Date: Mon, 28 Apr 2003 14:03:14 -0700 (PDT) From: Marcel Moolenaar <marcel@FreeBSD.org> To: Perforce Change Reviews <perforce@freebsd.org> Subject: PERFORCE change 29975 for review Message-ID: <200304282103.h3SL3Eep083867@repoman.freebsd.org>
next in thread | raw e-mail | index | archive | help
http://perforce.freebsd.org/chv.cgi?CH=29975 Change 29975 by marcel@marcel_nfs on 2003/04/28 14:02:44 Use the endian encoding/decoding functions to get 64-bit values out of the XSDT. They are very likely misaligned and handling of misaligned memory accesses by the kernel must not be depended upon. In other words: we coredump on ia64 because handling of unaligned memory access has a bug (a kernel from HEAD -- No EPC related fodder :-) This makes acpidump usable again on my bigsur. I can't recall having seen coredumps on pluto, which is where I would expect to see them... Affected files ... .. //depot/projects/ia64/usr.sbin/acpi/acpidump/acpi.c#11 edit Differences ... ==== //depot/projects/ia64/usr.sbin/acpi/acpidump/acpi.c#11 (text+ko) ==== @@ -28,6 +28,7 @@ * $FreeBSD: src/usr.sbin/acpi/acpidump/acpi.c,v 1.7 2003/04/15 05:15:50 njl Exp $ */ +#include <sys/endian.h> #include <sys/param.h> #include <sys/stat.h> @@ -474,7 +475,7 @@ { struct ACPIsdt *sp; - sp = acpi_map_physical(pa, sizeof(struct ACPIsdt)); + sp = acpi_map_physical(pa, SIZEOF_SDT_HDR); sp = acpi_map_physical(pa, sp->len); return (sp); } @@ -506,7 +507,7 @@ if (p != (void*)rsdp->body) printf(", "); if (rsdp->signature[0] == 'X') { - printf("0x%016x", *((u_int64_t*)p)); + printf("0x%016x", le64dec(p)); p += sizeof(u_int64_t); } else { printf("0x%08x", *((u_int32_t*)p)); @@ -521,7 +522,7 @@ p = (void*)rsdp->body; while (p < end) { if (rsdp->signature[0] == 'X') { - sdp = acpi_map_sdt(*((u_int64_t*)p)); + sdp = acpi_map_sdt(le64dec(p)); p += sizeof(u_int64_t); } else { sdp = acpi_map_sdt(*((u_int32_t*)p));
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200304282103.h3SL3Eep083867>