| Message ID | 20201204195149.611-1-nramas@linux.microsoft.com |
|---|---|
| Headers | show |
| Series | Carry forward IMA measurement log on kexec on ARM64 | expand |
Lakshmi Ramasubramanian <nramas@linux.microsoft.com> writes: > The function prototype for the functions defined in ima.c for powerpc > are given in the header file ima.h. But this header file is not > included in ima.c resulting in compilation errors such as given below. > > arch/powerpc/kexec/ima.c:56:5: error: no previous prototype for 'ima_get_kexec_buffer' [-Werror=missing-prototypes] > 56 | int ima_get_kexec_buffer(void **addr, size_t *size) > | ^~~~~~~~~~~~~~~~~~~~ > > The function parameters for remove_ima_buffer() and > arch_ima_add_kexec_buffer() are not described in the function header > resulting in warnings such as given below. > > arch/powerpc/kexec/ima.c:111: warning: Function parameter or member 'fdt' not described in 'remove_ima_buffer' > > Include ima.h in ima.c for powerpc. Describe the function parameters for > remove_ima_buffer() and arch_ima_add_kexec_buffer(). > > Co-developed-by: Prakhar Srivastava <prsriva@linux.microsoft.com> > Signed-off-by: Prakhar Srivastava <prsriva@linux.microsoft.com> > Signed-off-by: Lakshmi Ramasubramanian <nramas@linux.microsoft.com> > Reviewed-by: Mimi Zohar <zohar@linux.ibm.com> These warnings showed up when using `make W=1`, and this patch fixes them. Thanks! Reviewed-by: Thiago Jung Bauermann <bauerman@linux.ibm.com> -- Thiago Jung Bauermann IBM Linux Technology Center
Hello Lakshmi, Lakshmi Ramasubramanian <nramas@linux.microsoft.com> writes: > delete_fdt_mem_rsv() retrieves the memory reserve map entry, for > the given starting address and size, from the device tree blob, and > removes the entry from the device tree blob. This function is called > to free the resources reserved for the buffer used for carrying forward > the IMA measurement logs on kexec. This function does not have > architecture specific code, but is currently limited to powerpc. > > Move delete_fdt_mem_rsv() to "drivers/of/kexec_fdt.c" so that it is s/kexec_fdt.c/kexec.c/ > accessible for other architectures as well. > > Co-developed-by: Prakhar Srivastava <prsriva@linux.microsoft.com> > Signed-off-by: Prakhar Srivastava <prsriva@linux.microsoft.com> > Signed-off-by: Lakshmi Ramasubramanian <nramas@linux.microsoft.com> > --- > arch/powerpc/include/asm/kexec.h | 1 - > arch/powerpc/kexec/file_load.c | 32 ----------------- > drivers/of/Makefile | 1 + > drivers/of/kexec.c | 61 ++++++++++++++++++++++++++++++++ > include/linux/kexec.h | 5 +++ > 5 files changed, 67 insertions(+), 33 deletions(-) > create mode 100644 drivers/of/kexec.c > > diff --git a/arch/powerpc/include/asm/kexec.h b/arch/powerpc/include/asm/kexec.h > index 55d6ede30c19..7c223031ecdd 100644 > --- a/arch/powerpc/include/asm/kexec.h > +++ b/arch/powerpc/include/asm/kexec.h > @@ -126,7 +126,6 @@ int setup_purgatory(struct kimage *image, const void *slave_code, > int setup_new_fdt(const struct kimage *image, void *fdt, > unsigned long initrd_load_addr, unsigned long initrd_len, > const char *cmdline); > -int delete_fdt_mem_rsv(void *fdt, unsigned long start, unsigned long size); > > #ifdef CONFIG_PPC64 > struct kexec_buf; > diff --git a/arch/powerpc/kexec/file_load.c b/arch/powerpc/kexec/file_load.c > index 9a232bc36c8f..9efc98b1e2ae 100644 > --- a/arch/powerpc/kexec/file_load.c > +++ b/arch/powerpc/kexec/file_load.c > @@ -109,38 +109,6 @@ int setup_purgatory(struct kimage *image, const void *slave_code, > return 0; > } > > -/** > - * delete_fdt_mem_rsv - delete memory reservation with given address and size > - * > - * Return: 0 on success, or negative errno on error. > - */ > -int delete_fdt_mem_rsv(void *fdt, unsigned long start, unsigned long size) > -{ > - int i, ret, num_rsvs = fdt_num_mem_rsv(fdt); > - > - for (i = 0; i < num_rsvs; i++) { > - uint64_t rsv_start, rsv_size; > - > - ret = fdt_get_mem_rsv(fdt, i, &rsv_start, &rsv_size); > - if (ret) { > - pr_err("Malformed device tree.\n"); > - return -EINVAL; > - } > - > - if (rsv_start == start && rsv_size == size) { > - ret = fdt_del_mem_rsv(fdt, i); > - if (ret) { > - pr_err("Error deleting device tree reservation.\n"); > - return -EINVAL; > - } > - > - return 0; > - } > - } > - > - return -ENOENT; > -} > - > /* > * setup_new_fdt - modify /chosen and memory reservation for the next kernel > * @image: kexec image being loaded. > diff --git a/drivers/of/Makefile b/drivers/of/Makefile > index 6e1e5212f058..77d24712c0c8 100644 > --- a/drivers/of/Makefile > +++ b/drivers/of/Makefile > @@ -13,5 +13,6 @@ obj-$(CONFIG_OF_RESERVED_MEM) += of_reserved_mem.o > obj-$(CONFIG_OF_RESOLVE) += resolver.o > obj-$(CONFIG_OF_OVERLAY) += overlay.o > obj-$(CONFIG_OF_NUMA) += of_numa.o > +obj-$(CONFIG_OF_FLATTREE) += kexec.o Isn't this too broad? kexec.o will only be useful to kernel configs which enable CONFIG_KEXEC_FILE, so perhaps do: ifdef CONFIG_OF_FLATTREE ifdef CONFIG_KEXEC_FILE obj-y += kexec.o endif endif What do you think? > obj-$(CONFIG_OF_UNITTEST) += unittest-data/ > diff --git a/drivers/of/kexec.c b/drivers/of/kexec.c > new file mode 100644 > index 000000000000..b7d59105fcb8 > --- /dev/null > +++ b/drivers/of/kexec.c > @@ -0,0 +1,61 @@ > +// SPDX-License-Identifier: GPL-2.0+ > +/* > + * Copyright (C) 2020 Microsoft Corporation > + * > + * Author: Lakshmi Ramasubramanian (nramas@linux.microsoft.com) > + * > + * File: kexec.c > + * Defines kexec related functions. > + */ > + > +#define pr_fmt(fmt) "OF: kexec: " fmt > + > +#include <linux/kernel.h> > +#include <linux/slab.h> > +#include <linux/memblock.h> > +#include <linux/kexec.h> > +#include <linux/of.h> > +#include <linux/of_fdt.h> > +#include <linux/libfdt.h> > + > +/** > + * delete_fdt_mem_rsv - delete memory reservation with given address and size > + * > + * @fdt: Flattened Device Tree to update > + * @start: Starting address of the reservation to delete > + * @size: Size of the reservation to delete > + * > + * Return: 0 on success, or negative errno on error. > + */ > +int delete_fdt_mem_rsv(void *fdt, unsigned long start, unsigned long size) > +{ > + int i, ret, num_rsvs; > + > + if (!IS_ENABLED(CONFIG_KEXEC_FILE)) > + return 0; If you agree with my suggestion, then this IS_ENABLED() wouldn't be needed. > + > + num_rsvs = fdt_num_mem_rsv(fdt); > + for (i = 0; i < num_rsvs; i++) { > + uint64_t rsv_start, rsv_size; > + > + ret = fdt_get_mem_rsv(fdt, i, &rsv_start, &rsv_size); > + if (ret) { > + pr_err("Malformed device tree.\n"); > + return -EINVAL; > + } > + > + if (rsv_start == start && rsv_size == size) { > + ret = fdt_del_mem_rsv(fdt, i); > + if (ret) { > + pr_err("Error deleting device tree reservation.\n"); > + return -EINVAL; > + } > + > + pr_debug("Freed reserved memory at %lu of size %lu\n", > + start, size); > + return 0; > + } > + } > + > + return -ENOENT; > +} The function code is unchanged apart from the addition of the IS_ENABLED() and the pr_debug(), so that is good. > diff --git a/include/linux/kexec.h b/include/linux/kexec.h > index 9e93bef52968..d0234c4815da 100644 > --- a/include/linux/kexec.h > +++ b/include/linux/kexec.h > @@ -407,6 +407,11 @@ static inline int kexec_crash_loaded(void) { return 0; } > #define kexec_in_progress false > #endif /* CONFIG_KEXEC_CORE */ > > +#if defined(CONFIG_OF_FLATTREE) This would also change to require CONFIG_KEXEC_FILE. > +extern int delete_fdt_mem_rsv(void *fdt, unsigned long start, > + unsigned long size); > +#endif /* CONFIG_OF_FLATTREE */ > + > #endif /* !defined(__ASSEBMLY__) */ > > #endif /* LINUX_KEXEC_H */ -- Thiago Jung Bauermann IBM Linux Technology Center
Lakshmi Ramasubramanian <nramas@linux.microsoft.com> writes: > do_get_kexec_buffer() and get_addr_size_cells() are called by > ima_get_kexec_buffer(), ima_free_kexec_buffer, and remove_ima_buffer() > to retrieve the address and size of the buffer used for carrying > forward the IMA measurement log across kexec system call. These > functions correctly handle a device tree property that is a child node > of the root node, but not anything other than the immediate root > child nodes. > > Use the architecture independent functions get_ima_kexec_buffer() > and get_root_addr_size_cells() defined in "drivers/of/ima_kexec.c", s/ima_kexec.c/kexec.c/ > to get the address and size of the IMA measurement log buffer from > the device tree. Remove do_get_kexec_buffer() and get_addr_size_cells() > since they are not used anymore. > > Co-developed-by: Prakhar Srivastava <prsriva@linux.microsoft.com> > Signed-off-by: Prakhar Srivastava <prsriva@linux.microsoft.com> > Signed-off-by: Lakshmi Ramasubramanian <nramas@linux.microsoft.com> Looks good. Thanks! Reviewed-by: Thiago Jung Bauermann <bauerman@linux.ibm.com> > --- > arch/powerpc/kexec/ima.c | 58 +++++----------------------------------- > 1 file changed, 7 insertions(+), 51 deletions(-) > > diff --git a/arch/powerpc/kexec/ima.c b/arch/powerpc/kexec/ima.c > index a36c39db4b1a..906e8212435d 100644 > --- a/arch/powerpc/kexec/ima.c > +++ b/arch/powerpc/kexec/ima.c > @@ -13,40 +13,6 @@ > #include <linux/libfdt.h> > #include <asm/ima.h> > > -static int get_addr_size_cells(int *addr_cells, int *size_cells) > -{ > - struct device_node *root; > - > - root = of_find_node_by_path("/"); > - if (!root) > - return -EINVAL; > - > - *addr_cells = of_n_addr_cells(root); > - *size_cells = of_n_size_cells(root); > - > - of_node_put(root); > - > - return 0; > -} > - > -static int do_get_kexec_buffer(const void *prop, int len, unsigned long *addr, > - size_t *size) > -{ > - int ret, addr_cells, size_cells; > - > - ret = get_addr_size_cells(&addr_cells, &size_cells); > - if (ret) > - return ret; > - > - if (len < 4 * (addr_cells + size_cells)) > - return -ENOENT; > - > - *addr = of_read_number(prop, addr_cells); > - *size = of_read_number(prop + 4 * addr_cells, size_cells); > - > - return 0; > -} > - > /** > * ima_get_kexec_buffer - get IMA buffer from the previous kernel > * @addr: On successful return, set to point to the buffer contents. > @@ -56,16 +22,11 @@ static int do_get_kexec_buffer(const void *prop, int len, unsigned long *addr, > */ > int ima_get_kexec_buffer(void **addr, size_t *size) > { > - int ret, len; > + int ret; > unsigned long tmp_addr; > size_t tmp_size; > - const void *prop; > > - prop = of_get_property(of_chosen, "linux,ima-kexec-buffer", &len); > - if (!prop) > - return -ENOENT; > - > - ret = do_get_kexec_buffer(prop, len, &tmp_addr, &tmp_size); > + ret = get_ima_kexec_buffer(NULL, 0, &tmp_addr, &tmp_size); > if (ret) > return ret; > > @@ -89,7 +50,7 @@ int ima_free_kexec_buffer(void) > if (!prop) > return -ENOENT; > > - ret = do_get_kexec_buffer(prop->value, prop->length, &addr, &size); > + ret = get_ima_kexec_buffer(NULL, 0, &addr, &size); > if (ret) > return ret; > > @@ -98,7 +59,6 @@ int ima_free_kexec_buffer(void) > return ret; > > return memblock_free(addr, size); > - > } > > /** > @@ -112,19 +72,15 @@ int ima_free_kexec_buffer(void) > */ > void remove_ima_buffer(void *fdt, int chosen_node) > { > - int ret, len; > + int ret; > unsigned long addr; > size_t size; > - const void *prop; > > - prop = fdt_getprop(fdt, chosen_node, "linux,ima-kexec-buffer", &len); > - if (!prop) > + ret = get_ima_kexec_buffer(fdt, chosen_node, &addr, &size); > + if (ret) > return; > > - ret = do_get_kexec_buffer(prop, len, &addr, &size); > fdt_delprop(fdt, chosen_node, "linux,ima-kexec-buffer"); > - if (ret) > - return; > > ret = delete_fdt_mem_rsv(fdt, addr, size); > if (!ret) > @@ -191,7 +147,7 @@ int setup_ima_buffer(const struct kimage *image, void *fdt, int chosen_node) > if (!image->arch.ima_buffer_size) > return 0; > > - ret = get_addr_size_cells(&addr_cells, &size_cells); > + ret = get_root_addr_size_cells(&addr_cells, &size_cells); > if (ret) > return ret; -- Thiago Jung Bauermann IBM Linux Technology Center
Lakshmi Ramasubramanian <nramas@linux.microsoft.com> writes: > ima_get_kexec_buffer() retrieves the address and size of the buffer > used for carrying forward the IMA measurement logs on kexec from > the device tree. > > ima_free_kexec_buffer() removes the chosen node > "linux,ima-kexec-buffer" from the device tree, and frees the buffer > used for carrying forward the IMA measurement logs on kexec. > > These functions do not have architecture specific code, but are > currently limited to powerpc. Move ima_get_kexec_buffer() and > ima_free_kexec_buffer() to ima_kexec.c in IMA so that they are > accessible for other architectures as well. > > With the above change the functions in arch/powerpc/kexec/ima.c are > defined only when the kernel config CONFIG_IMA_KEXEC is enabled. > Update the Makefile to build arch/powerpc/kexec/ima.c only when > CONFIG_IMA_KEXEC is enabled and remove "#ifdef CONFIG_IMA_KEXEC" > in arch/powerpc/kexec/ima.c. Nice. > Co-developed-by: Prakhar Srivastava <prsriva@linux.microsoft.com> > Signed-off-by: Prakhar Srivastava <prsriva@linux.microsoft.com> > Signed-off-by: Lakshmi Ramasubramanian <nramas@linux.microsoft.com> > Reviewed-by: Mimi Zohar <zohar@linux.ibm.com> Reviewed-by: Thiago Jung Bauermann <bauerman@linux.ibm.com> Just a small nit below: > --- > arch/powerpc/include/asm/ima.h | 3 -- > arch/powerpc/kexec/Makefile | 7 +--- > arch/powerpc/kexec/ima.c | 50 ----------------------------- > security/integrity/ima/ima_kexec.c | 51 ++++++++++++++++++++++++++++++ > 4 files changed, 52 insertions(+), 59 deletions(-) > > diff --git a/arch/powerpc/include/asm/ima.h b/arch/powerpc/include/asm/ima.h > index a2fc71bc3b23..d8444d27f0d8 100644 > --- a/arch/powerpc/include/asm/ima.h > +++ b/arch/powerpc/include/asm/ima.h > @@ -6,9 +6,6 @@ > > struct kimage; > > -int ima_get_kexec_buffer(void **addr, size_t *size); > -int ima_free_kexec_buffer(void); > - > #ifdef CONFIG_IMA_KEXEC > int arch_ima_add_kexec_buffer(struct kimage *image, unsigned long load_addr, > size_t size); > diff --git a/arch/powerpc/kexec/Makefile b/arch/powerpc/kexec/Makefile > index 4aff6846c772..f54a9dbff4c8 100644 > --- a/arch/powerpc/kexec/Makefile > +++ b/arch/powerpc/kexec/Makefile > @@ -9,12 +9,7 @@ obj-$(CONFIG_PPC32) += relocate_32.o > > obj-$(CONFIG_KEXEC_FILE) += file_load.o ranges.o file_load_$(BITS).o elf_$(BITS).o > > -ifdef CONFIG_HAVE_IMA_KEXEC > -ifdef CONFIG_IMA > -obj-y += ima.o > -endif > -endif > - > +obj-$(CONFIG_IMA_KEXEC) += ima.o > > # Disable GCOV, KCOV & sanitizers in odd or sensitive code > GCOV_PROFILE_core_$(BITS).o := n > diff --git a/arch/powerpc/kexec/ima.c b/arch/powerpc/kexec/ima.c > index 68017123b07d..bf7084c0c4da 100644 > --- a/arch/powerpc/kexec/ima.c > +++ b/arch/powerpc/kexec/ima.c > @@ -13,55 +13,6 @@ > #include <linux/libfdt.h> > #include <asm/ima.h> With this patch, the following includes become unnecessary and can be removed from this file: #include <linux/of.h> #include <linux/memblock.h> > -/** > - * ima_get_kexec_buffer - get IMA buffer from the previous kernel > - * @addr: On successful return, set to point to the buffer contents. > - * @size: On successful return, set to the buffer size. > - * > - * Return: 0 on success, negative errno on error. > - */ > -int ima_get_kexec_buffer(void **addr, size_t *size) > -{ > - int ret; > - unsigned long tmp_addr; > - size_t tmp_size; > - > - ret = get_ima_kexec_buffer(NULL, 0, &tmp_addr, &tmp_size); > - if (ret) > - return ret; > - > - *addr = __va(tmp_addr); > - *size = tmp_size; > - > - return 0; > -} > - > -/** > - * ima_free_kexec_buffer - free memory used by the IMA buffer > - */ > -int ima_free_kexec_buffer(void) > -{ > - int ret; > - unsigned long addr; > - size_t size; > - struct property *prop; > - > - prop = of_find_property(of_chosen, "linux,ima-kexec-buffer", NULL); > - if (!prop) > - return -ENOENT; > - > - ret = get_ima_kexec_buffer(NULL, 0, &addr, &size); > - if (ret) > - return ret; > - > - ret = of_remove_property(of_chosen, prop); > - if (ret) > - return ret; > - > - return memblock_free(addr, size); > -} > - > -#ifdef CONFIG_IMA_KEXEC > /** > * arch_ima_add_kexec_buffer - do arch-specific steps to add the IMA buffer > * > @@ -154,4 +105,3 @@ int setup_ima_buffer(const struct kimage *image, void *fdt, int chosen_node) > > return 0; > } > -#endif /* CONFIG_IMA_KEXEC */ -- Thiago Jung Bauermann IBM Linux Technology Center
Lakshmi Ramasubramanian <nramas@linux.microsoft.com> writes: > Address and size of the buffer containing the IMA measurement log need > to be passed from the current kernel to the next kernel on kexec. > > Any existing "linux,ima-kexec-buffer" property in the device tree > needs to be removed and its corresponding memory reservation in > the currently running kernel needs to be freed. The address and > size of the current kernel's IMA measurement log need to be added > to the device tree's IMA kexec buffer node and memory for the buffer > needs to be reserved for the log to be carried over to the next kernel > on the kexec call. > > Add address and size fields to "struct kimage_arch" for ARM64 platform > to hold the address and size of the IMA measurement log buffer. Remove > any existing "linux,ima-kexec-buffer" property in the device tree and > free the corresponding memory reservation in the currently running > kernel. Add "linux,ima-kexec-buffer" property to the device tree and > reserve the memory for storing the IMA log that needs to be passed from > the current kernel to the next one. > > Update CONFIG_KEXEC_FILE to select CONFIG_HAVE_IMA_KEXEC to indicate > that the IMA measurement log information is present in the device tree > for ARM64. > > Co-developed-by: Prakhar Srivastava <prsriva@linux.microsoft.com> > Signed-off-by: Prakhar Srivastava <prsriva@linux.microsoft.com> > Signed-off-by: Lakshmi Ramasubramanian <nramas@linux.microsoft.com> Reviewed-by: Thiago Jung Bauermann <bauerman@linux.ibm.com> > --- > arch/arm64/Kconfig | 1 + > arch/arm64/include/asm/ima.h | 22 ++++++++++++++ > arch/arm64/include/asm/kexec.h | 5 ++++ > arch/arm64/kernel/Makefile | 1 + > arch/arm64/kernel/ima.c | 41 ++++++++++++++++++++++++++ > arch/arm64/kernel/machine_kexec_file.c | 8 +++++ > 6 files changed, 78 insertions(+) > create mode 100644 arch/arm64/include/asm/ima.h > create mode 100644 arch/arm64/kernel/ima.c > > diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig > index 1515f6f153a0..bcca4a467eda 100644 > --- a/arch/arm64/Kconfig > +++ b/arch/arm64/Kconfig > @@ -1094,6 +1094,7 @@ config KEXEC > config KEXEC_FILE > bool "kexec file based system call" > select KEXEC_CORE > + select HAVE_IMA_KEXEC > help > This is new version of kexec system call. This system call is > file based and takes file