diff options
190 files changed, 3360 insertions, 1600 deletions
diff --git a/Makefile.inc1 b/Makefile.inc1 index e341599ac505..81af336ecfa2 100644 --- a/Makefile.inc1 +++ b/Makefile.inc1 @@ -471,7 +471,7 @@ SUBDIR+= ${_DIR} # by calling 'makedb' in share/man. This is only relevant for # install/distribute so they build the whatis file after every manpage is # installed. -.if make(installworld) || make(install) +.if make(installworld) || make(install) || make(distributeworld) || make(distribute) SUBDIR+=.WAIT .endif SUBDIR+=etc @@ -27,6 +27,14 @@ NOTE TO PEOPLE WHO THINK THAT FreeBSD 16.x IS SLOW: world, or to merely disable the most expensive debugging functionality at runtime, run "ln -s 'abort:false,junk:false' /etc/malloc.conf".) +20251105: + pf(4) now supports nat64 via the af-to keyword. + +20251102: + Commit e5aa60d06958 changed the internal KAPI between + the NFS modules. As such, they all need to be rebuilt + from sources. __FreeBSD_version was bumped to 1600004 for this. + 20251025: flua(1) has moved to the new FreeBSD-flua package. If you use flua, you may want to install this package if it's not otherwise installed diff --git a/bin/date/date.1 b/bin/date/date.1 index b86a660a924d..f68892bd408d 100644 --- a/bin/date/date.1 +++ b/bin/date/date.1 @@ -29,7 +29,7 @@ .\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF .\" SUCH DAMAGE. .\" -.Dd September 1, 2025 +.Dd November 5, 2025 .Dt DATE 1 .Os .Sh NAME @@ -143,7 +143,8 @@ values are .Cm minutes , .Cm seconds , and -.Cm ns No Pq for nanoseconds . +.Cm ns +.Pq for nanoseconds . The date and time is formatted to the specified precision. When .Ar FMT @@ -172,7 +173,7 @@ Obsolete flag, accepted and ignored for compatibility. .It Fl R Use RFC 2822 date and time output format. This is equivalent to using -.Dq Li %a, %d %b %Y \&%T %z +.Ql %a, %d %b %Y \&%T %z as .Ar output_fmt while @@ -194,9 +195,7 @@ and can be specified in decimal, octal, or hex. Print the date and time of the last modification of .Ar filename . .It Fl u -Display or set the date in -.Tn UTC -(Coordinated Universal) time. +Display or set the date in UTC (Coordinated Universal) time. By default .Nm displays the time in the time zone described by @@ -328,7 +327,7 @@ The format string may contain any of the conversion specifications described in the .Xr strftime 3 manual page and -.Ql %N +.Ql \&%N for nanoseconds, as well as any arbitrary text. A newline .Pq Ql \en @@ -468,7 +467,7 @@ will display: .Dl "Sun Jan 4 04:15:24 GMT 1998" .Pp where it is currently -.Li "Mon Aug 4 04:15:24 BST 1997" . +.Ql "Mon Aug 4 04:15:24 BST 1997" . .Pp The command: .Pp @@ -493,29 +492,31 @@ will display the last Friday of the month: .Dl "Fri Aug 29 04:31:11 BST 1997" .Pp where it is currently -.Li "Mon Aug 4 04:31:11 BST 1997" . +.Ql "Mon Aug 4 04:31:11 BST 1997" . .Pp The command: .Pp .Dl "date 8506131627" .Pp sets the date to -.Dq Li "June 13, 1985, 4:27 PM" . +.Ql "June 13, 1985, 4:27 PM" . .Pp .Dl "date ""+%Y%m%d%H%M.%S""" .Pp may be used on one machine to print out the date suitable for setting on another. -.Qq ( Li "+%m%d%H%M%Y.%S" -for use on -.Tn Linux . ) +.Po Use +.Ql "+%m%d%H%M%Y.%S" +with GNU date on +Linux . +.Pc .Pp The command: .Pp .Dl "date 1432" .Pp sets the time to -.Li "2:32 PM" , +.Ql "2:32 PM" , without modifying the date. .Pp The command @@ -591,10 +592,10 @@ flag is compatible with .St -iso8601 . .Pp The -.Ql %N +.Ql \&%N conversion specification for nanoseconds is a non-standard extension. It is compatible with GNU date's -.Ql %N . +.Ql \&%N . .Sh HISTORY A .Nm @@ -615,6 +616,6 @@ flag was added in .Fx 12.0 . .Pp The -.Ql %N +.Ql \&%N conversion specification was added in .Fx 14.1 . diff --git a/cddl/contrib/opensolaris/cmd/dtrace/dtrace.1 b/cddl/contrib/opensolaris/cmd/dtrace/dtrace.1 index e263b936700d..456a9e319987 100644 --- a/cddl/contrib/opensolaris/cmd/dtrace/dtrace.1 +++ b/cddl/contrib/opensolaris/cmd/dtrace/dtrace.1 @@ -20,7 +20,7 @@ .\" .\" $FreeBSD$ .