svn commit: r304880 - head/sys/dev/iwm
Author: avos Date: Sat Aug 27 08:34:20 2016 New Revision: 304880 URL: https://svnweb.freebsd.org/changeset/base/304880 Log: iwm: add 'command accepted' debug notification (copied from wpi(4)). Now it should be easier to find out which command causes firmware panics when few commands are sent in a short period of time. Modified: head/sys/dev/iwm/if_iwm.c Modified: head/sys/dev/iwm/if_iwm.c == --- head/sys/dev/iwm/if_iwm.c Sat Aug 27 02:53:21 2016(r304879) +++ head/sys/dev/iwm/if_iwm.c Sat Aug 27 08:34:20 2016(r304880) @@ -3120,6 +3120,11 @@ iwm_cmd_done(struct iwm_softc *sc, struc return; /* Not a command ack. */ } + /* XXX wide commands? */ + IWM_DPRINTF(sc, IWM_DEBUG_CMD, + "cmd notification type 0x%x qid %d idx %d\n", + pkt->hdr.code, pkt->hdr.qid, pkt->hdr.idx); + data = &ring->data[pkt->hdr.idx]; /* If the command was mapped in an mbuf, free it. */ ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r304886 - in head/sys/modules: cloudabi32 cloudabi64
Author: ed
Date: Sat Aug 27 09:50:11 2016
New Revision: 304886
URL: https://svnweb.freebsd.org/changeset/base/304886
Log:
Properly use MACHINE_CPUARCH for finding cloudabi*_sysvec.c.
The build of the cloudabi32 kernel module currently fails for PC98. In
the case of PC98, we just want to use the code for i386.
Reported by: np
Modified:
head/sys/modules/cloudabi32/Makefile
head/sys/modules/cloudabi64/Makefile
Modified: head/sys/modules/cloudabi32/Makefile
==
--- head/sys/modules/cloudabi32/MakefileSat Aug 27 09:40:29 2016
(r304885)
+++ head/sys/modules/cloudabi32/MakefileSat Aug 27 09:50:11 2016
(r304886)
@@ -3,7 +3,7 @@
SYSDIR?=${.CURDIR}/../..
.PATH: ${SYSDIR}/compat/cloudabi32
-.PATH: ${SYSDIR}/${MACHINE}/cloudabi32
+.PATH: ${SYSDIR}/${MACHINE_CPUARCH}/cloudabi32
KMOD= cloudabi32
SRCS= cloudabi32_fd.c cloudabi32_module.c cloudabi32_poll.c \
Modified: head/sys/modules/cloudabi64/Makefile
==
--- head/sys/modules/cloudabi64/MakefileSat Aug 27 09:40:29 2016
(r304885)
+++ head/sys/modules/cloudabi64/MakefileSat Aug 27 09:50:11 2016
(r304886)
@@ -3,7 +3,7 @@
SYSDIR?=${.CURDIR}/../..
.PATH: ${SYSDIR}/compat/cloudabi64
-.PATH: ${SYSDIR}/${MACHINE}/cloudabi64
+.PATH: ${SYSDIR}/${MACHINE_CPUARCH}/cloudabi64
KMOD= cloudabi64
SRCS= cloudabi64_fd.c cloudabi64_module.c cloudabi64_poll.c \
___
svn-src-head@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r304891 - head/sys/dev/iwm
Author: avos
Date: Sat Aug 27 10:04:48 2016
New Revision: 304891
URL: https://svnweb.freebsd.org/changeset/base/304891
Log:
iwm: fix few comment typos.
Modified:
head/sys/dev/iwm/if_iwmreg.h
Modified: head/sys/dev/iwm/if_iwmreg.h
==
--- head/sys/dev/iwm/if_iwmreg.hSat Aug 27 10:00:36 2016
(r304890)
+++ head/sys/dev/iwm/if_iwmreg.hSat Aug 27 10:04:48 2016
(r304891)
@@ -3219,7 +3219,7 @@ enum iwm_sf_scenario {
#define IWM_SF_SINGLE_UNICAST_AGING_TIMER_DEF 400 /* 0.4 mSec */
#define IWM_SF_AGG_UNICAST_IDLE_TIMER_DEF 160 /* 150 uSec */
#define IWM_SF_AGG_UNICAST_AGING_TIMER_DEF 400 /* 0.4 mSec */
-#define IWM_SF_MCAST_IDLE_TIMER_DEF 160/* 150 mSec */
+#define IWM_SF_MCAST_IDLE_TIMER_DEF 160/* 150 uSec */
#define IWM_SF_MCAST_AGING_TIMER_DEF 400 /* 0.4 mSec */
#define IWM_SF_BA_IDLE_TIMER_DEF 160 /* 150 uSec */
#define IWM_SF_BA_AGING_TIMER_DEF 400 /* 0.4 mSec */
@@ -3244,7 +3244,7 @@ enum iwm_sf_scenario {
/**
* Smart Fifo configuration command.
- * @state: smart fifo state, types listed in iwm_sf_sate.
+ * @state: smart fifo state, types listed in iwm_sf_state.
* @watermark: Minimum allowed available free space in RXF for transient state.
* @long_delay_timeouts: aging and idle timer values for each scenario
* in long delay state.
___
svn-src-head@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r304892 - head/sys/arm64/arm64
Author: andrew
Date: Sat Aug 27 10:30:20 2016
New Revision: 304892
URL: https://svnweb.freebsd.org/changeset/base/304892
Log:
Print both the kernel read and write translation in DDB when asking for
a virtual to physical translation. These may be different, e.g. when a
page is mapped as read-only.
MFC after:1 month
Sponsored by: ABT Systems Ltd
Modified:
head/sys/arm64/arm64/machdep.c
Modified: head/sys/arm64/arm64/machdep.c
==
--- head/sys/arm64/arm64/machdep.c Sat Aug 27 10:04:48 2016
(r304891)
+++ head/sys/arm64/arm64/machdep.c Sat Aug 27 10:30:20 2016
(r304892)
@@ -1089,7 +1089,9 @@ DB_SHOW_COMMAND(vtop, db_show_vtop)
if (have_addr) {
phys = arm64_address_translate_s1e1r(addr);
- db_printf("Physical address reg: 0x%016lx\n", phys);
+ db_printf("Physical address reg (read): 0x%016lx\n", phys);
+ phys = arm64_address_translate_s1e1w(addr);
+ db_printf("Physical address reg (write): 0x%016lx\n", phys);
} else
db_printf("show vtop \n");
}
___
svn-src-head@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r304895 - head/usr.bin/netstat
Author: bde Date: Sat Aug 27 11:06:06 2016 New Revision: 304895 URL: https://svnweb.freebsd.org/changeset/base/304895 Log: Fix build without INET6 and with gcc. A function definition was ifdefed for INET6, but its protototype was not, and gcc detects the error. Modified: head/usr.bin/netstat/route.c Modified: head/usr.bin/netstat/route.c == --- head/usr.bin/netstat/route.cSat Aug 27 10:56:04 2016 (r304894) +++ head/usr.bin/netstat/route.cSat Aug 27 11:06:06 2016 (r304895) @@ -104,7 +104,9 @@ static int ifmap_size; static struct timespec uptime; static const char *netname4(in_addr_t, in_addr_t); +#ifdef INET6 static const char *netname6(struct sockaddr_in6 *, struct sockaddr_in6 *); +#endif static void p_rtable_sysctl(int, int); static void p_rtentry_sysctl(const char *name, struct rt_msghdr *); static int p_sockaddr(const char *name, struct sockaddr *, struct sockaddr *, ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r304907 - head/sys/modules
Author: bz
Date: Sat Aug 27 12:41:15 2016
New Revision: 304907
URL: https://svnweb.freebsd.org/changeset/base/304907
Log:
Do not try to build cloudabi32 for pc98.
Should unbreak tinderbox.
Modified:
head/sys/modules/Makefile
Modified: head/sys/modules/Makefile
==
--- head/sys/modules/Makefile Sat Aug 27 11:51:08 2016(r304906)
+++ head/sys/modules/Makefile Sat Aug 27 12:41:15 2016(r304907)
@@ -766,7 +766,7 @@ _epic= epic
_igb= igb
.endif
-.if ${MACHINE_CPUARCH} == "amd64" || ${MACHINE_CPUARCH} == "i386"
+.if ${MACHINE_CPUARCH} == "amd64" || ${MACHINE} == "i386"
_cloudabi32= cloudabi32
.endif
.if ${MACHINE_CPUARCH} == "aarch64" || ${MACHINE_CPUARCH} == "amd64"
___
svn-src-head@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r304908 - in head: lib/libnv lib/libnv/tests sys/conf sys/contrib/libnv
Author: oshogbo
Date: Sat Aug 27 13:37:30 2016
New Revision: 304908
URL: https://svnweb.freebsd.org/changeset/base/304908
Log:
Add cnv API.
cnv API is a set of functions for managing name/value pairs by cookie.
The cookie can be obtained by nvlist_next(), nvlist_get_parent() or
nvlist_get_pararr() function. This patch also includes unit tests.