descriptors as system call argument > diff --git a/arch/arm64/include/asm/ima.h b/arch/arm64/include/asm/ima.h > new file mode 100644 > index 000000000000..354adc8b444c > --- /dev/null > +++ b/arch/arm64/include/asm/ima.h > @@ -0,0 +1,22 @@ > +/* SPDX-License-Identifier: GPL-2.0-or-later */ > +/* > + * Copyright (C) 2019 Microsoft Corporation Shouldn't the year here be 2020? > + * > + * Author: Prakhar Srivastava <prsriva@linux.microsoft.com> > + * > + */ > +#ifndef _ASM_ARCH_IMA_H > +#define _ASM_ARCH_IMA_H > + > +struct kimage; > + > +#ifdef CONFIG_IMA_KEXEC > +extern int setup_ima_buffer(struct kimage *image, void *dtb, int off); > +#else > +static inline int setup_ima_buffer(struct kimage *image, void *dtb, int off) > +{ > + return 0; > +} > +#endif /* CONFIG_IMA_KEXEC */ > + > +#endif /* _ASM_ARCH_IMA_H */ <snip> > diff --git a/arch/arm64/kernel/ima.c b/arch/arm64/kernel/ima.c > new file mode 100644 > index 000000000000..d65bdbe1adf6 > --- /dev/null > +++ b/arch/arm64/kernel/ima.c > @@ -0,0 +1,41 @@ > +// SPDX-License-Identifier: GPL-2.0-or-later > +/* > + * Copyright (C) 2016 IBM Corporation Same question: shouldn't the year here be 2020? > + * > + * Authors: > + * Lakshmi Ramasubramanian <nramas@linux.microsoft.com> > + */ > + > +#include <linux/kexec.h> > +#include <linux/of.h> > +#include <linux/libfdt.h> > +#include <asm/ima.h> > + > +/** > + * setup_ima_buffer - add IMA buffer information to the fdt > + * > + * @image: kexec image being loaded. > + * @dtb: Flattened device tree for the next kernel. > + * @off: Offset to the chosen node. > + * > + * Return: 0 on success, or negative errno on error. > + */ > +int setup_ima_buffer(struct kimage *image, void *dtb, int off) > +{ > + int ret = 0; > + > + /* add ima-kexec-buffer */ > + if (image->arch.ima_buffer_size > 0) { > + ret = fdt_appendprop_addrrange(dtb, 0, off, > + "linux,ima-kexec-buffer", > + image->arch.ima_buffer_addr, > + image->arch.ima_buffer_size); > + if (ret) > + return (ret == -FDT_ERR_NOSPACE ? -ENOMEM : -EINVAL); > + > + ret = fdt_add_mem_rsv(dtb, image->arch.ima_buffer_addr, > + image->arch.ima_buffer_size); > + } > + > + return ret; > +} -- Thiago Jung Bauermann IBM Linux Technology Center
On 12/4/20 6:22 PM, Thiago Jung Bauermann wrote: > > Hello Lakshmi, > > Lakshmi Ramasubramanian <nramas@linux.microsoft.com> writes: > >> delete_fdt_mem_rsv() retrieves the memory reserve map entry, for >> the given starting address and size, from the device tree blob, and >> removes the entry from the device tree blob. This function is called >> to free the resources reserved for the buffer used for carrying forward >> the IMA measurement logs on kexec. This function does not have >> architecture specific code, but is currently limited to powerpc. >> >> Move delete_fdt_mem_rsv() to "drivers/of/kexec_fdt.c" so that it is > > s/kexec_fdt.c/kexec.c/ Missed that in the patch description. Will fix it. Thanks. >> accessible for other architectures as well. >> >> Co-developed-by: Prakhar Srivastava <prsriva@linux.microsoft.com> >> Signed-off-by: Prakhar Srivastava <prsriva@linux.microsoft.com> >> Signed-off-by: Lakshmi Ramasubramanian <nramas@linux.microsoft.com> >> --- >> arch/powerpc/include/asm/kexec.h | 1 - >> arch/powerpc/kexec/file_load.c | 32 ----------------- >> drivers/of/Makefile | 1 + >> drivers/of/kexec.c | 61 ++++++++++++++++++++++++++++++++ >> include/linux/kexec.h | 5 +++ >> 5 files changed, 67 insertions(+), 33 deletions(-) >> create mode 100644 drivers/of/kexec.c >> >> diff --git a/arch/powerpc/include/asm/kexec.h b/arch/powerpc/include/asm/kexec.h >> index 55d6ede30c19..7c223031ecdd 100644 >> --- a/arch/powerpc/include/asm/kexec.h >> +++ b/arch/powerpc/include/asm/kexec.h >> @@ -126,7 +126,6 @@ int setup_purgatory(struct kimage *image, const void *slave_code, >> int setup_new_fdt(const struct kimage *image, void *fdt, >> unsigned long initrd_load_addr, unsigned long initrd_len, >> const char *cmdline); >> -int delete_fdt_mem_rsv(void *fdt, unsigned long start, unsigned long size); >> >> #ifdef CONFIG_PPC64 >> struct kexec_buf; >> diff --git a/arch/powerpc/kexec/file_load.c b/arch/powerpc/kexec/file_load.c >> index 9a232bc36c8f..9efc98b1e2ae 100644 >> --- a/arch/powerpc/kexec/file_load.c >> +++ b/arch/powerpc/kexec/file_load.c >> @@ -109,38 +109,6 @@ int setup_purgatory(struct kimage *image, const void *slave_code, >> return 0; >> } >> >> -/** >> - * delete_fdt_mem_rsv - delete memory reservation with given address and size >> - * >> - * Return: 0 on success, or negative errno on error. >> - */ >> -int delete_fdt_mem_rsv(void *fdt, unsigned long start, unsigned long size) >> -{ >> - int i, ret, num_rsvs = fdt_num_mem_rsv(fdt); >> - >> - for (i = 0; i < num_rsvs; i++) { >> - uint64_t rsv_start, rsv_size; >> - >> - ret = fdt_get_mem_rsv(fdt, i, &rsv_start, &rsv_size); >> - if (ret) { >> - pr_err("Malformed device tree.\n"); >> - return -EINVAL; >> - } >> - >> - if (rsv_start == start && rsv_size == size) { >> - ret = fdt_del_mem_rsv(fdt, i); >> - if (ret) { >> - pr_err("Error deleting device tree reservation.\n"); >> - return -EINVAL; >> - } >> - >> - return 0; >> - } >> - } >> - >> - return -ENOENT; >> -} >> - >> /* >> * setup_new_fdt - modify /chosen and memory reservation for the next kernel >> * @image: kexec image being loaded. >> diff --git a/drivers/of/Makefile b/drivers/of/Makefile >> index 6e1e5212f058..77d24712c0c8 100644 >> --- a/drivers/of/Makefile >> +++ b/drivers/of/Makefile >> @@ -13,5 +13,6 @@ obj-$(CONFIG_OF_RESERVED_MEM) += of_reserved_mem.o >> obj-$(CONFIG_OF_RESOLVE) += resolver.o >> obj-$(CONFIG_OF_OVERLAY) += overlay.o >> obj-$(CONFIG_OF_NUMA) += of_numa.o >> +obj-$(CONFIG_OF_FLATTREE) += kexec.o > > Isn't this too broad? kexec.o will only be useful to kernel configs > which enable CONFIG_KEXEC_FILE, so perhaps do: > > ifdef CONFIG_OF_FLATTREE > ifdef CONFIG_KEXEC_FILE > obj-y += kexec.o > endif > endif > > What do you think? Per Rob's feedback on v9 patch set, I have moved all the architecture independent ima kexec functions to a single file "drivers/of/kexec.c" Since these functions are enabled on different kernel CONFIGs, I have used IS_ENABLED(CONFIG_XYZ) macro instead of "#ifdef" in the C file to conditionally compile. > >> obj-$(CONFIG_OF_UNITTEST) += unittest-data/ >> diff --git a/drivers/of/kexec.c b/drivers/of/kexec.c >> new file mode 100644 >> index 000000000000..b7d59105fcb8 >> --- /dev/null >> +++ b/drivers/of/kexec.c >> @@ -0,0 +1,61 @@ >> +// SPDX-License-Identifier: GPL-2.0+ >> +/* >> + * Copyright (C) 2020 Microsoft Corporation >> + * >> + * Author: Lakshmi Ramasubramanian (nramas@linux.microsoft.com) >> + * >> + * File: kexec.c >> + * Defines kexec related functions. >> + */ >> + >> +#define pr_fmt(fmt) "OF: kexec: " fmt >> + >> +#include <linux/kernel.h> >> +#include <linux/slab.h> >> +#include <linux/memblock.h> >> +#include <linux/kexec.h> >> +#include <linux/of.h> >> +#include <linux/of_fdt.h> >> +#include <linux/libfdt.h> >> + >> +/** >> + * delete_fdt_mem_rsv - delete memory reservation with given address and size >> + * >> + * @fdt: Flattened Device Tree to update >> + * @start: Starting address of the reservation to delete >> + * @size: Size of the reservation to delete >> + * >> + * Return: 0 on success, or negative errno on error. >> + */ >> +int delete_fdt_mem_rsv(void *fdt, unsigned long start, unsigned long size) >> +{ >> + int i, ret, num_rsvs; >> + >> + if (!IS_ENABLED(CONFIG_KEXEC_FILE)) >> + return 0; > > If you agree with my suggestion, then this IS_ENABLED() wouldn't be > needed. > Please see my response above. If there is a way to keep all the ima kexec functions in a single file and yet not use "#ifdef" in C file to conditionally compile, please let me know. I'll update. thanks, -lakshmi >> + >> + num_rsvs = fdt_num_mem_rsv(fdt); >> + for (i = 0; i < num_rsvs; i++) { >> + uint64_t rsv_start, rsv_size; >> + >> + ret = fdt_get_mem_rsv(fdt, i, &rsv_start, &rsv_size); >> + if (ret) { >> + pr_err("Malformed device tree.\n"); >> + return -EINVAL; >> + } >> + >> + if (rsv_start == start && rsv_size == size) { >> + ret = fdt_del_mem_rsv(fdt, i); >> + if (ret) { >> + pr_err("Error deleting device tree reservation.\n"); >> + return -EINVAL; >> + } >> + >> + pr_debug("Freed reserved memory at %lu of size %lu\n", >> + start, size); >> + return 0; >> + } >> + } >> + >> + return -ENOENT; >> +} > > The function code is unchanged apart from the addition of the > IS_ENABLED() and the pr_debug(), so that is good. > >> diff --git a/include/linux/kexec.h b/include/linux/kexec.h >> index 9e93bef52968..d0234c4815da 100644 >> --- a/include/linux/kexec.h >> +++ b/include/linux/kexec.h >> @@ -407,6 +407,11 @@ static inline int kexec_crash_loaded(void) { return 0; } >> #define kexec_in_progress false >> #endif /* CONFIG_KEXEC_CORE */ >> >> +#if defined(CONFIG_OF_FLATTREE) > > This would also change to require CONFIG_KEXEC_FILE. > >> +extern int delete_fdt_mem_rsv(void *fdt, unsigned long start, >> + unsigned long size); >> +#endif /* CONFIG_OF_FLATTREE */ >> + >> #endif /* !defined(__ASSEBMLY__) */ >> >> #endif /* LINUX_KEXEC_H */ > >
