From nobody Fri Oct 3 08:49:05 2025 Received: from smtp.kernel.org (aws-us-west-2-korg-mail-1.web.codeaurora.org [10.30.226.201]) (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 AD7471A5B9E for ; Wed, 3 Sep 2025 06:30:29 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=10.30.226.201 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1756881029; cv=none; b=Ftd9b/a0ivw/y/NGMIvlGoAowEE2dkumpm5ww/ji2czijBykB9waH87/Se5w2IbO5r1oWbbEJY7n/Kc4MODoirzsxk3sFGZYA0ihDpzbGmF6woGXvB0DAdBdpd3UTy7Wa1zVuKZgtfAzzcw/o+Qe49LF2kgHVMOH6NxHcanpxKk= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1756881029; c=relaxed/simple; bh=71VhAKjgW0E1aO+puSjE7uBqc/1v+0Vas8jZUkLzLuI=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=AJL2guHgH1dMcU05qL1KdKeQAojeFTFEAdR2HgGKV4gQUnnrSAxq/HpvKxrxonMQjBN8OZjE6ZtIuwZcQyUA30Uz/bnAXQe0E78qJ++aqGAE5Dc6SNsCfN2oYIVeJH2T9XiHl3LazsBVOV28M/DAOlFLYK29SgeVTkARo8cxLyw= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b=K12YYfWJ; arc=none smtp.client-ip=10.30.226.201 Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b="K12YYfWJ" Received: by smtp.kernel.org (Postfix) with ESMTPSA id 1168CC4CEF1; Wed, 3 Sep 2025 06:30:25 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=kernel.org; s=k20201202; t=1756881029; bh=71VhAKjgW0E1aO+puSjE7uBqc/1v+0Vas8jZUkLzLuI=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=K12YYfWJlksaS+EWHjzH/MCzJZAzvwF42oeLtOIsRgk4D+3RztSI8WhbMfCjqJxmR eplt3hMq0MVq+NIojWepYjQWaRZCMARGfsRSr0854+nCxMqRFbm6F9rDv/cbfQSrqT ys8u6+IBxtZD8sD9pLU/HvRfwzXwsf5Pr036ogTFzG8vV+vA0QwH+iQhpi5oUxZLC5 7sGTLV9Bj+sgMH/SG8aLnDrKZXk2f/NvNptrTHkZjXgEawFmwf/cwC/jR7DIft+I9F MBfH3Jt3v/VWqh1CkKzm1zgoqSXV8eDEb4vhx9rwaWivag2I3VH47wVc3MPjRYlRZa 7wbfsptG8q5MQ== From: Mike Rapoport To: Andrew Morton Cc: Alexander Graf , Baoquan He , Changyuan Lyu , Chris Li , Jason Gunthorpe , Mike Rapoport , Pasha Tatashin , Pratyush Yadav , kexec@lists.infradead.org, linux-mm@kvack.org, linux-kernel@vger.kernel.org Subject: [PATCH 1/2] kho: add support for preserving vmalloc allocations Date: Wed, 3 Sep 2025 09:30:17 +0300 Message-ID: <20250903063018.3346652-2-rppt@kernel.org> X-Mailer: git-send-email 2.50.1 In-Reply-To: <20250903063018.3346652-1-rppt@kernel.org> References: <20250903063018.3346652-1-rppt@kernel.org> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" From: "Mike Rapoport (Microsoft)" A vmalloc allocation is preserved using binary structure similar to global KHO memory tracker. It's a linked list of pages where each page is an array of physical address of pages in vmalloc area. kho_preserve_vmalloc() hands out the physical address of the head page to the caller. This address is used as the argument to kho_vmalloc_restore() to restore the mapping in the vmalloc address space and populate it with the preserved pages. Signed-off-by: Mike Rapoport (Microsoft) --- include/linux/kexec_handover.h | 12 +++ kernel/kexec_handover.c | 140 +++++++++++++++++++++++++++++++++ 2 files changed, 152 insertions(+) diff --git a/include/linux/kexec_handover.h b/include/linux/kexec_handover.h index 348844cffb13..b7bf3bf11019 100644 --- a/include/linux/kexec_handover.h +++ b/include/linux/kexec_handover.h @@ -42,8 +42,10 @@ struct kho_serialization; bool kho_is_enabled(void); =20 int kho_preserve_folio(struct folio *folio); +int kho_preserve_vmalloc(void *ptr, phys_addr_t *preservation); int