
---
target/ppc/cpu.c | 2 +-
target/ppc/cpu.h | 3 +++
2 files changed, 4 insertions(+), 1 deletion(-)

diff --git a/target/ppc/cpu.c b/target/ppc/cpu.c
index f933d9f2bd..d7b42bae52 100644
--- a/target/ppc/cpu.c
+++ b/target/ppc/cpu.c
@@ -112,7 +112,7 @@ static inline void fpscr_set_rounding_mode(CPUPPCState *env)

void ppc_store_fpscr(CPUPPCState *env, target_ulong val)
{
-val &= ~(FP_VX | FP_FEX);
+val &= FPSCR_MTFS_MASK;
if (val & FPSCR_IX) {
val |= FP_VX;
}
diff --git a/target/ppc/cpu.h b/target/ppc/cpu.h
index e946da5f3a..53463092ab 100644
--- a/target/ppc/cpu.h
+++ b/target/ppc/cpu.h
@@ -759,6 +759,9 @@ enum {
  FP_VXZDZ  | FP_VXIMZ  | FP_VXVC   | FP_VXSOFT | \
  FP_VXSQRT | FP_VXCVI)

+/* FPSCR bits that can be set by mtfsf, mtfsfi and mtfsb1 */
+#define FPSCR_MTFS_MASK (~((1ull << 11) | FP_VX | FP_FEX))


Instead of (1ull << 11) maybe PPC_BIT(52) is a bit clearer (or not 
depending on personal preference, so I don't mind either way.


Regards,
BALATON Zoltan


+
/*/
/* Vector status and control register */
#define VSCR_NJ 16 /* Vector non-java */