Submitted by: Adam Starak
Added:
head/lib/libnv/tests/cnv_tests.cc (contents, props changed)
head/sys/contrib/libnv/cnvlist.c (contents, props changed)
Modified:
head/lib/libnv/Makefile
head/lib/libnv/tests/Makefile
head/sys/conf/files
head/sys/contrib/libnv/nvlist.c
head/sys/contrib/libnv/nvlist_impl.h
Modified: head/lib/libnv/Makefile
==
--- head/lib/libnv/Makefile Sat Aug 27 12:41:15 2016(r304907)
+++ head/lib/libnv/Makefile Sat Aug 27 13:37:30 2016(r304908)
@@ -11,7 +11,8 @@ SHLIB_MAJOR= 0
.PATH: ${.CURDIR}/../../sys/contrib/libnv ${.CURDIR}/../../sys/sys
CFLAGS+=-I${.CURDIR}/../../sys -I${.CURDIR}
-SRCS= dnvlist.c
+SRCS= cnvlist.c
+SRCS+= dnvlist.c
SRCS+= msgio.c
SRCS+= nvlist.c
SRCS+= nvpair.c
Modified: head/lib/libnv/tests/Makefile
==
--- head/lib/libnv/tests/Makefile Sat Aug 27 12:41:15 2016
(r304907)
+++ head/lib/libnv/tests/Makefile Sat Aug 27 13:37:30 2016
(r304908)
@@ -1,6 +1,7 @@
# $FreeBSD$
ATF_TESTS_CXX= \
+ cnv_tests\
dnv_tests \
nv_array_tests \
nv_tests \
Added: head/lib/libnv/tests/cnv_tests.cc
==
--- /dev/null 00:00:00 1970 (empty, because file is newly added)
+++ head/lib/libnv/tests/cnv_tests.cc Sat Aug 27 13:37:30 2016
(r304908)
@@ -0,0 +1,1509 @@
+/*-
+ * Copyright (c) 2016 Adam Starak
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ *notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *notice, this list of conditions and the following disclaimer in the
+ *documentation and/or other materials provided with the distribution.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE AUTHORS AND CONTRIBUTORS ``AS IS'' AND
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHORS OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ *
+ * $FreeBSD$
+ */
+
+#include
+__FBSDID("$FreeBSD$");
+
+#include
+#include
+#include
+
+#include
+#include
+#include
+
+#definefd_is_valid(fd) (fcntl((fd), F_GETFL) != -1 || errno != EBADF)
+
+/* ATF cnvlist_get tests. */
+
+ATF_TEST_CASE_WITHOUT_HEAD(cnvlist_get_bool);
+ATF_TEST_CASE_BODY(cnvlist_get_bool)
+{
+ nvlist_t *nvl;
+ const char *key;
+ bool value;
+ void *cookie;
+ int type;
+
+ nvl = nvlist_create(0);
+ ATF_REQUIRE(nvl != NULL);
+ ATF_REQUIRE_EQ(nvlist_error(nvl), 0);
+ ATF_REQUIRE(nvlist_empty(nvl));
+
+ cookie = NULL;
+ key = "name";
+ value = true;
+
+ nvlist_add_bool(nvl, key, value);
+ ATF_REQUIRE_EQ(strcmp(key, nvlist_next(nvl, &type, &cookie)), 0);
+ ATF_REQUIRE_EQ(nvlist_error(nvl), 0);
+ ATF_REQUIRE_EQ(type, NV_TYPE_BOOL);
+ ATF_REQUIRE(!nvlist_empty(nvl));
+ ATF_REQUIRE(nvlist_exists(nvl, key));
+ ATF_REQUIRE(nvlist_exists_bool(nvl, key));
+
+ ATF_REQUIRE_EQ(cnvlist_get_bool(cookie), value);
+
+ ATF_REQUIRE_EQ(nvlist_next(nvl, &type, &cookie),
+ static_cast(NULL));
+
+ nvlist_destroy(nvl);
+}
+
+ATF_TEST_CASE_WITHOUT_HEAD(cnvlist_get_number);
+ATF_TEST_CASE_BODY(cnvlist_get_number)
+{
+ nvlist_t *nvl;
+ const char *key;
+ uint64_t value;
+ void *cookie;
+ int type;
+
+ nvl = nvlist_create(0);
+ ATF_REQUIRE(nvl != NULL);
+ ATF_REQUIRE_EQ(nvlist_error(nvl), 0);
+ ATF_REQUIRE(nvlist_empty(nvl));
+
+ cookie = NULL;
+ key = "name";
+ value = 420;
+
+ nvlist_add_number(nvl, key, value);
+ ATF_REQUIRE_EQ(strcmp(key, nvlist_n
svn commit: r304909 - in head: lib/libnv/tests sys/contrib/libnv
Author: oshogbo
Date: Sat Aug 27 13:40:27 2016
New Revision: 304909
URL: https://svnweb.freebsd.org/changeset/base/304909
Log:
Fix style issue in the cnv API.
Remove unused arguments in a macro.
Remove unused typedef.
Modified:
head/lib/libnv/tests/Makefile
head/lib/libnv/tests/cnv_tests.cc
head/sys/contrib/libnv/cnvlist.c
Modified: head/lib/libnv/tests/Makefile
==
--- head/lib/libnv/tests/Makefile Sat Aug 27 13:37:30 2016
(r304908)
+++ head/lib/libnv/tests/Makefile Sat Aug 27 13:40:27 2016
(r304909)
@@ -1,7 +1,7 @@
# $FreeBSD$
ATF_TESTS_CXX= \
- cnv_tests\
+ cnv_tests \
dnv_tests \
nv_array_tests \
nv_tests \
Modified: head/lib/libnv/tests/cnv_tests.cc
==
--- head/lib/libnv/tests/cnv_tests.cc Sat Aug 27 13:37:30 2016
(r304908)
+++ head/lib/libnv/tests/cnv_tests.cc Sat Aug 27 13:40:27 2016
(r304909)
@@ -188,8 +188,9 @@ ATF_TEST_CASE_BODY(cnvlist_get_nvlist)
ATF_REQUIRE(nvlist_exists(nvl, key));
ATF_REQUIRE(nvlist_exists_nvlist(nvl, key));
- /* Assuming nvlist_get_nvlist() is correct check if cnvlist returns the
-* same pointer.
+ /*
+* Assuming nvlist_get_nvlist() is correct check if cnvlist returns
+* the same pointer.
*/
result = cnvlist_get_nvlist(cookie);
ATF_REQUIRE_EQ(result, nvlist_get_nvlist(nvl, key));
@@ -499,7 +500,6 @@ ATF_TEST_CASE_BODY(cnvlist_get_nvlist_ar
ATF_TEST_CASE_WITHOUT_HEAD(cnvlist_get_descriptor_array);
ATF_TEST_CASE_BODY(cnvlist_get_descriptor_array)
{
-
nvlist_t *nvl;
size_t count, i, nitems;
const int *out_array;
@@ -730,7 +730,6 @@ ATF_TEST_CASE_BODY(cnvlist_take_nvlist)
ATF_REQUIRE(result == value);
/* Validate data inside nvlist. */
-
cookie = NULL;
ATF_REQUIRE_EQ(strcmp(subkey, nvlist_next(result, &type, &cookie)), 0);
ATF_REQUIRE_EQ(nvlist_error(value), 0);
Modified: head/sys/contrib/libnv/cnvlist.c
==
--- head/sys/contrib/libnv/cnvlist.cSat Aug 27 13:37:30 2016
(r304908)
+++ head/sys/contrib/libnv/cnvlist.cSat Aug 27 13:40:27 2016
(r304909)
@@ -53,14 +53,15 @@ __FBSDID("$FreeBSD$");
#include "nvlist_impl.h"
#include "nvpair_impl.h"
-#define CNVLIST_GET(ftype, type, nvtype) \
+#defineCNVLIST_GET(ftype, type, NVTYPE)
\
ftype \
cnvlist_get_##type(void *cookiep) \
{ \
\
-if (nvpair_type(cookiep) != NV_TYPE_##nvtype) \
- nvlist_report_missing(NV_TYPE_##nvtype, \
+ if (nvpair_type(cookiep) != NV_TYPE_##NVTYPE) { \
+ nvlist_report_missing(NV_TYPE_##NVTYPE, \
nvpair_name(cookiep)); \
+ } \
return (nvpair_get_##type(cookiep)); \
}
@@ -72,17 +73,18 @@ CNVLIST_GET(const nvlist_t *, nvlist, NV
CNVLIST_GET(int, descriptor, DESCRIPTOR)
#endif
-#undef CNVLIST_GET
+#undef CNVLIST_GET
-#define CNVLIST_GET_ARRAY(ftype, type, nvtype) \
+#defineCNVLIST_GET_ARRAY(ftype, type, NVTYPE)
\
ftype \
cnvlist_get_##type(void *cookiep, size_t *nitemsp) \
{ \
\
-if (nvpair_type(cookiep) != NV_TYPE_##nvtype) \
- nvlist_report_missing(NV_TYPE_##nvtype, \
+ if (nvpair_type(cookiep) != NV_TYPE_##NVTYPE) { \
+ nvlist_report_missing(NV_TYPE_##NVTYPE, \
nvpair_name(cookiep)); \
-return (nvpair_get_##type(cookiep, nitemsp)); \
+ } \
+ return (nvpair_get_##type(cookiep, nitemsp)); \
}
CNVLIST_GET_ARRAY(const bool *, bool_array, BOOL_ARRAY)
@@ -93,26 +95,27 @@ CNVLIST_GET_ARRAY(const nvlist_t * const
CNVLIST_GET_ARRAY(const int *, descriptor_array, DESCRIPTOR_ARRAY)
#endif
-#undef CNVLIST_GET_ARRAY
+#undef CNVLIST_GET_ARRAY
const void *
cnvlist_get_binary(void *cookiep, size_t *s
svn commit: r304910 - head/share/man/man9
Author: oshogbo Date: Sat Aug 27 13:47:52 2016 New Revision: 304910 URL: https://svnweb.freebsd.org/changeset/base/304910 Log: Introduce cnv man page. Submitted by: Adam Starak Reviewed by: cem@, wblock@ Differential Revision:https://reviews.freebsd.org/D7249 Added: head/share/man/man9/cnv.9 (contents, props changed) Modified: head/share/man/man9/Makefile Modified: head/share/man/man9/Makefile == --- head/share/man/man9/MakefileSat Aug 27 13:40:27 2016 (r304909) +++ head/share/man/man9/MakefileSat Aug 27 13:47:52 2016 (r304910) @@ -57,6 +57,7 @@ MAN= accept_filter.9 \ byteorder.9 \ casuword.9 \ cd.9 \ + cnv.9 \ condvar.9 \ config_intrhook.9 \ contigmalloc.9 \ @@ -617,6 +618,41 @@ MLINKS+=byteorder.9 be16dec.9 \ byteorder.9 le64dec.9 \ byteorder.9 le64enc.9 \ byteorder.9 le64toh.9 +MLINKS+=cnv.9 cnvlist.9 \ + cnv.9 cnvlist_free_binary.9 \ + cnv.9 cnvlist_free_bool.9 \ + cnv.9 cnvlist_free_bool_array.9 \ + cnv.9 cnvlist_free_descriptor.9 \ + cnv.9 cnvlist_free_descriptor_array.9 \ + cnv.9 cnvlist_free_null.9 \ + cnv.9 cnvlist_free_number.9 \ + cnv.9 cnvlist_free_number_array.9 \ + cnv.9 cnvlist_free_nvlist.9 \ + cnv.9 cnvlist_free_nvlist_array.9 \ + cnv.9 cnvlist_free_string.9 \ + cnv.9 cnvlist_free_string_array.9 \ + cnv.9 cnvlist_get_binary.9 \ + cnv.9 cnvlist_get_bool.9 \ + cnv.9 cnvlist_get_bool_array.9 \ + cnv.9 cnvlist_get_descriptor.9 \ + cnv.9 cnvlist_get_descriptor_array.9 \ + cnv.9 cnvlist_get_number.9 \ + cnv.9 cnvlist_get_number_array.9 \ + cnv.9 cnvlist_get_nvlist.9 \ + cnv.9 cnvlist_get_nvlist_array.9 \ + cnv.9 cnvlist_get_string.9 \ + cnv.9 cnvlist_get_string_array.9 \ + cnv.9 cnvlist_take_binary.9 \ + cnv.9 cnvlist_take_bool.9 \ + cnv.9 cnvlist_take_bool_array.9 \ + cnv.9 cnvlist_take_descriptor.9 \ + cnv.9 cnvlist_take_descriptor_array.9 \ + cnv.9 cnvlist_take_number.9 \ + cnv.9 cnvlist_take_number_array.9 \ + cnv.9 cnvlist_take_nvlist.9 \ + cnv.9 cnvlist_take_nvlist_array.9 \ + cnv.9 cnvlist_take_string.9 \ + cnv.9 cnvlist_take_string_array.9 MLINKS+=condvar.9 cv_broadcast.9 \ condvar.9 cv_broadcastpri.9 \ condvar.9 cv_destroy.9 \ Added: head/share/man/man9/cnv.9 == --- /dev/null 00:00:00 1970 (empty, because file is newly added) +++ head/share/man/man9/cnv.9 Sat Aug 27 13:47:52 2016(r304910) @@ -0,0 +1,199 @@ +.\" +.\" Copyright (c) 2016 Adam Starak +.\" All rights reserved. +.\" +.\" Redistribution and use in source and binary forms, with or without +.\" modification, are permitted provided that the following conditions +.\" are met: +.\" 1. Redistributions of source code must retain the above copyright +.\"notice, this list of conditions and the following disclaimer. +.\" 2. Redistributions in binary form must reproduce the above copyright +.\"notice, this list of conditions and the following disclaimer in the +.\"documentation and/or other materials provided with the distribution. +.\" +.\" THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND +.\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +.\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE +.\" ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE +.\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +.\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS +.\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) +.\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT +.\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY +.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF +.\" SUCH DAMAGE. +.\" +.\" $FreeBSD$ +.\" +.Dd July 26, 2016 +.Dt CNV 9 +.Os +.Sh NAME +.Nm cnvlist_get, +.Nm cnvlist_take, +.Nm cnvlist_free, +.Nd "API for managing name/value pairs by cookie." +.Sh LIBRARY +.Lb libnv +.Sh SYNOPSIS +.In sys/cnv.h +.Ft bool +.Fn cnvlist_get_bool "void *cookiep" +.Ft uint64_t +.Fn cnvlist_get_number "void *cookiep" +.Ft "const char *" +.Fn cnvlist_get_string "void *cookiep" +.Ft "const nvlist_t *" +.Fn cnvlist_get_nvlist "void *cookiep" +.Ft "const void *" +.Fn cnvlist_get_binary "void *cookiep" "size_t *sizep" +.Ft "const bool *" +.Fn cnvlist_get_bool_array "void *cookiep" "size_t *nitemsp" +.Ft "const uint64_t *" +.Fn cnvlist_get_number_array "void *cookiep" "size_t *nitemsp" +.Ft "const char * const *" +.Fn cnvlist_get_string_array "void *cookiep" "size_t *nitem
svn commit: r304911 - head/lib/libc/net
Author: ache
Date: Sat Aug 27 14:43:13 2016
New Revision: 304911
URL: https://svnweb.freebsd.org/changeset/base/304911
Log:
The formal behavior of qsort is unstable with regard to objects that
are equal. Unfortunately, RFC 3484 requires that otherwise equal objects
remain in the order supplied by the DNS server. The present code attempts
to deal with this by returning -1 for objects that are equal (i.e.,
returns that the first parameter is less then the second parameter).