kho_preserve_phys(phys_addr_t phys, size_t size); struct folio *kho_restore_folio(phys_addr_t phys); +void *kho_restore_vmalloc(phys_addr_t preservation); int kho_add_subtree(struct kho_serialization *ser, const char *name, void = *fdt); int kho_retrieve_subtree(const char *name, phys_addr_t *phys); =20 @@ -70,11 +72,21 @@ static inline int kho_preserve_phys(phys_addr_t phys, s= ize_t size) return -EOPNOTSUPP; } =20 +static inline int kho_preserve_vmalloc(void *ptr, phys_addr_t *preservatio= n) +{ + return -EOPNOTSUPP; +} + static inline struct folio *kho_restore_folio(phys_addr_t phys) { return NULL; } =20 +static inline void *kho_restore_vmalloc(phys_addr_t preservation) +{ + return NULL; +} + static inline int kho_add_subtree(struct kho_serialization *ser, const char *name, void *fdt) { diff --git a/kernel/kexec_handover.c b/kernel/kexec_handover.c index ecd1ac210dbd..a11ae79d6bc9 100644 --- a/kernel/kexec_handover.c +++ b/kernel/kexec_handover.c @@ -18,6 +18,7 @@ #include #include #include +#include =20 #include =20 @@ -733,6 +734,145 @@ int kho_preserve_phys(phys_addr_t phys, size_t size) } EXPORT_SYMBOL_GPL(kho_preserve_phys); =20 +struct kho_vmalloc_chunk; + +struct kho_vmalloc_hdr { + DECLARE_KHOSER_PTR(next, struct kho_vmalloc_chunk *); + unsigned int total_pages; /* only valid in the first chunk */ + unsigned int num_elms; +}; + +#define KHO_VMALLOC_SIZE \ + ((PAGE_SIZE - sizeof(struct kho_vmalloc_hdr)) / \ + sizeof(phys_addr_t)) + +struct kho_vmalloc_chunk { + struct kho_vmalloc_hdr hdr; + phys_addr_t phys[KHO_VMALLOC_SIZE]; +}; + +static_assert(sizeof(struct kho_vmalloc_chunk) =3D=3D PAGE_SIZE); + +static struct kho_vmalloc_chunk *new_vmalloc_chunk(struct kho_vmalloc_chun= k *cur) +{ + struct kho_vmalloc_chunk *chunk; + int err; + + chunk =3D kzalloc(PAGE_SIZE, GFP_KERNEL); + if (!chunk) + return NULL; + + err =3D kho_preserve_phys(virt_to_phys(chunk), PAGE_SIZE); + if (err) + goto err_free; + if (cur) + KHOSER_STORE_PTR(cur->hdr.next, chunk); + return chunk; + +err_free: + kfree(chunk); + return NULL; +} + +static void kho_vmalloc_free_chunks(struct kho_vmalloc_chunk *first_chunk) +{ + struct kho_mem_track *track =3D &kho_out.ser.track; + struct kho_vmalloc_chunk *chunk =3D first_chunk; + + while (chunk) { + unsigned long pfn =3D PHYS_PFN(virt_to_phys(chunk)); + struct kho_vmalloc_chunk *tmp =3D chunk; + + __kho_unpreserve(track, pfn, pfn + 1); + + chunk =3D KHOSER_LOAD_PTR(chunk->hdr.next); + kfree(tmp); + } +} + +int kho_preserve_vmalloc(void *ptr, phys_addr_t *preservation) +{ + struct kho_vmalloc_chunk *chunk, *first_chunk; + struct vm_struct *vm =3D find_vm_area(ptr); + int err; + + if (!vm) + return -EINVAL; + + /* we don't support HUGE_VMAP yet */ + if (get_vm_area_page_order(vm)) + return -EOPNOTSUPP; + + chunk =3D new_vmalloc_chunk(NULL); + if (!chunk) + return -ENOMEM; + first_chunk =3D chunk; + first_chunk->hdr.total_pages =3D vm->nr_pages; + + for (int i =3D 0; i < vm->nr_pages; i++) { + phys_addr_t phys =3D page_to_phys(vm->pages[i]); + + err =3D kho_preserve_phys(phys, PAGE_SIZE); + if (err) + goto err_free; + + chunk->phys[chunk->hdr.num_elms] =3D phys; + chunk->hdr.num_elms++; + if (chunk->hdr.num_elms =3D=3D ARRAY_SIZE(chunk->phys)) { + chunk =3D new_vmalloc_chunk(chunk); + if (!chunk) + goto err_free; + } + } + + *preservation =3D virt_to_phys(first_chunk); + return 