[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAJF2gTTRHzT+CEtb1LVYdfCorVUdLvCh_eMxrmC=xjdQ_JS6Sg@mail.gmail.com>
Date: Mon, 31 Jul 2023 08:49:17 +0800
From: Guo Ren <guoren@...nel.org>
To: Heiko Stübner <heiko@...ech.de>,
Arnd Bergmann <arnd@...db.de>
Cc: Emil Renner Berthing <emil.renner.berthing@...onical.com>,
Jisheng Zhang <jszhang@...nel.org>,
Prabhakar <prabhakar.csengg@...il.com>,
"Conor.Dooley" <conor.dooley@...rochip.com>,
Geert Uytterhoeven <geert+renesas@...der.be>,
Andrew Jones <ajones@...tanamicro.com>,
Paul Walmsley <paul.walmsley@...ive.com>,
Palmer Dabbelt <palmer@...belt.com>,
Albert Ou <aou@...s.berkeley.edu>,
Samuel Holland <samuel@...lland.org>,
linux-riscv@...ts.infradead.org,
Christoph Hellwig <hch@...radead.org>,
Rob Herring <robh+dt@...nel.org>,
Krzysztof Kozlowski <krzysztof.kozlowski+dt@...aro.org>,
devicetree@...r.kernel.org, linux-kernel@...r.kernel.org,
Linux-Renesas <linux-renesas-soc@...r.kernel.org>,
Biju Das <biju.das.jz@...renesas.com>,
"Lad, Prabhakar" <prabhakar.mahadev-lad.rj@...renesas.com>
Subject: Re: [PATCH v10 3/6] riscv: mm: dma-noncoherent: nonstandard cache
operations support
On Mon, Jul 31, 2023 at 4:36 AM Arnd Bergmann <arnd@...db.de> wrote:
>
> On Sun, Jul 30, 2023, at 17:42, Emil Renner Berthing wrote:
> > On Sun, 30 Jul 2023 at 17:11, Jisheng Zhang <jszhang@...nel.org> wrote:
>
> >> > +
> >> > static inline void arch_dma_cache_wback(phys_addr_t paddr, size_t size)
> >> > {
> >> > void *vaddr = phys_to_virt(paddr);
> >> >
> >> > +#ifdef CONFIG_RISCV_NONSTANDARD_CACHE_OPS
> >> > + if (unlikely(noncoherent_cache_ops.wback)) {
> >>
> >> I'm worried about the performance impact here.
> >> For unified kernel Image reason, RISCV_NONSTANDARD_CACHE_OPS will be
> >> enabled by default, so standard CMO and T-HEAD's CMO platform's
> >> performance will be impacted, because even an unlikely is put
> >> here, the check action still needs to be done.
> >
> > On IRC I asked why not use a static key so the overhead is just a
> > single nop when the standard CMO ops are available, but the consensus
> > seemed to be that the flushing would completely dominate this branch.
> > And on platforms with the standard CMO ops the branch be correctly
> > predicted anyway.
>
> Not just the flushing, but also loading back the invalidated
> cache lines afterwards is just very expensive. I don't think
> you would be able to measure a difference between the static
> key and a correctly predicted branch on any relevant usecase here.
Maybe we should move CMO & THEAD ops to the noncoherent_cache_ops, and
only keep one of them.
I prefer noncoherent_cache_ops, it's more maintance than ALTERNATIVE.
Heiko, How do you think about this?
>
> Arnd
--
Best Regards
Guo Ren
Powered by blists - more mailing lists