Unfortunately, the qsort API does not state that the first parameter
passed in is in any particular position in the list.
PR: 212122
Submitted by: herbie.robin...@stratus.com
MFC after: 3 days
Modified:
head/lib/libc/net/getaddrinfo.c
Modified: head/lib/libc/net/getaddrinfo.c
==
--- head/lib/libc/net/getaddrinfo.c Sat Aug 27 13:47:52 2016
(r304910)
+++ head/lib/libc/net/getaddrinfo.c Sat Aug 27 14:43:13 2016
(r304911)
@@ -224,6 +224,7 @@ struct ai_order {
struct policyqueue *aio_dstpolicy;
struct addrinfo *aio_ai;
int aio_matchlen;
+ int aio_initial_sequence;
};
static const ns_src default_dns_files[] = {
@@ -708,6 +709,7 @@ reorder(struct addrinfo *sentinel)
aio[i].aio_dstpolicy = match_addrselectpolicy(ai->ai_addr,
&policyhead);
set_source(&aio[i], &policyhead);
+ aio[i].aio_initial_sequence = i;
}
/* perform sorting. */
@@ -1066,6 +1068,23 @@ comp_dst(const void *arg1, const void *a
}
/* Rule 10: Otherwise, leave the order unchanged. */
+
+ /*
+* Note that qsort is unstable; so, we can't return zero and
+* expect the order to be unchanged.
+* That also means we can't depend on the current position of
+* dst2 being after dst1. We must enforce the initial order
+* with an explicit compare on the original position.
+* The qsort specification requires that "When the same objects
+* (consisting of width bytes, irrespective of their current
+* positions in the array) are passed more than once to the
+* comparison function, the results shall be consistent with one
+* another."
+* In other words, If A < B, then we must also return B > A.
+*/
+ if (dst2->aio_initial_sequence < dst1->aio_initial_sequence)
+ return(1);
+
return(-1);
}
___
svn-src-head@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r304912 - head/sys/sys
Author: oshogbo Date: Sat Aug 27 15:22:55 2016 New Revision: 304912 URL: https://svnweb.freebsd.org/changeset/base/304912 Log: Add missed header file for cnv.h . Submitted by: Adam Starak Reported by: ache@ Added: head/sys/sys/cnv.h (contents, props changed) Added: head/sys/sys/cnv.h == --- /dev/null 00:00:00 1970 (empty, because file is newly added) +++ head/sys/sys/cnv.h Sat Aug 27 15:22:55 2016(r304912) @@ -0,0 +1,113 @@ +/*- + * Copyright (c) 2016 Adam Starak + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * 1. Redistributions of source code must retain the above copyright + *notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + *notice, this list of conditions and the following disclaimer in the + *documentation and/or other materials provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED BY THE AUTHORS AND CONTRIBUTORS ``AS IS'' AND + * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHORS OR CONTRIBUTORS BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL + * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS + * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT + * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY + * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF + * SUCH DAMAGE. + * + * $FreeBSD$ + */ + +#ifndef_CNV_H_ +#define_CNV_H_ + +#include + +#ifndef _KERNEL +#include +#include +#include +#include +#endif + +#ifndef_NVLIST_T_DECLARED +#define_NVLIST_T_DECLARED +struct nvlist; + +typedef struct nvlist nvlist_t; +#endif + +__BEGIN_DECLS + +/* + * The cnvlist_get functions returns value associated with the given cookie. + * If it returns a pointer, the pointer represents internal buffer and should + * not be freed by the caller. + */ + +boolcnvlist_get_bool(void *cookiep); +uint64_tcnvlist_get_number(void *cookiep); +const char *cnvlist_get_string(void *cookiep); +const nvlist_t *cnvlist_get_nvlist(void *cookiep); +const void *cnvlist_get_binary(void *cookiep, size_t *sizep); +const bool *cnvlist_get_bool_array(void *cookiep, size_t *nitemsp); +const uint64_t *cnvlist_get_number_array(void *cookiep, size_t *nitemsp); +const char * const *cnvlist_get_string_array(void *cookiep, size_t *nitemsp); +const nvlist_t * const *cnvlist_get_nvlist_array(void *cookiep, size_t *nitemsp); +#ifndef _KERNEL +int cnvlist_get_descriptor(void *cookiep); +const int *cnvlist_get_descriptor_array(void *cookiep, size_t *nitemsp); +#endif + + +/* + * The cnvlist_take functions returns value associated with the given cookie and + * remove the given entry from the nvlist. + * The caller is responsible for freeing received data. + */ + +bool cnvlist_take_bool(nvlist_t *nvl, void *cookiep); +uint64_t cnvlist_take_number(nvlist_t *nvl, void *cookiep); +char*cnvlist_take_string(nvlist_t *nvl, void *cookiep); +nvlist_t*cnvlist_take_nvlist(nvlist_t *nvl, void *cookiep); +void*cnvlist_take_binary(nvlist_t *nvl, void *cookiep, size_t *sizep); +bool*cnvlist_take_bool_array(nvlist_t *nvl, void *cookiep, size_t *nitemsp); +uint64_t*cnvlist_take_number_array(nvlist_t *nvl, void *cookiep, size_t *nitemsp); +char **cnvlist_take_string_array(nvlist_t *nvl, void *cookiep, size_t *nitemsp); +nvlist_t **cnvlist_take_nvlist_array(nvlist_t *nvl, void *cookiep, size_t *nitemsp); +#ifndef _KERNEL +int cnvlist_take_descriptor(nvlist_t *nvl, void *cookiep); +int *cnvlist_take_descriptor_array(nvlist_t *nvl, void *cookiep, size_t *nitemsp); +#endif + +/* + * The cnvlist_free functions removes the given name/value pair from the nvlist based on cookie + * and frees memory associated with it. + */ + +void cnvlist_free_bool(nvlist_t *nvl, void *cookiep); +void cnvlist_free_number(nvlist_t *nvl, void *cookiep); +void cnvlist_free_string(nvlist_t *nvl, void *cookiep); +void cnvlist_free_nvlist(nvlist_t *nvl, void *cookiep); +void cnvlist_free_binary(nvlist_t *nvl, void *cookiep); +void cnvlist_free_descriptor(nvlist_t *nvl, void *cookiep); +void cnvlist_free_bool_array(nvlist_
Re: svn commit: r304857 - head/sys/netinet/tcp_stacks
On 08/27/16 at 01:47P, Lawrence Stewart wrote: > Pointy hat to: lstewart@ > > Apologies all for the breakage, thanks Hiren for fixing and apologies > also for missing your email. For some reason your reply to my commit did > not make it to my inbox and was filtered straight to my mailing list > folder which is annoying - I must have introduced a bug in my > .procmailrc at some point. No worries. :-) > > I don't understand why my buildkernel prior to commit succeeded though. > Is fastpath not build by default? Yeah, it's not since r301814. Cheers, Hiren pgpchIWM8G0ds.pgp Description: PGP signature
Re: svn commit: r304911 - head/lib/libc/net
On Sat, Aug 27, 2016 at 02:43:13PM +, Andrey A. Chernov wrote: > Author: ache > Date: Sat Aug 27 14:43:13 2016 > New Revision: 304911 > URL: https://svnweb.freebsd.org/changeset/base/304911 > > Log: > The formal behavior of qsort is unstable with regard to objects that > are equal. Unfortunately, RFC 3484 requires that otherwise equal objects > remain in the order supplied by the DNS server. The present code attempts > to deal with this by returning -1 for objects that are equal (i.e., > returns that the first parameter is less then the second parameter). > Unfortunately, the qsort API does not state that the first parameter > passed in is in any particular position in the list. Why not switch to mergesort which is stable by design? -- Brooks signature.asc Description: PGP signature
svn commit: r304915 - head/share/man/man9
Author: oshogbo Date: Sat Aug 27 18:08:25 2016 New Revision: 304915 URL: https://svnweb.freebsd.org/changeset/base/304915 Log: Bump date in the man page. Modified: head/share/man/man9/cnv.9 Modified: head/share/man/man9/cnv.9 == --- head/share/man/man9/cnv.9 Sat Aug 27 16:01:10 2016(r304914) +++ head/share/man/man9/cnv.9 Sat Aug 27 18:08:25 2016(r304915) @@ -25,7 +25,7 @@ .\" .\" $FreeBSD$ .\" -.Dd July 26, 2016 +.Dd August 27, 2016 .Dt CNV 9 .Os .Sh NAME ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r304916 - head/sys/kern
Author: kib
Date: Sat Aug 27 18:12:42 2016
New Revision: 304916
URL: https://svnweb.freebsd.org/changeset/base/304916
Log:
Consistently delimit each vnode description block with two blank
lines.
