Re: [RFC PATCH 1/4] hw: encode accessing CPU index in MemTxAttrs

From: Richard Henderson <richard.henderson@linaro.org>
To: "Alex Bennée" <alex.bennee@linaro.org>, qemu-devel@nongnu.org
Cc: qemu-arm@nongnu.org, Paolo Bonzini <pbonzini@redhat.com>
Subject: Re: [RFC PATCH 1/4] hw: encode accessing CPU index in MemTxAttrs
Date: Thu, 15 Sep 2022 09:05:13 +0100	[thread overview]
Message-ID: <f7de79e4-a0f4-c212-6960-afa0a82bc80a@linaro.org> (raw)
In-Reply-To: <20220914160955.812151-2-alex.bennee@linaro.org>

On 9/14/22 17:09, Alex Bennée wrote:
> @@ -1340,8 +1340,13 @@ static uint64_t io_readx(CPUArchState *env, CPUIOTLBEntry *iotlbentry,
>       uint64_t val;
>       bool locked = false;
>       MemTxResult r;
> +    MemTxAttrs attrs = iotlbentry->attrs;
>   
> -    section = iotlb_to_section(cpu, iotlbentry->addr, iotlbentry->attrs);
> +    /* encode the accessing CPU */
> +    attrs.requester_cpu = 1;
> +    attrs.requester_id = cpu->cpu_index;
> +
> +    section = iotlb_to_section(cpu, iotlbentry->addr, attrs);
>       mr = section->mr;
>       mr_offset = (iotlbentry->addr & TARGET_PAGE_MASK) + addr;
>       cpu->mem_io_pc = retaddr;

At first I was going to suggest that this be done in tlb_set_page_with_attrs, so that it 
could be done once and not duplicate code across read/write.

But then I got to thinking how this ought to interact with MEMTXATTRS_UNSPECIFIED, and now 
I think that we simply have to leave this to the cpu's tlb_fill routine, where it fills in 
(or doesn't) all of the other transaction attributes.

r~