On 12/5/20 11:51 AM, Thiago Jung Bauermann wrote: > > Lakshmi Ramasubramanian <nramas@linux.microsoft.com> writes: > >> do_get_kexec_buffer() and get_addr_size_cells() are called by >> ima_get_kexec_buffer(), ima_free_kexec_buffer, and remove_ima_buffer() >> to retrieve the address and size of the buffer used for carrying >> forward the IMA measurement log across kexec system call. These >> functions correctly handle a device tree property that is a child node >> of the root node, but not anything other than the immediate root >> child nodes. >> >> Use the architecture independent functions get_ima_kexec_buffer() >> and get_root_addr_size_cells() defined in "drivers/of/ima_kexec.c", > > s/ima_kexec.c/kexec.c/ Thanks for catching this. I'll fix it. > >> to get the address and size of the IMA measurement log buffer from >> the device tree. Remove do_get_kexec_buffer() and get_addr_size_cells() >> since they are not used anymore. >> >> Co-developed-by: Prakhar Srivastava <prsriva@linux.microsoft.com> >> Signed-off-by: Prakhar Srivastava <prsriva@linux.microsoft.com> >> Signed-off-by: Lakshmi Ramasubramanian <nramas@linux.microsoft.com> > > Looks good. Thanks! > > Reviewed-by: Thiago Jung Bauermann <bauerman@linux.ibm.com> Thanks Thiago. -lakshmi >> --- >> arch/powerpc/kexec/ima.c | 58 +++++----------------------------------- >> 1 file changed, 7 insertions(+), 51 deletions(-) >> >> diff --git a/arch/powerpc/kexec/ima.c b/arch/powerpc/kexec/ima.c >> index a36c39db4b1a..906e8212435d 100644 >> --- a/arch/powerpc/kexec/ima.c >> +++ b/arch/powerpc/kexec/ima.c >> @@ -13,40 +13,6 @@ >> #include <linux/libfdt.h> >> #include <asm/ima.h> >> >> -static int get_addr_size_cells(int *addr_cells, int *size_cells) >> -{ >> - struct device_node *root; >> - >> - root = of_find_node_by_path("/"); >> - if (!root) >> - return -EINVAL; >> - >> - *addr_cells = of_n_addr_cells(root); >> - *size_cells = of_n_size_cells(root); >> - >> - of_node_put(root); >> - >> - return 0; >> -} >> - >> -static int do_get_kexec_buffer(const void *prop, int len, unsigned long *addr, >> - size_t *size) >> -{ >> - int ret, addr_cells, size_cells; >> - >> - ret = get_addr_size_cells(&addr_cells, &size_cells); >> - if (ret) >> - return ret; >> - >> - if (len < 4 * (addr_cells + size_cells)) >> - return -ENOENT; >> - >> - *addr = of_read_number(prop, addr_cells); >> - *size = of_read_number(prop + 4 * addr_cells, size_cells); >> - >> - return 0; >> -} >> - >> /** >> * ima_get_kexec_buffer - get IMA buffer from the previous kernel >> * @addr: On successful return, set to point to the buffer contents. >> @@ -56,16 +22,11 @@ static int do_get_kexec_buffer(const void *prop, int len, unsigned long *addr, >> */ >> int ima_get_kexec_buffer(void **addr, size_t *size) >> { >> - int ret, len; >> + int ret; >> unsigned long tmp_addr; >> size_t tmp_size; >> - const void *prop; >> >> - prop = of_get_property(of_chosen, "linux,ima-kexec-buffer", &len); >> - if (!prop) >> - return -ENOENT; >> - >> - ret = do_get_kexec_buffer(prop, len, &tmp_addr, &tmp_size); >> + ret = get_ima_kexec_buffer(NULL, 0, &tmp_addr, &tmp_size); >> if (ret) >> return ret; >> >> @@ -89,7 +50,7 @@ int ima_free_kexec_buffer(void) >> if (!prop) >> return -ENOENT; >> >> - ret = do_get_kexec_buffer(prop->value, prop->length, &addr, &size); >> + ret = get_ima_kexec_buffer(NULL, 0, &addr, &size); >> if (ret) >> return ret; >> >> @@ -98,7 +59,6 @@ int ima_free_kexec_buffer(void) >> return ret; >> >> return memblock_free(addr, size); >> - >> } >> >> /** >> @@ -112,19 +72,15 @@ int ima_free_kexec_buffer(void) >> */ >> void remove_ima_buffer(void *fdt, int chosen_node) >> { >> - int ret, len; >> + int ret; >> unsigned long addr; >> size_t size; >> - const void *prop; >> >> - prop = fdt_getprop(fdt, chosen_node, "linux,ima-kexec-buffer", &len); >> - if (!prop) >> + ret = get_ima_kexec_buffer(fdt, chosen_node, &addr, &size); >> + if (ret) >> return; >> >> - ret = do_get_kexec_buffer(prop, len, &addr, &size); >> fdt_delprop(fdt, chosen_node, "linux,ima-kexec-buffer"); >> - if (ret) >> - return; >> >> ret = delete_fdt_mem_rsv(fdt, addr, size); >> if (!ret) >> @@ -191,7 +147,7 @@ int setup_ima_buffer(const struct kimage *image, void *fdt, int chosen_node) >> if (!image->arch.ima_buffer_size) >> return 0; >> >> - ret = get_addr_size_cells(&addr_cells, &size_cells); >> + ret = get_root_addr_size_cells(&addr_cells, &size_cells); >> if (ret) >> return ret; > >
On 12/5/20 1:02 PM, Thiago Jung Bauermann wrote: > > Lakshmi Ramasubramanian <nramas@linux.microsoft.com> writes: > >> ima_get_kexec_buffer() retrieves the address and size of the buffer >> used for carrying forward the IMA measurement logs on kexec from >> the device tree. >> >> ima_free_kexec_buffer() removes the chosen node >> "linux,ima-kexec-buffer" from the device tree, and frees the buffer >> used for carrying forward the IMA measurement logs on kexec. >> >> These functions do not have architecture specific code, but are >> currently limited to powerpc. Move ima_get_kexec_buffer() and >> ima_free_kexec_buffer() to ima_kexec.c in IMA so that they are >> accessible for other architectures as well. >> >> With the above change the functions in arch/powerpc/kexec/ima.c are >> defined only when the kernel config CONFIG_IMA_KEXEC is enabled. >> Update the Makefile to build arch/powerpc/kexec/ima.c only when >> CONFIG_IMA_KEXEC is enabled and remove "#ifdef CONFIG_IMA_KEXEC" >> in arch/powerpc/kexec/ima.c. > > Nice. > >> Co-developed-by: Prakhar Srivastava <prsriva@linux.microsoft.com> >> Signed-off-by: Prakhar Srivastava <prsriva@linux.microsoft.com> >> Signed-off-by: Lakshmi Ramasubramanian <nramas@linux.microsoft.com> >> Reviewed-by: Mimi Zohar <zohar@linux.ibm.com> > > Reviewed-by: Thiago Jung Bauermann <bauerman@linux.ibm.com> > > Just a small nit below: > >> --- >> arch/powerpc/include/asm/ima.h | 3 -- >> arch/powerpc/kexec/Makefile | 7 +--- >> arch/powerpc/kexec/ima.c | 50 ----------------------------- >> security/integrity/ima/ima_kexec.c | 51 ++++++++++++++++++++++++++++++ >> 4 files changed, 52 insertions(+), 59 deletions(-) >> >> diff --git a/arch/powerpc/include/asm/ima.h b/arch/powerpc/include/asm/ima.h >> index a2fc71bc3b23..d8444d27f0d8 100644 >> --- a/arch/powerpc/include/asm/ima.h >> +++ b/arch/powerpc/include/asm/ima.h >> @@ -6,9 +6,6 @@ >> >> struct kimage; >> >> -int ima_get_kexec_buffer(void **addr, size_t *size); >> -int ima_free_kexec_buffer(void); >> - >> #ifdef CONFIG_IMA_KEXEC >> int arch_ima_add_kexec_buffer(struct kimage *image, unsigned long load_addr, >> size_t size); >> diff --git a/arch/powerpc/kexec/Makefile b/arch/powerpc/kexec/Makefile >> index 4aff6846c772..f54a9dbff4c8 100644 >> --- a/arch/powerpc/kexec/Makefile >> +++ b/arch/powerpc/kexec/Makefile >> @@ -9,12 +9,7 @@ obj-$(CONFIG_PPC32) += relocate_32.o >> >> obj-$(CONFIG_KEXEC_FILE) += file_load.o ranges.o file_load_$(BITS).o elf_$(BITS).o >> >> -ifdef CONFIG_HAVE_IMA_KEXEC >> -ifdef CONFIG_IMA >> -obj-y += ima.o >> -endif >> -endif >> - >> +obj-$(CONFIG_IMA_KEXEC) += ima.o >> >> # Disable GCOV, KCOV & sanitizers in odd or sensitive code >> GCOV_PROFILE_core_$(BITS).o := n >> diff --git a/arch/powerpc/kexec/ima.c b/arch/powerpc/kexec/ima.c >> index 68017123b07d..bf7084c0c4da 100644 >> --- a/arch/powerpc/kexec/ima.c >> +++ b/arch/powerpc/kexec/ima.c >> @@ -13,55 +13,6 @@ >> #include <linux/libfdt.h> >> #include <asm/ima.h> > > With this patch, the following includes become unnecessary and can be > removed from this file: > > #include <linux/of.h> > #include <linux/memblock.h> > Sure - will make this change. thanks, -lakshmi >> -/** >> - * ima_get_kexec_buffer - get IMA buffer from the previous kernel >> - * @addr: On successful return, set to point to the buffer contents. >> - * @size: On successful return, set to the buffer size. >> - * >> - * Return: 0 on success, negative errno on error. >> - */ >> -int ima_get_kexec_buffer(void **addr, size_t *size) >> -{ >> - int ret; >> - unsigned long tmp_addr; >> - size_t tmp_size; >> - >> - ret = get_ima_kexec_buffer(NULL, 0, &tmp_addr, &tmp_size); >> - if (ret) >> - return ret; >> - >> - *addr = __va(tmp_addr); >> - *size = tmp_size; >> - >> - return 0; >> -} >> - >> -/** >> - * ima_free_kexec_buffer - free memory used by the IMA buffer >> - */ >> -int ima_free_kexec_buffer(void) >> -{ >> - int ret; >> - unsigned long addr; >> - size_t size; >> - struct property *prop; >> - >> - prop = of_find_property(of_chosen, "linux,ima-kexec-buffer", NULL); >> - if (!prop) >> - return -ENOENT; >> - >> - ret = get_ima_kexec_buffer(NULL, 0, &addr, &size); >> - if (ret) >> - return ret; >> - >> - ret = of_remove_property(of_chosen, prop); >> - if (ret) >> - return ret; >> - >> - return memblock_free(addr, size); >> -} >> - >> -#ifdef CONFIG_IMA_KEXEC >> /** >> * arch_ima_add_kexec_buffer - do arch-specific steps to add the IMA buffer >> * >> @@ -154,4 +105,3 @@ int setup_ima_buffer(const struct kimage *image, void *fdt, int chosen_node) >> >> return 0; >> } >> -#endif /* CONFIG_IMA_KEXEC */ >
