svn commit: r365739 - in head: share/man/man5 share/mk stand tools/build/options
Author: bdragon
Date: Tue Sep 15 04:22:28 2020
New Revision: 365739
URL: https://svnweb.freebsd.org/changeset/base/365739
Log:
[PowerPC] Remove obsolete MK_LOADER_FORCE_LE
In D12421, the ability to compile stand/ in little-endian was added, with the
intention to extend loader.kboot to run in Petitboot.
However, no further work was done, as the kernel then gained self-execution
capabilities as Petitboot was taught to load FreeBSD kernels directly.
The FreeBSD installer on powerpc64 (on POWER8 and POWER9) uses
/boot/etc/kboot.conf instead of loader.
As this option does nothing but cause stand/ to be miscompiled and actively
causes confusion, remove it.
(I have a functioning petitboot loader in my local tree, however, it turned
out to be quite inconvient to use due to the current petitboot plugin design
so I put it on hold.)
Reviewed by: emaste, imp, jhibbits
Sponsored by: Tag1 Consulting, Inc.
Differential Revision:https://reviews.freebsd.org/D26430
Deleted:
head/tools/build/options/WITH_LOADER_FORCE_LE
Modified:
head/share/man/man5/src.conf.5
head/share/mk/src.opts.mk
head/stand/defs.mk
Modified: head/share/man/man5/src.conf.5
==
--- head/share/man/man5/src.conf.5 Tue Sep 15 00:22:30 2020
(r365738)
+++ head/share/man/man5/src.conf.5 Tue Sep 15 04:22:28 2020
(r365739)
@@ -1,6 +1,6 @@
.\" DO NOT EDIT-- this file is @generated by tools/build/options/makeman.
.\" $FreeBSD$
-.Dd September 11, 2020
+.Dd September 14, 2020
.Dt SRC.CONF 5
.Os
.Sh NAME
@@ -983,9 +983,6 @@ with support for verification based on certificates ob
.It Va WITH_LOADER_FIREWIRE
Enable firewire support in /boot/loader on x86. This option is a nop
on all other platforms.
-.It Va WITH_LOADER_FORCE_LE
-Set to force the powerpc boot loader to launch the kernel in little
-endian mode.
.It Va WITHOUT_LOADER_GELI
Disable inclusion of GELI crypto support in the boot chain binaries.
.Pp
Modified: head/share/mk/src.opts.mk
==
--- head/share/mk/src.opts.mk Tue Sep 15 00:22:30 2020(r365738)
+++ head/share/mk/src.opts.mk Tue Sep 15 04:22:28 2020(r365739)
@@ -211,7 +211,6 @@ __DEFAULT_NO_OPTIONS = \
HESIOD \
LIBSOFT \
LOADER_FIREWIRE \
-LOADER_FORCE_LE \
LOADER_VERBOSE \
LOADER_VERIEXEC_PASS_MANIFEST \
MALLOC_PRODUCTION \
Modified: head/stand/defs.mk
==
--- head/stand/defs.mk Tue Sep 15 00:22:30 2020(r365738)
+++ head/stand/defs.mk Tue Sep 15 04:22:28 2020(r365739)
@@ -175,12 +175,6 @@ DD=dd ${DD_NOSTATUS}
CFLAGS+= -G0 -fno-pic -mno-abicalls
.endif
-.if ${MK_LOADER_FORCE_LE} != "no"
-.if ${MACHINE_ARCH} == "powerpc64"
-CFLAGS+= -mlittle-endian
-.endif
-.endif
-
#
# Have a sensible default
#
___
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: r365737 - head/sys/geom/eli
Author: imp
Date: Mon Sep 14 23:51:14 2020
New Revision: 365737
URL: https://svnweb.freebsd.org/changeset/base/365737
Log:
We don't need the sc_ekeys_lock in standalone environment.
When we bring in geli into the boot loader, we are single threaded so
we don't have to worry about locking. We have no mutexes, and don't need
to use them, so comment it out.
MFC After: 3 days
Modified:
head/sys/geom/eli/g_eli.h
Modified: head/sys/geom/eli/g_eli.h
==
--- head/sys/geom/eli/g_eli.h Mon Sep 14 23:49:17 2020(r365736)
+++ head/sys/geom/eli/g_eli.h Mon Sep 14 23:51:14 2020(r365737)
@@ -181,7 +181,9 @@ struct g_eli_softc {
uint8_t sc_ekey[G_ELI_DATAKEYLEN];
TAILQ_HEAD(, g_eli_key) sc_ekeys_queue;
RB_HEAD(g_eli_key_tree, g_eli_key) sc_ekeys_tree;
+#ifndef _STANDALONE
struct mtx sc_ekeys_lock;
+#endif
uint64_t sc_ekeys_total;
uint64_t sc_ekeys_allocated;
u_intsc_ealgo;
___
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: r365735 - head/sys/dev/iicbus
Author: imp
Date: Mon Sep 14 23:30:04 2020
New Revision: 365735
URL: https://svnweb.freebsd.org/changeset/base/365735
Log:
Don't do the busy dance in icee_open/close
We don't need to do the busy dance for this driver. It's handled by
destroy_dev() entirely. Since all we did was busy/unbusy in
open/close, just delete them. We therefore don't need to track closes
either.
Reviewed by: ian@
Differential Revision: https://reviews.freebsd.org/D26431
Modified:
head/sys/dev/iicbus/icee.c
Modified: head/sys/dev/iicbus/icee.c
==
--- head/sys/dev/iicbus/icee.c Mon Sep 14 23:27:51 2020(r365734)
+++ head/sys/dev/iicbus/icee.c Mon Sep 14 23:30:04 2020(r365735)
@@ -113,17 +113,12 @@ static struct ofw_compat_data compat_data[] = {
#define CDEV2SOFTC(dev)((dev)->si_drv1)
/* cdev routines */
-static d_open_t icee_open;
-static d_close_t icee_close;
static d_read_t icee_read;
static d_write_t icee_write;
static struct cdevsw icee_cdevsw =
{
.d_version = D_VERSION,
- .d_flags = D_TRACKCLOSE,
- .d_open = icee_open,
- .d_close = icee_close,
.d_read = icee_read,
.d_write = icee_write
};
@@ -227,28 +222,6 @@ icee_detach(device_t dev)
struct icee_softc *sc = device_get_softc(dev);
destroy_dev(sc->cdev);
- return (0);
-}
-
-static int
-icee_open(struct cdev *dev, int oflags, int devtype, struct thread *td)
-{
- struct icee_softc *sc;
-
- sc = CDEV2SOFTC(dev);
- if (device_get_state(sc->dev) < DS_BUSY)
- device_busy(sc->dev);
-
- return (0);
-}
-
-static int
-icee_close(struct cdev *dev, int fflag, int devtype, struct thread *td)
-{
- struct icee_softc *sc;
-
- sc = CDEV2SOFTC(dev);
- device_unbusy(sc->dev);
return (0);
}
___
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: r365734 - head/sys/sys
Author: imp Date: Mon Sep 14 23:27:51 2020 New Revision: 365734 URL: https://svnweb.freebsd.org/changeset/base/365734 Log: Tweak what's visible in the standalone environment. We define offsetof in stand.h typically, but when this is included we can define it multiple times. However, we don't define bool in stand.h at the moment, so allow it to be defined inside types.h when we're building for the standalone environment. MFC After: 3 days Modified: head/sys/sys/types.h Modified: head/sys/sys/types.h == --- head/sys/sys/types.hMon Sep 14 22:42:17 2020(r365733) +++ head/sys/sys/types.hMon Sep 14 23:27:51 2020(r365734) @@ -292,6 +292,10 @@ typedef__uint64_t uoff_t; typedefcharvm_memattr_t; /* memory attribute codes */ typedefstruct vm_page *vm_page_t; +#define offsetof(type, field) __offsetof(type, field) +#endif /* _KERNEL */ + +#ifdefined(_KERNEL) || defined(_STANDALONE) #if !defined(__bool_true_false_are_defined) && !defined(__cplusplus) #define__bool_true_false_are_defined 1 #definefalse 0 @@ -301,10 +305,7 @@ typedefint _Bool; #endif typedef_Bool bool; #endif /* !__bool_true_false_are_defined && !__cplusplus */ - -#define offsetof(type, field) __offsetof(type, field) - -#endif /* _KERNEL */ +#endif /* KERNEL || _STANDALONE */ /* * The following are all things that really shouldn't exist in this header, ___ 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: r365732 - head/sys/dev/cxgbe/common
Author: np
Date: Mon Sep 14 22:15:54 2020
New Revision: 365732
URL: https://svnweb.freebsd.org/changeset/base/365732
Log:
cxgbe(4): Get the count of FCS errors from the MAC and not MPS for T6 ports.
