From: David Gibson Subject: dtc: Automatically pick a sensible boot_cpuid_phys Date: Fri, 19 Feb 2010 15:50:50 +1100 Message-ID: <20100219045050.GJ29038@yookeroo> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: Content-Disposition: inline List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: devicetree-discuss-bounces+gldd-devicetree-discuss=m.gmane.org-uLR06cmDAlY/bJ5BZ2RsiQ@public.gmane.org Errors-To: devicetree-discuss-bounces+gldd-devicetree-discuss=m.gmane.org-uLR06cmDAlY/bJ5BZ2RsiQ@public.gmane.org To: Jon Loeliger Cc: devicetree-discuss-uLR06cmDAlY/bJ5BZ2RsiQ@public.gmane.org List-Id: devicetree@vger.kernel.org Currently, when in -Idts -Odtb or -Ifs -Odtb modes, dtc always defaults to using 0 as the value for the boot_cpuid_phys header field. That's correct quite often, but there are some systems where there is no CPU with hardware ID of 0, or where we don't want to use the CPU with hardware ID 0 at all (e.g. for AMP-style partitioning). The only way to override this default currently, is with the -b command line option. This patch improves dtc to instead base the default boot_cpuid_phys value on the reg property of the first listed subnode of /cpus. This means that dtc will get boot_cpuid_phys correct by default in a greater proportion of cases (since the boot cpu is usually listed first, and this way at least the boot_cpuid_phys default will match some existing cpu node). If the node doesn't exist or has an invalid 'reg' property (missing or not 4 bytes in length), then boot_cpuid_phys is set to 0. Signed-off-by: David Gibson Index: dtc/livetree.c =================================================================== --- dtc.orig/livetree.c 2010-02-19 15:48:36.145429911 +1100 +++ dtc/livetree.c 2010-02-19 15:48:38.109450465 +1100 @@ -318,3 +318,26 @@ cell_t get_node_phandle(struct node *roo return node->phandle; } + +uint32_t guess_boot_cpuid(struct node *tree) +{ + struct node *cpus, *bootcpu; + struct property *reg; + + cpus = get_node_by_path(tree, "/cpus"); + if (!cpus) + return 0; + + + bootcpu = cpus->children; + if (!bootcpu) + return 0; + + reg = get_property(bootcpu, "reg"); + if (!reg || (reg->val.len != sizeof(uint32_t))) + return 0; + + /* FIXME: Sanity check node? */ + + return propval_cell(reg); +} Index: dtc/dtc.h =================================================================== --- dtc.orig/dtc.h 2010-02-19 15:48:36.113452731 +1100 +++ dtc/dtc.h 2010-02-19 15:48:38.109450465 +1100 @@ -178,6 +178,8 @@ struct node *get_node_by_phandle(struct struct node *get_node_by_ref(struct node *tree, const char *ref); cell_t get_node_phandle(struct node *root, struct node *node); +uint32_t guess_boot_cpuid(struct node *tree); + /* Boot info (tree plus memreserve information */ struct reserve_info { Index: dtc/dtc-parser.y =================================================================== --- dtc.orig/dtc-parser.y 2010-02-19 15:48:36.189449547 +1100 +++ dtc/dtc-parser.y 2010-02-19 15:48:38.109450465 +1100 @@ -85,7 +85,8 @@ static unsigned long long eval_literal(c sourcefile: DT_V1 ';' memreserves devicetree { - the_boot_info = build_boot_info($3, $4, 0); + the_boot_info = build_boot_info($3, $4, + guess_boot_cpuid($4)); } ; Index: dtc/fstree.c =================================================================== --- dtc.orig/fstree.c 2010-02-19 15:48:36.153429392 +1100 +++ dtc/fstree.c 2010-02-19 15:48:38.113439869 +1100 @@ -87,6 +87,6 @@ struct boot_info *dt_from_fs(const char tree = read_fstree(dirname); tree = name_node(tree, "", NULL); - return build_boot_info(NULL, tree, 0); + return build_boot_info(NULL, tree, guess_boot_cpuid(tree)); } Index: dtc/tests/boot-cpuid.dts =================================================================== --- /dev/null 1970-01-01 00:00:00.000000000 +0000 +++ dtc/tests/boot-cpuid.dts 2010-02-19 15:48:38.113439869 +1100 @@ -0,0 +1,16 @@ +/dts-v1/; + +/ { + cpus { + cpu@10 { + device_type = "cpu"; + compatible = "fake-cpu"; + reg = <0x10>; + }; + cpu@11 { + device_type = "cpu"; + compatible = "fake-cpu"; + reg = <0x11>; + }; + }; +}; Index: dtc/tests/run_tests.sh =================================================================== --- dtc.orig/tests/run_tests.sh 2010-02-19 15:48:37.469450760 +1100 +++ dtc/tests/run_tests.sh 2010-02-19 15:49:29.099425093 +1100 @@ -243,12 +243,26 @@ dtc_tests () { run_test incbin incbin.test.dtb # Check boot_cpuid_phys handling - run_dtc_test -I dts -O dtb -b 17 -o boot_cpuid.test.dtb empty.dts - run_test boot-cpuid boot_cpuid.test.dtb 17 - run_dtc_test -I dtb -O dtb -b 17 -o boot_cpuid_test_tree1.test.dtb test_tree1.dtb - run_test boot-cpuid boot_cpuid_test_tree1.test.dtb 17 - run_dtc_test -I dtb -O dtb -o boot_cpuid_preserved_test_tree1.test.dtb boot_cpuid_test_tree1.test.dtb - run_test dtbs_equal_ordered boot_cpuid_preserved_test_tree1.test.dtb boot_cpuid_test_tree1.test.dtb + run_dtc_test -I dts -O dtb -o boot_cpuid.test.dtb boot-cpuid.dts + run_test boot-cpuid boot_cpuid.test.dtb 16 + + run_dtc_test -I dts -O dtb -b 17 -o boot_cpuid_17.test.dtb boot-cpuid.dts + run_test boot-cpuid boot_cpuid_17.test.dtb 17 + + run_dtc_test -I dtb -O dtb -o preserve_boot_cpuid.test.dtb boot_cpuid.test.dtb + run_test boot-cpuid preserve_boot_cpuid.test.dtb 16 + run_test dtbs_equal_ordered preserve_boot_cpuid.test.dtb boot_cpuid.test.dtb + + run_dtc_test -I dtb -O dtb -o preserve_boot_cpuid_17.test.dtb boot_cpuid_17.test.dtb + run_test boot-cpuid preserve_boot_cpuid_17.test.dtb 17 + run_test dtbs_equal_ordered preserve_boot_cpuid_17.test.dtb boot_cpuid_17.test.dtb + + run_dtc_test -I dtb -O dtb -b17 -o override17_boot_cpuid.test.dtb boot_cpuid.test.dtb + run_test boot-cpuid override17_boot_cpuid.test.dtb 17 + + run_dtc_test -I dtb -O dtb -b0 -o override0_boot_cpuid_17.test.dtb boot_cpuid_17.test.dtb + run_test boot-cpuid override0_boot_cpuid_17.test.dtb 0 + # Check -Oasm mode for tree in test_tree1.dts escapes.dts references.dts path-references.dts \ -- David Gibson | I'll have my music baroque, and my code david AT gibson.dropbear.id.au | minimalist, thank you. NOT _the_ _other_ | _way_ _around_! http://www.ozlabs.org/~dgibson