0; + +err_free: + kho_vmalloc_free_chunks(first_chunk); + return err; +} +EXPORT_SYMBOL_GPL(kho_preserve_vmalloc); + +void *kho_restore_vmalloc(phys_addr_t preservation) +{ + struct kho_vmalloc_chunk *chunk =3D phys_to_virt(preservation); + unsigned int idx =3D 0, nr =3D 0; + struct page **pages; + void *ptr; + + nr =3D chunk->hdr.total_pages; + pages =3D kvmalloc_array(nr, sizeof(*pages), GFP_KERNEL); + if (!pages) + return NULL; + + while (chunk) { + struct page *page; + + for (int i =3D 0; i < chunk->hdr.num_elms; i++) { + page =3D phys_to_page(chunk->phys[i]); + kho_restore_page(page, 0); + pages[idx++] =3D page; + } + + page =3D virt_to_page(chunk); + chunk =3D KHOSER_LOAD_PTR(chunk->hdr.next); + kho_restore_page(page, 0); + __free_page(page); + } + + ptr =3D vmap(pages, nr, VM_MAP_PUT_PAGES, PAGE_KERNEL); + if (!ptr) + goto err_free_pages_array; + + return ptr; + +err_free_pages_array: + kvfree(pages); + return NULL; +} +EXPORT_SYMBOL_GPL(kho_restore_vmalloc); + /* Handling for debug/kho/out */ =20 static struct dentry *debugfs_root; --=20 2.50.1 From nobody Fri Oct 3 08:49:05 2025 Received: from smtp.kernel.org (aws-us-west-2-korg-mail-1.web.codeaurora.org [10.30.226.201]) (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 629331A5B9E for ; Wed, 3 Sep 2025 06:30:33 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=10.30.226.201 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1756881034; cv=none; b=fbMyskqmnN/RRtIka3pRxOUkyTcKICWk4yHSETGNtEAKrUSRfg1tc3iuLG8tjqNvKERI7pSd+m/IxFfofS8xglgvjviPzbE5SlcBWKnApy30LK3zD6BnSXVY9xsgfdrxELcV6EgO25rwe8VlZweLZj+aC7BXe285nzSXtcjzt70= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1756881034; c=relaxed/simple; bh=C3NlxNYz6bfCABhEIee3Vq9GbnEKAONZKNFONpPiN3c=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=BOPscpSiJFdWSjJkS4zRyuBlRLcx9JVLE4MUyQpqeVQPuH42MuXMYAK6jlbcTR7mYrH67EbFeBw1EVxlM3U5Iw4FTtOOG7zY4xfATSJGNyPCdA4XC/PNMC6wQxCaZkXtxHDiAjgRAuNBE3kks9e/CAJ4QFdsoWa6Bk1wj42eoNo= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b=aC+N1oXw; arc=none smtp.client-ip=10.30.226.201 Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b="aC+N1oXw" Received: by smtp.kernel.org (Postfix) with ESMTPSA id B792CC4CEF0; Wed, 3 Sep 2025 06:30:29 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=kernel.org; s=k20201202; t=1756881032; bh=C3NlxNYz6bfCABhEIee3Vq9GbnEKAONZKNFONpPiN3c=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=aC+N1oXwxBgKXwGUcqK/EB5vl94l2OaSNEWh5MVcKe9/E3ZEaR3Il3xvniELhXC/u QEjwv5BvLPs/0LpPZgTeF9956fEiZqkVkJH1CteKyNspczWHXrTsgbwdSqpVSW7y3a bjV4uwAzs5j3qtLiKFRj36ghUYj4cbpmvKNvJwjfPg3Gc2t600hWq/Zw42Zd6u1Vow gPAObDgd8sIX6fQ95Fa2bXboBlxUlAUAR/q0sZ1FN088T0nnV3BFvMMv1XeM4kKNfR DkTEZ0nX/vogWZzUXpbt7gdX7X7Hlm9WfJSpdtksE5qf5hUiDk92XCB/enYUB7ESzO jtNDZKFvGUnIw== From: Mike Rapoport To: Andrew Morton Cc: Alexander Graf , Baoquan He , Changyuan Lyu , Chris Li , Jason Gunthorpe , Mike Rapoport , Pasha Tatashin , Pratyush Yadav , kexec@lists.infradead.org, linux-mm@kvack.org, linux-kernel@vger.kernel.org Subject: [PATCH 2/2] lib/test_kho: use kho_preserve_vmalloc instead of storing addresses in fdt Date: Wed, 3 Sep 2025 09:30:18 +0300 Message-ID: <20250903063018.3346652-3-rppt@kernel.org> X-Mailer: git-send-email 2.50.1 In-Reply-To: <20250903063018.3346652-1-rppt@kernel.org> References: <20250903063018.3346652-1-rppt@kernel.org> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" From: "Mike Rapoport (Microsoft)" KHO test stores physical addresses of the preserved folios directly in fdt. Use kho_preserve_vmalloc() instead of it and kho_restore_vmalloc() to retrieve the addresses after kexec. This makes the test more scalable from one side and adds tests coverage for kho_preserve_vmalloc() from the other. Signed-off-by: Mike Rapoport (Microsoft) --- lib/test_kho.c | 30 +++++++++++++++++++++++------- 1 file changed, 23 insertions(+), 7 deletions(-) diff --git a/lib/test_kho.c b/lib/test_kho.c index c2eb899c3b45..10045f5979a0 100644 --- a/lib/test_kho.c +++ b/lib/test_kho.c @@ -32,6 +32,7 @@ module_param(max_mem, long, 0644); struct kho_test_state { unsigned int nr_folios; struct folio **folios; + phys_addr_t *folios_info; struct folio *fdt; __wsum csum; }; @@ -68,13 +69,17 @@ static struct notifier_block kho_test_nb =3D { static int kho_test_save_data(struct kho_test_state *state, void *fdt) { phys_addr_t *folios_info __free(kvfree) =3D NULL; + phys_addr_t folios_info_phys; int err =3D 0; =20 - folios_info =3D kvmalloc_array(state->nr_folios, sizeof(*folios_info), - GFP_KERNEL); + folios_info =3D vmalloc_array(state->nr_folios, sizeof(*folios_info)); if (!folios_info) return -ENOMEM; =20 + err =3D kho_preserve_vmalloc(folios_info, &folios_info_phys); + if (err) + return err; + for (int i =3D 0; i < state->nr_folios; i++) { struct folio *folio =3D state->folios[i]; unsigned int order =3D folio_order(folio); @@ -89,11 +94,14 @@ static int kho_test_save_data(struct kho_test_state *st= ate, void *fdt) err |=3D fdt_begin_node(fdt, "data"); err |=3D fdt_property(fdt, "nr_folios", &state->nr_folios, sizeof(state->nr_folios)); - err |=3D fdt_property(fdt, "folios_info", folios_info, - state->nr_folios * sizeof(*folios_info)); + err |=3D fdt_property(fdt, "folios_info", &folios_info_phys, + sizeof(folios_info_phys)); err |=3D fdt_property(fdt, "csum", &state->csum, sizeof(state->csum)); err |=3D fdt_end_node(fdt); =20 + if (!err) + state->folios_info =3D no_free_ptr(folios_info); + return err; } =20 @@ -197,7 +205,8 @@ static int kho_test_save(void) static int kho_test_restore_data(const void *fdt, int node) { const unsigned int *nr_folios; - const phys_addr_t *folios_info; + const phys_addr_t *folios_info_phys; + phys_addr_t *folios_info; const __wsum *old_csum; __wsum csum =3D 0; int len; @@ -212,8 +221,12 @@ static int kho_test_restore_data(const void *fdt, int = node) if (!old_csum || len !=3D sizeof(*old_csum)) return -EINVAL; =20 - folios_info =3D fdt_getprop(fdt, node, "folios_info", &len); - if (!folios_info || len !=3D sizeof(*folios_info) * *nr_folios) + folios_info_phys =3D fdt_getprop(fdt, node, "folios_info", &len); + if (!folios_info_phys || len !=3D sizeof(*folios_info_phys)) + return -EINVAL; + + folios_info =3D kho_restore_vmalloc(*folios_info_phys); + if (!folios_info) return -EINVAL; =20 for (int i =3D 0; i < *nr_folios; i++) { @@ -233,6 +246,8 @@ static int kho_test_restore_data(const void *fdt, int n= ode) folio_put(folio); } =20 + vfree(folios_info); + if (csum !=3D *old_csum) return -EINVAL; =20 @@ -291,6 +306,7 @@ static void kho_test_cleanup(void) folio_put(kho_test_state.folios[i]); =20 kvfree(kho_test_state.folios); + vfree(kho_test_state.folios_info); } =20 static void __exit kho_test_exit(void) --=20 2.50.1