The MPS register on the T6 counts something other than FCS errors despite its
name.
MFC after:3 days
Sponsored by: Chelsio Communications
Modified:
head/sys/dev/cxgbe/common/t4_hw.c
Modified: head/sys/dev/cxgbe/common/t4_hw.c
==
--- head/sys/dev/cxgbe/common/t4_hw.c Mon Sep 14 21:33:11 2020
(r365731)
+++ head/sys/dev/cxgbe/common/t4_hw.c Mon Sep 14 22:15:54 2020
(r365732)
@@ -6853,6 +6853,7 @@ void t4_get_port_stats_offset(struct adapter *adap, in
void t4_get_port_stats(struct adapter *adap, int idx, struct port_stats *p)
{
u32 bgmap = adap2pinfo(adap, idx)->mps_bg_map;
+ struct link_config *lc = >port[idx]->link_cfg;
u32 stat_ctl = t4_read_reg(adap, A_MPS_STAT_CTL);
#define GET_STAT(name) \
@@ -6902,7 +6903,6 @@ void t4_get_port_stats(struct adapter *adap, int idx,
p->rx_ucast_frames = GET_STAT(RX_PORT_UCAST);
p->rx_too_long = GET_STAT(RX_PORT_MTU_ERROR);
p->rx_jabber= GET_STAT(RX_PORT_MTU_CRC_ERROR);
- p->rx_fcs_err = GET_STAT(RX_PORT_CRC_ERROR);
p->rx_len_err = GET_STAT(RX_PORT_LEN_ERROR);
p->rx_symbol_err= GET_STAT(RX_PORT_SYM_ERROR);
p->rx_runt = GET_STAT(RX_PORT_LESS_64B);
@@ -6922,6 +6922,26 @@ void t4_get_port_stats(struct adapter *adap, int idx,
p->rx_ppp6 = GET_STAT(RX_PORT_PPP6);
p->rx_ppp7 = GET_STAT(RX_PORT_PPP7);
+ /*
+* The T6's MPS's RX_PORT_CRC_ERROR register doesn't actually count CRC
+* errors so get that information from the MAC instead. Which MAC is in
+* use depends on speed and FEC. The MAC counters clear on reset or
+* link state change so we are only reporting errors for this
+* incarnation of the link here.
+*/
+ if (chip_id(adap) != CHELSIO_T6)
+ p->rx_fcs_err = GET_STAT(RX_PORT_CRC_ERROR);
+ else if (lc->link_ok) {
+ if (lc->speed > 25000 ||
+ (lc->speed == 25000 && lc->fec == FEC_RS)) {
+ p->rx_fcs_err = t4_read_reg64(adap, T5_PORT_REG(idx,
+ A_MAC_PORT_AFRAMECHECKSEQUENCEERRORS));
+ } else {
+ p->rx_fcs_err = t4_read_reg64(adap, T5_PORT_REG(idx,
+ A_MAC_PORT_MTIP_1G10G_RX_CRCERRORS));
+ }
+ }
+
if (chip_id(adap) >= CHELSIO_T5) {
if (stat_ctl & F_COUNTPAUSESTATRX) {
p->rx_frames -= p->rx_pause;
@@ -10757,6 +10777,12 @@ void t4_clr_port_stats(struct adapter *adap, int idx)
t4_write_reg(adap,
A_MPS_STAT_RX_BG_0_MAC_TRUNC_FRAME_L + i * 8, 0);
}
+ if (chip_id(adap) == CHELSIO_T6) {
+ t4_write_reg64(adap, T5_PORT_REG(idx,
+ A_MAC_PORT_AFRAMECHECKSEQUENCEERRORS), 0);
+ t4_write_reg64(adap, T5_PORT_REG(idx,
+ A_MAC_PORT_MTIP_1G10G_RX_CRCERRORS), 0);
+ }
}
/**
___
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: r365729 - head/sys/dev/usb
Author: ian Date: Mon Sep 14 17:33:28 2020 New Revision: 365729 URL: https://svnweb.freebsd.org/changeset/base/365729 Log: Add product ID strings for a couple Microchip usb hubs. Also, update the vendor ID string to say just "Microchip Technology" -- the buyout of Standard Microsystems happened in 2012 and the SMC/SMSC names are pretty much retired at this point. PR: 241406 Modified: head/sys/dev/usb/usbdevs Modified: head/sys/dev/usb/usbdevs == --- head/sys/dev/usb/usbdevsMon Sep 14 17:21:52 2020(r365728) +++ head/sys/dev/usb/usbdevsMon Sep 14 17:33:28 2020(r365729) @@ -103,7 +103,7 @@ vendor CREATIVE 0x041e Creative Labs vendor NOKIA 0x0421 Nokia vendor ADI 0x0422 ADI Systems vendor CATC0x0423 Computer Access Technology -vendor SMC20x0424 Microchip (Standard Microsystems) +vendor SMC20x0424 Microchip Technology vendor MOTOROLA_HK 0x0425 Motorola HK vendor GRAVIS 0x0428 Advanced Gravis Computer vendor CIRRUSLOGIC 0x0429 Cirrus Logic @@ -4427,13 +4427,15 @@ product SMART2 G2MEMKEY 0x1700 G2 Memory Key product SMARTBRIDGES SMARTLINK 0x0001 SmartLink USB Ethernet product SMARTBRIDGES SMARTNIC 0x0003 smartNIC 2 PnP Ethernet -/* SMC products */ +/* Microchip Technology (formerly SMC) products */ product SMC 2102USB0x0100 10Mbps Ethernet product SMC 2202USB0x0200 10/100 Ethernet product SMC 2206USB0x0201 EZ Connect USB Ethernet product SMC 2862WG 0xee13 EZ Connect Wireless Adapter product SMC2 2020HUB 0x2020 USB Hub +product SMC2 2513HUB 0x2513 USB Hub product SMC2 2514HUB 0x2514 USB Hub +product SMC2 2517HUB 0x2517 USB Hub product SMC3 2662WUSB 0xa002 2662W-AR Wireless product SMC2 LAN7800_ETH 0x7800 USB/Ethernet product SMC2 LAN7801_ETH 0x7801 USB/Ethernet ___ 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: r365720 - head/sys/conf
On Mon, Sep 14, 2020 at 8:45 AM Gordon Tetlow wrote:
> Author: gordon
> Date: Mon Sep 14 14:45:30 2020
> New Revision: 365720
> URL: https://svnweb.freebsd.org/changeset/base/365720
>
> Log:
> Partially revert r346018 and use the if/then construct instead of shell.