Sponsored by: The FreeBSD Foundation
MFC after:3 days
Modified:
head/sys/kern/vnode_if.src
Modified: head/sys/kern/vnode_if.src
==
--- head/sys/kern/vnode_if.src Sat Aug 27 18:08:25 2016(r304915)
+++ head/sys/kern/vnode_if.src Sat Aug 27 18:12:42 2016(r304916)
@@ -59,6 +59,7 @@ vop_islocked {
IN struct vnode *vp;
};
+
%% lookup dvp L L L
%% lookup vpp - L -
@@ -72,6 +73,7 @@ vop_lookup {
IN struct componentname *cnp;
};
+
%% cachedlookupdvp L L L
%% cachedlookupvpp - L -
@@ -83,6 +85,7 @@ vop_cachedlookup {
IN struct componentname *cnp;
};
+
%% create dvp E E E
%% create vpp - L -
%! create postvop_create_post
@@ -177,12 +180,14 @@ vop_setattr {
IN struct ucred *cred;
};
+
%% markatime vp L L L
vop_markatime {
IN struct vnode *vp;
};
+
%% readvp L L L
%! readpostvop_read_post
@@ -642,6 +647,7 @@ vop_allocate {
INOUT off_t *len;
};
+
%% advise vp U U U
vop_advise {
@@ -651,6 +657,7 @@ vop_advise {
IN int advice;
};
+
%% unp_bindvp E E E
vop_unp_bind {
@@ -658,6 +665,7 @@ vop_unp_bind {
IN struct socket *socket;
};
+
%% unp_connect vp L L L
vop_unp_connect {
@@ -665,30 +673,35 @@ vop_unp_connect {
OUT struct socket **socket;
};
+
%% unp_detach vp = = =
vop_unp_detach {
IN struct vnode *vp;
};
+
%% is_text vp L L L
vop_is_text {
IN struct vnode *vp;
};
+
%% set_textvp E E E
vop_set_text {
IN struct vnode *vp;
};
+
%% vop_unset_text vp E E E
vop_unset_text {
IN struct vnode *vp;
};
+
%% get_writecount vp L L L
vop_get_writecount {
@@ -696,6 +709,7 @@ vop_get_writecount {
OUT int *writecount;
};
+
%% add_writecount vp E E E
vop_add_writecount {
@@ -703,6 +717,7 @@ vop_add_writecount {
IN int inc;
};
+
%% fdatasync vp L L L
vop_fdatasync {
___
svn-src-head@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
Re: svn commit: r304911 - head/lib/libc/net
On 27.08.2016 20:37, Brooks Davis wrote: > On Sat, Aug 27, 2016 at 02:43:13PM +, Andrey A. Chernov wrote: >> Author: ache >> Date: Sat Aug 27 14:43:13 2016 >> New Revision: 304911 >> URL: https://svnweb.freebsd.org/changeset/base/304911 >> >> Log: >> The formal behavior of qsort is unstable with regard to objects that >> are equal. Unfortunately, RFC 3484 requires that otherwise equal objects >> remain in the order supplied by the DNS server. The present code attempts >> to deal with this by returning -1 for objects that are equal (i.e., >> returns that the first parameter is less then the second parameter). >> Unfortunately, the qsort API does not state that the first parameter >> passed in is in any particular position in the list. > > Why not switch to mergesort which is stable by design? megresort() is 1) slower 2) requires additional memory allocated. signature.asc Description: OpenPGP digital signature
svn commit: r304918 - head/sys/cam/ata
Author: mav
Date: Sat Aug 27 19:51:37 2016
New Revision: 304918
URL: https://svnweb.freebsd.org/changeset/base/304918
Log:
Decode some new ATA commands found in ACS-3.
MFC after:1 week
Modified:
head/sys/cam/ata/ata_all.c
Modified: head/sys/cam/ata/ata_all.c
==
--- head/sys/cam/ata/ata_all.c Sat Aug 27 18:51:04 2016(r304917)
+++ head/sys/cam/ata/ata_all.c Sat Aug 27 19:51:37 2016(r304918)
@@ -88,6 +88,7 @@ ata_op_string(struct ata_cmd *cmd)
}
return "DSM";
case 0x08: return ("DEVICE_RESET");
+ case 0x0b: return ("REQUEST_SENSE_DATA_EXT");
case 0x20: return ("READ");
case 0x24: return ("READ48");
case 0x25: return ("READ_DMA48");
@@ -120,6 +121,12 @@ ata_op_string(struct ata_cmd *cmd)
case 0x47: return ("READ_LOG_DMA_EXT");
case 0x4a: return ("ZAC_MANAGEMENT_IN");
case 0x51: return ("CONFIGURE_STREAM");
+ case 0x57: return ("WRITE_LOG_DMA_EXT");
+ case 0x5b: return ("TRUSTED_NON_DATA");
+ case 0x5c: return ("TRUSTED_RECEIVE");
+ case 0x5d: return ("TRUSTED_RECEIVE_DMA");
+ case 0x5e: return ("TRUSTED_SEND");
+ case 0x5f: return ("TRUSTED_SEND_DMA");
case 0x60: return ("READ_FPDMA_QUEUED");
case 0x61: return ("WRITE_FPDMA_QUEUED");
case 0x63:
@@ -160,9 +167,12 @@ ata_op_string(struct ata_cmd *cmd)
}
return ("SEP_ATTN");
case 0x70: return ("SEEK");
+ case 0x77: return ("SET_DATE_TIME_EXT");
+ case 0x78: return ("ACCESSIBLE_MAX_ADDRESS_CONFIGURATION");
case 0x87: return ("CFA_TRANSLATE_SECTOR");
case 0x90: return ("EXECUTE_DEVICE_DIAGNOSTIC");
case 0x92: return ("DOWNLOAD_MICROCODE");
+ case 0x93: return ("DOWNLOAD_MICROCODE_DMA");
case 0x9a: return ("ZAC_MANAGEMENT_OUT");
case 0xa0: return ("PACKET");
case 0xa1: return ("ATAPI_IDENTIFY");
@@ -180,6 +190,7 @@ ata_op_string(struct ata_cmd *cmd)
}
return ("SMART");
case 0xb1: return ("DEVICE CONFIGURATION");
+ case 0xb4: return ("SANITIZE_DEVICE");
case 0xc0: return ("CFA_ERASE");
case 0xc4: return ("READ_MUL");
case 0xc5: return ("WRITE_MUL");
___
svn-src-head@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r304920 - head/contrib/ncurses/ncurses/tinfo
Author: dim
Date: Sat Aug 27 20:33:19 2016
New Revision: 304920
URL: https://svnweb.freebsd.org/changeset/base/304920
Log:
In ncurses baudrate definitions, avoid warnings about implicit
conversions from int to short changing the values. This applies to
B38400 and higher, since their values do not fit into a short.
However, since the wrapped values are still unique, and they only serve
as keys, there is no problem in adding a cast to silence the warnings.
This also avoids changing the ABI, which would happen if we changed
NCURSES_OSPEED to int.
Discussed with: Thomas Dickey
MFC after:1 week
Modified:
head/contrib/ncurses/ncurses/tinfo/lib_baudrate.c
Modified: head/contrib/ncurses/ncurses/tinfo/lib_baudrate.c
==
--- head/contrib/ncurses/ncurses/tinfo/lib_baudrate.c Sat Aug 27 20:08:34
2016(r304919)
+++ head/contrib/ncurses/ncurses/tinfo/lib_baudrate.c Sat Aug 27 20:33:19
2016(r304920)
@@ -94,7 +94,7 @@ struct speed {
int sp;/* the actual speed */
};
-#define DATA(number) { B##number, number }
+#define DATA(number) { (NCURSES_OSPEED)B##number, number }
static struct speed const speeds[] =
{
___
svn-src-head@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r304927 - head/sys/kern
Author: mjg
Date: Sat Aug 27 22:43:41 2016
New Revision: 304927
URL: https://svnweb.freebsd.org/changeset/base/304927
Log:
vfs: provide a common exit point in namei for error cases
This shortens the function, adds the SDT_PROBE use for error cases and
consistenly unrefs rootdir last.
Reviewed by: kib
MFC after:2 weeks
Modified:
head/sys/kern/vfs_lookup.c
Modified: head/sys/kern/vfs_lookup.c
==
--- head/sys/kern/vfs_lookup.c Sat Aug 27 22:23:10 2016(r304926)
+++ head/sys/kern/vfs_lookup.c Sat Aug 27 22:43:41 2016(r304927)
@@ -295,21 +295,15 @@ namei(struct nameidata *ndp)
if (error != 0) {
if (dp != NULL)
vrele(dp);
- vrele(ndp->ni_rootdir);
- namei_cleanup_cnp(cnp);
- return (error);
+ goto out;
}
SDT_PROBE3(vfs, namei, lookup, entry, dp, cnp->cn_pnbuf,
cnp->cn_flags);
for (;;) {
ndp->ni_startdir = dp;
error = lookup(ndp);
- if (error != 0) {
- vrele(ndp->ni_rootdir);
- namei_cleanup_cnp(cnp);
- SDT_PROBE2(vfs, namei, lookup, return, error, NULL);
- return (error);
- }
+ if (error != 0)
+ goto out;
/*
* If not a symbolic link, we're done.