On 12/5/20 1:44 PM, Thiago Jung Bauermann wrote: > > Lakshmi Ramasubramanian <nramas@linux.microsoft.com> writes: > >> Address and size of the buffer containing the IMA measurement log need >> to be passed from the current kernel to the next kernel on kexec. >> >> Any existing "linux,ima-kexec-buffer" property in the device tree >> needs to be removed and its corresponding memory reservation in >> the currently running kernel needs to be freed. The address and >> size of the current kernel's IMA measurement log need to be added >> to the device tree's IMA kexec buffer node and memory for the buffer >> needs to be reserved for the log to be carried over to the next kernel >> on the kexec call. >> >> Add address and size fields to "struct kimage_arch" for ARM64 platform >> to hold the address and size of the IMA measurement log buffer. Remove >> any existing "linux,ima-kexec-buffer" property in the device tree and >> free the corresponding memory reservation in the currently running >> kernel. Add "linux,ima-kexec-buffer" property to the device tree and >> reserve the memory for storing the IMA log that needs to be passed from >> the current kernel to the next one. >> >> Update CONFIG_KEXEC_FILE to select CONFIG_HAVE_IMA_KEXEC to indicate >> that the IMA measurement log information is present in the device tree >> for ARM64. >> >> Co-developed-by: Prakhar Srivastava <prsriva@linux.microsoft.com> >> Signed-off-by: Prakhar Srivastava <prsriva@linux.microsoft.com> >> Signed-off-by: Lakshmi Ramasubramanian <nramas@linux.microsoft.com> > > Reviewed-by: Thiago Jung Bauermann <bauerman@linux.ibm.com> > >> --- >> arch/arm64/Kconfig | 1 + >> arch/arm64/include/asm/ima.h | 22 ++++++++++++++ >> arch/arm64/include/asm/kexec.h | 5 ++++ >> arch/arm64/kernel/Makefile | 1 + >> arch/arm64/kernel/ima.c | 41 ++++++++++++++++++++++++++ >> arch/arm64/kernel/machine_kexec_file.c | 8 +++++ >> 6 files changed, 78 insertions(+) >> create mode 100644 arch/arm64/include/asm/ima.h >> create mode 100644 arch/arm64/kernel/ima.c >> >> diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig >> index 1515f6f153a0..bcca4a467eda 100644 >> --- a/arch/arm64/Kconfig >> +++ b/arch/arm64/Kconfig >> @@ -1094,6 +1094,7 @@ config KEXEC >> config KEXEC_FILE >> bool "kexec file based system call" >> select KEXEC_CORE >> + select HAVE_IMA_KEXEC >> help >> This is new version of kexec system call. This system call is >> file based and takes file descriptors as system call argument >> diff --git a/arch/arm64/include/asm/ima.h b/arch/arm64/include/asm/ima.h >> new file mode 100644 >> index 000000000000..354adc8b444c >> --- /dev/null >> +++ b/arch/arm64/include/asm/ima.h >> @@ -0,0 +1,22 @@ >> +/* SPDX-License-Identifier: GPL-2.0-or-later */ >> +/* >> + * Copyright (C) 2019 Microsoft Corporation > > Shouldn't the year here be 2020? Yes - will update. >> + * >> + * Author: Prakhar Srivastava <prsriva@linux.microsoft.com> >> + * >> + */ >> +#ifndef _ASM_ARCH_IMA_H >> +#define _ASM_ARCH_IMA_H >> + >> +struct kimage; >> + >> +#ifdef CONFIG_IMA_KEXEC >> +extern int setup_ima_buffer(struct kimage *image, void *dtb, int off); >> +#else >> +static inline int setup_ima_buffer(struct kimage *image, void *dtb, int off) >> +{ >> + return 0; >> +} >> +#endif /* CONFIG_IMA_KEXEC */ >> + >> +#endif /* _ASM_ARCH_IMA_H */ > > <snip> > >> diff --git a/arch/arm64/kernel/ima.c b/arch/arm64/kernel/ima.c >> new file mode 100644 >> index 000000000000..d65bdbe1adf6 >> --- /dev/null >> +++ b/arch/arm64/kernel/ima.c >> @@ -0,0 +1,41 @@ >> +// SPDX-License-Identifier: GPL-2.0-or-later >> +/* >> + * Copyright (C) 2016 IBM Corporation > > Same question: shouldn't the year here be 2020? Yes. Also, should be "Copyright (C) 2020 Microsoft Corporation" Will make these changes. thanks, -lakshmi > >> + * >> + * Authors: >> + * Lakshmi Ramasubramanian <nramas@linux.microsoft.com> >> + */ >> + >> +#include <linux/kexec.h> >> +#include <linux/of.h> >> +#include <linux/libfdt.h> >> +#include <asm/ima.h> >> + >> +/** >> + * setup_ima_buffer - add IMA buffer information to the fdt >> + * >> + * @image: kexec image being loaded. >> + * @dtb: Flattened device tree for the next kernel. >> + * @off: Offset to the chosen node. >> + * >> + * Return: 0 on success, or negative errno on error. >> + */ >> +int setup_ima_buffer(struct kimage *image, void *dtb, int off) >> +{ >> + int ret = 0; >> + >> + /* add ima-kexec-buffer */ >> + if (image->arch.ima_buffer_size > 0) { >> + ret = fdt_appendprop_addrrange(dtb, 0, off, >> + "linux,ima-kexec-buffer", >> + image->arch.ima_buffer_addr, >> + image->arch.ima_buffer_size); >> + if (ret) >> + return (ret == -FDT_ERR_NOSPACE ? -ENOMEM : -EINVAL); >> + >> + ret = fdt_add_mem_rsv(dtb, image->arch.ima_buffer_addr, >> + image->arch.ima_buffer_size); >> + } >> + >> + return ret; >> +} >
On 12/6/20 5:50 PM, Lakshmi Ramasubramanian wrote: Hi Thiago, > On 12/4/20 6:22 PM, Thiago Jung Bauermann wrote >> >> Hello Lakshmi, >> >> Lakshmi Ramasubramanian <nramas@linux.microsoft.com> writes: >> >>> delete_fdt_mem_rsv() retrieves the memory reserve map entry, for >>> the given starting address and size, from the device tree blob, and >>> removes the entry from the device tree blob. This function is called >>> to free the resources reserved for the buffer used for carrying forward >>> the IMA measurement logs on kexec. This function does not have >>> architecture specific code, but is currently limited to powerpc. >>> >>> Move delete_fdt_mem_rsv() to "drivers/of/kexec_fdt.c" so that it is >> >> s/kexec_fdt.c/kexec.c/ > > Missed that in the patch description. Will fix it. Thanks. > >>> accessible for other architectures as well. >>> >>> Co-developed-by: Prakhar Srivastava <prsriva@linux.microsoft.com> >>> Signed-off-by: Prakhar Srivastava <prsriva@linux.microsoft.com> >>> Signed-off-by: Lakshmi Ramasubramanian <nramas@linux.microsoft.com> >>> --- >>> arch/powerpc/include/asm/kexec.h | 1 - >>> arch/powerpc/kexec/file_load.c | 32 ----------------- >>> drivers/of/Makefile | 1 + >>> drivers/of/kexec.c | 61 ++++++++++++++++++++++++++++++++ >>> include/linux/kexec.h | 5 +++ >>> 5 files changed, 67 insertions(+), 33 deletions(-) >>> create mode 100644 drivers/of/kexec.c >>> >>> diff --git a/arch/powerpc/include/asm/kexec.h >>> b/arch/powerpc/include/asm/kexec.h >>> index 55d6ede30c19..7c223031ecdd 100644 >>> --- a/arch/powerpc/include/asm/kexec.h >>> +++ b/arch/powerpc/include/asm/kexec.h >>> @@ -126,7 +126,6 @@ int setup_purgatory(struct kimage *image, const >>> void *slave_code, >>> int setup_new_fdt(const struct kimage *image, void *fdt, >>> unsigned long initrd_load_addr, unsigned long initrd_len, >>> const char *cmdline); >>> -int delete_fdt_mem_rsv(void *fdt, unsigned long start, unsigned long >>> size); >>> #ifdef CONFIG_PPC64 >>> struct kexec_buf; >>> diff --git a/arch/powerpc/kexec/file_load.c >>> b/arch/powerpc/kexec/file_load.c >>> index 9a232bc36c8f..9efc98b1e2ae 100644 >>> --- a/arch/powerpc/kexec/file_load.c >>> +++ b/arch/powerpc/kexec/file_load.c >>> @@ -109,38 +109,6 @@ int setup_purgatory(struct kimage *image, const >>> void *slave_code, >>> return 0; >>> } >>> -/** >>> - * delete_fdt_mem_rsv - delete memory reservation with given address >>> and size >>> - * >>> - * Return: 0 on success, or negative errno on error. >>> - */ >>> -int delete_fdt_mem_rsv(void *fdt, unsigned long start, unsigned long >>> size) >>> -{ >>> - int i, ret, num_rsvs = fdt_num_mem_rsv(fdt); >>> - >>> - for (i = 0; i < num_rsvs; i++) { >>> - uint64_t rsv_start, rsv_size; >>> - >>> - ret = fdt_get_mem_rsv(fdt, i, &rsv_start, &rsv_size); >>> - if (ret) { >>> - pr_err("Malformed device tree.\n"); >>> - return -EINVAL; >>> - } >>> - >>> - if (rsv_start == start && rsv_size == size) { >>> - ret = fdt_del_mem_rsv(fdt, i); >>> - if (ret) { >>> - pr_err("Error deleting device tree reservation.\n"); >>> - return -EINVAL; >>> - } >>> - >>> - return 0; >>> - } >>> - } >>> - >>> - return -ENOENT; >>> -} >>> - >>> /* >>> * setup_new_fdt - modify /chosen and memory reservation for the >>> next kernel >>> * @image: kexec image being loaded. >>> diff --git a/drivers/of/Makefile b/drivers/of/Makefile >>> index 6e1e5212f058..77d24712c0c8 100644 >>> --- a/drivers/of/Makefile >>> +++ b/drivers/of/Makefile >>> @@ -13,5 +13,6 @@ obj-$(CONFIG_OF_RESERVED_MEM) += of_reserved_mem.o >>> obj-$(CONFIG_OF_RESOLVE) += resolver.o >>> obj-$(CONFIG_OF_OVERLAY) += overlay.o >>> obj-$(CONFIG_OF_NUMA) += of_numa.o >>> +obj-$(CONFIG_OF_FLATTREE) += kexec.o >> >> Isn't this too broad? kexec.o will only be useful to kernel configs >> which enable CONFIG_KEXEC_FILE, so perhaps do: >> >> ifdef CONFIG_OF_FLATTREE >> ifdef CONFIG_KEXEC_FILE >> obj-y += kexec.o >> endif >> endif >> >> What do you think? > > Per Rob's feedback on v9 patch set, I have moved all the architecture > independent ima kexec functions to a single file "drivers/of/kexec.c" > > Since these functions are enabled on different kernel CONFIGs, I have > used IS_ENABLED(CONFIG_XYZ) macro instead of "#ifdef" in the C file to > conditionally compile. Per Rob's feedback on the v9 patch, I'll keep the ima kexec functions in a single file (in "drivers/of/kexec.c") and use IS_ENABLED() macro to handle the function calls. I'll make the other changes you'd suggested on v10 patches and will post v11 patch set shortly. thanks, -lakshmi > >> >>> obj-$(CONFIG_OF_UNITTEST) += unittest-data/ >>> diff --git a/drivers/of/kexec.c b/drivers/of/kexec.c >>> new file mode 100644 >>> index 000000000000..b7d59105fcb8 >>> --- /dev/null >>> +++ b/drivers/of/kexec.c >>> @@ -0,0 +1,61 @@ >>> +// SPDX-License-Identifier: GPL-2.0+ >>> +/* >>> + * Copyright (C) 2020 Microsoft Corporation >>> + * >>> + * Author: Lakshmi Ramasubramanian (nramas@linux.microsoft.com) >>> + * >>> + * File: kexec.c >>> + * Defines kexec related functions. >>> + */ >>> + >>> +#define pr_fmt(fmt) "OF: kexec: " fmt >>> + >>> +#include <linux/kernel.h> >>> +#include <linux/slab.h> >>> +#include <linux/memblock.h> >>> +#include <linux/kexec.h> >>> +#include <linux/of.h> >>> +#include <linux/of_fdt.h> >>> +#include <linux/libfdt.h> >>> + >>> +/** >>> + * delete_fdt_mem_rsv - delete memory reservation with given address >>> and size >>> + * >>> + * @fdt: Flattened Device Tree to update >>> + * @start: Starting address of the reservation to delete >>> + * @size: Size of the reservation to delete >>> + * >>> + * Return: 0 on success, or negative errno on error. >>> + */ >>> +int delete_fdt_mem_rsv(void *fdt, unsigned long start, unsigned long >>> size) >>> +{ >>> + int i, ret, num_rsvs; >>> + >>> + if (!IS_ENABLED(CONFIG_KEXEC_FILE)) >>> + return 0; >> >> If you agree with my suggestion, then this IS_ENABLED() wouldn't be >> needed. >> > > Please see my response above. > > If there is a way to keep all the ima kexec functions in a single file > and yet not use "#ifdef" in C file to conditionally compile, please let > me know. I'll update. > > thanks, > -lakshmi > >>> + >>> + num_rsvs = fdt_num_mem_rsv(fdt); >>> + for (i = 0; i < num_rsvs; i++) { >>> + uint64_t rsv_start, rsv_size; >>> + >>> + ret = fdt_get_mem_rsv(fdt, i, &rsv_start, &rsv_size); >>> + if (ret) { >>> + pr_err("Malformed device tree.\n"); >>> + return -EINVAL; >>> + } >>> + >>> + if (rsv_start == start && rsv_size == size) { >>> + ret = fdt_del_mem_rsv(fdt, i); >>> + if (ret) { >>> + pr_err("Error deleting device tree reservation.\n"); >>> + return -EINVAL; >>> + } >>> + >>> + pr_debug("Freed reserved memory at %lu of size %lu\n", >>> + start, size); >>> + return 0; >>> + } >>> + } >>> + >>> + return -ENOENT; >>> +} >> >> The function code is unchanged apart from the addition of the >> IS_ENABLED() and the pr_debug(), so that is good. >> >>> diff --git a/include/linux/kexec.h b/include/linux/kexec.h >>> index 9e93bef52968..d0234c4815da 100644 >>> --- a/include/linux/kexec.h >>> +++ b/include/linux/kexec.h >>> @@ -407,6 +407,11 @@ static inline int kexec_crash_loaded(void) { >>> return 0; } >>> #define kexec_in_progress false >>> #endif /* CONFIG_KEXEC_CORE */ >>> +#if defined(CONFIG_OF_FLATTREE) >> >> This would also change to require CONFIG_KEXEC_FILE. >> >>> +extern int delete_fdt_mem_rsv(void *fdt, unsigned long start, >>> + unsigned long size); >>> +#endif /* CONFIG_OF_FLATTREE */ >>> + >>> #endif /* !defined(__ASSEBMLY__) */ >>> #endif /* LINUX_KEXEC_H */ >> >>
Hi Lakshmi, Lakshmi Ramasubramanian <nramas@linux.microsoft.com> writes: > On 12/6/20 5:50 PM, Lakshmi Ramasubramanian wrote: > > Hi Thiago, > >> On 12/4/20 6:22 PM, Thiago Jung Bauermann wrote >>> >>> Hello Lakshmi, >>> >>> Lakshmi Ramasubramanian <nramas@linux.microsoft.com> writes: >>> >>>> delete_fdt_mem_rsv() retrieves the memory reserve map entry, for >>>> the given starting address and size, from the device tree blob, and >>>> removes the entry from the device tree blob. This function is called >>>> to free the resources reserved for the buffer used for carrying forward >>>> the IMA measurement logs on kexec. This function does not have >>>> architecture specific code, but is currently limited to powerpc. >>>> >>>> Move delete_fdt_mem_rsv() to "drivers/of/kexec_fdt.c" so that it is >>> >>> s/kexec_fdt.c/kexec.c/ >> Missed that in the patch description. Will fix it. Thanks. >> >>>> accessible for other architectures as well. >>>> >>>> Co-developed-by: Prakhar Srivastava <prsriva@linux.microsoft.com> >>>> Signed-off-by: Prakhar Srivastava <prsriva@linux.microsoft.com> >>>> Signed-off-by: Lakshmi Ramasubramanian <nramas@linux.microsoft.com> >>>> --- >>>> arch/powerpc/include/asm/kexec.h | 1 - >>>> arch/powerpc/kexec/file_load.c | 32 ----------------- >>>> drivers/of/Makefile | 1 + >>>> drivers/of/kexec.c | 61 ++++++++++++++++++++++++++++++++ >>>> include/linux/kexec.h | 5 +++ >>>> 5 files changed, 67 insertions(+), 33 deletions(-) >>>> create mode 100644 drivers/of/kexec.c >>>> >>>> diff --git a/arch/powerpc/include/asm/kexec.h >>>> b/arch/powerpc/include/asm/kexec.h >>>> index 55d6ede30c19..7c223031ecdd 100644 >>>> --- a/arch/powerpc/include/asm/kexec.h >>>> +++ b/arch/powerpc/include/asm/kexec.h >>>> @@ -126,7 +126,6 @@ int setup_purgatory(struct kimage *image, const void >>>> *slave_code, >>>> int setup_new_fdt(const struct kimage *image, void *fdt, >>>> unsigned long initrd_load_addr, unsigned long initrd_len, >>>> const char *cmdline); >>>> -int delete_fdt_mem_rsv(void *fdt, unsigned long start, unsigned long size); >>>> #ifdef CONFIG_PPC64 >>>> struct kexec_buf; >>>> diff --git a/arch/powerpc/kexec/file_load.c b/arch/powerpc/kexec/file_load.c >>>> index 9a232bc36c8f..9efc98b1e2ae 100644 >>>> --- a/arch/powerpc/kexec/file_load.c >>>> +++ b/arch/powerpc/kexec/file_load.c >>>> @@ -109,38 +109,6 @@ int setup_purgatory(struct kimage *image, const void >>>> *slave_code, >>>> return 0; >>>> } >>>> -/** >>>> - * delete_fdt_mem_rsv - delete memory reservation with given address and >>>> size >>>> - * >>>> - * Return: 0 on success, or negative errno on error. >>>> - */ >>>> -int delete_fdt_mem_rsv(void *fdt, unsigned long start, unsigned long size) >>>> -{ >>>> - int i, ret, num_rsvs = fdt_num_mem_rsv(fdt); >>>> - >>>> - for (i = 0; i < num_rsvs; i++) { >>>> - uint64_t rsv_start, rsv_size; >>>> - >>>> - ret = fdt_get_mem_rsv(fdt, i, &rsv_start, &rsv_size); >>>> - if (ret) { >>>> - pr_err("Malformed device tree.\n"); >>>> - return -EINVAL; >>>> - } >>>> - >>>> - if (rsv_start == start && rsv_size == size) { >>>> - ret = fdt_del_mem_rsv(fdt, i); >>>> - if (ret) { >>>> - pr_err("Error deleting device tree reservation.\n"); >>>> - return -EINVAL; >>>> - } >>>> - >>>> - return 0; >>>> - } >>>> - } >>>> - >>>> - return -ENOENT; >>>> -} >>>> - >>>> /* >>>> * setup_new_fdt - modify /chosen and memory reservation for the next >>>> kernel >>>> * @image: kexec image being loaded. >>>> diff --git a/drivers/of/Makefile b/drivers/of/Makefile >>>> index 6e1e5212f058..77d24712c0c8 100644 >>>> --- a/drivers/of/Makefile >>>> +++ b/drivers/of/Makefile >>>> @@ -13,5 +13,6 @@ obj-$(CONFIG_OF_RESERVED_MEM) += of_reserved_mem.o >>>> obj-$(CONFIG_OF_RESOLVE) += resolver.o >>>> obj-$(CONFIG_OF_OVERLAY) += overlay.o >>>> obj-$(CONFIG_OF_NUMA) += of_numa.o >>>> +obj-$(CONFIG_OF_FLATTREE) += kexec.o >>> >>> Isn't this too broad? kexec.o will only be useful to kernel configs >>> which enable CONFIG_KEXEC_FILE, so perhaps do: >>> >>> ifdef CONFIG_OF_FLATTREE >>> ifdef CONFIG_KEXEC_FILE >>> obj-y += kexec.o >>> endif >>> endif >>> >>> What do you think? >> Per Rob's feedback on v9 patch set, I have moved all the architecture >> independent ima kexec functions to a single file "drivers/of/kexec.c" >> Since these functions are enabled on different kernel CONFIGs, I have >> used IS_ENABLED(CONFIG_XYZ) macro instead of "#ifdef" in the C file to >> conditionally compile. > Per Rob's feedback on the v9 patch, I'll keep the ima kexec functions in a > single file (in "drivers/of/kexec.c") and use IS_ENABLED() macro to handle the > function calls. > > I'll make the other changes you'd suggested on v10 patches and will post v11 > patch set shortly. From a cursory look at the use of functions in this file, I got the impression that there wouldn't be any reference to them in kernel configs that didn't have CONFIG_KEXEC_FILE enabled, which is why I suggested the change above. I think you can make it without any other changes to the code. I could be wrong though, and there could be some config which tried to use some of these functions even when CONFIG_KEXEC_FILE is disabled. In that case, the customary way to resolve it is to provide static inline stub versions in a header file (not in a .c file) of just those functions that are needed. The reason why placing stub functions in header files is better is that then the compiler has visibility of the dummy function when compiling the source file which uses the function, and is able to eliminate the dead code that arises from the function always returning one value. Using IS_ENABLED() to do an early return as the first operation in the function in a separate .c file means that the compiler doesn't know anything and has to put a jump to the dummy function (only to immediately return), and retain the code that deals with the possibility of different values being returned. It's not a big deal in this case because none of these functions is in a hot path, but it does make the kernel text a tiny bit bigger than necessary. -- Thiago Jung Bauermann IBM Linux Technology Center