>
> There are a couple of places in the tree that directly parse the
> newvers.sh
> script looking for the BRANCH variable. I found two locations, one in
> release/Makefile and the other in bin/freebsd-version/Makefile.
>
> While there is a good argument that BRANCH_OVERRIDE should properly
> propagate in those circumstances and the new behavior is thus better, the
> reality is this change broke freebsd-update's ability to find timestamps
> in
> binaries and resulted in a large number of gratuitous changes.
>
Where? I fixed all those ages ago.
Warner
> Reported by: freebsd-update
> Discussed with: cperciva
> MFC after:1 day
>
> Modified:
> head/sys/conf/newvers.sh
>
> Modified: head/sys/conf/newvers.sh
>
> ==
> --- head/sys/conf/newvers.shMon Sep 14 14:24:54 2020(r365719)
> +++ head/sys/conf/newvers.shMon Sep 14 14:45:30 2020(r365720)
> @@ -54,7 +54,10 @@
>
> TYPE="FreeBSD"
> REVISION="13.0"
> -BRANCH="${BRANCH_OVERRIDE:-CURRENT}"
> +BRANCH="CURRENT"
> +if [ -n "${BRANCH_OVERRIDE}" ]; then
> + BRANCH=${BRANCH_OVERRIDE}
> +fi
> RELEASE="${REVISION}-${BRANCH}"
> VERSION="${TYPE} ${RELEASE}"
>
>
___
svn-src-head@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r365727 - in head/usr.bin: . gprof
Author: andrew
Date: Mon Sep 14 16:18:53 2020
New Revision: 365727
URL: https://svnweb.freebsd.org/changeset/base/365727
Log:
Cleanups for gprof:
* Remove identical or almost identical headers
* Only build aout.c on amd64 and i386. None of the the other current
architectures ever supported running a.out binaries
* Enable on all architectures
Sponsored by: Innovate UK
Differential Revision:https://reviews.freebsd.org/D26369
Deleted:
head/usr.bin/gprof/amd64.h
head/usr.bin/gprof/arm.h
head/usr.bin/gprof/i386.h
head/usr.bin/gprof/mips.h
head/usr.bin/gprof/powerpc.h
head/usr.bin/gprof/riscv.h
head/usr.bin/gprof/sparc64.h
Modified:
head/usr.bin/Makefile
head/usr.bin/gprof/Makefile
head/usr.bin/gprof/gprof.h
Modified: head/usr.bin/Makefile
==
--- head/usr.bin/Makefile Mon Sep 14 16:12:28 2020(r365726)
+++ head/usr.bin/Makefile Mon Sep 14 16:18:53 2020(r365727)
@@ -262,11 +262,7 @@ SUBDIR.${MK_TOOLCHAIN}+= ctags
SUBDIR.${MK_TOOLCHAIN}+= cxxfilt
SUBDIR.${MK_TOOLCHAIN}+= objcopy
SUBDIR.${MK_TOOLCHAIN}+= file2c
-# ARM64TODO gprof does not build
-# RISCVTODO gprof does not build
-.if ${MACHINE_ARCH} != "aarch64"
SUBDIR.${MK_TOOLCHAIN}+= gprof
-.endif
SUBDIR.${MK_TOOLCHAIN}+= indent
SUBDIR.${MK_TOOLCHAIN}+= lex
SUBDIR.${MK_TOOLCHAIN}+= mkstr
Modified: head/usr.bin/gprof/Makefile
==
--- head/usr.bin/gprof/Makefile Mon Sep 14 16:12:28 2020(r365726)
+++ head/usr.bin/gprof/Makefile Mon Sep 14 16:18:53 2020(r365727)
@@ -5,8 +5,7 @@ PROG= gprof
SRCS= gprof.c arcs.c dfn.c elf.c lookup.c hertz.c \
printgprof.c printlist.c kernel.c
-.if ${MACHINE_ARCH} != "aarch64" && ${MACHINE_CPUARCH} != "riscv" && \
-${MACHINE_ARCH} != "s390x"
+.if ${MACHINE_CPUARCH} == "amd64" || ${MACHINE_CPUARCH} == "i386"
SRCS+= aout.c
CFLAGS+= -DWITH_AOUT
.endif
Modified: head/usr.bin/gprof/gprof.h
==
--- head/usr.bin/gprof/gprof.h Mon Sep 14 16:12:28 2020(r365726)
+++ head/usr.bin/gprof/gprof.h Mon Sep 14 16:18:53 2020(r365727)
@@ -39,27 +39,14 @@
#include
#include
-#if __amd64__
-# include "amd64.h"
-#endif
-#if __arm__
-# include "arm.h"
-#endif
-#if __i386__
-# include "i386.h"
-#endif
-#if __mips__
-# include "mips.h"
-#endif
-#if __powerpc__
-# include "powerpc.h"
-#endif
-#if __sparc64__
-# include "sparc64.h"
-#endif
-#if __riscv
-#include "riscv.h"
-#endif
+/*
+ * offset (in bytes) of the code from the entry address of a routine.
+ * (see asgnsamples for use and explanation.)
+ */
+#define OFFSET_OF_CODE 0
+
+enum opermodes { dummy };
+typedef enum opermodes operandenum;
/*
* booleans
___
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: r365726 - in head: cddl/lib/libzfs cddl/lib/libzpool lib/libpmc share/mk sys/conf tests/sys/kern usr.bin/gcore
> On 14 Sep 2020, at 17:12, Andrew Turner wrote: > > Author: andrew > Date: Mon Sep 14 16:12:28 2020 > New Revision: 365726 > URL: https://svnweb.freebsd.org/changeset/base/365726 > > Log: > Use MACHINE_CPUARCH when checking for arm64 > > Use MACHINE_CPUARCH with arm64 (aarch64) when we build code that could run > on any 64-bit Arm instruction set. This will simplify checks in downstream > consumers targeting prototype instruction sets. > > The only place we check for MACHINE_ARCH == aarch64 is when building the > device tree blobs. As these are targeting current generation ISAs. > > Sponsored by:Innovate UK > Differential Revision: https://reviews.freebsd.org/D26370 I missed: Reviewed by: brooks Andrew ___ 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: r365726 - in head: cddl/lib/libzfs cddl/lib/libzpool lib/libpmc share/mk sys/conf tests/sys/kern usr.bin/gcore
Author: andrew
Date: Mon Sep 14 16:12:28 2020
New Revision: 365726
URL: https://svnweb.freebsd.org/changeset/base/365726
Log:
Use MACHINE_CPUARCH when checking for arm64
Use MACHINE_CPUARCH with arm64 (aarch64) when we build code that could run
on any 64-bit Arm instruction set. This will simplify checks in downstream
consumers targeting prototype instruction sets.
