DMARC-Filter: OpenDMARC Filter v1.4.2 smtp.subspace.kernel.org 6AED7606A326
From: Michal Wilczynski <m.wilczynski@samsung.com>
Date: Fri, 06 Nov 2026 15:42:46 +0100
Subject: [PATCH v2 06/15] lib/lz4: backport upstream's aarch64 fast decode
 loop enablement
Precedence: list
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Message-Id: <20261106-lz4-vendor-upstream-v2-6-ee04c3f0d7dc@samsung.com>
In-Reply-To: <20261106-lz4-vendor-upstream-v2-0-ee04c3f0d7dc@samsung.com>
To: Michal Wilczynski <m.wilczynski@samsung.com>,  Nathan Chancellor
	<nathan@kernel.org>,  Nick Desaulniers <ndesaulniers@google.com>,  Bill
	Wendling <morbo@google.com>, Justin Stitt <justinstitt@google.com>,  Russell
	King <linux@armlinux.org.uk>,  Thomas Bogendoerfer
	<tsbogend@alpha.franken.de>,  "James E.J. Bottomley"
	<James.Bottomley@HansenPartnership.com>,  Helge Deller <deller@gmx.de>,
	Heiko Carstens <hca@linux.ibm.com>,  Vasily Gorbik <gor@linux.ibm.com>, 
	Alexander Gordeev <agordeev@linux.ibm.com>,  Christian Borntraeger
	<borntraeger@linux.ibm.com>,  Sven Schnelle <svens@linux.ibm.com>, Thomas
	Gleixner <tglx@kernel.org>,  Ingo Molnar <mingo@redhat.com>, Borislav Petkov
	<bp@alien8.de>,  Dave Hansen <dave.hansen@linux.intel.com>, x86@kernel.org, 
	"H. Peter Anvin" <hpa@zytor.com>, Minchan Kim <minchan@kernel.org>,  Sergey
	Senozhatsky <senozhatsky@chromium.org>, Jens Axboe <axboe@kernel.dk>, 
	Andrew Morton <akpm@linux-foundation.org>
Cc: linux-kernel@vger.kernel.org, llvm@lists.linux.dev,
	linux-arm-kernel@lists.infradead.org, linux-mips@vger.kernel.org, 
	linux-parisc@vger.kernel.org, linux-s390@vger.kernel.org, 
	linux-block@vger.kernel.org, Yann Collet <yann.collet.73@gmail.com>,  Nick
	Terrell <terrelln@meta.com>, Gao Xiang <xiang@kernel.org>,  Chao Yu
	<chao@kernel.org>, Jaegeuk Kim <jaegeuk@kernel.org>,  Eric Biggers
	<ebiggers@kernel.org>,  Herbert Xu <herbert@gondor.apana.org.au>,  David
	Laight <david.laight.linux@gmail.com>,  Sungguk <sungguk.na@samsung.com>,
	Jaehoon Chung <jh80.chung@samsung.com>,  Marek Szyprowski
	<m.szyprowski@samsung.com>, linux-erofs@lists.ozlabs.org, 
	linux-f2fs-devel@lists.sourceforge.net, linux-crypto@vger.kernel.org
Content-Type: text/plain; charset="utf-8"
References: <20261106-lz4-vendor-upstream-v2-0-ee04c3f0d7dc@samsung.com>
	<CGME20261106144258eucas1p1f59ead3e337482e074bfb26f4553e70d@eucas1p1.samsung.com>

v1.10.0 leaves LZ4_FAST_DEC_LOOP off for aarch64 built with clang, after
a slowdown reported on some Android phones in 2019 (upstream PR #707).
An arm64 kernel built with clang therefore decodes with the slower loop.

Upstream has since enabled it for all aarch64, in three unreleased
commits:

commit 0aa33884faf5 ("[LZ4] Limit disabling LZ4_FAST_DEC_LOOP to aarch64+clang+android")
commit 51f8af506b4a ("Fix missing endif")
commit ed86bc1ed802 ("AArch64: Enable LZ4_FAST_DEC_LOOP for Android")

Arm measured +3.5% to +15.6% on Cortex-A510 to X4 with clang. Take the
lz4.c hunks of all three. This goes away at the next re-sync.

Signed-off-by: Michal Wilczynski <m.wilczynski@samsung.com>
---
 lib/lz4/upstream/lz4.c | 7 +------
 1 file changed, 1 insertion(+), 6 deletions(-)

diff --git a/lib/lz4/upstream/lz4.c b/lib/lz4/upstream/lz4.c
index 0d474be0236a698d297ab9a430f5082485eb3709..ee69d57aae112f4057c050bbdab4cc0f2d30f9d9 100644
--- a/lib/lz4/upstream/lz4.c
+++ b/lib/lz4/upstream/lz4.c
@@ -478,12 +478,7 @@ static const int      dec64table[8] = {0, 0, 0, -1, -4,  1, 2, 3};
 #ifndef LZ4_FAST_DEC_LOOP
 #  if defined __i386__ || defined _M_IX86 || defined __x86_64__ || defined _M_X64
 #    define LZ4_FAST_DEC_LOOP 1
-#  elif defined(__aarch64__) && defined(__APPLE__)
-#    define LZ4_FAST_DEC_LOOP 1
-#  elif defined(__aarch64__) && !defined(__clang__)
-     /* On non-Apple aarch64, we disable this optimization for clang because
-      * on certain mobile chipsets, performance is reduced with clang. For
-      * more information refer to https://github.com/lz4/lz4/pull/707 */
+#  elif defined(__aarch64__)
 #    define LZ4_FAST_DEC_LOOP 1
 #  else
 #    define LZ4_FAST_DEC_LOOP 0

-- 
2.34.1