\" -.Dd July 30, 2025 +.Dd November 4, 2025 .Dt DTRACE 1 .Os .Sh NAME @@ -1292,6 +1292,7 @@ in .Xr cpp 1 , .Xr dwatch 1 , .Xr dtrace_audit 4 , +.Xr dtrace_callout_execute 4 , .Xr dtrace_dtrace 4 , .Xr dtrace_fbt 4 , .Xr dtrace_io 4 , @@ -1305,6 +1306,7 @@ in .Xr dtrace_tcp 4 , .Xr dtrace_udp 4 , .Xr dtrace_udplite 4 , +.Xr dtrace_vfs 4 , .Xr elf 5 , .Xr d 7 , .Xr tracing 7 , diff --git a/cddl/lib/libdtrace/io.d b/cddl/lib/libdtrace/io.d index d576f57476ce..484e6416bac7 100644 --- a/cddl/lib/libdtrace/io.d +++ b/cddl/lib/libdtrace/io.d @@ -73,7 +73,7 @@ translator bufinfo_t < struct bio *B > { b_lblkno = 0; b_resid = B->bio_resid; b_bufsize = 0; /* XXX gnn */ - b_error = B->bio_error; + b_error = B->bio_exterr.error; }; /* diff --git a/contrib/bsddialog/CHANGELOG b/contrib/bsddialog/CHANGELOG index 7800098644d7..503f70c73daa 100644 --- a/contrib/bsddialog/CHANGELOG +++ b/contrib/bsddialog/CHANGELOG @@ -1,3 +1,13 @@ +2025-08-15 Version 1.1 + + Library: + * add: bsddialog_slider(). + Thanks to https://gitlab.com/alfix/bsddialog/-/merge_requests/7 + + Utility: + * add: --slider dialog. + Thanks to https://gitlab.com/alfix/bsddialog/-/merge_requests/7 + 2025-06-22 Version 1.0.5 Manual: diff --git a/contrib/bsddialog/Makefile b/contrib/bsddialog/Makefile index 335b693470e6..7eb5938ff291 100644 --- a/contrib/bsddialog/Makefile +++ b/contrib/bsddialog/Makefile @@ -4,7 +4,7 @@ # Written in 2023 by Alfonso Sabato Siciliano OUTPUT = bsddialog -export VERSION=1.0.5 +export VERSION=1.1 .CURDIR ?= ${CURDIR} LIBPATH = ${.CURDIR}/lib LIBBSDDIALOG = ${LIBPATH}/libbsddialog.so diff --git a/contrib/bsddialog/README.md b/contrib/bsddialog/README.md index 5a25109775fe..f7552c303c09 100644 --- a/contrib/bsddialog/README.md +++ b/contrib/bsddialog/README.md @@ -1,4 +1,4 @@ -# BSDDialog 1.0.5 +# BSDDialog 1.1 This project provides **bsddialog** and **libbsddialog**, an utility and a library to build scripts and tools with TUI dialogs and widgets. @@ -31,7 +31,8 @@ Output: --calendar, --checklist, --datebox, --form, --gauge, --infobox, --inputbox, --menu, --mixedform, --mixedgauge, --msgbox, --passwordbox, --passwordform, ---pause, --radiolist, --rangebox, --textbox, --timebox, --treeview, --yesno. +--pause, --radiolist, --rangebox, --slider, --textbox, --timebox, --treeview, +--yesno. **Manual** @@ -69,6 +70,7 @@ in the _Public Domain_ to build new projects: % sh ./examples_utility/pause.sh % sh ./examples_utility/radiolist.sh % sh ./examples_utility/rangebox.sh +% sh ./examples_utility/slider.sh % sh ./examples_utility/timebox.sh % sh ./examples_utility/yesno.sh ``` @@ -106,6 +108,7 @@ in the _Public Domain_ to build new projects: % ./pause % ./radiolist % ./rangebox +% ./slider % ./theme % ./timebox % ./yesno diff --git a/contrib/bsddialog/examples_library/compile b/contrib/bsddialog/examples_library/compile index 1a68313090f6..b5e08f740784 100755 --- a/contrib/bsddialog/examples_library/compile +++ b/contrib/bsddialog/examples_library/compile @@ -12,7 +12,7 @@ set -x libpath=../lib examples="menu checklist radiolist mixedlist theme infobox yesno msgbox \ -datebox form timebox rangebox pause calendar gauge mixedgauge textbox" +datebox form timebox rangebox pause calendar gauge mixedgauge textbox slider" rm -f $examples diff --git a/contrib/bsddialog/examples_library/slider.c b/contrib/bsddialog/examples_library/slider.c new file mode 100644 index 000000000000..2292f0e8efed --- /dev/null +++ b/contrib/bsddialog/examples_library/slider.c @@ -0,0 +1,44 @@ +/*- + * SPDX-License-Identifier: CC0-1.0 + * + * Written in 2025 by Alfonso Sabato Siciliano. + * To the extent possible under law, the author has dedicated all copyright + * and related and neighboring rights to this software to the public domain + * worldwide. This software is distributed without any warranty, see: + * <http://creativecommons.org/publicdomain/zero/1.0/>. + */ + +#include <bsddialog.h> +#include <stdio.h> + +int main() +{ + int output; + unsigned long start, end, blocks[2][2]; + struct bsddialog_conf conf; + + start = 20; + end = 70; + blocks[0][0] = 5; + blocks[0][1] = 10; + blocks[1][0] = 80; + blocks[1][1] = 90; + + if (bsddialog_init() == BSDDIALOG_ERROR) { + printf("Error: %s\n", bsddialog_geterror()); + return (1); + } + bsddialog_initconf(&conf); + conf.title = "slider"; + + output = bsddialog_slider(&conf, "Example", 0, 0, "GiB", 100, &start, + &end, false, 2, blocks); + bsddialog_end(); + if (output == BSDDIALOG_ERROR) { + printf("Error: %s\n", bsddialog_geterror()); + return (1); + } + printf("Start: %lu, End: %lu\n", start, end); + + return (0); +} diff --git a/contrib/bsddialog/examples_utility/slider.sh b/contrib/bsddialog/examples_utility/slider.sh new file mode 100644 index 000000000000..bd037e13c977 --- /dev/null +++ b/contrib/bsddialog/examples_utility/slider.sh @@ -0,0 +1,34 @@ +#!