The only place we check for MACHINE_ARCH == aarch64 is when building the
device tree blobs. As these are targeting current generation ISAs.
Sponsored by: Innovate UK
Differential Revision:https://reviews.freebsd.org/D26370
Modified:
head/cddl/lib/libzfs/Makefile
head/cddl/lib/libzpool/Makefile
head/lib/libpmc/Makefile
head/share/mk/bsd.endian.mk
head/share/mk/bsd.sys.mk
head/sys/conf/dtb.build.mk
head/tests/sys/kern/Makefile
head/usr.bin/gcore/Makefile
Modified: head/cddl/lib/libzfs/Makefile
==
--- head/cddl/lib/libzfs/Makefile Mon Sep 14 15:58:10 2020
(r365725)
+++ head/cddl/lib/libzfs/Makefile Mon Sep 14 16:12:28 2020
(r365726)
@@ -82,7 +82,7 @@ CFLAGS += -DHAVE_SSE2
ARCH_C += zfs_fletcher_avx512.c
CFLAGS+= -DHAVE_AVX2 -DHAVE_AVX -D__x86_64 -DHAVE_AVX512F
.endif
-.if ${MACHINE_ARCH} == "aarch64"
+.if ${MACHINE_CPUARCH} == "aarch64"
ARCH_C += zfs_fletcher_aarch64_neon.c
.endif
Modified: head/cddl/lib/libzpool/Makefile
==
--- head/cddl/lib/libzpool/Makefile Mon Sep 14 15:58:10 2020
(r365725)
+++ head/cddl/lib/libzpool/Makefile Mon Sep 14 16:12:28 2020
(r365726)
@@ -187,7 +187,7 @@ ARCH_C += zfs_fletcher_avx512.c
CFLAGS+= -DHAVE_AVX2 -DHAVE_AVX -D__x86_64 -DHAVE_AVX512F \
-DHAVE_AVX512BW
.endif
-.if ${MACHINE_ARCH} == "aarch64"
+.if ${MACHINE_CPUARCH} == "aarch64"
ARCH_C += zfs_fletcher_aarch64_neon.c
.endif
Modified: head/lib/libpmc/Makefile
==
--- head/lib/libpmc/MakefileMon Sep 14 15:58:10 2020(r365725)
+++ head/lib/libpmc/MakefileMon Sep 14 16:12:28 2020(r365726)
@@ -7,10 +7,10 @@ INCS= pmc.h pmclog.h pmcformat.h
CFLAGS+= -I${SRCTOP}/${RELDIR:H}/libpmcstat
-.if ${MACHINE_ARCH} == "aarch64" || ${MACHINE_ARCH} == "amd64" || \
+.if ${MACHINE_CPUARCH} == "aarch64" || ${MACHINE_ARCH} == "amd64" || \
${MACHINE_ARCH} == "i386"
-.if ${MACHINE_ARCH} == "aarch64"
+.if ${MACHINE_CPUARCH} == "aarch64"
EVENT_ARCH="arm64"
.elif ${MACHINE_ARCH} == "amd64" || ${MACHINE_ARCH} == "i386"
EVENT_ARCH="x86"
Modified: head/share/mk/bsd.endian.mk
==
--- head/share/mk/bsd.endian.mk Mon Sep 14 15:58:10 2020(r365725)
+++ head/share/mk/bsd.endian.mk Mon Sep 14 16:12:28 2020(r365726)
@@ -1,6 +1,6 @@
# $FreeBSD$
-.if ${MACHINE_ARCH} == "aarch64" || \
+.if ${MACHINE_CPUARCH} == "aarch64" || \
${MACHINE_ARCH} == "amd64" || \
${MACHINE_ARCH} == "i386" || \
(${MACHINE} == "arm" && ${MACHINE_ARCH:Marm*eb*} == "") || \
Modified: head/share/mk/bsd.sys.mk
==
--- head/share/mk/bsd.sys.mkMon Sep 14 15:58:10 2020(r365725)
+++ head/share/mk/bsd.sys.mkMon Sep 14 16:12:28 2020(r365726)
@@ -187,7 +187,7 @@ CWARNFLAGS+=
-Wno-error=aggressive-loop-optimizations
.endif
# GCC's own arm_neon.h triggers various warnings
-.if ${MACHINE_ARCH} == "aarch64"
+.if ${MACHINE_CPUARCH} == "aarch64"
CWARNFLAGS+= -Wno-system-headers
.endif
.endif # gcc
Modified: head/sys/conf/dtb.build.mk
==
--- head/sys/conf/dtb.build.mk Mon Sep 14 15:58:10 2020(r365725)
+++ head/sys/conf/dtb.build.mk Mon Sep 14 16:12:28 2020(r365726)
@@ -22,7 +22,7 @@ SYSDIR= ${S}
.for _dts in ${DTS}
# DTB for aarch64 needs to preserve the immediate parent of the .dts, because
# these DTS are vendored and should be installed into their vendored directory.
-.if ${MACHINE_ARCH} == "aarch64"
+.if ${MACHINE_CPUARCH} == "aarch64"
DTB+= ${_dts:R:S/$/.dtb/}
.else
DTB+= ${_dts:T:R:S/$/.dtb/}
Modified: head/tests/sys/kern/Makefile
==
--- head/tests/sys/kern/MakefileMon Sep 14 15:58:10 2020
(r365725)
+++ head/tests/sys/kern/MakefileMon Sep 14 16:12:28 2020
(r365726)
@@ -57,7 +57,7 @@ LIBADD.mqueue_test+= rt
.if ${MACHINE_ARCH} == "amd64" || \
${MACHINE_ARCH} == "i386" || \
-${MACHINE_ARCH} == "aarch64"
+${MACHINE_CPUARCH} == "aarch64"
ATF_TESTS_C+= libkern_crc32
.if ${MACHINE_ARCH} == "amd64" || ${MACHINE_ARCH} == "i386"
LDADD.libkern_crc32+=
svn commit: r365725 - head/release
Author: gjb
Date: Mon Sep 14 15:58:10 2020
New Revision: 365725
URL: https://svnweb.freebsd.org/changeset/base/365725
Log:
Fix setting RELENGDIR when release.sh is called from an external
script.
