powerpc/fadump: process the crashdump by exporting it as /proc/vmcore

Add support in the kernel to process the crash'ed kernel's memory
preserved during MPIPL and export it as /proc/vmcore file for the
userland scripts to filter and analyze it later.

Signed-off-by: Hari Bathini <hbathini@linux.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/156821351482.5656.6255805804744333073.stgit@hbathini.in.ibm.com

diff --git a/arch/powerpc/platforms/powernv/opal-fadump.c b/arch/powerpc/platforms/powernv/opal-fadump.c
index a4b96a7..5ace7c4 100644 (file)
--- a/arch/powerpc/platforms/powernv/opal-fadump.c
+++ b/arch/powerpc/platforms/powernv/opal-fadump.c
@@ -12,6 +12,7 @@
 #include <linux/of_fdt.h>
 #include <linux/libfdt.h>
 #include <linux/mm.h>
+#include <linux/crash_dump.h>
 
 #include <asm/page.h>
 #include <asm/opal.h>
@@ -19,6 +20,7 @@
 
 #include "opal-fadump.h"
 
+static const struct opal_fadump_mem_struct *opal_fdm_active;
 static struct opal_fadump_mem_struct *opal_fdm;
 
 static int opal_fadump_unregister(struct fw_dump *fadump_conf);
@@ -39,6 +41,37 @@ static void opal_fadump_update_config(struct fw_dump *fadump_conf,
        fadump_conf->fadumphdr_addr = fdm->fadumphdr_addr;
 }
 
+/*
+ * This function is called in the capture kernel to get configuration details
+ * from metadata setup by the first kernel.
+ */
+static void opal_fadump_get_config(struct fw_dump *fadump_conf,
+                                  const struct opal_fadump_mem_struct *fdm)
+{
+       int i;
+
+       if (!fadump_conf->dump_active)
+               return;
+
+       fadump_conf->boot_memory_size = 0;
+
+       pr_debug("Boot memory regions:\n");
+       for (i = 0; i < fdm->region_cnt; i++) {
+               pr_debug("\t%d. base: 0x%llx, size: 0x%llx\n",
+                        (i + 1), fdm->rgn[i].src, fdm->rgn[i].size);
+
+               fadump_conf->boot_memory_size += fdm->rgn[i].size;
+       }
+
+       /*
+        * Start address of reserve dump area (permanent reservation) for
+        * re-registering FADump after dump capture.
+        */
+       fadump_conf->reserve_dump_area_start = fdm->rgn[0].dest;
+
+       opal_fadump_update_config(fadump_conf, fdm);
+}
+
 /* Initialize kernel metadata */
 static void opal_fadump_init_metadata(struct opal_fadump_mem_struct *fdm)
 {
@@ -209,24 +242,97 @@ static void opal_fadump_cleanup(struct fw_dump *fadump_conf)
                pr_warn("Could not reset (%llu) kernel metadata tag!\n", ret);
 }
 
+/*
+ * Convert CPU state data saved at the time of crash into ELF notes.
+ *
+ * Append crashing CPU's register data saved by the kernel in the PT_NOTE.
+ */
+static int __init
+opal_fadump_build_cpu_notes(struct fw_dump *fadump_conf,
+                           struct fadump_crash_info_header *fdh)
+{
+       u32 num_cpus = 1, *note_buf;
+       int rc;
+
+       if (fdh->crashing_cpu == FADUMP_CPU_UNKNOWN)
+               return -ENODEV;
+
+       /* Allocate CPU notes buffer to hold crashing cpu notes. */
+       rc = fadump_setup_cpu_notes_buf(num_cpus);
+       if (rc != 0)
+               return rc;
+
+       note_buf = (u32 *)fadump_conf->cpu_notes_buf_vaddr;
+       note_buf = fadump_regs_to_elf_notes(note_buf, &(fdh->regs));
+       final_note(note_buf);
+
+       pr_debug("Updating elfcore header (%llx) with cpu notes\n",
+                fdh->elfcorehdr_addr);
+       fadump_update_elfcore_header(__va(fdh->elfcorehdr_addr));
+       return 0;
+}
+
 static int __init opal_fadump_process(struct fw_dump *fadump_conf)
 {
-       return -EINVAL;
+       struct fadump_crash_info_header *fdh;
+       int rc = -EINVAL;
+
+       if (!opal_fdm_active || !fadump_conf->fadumphdr_addr)
+               return rc;
+
+       /* Validate the fadump crash info header */
+       fdh = __va(fadump_conf->fadumphdr_addr);
+       if (fdh->magic_number != FADUMP_CRASH_INFO_MAGIC) {
+               pr_err("Crash info header is not valid.\n");
+               return rc;
+       }
+
+       rc = opal_fadump_build_cpu_notes(fadump_conf, fdh);
+       if (rc)
+               return rc;
+
+       /*
+        * We are done validating dump info and elfcore header is now ready
+        * to be exported. set elfcorehdr_addr so that vmcore module will
+        * export the elfcore header through '/proc/vmcore'.
+        */
+       elfcorehdr_addr = fdh->elfcorehdr_addr;
+
+       return rc;
 }
 
