Date: Mon, 29 Jun 2009 17:33:12 +0200 From: Michael Gmelin <freebsdusb@bindone.de> To: Rui Paulo <rpaulo@freebsd.org>, freebsd-acpi@freebsd.org Cc: "Paul B. Mahol" <onemda@gmail.com> Subject: New patchset for acpi_wmi/acpi_hp Message-ID: <4A48DEB8.1040808@bindone.de>
next in thread | raw e-mail | index | archive | help
[-- Attachment #1 --]
Hello,
please find attached more patches against CURRENT of today. These
patches include the patches I've sent on the 27th.
So the complete feature list of this patchset is:
acpi_wmi_if:
Document different semantics for ACPI_WMI_PROVIDES_GUID_STRING_METHOD
acpi_wmi.c:
Modify acpi_wmi_provides_guid_string_method to return absolut number of
instances known for the given GUID.
acpi_hp.c:
- sysctl dev.acpi_hp.0.verbose to toggle debug output
- A modification so this can deal with different array lengths
when reading the CMI BIOS - now it works ok on HP Compaq nx7300
as well.
- Change behaviour to query only max_instance-1 CMI BIOS instances,
because all HPs seen so far are broken in that respect
(or there is a fundamental misunderstanding on my side, possible
as well). This way a disturbing ACPI Error Field exceeds Buffer
message is avoided.
- New bit to set on dev.acpi_hp.0.cmi_detail (0x8) to
also query the highest guid instance of CMI bios
acpi_hp.4:
- Document dev.acpi_hp.0.verbose sysctl in man page
- Document new bit for dev.acpi_hp.0.cmi_detail
- Add a section to manpage about hardware that has been reported
to work ok
Installation instructions (against latest CURRENT):
patch -d /usr/src < /path/to/acpi_wmi_acpi_hp.patch
cd /usr/src/sys/modules/acpi/acpi_wmi
make all && make install
cd /usr/src/sys/modules/acpi/acpi_hp
make all && make install
cd /usr/src/share/man/man4
make all && make install
cheers
Michael
[-- Attachment #2 --]
--- share/man/man4/acpi_hp.4.orig 2009-06-26 13:03:16.331066657 +0200
+++ share/man/man4/acpi_hp.4 2009-06-29 17:19:18.544247949 +0200
@@ -165,6 +165,9 @@
Show a list of valid options for the BIOS setting
.It Li 0x04
Show additional flags of BIOS setting (ReadOnly etc.)
+.It Li 0x08
+Query highest BIOS entry instance. This is broken on many HP models and
+therefore disabled by default.
.El
.It Va dev.acpi_hp.0.verbose
(read-only)
--- sys/dev/acpi_support/acpi_hp.c.orig 2009-06-26 12:54:46.509994426 +0200
+++ sys/dev/acpi_support/acpi_hp.c 2009-06-29 17:18:30.928244238 +0200
@@ -107,6 +107,7 @@
#define ACPI_HP_CMI_DETAIL_PATHS 0x01
#define ACPI_HP_CMI_DETAIL_ENUMS 0x02
#define ACPI_HP_CMI_DETAIL_FLAGS 0x04
+#define ACPI_HP_CMI_DETAIL_SHOW_MAX_INSTANCE 0x08
struct acpi_hp_inst_seq_pair {
UINT32 sequence; /* sequence number as suggested by cmi bios */
@@ -505,9 +506,10 @@
sc->has_notify = 1;
}
}
- if (ACPI_WMI_PROVIDES_GUID_STRING(sc->wmi_dev, ACPI_HP_WMI_CMI_GUID)) {
+ if ((sc->has_cmi =
+ ACPI_WMI_PROVIDES_GUID_STRING(sc->wmi_dev, ACPI_HP_WMI_CMI_GUID)
+ )) {