Sponsored by: Rubicon Communications, LLC (netgate.com)
Modified:
head/release/release.sh
Modified: head/release/release.sh
==
--- head/release/release.sh Mon Sep 14 15:48:30 2020(r365724)
+++ head/release/release.sh Mon Sep 14 15:58:10 2020(r365725)
@@ -57,7 +57,9 @@ usage() {
env_setup() {
# The directory within which the release will be built.
CHROOTDIR="/scratch"
- RELENGDIR="$(dirname $(realpath ${0}))"
+ if [ -z "${RELENGDIR}" ]; then
+ export RELENGDIR="$(dirname $(realpath ${0}))"
+ fi
# The default version control system command to obtain the sources.
for _dir in /usr/bin /usr/local/bin; do
___
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: r365724 - in head/stand: ficl ficl/powerpc powerpc/ofw
Author: bdragon
Date: Mon Sep 14 15:48:30 2020
New Revision: 365724
URL: https://svnweb.freebsd.org/changeset/base/365724
Log:
stand/ficl 64-bit compatibility
Currently, the only thing that prevents a functioning 64-bit FICL build is
a few integer types that were intended to be fixed-width.
Changing them to C99 integer types allows building a functioning 64-bit
FICL.
While this isn't applicable to the default settings of any in-tree loaders,
it is necessary for a future Petitboot loader, due to the requirement that
it be compiled as a 64-bit program.
Reviewed by: tsoome, imp (earlier revision)
Sponsored by: Tag1 Consulting, Inc.
Differential Revision:https://reviews.freebsd.org/D26364
Modified:
head/stand/ficl/ficl.h
head/stand/ficl/powerpc/sysdep.h
head/stand/powerpc/ofw/main.c
Modified: head/stand/ficl/ficl.h
==
--- head/stand/ficl/ficl.h Mon Sep 14 15:20:37 2020(r365723)
+++ head/stand/ficl/ficl.h Mon Sep 14 15:48:30 2020(r365724)
@@ -249,7 +249,7 @@ typedef struct ficl_system_info FICL_SYSTEM_INFO;
** complement of false... that unifies logical and bitwise operations
** nicely.
*/
-#define FICL_TRUE ((unsigned long)~(0L))
+#define FICL_TRUE ((FICL_UNS)~(0LL))
#define FICL_FALSE (0)
#define FICL_BOOL(x) ((x) ? FICL_TRUE : FICL_FALSE)
Modified: head/stand/ficl/powerpc/sysdep.h
==
--- head/stand/ficl/powerpc/sysdep.hMon Sep 14 15:20:37 2020
(r365723)
+++ head/stand/ficl/powerpc/sysdep.hMon Sep 14 15:48:30 2020
(r365724)
@@ -79,15 +79,15 @@
** System dependent data type declarations...
*/
#if !defined INT32
-#define INT32 int
+#define INT32 int32_t
#endif
#if !defined UNS32
-#define UNS32 unsigned int
+#define UNS32 uint32_t
#endif
#if !defined UNS16
-#define UNS16 unsigned short
+#define UNS16 uint16_t
#endif
#if !defined UNS8
@@ -367,6 +367,9 @@ typedef struct
*/
#if !defined FICL_ALIGN
#define FICL_ALIGN 2
+#endif
+
+#if !defined FICL_ALIGN_ADD
#define FICL_ALIGN_ADD ((1 << FICL_ALIGN) - 1)
#endif
Modified: head/stand/powerpc/ofw/main.c
==
--- head/stand/powerpc/ofw/main.c Mon Sep 14 15:20:37 2020
(r365723)
+++ head/stand/powerpc/ofw/main.c Mon Sep 14 15:48:30 2020
(r365724)
@@ -63,7 +63,7 @@ init_heap(void)
{
bzero(heap, HEAP_SIZE);
- setheap(heap, (void *)((int)heap + HEAP_SIZE));
+ setheap(heap, (void *)((uintptr_t)heap + HEAP_SIZE));
}
uint64_t
___
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: r365723 - head/sys/powerpc/powerpc
Author: bdragon
Date: Mon Sep 14 15:20:37 2020
New Revision: 365723
URL: https://svnweb.freebsd.org/changeset/base/365723
Log:
[PowerPC] Make cpu frequency detection endian-independent
On ibm,extended-clock-frequency, ensure we be64toh() the value.
On clock-frequency, remove the right-shifting hack (which was needed due to
reading a 32 bit value into a 64 bit variable) and switch to OF_getencprop()
for reading (which will handle endian conversion internally.)
Reviewed by: jhibbits (in irc)
Sponsored by: Tag1 Consulting, Inc.
Modified:
head/sys/powerpc/powerpc/cpu.c
Modified: head/sys/powerpc/powerpc/cpu.c
==
--- head/sys/powerpc/powerpc/cpu.c Mon Sep 14 14:53:09 2020
(r365722)
+++ head/sys/powerpc/powerpc/cpu.c Mon Sep 14 15:20:37 2020
(r365723)
@@ -72,6 +72,7 @@
#include
#include
#include
+#include
#include
#include
@@ -358,6 +359,7 @@ cpu_est_clockrate(int cpu_id, uint64_t *cps)
uint16_tvers;
register_t msr;
phandle_t cpu, dev, root;
+ uint32_tfreq32;
int res = 0;
charbuf[8];
@@ -428,10 +430,11 @@ cpu_est_clockrate(int cpu_id, uint64_t *cps)
return (ENOENT);
if (OF_getprop(cpu, "ibm,extended-clock-frequency",
cps, sizeof(*cps)) >= 0) {
+ *cps = be64toh(*cps);
return (0);
- } else if (OF_getprop(cpu, "clock-frequency", cps,
- sizeof(cell_t)) >= 0) {
- *cps >>= 32;
+ } else if (OF_getencprop(cpu, "clock-frequency",
+ , sizeof(freq32)) >= 0) {
+ *cps = freq32;
return (0);
} else {
return (ENOENT);
___
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: r365720 - head/sys/conf
Author: gordon
Date: Mon Sep 14 14:45:30 2020
New Revision: 365720
URL: https://svnweb.freebsd.org/changeset/base/365720
Log:
Partially revert r346018 and use the if/then construct instead of shell.
There are a couple of places in the tree that directly parse the newvers.sh
script looking for the BRANCH variable. I found two locations, one in
release/Makefile and the other in bin/freebsd-version/Makefile.
While there is a good argument that BRANCH_OVERRIDE should properly
propagate in those circumstances and the new behavior is thus better, the
reality is this change broke freebsd-update's ability to find timestamps in
binaries and resulted in a large number of gratuitous changes.
