Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752579AbdGFWN5 (ORCPT ); Thu, 6 Jul 2017 18:13:57 -0400 Received: from mail-it0-f65.google.com ([209.85.214.65]:35312 "EHLO mail-it0-f65.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751943AbdGFWN4 (ORCPT ); Thu, 6 Jul 2017 18:13:56 -0400 MIME-Version: 1.0 In-Reply-To: <20170706215233.11329-3-ross.zwisler@linux.intel.com> References: <20170706215233.11329-1-ross.zwisler@linux.intel.com> <20170706215233.11329-3-ross.zwisler@linux.intel.com> From: "Rafael J. Wysocki" Date: Fri, 7 Jul 2017 00:13:54 +0200 X-Google-Sender-Auth: GiTXclNfrH_wi5KCb2yScYxrFuc Message-ID: Subject: Re: [RFC v2 2/5] acpi: HMAT support in acpi_parse_entries_array() To: Ross Zwisler Cc: Linux Kernel Mailing List , "Anaczkowski, Lukasz" , "Box, David E" , "Kogut, Jaroslaw" , "Lahtinen, Joonas" , "Moore, Robert" , "Nachimuthu, Murugasamy" , "Odzioba, Lukasz" , "Rafael J. Wysocki" , "Rafael J. Wysocki" , "Schmauss, Erik" , "Verma, Vishal L" , "Zheng, Lv" , Andrew Morton , Dan Williams , Dave Hansen , Greg Kroah-Hartman , Jerome Glisse , Len Brown , Tim Chen , "devel@acpica.org" , ACPI Devel Maling List , Linux Memory Management List , "linux-nvdimm@lists.01.org" Content-Type: text/plain; charset="UTF-8" Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 5339 Lines: 147 On Thu, Jul 6, 2017 at 11:52 PM, Ross Zwisler wrote: > The current implementation of acpi_parse_entries_array() assumes that each > subtable has a standard ACPI subtable entry of type struct > acpi_sutbable_header. This standard subtable header has a one byte length > followed by a one byte type. > > The HMAT subtables have to allow for a longer length so they have subtable > headers of type struct acpi_hmat_structure which has a 2 byte type and a 4 > byte length. > > Enhance the subtable parsing in acpi_parse_entries_array() so that it can > handle these new HMAT subtables. > > Signed-off-by: Ross Zwisler > --- > drivers/acpi/numa.c | 2 +- > drivers/acpi/tables.c | 52 ++++++++++++++++++++++++++++++++++++++++----------- > 2 files changed, 42 insertions(+), 12 deletions(-) > > diff --git a/drivers/acpi/numa.c b/drivers/acpi/numa.c > index edb0c79..917f1cc 100644 > --- a/drivers/acpi/numa.c > +++ b/drivers/acpi/numa.c > @@ -443,7 +443,7 @@ int __init acpi_numa_init(void) > * So go over all cpu entries in SRAT to get apicid to node mapping. > */ > > - /* SRAT: Static Resource Affinity Table */ > + /* SRAT: System Resource Affinity Table */ > if (!acpi_table_parse(ACPI_SIG_SRAT, acpi_parse_srat)) { > struct acpi_subtable_proc srat_proc[3]; > This change is unrelated to the rest of the patch. Maybe send it separately? > diff --git a/drivers/acpi/tables.c b/drivers/acpi/tables.c > index ff42539..7979171 100644 > --- a/drivers/acpi/tables.c > +++ b/drivers/acpi/tables.c > @@ -218,6 +218,33 @@ void acpi_table_print_madt_entry(struct acpi_subtable_header *header) > } > } > > +static unsigned long __init > +acpi_get_entry_type(char *id, void *entry) > +{ > + if (!strncmp(id, ACPI_SIG_HMAT, 4)) > + return ((struct acpi_hmat_structure *)entry)->type; > + else > + return ((struct acpi_subtable_header *)entry)->type; > +} I slightly prefer to use ? : in similar situations. > + > +static unsigned long __init > +acpi_get_entry_length(char *id, void *entry) > +{ > + if (!strncmp(id, ACPI_SIG_HMAT, 4)) > + return ((struct acpi_hmat_structure *)entry)->length; > + else > + return ((struct acpi_subtable_header *)entry)->length; > +} > + > +static unsigned long __init > +acpi_get_subtable_header_length(char *id) > +{ > + if (!strncmp(id, ACPI_SIG_HMAT, 4)) > + return sizeof(struct acpi_hmat_structure); > + else > + return sizeof(struct acpi_subtable_header); > +} > + > /** > * acpi_parse_entries_array - for each proc_num find a suitable subtable > * > @@ -242,10 +269,10 @@ acpi_parse_entries_array(char *id, unsigned long table_size, > struct acpi_subtable_proc *proc, int proc_num, > unsigned int max_entries) > { > - struct acpi_subtable_header *entry; > - unsigned long table_end; > + unsigned long table_end, subtable_header_length; > int count = 0; > int errs = 0; > + void *entry; > int i; > > if (acpi_disabled) > @@ -263,19 +290,23 @@ acpi_parse_entries_array(char *id, unsigned long table_size, > } > > table_end = (unsigned long)table_header + table_header->length; > + subtable_header_length = acpi_get_subtable_header_length(id); > > /* Parse all entries looking for a match. */ > > - entry = (struct acpi_subtable_header *) > - ((unsigned long)table_header + table_size); > + entry = (void *)table_header + table_size; > + > + while (((unsigned long)entry) + subtable_header_length < table_end) { > + unsigned long entry_type, entry_length; > > - while (((unsigned long)entry) + sizeof(struct acpi_subtable_header) < > - table_end) { > if (max_entries && count >= max_entries) > break; > > + entry_type = acpi_get_entry_type(id, entry); > + entry_length = acpi_get_entry_length(id, entry); > + > for (i = 0; i < proc_num; i++) { > - if (entry->type != proc[i].id) > + if (entry_type != proc[i].id) > continue; > if (!proc[i].handler || > (!errs && proc[i].handler(entry, table_end))) { > @@ -290,16 +321,15 @@ acpi_parse_entries_array(char *id, unsigned long table_size, > count++; > > /* > - * If entry->length is 0, break from this loop to avoid > + * If entry_length is 0, break from this loop to avoid > * infinite loop. > */ > - if (entry->length == 0) { > + if (entry_length == 0) { > pr_err("[%4.4s:0x%02x] Invalid zero length\n", id, proc->id); > return -EINVAL; > } > > - entry = (struct acpi_subtable_header *) > - ((unsigned long)entry + entry->length); > + entry += entry_length; > } > > if (max_entries && count > max_entries) { > -- The rest is fine by me. Thanks, Rafael