Re: [PATCH] target/arm: Restructure has_vfp_d32 test

2023-06-23 Thread Peter Maydell
On Mon, 19 Jun 2023 at 15:02, Richard Henderson wrote: > > One cannot test for feature aa32_simd_r32 without first > testing if AArch32 mode is supported at all. This leads to > > qemu-system-aarch64: ARM CPUs must have both VFP-D32 and Neon or neither > > for Apple M1 cpus. > > We already have

Re: [PATCH] target/arm: Restructure has_vfp_d32 test

2023-06-20 Thread Richard Henderson
On 6/20/23 07:29, Philippe Mathieu-Daudé wrote: On 19/6/23 16:02, Richard Henderson wrote: One cannot test for feature aa32_simd_r32 without first testing if AArch32 mode is supported at all.  This leads to qemu-system-aarch64: ARM CPUs must have both VFP-D32 and Neon or neither for Apple M1

Re: [PATCH] target/arm: Restructure has_vfp_d32 test

2023-06-19 Thread Philippe Mathieu-Daudé
On 19/6/23 16:02, Richard Henderson wrote: One cannot test for feature aa32_simd_r32 without first testing if AArch32 mode is supported at all. This leads to qemu-system-aarch64: ARM CPUs must have both VFP-D32 and Neon or neither for Apple M1 cpus. We already have a check for ARMv8-A never

Re: [PATCH] target/arm: Restructure has_vfp_d32 test

2023-06-19 Thread Mads Ynddal
> On 19 Jun 2023, at 16.02, Richard Henderson > wrote: > > One cannot test for feature aa32_simd_r32 without first > testing if AArch32 mode is supported at all. This leads to > > qemu-system-aarch64: ARM CPUs must have both VFP-D32 and Neon or neither > > for Apple M1 cpus. > > We

Re: [PATCH] target/arm: Restructure has_vfp_d32 test

2023-06-19 Thread Cédric Le Goater
On 6/19/23 16:02, Richard Henderson wrote: One cannot test for feature aa32_simd_r32 without first testing if AArch32 mode is supported at all. This leads to qemu-system-aarch64: ARM CPUs must have both VFP-D32 and Neon or neither for Apple M1 cpus. The orangepi5 board I use didn't have

Re: [PATCH] target/arm: Restructure has_vfp_d32 test

2023-06-19 Thread Richard Henderson
On 6/19/23 18:07, Philippe Mathieu-Daudé wrote: We already have a check for ARMv8-A never setting vfp-d32 true, ... gah! "false". r~

Re: [PATCH] target/arm: Restructure has_vfp_d32 test

2023-06-19 Thread Philippe Mathieu-Daudé
On 19/6/23 16:02, Richard Henderson wrote: One cannot test for feature aa32_simd_r32 without first testing if AArch32 mode is supported at all. This leads to qemu-system-aarch64: ARM CPUs must have both VFP-D32 and Neon or neither for Apple M1 cpus. We already have a check for ARMv8-A never

[PATCH] target/arm: Restructure has_vfp_d32 test

2023-06-19 Thread Richard Henderson
One cannot test for feature aa32_simd_r32 without first testing if AArch32 mode is supported at all. This leads to qemu-system-aarch64: ARM CPUs must have both VFP-D32 and Neon or neither for Apple M1 cpus. We already have a check for ARMv8-A never setting vfp-d32 true, so restructure the code