Reported by: freebsd-update
Discussed with: cperciva
MFC after:1 day
Modified:
head/sys/conf/newvers.sh
Modified: head/sys/conf/newvers.sh
==
--- head/sys/conf/newvers.shMon Sep 14 14:24:54 2020(r365719)
+++ head/sys/conf/newvers.shMon Sep 14 14:45:30 2020(r365720)
@@ -54,7 +54,10 @@
TYPE="FreeBSD"
REVISION="13.0"
-BRANCH="${BRANCH_OVERRIDE:-CURRENT}"
+BRANCH="CURRENT"
+if [ -n "${BRANCH_OVERRIDE}" ]; then
+ BRANCH=${BRANCH_OVERRIDE}
+fi
RELEASE="${REVISION}-${BRANCH}"
VERSION="${TYPE} ${RELEASE}"
___
svn-src-head@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r365719 - head/sys/dev/mlx5/mlx5_en
Author: hselasky Date: Mon Sep 14 14:24:54 2020 New Revision: 365719 URL: https://svnweb.freebsd.org/changeset/base/365719 Log: Poll statistics more frequently in mlx5en(4). This makes traffic steering algorithms more accurate. MFC after:1 week Submitted by: gallatin @ Sponsored by: Mellanox Technologies // NVIDIA Networking Modified: head/sys/dev/mlx5/mlx5_en/mlx5_en_main.c Modified: head/sys/dev/mlx5/mlx5_en/mlx5_en_main.c == --- head/sys/dev/mlx5/mlx5_en/mlx5_en_main.cMon Sep 14 12:22:19 2020 (r365718) +++ head/sys/dev/mlx5/mlx5_en/mlx5_en_main.cMon Sep 14 14:24:54 2020 (r365719) @@ -1073,7 +1073,7 @@ mlx5e_update_stats(void *arg) queue_work(priv->wq, >update_stats_work); - callout_reset(>watchdog, hz, _update_stats, priv); + callout_reset(>watchdog, hz / 4, _update_stats, priv); } static void ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r365713 - in head: lib/libbe/tests sbin/bectl/tests
Author: trasz
Date: Mon Sep 14 10:35:39 2020
New Revision: 365713
URL: https://svnweb.freebsd.org/changeset/base/365713
Log:
Skip armv7 bectl tests on CI; they deadlock (""KSTACK_PAGES is 2").
PR: 249229
Sponsored by: DARPA
Modified:
head/lib/libbe/tests/be_create.sh
head/sbin/bectl/tests/bectl_test.sh
Modified: head/lib/libbe/tests/be_create.sh
==
--- head/lib/libbe/tests/be_create.sh Mon Sep 14 10:17:07 2020
(r365712)
+++ head/lib/libbe/tests/be_create.sh Mon Sep 14 10:35:39 2020
(r365713)
@@ -93,6 +93,11 @@ libbe_create_body()
atf_skip "https://bugs.freebsd.org/249055;
fi
+ if [ "$(atf_config_get ci false)" = "true" ] && \
+ [ "$(uname -p)" = "armv7" ]; then
+ atf_skip "https://bugs.freebsd.org/249229;
+ fi
+
cwd=$(atf_get_srcdir)
zpool=$(make_zpool_name)
disk=${cwd}/disk.img
Modified: head/sbin/bectl/tests/bectl_test.sh
==
--- head/sbin/bectl/tests/bectl_test.sh Mon Sep 14 10:17:07 2020
(r365712)
+++ head/sbin/bectl/tests/bectl_test.sh Mon Sep 14 10:35:39 2020
(r365713)
@@ -98,6 +98,11 @@ bectl_create_body()
atf_skip "https://bugs.freebsd.org/249055;
fi
+ if [ "$(atf_config_get ci false)" = "true" ] && \
+ [ "$(uname -p)" = "armv7" ]; then
+ atf_skip "https://bugs.freebsd.org/249229;
+ fi
+
cwd=$(realpath .)
zpool=$(make_zpool_name)
disk=${cwd}/disk.img
@@ -153,6 +158,11 @@ bectl_destroy_body()
atf_skip "https://bugs.freebsd.org/249055;
fi
+ if [ "$(atf_config_get ci false)" = "true" ] && \
+ [ "$(uname -p)" = "armv7" ]; then
+ atf_skip "https://bugs.freebsd.org/249229;
+ fi
+
cwd=$(realpath .)
zpool=$(make_zpool_name)
disk=${cwd}/disk.img
@@ -238,6 +248,11 @@ bectl_export_import_body()
atf_skip "https://bugs.freebsd.org/249055;
fi
+ if [ "$(atf_config_get ci false)" = "true" ] && \
+ [ "$(uname -p)" = "armv7" ]; then
+ atf_skip "https://bugs.freebsd.org/249229;
+ fi
+
cwd=$(realpath .)
zpool=$(make_zpool_name)
disk=${cwd}/disk.img
@@ -271,6 +286,11 @@ bectl_list_body()
atf_skip "https://bugs.freebsd.org/249055;
fi
+ if [ "$(atf_config_get ci false)" = "true" ] && \
+ [ "$(uname -p)" = "armv7" ]; then
+ atf_skip "https://bugs.freebsd.org/249229;
+ fi
+
cwd=$(realpath .)
zpool=$(make_zpool_name)
disk=${cwd}/disk.img
@@ -311,6 +331,11 @@ bectl_mount_body()
atf_skip "https://bugs.freebsd.org/249055;
fi
+ if [ "$(atf_config_get ci false)" = "true" ] && \
+ [ "$(uname -p)" = "armv7" ]; then
+ atf_skip "https://bugs.freebsd.org/249229;
+ fi
+
cwd=$(realpath .)
zpool=$(make_zpool_name)
disk=${cwd}/disk.img
@@ -350,6 +375,11 @@ bectl_rename_body()
atf_skip "https://bugs.freebsd.org/249055;
fi
+ if [ "$(atf_config_get ci false)" = "true" ] && \
+ [ "$(uname -p)" = "armv7" ]; then
+ atf_skip "https://bugs.freebsd.org/249229;
+ fi
+
cwd=$(realpath .)
zpool=$(make_zpool_name)
disk=${cwd}/disk.img
@@ -382,6 +412,11 @@ bectl_jail_body()
atf_skip "https://bugs.freebsd.org/249055;
fi
+ if [ "$(atf_config_get ci false)" = "true" ] && \
+ [ "$(uname -p)" = "armv7" ]; then
+ atf_skip "https://bugs.freebsd.org/249229;
+ fi
+
cwd=$(realpath .)
zpool=$(make_zpool_name)
disk=${cwd}/disk.img
@@ -455,6 +490,11 @@ bectl_jail_cleanup()
if [ "$(atf_config_get ci false)" = "true" ] && \
[ "$(uname -p)" = "i386" ]; then
atf_skip "https://bugs.freebsd.org/249055;
+ fi
+
+ if [ "$(atf_config_get ci false)" = "true" ] && \
+ [ "$(uname -p)" = "armv7" ]; then
+ atf_skip "https://bugs.freebsd.org/249229;
fi
zpool=$(get_zpool_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: r365712 - head/sys/kern
Author: trasz
Date: Mon Sep 14 10:17:07 2020
New Revision: 365712
URL: https://svnweb.freebsd.org/changeset/base/365712
Log:
Move td_softdep_cleanup() from userret() to ast(); it's infrequent
at best. The schedule_cleanup() function already sets TDF_ASTPENDING.