On 12/11/20 10:19 AM, Thiago Jung Bauermann wrote: > > Hi Lakshmi, > > Lakshmi Ramasubramanian <nramas@linux.microsoft.com> writes: > >> On 12/6/20 5:50 PM, Lakshmi Ramasubramanian wrote: >> >> Hi Thiago, >> >>> On 12/4/20 6:22 PM, Thiago Jung Bauermann wrote >>>> >>>> Hello Lakshmi, >>>> >>>> Lakshmi Ramasubramanian <nramas@linux.microsoft.com> writes: >>>> >>>>> delete_fdt_mem_rsv() retrieves the memory reserve map entry, for >>>>> the given starting address and size, from the device tree blob, and >>>>> removes the entry from the device tree blob. This function is called >>>>> to free the resources reserved for the buffer used for carrying forward >>>>> the IMA measurement logs on kexec. This function does not have >>>>> architecture specific code, but is currently limited to powerpc. >>>>> >>>>> Move delete_fdt_mem_rsv() to "drivers/of/kexec_fdt.c" so that it is >>>> >>>> s/kexec_fdt.c/kexec.c/ >>> Missed that in the patch description. Will fix it. Thanks. >>> >>>>> accessible for other architectures as well. >>>>> >>>>> Co-developed-by: Prakhar Srivastava <prsriva@linux.microsoft.com> >>>>> Signed-off-by: Prakhar Srivastava <prsriva@linux.microsoft.com> >>>>> Signed-off-by: Lakshmi Ramasubramanian <nramas@linux.microsoft.com> >>>>> --- >>>>> arch/powerpc/include/asm/kexec.h | 1 - >>>>> arch/powerpc/kexec/file_load.c | 32 ----------------- >>>>> drivers/of/Makefile | 1 + >>>>> drivers/of/kexec.c | 61 ++++++++++++++++++++++++++++++++ >>>>> include/linux/kexec.h | 5 +++ >>>>> 5 files changed, 67 insertions(+), 33 deletions(-) >>>>> create mode 100644 drivers/of/kexec.c >>>>> >>>>> diff --git a/arch/powerpc/include/asm/kexec.h >>>>> b/arch/powerpc/include/asm/kexec.h >>>>> index 55d6ede30c19..7c223031ecdd 100644 >>>>> --- a/arch/powerpc/include/asm/kexec.h >>>>> +++ b/arch/powerpc/include/asm/kexec.h >>>>> @@ -126,7 +126,6 @@ int setup_purgatory(struct kimage *image, const void >>>>> *slave_code, >>>>> int setup_new_fdt(const struct kimage *image, void *fdt, >>>>> unsigned long initrd_load_addr, unsigned long initrd_len, >>>>> const char *cmdline); >>>>> -int delete_fdt_mem_rsv(void *fdt, unsigned long start, unsigned long size); >>>>> #ifdef CONFIG_PPC64 >>>>> struct kexec_buf; >>>>> diff --git a/arch/powerpc/kexec/file_load.c b/arch/powerpc/kexec/file_load.c >>>>> index 9a232bc36c8f..9efc98b1e2ae 100644 >>>>> --- a/arch/powerpc/kexec/file_load.c >>>>> +++ b/arch/powerpc/kexec/file_load.c >>>>> @@ -109,38 +109,6 @@ int setup_purgatory(struct kimage *image, const void >>>>> *slave_code, >>>>> return 0; >>>>> } >>>>> -/** >>>>> - * delete_fdt_mem_rsv - delete memory reservation with given address and >>>>> size >>>>> - * >>>>> - * Return: 0 on success, or negative errno on error. >>>>> - */ >>>>> -int delete_fdt_mem_rsv(void *fdt, unsigned long start, unsigned long size) >>>>> -{ >>>>> - int i, ret, num_rsvs = fdt_num_mem_rsv(fdt); >>>>> - >>>>> - for (i = 0; i < num_rsvs; i++) { >>>>> - uint64_t rsv_start, rsv_size; >>>>> - >>>>> - ret = fdt_get_mem_rsv(fdt, i, &rsv_start, &rsv_size); >>>>> - if (ret) { >>>>> - pr_err("Malformed device tree.\n"); >>>>> - return -EINVAL; >>>>> - } >>>>> - >>>>> - if (rsv_start == start && rsv_size == size) { >>>>> - ret = fdt_del_mem_rsv(fdt, i); >>>>> - if (ret) { >>>>> - pr_err("Error deleting device tree reservation.\n"); >>>>> - return -EINVAL; >>>>> - } >>>>> - >>>>> - return 0; >>>>> - } >>>>> - } >>>>> - >>>>> - return -ENOENT; >>>>> -} >>>>> - >>>>> /* >>>>> * setup_new_fdt - modify /chosen and memory reservation for the next >>>>> kernel >>>>> * @image: kexec image being loaded. >>>>> diff --git a/drivers/of/Makefile b/drivers/of/Makefile >>>>> index 6e1e5212f058..77d24712c0c8 100644 >>>>> --- a/drivers/of/Makefile >>>>> +++ b/drivers/of/Makefile >>>>> @@ -13,5 +13,6 @@ obj-$(CONFIG_OF_RESERVED_MEM) += of_reserved_mem.o >>>>> obj-$(CONFIG_OF_RESOLVE) += resolver.o >>>>> obj-$(CONFIG_OF_OVERLAY) += overlay.o >>>>> obj-$(CONFIG_OF_NUMA) += of_numa.o >>>>> +obj-$(CONFIG_OF_FLATTREE) += kexec.o >>>> >>>> Isn't this too broad? kexec.o will only be useful to kernel configs >>>> which enable CONFIG_KEXEC_FILE, so perhaps do: >>>> >>>> ifdef CONFIG_OF_FLATTREE >>>> ifdef CONFIG_KEXEC_FILE >>>> obj-y += kexec.o >>>> endif >>>> endif >>>> >>>> What do you think? >>> Per Rob's feedback on v9 patch set, I have moved all the architecture >>> independent ima kexec functions to a single file "drivers/of/kexec.c" >>> Since these functions are enabled on different kernel CONFIGs, I have >>> used IS_ENABLED(CONFIG_XYZ) macro instead of "#ifdef" in the C file to >>> conditionally compile. >> Per Rob's feedback on the v9 patch, I'll keep the ima kexec functions in a >> single file (in "drivers/of/kexec.c") and use IS_ENABLED() macro to handle the >> function calls. >> >> I'll make the other changes you'd suggested on v10 patches and will post v11 >> patch set shortly. > >>From a cursory look at the use of functions in this file, I got the > impression that there wouldn't be any reference to them in kernel > configs that didn't have CONFIG_KEXEC_FILE enabled, which is why I > suggested the change above. I think you can make it without any other > changes to the code. > > I could be wrong though, and there could be some config which tried to > use some of these functions even when CONFIG_KEXEC_FILE is disabled. In > that case, the customary way to resolve it is to provide static inline > stub versions in a header file (not in a .c file) of just those > functions that are needed. > > The reason why placing stub functions in header files is better is that > then the compiler has visibility of the dummy function when compiling > the source file which uses the function, and is able to eliminate the > dead code that arises from the function always returning one value. I agree with you Thiago. Is there a way to keep all the relevant functions in a single C file, not use "#ifdef" in C file, and follow the coding pattern you've described above (i mean, "defining a stub function in a header file when the config conditions are not met")? > > Using IS_ENABLED() to do an early return as the first operation in the > function in a separate .c file means that the compiler doesn't know > anything and has to put a jump to the dummy function (only to > immediately return), and retain the code that deals with the possibility > of different values being returned. > > It's not a big deal in this case because none of these functions is in a > hot path, but it does make the kernel text a tiny bit bigger than > necessary. > I agree. thanks, -lakshmi
Lakshmi Ramasubramanian <nramas@linux.microsoft.com> writes: > On 12/11/20 10:19 AM, Thiago Jung Bauermann wrote: >> Hi Lakshmi, >> Lakshmi Ramasubramanian <nramas@linux.microsoft.com> writes: >> >>> On 12/6/20 5:50 PM, Lakshmi Ramasubramanian wrote: >>> >>> Hi Thiago, >>> >>>> On 12/4/20 6:22 PM, Thiago Jung Bauermann wrote >>>>> >>>>> Hello Lakshmi, >>>>> >>>>> Lakshmi Ramasubramanian <nramas@linux.microsoft.com> writes: >>>>> >>>>>> delete_fdt_mem_rsv() retrieves the memory reserve map entry, for >>>>>> the given starting address and size, from the device tree blob, and >>>>>> removes the entry from the device tree blob. This function is called >>>>>> to free the resources reserved for the buffer used for carrying forward >>>>>> the IMA measurement logs on kexec. This function does not have >>>>>> architecture specific code, but is currently limited to powerpc. >>>>>> >>>>>> Move delete_fdt_mem_rsv() to "drivers/of/kexec_fdt.c" so that it is >>>>> >>>>> s/kexec_fdt.c/kexec.c/ >>>> Missed that in the patch description. Will fix it. Thanks. >>>> >>>>>> accessible for other architectures as well. >>>>>> >>>>>> Co-developed-by: Prakhar Srivastava <prsriva@linux.microsoft.com> >>>>>> Signed-off-by: Prakhar Srivastava <prsriva@linux.microsoft.com> >>>>>> Signed-off-by: Lakshmi Ramasubramanian <nramas@linux.microsoft.com> >>>>>> --- >>>>>> arch/powerpc/include/asm/kexec.h | 1 - >>>>>> arch/powerpc/kexec/file_load.c | 32 ----------------- >>>>>> drivers/of/Makefile | 1 + >>>>>> drivers/of/kexec.c | 61 ++++++++++++++++++++++++++++++++ >>>>>> include/linux/kexec.h | 5 +++ >>>>>> 5 files changed, 67 insertions(+), 33 deletions(-) >>>>>> create mode 100644 drivers/of/kexec.c >>>>>> >>>>>> diff --git a/arch/powerpc/include/asm/kexec.h >>>>>> b/arch/powerpc/include/asm/kexec.h >>>>>> index 55d6ede30c19..7c223031ecdd 100644 >>>>>> --- a/arch/powerpc/include/asm/kexec.h >>>>>> +++ b/arch/powerpc/include/asm/kexec.h >>>>>> @@ -126,7 +126,6 @@ int setup_purgatory(struct kimage *image, const void >>>>>> *slave_code, >>>>>> int setup_new_fdt(const struct kimage *image, void *fdt, >>>>>> unsigned long initrd_load_addr, unsigned long initrd_len, >>>>>> const char *cmdline); >>>>>> -int delete_fdt_mem_rsv(void *fdt, unsigned long start, unsigned long size); >>>>>> #ifdef CONFIG_PPC64 >>>>>> struct kexec_buf; >>>>>> diff --git a/arch/powerpc/kexec/file_load.c b/arch/powerpc/kexec/file_load.c >>>>>> index 9a232bc36c8f..9efc98b1e2ae 100644 >>>>>> --- a/arch/powerpc/kexec/file_load.c >>>>>> +++ b/arch/powerpc/kexec/file_load.c >>>>>> @@ -109,38 +109,6 @@ int setup_purgatory(struct kimage *image, const void >>>>>> *slave_code, >>>>>> return 0; >>>>>> } >>>>>> -/** >>>>>> - * delete_fdt_mem_rsv - delete memory reservation with given address and >>>>>> size >>>>>> - * >>>>>> - * Return: 0 on success, or negative errno on error. >>>>>> - */ >>>>>> -int delete_fdt_mem_rsv(void *fdt, unsigned long start, unsigned long size) >>>>>> -{ >>>>>> - int i, ret, num_rsvs = fdt_num_mem_rsv(fdt); >>>>>> - >>>>>> - for (i = 0; i < num_rsvs; i++) { >>>>>> - uint64_t rsv_start, rsv_size; >>>>>> - >>>>>> - ret = fdt_get_mem_rsv(fdt, i, &rsv_start, &rsv_size); >>>>>> - if (ret) { >>>>>> - pr_err("Malformed device tree.