From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from smtp.kernel.org (aws-us-west-2-korg-mail-alma10-1.taild15c8.ts.net [100.103.45.18]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 5AEA72EA468 for ; Wed, 15 Jul 2026 02:17:58 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=100.103.45.18 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1784081882; cv=none; b=ne9Zh9mQI8/bigQrULSnTL2hu0i3sRin7mhJRIA6s9vcv3Vm2lB8YzCOCqXELScAYKNS2krLqN7JNgPx+jxRnJUPIc7STQNc0Wh4M95CeVs5C8CUDdac6HkO9k/nZhCOe/mAfq9ydjV2nPWz3wEuDTElrKwVjVldAsp9mV/3AYE= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1784081882; c=relaxed/simple; bh=TjSGEkiguLi6oGg/b/qvtoM7TSMoC2TK3Crx6qZ7vWE=; h=Date:To:From:Subject:Message-Id; b=OoktOWKz8KMhulWjtDxBQtaZAeOnhr3+3SZCatTwhjMsovmpBiy6zm/4cF9dKdf+H/md2RWGB2oWJDHxYUm1OipextdEf2uYrxC2Vowm2BMdrT74d+g6ctpi1WCvNRGbwk6JVHf4sQtDusjil/SsCxTNEPbPgZ/dNieSVu8pRyc= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=linux-foundation.org header.i=@linux-foundation.org header.b=dRdNagGH; arc=none smtp.client-ip=100.103.45.18 Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=linux-foundation.org header.i=@linux-foundation.org header.b="dRdNagGH" Received: by smtp.kernel.org (Postfix) with ESMTPSA id DED441F000E9; Wed, 15 Jul 2026 02:17:57 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=linux-foundation.org; s=korg; t=1784081878; bh=cnhIPnnxjrWSsYjB7sUvL+66uiVYQPEQnW8nf6TXMqA=; h=Date:To:From:Subject; b=dRdNagGHHyqGsK4XAGdOQHJUbItg8m3ZM6T1TZ84/F3YonnBt8NxfD/dFpEx6jsto uBjfRNmozmNva+HI4YVCDJT/6UZypYnGtTpPne7VolkuEFn1WVIyMFqvqhWdURHpyQ UJCjo8V2PDsubTUK7zx90pcOPyBWezFWGww0AuSQ= Date: Tue, 14 Jul 2026 19:17:57 -0700 To: mm-commits@vger.kernel.org,wei.liu@kernel.org,vbabka@kernel.org,tzimmermann@suse.de,surenb@google.com,shuah@kernel.org,rppt@kernel.org,oleg@redhat.com,ogabbay@kernel.org,mripard@kernel.org,mhocko@suse.com,maarten.lankhorst@linux.intel.com,lyude@redhat.com,longli@microsoft.com,ljs@kernel.org,lizhi.hou@amd.com,liam@infradead.org,leon@kernel.org,kys@microsoft.com,kees@kernel.org,jgg@nvidia.com,haiyangz@microsoft.com,decui@microsoft.com,david@kernel.org,dakr@kernel.org,corbet@lwn.net,airlied@gmail.com,skinsburskii@gmail.com,akpm@linux-foundation.org From: Andrew Morton Subject: + drm-nouveau-use-hmm_range_fault_unlocked_timeout-for-svm-faults-fix.patch added to mm-unstable branch Message-Id: <20260715021757.DED441F000E9@smtp.kernel.org> Precedence: bulk X-Mailing-List: mm-commits@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: The patch titled Subject: drm-nouveau-use-hmm_range_fault_unlocked_timeout-for-svm-faults-fix has been added to the -mm mm-unstable branch. Its filename is drm-nouveau-use-hmm_range_fault_unlocked_timeout-for-svm-faults-fix.patch This patch will shortly appear at https://git.kernel.org/pub/scm/linux/kernel/git/akpm/25-new.git/tree/patches/drm-nouveau-use-hmm_range_fault_unlocked_timeout-for-svm-faults-fix.patch This patch will later appear in the mm-unstable branch at git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm Before you just go and hit "reply", please: a) Consider who else should be cc'ed b) Prefer to cc a suitable mailing list as well c) Ideally: find the original patch on the mailing list and do a reply-to-all to that, adding suitable additional cc's *** Remember to use Documentation/process/submit-checklist.rst when testing your code *** The -mm tree is included into linux-next via various branches at git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm and is updated there most days ------------------------------------------------------ From: Stanislav Kinsburskii