Reviewed by: kib, mckusick
Tested by:pho
MFC after:2 weeks
Sponsored by: DARPA
Differential Revision:https://reviews.freebsd.org/D26375
Modified:
head/sys/kern/subr_trap.c
Modified: head/sys/kern/subr_trap.c
==
--- head/sys/kern/subr_trap.c Mon Sep 14 10:14:03 2020(r365711)
+++ head/sys/kern/subr_trap.c Mon Sep 14 10:17:07 2020(r365712)
@@ -134,9 +134,6 @@ userret(struct thread *td, struct trapframe *frame)
KTRUSERRET(td);
#endif
- td_softdep_cleanup(td);
- MPASS(td->td_su == NULL);
-
/*
* Charge system time if profiling.
*/
@@ -188,8 +185,6 @@ userret(struct thread *td, struct trapframe *frame)
("userret: Returning with preallocated vnode"));
KASSERT((td->td_flags & (TDF_SBDRY | TDF_SEINTR | TDF_SERESTART)) == 0,
("userret: Returning with stop signals deferred"));
- KASSERT(td->td_su == NULL,
- ("userret: Returning with SU cleanup request not handled"));
KASSERT(td->td_vslock_sz == 0,
("userret: Returning with vslock-wired space"));
#ifdef VIMAGE
@@ -278,6 +273,9 @@ ast(struct trapframe *framep)
ktrcsw(0, 1, __func__);
#endif
}
+
+ td_softdep_cleanup(td);
+ MPASS(td->td_su == NULL);
/*
* If this thread tickled GEOM, we need to wait for the giggling to
___
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: r365711 - in head/sys: geom kern
Author: trasz
Date: Mon Sep 14 10:14:03 2020
New Revision: 365711
URL: https://svnweb.freebsd.org/changeset/base/365711
Log:
Move TDP_GEOM check from userret() to ast(); this code path is quite
infrequent.
Reviewed by: kib
No objections:mav
Tested by:pho
MFC after:2 weeks
Sponsored by: DARPA
Differential Revision:https://reviews.freebsd.org/D26374
Modified:
head/sys/geom/geom_event.c
head/sys/kern/subr_trap.c
Modified: head/sys/geom/geom_event.c
==
--- head/sys/geom/geom_event.c Mon Sep 14 09:44:24 2020(r365710)
+++ head/sys/geom/geom_event.c Mon Sep 14 10:14:03 2020(r365711)
@@ -374,6 +374,9 @@ g_post_event_x(g_event_t *func, void *arg, int flag, i
if (epp != NULL)
*epp = ep;
curthread->td_pflags |= TDP_GEOM;
+ thread_lock(curthread);
+ curthread->td_flags |= TDF_ASTPENDING;
+ thread_unlock(curthread);
return (0);
}
Modified: head/sys/kern/subr_trap.c
==
--- head/sys/kern/subr_trap.c Mon Sep 14 09:44:24 2020(r365710)
+++ head/sys/kern/subr_trap.c Mon Sep 14 10:14:03 2020(r365711)
@@ -138,13 +138,6 @@ userret(struct thread *td, struct trapframe *frame)
MPASS(td->td_su == NULL);
/*
-* If this thread tickled GEOM, we need to wait for the giggling to
-* stop before we return to userland
-*/
- if (__predict_false(td->td_pflags & TDP_GEOM))
- g_waitidle();
-
- /*
* Charge system time if profiling.
*/
if (__predict_false(p->p_flag & P_PROFIL))
@@ -285,6 +278,13 @@ ast(struct trapframe *framep)
ktrcsw(0, 1, __func__);
#endif
}
+
+ /*
+* If this thread tickled GEOM, we need to wait for the giggling to
+* stop before we return to userland
+*/
+ if (__predict_false(td->td_pflags & TDP_GEOM))
+ g_waitidle();
#ifdef DIAGNOSTIC
if (p->p_numthreads == 1 && (flags & TDF_NEEDSIGCHK) == 0) {
___
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: r365710 - head/sys/kern
Author: trasz
Date: Mon Sep 14 09:44:24 2020
New Revision: 365710
URL: https://svnweb.freebsd.org/changeset/base/365710
Log:
Move racct/rctl throttling from userret() to ast(). There's no reason
for it to sit in the syscall fast path.
Reviewed by: kib
MFC after:2 weeks
Sponsored by: DARPA
Differential Revision:https://reviews.freebsd.org/D26368
Modified:
head/sys/kern/kern_racct.c
head/sys/kern/subr_trap.c
Modified: head/sys/kern/kern_racct.c
==
--- head/sys/kern/kern_racct.c Mon Sep 14 08:59:16 2020(r365709)
+++ head/sys/kern/kern_racct.c Mon Sep 14 09:44:24 2020(r365710)
@@ -1146,6 +1146,8 @@ racct_proc_throttle(struct proc *p, int timeout)
FOREACH_THREAD_IN_PROC(p, td) {
thread_lock(td);
+ td->td_flags |= TDF_ASTPENDING;
+
switch (td->td_state) {
case TDS_RUNQ:
/*
Modified: head/sys/kern/subr_trap.c
==
--- head/sys/kern/subr_trap.c Mon Sep 14 08:59:16 2020(r365709)
+++ head/sys/kern/subr_trap.c Mon Sep 14 09:44:24 2020(r365710)
@@ -206,10 +206,6 @@ userret(struct thread *td, struct trapframe *frame)
__func__, td, p->p_pid, td->td_name, curvnet,
(td->td_vnet_lpush != NULL) ? td->td_vnet_lpush : "N/A"));
#endif
-#ifdef RACCT
- if (__predict_false(racct_enable && p->p_throttled != 0))
- racct_proc_throttled(p);
-#endif
}
/*
@@ -361,6 +357,11 @@ ast(struct trapframe *framep)
td->td_pflags &= ~TDP_OLDMASK;
kern_sigprocmask(td, SIG_SETMASK, >td_oldsigmask, NULL, 0);
}
+
+#ifdef RACCT
+ if (__predict_false(racct_enable && p->p_throttled != 0))
+ racct_proc_throttled(p);
+#endif
userret(td, framep);
}
___
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: r365709 - head/sys/dev/gpio
Author: andrew
Date: Mon Sep 14 08:59:16 2020
New Revision: 365709
URL: https://svnweb.freebsd.org/changeset/base/365709
Log:
Allow for interrupts on pl061 children
Add enough infrastructure for interrupts on children of the pl061 GPIO
controller. As gpiobus already provided these the pl061 driver also needs
to pass requests up the newbus hierarchy.
Currently there are no children that expect to configure interrupts, however
this is expected to change to support the ACPI Event Information interface.