*/
@@ -383,18 +377,16 @@ namei(struct nameidata *ndp)
if (*(cnp->cn_nameptr) == '/') {
vrele(dp);
error = namei_handle_root(ndp, &dp);
- if (error != 0) {
- vrele(ndp->ni_rootdir);
- namei_cleanup_cnp(cnp);
- return (error);
- }
+ if (error != 0)
+ goto out;
}
}
- vrele(ndp->ni_rootdir);
- namei_cleanup_cnp(cnp);
vput(ndp->ni_vp);
ndp->ni_vp = NULL;
vrele(ndp->ni_dvp);
+out:
+ vrele(ndp->ni_rootdir);
+ namei_cleanup_cnp(cnp);
SDT_PROBE2(vfs, namei, lookup, return, error, NULL);
return (error);
}
___
svn-src-head@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r304928 - in head/lib/libc: amd64/sys i386/sys sys
Author: kib
Date: Sat Aug 27 23:03:23 2016
New Revision: 304928
URL: https://svnweb.freebsd.org/changeset/base/304928
Log:
Do not obliterate errno value in the main thread during ptrace(2) call on x86.
Since ptrace(2) syscall can return -1 for non-error situations, libc
wrappers set errno to 0 before performing the syscall, as the service
to the caller. On both i386 and amd64, the errno symbol was directly
referenced, which only works correctly in single-threaded process.
Change assembler wrappers for ptrace(2) to get current thread errno
location by calling __error(). Allow __error interposing, as
currently allowed in cerror().
Sponsored by: The FreeBSD Foundation
MFC after:1 week
Modified:
head/lib/libc/amd64/sys/ptrace.S
head/lib/libc/i386/sys/ptrace.S
head/lib/libc/sys/ptrace.2
Modified: head/lib/libc/amd64/sys/ptrace.S
==
--- head/lib/libc/amd64/sys/ptrace.SSat Aug 27 22:43:41 2016
(r304927)
+++ head/lib/libc/amd64/sys/ptrace.SSat Aug 27 23:03:23 2016
(r304928)
@@ -38,14 +38,16 @@ __FBSDID("$FreeBSD$");
#include "SYS.h"
+ .globl CNAME(__error)
+ .type CNAME(__error),@function
+
ENTRY(ptrace)
- xorl%eax,%eax
#ifdef PIC
- movqPIC_GOT(CNAME(errno)),%r8
- movl%eax,(%r8)
+ callq PIC_PLT(CNAME(__error))
#else
- movl%eax,CNAME(errno)(%rip)
+ callq CNAME(__error)
#endif
+ movl$0,(%rax)
mov $SYS_ptrace,%eax
KERNCALL
jb HIDENAME(cerror)
Modified: head/lib/libc/i386/sys/ptrace.S
==
--- head/lib/libc/i386/sys/ptrace.S Sat Aug 27 22:43:41 2016
(r304927)
+++ head/lib/libc/i386/sys/ptrace.S Sat Aug 27 23:03:23 2016
(r304928)
@@ -38,16 +38,18 @@ __FBSDID("$FreeBSD$");
#include "SYS.h"
+ .globl CNAME(__error)
+ .type CNAME(__error),@function
+
ENTRY(ptrace)
- xorl%eax,%eax
#ifdef PIC
-PIC_PROLOGUE
-movlPIC_GOT(CNAME(errno)),%edx
-movl%eax,(%edx)
-PIC_EPILOGUE
+ PIC_PROLOGUE
+ callPIC_PLT(CNAME(__error))
+ PIC_EPILOGUE
#else
-movl%eax,CNAME(errno)
+ callCNAME(__error)
#endif
+ movl$0,(%eax)
mov $SYS_ptrace,%eax
KERNCALL
jb HIDENAME(cerror)
Modified: head/lib/libc/sys/ptrace.2
==
--- head/lib/libc/sys/ptrace.2 Sat Aug 27 22:43:41 2016(r304927)
+++ head/lib/libc/sys/ptrace.2 Sat Aug 27 23:03:23 2016(r304928)
@@ -2,7 +2,7 @@
.\"$NetBSD: ptrace.2,v 1.2 1995/02/27 12:35:37 cgd Exp $
.\"
.\" This file is in the public domain.
-.Dd July 28, 2016
+.Dd August 28, 2016
.Dt PTRACE 2
.Os
.Sh NAME
@@ -906,7 +906,13 @@ to return
\-1
as a non-error value; to disambiguate,
.Va errno
-can be set to 0 before the call and checked afterwards.
+is set to 0 in the libc wrapper for the
+.Fn ptrace
+system call and
+.Fn ptrace
+callers can reliably check
+.Va errno
+for non-zero value afterwards.
.Sh ERRORS
The
.Fn ptrace
___
svn-src-head@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r304931 - head/lib/libc/amd64/sys
Author: kib
Date: Sun Aug 28 00:26:06 2016
New Revision: 304931
URL: https://svnweb.freebsd.org/changeset/base/304931
Log:
Follow ABI when calling __error from the ptrace(2) wrapper.
In particular, preserve syscall arguments on stack, since callee is
not required to preserve arg-passing registers. Align stack.
Sponsored by: The FreeBSD Foundation
MFC after:1 week
Modified:
head/lib/libc/amd64/sys/ptrace.S
Modified: head/lib/libc/amd64/sys/ptrace.S
==
--- head/lib/libc/amd64/sys/ptrace.SSun Aug 28 00:21:24 2016
(r304930)
+++ head/lib/libc/amd64/sys/ptrace.SSun Aug 28 00:26:06 2016
(r304931)
@@ -42,11 +42,21 @@ __FBSDID("$FreeBSD$");
.type CNAME(__error),@function
ENTRY(ptrace)
+ pushq %rdi/* align stack */
+ pushq %rdi
+ pushq %rsi
+ pushq %rdx
+ pushq %rcx
#ifdef PIC
callq PIC_PLT(CNAME(__error))
#else
callq CNAME(__error)
#endif
+ popq%rcx
+ popq%rdx
+ popq%rsi
+ popq%rdi
+ popq%rdi
movl$0,(%rax)
mov $SYS_ptrace,%eax
KERNCALL
___
svn-src-head@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
Re: svn commit: r304928 - in head/lib/libc: amd64/sys i386/sys sys
On 28.08.2016 2:03, Konstantin Belousov wrote: > Since ptrace(2) syscall can return -1 for non-error situations, libc > wrappers set errno to 0 before performing the syscall, as the service > to the caller. Both C99 and POSIX directly prohibits any standard function to set errno to 0. ptrace() should either choose other errno to indicate non-error situation or change return -1 to something else. ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
Re: svn commit: r304928 - in head/lib/libc: amd64/sys i386/sys sys
On 28.08.2016 3:38, Andrey Chernov wrote: > On 28.08.2016 2:03, Konstantin Belousov wrote: >> Since ptrace(2) syscall can return -1 for non-error situations, libc >> wrappers set errno to 0 before performing the syscall, as the service >> to the caller. > > Both C99 and POSIX directly prohibits any standard function to set errno > to 0. ptrace() should either choose other errno to indicate non-error > situation or change return -1 to something else. > ...and don't touch errno. POSIX: "No function in this volume of POSIX.1-2008 shall set errno to zero." > On both i386 and amd64, the errno symbol was directly > referenced, which only works correctly in single-threaded process. POSIX: "For each thread of a process, the value of errno shall not be affected by function calls or assignments to errno by other threads." ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
Re: svn commit: r304928 - in head/lib/libc: amd64/sys i386/sys sys
On Sun, Aug 28, 2016 at 03:38:10AM +0300, Andrey Chernov wrote: > On 28.08.2016 2:03, Konstantin Belousov wrote: > > Since ptrace(2) syscall can return -1 for non-error situations, libc > > wrappers set errno to 0 before performing the syscall, as the service > > to the caller. > > Both C99 and POSIX directly prohibits any standard function to set errno > to 0. ptrace() should either choose other errno to indicate non-error > situation or change return -1 to something else. > ptrace(2) is not a standard function. And, we cannot break ABI for the syscall. ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
Re: svn commit: r304928 - in head/lib/libc: amd64/sys i386/sys sys
On 28.08.2016 3:56, Konstantin Belousov wrote: > On Sun, Aug 28, 2016 at 03:38:10AM +0300, Andrey Chernov wrote: >> On 28.08.2016 2:03, Konstantin Belousov wrote: >>> Since ptrace(2) syscall can return -1 for non-error situations, libc >>> wrappers set errno to 0 before performing the syscall, as the service >>> to the caller. >> >> Both C99 and POSIX directly prohibits any standard function to set errno >> to 0. ptrace() should either choose other errno to indicate non-error >> situation or change return -1 to something else. >> > ptrace(2) is not a standard function. > And, we cannot break ABI for the syscall. > C99 statement sounds stricter: "The value of errno is zero at program startup, but is never set to zero by any library function. 176)" And syscall is not different from library function from C99 point of view. ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
Re: svn commit: r304928 - in head/lib/libc: amd64/sys i386/sys sys
On Sun, Aug 28, 2016 at 03:50:04AM +0300, Andrey Chernov wrote: > On 28.08.2016 3:38, Andrey Chernov wrote: > > On 28.08.2016 2:03, Konstantin Belousov wrote: > >> Since ptrace(2) syscall can return -1 for non-error situations, libc > >> wrappers set errno to 0 before performing the syscall, as the service > >> to the caller. > > > > Both C99 and POSIX directly prohibits any standard function to set errno > > to 0. ptrace() should either choose other errno to indicate non-error > > situation or change return -1 to something else. > > > ...and don't touch errno. > > POSIX: "No function in this volume of POSIX.1-2008 shall set errno to zero." I am quite curious where ptrace(2) is defined by POSIX. > > > On both i386 and amd64, the errno symbol was directly > > referenced, which only works correctly in single-threaded process. > > POSIX: "For each thread of a process, the value of errno shall not be > affected by function calls or assignments to errno by other threads." And ? What should the citation add new to the substance of the code change ? On Sun, Aug 28, 2016 at 04:04:00AM +0300, Andrey Chernov wrote: > On 28.08.2016 3:56, Konstantin Belousov wrote: > > On Sun, Aug 28, 2016 at 03:38:10AM +0300, Andrey Chernov wrote: > >> On 28.08.2016 2:03, Konstantin Belousov wrote: > >>> Since ptrace(2) syscall can return -1 for non-error situations, libc > >>> wrappers set errno to 0 before performing the syscall, as the service > >>> to the caller. > >> > >> Both C99 and POSIX directly prohibits any standard function to set errno > >> to 0. ptrace() should either choose other errno to indicate non-error > >> situation or change return -1 to something else. > >> > > ptrace(2) is not a standard function. > > And, we cannot break ABI for the syscall. > > > > C99 statement sounds stricter: > "The value of errno is zero at program startup, but is never set to zero > by any library function. 176)" > And syscall is not different from library function from C99 point of view. Point me to a single line in C99 which mentions ptrace(). Do you understand what did the commit changed, and what it did not ? Setting errno to zero before the syscall was the existing behaviour before the change, and I did not modified anything there. But previous wrapper set errno to zero in main thread even if called from some other thread, which was the bug fixed. ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
Re: svn commit: r304928 - in head/lib/libc: amd64/sys i386/sys sys
