mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Xi Ruoyao <xry111@xry111.site>
To: Tiezhu Yang <yangtiezhu@loongson.cn>,
	Huacai Chen <chenhuacai@kernel.org>
Cc: loongarch@lists.linux.dev, linux-kernel@vger.kernel.org
Subject: Re: [RFC PATCH] LoongArch: Do not include larchintrin.h
Date: Tue, 27 May 2025 11:47:19 +0800	[thread overview]
Message-ID: <19f4722d5cdb90a207129ee675c7278423cd328c.camel@xry111.site> (raw)
In-Reply-To: <a918b221-b7f3-9994-9a7a-d10345aa30df@loongson.cn>

On Tue, 2025-05-27 at 11:17 +0800, Tiezhu Yang wrote:
> On 2025/5/21 下午1:41, Xi Ruoyao wrote:
> > On Tue, 2025-05-20 at 14:49 +0800, Tiezhu Yang wrote:
> > > larchintrin.h is a system header of compiler, include it in the
> > > kernel header may lead to the fatal error "'larchintrin.h' file
> > > not found".
> > > 
> > > There are two related cases so far:
> > > 
> > > (1) When compiling samples/bpf, it has been fixed in the latest
> > > kernel [1].
> > > 
> > > (2) When running bcc script, it has been fixed in the latest
> > > bcc [2] [3], like this:
> > > 
> > > $ /usr/share/bcc/tools/filetop
> > > In file included from <built-in>:4:
> > > In file included from /virtual/include/bcc/helpers.h:54:
> > > In file included from arch/loongarch/include/asm/page.h:7:
> > > In file included from arch/loongarch/include/asm/addrspace.h:9:
> > > arch/loongarch/include/asm/loongarch.h:11:10: fatal error:
> > > 'larchintrin.h' file not found
> > >     11 | #include <larchintrin.h>
> > >        |          ^~~~~~~~~~~~~~~
> > > 1 error generated.
> > > 
> > > Maybe there are same errors for the other unknown projects, it is
> > > annoyance to add the include path each time. In order to avoid
> > > such
> > > errors once and for all, do not include larchintrin.h, just use
> > > the
> > > builtin functions directly.
> > 
> > Sorry, but in GCC those builtin functions are not documented and may
> > subject to change in the future.  Only the larchintrin.h interface
> > is
> > documented.
> 
> AFAICT, the LoongArch Base Built-in Functions are listed in the GCC
> documentation [1], they will not be changed easily and frequently in
> my opinion.
> 
> __builtin_loongarch_cpucfg()
> __builtin_loongarch_csrrd_w()
> __builtin_loongarch_csrrd_d()
> __builtin_loongarch_csrwr_w()
> __builtin_loongarch_csrwr_d()
> __builtin_loongarch_csrxchg_w()
> __builtin_loongarch_csrxchg_d()
> __builtin_loongarch_iocsrrd_w()
> __builtin_loongarch_iocsrrd_d()
> __builtin_loongarch_iocsrwr_w()
> __builtin_loongarch_iocsrwr_d()
> 
> > Thus if you don't want to rely on GCC for those operations, you may
> > need
> > to write inline asm...
> 
> so these builtin functions can be used directly and safely.

Oops, I mistakenly believed they were like __builtin_lsx_* which are not
documented.

So yes they can be used directly.

-- 
Xi Ruoyao <xry111@xry111.site>
School of Aerospace Science and Technology, Xidian University

  reply	other threads:[~2025-05-27  3:47 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-05-20  6:49 Tiezhu Yang
2025-05-21  2:14 ` Hengqi Chen
2025-05-21  5:41 ` Xi Ruoyao
2025-05-27  3:17   ` Tiezhu Yang
2025-05-27  3:47     ` Xi Ruoyao [this message]
2025-05-28 13:54       ` Huacai Chen

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=19f4722d5cdb90a207129ee675c7278423cd328c.camel@xry111.site \
    --to=xry111@xry111.site \
    --cc=chenhuacai@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=loongarch@lists.linux.dev \
    --cc=yangtiezhu@loongson.cn \
    /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

all inboxes | Powered by JetHome®