From mboxrd@z Thu Jan 1 00:00:00 1970 Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.remote-ip=100.103.45.18 ARC-Seal: i=1; d=subspace.kernel.org; s=arc20261015; a=rsa-sha256; cv=none; t=1794924307; b=lQwm1uPZ0GHvNtmLy6VEPJVdE01m7gf2nyx9WyqPJNZmkK5wqn39LufdMI5XAbhhB7iQ WFmipSca9zFADn+9xtDhxe+WL2xhsLDN0TqZ75WNHdQLEkNDCPsc3EwD8YyOW64J+cQ4v fG/1bhIdPTyLkK5OuxhugGrxEQAb5H2LRtF0t8zK4tU50N3nfAHwiyPgrLSQN1dcO9AwZ lMXqMfqYN6bnfQDzqZfr+qzrgh+JRiUL7hM0ahLSDEetLlngcguc+4jASIG5qk1If7m+O l2P3xVEcmQ6wuguht/ZBiqUmtvvpsOzV7ujR3LPhr/WvLo9ukBqolfIWQ5DFMvDeGcw== ARC-Message-Signature: i=1; d=subspace.kernel.org; s=arc20261015; a=rsa-sha256; c=relaxed/relaxed; t=1794924307; h=Received:DKIM-Signature:Message-ID:From:Subject:Reply-To:To:Cc: In-Reply-To:References:Content-Type:Content-Transfer-Encoding:Date; bh=coQw/pjKP5s+yblH7PMp+lJnV2E2rZhvzsYmHLQwVOQ=; b=jvrfvc6Sg9ET9Qa1DyVOeoD0hlg8kpCTd6eSExhOcpSOz3sZjTSF3B27I2YbwPLnJVpY MMDwXPJE6AfCFoR2UWWlcMJMIe6yNld118W4z8wQLHon4XVw1kvftXbetn/tmpdgKYrdM 9TCbEjq8f8BPeSlZRRh2aBE8Gl5Ygyl4IwM5Lx7CgOIumoT98wuCx/jbodFx4aIDTzklU wGXwMkWZSEI1rMA4JHDnumq4//smUGk1ZHKmPztfW0ZlNxFF5wPvuLdsbwR80mdiPb3xh VTQN311UThwna9QaaKLtpDpNg9LXFtUYZ6amj/b0U45vYEmbbxWX0Jjk0PNJLf2taCg== ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dkim=pass header.d=kernel.org header.i=@kernel.org header.a=rsa-sha256 header.s=k20260515 header.b=QLFf8uoQ; arc=none smtp.remote-ip=100.103.45.18 Received: from smtp.kernel.org (quasi.space.kernel.org [100.103.45.18]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature ECDSA (prime256v1)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id D8CA3606A896; Tue, 17 Nov 2026 14:05:07 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.a=rsa-sha256 header.s=k20260515 header.b=QLFf8uoQ Received: by smtp.kernel.org (Postfix) with ESMTPSA id 186FF1F000C7; Tue, 17 Nov 2026 14:05:07 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=kernel.org; s=k20260515; t=1794924307; bh=coQw/pjKP5s+yblH7PMp+lJnV2E2rZhvzsYmHLQwVOQ=; h=From:Subject:Reply-To:To:Cc:In-Reply-To:References:Date; b=QLFf8uoQaCvzNQPhWl3jEanWjUD657Ksl7ZzS5r6R7+f8bA6ToKxQIRiJCsQdmWOE F4gvMM6X2hyBkb7LYdG0JgNyquET78KF66d4PndQFObVMTQ9Yxg3rWjrq8jau1GuWq 2xKL0Pq2eNUeU3ckpj11fTkpHbq9k/gf8Uo3m7u+6Z2gRo+oyW4+fyyoGp9HuLGMkW 5C5i27F7g4UPIeMzwq41XRC3h+hoSdOLGddN+T6kDspsr5xAZCq/TTrgBXjx/uY8MW mXDNHkk1ljNz5dtPfbnnIIgy2xsMTsnqOWnqLUJpvPDolLko4etqUqc/W2c9F5vYLu Y7PHKbEd1+BrQ== Message-ID: From: sashiko-bot@kernel.org Subject: Re: [PATCH RFC 6/6] ARM: add the SiRF AT550 platform and YPL514si board Reply-To: sashiko-reviews@lists.linux.dev To: yy-Local-LLM , devicetree@vger.kernel.org Cc: linux-input@vger.kernel.org, robh@kernel.org, conor+dt@kernel.org, dmitry.torokhov@gmail.com In-Reply-To: <20261117-at550-rfc-v1-v1-6-5879573cb825@gmail.com> References: <20261117-at550-rfc-v1-v1-0-5879573cb825@gmail.com> <20261117-at550-rfc-v1-v1-6-5879573cb825@gmail.com> Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable Date: Tue, 17 Nov 2026 14:05:06 +0000 X-Mailing-List: devicetree@vger.kernel.org List-Id: List-Post: List-Help: List-Subscribe: List-Unsubscribe: List-Owner: Precedence: list Thank you for your contribution! Sashiko AI review found 5 potential issues to consider: - [High] ARM: at550: non-real name yy-Local-LLM in Signed-off-by and MAINTAINERS - [High] ARM: at550: ARCH_AT550 selects CPU_V6 without !SMP dependency in Kconfig - [High] ARM: dts: at550: /memreserve/ causes ioremap_wc() failure in simplefb_probe() - [Low] ARM: at550_defconfig: missing CONFIG_ARCH_MULTI_V7 disablement - [Low] ARM: dts: unsorted subdir-y entry and missing newline in Makefile -- commit 77bddca305316e26eed799b27f4f024275217704 Author: yy-Local-LLM > ARM: add the SiRF AT550 platform and YPL514si board > > Add a minimal DT machine using standard IRQ, timer and simplefb paths. > Describe the SoC and the firmware-configured YPL514si display. Leave > usable memory and the divided counter rate for the bootloader to > populate. Keep optional peripheral nodes disabled until firmware hands > them over. Disable unvalidated board suspend; limit this board to SD PIO > at 25 MHz. Document the RAM-boot contract and the missing > cold-boot/clock-tree support. > > Signed-off-by: yy-Local-LLM [Severity: High] Documentation/process/submitting-patches.rst requires a real full human name in the Signed-off-by trailer and in MAINTAINERS rather than a handle or bot name (yy-Local-LLM). [ ... ] > diff --git a/arch/arm/boot/dts/Makefile b/arch/arm/boot/dts/Makefile > index 28fba538d5520..a4844a80b8d08 100644 > --- a/arch/arm/boot/dts/Makefile > +++ b/arch/arm/boot/dts/Makefile > @@ -40,3 +40,5 @@ subdir-y +=3D vt8500 > subdir-y +=3D xen > subdir-y +=3D xilinx > subdir-y +=3D zte > + > +subdir-y +=3D sirf > \ No newline at end of file [Severity: Low] This isn't a bug, but the vendor subdirectories in this Makefile are kept in alphabetical order (between sigmastar and socionext), without a blank line, and the file is missing a trailing newline. [ ... ] > diff --git a/arch/arm/boot/dts/sirf/at550-ypl514si.dts b/arch/arm/boot/dt= s/sirf/at550-ypl514si.dts > new file mode 100644 > index 0000000000000..d45328fa09e82 > --- /dev/null > +++ b/arch/arm/boot/dts/sirf/at550-ypl514si.dts > @@ -0,0 +1,52 @@ > +// SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) > +/dts-v1/; > +/memreserve/ 0xc0100000 0x40000; > +#include "at550.dtsi" > + > +/ { > + model =3D "Yupiteru YPL514si"; > + compatible =3D "yupiteru,ypl514si", "sirf,at550"; > + > + memory@c0000000 { > + device_type =3D "memory"; > + /* The bootloader supplies the region allocated for Linux. */ > + reg =3D <0xc0000000 0>; > + }; > + > + chosen { > + #address-cells =3D <1>; > + #size-cells =3D <1>; > + ranges; > + > + framebuffer@c0100000 { > + compatible =3D "simple-framebuffer"; > + reg =3D <0xc0100000 