 static void opal_fadump_region_show(struct fw_dump *fadump_conf,
                                    struct seq_file *m)
 {
-       const struct opal_fadump_mem_struct *fdm_ptr = opal_fdm;
+       const struct opal_fadump_mem_struct *fdm_ptr;
        u64 dumped_bytes = 0;
        int i;
 
+       if (fadump_conf->dump_active)
+               fdm_ptr = opal_fdm_active;
+       else
+               fdm_ptr = opal_fdm;
+
        for (i = 0; i < fdm_ptr->region_cnt; i++) {
+               /*
+                * Only regions that are registered for MPIPL
+                * would have dump data.
+                */
+               if ((fadump_conf->dump_active) &&
+                   (i < fdm_ptr->registered_regions))
+                       dumped_bytes = fdm_ptr->rgn[i].size;
+
                seq_printf(m, "DUMP: Src: %#016llx, Dest: %#016llx, ",
                           fdm_ptr->rgn[i].src, fdm_ptr->rgn[i].dest);
                seq_printf(m, "Size: %#llx, Dumped: %#llx bytes\n",
                           fdm_ptr->rgn[i].size, dumped_bytes);
        }
+
+       /* Dump is active. Show reserved area start address. */
+       if (fadump_conf->dump_active) {
+               seq_printf(m, "\nMemory above %#016lx is reserved for saving crash dump\n",
+                          fadump_conf->reserve_dump_area_start);
+       }
 }
 
 static void opal_fadump_trigger(struct fadump_crash_info_header *fdh,
@@ -257,7 +363,11 @@ static struct fadump_ops opal_fadump_ops = {
 
 void __init opal_fadump_dt_scan(struct fw_dump *fadump_conf, u64 node)
 {
+       const __be32 *prop;
        unsigned long dn;
+       u64 addr = 0;
+       s64 ret;
+
 
        /*
         * Check if Firmware-Assisted Dump is supported. if yes, check
@@ -276,4 +386,37 @@ void __init opal_fadump_dt_scan(struct fw_dump *fadump_conf, u64 node)
 
        fadump_conf->ops                = &opal_fadump_ops;
        fadump_conf->fadump_supported   = 1;
+
+       /*
+        * Check if dump has been initiated on last reboot.
+        */
+       prop = of_get_flat_dt_prop(dn, "mpipl-boot", NULL);
+       if (!prop)
+               return;
+
+       ret = opal_mpipl_query_tag(OPAL_MPIPL_TAG_KERNEL, &addr);
+       if ((ret != OPAL_SUCCESS) || !addr) {
+               pr_err("Failed to get Kernel metadata (%lld)\n", ret);
+               return;
+       }
+
+       addr = be64_to_cpu(addr);
+       pr_debug("Kernel metadata addr: %llx\n", addr);
+
+       opal_fdm_active = __va(addr);
+       if (opal_fdm_active->version != OPAL_FADUMP_VERSION) {
+               pr_warn("Supported kernel metadata version: %u, found: %d!\n",
+                       OPAL_FADUMP_VERSION, opal_fdm_active->version);
+               pr_warn("WARNING: Kernel metadata format mismatch identified! Core file maybe corrupted..\n");
+       }
+
+       /* Kernel regions not registered with f/w for MPIPL */
+       if (opal_fdm_active->registered_regions == 0) {
+               opal_fdm_active = NULL;
+               return;
+       }
+
+       pr_info("Firmware-assisted dump is active.\n");
+       fadump_conf->dump_active = 1;
+       opal_fadump_get_config(fadump_conf, opal_fdm_active);
 }