/bin/sh +#- +# SPDX-License-Identifier: CC0-1.0 +# +# Written in 2025 by Braulio Rivas. +# +# To the extent possible under law, the author has dedicated all copyright +# and related and neighboring rights to this software to the public domain +# worldwide. This software is distributed without any warranty, see: +# <http://creativecommons.org/publicdomain/zero/1.0/>. + +: ${BSDDIALOG_ERROR=255} +: ${BSDDIALOG_OK=0} +: ${BSDDIALOG_CANCEL=1} +: ${BSDDIALOG_ESC=5} + +STARTEND=$(./bsddialog --slider "Choose a new partition location" 0 0 MiB \ + 30000 5000 6000 0 1000 3000 25000 30000 \ +3>&1 1>&2 2>&3 3>&-) + +case $? in + $BSDDIALOG_ERROR ) + exit 1 + ;; + $BSDDIALOG_ESC ) + echo "[ESC]" + ;; + $BSDDIALOG_CANCEL ) + echo "[Cancel]" + ;; + $BSDDIALOG_OK ) + echo "[OK] $STARTEND" + ;; +esac diff --git a/contrib/bsddialog/lib/Makefile b/contrib/bsddialog/lib/Makefile index c728541a9f7a..b5a3c1a8afc4 100644 --- a/contrib/bsddialog/lib/Makefile +++ b/contrib/bsddialog/lib/Makefile @@ -8,7 +8,7 @@ LIBRARY_SO = lib${LIBRARY:=.so} LIBRARY_A = lib${LIBRARY:=.a} HEADERS = bsddialog.h bsddialog_theme.h bsddialog_progressview.h SOURCES = barbox.c datebox.c formbox.c libbsddialog.c lib_util.c \ - menubox.c messagebox.c textbox.c theme.c timebox.c + menubox.c messagebox.c slider.c textbox.c theme.c timebox.c OBJECTS = ${SOURCES:.c=.o} PREFIX = /usr/local diff --git a/contrib/bsddialog/lib/bsddialog.h b/contrib/bsddialog/lib/bsddialog.h index fc59071c6fa0..4b387d993363 100644 --- a/contrib/bsddialog/lib/bsddialog.h +++ b/contrib/bsddialog/lib/bsddialog.h @@ -30,7 +30,7 @@ #include <stdbool.h> -#define LIBBSDDIALOG_VERSION "1.0.5" +#define LIBBSDDIALOG_VERSION "1.1" /* Return values */ #define BSDDIALOG_ERROR -1 @@ -245,6 +245,12 @@ bsddialog_rangebox(struct bsddialog_conf *conf, const char *text, int rows, int cols, int min, int max, int *value); int +bsddialog_slider(struct bsddialog_conf *conf, const char *text, int rows, + int cols, const char *unit, unsigned long length, unsigned long *start, + unsigned long *end, bool resize, unsigned int nblocks, + unsigned long blocks[][2]); + +int bsddialog_textbox(struct bsddialog_conf *conf, const char *file, int rows, int cols); diff --git a/contrib/bsddialog/lib/slider.c b/contrib/bsddialog/lib/slider.c new file mode 100644 index 000000000000..534253470849 --- /dev/null +++ b/contrib/bsddialog/lib/slider.c @@ -0,0 +1,670 @@ +/*- + * SPDX-License-Identifier: BSD-2-Clause + * + * Copyright (c) 2025 Braulio Rivas + * + * 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. + */ + +#include <curses.h> +#include <stdlib.h> +#include <string.h> + +#include "bsddialog.h" +#include "bsddialog_theme.h" +#include "lib_util.h" + +#define MINHSLIDER 13 +#define MINWSLIDER 36 + +#define NULLWIN -1 +#define START_WIN 0 +#define END_WIN 1 +#define STEP_WIN 2 +#define SLIDER_WIN 3 +#define NWIN 4 + +enum operation { + MOVERIGHT, + MOVEFARRIGHT, + MOVEFASTRIGHT, + MOVELEFT, + MOVEFARLEFT, + MOVEFASTLEFT, + INCREASELEFT, + DECREASELEFT, + INCREASERIGHT, + DECREASERIGHT, + INCREASESTEP, + DECREASESTEP, +}; + +struct sliderctl { + enum operation op; + unsigned long (*spaces)[2]; + int nspaces; /* api unsigned, but segfault handlesliderctl():MOVELEFT */ + unsigned long length; + unsigned long *start; + unsigned long *end; + unsigned long step; +}; + +static int crashes(long x, long y, long a, long b) +{ + return ((x <= a && a <= y) || (x <= b && b <= y)); +} + +static int fits(long x, long y, long a, long b) +{ + return ((x <= a) && (b <= y)); +} + +static void handlesliderctl(struct sliderctl *sliderctl) +{ + int i, step, tmpstep; + unsigned long x, y, size, old_start, old_end; + signed long new_start, new_end; + + step = sliderctl->step; + old_start = *(sliderctl->start); + new_start = old_start; + old_end = *(sliderctl->end); + new_end = old_end; + size = old_end - old_start + 1; + + switch (sliderctl->op) { + case MOVERIGHT: + new_start = old_start + step; + new_end = old_end + step; + + for (i = 0; i < sliderctl->nspaces; i++) { + x = (sliderctl->spaces)[i][0]; + y = (sliderctl->spaces)[i][1]; + + if (crashes(x, y, new_start, new_end)) { + new_start = y + 1; + new_end = new_start + size - 