On 28.08.2016 4:04, Andrey Chernov wrote: > On 28.08.2016 3:56, Konstantin Belousov wrote: >> On Sun, Aug 28, 2016 at 03:38:10AM +0300, Andrey Chernov wrote: >>> On 28.08.2016 2:03, Konstantin Belousov wrote: Since ptrace(2) syscall can return -1 for non-error situations, libc wrappers set errno to 0 before performing the syscall, as the service to the caller. >>> >>> Both C99 and POSIX directly prohibits any standard function to set errno >>> to 0. ptrace() should either choose other errno to indicate non-error >>> situation or change return -1 to something else. >>> >> ptrace(2) is not a standard function. >> > > C99 statement sounds stricter: > "The value of errno is zero at program startup, but is never set to zero > by any library function. 176)" > And syscall is not different from library function from C99 point of view. > >> And, we cannot break ABI for the syscall. We can fix already broken (from standards point of view) ABI for the syscall. ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
Re: svn commit: r304928 - in head/lib/libc: amd64/sys i386/sys sys
On 28.08.2016 4:15, Konstantin Belousov wrote: >> POSIX: "No function in this volume of POSIX.1-2008 shall set errno to zero." > I am quite curious where ptrace(2) is defined by POSIX. POSIX just repeats C99 statement for its own functions, supporting this rule too, but C99 rule is more general and related to any library functions. >> POSIX: "For each thread of a process, the value of errno shall not be >> affected by function calls or assignments to errno by other threads." > And ? What should the citation add new to the substance > of the code change ? This is for your comment "On both i386 and amd64, the errno symbol was directly referenced, which only works correctly in single-threaded process." >> C99 statement sounds stricter: >> "The value of errno is zero at program startup, but is never set to zero >> by any library function. 176)" >> And syscall is not different from library function from C99 point of view. > Point me to a single line in C99 which mentions ptrace(). > > Do you understand what did the commit changed, and what it did not ? > Setting errno to zero before the syscall was the existing behaviour > before the change, and I did not modified anything there. But previous > wrapper set errno to zero in main thread even if called from some other > thread, which was the bug fixed. If you may notice, I don't blame you and don't say that you introduce setting errno to 0. I just want to bring your attention to the problem while you are in that area and familiar with it. ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
Re: svn commit: r304928 - in head/lib/libc: amd64/sys i386/sys sys
On 28.08.2016 4:25, Andrey Chernov wrote: >> Point me to a single line in C99 which mentions ptrace(). Already done: ptrace == "any library function". ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
Re: svn commit: r304928 - in head/lib/libc: amd64/sys i386/sys sys
On 28.08.2016 4:37, Andrey Chernov wrote: > On 28.08.2016 4:25, Andrey Chernov wrote: >>> Point me to a single line in C99 which mentions ptrace(). > > Already done: ptrace == "any library function". To elaborate it more, C99 does not have finite list of library functions, but it says about standard libraries in general (not user libraries), and ptrace() belongs to standard library according to its manpage: Standard C Library (libc, -lc) ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
Re: svn commit: r304928 - in head/lib/libc: amd64/sys i386/sys sys
On Sun, Aug 28, 2016 at 04:25:46AM +0300, Andrey Chernov wrote: > On 28.08.2016 4:15, Konstantin Belousov wrote: > >> POSIX: "No function in this volume of POSIX.1-2008 shall set errno to > >> zero." > > I am quite curious where ptrace(2) is defined by POSIX. > > POSIX just repeats C99 statement for its own functions, supporting this > rule too, but C99 rule is more general and related to any library functions. > > >> POSIX: "For each thread of a process, the value of errno shall not be > >> affected by function calls or assignments to errno by other threads." > > And ? What should the citation add new to the substance > > of the code change ? > > This is for your comment "On both i386 and amd64, the errno symbol was > directly referenced, which only works correctly in single-threaded > process." I still do not understand what you want to say there. Errno as the symbol existing in the symbol table of libc, gives 'POSIX errno' value for the main thread. Preprocessor definition converts C language accesses to errno into some indirections which result in accesses to per-thread errno location. The bug in x86 asm code was due to direct usage of errno. What POSIX requires from the C-level errno symbol does not define a semantic for the memory location pointed to by the errno sym-table symbol. And amusingly, all other arches did it right, except aarch64 and risc-v copied from aarch64. They lack the wrapper at all, I wrote aarch64 ptrace.S already. > > >> C99 statement sounds stricter: > >> "The value of errno is zero at program startup, but is never set to zero > >> by any library function. 176)" > >> And syscall is not different from library function from C99 point of view. > > Point me to a single line in C99 which mentions ptrace(). > > > > Do you understand what did the commit changed, and what it did not ? > > Setting errno to zero before the syscall was the existing behaviour > > before the change, and I did not modified anything there. But previous > > wrapper set errno to zero in main thread even if called from some other > > thread, which was the bug fixed. > > If you may notice, I don't blame you and don't say that you introduce > setting errno to 0. I just want to bring your attention to the problem > while you are in that area and familiar with it. > I know that POSIX requires that POSIX-defined functions did not modified errno except on error, but it cannot require anything from functions which are not defined by the standard. I agree that it would be more consistent for ptrace(2) to not do that as well, but the behaviour is already there for 35 years and I do not view the 'consistency' as a serious reason to break ABI and introduce random failures for innocent consumers of it. On Sun, Aug 28, 2016 at 04:37:11AM +0300, Andrey Chernov wrote: > On 28.08.2016 4:25, Andrey Chernov wrote: > >> Point me to a single line in C99 which mentions ptrace(). > > Already done: ptrace == "any library function". *Shaking head* 'Library functions' references in the context of C99/C11 are implicitely limited to the functions defined by the chapter 7 Library of the standard. To play this sillyness to the end, please answer whether I am allowed to provide static functions definitions in the libraries headers ? E.g. clause 7.1.2 6 of C11 says: Any declaration of a library function shall have external linkage. ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
Re: svn commit: r304436 - in head: . sys/netinet
On 26 August 2016 at 17:46, Slawa Olhovchenkov wrote:
> On Fri, Aug 26, 2016 at 04:55:34PM -0700, Adrian Chadd wrote:
>
>> Hi,
>>
>> I use the kernel lock profiling debugging,
>
> I am already have 100% utilise all CPU cores, I think this is drop
> performance?
>
>> but you can use dtrace to
>> get an idea:
>>
>> dtrace -n 'lockstat:::adaptive-block { @[stack()] = sum(arg1); }'
>
> How to interpret results (how to distinct lock contention from lock
> cost/overhead (LOCK CMPXCGQ is very expensive))?
well, paste the results? :)
the lock is expensive because it's contended. :)
-a
>> (https://wiki.freebsd.org/DTrace/One-Liners)
>>
>>
>>
>> -adrian
___
svn-src-head@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
Re: svn commit: r304928 - in head/lib/libc: amd64/sys i386/sys sys
On 28.08.2016 4:52, Konstantin Belousov wrote: POSIX: "For each thread of a process, the value of errno shall not be affected by function calls or assignments to errno by other threads." >>> And ? What should the citation add new to the substance >>> of the code change ? >> >> This is for your comment "On both i386 and amd64, the errno symbol was >> directly referenced, which only works correctly in single-threaded >> process." > I still do not understand what you want to say there. Errno as the > symbol existing in the symbol table of libc, gives 'POSIX errno' value > for the main thread. Preprocessor definition converts C language > accesses to errno into some indirections which result in accesses to > per-thread errno location. The bug in x86 asm code was due to direct > usage of errno. This particular quote is not describing a problem, it supports your change. > I know that POSIX requires that POSIX-defined functions did not modified > errno except on error, POSIX don't say it. You may modify errno to any value besides 0 while returning success from the function excepting only those functions where POSIX directly states they can't modify errno. I.e. only 0 is disallowed in all cases. > I agree that it would be more > consistent for ptrace(2) to not do that as well, but the behaviour is > already there for 35 years and I do not view the 'consistency' as a serious > reason to break ABI and introduce random failures for innocent consumers > of it. How hard it will be to bring ptrace() to what C99 expects? Perhaps now time is suited well to change some obsoleted things. >> Already done: ptrace == "any library function". > *Shaking head* > > 'Library functions' references in the context of C99/C11 are implicitely > limited to the functions defined by the chapter 7 Library of the standard. No, they are limited to the libraries described in ISO/IEC International Standards family which have C standard library among them. C standard library described in ANSI C standard which is: ISO/IEC (1999). ISO/IEC 9899:1999(E): Programming Languages. libc is C standard library with extensions, and C99 directly says about them: "conforming implementation may have extensions (including additional library functions), provided they do not alter the behavior of any strictly conforming program.3)" ptrace() is extension (additional library function) so can't set errno to 0 (it breaks strictly conforming program). > To play this sillyness to the end, please answer whether I am allowed to > provide static functions definitions in the libraries headers ? E.g. clause > 7.1.2 6 of C11 says: > Any declaration of a library function shall have external linkage. You can, as long as your static function (i.e. extension) do not alter the behavior of any strictly conforming program (see the quote above). ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r304932 - head/sys/mips/conf
