From: "Philippe Mathieu-Daudé" <philmd@linaro.org>
To: Richard Henderson <richard.henderson@linaro.org>, qemu-devel@nongnu.org
Subject: Re: [PATCH v2 3/3] softmmu: Introduce cpu_address_space_sync
Date: Fri, 15 Sep 2023 18:57:59 +0200 [thread overview]
Message-ID: <fc6589b5-afc7-8fd1-a53b-ca600ab03100@linaro.org> (raw)
In-Reply-To: <20230915163254.123338-4-richard.henderson@linaro.org>
On 15/9/23 18:32, Richard Henderson wrote:
> Synchronously check and update the address space for the
> current cpu for any slow path access.
>
> Resolves: https://gitlab.com/qemu-project/qemu/-/issues/1866
> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
> ---
> include/exec/memory.h | 6 ++++++
> accel/tcg/cputlb.c | 2 ++
> softmmu/physmem.c | 20 ++++++++++++++++++++
> 3 files changed, 28 insertions(+)
>
> diff --git a/include/exec/memory.h b/include/exec/memory.h
> index 68284428f8..7ec842076f 100644
> --- a/include/exec/memory.h
> +++ b/include/exec/memory.h
> @@ -2780,6 +2780,12 @@ void address_space_cache_destroy(MemoryRegionCache *cache);
> IOMMUTLBEntry address_space_get_iotlb_entry(AddressSpace *as, hwaddr addr,
> bool is_write, MemTxAttrs attrs);
>
> +/*
> + * Ensure all cpu address spaces are up-to-date.
> + * Return true if changes made and tlb flushed.
No return value, stale comment?
> + */
> +void cpu_address_space_sync(CPUState *cpu);
> +
> /* address_space_translate: translate an address range into an address space
> * into a MemoryRegion and an address range into that section. Should be
> * called from an RCU critical section, to avoid that the last reference
> diff --git a/accel/tcg/cputlb.c b/accel/tcg/cputlb.c
> index 3270f65c20..91be3f3064 100644
> --- a/accel/tcg/cputlb.c
> +++ b/accel/tcg/cputlb.c
> @@ -1827,6 +1827,8 @@ static bool mmu_lookup(CPUArchState *env, vaddr addr, MemOpIdx oi,
> l->page[1].size = 0;
> crosspage = (addr ^ l->page[1].addr) & TARGET_PAGE_MASK;
>
> + cpu_address_space_sync(env_cpu(env));
> +
> if (likely(!crosspage)) {
> mmu_lookup1(env, &l->page[0], l->mmu_idx, type, ra);
>
> diff --git a/softmmu/physmem.c b/softmmu/physmem.c
> index e1c535380a..5a89caa257 100644
> --- a/softmmu/physmem.c
> +++ b/softmmu/physmem.c
> @@ -2536,6 +2536,26 @@ static void tcg_commit(MemoryListener *listener)
> }
> }
>
> +void cpu_address_space_sync(CPUState *cpu)
> +{
> + int i, n = cpu->num_ases;
> + bool need_flush = false;
> +
> + for (i = 0; i < n; ++i) {
> + CPUAddressSpace *cpuas = &cpu->cpu_ases[i];
> + uint32_t gen = qatomic_load_acquire(&cpuas->layout_gen);
> +
> + if (cpuas->commit_gen != gen) {
> + cpuas->commit_gen = gen;
> + cpuas->memory_dispatch = address_space_to_dispatch(cpuas->as);
> + need_flush = true;
> + }
> + }
> + if (need_flush) {
> + tlb_flush(cpu);
> + }
> +}
> +
> static void memory_map_init(void)
> {
> system_memory = g_malloc(sizeof(*system_memory));
next prev parent reply other threads:[~2023-09-15 16:58 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-09-15 16:32 [PATCH v2 0/3] accel/tcg: Alternate fix for #1866 Richard Henderson
2023-09-15 16:32 ` [PATCH v2 1/3] softmmu: Use cpu->created in tcg_commit Richard Henderson
2023-09-15 16:55 ` Philippe Mathieu-Daudé
2023-09-15 16:32 ` [PATCH v2 2/3] softmmu: Introduce AddressSpaceDispatch generation numbers Richard Henderson
2023-09-15 16:32 ` [PATCH v2 3/3] softmmu: Introduce cpu_address_space_sync Richard Henderson
2023-09-15 16:57 ` Philippe Mathieu-Daudé [this message]
2023-09-15 17:19 ` Richard Henderson
2023-09-15 16:55 ` [PATCH v2 0/3] accel/tcg: Alternate fix for #1866 Philippe Mathieu-Daudé
2023-09-15 17:18 ` Richard Henderson
2023-09-15 18:54 ` Richard Henderson
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=fc6589b5-afc7-8fd1-a53b-ca600ab03100@linaro.org \
--to=philmd@linaro.org \
--cc=qemu-devel@nongnu.org \
--cc=richard.henderson@linaro.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).