1; + break; + } + } + break; + case MOVELEFT: + new_start = old_start - step; + new_end = old_end - step; + + for (i = sliderctl->nspaces - 1; i >= 0; i--) { + x = (sliderctl->spaces)[i][0]; + y = (sliderctl->spaces)[i][1]; + + if (crashes(x, y, new_start, new_end)) { + new_end = x - 1; + new_start = new_end - size + 1; + break; + } + } + break; + case INCREASELEFT: + new_start = old_start + step; + break; + case DECREASELEFT: + new_start = old_start - step; + for (i = 0; i < sliderctl->nspaces; i++) { + x = (sliderctl->spaces)[i][0]; + y = (sliderctl->spaces)[i][1]; + + if (crashes(x, y, new_start, new_end)) { + new_start = old_start; + break; + } + } + break; + case INCREASERIGHT: + new_end = old_end + step; + for (i = 0; i < sliderctl->nspaces; i++) { + x = (sliderctl->spaces)[i][0]; + y = (sliderctl->spaces)[i][1]; + + if (crashes(x, y, new_start, new_end)) { + new_end = old_end; + break; + } + } + break; + case DECREASERIGHT: + new_end = old_end - step; + break; + case MOVEFARLEFT: + new_start = 0; + new_end = size - 1; + for (i = 0; i < sliderctl->nspaces; i++) { + x = (sliderctl->spaces)[i][0]; + y = (sliderctl->spaces)[i][1]; + + if (crashes(x, y, new_start, new_end)) { + new_start = y + 1; + new_end = new_start + size - 1; + break; + } + } + break; + case MOVEFARRIGHT: + new_end = (sliderctl->length) - 1; + new_start = new_end - size + 1; + for (i = sliderctl->nspaces - 1; i >= 0; i--) { + x = (sliderctl->spaces)[i][0]; + y = (sliderctl->spaces)[i][1]; + + if (crashes(x, y, new_start, new_end)) { + new_end = x - 1; + new_start = new_end - size + 1; + break; + } + } + break; + case MOVEFASTLEFT: + if (size < 10) { + tmpstep = 1; + } else { + tmpstep = ((sliderctl->length) * 10) / 100; + } + new_start = old_start - tmpstep; + new_end = old_end - tmpstep; + + for (i = sliderctl->nspaces - 1; i >= 0; i--) { + x = (sliderctl->spaces)[i][0]; + y = (sliderctl->spaces)[i][1]; + + if (crashes(x, y, new_start, new_end)) { + new_end = x - 1; + new_start = new_end - size + 1; + break; + } + } + break; + case MOVEFASTRIGHT: + if (size < 10) { + tmpstep = 1; + } else { + tmpstep = ((sliderctl->length) * 10) / 100; + } + new_start = old_start + tmpstep; + new_end = old_end + tmpstep; + + for (i = 0; i < sliderctl->nspaces; i++) { + x = (sliderctl->spaces)[i][0]; + y = (sliderctl->spaces)[i][1]; + + if (crashes(x, y, new_start, new_end)) { + new_start = y + 1; + new_end = new_start + size - 1; + break; + } + } + break; + case INCREASESTEP: + ++step; + break; + case DECREASESTEP: + if (step > 1) { + --step; + } + break; + } + + if (fits(0, (sliderctl->length) - 1, new_start, new_end) != 1) { + new_start = old_start; + new_end = old_end; + } + + if (new_start > new_end) { + new_start = old_start; + new_end = old_end; + } + + sliderctl->step = step; + + *(sliderctl->start) = new_start; + *(sliderctl->end) = new_end; +} + +static void +drawsquare(struct bsddialog_conf *conf, WINDOW *win, enum elevation elev, + bool focus, const char *fmt, unsigned long value) +{ + int h, l, w; + + getmaxyx(win, h, w); + draw_borders(conf, win, elev); + if (focus) { + l = 2 + w % 2; + wattron(win, t.dialog.arrowcolor); + mvwhline(win, 0, w / 2 - l / 2, UARROW(conf), l); + mvwhline(win, h - 1, w / 2 - l / 2, DARROW(conf), l); + wattroff(win, t.dialog.arrowcolor); + } + + if (focus) + wattron(win, t.menu.f_namecolor); + + mvwprintw(win, 1, 1, fmt, value); + + if (focus) + wattroff(win, t.menu.f_namecolor); + + wnoutrefresh(win); +} + +static void +print_slider(struct bsddialog_conf *conf, WINDOW *win, + unsigned long spaces[][2], int nspaces, unsigned long length, + unsigned long *start, unsigned long *end, bool active) +{ + int i, y, x, l, height, width; + unsigned long s, e; + chtype ch; + + getmaxyx(win, height, width); + wclear(win); + draw_borders(conf, win, RAISED); + + if (active) { + wattron(win, t.dialog.arrowcolor); + mvwvline(win, 1, 0, LARROW(conf), 1); + mvwvline(win, 1, width - 1, RARROW(conf), 1); + wattroff(win, t.dialog.arrowcolor); + } + + y = height / 2; + width -= 1; + + ch = ' ' | bsddialog_color(BSDDIALOG_RED, BSDDIALOG_RED, 0); + for (i = 0; i < nspaces; i++) { + s = spaces[i][0]; + e = spaces[i][1]; + + x = (s * width) / length; + l = ((e - s) * width) / length; + + if ((e - s) == 0) { + l = 0; + } else if (l == 0) { + l = 1; + } + + mvwhline(win, y, x + 1, ch, l); + } + + ch = ' ' | t.bar.f_color; + s = ((*start) * width) / length; + l = (((*end) - (*start)) * width) / length; + if ((*end - *start) == 0) { + l = 0; + } else if (l == 0) { + l = 1; + } + mvwhline(win, y, s + 1, ch, l); + + wnoutrefresh(win); +} + +static int +slider_draw(struct dialog *d, bool redraw, WINDOW *start_win, WINDOW *end_win, + WINDOW *size_win, WINDOW *step_win, WINDOW *slider_win, const char *unit) +{ + char *buf; + int yslider, xslider; + + if (redraw) { + hide_dialog(d); + refresh(); /* Important for decreasing screen */ + } + if (dialog_size_position(d, MINHSLIDER, MINWSLIDER, NULL) != 0) + return (BSDDIALOG_ERROR); + if (draw_dialog(d) != 0) /* doupdate in main loop */ + return (BSDDIALOG_ERROR); + if (redraw) + refresh(); /* Important to fix grey lines expanding screen */ + TEXTPAD(d, MINHSLIDER + HBUTTONS); + + yslider = d->y + d->h - 15; + xslider = d->x + d->w / 2 - 17; + asprintf(&buf, "Start (%s)", unit); + mvwaddstr(d->widget, d->h - 16, d->w / 2 - 17, buf); + free(buf); + update_box(d->conf, start_win, yslider, xslider, 3, 17, RAISED); + asprintf(&buf, "End (%s)", unit); + mvwaddstr(d->widget, d->h - 16, d->w / 2, buf); + free(buf); + update_box(d->conf, end_win, yslider, xslider + 17, 3, 17, RAISED); + asprintf(&buf, "Size (%s)", unit); + mvwaddstr(d->widget, d->h - 12, d->w / 2 - 17, buf); + free(buf); + update_box(d->conf, size_win, yslider + 4, xslider, 3, 17, RAISED); + asprintf(&buf, "Step (%s)", unit); + mvwaddstr(d->widget, d->h - 12, d->w / 2, buf); + free(buf); + update_box(d->conf, step_win, yslider + 4, xslider + 17, 3, 17, RAISED); + + update_box(d->conf, slider_win, yslider + 7, xslider, 3, 34, RAISED); + wnoutrefresh(d->widget); + + return (0); +} + +/* API */ +int +bsddialog_slider(struct bsddialog_conf *conf, const char *text, int rows, + int cols, const char *unit, unsigned long length, unsigned long *start, + unsigned long *end, bool resize, unsigned int nblocks, + unsigned long blocks[][2]) +{ + struct sliderctl ctl; + bool loop, focusbuttons; + int retval, sel; + wint_t input; + unsigned long size; + WINDOW *start_win, *end_win, *size_win, *step_win, *slider_win; + struct dialog dialog; + + CHECK_PTR(start); + CHECK_PTR(end); + + ctl.spaces = blocks; + ctl.nspaces = nblocks; + ctl.length = length; + ctl.start = start; + ctl.end = end; + ctl.step = 1; + + if (prepare_dialog(conf, text, rows, cols, &dialog) != 0) + return (BSDDIALOG_ERROR); + set_buttons(&dialog, true, OK_LABEL, CANCEL_LABEL); + + if ((start_win = newwin(1, 1, 1, 1)) == NULL) + RETURN_ERROR("Cannot build WINDOW for start"); + wbkgd(start_win, t.dialog.color); + + if ((end_win = newwin(1, 1, 1, 1)) == NULL) + RETURN_ERROR("Cannot build WINDOW for end"); + wbkgd(end_win, t.dialog.color); + + if ((step_win = newwin(1, 1, 1, 1)) == NULL) + RETURN_ERROR("Cannot build WINDOW for step"); + wbkgd(step_win, t.dialog.color); + + if ((size_win = newwin(1, 1, 1, 1)) == NULL) + RETURN_ERROR("Cannot build WINDOW for size"); + wbkgd(size_win, t.dialog.color); + + if ((slider_win = newwin(1, 1, 1, 1)) == NULL) + RETURN_ERROR("Cannot build WINDOW for slider"); + wbkgd(slider_win, t.dialog.color); + + if (slider_draw(&dialog, false, start_win, end_win, size_win, step_win, + slider_win, unit) != 0) + return (BSDDIALOG_ERROR); + + sel = NULLWIN; + loop = focusbuttons = true; + while (loop) { + size = *(ctl.end) - *(ctl.start) + 1; + drawsquare(conf, start_win, RAISED, sel == START_WIN, "%15lu", *start); + drawsquare(conf, end_win, RAISED, sel == END_WIN, "%15lu", *end); + drawsquare(conf, size_win, RAISED, 0, "%15lu", size); + drawsquare(conf, step_win, RAISED, sel == STEP_WIN, "%15d", ctl.step); + print_slider(conf, slider_win, blocks, nblocks, length, start, + end, sel == SLIDER_WIN); + doupdate(); + + if (get_wch(&input) == ERR) + continue; + switch (input) { + case KEY_ENTER: + case 10: /* Enter */ + if (focusbuttons || conf->button.always_active) { + retval = BUTTONVALUE(dialog.bs); + loop = false; + } + break; + case 27: /* Esc */ + if (conf->key.enable_esc) { + retval = BSDDIALOG_ESC; + loop = false; + } + break; + case '\t': /* TAB */ + if (focusbuttons) { + dialog.bs.curr++; + if (dialog.bs.curr >= (int)dialog.bs.nbuttons) { + focusbuttons = false; + sel = START_WIN; + dialog.bs.curr = + conf->button.always_active ? 