On 9/20/21 1:08 AM, Peter Maydell wrote:
On Mon, 20 Sept 2021 at 02:29, Richard Henderson
<richard.hender...@linaro.org> wrote:
On 8/26/21 6:45 AM, Peter Maydell wrote:
I don't think you should need to special case AArch64 vs AArch32 like this;
you can do
env->exception.vaddress = pc;
env->exception.fsr = the_fsr;
raise_exception(env, EXCP_PREFETCH_ABORT, syn_pcalignment(), target_el);
for both. AArch64/AArch32-Hyp exception entry will ignore exception.fsr,
and AArch32-not-Hyp entry will ignore exception.syndrome.
Not true. The latter case still requires syndrome with EC_INSNABORT, etc.
For AArch32-not-Hyp ? Syndrome doesn't matter at all in that case
(only Hyp mode and AArch64 have syndrome registers); it just needs
to take the prefetch abort exception, which you get by using
EXCP_PREFETCH_ABORT.
In which case I have incorrect asserts over in cpu_loop. Sigh.
r~