0x3fc00>; > + width =3D <480>; > + height =3D <272>; > + stride =3D <960>; > + format =3D "r5g6b5"; > + status =3D "disabled"; > + }; > + }; > +}; [Severity: High] Will reserving the framebuffer with /memreserve/ cause simplefb_probe() to fail on 32-bit arm? /memreserve/ only calls memblock_reserve() without setting MEMBLOCK_NOMAP, so map_lowmem() still maps 0xc0100000 into the kernel linear mapping when memory@c0000000 covers that range. When simplefb_probe() then calls ioremap_wc(), __arm_ioremap_pfn_caller() hits a WARN_ON() and returns NULL to reject the mismatched memory attribute alias, failing probe with -ENOMEM: arch/arm/mm/ioremap.c:__arm_ioremap_pfn_caller() { ... if (WARN_ON(memblock_is_map_memory(PFN_PHYS(pfn)) && mtype !=3D MT_MEMORY_RW)) return NULL; ... } Reserving the region via a /reserved-memory child node with the no-map property excludes it from the linear mapping so ioremap_wc() succeeds. [ ... ] > diff --git a/arch/arm/configs/at550_defconfig b/arch/arm/configs/at550_de= fconfig > new file mode 100644 > index 0000000000000..698b2f67574d0 > --- /dev/null > +++ b/arch/arm/configs/at550_defconfig > @@ -0,0 +1,34 @@ > +# CONFIG_LOCALVERSION_AUTO is not set > +CONFIG_SYSVIPC=3Dy > +CONFIG_NO_HZ_IDLE=3Dy > +CONFIG_HIGH_RES_TIMERS=3Dy > +CONFIG_BLK_DEV_INITRD=3Dy > +CONFIG_ARCH_MULTI_V6=3Dy > +CONFIG_ARCH_AT550=3Dy [Severity: Low] Should at550_defconfig explicitly disable CONFIG_ARCH_MULTI_V7? ARCH_MULTI_V7 defaults to y in arch/arm/Kconfig.platforms, so omitting "# CONFIG_ARCH_MULTI_V7 is not set" leaves CPU_V7, HAVE_SMP, multi-CPU/TLB dispatch, and a 64-byte L1 cache line shift enabled for this single-core ARM1136 target. [ ... ] > diff --git a/arch/arm/mach-at550/Kconfig b/arch/arm/mach-at550/Kconfig > new file mode 100644 > index 0000000000000..c6380e529641d > --- /dev/null > +++ b/arch/arm/mach-at550/Kconfig > @@ -0,0 +1,13 @@ > +# SPDX-License-Identifier: GPL-2.0-only > +config ARCH_AT550 > + bool "SiRF AT550 platforms" > + depends on ARCH_MULTI_V6 > + select ARM_ERRATA_411920 > + select AT550_IRQ > + select CLKSRC_AT550 > + select CPU_V6 [Severity: High] Does selecting CPU_V6 without a !SMP dependency break multiplatform SMP builds? In arch/arm/mm/Kconfig, CPU_V6 depends on !SMP. Enabling CONFIG_ARCH_AT550 in a configuration with CONFIG_SMP=3Dy (such as allmodconfig or a combined v6/v7 multiplatform kernel) triggers an unmet direct dependency warning for CPU_V6 and compiles out 1-byte and 2-byte __arch_xchg() and 2-byte __cmpxchg() in arch/arm/include/asm/cmpxchg.h, causing link failures with undefined references to __bad_xchg and __bad_cmpxchg (and smp_on_up when CONFIG_SMP_ON_UP=3Dn). Other ARM1136 platforms either select CPU_ARM1136R1 (which selects CPU_V6K) or depend on !SMP when selecting CPU_ARM1136R0 / CPU_V6. --=20 Sashiko AI review =C2=B7 https://sashiko.dev/#/patchset/20261117-at550-rfc-= v1-v1-0-5879573cb825@gmail.com?part=3D6