From: Nick Desaulniers <ndesaulniers@google.com>
To: Arnd Bergmann <arnd@kernel.org>, Russell King <linux@armlinux.org.uk>
Cc: Tom Rix <trix@redhat.com>,
linux-arm-kernel@lists.infradead.org,
linux-kernel@vger.kernel.org, llvm@lists.linux.dev,
Miguel Ojeda <ojeda@kernel.org>, Ard Biesheuvel <ardb@kernel.org>,
Gary Guo <gary@garyguo.net>,
Craig Topper <craig.topper@sifive.com>,
Philip Reames <listmail@philipreames.com>,
jh@jhauser.us, Nick Desaulniers <ndesaulniers@google.com>,
Nathan Chancellor <nathan@kernel.org>
Subject: [PATCH] ARM: NWFPE: avoid compiler-generated __aeabi_uldivmod
Date: Mon, 10 Oct 2022 15:53:42 -0700 [thread overview]
Message-ID: <20221010225342.3903590-1-ndesaulniers@google.com> (raw)
In-Reply-To: <CAKwvOdnQ-tWa0oY0CKAwJD5WQDgVBvxDtf6=E3hS8Ytz5J1OKg@mail.gmail.com>
clang-15's ability to elide loops completely became more aggressive when
it can deduce how a variable is being updated in a loop. Counting down
one variable by an increment of another can be replaced by a modulo
operation.
For 64b variables on 32b ARM EABI targets, this can result in the
compiler generating calls to __aeabi_uldivmod, which it does for a do
while loop in float64_rem().
For the kernel, we'd generally prefer that developers not open code 64b
division via binary / operators and instead use the more explicit
helpers from div64.h. On arm-linux-gnuabi targets, failure to do so can
result in linkage failures due to undefined references to
__aeabi_uldivmod().
While developers can avoid open coding divisions on 64b variables, the
compiler doesn't know that the Linux kernel has a partial implementation
of a compiler runtime (--rtlib) to enforce this convention.
It's also undecidable for the compiler whether the code in question
would be faster to execute the loop vs elide it and do the 64b division.
While I actively avoid using the internal -mllvm command line flags, I
think we get better code than using barrier() here, which will force
reloads+spills in the loop for all toolchains.
Link: https://github.com/ClangBuiltLinux/linux/issues/1666
Reported-by: Nathan Chancellor <nathan@kernel.org>
Signed-off-by: Nick Desaulniers <ndesaulniers@google.com>
---
arch/arm/nwfpe/Makefile | 6 ++++++
1 file changed, 6 insertions(+)
diff --git a/arch/arm/nwfpe/Makefile b/arch/arm/nwfpe/Makefile
index 303400fa2cdf..2aec85ab1e8b 100644
--- a/arch/arm/nwfpe/Makefile
+++ b/arch/arm/nwfpe/Makefile
@@ -11,3 +11,9 @@ nwfpe-y += fpa11.o fpa11_cpdo.o fpa11_cpdt.o \
entry.o
nwfpe-$(CONFIG_FPE_NWFPE_XP) += extended_cpdo.o
+
+# Try really hard to avoid generating calls to __aeabi_uldivmod() from
+# float64_rem() due to loop elision.
+ifdef CONFIG_CC_IS_CLANG
+CFLAGS_softfloat.o += -mllvm -replexitval=never
+endif
--
2.38.0.rc2.412.g84df46c1b4-goog
next prev parent reply other threads:[~2022-10-10 22:53 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-07-16 0:16 [PATCH] arm: lib: implement aeabi_uldivmod via div64_u64_rem Nick Desaulniers
2022-07-16 9:46 ` Arnd Bergmann
2022-10-10 21:23 ` Nick Desaulniers
2022-10-10 22:13 ` Arnd Bergmann
2022-10-10 22:34 ` Nick Desaulniers
2022-10-10 22:53 ` Nick Desaulniers [this message]
2022-10-11 11:01 ` [PATCH] ARM: NWFPE: avoid compiler-generated __aeabi_uldivmod Arnd Bergmann
2022-10-13 17:04 ` Nathan Chancellor
2022-07-18 16:50 ` [PATCH] arm: lib: implement aeabi_uldivmod via div64_u64_rem kernel test robot
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=20221010225342.3903590-1-ndesaulniers@google.com \
--to=ndesaulniers@google.com \
--cc=ardb@kernel.org \
--cc=arnd@kernel.org \
--cc=craig.topper@sifive.com \
--cc=gary@garyguo.net \
--cc=jh@jhauser.us \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux@armlinux.org.uk \
--cc=listmail@philipreames.com \
--cc=llvm@lists.linux.dev \
--cc=nathan@kernel.org \
--cc=ojeda@kernel.org \
--cc=trix@redhat.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).