From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mta1.migadu.com (out-154.mta1.migadu.com [95.215.58.154]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id AA88E39185C for ; Wed, 14 Oct 2026 16:41:30 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=95.215.58.154 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1791996092; cv=none; b=vFjkqES01VPE+i1Wa4BdU7Pu6aASveDTeVkXUc6O1oi56sld7rJCcSvlwircxV8GxT1C2XMyBGSdFcI4qcWSfNnnrHQ/xb905yns8uTTeaod0o8Y4h67nuG/tsT8FiLp/W/q9aYjk6G1Gv0vdnqor0iO53wZFaHiudpW6ycgdsU= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1791996092; c=relaxed/simple; bh=kQ75PMv+BY9OnLoeEVPyFqxTAPL+KhKo+I77xMhcgcc=; h=Message-ID:Date:MIME-Version:Subject:To:Cc:References:From: In-Reply-To:Content-Type; b=LvWwmPPsLcy4IVKwTzEA+yW8Ql03tOKkW2KEMb8Q03vU8D2R+5M7MboJPbCFW6NesSHAejrD+oh16o6hIIhLONqKJeWr1szjnE5S0t0Cj0ZL6MBRukbEiG21uXSmFx3H2+1soY7mJIM8kgBrWdGcwEqQQ1cws6N/VzGnGA7JLvw= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dmarc=pass (p=quarantine dis=none) header.from=packett.cool; spf=pass smtp.mailfrom=packett.cool; dkim=pass (2048-bit key) header.d=packett.cool header.i=@packett.cool header.b=MsOieU8G; arc=none smtp.client-ip=95.215.58.154 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=quarantine dis=none) header.from=packett.cool Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=packett.cool Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=packett.cool header.i=@packett.cool header.b="MsOieU8G" X-Envelope-To: linux-kernel@vger.kernel.org DKIM-Signature: a=rsa-sha256; bh=kQ75PMv+BY9OnLoeEVPyFqxTAPL+KhKo+I77xMhcgcc=; c=simple/simple; d=packett.cool; h=from:to:subject:date:message-id:mime-version:content-type; s=key1; t=1791996088; v=1; x=1792600888; b=MsOieU8GejELPQB8oXQZ702KEehwpVk1NSUAGj509KZOkxXo5eNCJepaJ/YuteKifIZQYZkI LAOlwOis9kUDG7Nmyg3M19OQIHYRwjOeuVG2GmHB3aF3yB3J0vLmSncpSBgfV6fjZ6qrd+60w6F ob9OFUTVW8QwncRsah8TiUOe/JxE3ouS3l/2ax47QVaUjqJwhwF/m/ywYMoFxM7wMAuy94qe3/h hleDlYsc2Sjcch8KM51WPNmB3Bn1BYxS0II3bJS2+TZRzDKBCaZgZ0CrUnTLF6d4ekOedWJ4VZR 7wBQPdJVxe4CD6D80pW1aUKuwL/fd+awHvVz8j2qHHLQw== X-Envelope-To: linux-kernel@vger.kernel.org Received: by smtp.migadu.com with ESMTPS id 454a9d4a808200d2; Wed, 14 Oct 2026 16:41:28 +0000 X-Mizu-Trace-ID: 454a9d4a808200d2 X-Migadu-Flow: FLOW_OUT Message-ID: <12d14e1d-8355-4b3b-9eef-bd5e90dfd19f@packett.cool> Date: Wed, 14 Oct 2026 13:41:19 -0300 Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 User-Agent: Mozilla Thunderbird Subject: Re: [PATCH v1] arm64: dts: qcom: talos: Make the ADSP remote heap no-map To: Jianping Li , Bjorn Andersson , Konrad Dybcio Cc: Rob Herring , Krzysztof Kozlowski , Conor Dooley , linux-arm-msm@vger.kernel.org, devicetree@vger.kernel.org, linux-kernel@vger.kernel.org, ekansh.gupta@oss.qualcomm.com, quic_chennak@quicinc.com References: <20261014083820.468-1-jianping.li@oss.qualcomm.com> Content-Language: en-US From: Val Packett In-Reply-To: <20261014083820.468-1-jianping.li@oss.qualcomm.com> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 8bit On 10/14/26 5:38 AM, Jianping Li wrote: > The ADSP remote heap is declared as a reusable CMA pool. FastRPC > assigns the entire region to the DSP through qcom_scm_assign_mem() > during probe without allocating the pages from CMA. Consequently, > the pages remain available for movable HLOS allocations even after > HLOS access to the region has been revoked. > > On Talos, GPU stress testing triggers a synchronous external abort > when zram accesses pages within the ADSP remote heap. > > Replace reusable with no-map to prevent the region from being reused > for normal kernel allocations and exclude it from the kernel's > standard memory mapping. > > Fixes: b0b51a5f1911f ("arm64: dts: qcom: talos: Add memory-region for audio PD") > Link: https://lore.kernel.org/all/20260424-talosaudio-v3-1-9e2ad5d78a2e@oss.qualcomm.com/ > Signed-off-by: Jianping Li > --- > arch/arm64/boot/dts/qcom/talos.dtsi | 2 +- This also applies to: glymur, agatti, sm8250, kodiak, sm8750.. (at least those I've found by grepping for the adsp_rpc_remote_heap_mem label) I've been hitting the SEA on a kodiak phone (sm7325-motorola-dubai) constantly with just light everyday usage: Internal error: synchronous external abort: 0000000096000010 [#1]  SMP CPU: 7 UID: 10000 PID: 26485 Comm: pool-24 Tainted: G   M    W         7.3.0-rc6-next-20261009-uwu+ #167 PREEMPT(full) Tainted: [M]=MACHINE_CHECK, [W]=WARN Hardware name: motorola Motorola Edge 30 (Tianma)/Motorola Edge 30 (Tianma), BIOS 2026.07-gca7686425f32-dirty 07/01/2026 pstate: 60400005 (nZCv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--) pc : clear_page+0x30/0x70 lr : prep_new_page+0x94/0x178 ... Call trace:  clear_page+0x30/0x70 (P)  get_page_from_freelist+0x1bd8/0x1c00  __alloc_frozen_pages_noprof+0x1cc/0x3f8  alloc_pages_mpol+0x124/0x210  vma_alloc_folio_noprof+0x124/0x200  vma_alloc_zeroed_movable_folio+0x3c/0x50  do_pte_missing+0x9e0/0xf80  ... This no-map does seem to fix it. Thanks, ~val