Sponsored by: Innovate UK
Modified:
head/sys/dev/gpio/pl061.c
Modified: head/sys/dev/gpio/pl061.c
==
--- head/sys/dev/gpio/pl061.c Mon Sep 14 08:51:18 2020(r365708)
+++ head/sys/dev/gpio/pl061.c Mon Sep 14 08:59:16 2020(r365709)
@@ -553,6 +553,11 @@ static device_method_t pl061_methods[] = {
DEVMETHOD(device_attach,pl061_attach),
DEVMETHOD(device_detach,pl061_detach),
+ /* Bus interface */
+ DEVMETHOD(bus_setup_intr, bus_generic_setup_intr),
+ DEVMETHOD(bus_activate_resource,bus_generic_activate_resource),
+ DEVMETHOD(bus_deactivate_resource,
bus_generic_deactivate_resource),
+
/* GPIO protocol */
DEVMETHOD(gpio_get_bus, pl061_get_bus),
DEVMETHOD(gpio_pin_max, pl061_pin_max),
___
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: r365708 - head/sbin/pfctl/tests
Author: arichardson
Date: Mon Sep 14 08:51:18 2020
New Revision: 365708
URL: https://svnweb.freebsd.org/changeset/base/365708
Log:
pfctl_test: avoid 200 calls to atf_get_srcdir
I have been trying to reduce the time that testsuite runs take for CheriBSD
on QEMU (currently about 22 hours). One of the slowest tests is pfctl_test:
Just listing the available test cases currently takes 98 seconds on a
CheriBSD RISC-V system due to all the processes being spawned. This trivial
patch reduces the time to 92 seconds. The better solution would be to
rewrite the test in C/C++ which I may do as a follow-up change.
Reviewed By: kp
Differential Revision: https://reviews.freebsd.org/D26417
Modified:
head/sbin/pfctl/tests/pfctl_test.sh
Modified: head/sbin/pfctl/tests/pfctl_test.sh
==
--- head/sbin/pfctl/tests/pfctl_test.sh Mon Sep 14 06:26:36 2020
(r365707)
+++ head/sbin/pfctl/tests/pfctl_test.sh Mon Sep 14 08:51:18 2020
(r365708)
@@ -24,21 +24,22 @@ pftests="0001 0002 0003 0004 0005 0006 0007 0008 0009
0079 0081 0082 0084 0085 0087 0088 0089 0090 0091 0092 0094 0095 0096
0097 0098 0100 0101 0102 0104 1001 1002 1003 1004 1005"
-. $(atf_get_srcdir)/files/pfctl_test_descr.sh
+atf_srcdir=$(atf_get_srcdir)
+. "${atf_srcdir}/files/pfctl_test_descr.sh"
for i in ${pftests} ; do
atf_test_case "pf${i}"
eval "pf${i}_head () { atf_set descr \"$(pf${i}_descr)\" ; }"
eval "pf${i}_body () { \
kldstat -m pf || atf_skip \"pf(4) is not loaded\" && \
- cd $(atf_get_srcdir)/files && \
+ cd ${atf_srcdir}/files && \
atf_check -o file:pf${i}.ok \
pfctl -o none -nvf - < pf${i}.in ; }"
atf_test_case "selfpf${i}"
eval "selfpf${i}_head () { atf_set descr \"self$(pf${i}_descr)\" ; }"
eval "selfpf${i}_body () { \
- cd $(atf_get_srcdir)/files && \
+ cd ${atf_srcdir}/files && \
atf_check -o file:pf${i}.ok \
pfctl -o none -nvf - < pf${i}.ok ; }"
done
___
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: r365707 - head/share/man/man9
Author: scottl Date: Mon Sep 14 06:26:36 2020 New Revision: 365707 URL: https://svnweb.freebsd.org/changeset/base/365707 Log: Update bus_dma.9 for the expanded template API. Add some clarifying language about the operation of parent tags in templates. Modified: head/share/man/man9/bus_dma.9 Modified: head/share/man/man9/bus_dma.9 == --- head/share/man/man9/bus_dma.9 Mon Sep 14 05:58:12 2020 (r365706) +++ head/share/man/man9/bus_dma.9 Mon Sep 14 06:26:36 2020 (r365707) @@ -63,6 +63,8 @@ .Nm bus_dma_template_init , .Nm bus_dma_template_tag , .Nm bus_dma_template_clone , +.Nm bus_dma_template_fill, +.Nm BUS_DMA_TEMPLATE_FILL, .Nm bus_dmamap_create , .Nm bus_dmamap_destroy , .Nm bus_dmamap_load , @@ -90,19 +92,29 @@ .Fn bus_dma_tag_destroy "bus_dma_tag_t dmat" .Ft void .Fo bus_dma_template_init -.Fa "bus_dma_template_t template" +.Fa "bus_dma_template_t *template" .Fa "bus_dma_tag_t parent" .Fc .Ft int .Fo bus_dma_template_tag -.Fa "bus_dma_template_t template" +.Fa "bus_dma_template_t *template" .Fa "bus_dma_tag_t *dmat" .Fc .Ft void .Fo bus_dma_template_clone -.Fa "bus_dma_template_t template" +.Fa "bus_dma_template_t *template" .Fa "bus_dma_tag_t dmat" .Fc +.Ft void +.Fo bus_dma_template_fill +.Fa "bus_dma_template_t *template" +.Fa "bus_dma_param_t params[]" +.Fa "u_int count" +.Fc +.Fo BUS_DMA_TEMPLATE_FILL +.Fa "bus_dma_template_t *template" +.Fa "bus_dma_param_t param ..." +.Fc .Ft int .Fn bus_dmamap_create "bus_dma_tag_t dmat" "int flags" "bus_dmamap_t *mapp" .Ft int @@ -697,8 +709,12 @@ Initializes a .Fa bus_dma_template_t structure. If the .Fa parent -argument is non-NULL, values from this tag will be copied into the template, -replacing any defaults. +argument is non-NULL, this parent tag is associated with the template and +will be compiled into the dma tag that is later created. The values +of the parent are not copied into the template. During tag creation in +.Fn bus_dma_tag_template , +any parameters from the parent tag that are more restrictive than what is +in the provided template will overwrite what goes into the new tag. .It Fn bus_dma_template_tag "*template" "*dmat" Unpacks a template into a tag, and returns the tag via the .Fa dmat . @@ -712,6 +728,31 @@ The template does not need to be initialized first. A be overwritten by the values contained in the tag. When paired with .Fn bus_dma_template_tag , this function is useful for creating copies of tags. +.It Fn bus_dma_template_fill "*template" "params[]" "count" +Fills in the selected fields of the template with the keyed values from the +.Fa params +array. This is not meant to be called directly, use +.Fn BUS_DMA_TEMPLATE_FILL +instead. +.It Fn BUS_DMA_TEMPLATE_FILL "*template" "param ..." +Fills in the selected fields of the template with a variable number of +key-value parameters. The macros listed below take an argument of the +specified type and encapsulate it into a key-value structure that is directly +usable as a parameter argument. Muliple parameters may be provided at once. +.Bd -literal + BD_PARENT() void * + BD_ALIGNMENT() uintmax_t + BD_BOUNDARY() uintmax_t + BD_LOWADDR()vm_paddr_t + BD_HIGHADDR() vm_paddr_t + BD_MAXSIZE()uintmax_t + BD_NSEGMENTS() uintmax_t + BD_MAXSEGSIZE() uintmax_t + BD_FLAGS() uintmax_t + BD_LOCKFUNC() void * + BD_LOCKFUNCARG() void * +.Ed +.Pp .It Fn bus_dmamap_create "dmat" "flags" "*mapp" Allocates and initializes a DMA map. Arguments are as follows: ___ 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"
