From: Richard Henderson <richard.henderson@linaro.org>
To: Taylor Simpson <tsimpson@quicinc.com>,
Richard Henderson <rth@twiddle.net>,
"qemu-devel@nongnu.org" <qemu-devel@nongnu.org>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Subject: Re: [PATCH] Add support for a helper with 7 arguments
Date: Sun, 9 Feb 2020 10:17:56 -0800 [thread overview]
Message-ID: <fc6876dc-1a15-152c-5585-4ab06deb60fc@linaro.org> (raw)
In-Reply-To: <SN6PR02MB4895FE0821C3927B41636AB0DE1E0@SN6PR02MB4895.namprd02.prod.outlook.com>
On 2/8/20 5:08 AM, Taylor Simpson wrote:
>> {
>> r6 = memb(r1)
>> r7 = memb(r2)
>> }
>>
>> qemu_ld t0, r1, MO_UB, mmu_idx
>> qemu_ld t1, r2, MO_UB, mmu_idx
>> mov r6, t0
>> mov r7, t1
>>
>
> Here is the TCG we generate currently.
> movi_i32 tmp0,$0x0
> add_i32 loc2,r1,tmp0
> qemu_ld_i32 loc3,loc2,sb,0
> mov_i32 new_value,loc3
> movi_i32 tmp0,$0x0
> add_i32 loc2,r2,tmp0
> qemu_ld_i32 loc3,loc2,sb,0
> mov_i32 new_value,loc3
> mov_i32 r6,new_value
> mov_i32 r7,new_value
> I could work on eliminating the add of zero and the extra copies. Is TCG
> able to optimize these before emitting the host code?
We can optimize them.
However, you should prefer to use tcg_gen_addi_* over tcg_gen_add_* (etc) when
you know that one operand is constant. This will optimize away the add zero
immediately as opposed to allocating memory and walking the data structures to
eliminate it later.
Why are you using a local temporary for EA? That should be dead immediately
after this slot is complete.
What's with two temporaries both named "new_value"?
>> For a conditional load in slot 0, we can load directly into the final
>> destination register and skip the temporary.
>
> In general, there will be lots of checks we would need to perform before
> concluding that an instruction can write directly into the destination. For
> example, we have to make sure no other instruction later in the packet reads
> r7.
Which is of course all trivial for slot 0, being last.
r~
next prev parent reply other threads:[~2020-02-09 18:18 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-02-05 22:41 [PATCH] Add support for a helper with 7 arguments Taylor Simpson
2020-02-06 6:02 ` Richard Henderson
2020-02-06 10:28 ` Richard Henderson
2020-02-06 14:03 ` Taylor Simpson
2020-02-06 15:35 ` Richard Henderson
2020-02-06 17:52 ` Taylor Simpson
2020-02-07 0:27 ` Richard Henderson
2020-02-07 4:46 ` Taylor Simpson
2020-02-07 8:53 ` Richard Henderson
2020-02-07 11:59 ` Taylor Simpson
2020-02-07 12:14 ` Richard Henderson
2020-02-07 12:43 ` Taylor Simpson
2020-02-07 15:49 ` Richard Henderson
2020-02-09 5:08 ` Taylor Simpson
2020-02-09 18:17 ` Richard Henderson [this message]
2020-02-09 20:51 ` Taylor Simpson
2020-02-10 4:54 ` Taylor Simpson
2020-02-10 16:33 ` 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=fc6876dc-1a15-152c-5585-4ab06deb60fc@linaro.org \
--to=richard.henderson@linaro.org \
--cc=pbonzini@redhat.com \
--cc=qemu-devel@nongnu.org \
--cc=rth@twiddle.net \
--cc=tsimpson@quicinc.com \
/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).