\n"); >>>>>> - return -EINVAL; >>>>>> - } >>>>>> - >>>>>> - if (rsv_start == start && rsv_size == size) { >>>>>> - ret = fdt_del_mem_rsv(fdt, i); >>>>>> - if (ret) { >>>>>> - pr_err("Error deleting device tree reservation.\n"); >>>>>> - return -EINVAL; >>>>>> - } >>>>>> - >>>>>> - return 0; >>>>>> - } >>>>>> - } >>>>>> - >>>>>> - return -ENOENT; >>>>>> -} >>>>>> - >>>>>> /* >>>>>> * setup_new_fdt - modify /chosen and memory reservation for the next >>>>>> kernel >>>>>> * @image: kexec image being loaded. >>>>>> diff --git a/drivers/of/Makefile b/drivers/of/Makefile >>>>>> index 6e1e5212f058..77d24712c0c8 100644 >>>>>> --- a/drivers/of/Makefile >>>>>> +++ b/drivers/of/Makefile >>>>>> @@ -13,5 +13,6 @@ obj-$(CONFIG_OF_RESERVED_MEM) += of_reserved_mem.o >>>>>> obj-$(CONFIG_OF_RESOLVE) += resolver.o >>>>>> obj-$(CONFIG_OF_OVERLAY) += overlay.o >>>>>> obj-$(CONFIG_OF_NUMA) += of_numa.o >>>>>> +obj-$(CONFIG_OF_FLATTREE) += kexec.o >>>>> >>>>> Isn't this too broad? kexec.o will only be useful to kernel configs >>>>> which enable CONFIG_KEXEC_FILE, so perhaps do: >>>>> >>>>> ifdef CONFIG_OF_FLATTREE >>>>> ifdef CONFIG_KEXEC_FILE >>>>> obj-y += kexec.o >>>>> endif >>>>> endif >>>>> >>>>> What do you think? >>>> Per Rob's feedback on v9 patch set, I have moved all the architecture >>>> independent ima kexec functions to a single file "drivers/of/kexec.c" >>>> Since these functions are enabled on different kernel CONFIGs, I have >>>> used IS_ENABLED(CONFIG_XYZ) macro instead of "#ifdef" in the C file to >>>> conditionally compile. >>> Per Rob's feedback on the v9 patch, I'll keep the ima kexec functions in a >>> single file (in "drivers/of/kexec.c") and use IS_ENABLED() macro to handle the >>> function calls. >>> >>> I'll make the other changes you'd suggested on v10 patches and will post v11 >>> patch set shortly. >> >>>From a cursory look at the use of functions in this file, I got the >> impression that there wouldn't be any reference to them in kernel >> configs that didn't have CONFIG_KEXEC_FILE enabled, which is why I >> suggested the change above. I think you can make it without any other >> changes to the code. >> I could be wrong though, and there could be some config which tried to >> use some of these functions even when CONFIG_KEXEC_FILE is disabled. In >> that case, the customary way to resolve it is to provide static inline >> stub versions in a header file (not in a .c file) of just those >> functions that are needed. >> The reason why placing stub functions in header files is better is that >> then the compiler has visibility of the dummy function when compiling >> the source file which uses the function, and is able to eliminate the >> dead code that arises from the function always returning one value. > > I agree with you Thiago. > > Is there a way to keep all the relevant functions in a single C file, not use > "#ifdef" in C file, and follow the coding pattern you've described above (i > mean, "defining a stub function in a header file when the config conditions are > not met")? Like I said above, my impression is that you don't need any ifdefs since my understanding is that all funcitons in the C file are only used when CONFIG_KEXEC_FILE is set. Do you anticipate (or have found) problems with that? -- Thiago Jung Bauermann IBM Linux Technology Center
On kexec file load Integrity Measurement Architecture (IMA) subsystem may verify the IMA signature of the kernel and initramfs, and measure it. The command line parameters passed to the kernel in the kexec call may also be measured by IMA. A remote attestation service can verify a TPM quote based on the TPM event log, the IMA measurement list, and the TPM PCR data. This can be achieved only if the IMA measurement log is carried over from the current kernel to the next kernel across the kexec call. powerpc already supports carrying forward the IMA measurement log on kexec. This patch set adds support for carrying forward the IMA measurement log on kexec on ARM64. This patch set moves the platform independent code defined for powerpc such that it can be reused for other platforms as well. A chosen node "linux,ima-kexec-buffer" is added to the DTB for ARM64 to hold the address and the size of the memory reserved to carry the IMA measurement log. This patch set has been tested for ARM64 platform using QEMU. I would like help from the community for testing this change on powerpc. Thanks. This patch set is based on commit e87297fa080a ("Merge tag 'drm-fixes-2020-12-04' of git://anongit.freedesktop.org/drm/drm") in https://github.com/torvalds/linux "master" branch. Changelog: v10 - Moved delete_fdt_mem_rsv(), remove_ima_buffer(), get_ima_kexec_buffer, and get_root_addr_size_cells() to drivers/of/kexec.c - Moved arch_ima_add_kexec_buffer() to security/integrity/ima/ima_kexec.c - Conditionally define IMA buffer fields in struct kimage_arch v9 - Moved delete_fdt_mem_rsv() to drivers/of/kexec_fdt.c - Defined a new function get_ima_kexec_buffer() in drivers/of/ima_kexec.c to replace do_get_kexec_buffer() - Changed remove_ima_kexec_buffer() to the original function name remove_ima_buffer() - Moved remove_ima_buffer() to drivers/of/ima_kexec.c - Moved ima_get_kexec_buffer() and ima_free_kexec_buffer() to security/integrity/ima/ima_kexec.c v8: - Moved remove_ima_kexec_buffer(), do_get_kexec_buffer(), and delete_fdt_mem_rsv() to drivers/of/fdt.c - Moved ima_dump_measurement_list() and ima_add_kexec_buffer() back to security/integrity/ima/ima_kexec.c v7: - Renamed remove_ima_buffer() to remove_ima_kexec_buffer() and moved this function definition to kernel. - Moved delete_fdt_mem_rsv() definition to kernel - Moved ima_dump_measurement_list() and ima_add_kexec_buffer() to a new file namely ima_kexec_fdt.c in IMA v6: - Remove any existing FDT_PROP_IMA_KEXEC_BUFFER property in the device tree and also its corresponding memory reservation in the currently running kernel. - Moved the function remove_ima_buffer() defined for powerpc to IMA and renamed the function to ima_remove_kexec_buffer(). Also, moved delete_fdt_mem_rsv() from powerpc to IMA. v5: - Merged get_addr_size_cells() and do_get_kexec_buffer() into a single function when moving the arch independent code from powerpc to IMA - Reverted the change to use FDT functions in powerpc code and added back the original code in get_addr_size_cells() and do_get_kexec_buffer() for powerpc. - Added fdt_add_mem_rsv() for ARM64 to reserve the memory for the IMA log buffer during kexec. - Fixed the warning reported by kernel test bot for ARM64 arch_ima_add_kexec_buffer() - moved this function to a new file namely arch/arm64/kernel/ima_kexec.c v4: - Submitting the patch series on behalf of the original author Prakhar Srivastava <prsriva@linux.microsoft.com> - Moved FDT_PROP_IMA_KEXEC_BUFFER ("linux,ima-kexec-buffer") to libfdt.h so that it can be shared by multiple platforms. v3: Breakup patches further into separate patches. - Refactoring non architecture specific code out of powerpc - Update powerpc related code to use fdt functions - Update IMA buffer read related code to use of functions - Add support to store the memory information of the IMA measurement logs to be carried forward. - Update the property strings to align with documented nodes https://github.com/devicetree-org/dt-schema/pull/46 v2: Break patches into separate patches. - Powerpc related Refactoring - Updating the docuemntation for chosen node - Updating arm64 to support IMA buffer pass v1: Refactoring carrying over IMA measuremnet logs over Kexec. This patch moves the non-architecture specific code out of powerpc and adds to security/ima.(Suggested by Thiago) Add Documentation regarding the ima-kexec-buffer node in the chosen node documentation v0: Add a layer of abstraction to use the memory reserved by device tree for ima buffer pass. Add support for ima buffer pass using reserved memory for arm64 kexec. Update the arch sepcific code path in kexec file load to store the ima buffer in the reserved memory. The same reserved memory is read on kexec or cold boot. Lakshmi Ramasubramanian (8): powerpc: fix compiler warnings and errors powerpc: Move delete_fdt_mem_rsv() to drivers/of/kexec.c powerpc: Move ima buffer functions to drivers/of/kexec.c powerpc: Use ima kexec node functions powerpc: Move remove_ima_buffer() to drivers/of/kexec.c powerpc: Move ima_get_kexec_buffer() and ima_free_kexec_buffer() to ima powerpc: Move arch_ima_add_kexec_buffer to ima arm64: Add IMA log information in kimage used for kexec arch/arm64/Kconfig | 1 + arch/arm64/include/asm/ima.h | 22 ++++ arch/arm64/include/asm/kexec.h | 5 + arch/arm64/kernel/Makefile | 1 + arch/arm64/kernel/ima.c | 41 +++++++ arch/arm64/kernel/machine_kexec_file.c | 8 ++ arch/powerpc/include/asm/ima.h | 14 +-- arch/powerpc/include/asm/kexec.h | 1 - arch/powerpc/kexec/Makefile | 7 +- arch/powerpc/kexec/file_load.c | 32 ----- arch/powerpc/kexec/ima.c | 137 +-------------------- drivers/of/Makefile | 1 + drivers/of/kexec.c | 160 +++++++++++++++++++++++++ include/linux/kexec.h | 9 ++ security/integrity/ima/ima_kexec.c | 73 +++++++++++ 15 files changed, 326 insertions(+), 186 deletions(-) create mode 100644 arch/arm64/include/asm/ima.h create mode 100644 arch/arm64/kernel/ima.c create mode 100644 drivers/of/kexec.c