0 : -1; + } + } else { + sel++; + if ((sel + 1) > NWIN) { + focusbuttons = true; + sel = NULLWIN; + dialog.bs.curr = 0; + } + } + DRAW_BUTTONS(dialog); + break; + case KEY_CTRL('n'): + case KEY_RIGHT: + if (focusbuttons) { + dialog.bs.curr++; + if (dialog.bs.curr >= (int)dialog.bs.nbuttons) { + focusbuttons = false; + sel = START_WIN; + dialog.bs.curr = + conf->button.always_active ? 0 : -1; + } + } else if (sel == SLIDER_WIN) { + ctl.op = MOVERIGHT; + handlesliderctl(&ctl); + } else { + sel++; + } + DRAW_BUTTONS(dialog); + break; + case KEY_CTRL('p'): + case KEY_LEFT: + if (focusbuttons) { + dialog.bs.curr--; + if (dialog.bs.curr < 0) { + focusbuttons = false; + sel = SLIDER_WIN; + dialog.bs.curr = + conf->button.always_active ? 0 : -1; + } + } else if (sel == SLIDER_WIN) { + ctl.op = MOVELEFT; + handlesliderctl(&ctl); + } else if (sel == END_WIN) { + sel = START_WIN; + } else { + focusbuttons = true; + sel = NULLWIN; + dialog.bs.curr = 0; + } + DRAW_BUTTONS(dialog); + break; + case KEY_UP: + if (focusbuttons) { + sel = SLIDER_WIN; + focusbuttons = false; + dialog.bs.curr = + conf->button.always_active ? 0 : -1; + DRAW_BUTTONS(dialog); + } else if (sel == START_WIN) { + if (resize) { + ctl.op = INCREASELEFT; + } else { + ctl.op = MOVERIGHT; + } + handlesliderctl(&ctl); + } else if (sel == END_WIN) { + if (resize) { + ctl.op = INCREASERIGHT; + } else { + ctl.op = MOVERIGHT; + } + handlesliderctl(&ctl); + } else if (sel == STEP_WIN) { + ctl.op = INCREASESTEP; + handlesliderctl(&ctl); + } + break; + case KEY_DOWN: + if (focusbuttons) { + break; + } else if (sel == START_WIN) { + if (resize) { + ctl.op = DECREASELEFT; + } else { + ctl.op = MOVELEFT; + } + handlesliderctl(&ctl); + } else if (sel == END_WIN) { + if (resize) { + ctl.op = DECREASERIGHT; + } else { + ctl.op = MOVELEFT; + } + handlesliderctl(&ctl); + } else if (sel == STEP_WIN) { + ctl.op = DECREASESTEP; + handlesliderctl(&ctl); + } + break; + case '-': + if (focusbuttons) { + break; + } else if (sel == START_WIN) { + if (resize) { + ctl.op = DECREASELEFT; + } else { + ctl.op = MOVELEFT; + } + handlesliderctl(&ctl); + } else if (sel == END_WIN) { + if (resize) { + ctl.op = DECREASERIGHT; + } else { + ctl.op = MOVELEFT; + } + handlesliderctl(&ctl); + } else if (sel == STEP_WIN) { + ctl.op = DECREASESTEP; + handlesliderctl(&ctl); + } + break; + case '+': + if (focusbuttons) { + break; + } else if (sel == START_WIN) { + if (resize) { + ctl.op = INCREASELEFT; + } else { + ctl.op = MOVERIGHT; + } + handlesliderctl(&ctl); + } else if (sel == END_WIN) { + if (resize) { + ctl.op = INCREASERIGHT; + } else { + ctl.op = MOVERIGHT; + } + handlesliderctl(&ctl); + } else if (sel == STEP_WIN) { + ctl.op = INCREASESTEP; + handlesliderctl(&ctl); + } + break; + case KEY_HOME: + if (focusbuttons) { + break; + } else if (sel == SLIDER_WIN) { + ctl.op = MOVEFARLEFT; + handlesliderctl(&ctl); + } + break; + case KEY_END: + if (focusbuttons) { + break; + } else if (sel == SLIDER_WIN) { + ctl.op = MOVEFARRIGHT; + handlesliderctl(&ctl); + } + break; + case KEY_PPAGE: + if (focusbuttons) { + break; + } else if (sel == SLIDER_WIN) { + ctl.op = MOVEFASTLEFT; + handlesliderctl(&ctl); + } + break; + case KEY_NPAGE: + if (focusbuttons) { + break; + } else if (sel == SLIDER_WIN) { + ctl.op = MOVEFASTRIGHT; + handlesliderctl(&ctl); + } + break; + case KEY_F(1): + if (conf->key.f1_file == NULL && + conf->key.f1_message == NULL) + break; + if (f1help_dialog(conf) != 0) + return (BSDDIALOG_ERROR); + if (slider_draw(&dialog, true, start_win, end_win, size_win, + step_win, slider_win, unit) != 0) + return (BSDDIALOG_ERROR); + break; + case KEY_CTRL('l'): + case KEY_RESIZE: + if (slider_draw(&dialog, true, start_win, end_win, size_win, + step_win, slider_win, unit) != 0) + return (BSDDIALOG_ERROR); + break; + default: + if (shortcut_buttons(input, &dialog.bs)) { + DRAW_BUTTONS(dialog); + doupdate(); + retval = BUTTONVALUE(dialog.bs); + loop = false; + } + } + } + + delwin(start_win); + delwin(end_win); + delwin(step_win); + delwin(slider_win); + end_dialog(&dialog); + + return (retval); +} diff --git a/contrib/bsddialog/utility/bsddialog.1 b/contrib/bsddialog/utility/bsddialog.1 index 0ec2a96952bd..d03a4ce8881a 100644 --- a/contrib/bsddialog/utility/bsddialog.1 +++ b/contrib/bsddialog/utility/bsddialog.1 @@ -22,7 +22,7 @@ .\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF .