From: "Nicholas Piggin" <npiggin@gmail.com>
To: "Richard Henderson" <richard.henderson@linaro.org>,
<qemu-devel@nongnu.org>
Cc: <qemu-ppc@nongnu.org>, <balaton@eik.bme.hu>
Subject: Re: [PATCH 1/4] target/ppc/mem_helper.c: Remove a conditional from dcbz_common()
Date: Wed, 17 Jul 2024 21:36:15 +1000 [thread overview]
Message-ID: <D2RS7N16U915.3TZB6FC33BXQB@gmail.com> (raw)
In-Reply-To: <20240702234659.2106870-2-richard.henderson@linaro.org>
On Wed Jul 3, 2024 at 9:46 AM AEST, Richard Henderson wrote:
> From: BALATON Zoltan <balaton@eik.bme.hu>
>
> Instead of passing a bool and select a value within dcbz_common() let
> the callers pass in the right value to avoid this conditional
> statement. On PPC dcbz is often used to zero memory and some code uses
> it a lot. This change improves the run time of a test case that copies
> memory with a dcbz call in every iteration from 6.23 to 5.83 seconds.
>
Reviewed-by: Nicholas Piggin <npiggin@gmail.com>
> Signed-off-by: BALATON Zoltan <balaton@eik.bme.hu>
> Message-Id: <20240622204833.5F7C74E6000@zero.eik.bme.hu>
> Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
> ---
> target/ppc/mem_helper.c | 7 +++----
> 1 file changed, 3 insertions(+), 4 deletions(-)
>
> diff --git a/target/ppc/mem_helper.c b/target/ppc/mem_helper.c
> index f88155ad45..361fd72226 100644
> --- a/target/ppc/mem_helper.c
> +++ b/target/ppc/mem_helper.c
> @@ -271,12 +271,11 @@ void helper_stsw(CPUPPCState *env, target_ulong addr, uint32_t nb,
> }
>
> static void dcbz_common(CPUPPCState *env, target_ulong addr,
> - uint32_t opcode, bool epid, uintptr_t retaddr)
> + uint32_t opcode, int mmu_idx, uintptr_t retaddr)
> {
> target_ulong mask, dcbz_size = env->dcache_line_size;
> uint32_t i;
> void *haddr;
> - int mmu_idx = epid ? PPC_TLB_EPID_STORE : ppc_env_mmu_index(env, false);
>
> #if defined(TARGET_PPC64)
> /* Check for dcbz vs dcbzl on 970 */
> @@ -309,12 +308,12 @@ static void dcbz_common(CPUPPCState *env, target_ulong addr,
>
> void helper_dcbz(CPUPPCState *env, target_ulong addr, uint32_t opcode)
> {
> - dcbz_common(env, addr, opcode, false, GETPC());
> + dcbz_common(env, addr, opcode, ppc_env_mmu_index(env, false), GETPC());
> }
>
> void helper_dcbzep(CPUPPCState *env, target_ulong addr, uint32_t opcode)
> {
> - dcbz_common(env, addr, opcode, true, GETPC());
> + dcbz_common(env, addr, opcode, PPC_TLB_EPID_STORE, GETPC());
> }
>
> void helper_icbi(CPUPPCState *env, target_ulong addr)
next prev parent reply other threads:[~2024-07-17 11:36 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-07-02 23:46 [PATCH 0/4] target/ppc: Cleanups for dcbz Richard Henderson
2024-07-02 23:46 ` [PATCH 1/4] target/ppc/mem_helper.c: Remove a conditional from dcbz_common() Richard Henderson
2024-07-17 11:36 ` Nicholas Piggin [this message]
2024-07-02 23:46 ` [PATCH 2/4] target/ppc: Hoist dcbz_size out of dcbz_common Richard Henderson
2024-07-17 11:23 ` Nicholas Piggin
2024-07-02 23:46 ` [PATCH 3/4] target/ppc: Split out helper_dbczl for 970 Richard Henderson
2024-07-17 11:33 ` Nicholas Piggin
2024-07-02 23:46 ` [PATCH 4/4] target/ppc: Merge helper_{dcbz,dcbzep} Richard Henderson
2024-07-17 11:35 ` Nicholas Piggin
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=D2RS7N16U915.3TZB6FC33BXQB@gmail.com \
--to=npiggin@gmail.com \
--cc=balaton@eik.bme.hu \
--cc=qemu-devel@nongnu.org \
--cc=qemu-ppc@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).