Author: adrian Date: Sun Aug 28 04:03:52 2016 New Revision: 304932 URL: https://svnweb.freebsd.org/changeset/base/304932 Log: [mips] QCA955x config file changes to bring things into alignment with the other atheros config files. * ALT_BREAK_TO_DEBUGGER, because we get random breaks when unplugging USB to serial adapters * Build more modules! Modified: head/sys/mips/conf/QCA955X_BASE Modified: head/sys/mips/conf/QCA955X_BASE == --- head/sys/mips/conf/QCA955X_BASE Sun Aug 28 00:26:06 2016 (r304931) +++ head/sys/mips/conf/QCA955X_BASE Sun Aug 28 04:03:52 2016 (r304932) @@ -17,7 +17,7 @@ cpu CPU_MIPS74K makeoptionsKERNLOADADDR=0x8005 optionsHZ=1000 -optionsBREAK_TO_DEBUGGER +#options BREAK_TO_DEBUGGER optionsALT_BREAK_TO_DEBUGGER # options BOOTVERBOSE=10 @@ -26,8 +26,7 @@ files "../atheros/files.ar71xx" hints "QCA955X_BASE.hints" makeoptionsDEBUG=-g#Build kernel with gdb(1) debug symbols -# makeoptions MODULES_OVERRIDE="gpio ar71xx if_gif if_gre if_bridge bridgestp usb wlan wlan_xauth wlan_acl wlan_wep wlan_tkip wlan_ccmp wlan_rssadapt wlan_amrr ath ath_ahb hwpmc" -makeoptionsMODULES_OVERRIDE="if_vlan ipfw if_gre if_gif if_bridge bridgestp" +makeoptionsMODULES_OVERRIDE="gpio ar71xx if_gif if_vlan if_gre if_tap if_tun if_bridge bridgestp usb wlan wlan_xauth wlan_acl wlan_wep wlan_tkip wlan_ccmp wlan_rssadapt wlan_amrr hwpmc ipfw ipfw_nat libalias ipfw_nptv6 urtwn urtwnfw otus otusfw" optionsDDB optionsKDB ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r304933 - head/sys/mips/conf
Author: adrian Date: Sun Aug 28 04:06:36 2016 New Revision: 304933 URL: https://svnweb.freebsd.org/changeset/base/304933 Log: [mips] make AR933X boards use ALT_BREAK_TO_DEBUGGER like the rest of the Atheros boards. Modified: head/sys/mips/conf/std.AR933X Modified: head/sys/mips/conf/std.AR933X == --- head/sys/mips/conf/std.AR933X Sun Aug 28 04:03:52 2016 (r304932) +++ head/sys/mips/conf/std.AR933X Sun Aug 28 04:06:36 2016 (r304933) @@ -25,6 +25,7 @@ makeoptions MODULES_OVERRIDE="gpio ar71x optionsDDB optionsKDB optionsALQ +options ALT_BREAK_TO_DEBUGGER optionsSCHED_4BSD #4BSD scheduler optionsINET#InterNETworking ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r304934 - head/sys/mips/conf
Author: adrian Date: Sun Aug 28 04:06:57 2016 New Revision: 304934 URL: https://svnweb.freebsd.org/changeset/base/304934 Log: Since i'm using carambola 2 for more debugging, enable ARGE_DEBUG by default. Modified: head/sys/mips/conf/CARAMBOLA2 Modified: head/sys/mips/conf/CARAMBOLA2 == --- head/sys/mips/conf/CARAMBOLA2 Sun Aug 28 04:06:36 2016 (r304933) +++ head/sys/mips/conf/CARAMBOLA2 Sun Aug 28 04:06:57 2016 (r304934) @@ -31,6 +31,7 @@ options AR71XX_REALMEM=(64*1024* #deviceiic # Options required for miiproxy and mdiobus +optionsARGE_DEBUG optionsARGE_MDIO # Export an MDIO bus separate from arge device miiproxy# MDIO bus <-> MII PHY rendezvous ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r304936 - head/sys/mips/conf
Author: adrian Date: Sun Aug 28 04:08:04 2016 New Revision: 304936 URL: https://svnweb.freebsd.org/changeset/base/304936 Log: Use ALT_BREAK_TO_DEBUGGER as well. Modified: head/sys/mips/conf/std.AR934X Modified: head/sys/mips/conf/std.AR934X == --- head/sys/mips/conf/std.AR934X Sun Aug 28 04:07:34 2016 (r304935) +++ head/sys/mips/conf/std.AR934X Sun Aug 28 04:08:04 2016 (r304936) @@ -25,6 +25,7 @@ makeoptions MODULES_OVERRIDE="gpio ar71x optionsDDB optionsKDB +optionsALT_BREAK_TO_DEBUGGER optionsALQ optionsSCHED_4BSD #4BSD scheduler ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r304935 - head/sys/mips/conf
Author: adrian Date: Sun Aug 28 04:07:34 2016 New Revision: 304935 URL: https://svnweb.freebsd.org/changeset/base/304935 Log: .. and since others are using wdr3600/wdr4300 right now and are having arge issues with spammed interrupts / link issues, flip on ARGE_DEBUG too. Modified: head/sys/mips/conf/TL-WDR4300 Modified: head/sys/mips/conf/TL-WDR4300 == --- head/sys/mips/conf/TL-WDR4300 Sun Aug 28 04:06:57 2016 (r304934) +++ head/sys/mips/conf/TL-WDR4300 Sun Aug 28 04:07:34 2016 (r304935) @@ -24,6 +24,7 @@ options AR71XX_REALMEM=(128*1024 #deviceiic # Options required for miiproxy and mdiobus +optionsARGE_DEBUG optionsARGE_MDIO # Export an MDIO bus separate from arge device miiproxy# MDIO bus <-> MII PHY rendezvous ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r304937 - head/sys/mips/conf
Author: adrian Date: Sun Aug 28 04:08:26 2016 New Revision: 304937 URL: https://svnweb.freebsd.org/changeset/base/304937 Log: [mips] migrate the QCA955X_BASE file to std.QCA955X, to bring this in line with other board defaults. Added: head/sys/mips/conf/std.QCA955X - copied, changed from r304936, head/sys/mips/conf/QCA955X_BASE Deleted: head/sys/mips/conf/QCA955X_BASE Modified: head/sys/mips/conf/AP135 head/sys/mips/conf/DIR-655A1 head/sys/mips/conf/TL-ARCHERC7V2 head/sys/mips/conf/TL-WR1043NDv2 Modified: head/sys/mips/conf/AP135 == --- head/sys/mips/conf/AP135Sun Aug 28 04:08:04 2016(r304936) +++ head/sys/mips/conf/AP135Sun Aug 28 04:08:26 2016(r304937) @@ -15,7 +15,7 @@ #NO_UNIVERSE # Include the default QCA955x parameters -include "QCA955X_BASE" +include "std.QCA955X" ident AP135 Modified: head/sys/mips/conf/DIR-655A1 == --- head/sys/mips/conf/DIR-655A1Sun Aug 28 04:08:04 2016 (r304936) +++ head/sys/mips/conf/DIR-655A1Sun Aug 28 04:08:26 2016 (r304937) @@ -10,7 +10,7 @@ #NO_UNIVERSE # Include the default QCA955x parameters -include "QCA955X_BASE" +include "std.QCA955X" ident DIR-655A1 Modified: head/sys/mips/conf/TL-ARCHERC7V2 == --- head/sys/mips/conf/TL-ARCHERC7V2Sun Aug 28 04:08:04 2016 (r304936) +++ head/sys/mips/conf/TL-ARCHERC7V2Sun Aug 28 04:08:26 2016 (r304937) @@ -15,7 +15,7 @@ #NO_UNIVERSE # Include the default QCA955x parameters -include "QCA955X_BASE" +include "std.QCA955X" ident TL-ARCHERC7V2 Modified: head/sys/mips/conf/TL-WR1043NDv2 == --- head/sys/mips/conf/TL-WR1043NDv2Sun Aug 28 04:08:04 2016 (r304936) +++ head/sys/mips/conf/TL-WR1043NDv2Sun Aug 28 04:08:26 2016 (r304937) @@ -10,7 +10,7 @@ #NO_UNIVERSE # Include the default QCA955x parameters -include "QCA955X_BASE" +include "std.QCA955X" ident TL-WR1043NDv2 Copied and modified: head/sys/mips/conf/std.QCA955X (from r304936, head/sys/mips/conf/QCA955X_BASE) == --- head/sys/mips/conf/QCA955X_BASE Sun Aug 28 04:08:04 2016 (r304936, copy source) +++ head/sys/mips/conf/std.QCA955X Sun Aug 28 04:08:26 2016 (r304937) @@ -12,7 +12,7 @@ # machine mips mips -ident QCA955X_BASE +#ident std.QCA955X cpuCPU_MIPS74K makeoptionsKERNLOADADDR=0x8005 optionsHZ=1000 ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r304938 - head/sys/mips/conf
Author: adrian Date: Sun Aug 28 04:10:23 2016 New Revision: 304938 URL: https://svnweb.freebsd.org/changeset/base/304938 Log: [mips] move AR91XX_BASE to std.AR91XX. Added: head/sys/mips/conf/std.AR91XX - copied, changed from r304937, head/sys/mips/conf/AR91XX_BASE Deleted: head/sys/mips/conf/AR91XX_BASE Modified: head/sys/mips/conf/TP-WN1043ND Modified: head/sys/mips/conf/TP-WN1043ND == --- head/sys/mips/conf/TP-WN1043ND Sun Aug 28 04:08:26 2016 (r304937) +++ head/sys/mips/conf/TP-WN1043ND Sun Aug 28 04:10:23 2016 (r304938) @@ -7,7 +7,7 @@ #NO_UNIVERSE # Include the default AR913x parameters common to all AR913x SoC users. -include "AR91XX_BASE" +include "std.AR91XX" ident TP-WN1043ND Copied and modified: head/sys/mips/conf/std.AR91XX (from r304937, head/sys/mips/conf/AR91XX_BASE) == --- head/sys/mips/conf/AR91XX_BASE Sun Aug 28 04:08:26 2016 (r304937, copy source) +++ head/sys/mips/conf/std.AR91XX Sun Aug 28 04:10:23 2016 (r304938) @@ -11,7 +11,7 @@ # machine mips mips -ident AR91XX_BASE +#ident std.AR91XX cpuCPU_MIPS24K makeoptionsKERNLOADADDR=0x8005 optionsHZ=1000 ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
Re: svn commit: r304912 - head/sys/sys
... ===> lib/libnv (obj,all,install) In file included from /usr/home/adrian/work/freebsd/head-embedded/src/lib/libnv/../../sys/contrib/libnv/cnvlist.c:49:0: /usr/home/adrian/work/freebsd/head-embedded/src/lib/libnv/../../sys/sys/cnv.h:107:6: error: redundant redeclaration of 'cnvlist_free_descriptor' [-Werror=redundant-decls] void cnvlist_free_descriptor(nvlist_t *nvl, void *cookiep); ^ /usr/home/adrian/work/freebsd/head-embedded/src/lib/libnv/../../sys/sys/cnv.h:101:6: note: previous declaration of 'cnvlist_free_descriptor' was here void cnvlist_free_descriptor(nvlist_t *nvl, void *cookiep); ^ cc1: all warnings being treated as errors --- cnvlist.o --- *** [cnvlist.o] Error code 1 make[4]: stopped in /usr/home/adrian/work/freebsd/head-embedded/src/lib/libnv 1 error -a On 27 August 2016 at 08:22, Mariusz Zaborski wrote: > Author: oshogbo > Date: Sat Aug 27 15:22:55 2016 > New Revision: 304912 > URL: https://svnweb.freebsd.org/changeset/base/304912 > > Log: > Add missed header file for cnv.h . > > Submitted by: Adam Starak > Reported by: ache@ > > Added: > head/sys/sys/cnv.h (contents, props changed) > > Added: head/sys/sys/cnv.h > == > --- /dev/null 00:00:00 1970 (empty, because file is newly added) > +++ head/sys/sys/cnv.h Sat Aug 27 15:22:55 2016(r304912) > @@ -0,0 +1,113 @@ > +/*- > + * Copyright (c) 2016 Adam Starak > + * All rights reserved. > + * > + * Redistribution and use in source and binary forms, with or without > + * modification, are permitted provided that the following conditions > + * are met: > + * 1. Redistributions of source code must retain the above copyright > + *notice, this list of conditions and the following disclaimer. > + * 2. Redistributions in binary form must reproduce the above copyright > + *notice, this list of conditions and the following disclaimer in the > + *documentation and/or other materials provided with the distribution. > + * > + * THIS SOFTWARE IS PROVIDED BY THE AUTHORS AND CONTRIBUTORS ``AS IS'' AND > + * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE > + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE > + * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHORS OR