device_printf(dev, "HP CMI GUID detected\n");
- sc->has_cmi = 1;
}
if (sc->has_cmi) {
@@ -772,6 +774,10 @@
arg?1:0));
case ACPI_HP_METHOD_CMI_DETAIL:
sc->cmi_detail = arg;
+ if ((arg & ACPI_HP_CMI_DETAIL_SHOW_MAX_INSTANCE) !=
+ (oldarg & ACPI_HP_CMI_DETAIL_SHOW_MAX_INSTANCE)) {
+ sc->cmi_order_size = -1;
+ }
break;
case ACPI_HP_METHOD_VERBOSE:
sc->verbose = arg;
@@ -1122,6 +1128,7 @@
struct acpi_hp_softc *sc;
int pos, i, l, ret;
UINT8 instance;
+ UINT8 maxInstance;
UINT32 sequence;
int linesize = 1025;
char line[linesize];
@@ -1138,14 +1145,20 @@
else {
if (!sbuf_done(&sc->hpcmi_sbuf)) {
if (sc->cmi_order_size < 0) {
+ maxInstance = sc->has_cmi;
+ if (!(sc->cmi_detail &
+ ACPI_HP_CMI_DETAIL_SHOW_MAX_INSTANCE) &&
+ maxInstance > 0) {
+ maxInstance--;
+ }
sc->cmi_order_size = 0;
- for (instance = 0; instance < 128;
+ for (instance = 0; instance < maxInstance;
++instance) {
if (acpi_hp_get_cmi_block(sc->wmi_dev,
ACPI_HP_WMI_CMI_GUID, instance,
line, linesize, &sequence,
sc->cmi_detail)) {
- instance = 128;
+ instance = maxInstance;
}
else {
pos = sc->cmi_order_size;
--- sys/dev/acpi_support/acpi_wmi.c.orig 2009-06-21 22:27:26.897414000 +0200
+++ sys/dev/acpi_support/acpi_wmi.c 2009-06-29 17:17:39.554824991 +0200
@@ -326,11 +326,13 @@
static int
acpi_wmi_provides_guid_string_method(device_t dev, const char *guid_string)
{
+ struct wmi_info *winfo;
int ret;
ACPI_FUNCTION_TRACE((char *)(uintptr_t)__func__);
ACPI_SERIAL_BEGIN(acpi_wmi);
- ret = (acpi_wmi_lookup_wmi_info_by_guid_string(guid_string) == NULL)?0:1;
+ winfo = acpi_wmi_lookup_wmi_info_by_guid_string(guid_string);
+ ret = (winfo == NULL)?0:winfo->ginfo.max_instance+1;
ACPI_SERIAL_END(acpi_wmi);
return (ret);
--- sys/dev/acpi_support/acpi_wmi_if.m.orig 2009-06-21 22:27:31.119098000 +0200
+++ sys/dev/acpi_support/acpi_wmi_if.m 2009-06-29 17:18:38.217246332 +0200
@@ -46,6 +46,7 @@
#
# Check if given GUID exists in WMI
+# Returns number of instances (max_instace+1) or 0 if guid doesn't exist
#
# device_t dev: Device to probe
# const char* guid_string: String form of the GUID
[-- Attachment #3 --]
VHJ ZksFW+NWwuܤql3`3mgG4Jٕ2I}3`G U.C`գ6+b<
I 8؎5X)I gɽss_Zs?K?AۈptZm.PK.u
ˢu]ǃFI-[ ^vm8m^
^u
q5"L3p/ ī,qٜRea4,e$=L,}@<Z^kluk<y8sBvQMaĉnh q_0 pzx)Ѥ@[Fs?^ oa§zhyo 1nx2S^=AK[1Zɼݦ(:]Vqm |=]5 t0n5{akcMb9$!^<o0pt|guFG3vq4: jƙ+\{~<;zٵ=r_X0
{L5 ͒uA6)f)0j]#m1>_ Z/<?t=!]a Cqis?eQ[xPǷNAn\\U
G+](e s8-B^GR^>M VF>MG0lQ6eopH8qx o7CtQkH4rSq<SPJ ~2vI.c{
һxTqiy} hn /&okiQLxOVhn{i']|ӻ|y>m+H 8Jv1^1,@<SVqڄ &$<;,E[ʲmo7#HێQJLGpN)Zj7-Y0 -'u:^O$'!%K7ډ@lT=@*L̓?Iߡ^z8{{ G>R*A(c Z&fK=zA7/t/F)ǤЌg"Zl;>VOovWrh#q7 2ɭNŧw?tlUZw}^0+S0ͮuMhw<b9T;-{]c8rC+Ӳed
jɡfaAjPjj ߄:0,@bK^ d-p
_$U""T4R j ǣ.{k]e MQϰw?:c/{^N#ic8^Xi V}ӫóg0:;@JE -rz>iJ6讇I?4ԺN"A9N'XgQ뤾ۆI=v-ˬv7|׀5bY;\::~KgZ.p^)}5:`"R f &1O\c*yډY H1[>C%]ʫn,
b!Gܵ1ܰ?<`c<'GgB]u)_^,Xo@8Uo 89!#lR)Α9:GRH I#ss'?}?J]k#XT$PG,k+Tpl燎8`U
_M P
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4A48DEB8.1040808>