\" SUCH DAMAGE. .\" -.Dd June 22, 2025 +.Dd June 24, 2025 .Dt BSDDIALOG 1 .Os .Sh NAME @@ -851,6 +851,10 @@ utility first appeared in was written by .An Alfonso Sabato Siciliano .Aq Mt asiciliano@FreeBSD.org . +.An Braulio Rivas +.Aq Mt brauliorivas@FreeBSD.org +implemented the slider dialog for the +.Dq GSoC 2025 Full Disk Administration Tool for FreeBSD . .Pp .Nm bsddialog provides also a subset of the functionality described in the diff --git a/contrib/bsddialog/utility/util.h b/contrib/bsddialog/utility/util.h index d1f7793c9755..5e5cba3f82f7 100644 --- a/contrib/bsddialog/utility/util.h +++ b/contrib/bsddialog/utility/util.h @@ -108,6 +108,7 @@ int passwordform_builder(BUILDER_ARGS); int pause_builder(BUILDER_ARGS); int radiolist_builder(BUILDER_ARGS); int rangebox_builder(BUILDER_ARGS); +int slider_builder(BUILDER_ARGS); int textbox_builder(BUILDER_ARGS); int timebox_builder(BUILDER_ARGS); int treeview_builder(BUILDER_ARGS); diff --git a/contrib/bsddialog/utility/util_builders.c b/contrib/bsddialog/utility/util_builders.c index 0a968d4319f9..9784b1a5e563 100644 --- a/contrib/bsddialog/utility/util_builders.c +++ b/contrib/bsddialog/utility/util_builders.c @@ -608,12 +608,12 @@ int form_builder(BUILDER_ARGS) exit_error(false, "cannot allocate memory for form items"); j = 0; for (i = 0; i < nitems; i++) { - items[i].label = argv[j++]; + items[i].label = argv[j++]; items[i].ylabel = (unsigned int)strtoul(argv[j++], NULL, 10); items[i].xlabel = (unsigned int)strtoul(argv[j++], NULL, 10); - items[i].init = argv[j++]; - items[i].yfield = (unsigned int)strtoul(argv[j++], NULL, 10); - items[i].xfield = (unsigned int)strtoul(argv[j++], NULL, 10); + items[i].init = argv[j++]; + items[i].yfield = (unsigned int)strtoul(argv[j++], NULL, 10); + items[i].xfield = (unsigned int)strtoul(argv[j++], NULL, 10); fieldlen = (int)strtol(argv[j++], NULL, 10); if (fieldlen == 0) @@ -651,15 +651,15 @@ int inputbox_builder(BUILDER_ARGS) if (argc > 1) error_args("--inputbox", argc - 1, argv + 1); - item.label = ""; - item.ylabel = 0; - item.xlabel = 0; - item.init = argc > 0 ? argv[0] : ""; - item.yfield = 0; - item.xfield = 0; + item.label = ""; + item.ylabel = 0; + item.xlabel = 0; + item.init = argc > 0 ? argv[0] : ""; + item.yfield = 0; + item.xfield = 0; item.fieldlen = 1; item.maxvaluelen = opt->max_input_form; - item.flags = BSDDIALOG_FIELDNOCOLOR; + item.flags = BSDDIALOG_FIELDNOCOLOR; item.flags |= BSDDIALOG_FIELDCURSOREND; item.flags |= BSDDIALOG_FIELDEXTEND; item.bottomdesc = ""; @@ -691,12 +691,12 @@ int mixedform_builder(BUILDER_ARGS) exit_error(false, "cannot allocate memory for form items"); j = 0; for (i = 0; i < nitems; i++) { - items[i].label = argv[j++]; + items[i].label = argv[j++]; items[i].ylabel = (unsigned int)strtoul(argv[j++], NULL, 10); items[i].xlabel = (unsigned int)strtoul(argv[j++], NULL, 10); - items[i].init = argv[j++]; - items[i].yfield = (unsigned int)strtoul(argv[j++], NULL, 10); - items[i].xfield = (unsigned int)strtoul(argv[j++], NULL, 10); + items[i].init = argv[j++]; + items[i].yfield = (unsigned int)strtoul(argv[j++], NULL, 10); + items[i].xfield = (unsigned int)strtoul(argv[j++], NULL, 10); fieldlen = (int)strtol(argv[j++], NULL, 10); if (fieldlen == 0) items[i].fieldlen = strcols(items[i].init); @@ -737,13 +737,13 @@ int passwordbox_builder(BUILDER_ARGS) if (argc > 1) error_args("--passwordbox", argc - 1, argv + 1); - item.label = ""; - item.ylabel = 0; - item.xlabel = 0; - item.init = argc > 0 ? argv[0] : ""; - item.yfield = 0; - item.xfield = 0; - item.fieldlen = 1; + item.label = ""; + item.ylabel = 0; + item.xlabel = 0; + item.init = argc > 0 ? argv[0] : ""; + item.yfield = 0; + item.xfield = 0; + item.fieldlen = 1; item.maxvaluelen = opt->max_input_form; item.flags = BSDDIALOG_FIELDHIDDEN; item.flags |= BSDDIALOG_FIELDNOCOLOR; @@ -779,12 +779,12 @@ int passwordform_builder(BUILDER_ARGS) exit_error(false, "cannot allocate memory for form items"); j = 0; for (i = 0; i < nitems; i++) { - items[i].label = argv[j++]; + items[i].label = argv[j++]; items[i].ylabel = (unsigned int)strtoul(argv[j++], NULL, 10); items[i].xlabel = (unsigned int)strtoul(argv[j++], NULL, 10); - items[i].init = argv[j++]; - items[i].yfield = (unsigned int)strtoul(argv[j++], NULL, 10); - items[i].xfield = (unsigned int)strtoul(argv[j++], NULL, 