CONTRIBUTORS BE LIABLE > + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL > + * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS > + * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) > + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT > + * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY > + * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF > + * SUCH DAMAGE. > + * > + * $FreeBSD$ > + */ > + > +#ifndef_CNV_H_ > +#define_CNV_H_ > + > +#include > + > +#ifndef _KERNEL > +#include > +#include > +#include > +#include > +#endif > + > +#ifndef_NVLIST_T_DECLARED > +#define_NVLIST_T_DECLARED > +struct nvlist; > + > +typedef struct nvlist nvlist_t; > +#endif > + > +__BEGIN_DECLS > + > +/* > + * The cnvlist_get functions returns value associated with the given cookie. > + * If it returns a pointer, the pointer represents internal buffer and should > + * not be freed by the caller. > + */ > + > +boolcnvlist_get_bool(void *cookiep); > +uint64_tcnvlist_get_number(void *cookiep); > +const char *cnvlist_get_string(void *cookiep); > +const nvlist_t *cnvlist_get_nvlist(void *cookiep); > +const void *cnvlist_get_binary(void *cookiep, size_t *sizep); > +const bool *cnvlist_get_bool_array(void *cookiep, size_t > *nitemsp); > +const uint64_t *cnvlist_get_number_array(void *cookiep, size_t > *nitemsp); > +const char * const *cnvlist_get_string_array(void *cookiep, size_t > *nitemsp); > +const nvlist_t * const *cnvlist_get_nvlist_array(void *cookiep, size_t > *nitemsp); > +#ifndef _KERNEL > +int cnvlist_get_descriptor(void *cookiep); > +const int *cnvlist_get_descriptor_array(void *cookiep, size_t > *nitemsp); > +#endif > + > + > +/* > + * The cnvlist_take functions returns value associated with the given cookie > and > + * remove the given entry from the nvlist. > + * The caller is responsible for freeing received data. > + */ > + > +bool cnvlist_take_bool(nvlist_t *nvl, void *cookiep); > +uint64_t cnvlist_take_number(nvlist_t *nvl, void *cookiep); > +char*cnvlist_take_string(nvlist_t *nvl, void *cookiep); > +nvlist_t*cnvlist_take_nvlist(nvlist_t *nvl, void *cookiep); > +void*cnvlist_take_binary(nvlist_t *nvl, void *cookiep, > size_t *sizep); > +bool*cnvlist_take_bool_array(nvlist_t *nvl, void > *cookiep, size_t *nite
Re: svn commit: r304928 - in head/lib/libc: amd64/sys i386/sys sys
On 28.08.2016 5:33, Andrey Chernov wrote: > "conforming implementation may have extensions (including additional > library functions), provided they do not alter the behavior of any > strictly conforming program.3)" > > ptrace() is extension (additional library function) so can't set errno > to 0 (it breaks strictly conforming program). Sorry for misguiding in this particular part. I confuse the "strictly conforming program" with the "conforming program" term used there too. "Strictly conforming program" can't call ptrace(). In any case our own ptrace(2) manpage suggest to set errno to 0 manually _before_ ptrace() call and do not relay on ptrace() to do it by itself: "Some requests can cause ptrace() to return -1 as a non-error value; to disambiguate, errno can be set to 0 before the call and checked afterwards." It will be better to stay common policy about errno even for extensions to not keep exclusions in the mind. ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r304940 - head/sys/dev/uart
Author: jhibbits
Date: Sun Aug 28 04:40:27 2016
New Revision: 304940
URL: https://svnweb.freebsd.org/changeset/base/304940
Log:
Fix UART PPS capture mode printing
* Add breaks to prevent fallthrough and printing of multiple modes.
* Only check the mode, mask out all other bits.
Modified:
head/sys/dev/uart/uart_core.c
Modified: head/sys/dev/uart/uart_core.c
==
--- head/sys/dev/uart/uart_core.c Sun Aug 28 04:15:20 2016
(r304939)
+++ head/sys/dev/uart/uart_core.c Sun Aug 28 04:40:27 2016
(r304940)
@@ -97,15 +97,19 @@ uart_pps_print_mode(struct uart_softc *s
{
device_printf(sc->sc_dev, "PPS capture mode: ");
- switch(sc->sc_pps_mode) {
+ switch(sc->sc_pps_mode & UART_PPS_SIGNAL_MASK) {
case UART_PPS_DISABLED:
printf("disabled");
+ break;
case UART_PPS_CTS:
printf("CTS");
+ break;
case UART_PPS_DCD:
printf("DCD");
+ break;
default:
printf("invalid");
+ break;
}
if (sc->sc_pps_mode & UART_PPS_INVERT_PULSE)
printf("-Inverted");
___
svn-src-head@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r304941 - head/tools/tools/net80211/wlanwds
Author: adrian
Date: Sun Aug 28 04:54:02 2016
New Revision: 304941
URL: https://svnweb.freebsd.org/changeset/base/304941
Log:
[net80211] fix up wlanwds to compile without warnings on clang.
Modified:
head/tools/tools/net80211/wlanwds/wlanwds.c
Modified: head/tools/tools/net80211/wlanwds/wlanwds.c
==
--- head/tools/tools/net80211/wlanwds/wlanwds.c Sun Aug 28 04:40:27 2016
(r304940)
+++ head/tools/tools/net80211/wlanwds/wlanwds.c Sun Aug 28 04:54:02 2016
(r304941)
@@ -182,7 +182,7 @@ static int
getparent(const char *ifname, char parent[IFNAMSIZ+1])
{
char oid[256];
- int parentlen;
+ size_t parentlen;
/* fetch parent interface name */
snprintf(oid, sizeof(oid), "net.wlan.%s.%%parent", ifname+4);
@@ -239,7 +239,7 @@ iswdsvap(int s, const char *ifname)
* to have already verified this is possible.
*/
static void
-getbssid(int s, const char *ifname, char bssid[IEEE80211_ADDR_LEN])
+getbssid(int s, const char *ifname, uint8_t bssid[IEEE80211_ADDR_LEN])
{
struct ieee80211req ireq;
@@ -261,7 +261,7 @@ static void
scanforvaps(int s)
{
char ifname[IFNAMSIZ+1];
- char bssid[IEEE80211_ADDR_LEN];
+ uint8_t bssid[IEEE80211_ADDR_LEN];
int i;
/* XXX brutal; should just walk sysctl tree */
___
svn-src-head@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
Re: svn commit: r304928 - in head/lib/libc: amd64/sys i386/sys sys
On Sun, 28 Aug 2016, Andrey Chernov wrote: On 28.08.2016 4:52, Konstantin Belousov wrote: POSIX: "For each thread of a process, the value of errno shall not be affected by function calls or assignments to errno by other threads." And ? What should the citation add new to the substance of the code change ? This is for your comment "On both i386 and amd64, the errno symbol was directly referenced, which only works correctly in single-threaded process." I still do not understand what you want to say there. Errno as the symbol existing in the symbol table of libc, gives 'POSIX errno' value for the main thread. Preprocessor definition converts C language accesses to errno into some indirections which result in accesses to per-thread errno location. The bug in x86 asm code was due to direct usage of errno. This particular quote is not describing a problem, it supports your change. I know that POSIX requires that POSIX-defined functions did not modified errno except on error, POSIX don't say it. You may modify errno to any value besides 0 while returning success from the function excepting only those functions where POSIX directly states they can't modify errno. I.e. only 0 is disallowed in all cases. POSIX seems to be very deficient in stating which functions can't modify errno. It doesn't say this clearly even for strtol(). C90 and later C standards say this clearly for strtol() but not many other functions since not many other functions need this in plain C. POSIX only says this indirectly even for strtol() by saying that it defers to the C standard unless stated otherwise. But POSIX has many functions that need this statement, starting with read() with sizes > SSIZE_MAX on implementations that support such sizes. (read() is very hard to use if you use such sizes with it. v7 had the correct arg type (int) to disallow such sizes. Now read() is only safe to use by avoiding such sizes. This is easy to do in applications but not in libraries.) I agree that it would be more consistent for ptrace(2) to not do that as well, but the behaviour is already there for 35 years and I do not view the 'consistency' as a serious reason to break ABI and introduce random failures for innocent consumers of it. How hard it will be to bring ptrace() to what C99 expects? Perhaps now time is suited well to change some obsoleted things. This should be safe to change, since portable applications like gdb can't assume that the implemementation clobbers errno for them. Even FreeBSD's man page doesn't document the FreeBSD behaviour. It documents, with poor wording, that applications must set errno as usual: %%% RETURN VALUES Some requests can cause ptrace() to return -1 as a non-error value; to disambiguate, errno can be set to 0 before the call and checked afterwards. %%% The poor wording is just "errno can be set to 0". It _must_ be set to 0. Also, the function gurantees to not clobber errno so that this checking is guaranteed to work. FreeBSD's man page for strtol says nothing at all about either setting errno before calls or the C90+ guarantees that make this useful. "conforming implementation may have extensions (including additional library functions), provided they do not alter the behavior of any strictly conforming program.3)" ptrace() is extension (additional library function) so can't set errno to 0 (it breaks strictly conforming program). Use of ptrace() makes a program very far from stricty conforming. Only quality of implementation requires ptrace() to follow the usual rules. Bruce ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r304944 - head/sys/contrib/dev/ath/ath_hal/ar9300
Author: adrian Date: Sun Aug 28 05:42:03 2016 New Revision: 304944 URL: https://svnweb.freebsd.org/changeset/base/304944 Log: [ath_hal] add USB PHY control registers for hornet/AR9330. Obtained from:linux ath9k Modified: head/sys/contrib/dev/ath/ath_hal/ar9300/ar9300phy.h Modified: head/sys/contrib/dev/ath/ath_hal/ar9300/ar9300phy.h == --- head/sys/contrib/dev/ath/ath_hal/ar9300/ar9300phy.h Sun Aug 28 05:18:39 2016(r304943) +++ head/sys/contrib/dev/ath/ath_hal/ar9300/ar9300phy.h Sun Aug 28 05:42:03 2016(r304944) @@ -1883,6 +1883,12 @@ #define AR_PHY_PLL_BB_DPLL4 AR_PHY_65NM(overlay_0x16180.Osprey.ch0_bb_dpll4) /* + * Wasp/Hornet PHY USB PLL control + */ +#define AR_PHY_USB_CTRL1 0x16c84 +#define AR_PHY_USB_CTRL2 0x16c88 + +/* * PMU Register Map */ #define AR_PHY_PMU(_field) offsetof(struct pmu_reg, _field) ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