Subject: drm-nouveau-use-hmm_range_fault_unlocked_timeout-for-svm-faults-fix Date: Tue, 14 Jul 2026 15:21:35 -0700 nouveau_range_fault() now uses hmm_range_fault_unlocked_timeout() for the HMM fault path. Pass the remaining SVM fault timeout into HMM as a positive relative timeout instead of open-coding the HMM retry loop in nouveau. Keep the outer absolute timeout around nouveau's mmu_interval_read_retry() loop. HMM's timeout only bounds retries while HMM is walking and faulting the range. Once HMM returns successfully, a continuous stream of mmu-notifier invalidations can still make mmu_interval_read_retry() restart the operation. This path runs from the GPU fault worker, so fatal signals for the faulting userspace task cannot be relied on to break that outer loop. Use time_after_eq() before the HMM call so the remaining timeout passed to HMM is always positive. This preserves the old timeout behavior at the expiry boundary and avoids passing 0, which means retry indefinitely, to hmm_range_fault_unlocked_timeout(). Link: https://lore.kernel.org/178406769581.1106335.2090301352195900623.stgit@skinsburskii Signed-off-by: Stanislav Kinsburskii Cc: Danilo Krummrich Cc: Dave Airlie Cc: David Hildenbrand Cc: Dexuan Cui Cc: Haiyang Zhang Cc: Jason Gunthorpe Cc: Jonathan Corbet Cc: Kees Cook Cc: K. Y. Srinivasan Cc: Leon Romanovsky Cc: Liam R. Howlett Cc: Lizhi Hou Cc: Long Li Cc: Lorenzo Stoakes Cc: Lyude Cc: Maarten Lankhorst Cc: Maxime Ripard Cc: Michal Hocko Cc: Mike Rapoport Cc: Oded Gabbay Cc: Oleg Nesterov Cc: Shuah Khan Cc: Suren Baghdasaryan Cc: Thomas Zimemrmann Cc: Vlastimil Babka Cc: Wei Liu Signed-off-by: Andrew Morton --- drivers/gpu/drm/nouveau/nouveau_svm.c | 14 ++++++++++---- 1 file changed, 10 insertions(+), 4 deletions(-) --- a/drivers/gpu/drm/nouveau/nouveau_svm.c~drm-nouveau-use-hmm_range_fault_unlocked_timeout-for-svm-faults-fix +++ a/drivers/gpu/drm/nouveau/nouveau_svm.c @@ -678,14 +678,20 @@ static int nouveau_range_fault(struct no range.end = notifier->notifier.interval_tree.last + 1; while (true) { - if (time_after(jiffies, timeout)) { + long remaining = timeout - jiffies; + + /* + * The HMM timeout only bounds retries while HMM is walking and + * faulting the range. This fault is handled by a kernel worker, + * so fatal signals from the faulting process cannot stop an + * endless stream of invalidations here. + */ + if (time_after_eq(jiffies, timeout)) { ret = -EBUSY; goto out; } - ret = hmm_range_fault_unlocked_timeout(&range, - max(timeout - jiffies, - 1L)); + ret = hmm_range_fault_unlocked_timeout(&range, remaining); if (ret) goto out; _ Patches currently in -mm which might be from skinsburskii@gmail.com are lib-test_hmm-use-device-devt-for-coherent-device-range-selection.patch mm-hmm-move-page-fault-handling-out-of-walk-callbacks.patch mm-hmm-add-hmm_range_fault_unlocked_timeout-for-mmap-lock-drop-support.patch mm-hmm-add-hmm_range_fault_unlocked_timeout-for-mmap-lock-drop-support-fix.patch selftests-mm-add-hmm-test-for-mmap-lock-dropping-faults.patch mshv-use-hmm_range_fault_unlocked_timeout-for-region-faults.patch drm-nouveau-use-hmm_range_fault_unlocked_timeout-for-svm-faults.patch drm-nouveau-use-hmm_range_fault_unlocked_timeout-for-svm-faults-fix.patch rdma-umem-use-hmm_range_fault_unlocked_timeout-for-odp-faults.patch accel-amdxdna-use-hmm_range_fault_unlocked_timeout-for-range-population.patch accel-amdxdna-use-hmm_range_fault_unlocked_timeout-for-range-population-fix.patch drm-gpusvm-use-hmm_range_fault_unlocked_timeout-for-range-faults.patch drm-gpusvm-use-hmm_range_fault_unlocked_timeout-for-range-faults-fix.patch