10); + items[i].init = argv[j++]; + items[i].yfield = (unsigned int)strtoul(argv[j++], NULL, 10); + items[i].xfield = (unsigned int)strtoul(argv[j++], NULL, 10); fieldlen = (int)strtol(argv[j++], NULL, 10); items[i].fieldlen = abs(fieldlen); @@ -809,3 +809,41 @@ int passwordform_builder(BUILDER_ARGS) return (output); } + +int slider_builder(BUILDER_ARGS) +{ + bool resize; + int output; + char *unit; + unsigned int i, nblocks; + unsigned long length, start, end, (*blocks)[2]; + + if (argc < 5) + exit_error(true, "--slider requires: <unit> <lenght> <start> <end> <resize>"); + unit = argv[0]; + length = strtoul(argv[1], NULL, 10); + start = strtoul(argv[2], NULL, 10); + end = strtoul(argv[3], NULL, 10); + resize = strtoul(argv[4], NULL, 10) == 0 ? false : true; + + argc -= 5; + argv += 5; + if (argc & 1) + exit_error(true, "bad [<start_block> <end_block> ...] number"); + nblocks = argc / 2; + if ((blocks = malloc(nblocks * sizeof(*blocks))) == NULL) + exit_error(false, "Cannot allocate memory for blocks"); + for (i = 0; i < nblocks; i++) { + blocks[i][0] = strtoul(argv[2 * i], NULL, 10); + blocks[i][1] = strtoul(argv[2 * i + 1], NULL, 10); + } + + output = bsddialog_slider(conf, text, rows, cols, unit, length, &start, &end, + resize, nblocks, blocks); + free(blocks); + + if (output != BSDDIALOG_ERROR) + dprintf(opt->output_fd, "%lu %lu", start, end); + + return (output); +} diff --git a/contrib/bsddialog/utility/util_cli.c b/contrib/bsddialog/utility/util_cli.c index 01b6fc31f065..7e76ec84654f 100644 --- a/contrib/bsddialog/utility/util_cli.c +++ b/contrib/bsddialog/utility/util_cli.c @@ -143,7 +143,8 @@ enum OPTS { TEXTBOX, TIMEBOX, TREEVIEW, - YESNO + YESNO, + SLIDER, }; /* options descriptor */ @@ -264,6 +265,7 @@ static struct option longopts[] = { {"pause", no_argument, NULL, PAUSE}, {"radiolist", no_argument, NULL, RADIOLIST}, {"rangebox", no_argument, NULL, RANGEBOX}, + {"slider", no_argument, NULL, SLIDER}, {"textbox", no_argument, NULL, TEXTBOX}, {"timebox", no_argument, NULL, TIMEBOX}, {"treeview", no_argument, NULL, TREEVIEW}, @@ -801,6 +803,13 @@ parseargs(int argc, char **argv, struct bsddialog_conf *conf, opt->name = "--rangebox"; opt->dialogbuilder = rangebox_builder; break; + case SLIDER: + if (opt->dialogbuilder != NULL) + exit_error(true, "%s and --slider without " + "--and-dialog", opt->name); + opt->name = "--slider"; + opt->dialogbuilder = slider_builder; + break; case TEXTBOX: if (opt->dialogbuilder != NULL) exit_error(true, "%s and --textbox without " diff --git a/contrib/tzcode/localtime.c b/contrib/tzcode/localtime.c index 1668475ea646..58099d234e2b 100644 --- a/contrib/tzcode/localtime.c +++ b/contrib/tzcode/localtime.c @@ -1583,15 +1583,15 @@ tzdata_is_fresh(void) struct timespec now; if (clock_gettime(CLOCK_MONOTONIC, &now) < 0) - return 0; + return 1; - if ((now.tv_sec - last_checked >= __tz_change_interval) || - (last_checked > now.tv_sec)) { + if (last_checked == 0 || last_checked > now.tv_sec || + now.tv_sec - last_checked >= __tz_change_interval) { last_checked = now.tv_sec; - return 1; + return 0; } - return 0; + return 1; } #endif /* DETECT_TZ_CHANGES */ @@ -1642,7 +1642,7 @@ tzset_unlocked_name(char const *name) ? lcl_is_set < 0 : 0 < lcl_is_set && strcmp(lcl_TZname, name) == 0) #ifdef DETECT_TZ_CHANGES - if (tzdata_is_fresh() == 0) + if (tzdata_is_fresh()) #endif /* DETECT_TZ_CHANGES */ return; # ifdef ALL_STATE diff --git a/crypto/openssh/misc.c b/crypto/openssh/misc.c index dd0bd032ae3c..1ed71646ebe4 100644 --- a/crypto/openssh/misc.c +++ b/crypto/openssh/misc.c @@ -2535,8 +2535,10 @@ format_absolute_time(uint64_t t, char *buf, size_t len) time_t tt = t > SSH_TIME_T_MAX ? SSH_TIME_T_MAX : t; struct tm tm; - localtime_r(&tt, &tm); - strftime(buf, len, "%Y-%m-%dT%H:%M:%S", &tm); + if (localtime_r(&tt, &tm) == NULL) + strlcpy(buf, "UNKNOWN-TIME", len); + else + strftime(buf, len, "%Y-%m-%dT%H:%M:%S", &tm); } /* diff --git a/etc/mtree/BSD.include.dist b/etc/mtree/BSD.include.dist index ea333a38d889..56d3a823a8f2 100644 --- a/etc/mtree/BSD.include.dist +++ b/etc/mtree/BSD.include.dist @@ -2,17 +2,17 @@ # Please see the file src/etc/mtree/README before making changes to this file. # -/set type=dir uname=root gname=wheel mode=0755 +/set type=dir uname=root gname=wheel mode=0755 tags=package=clibs-dev . arpa .. - atf-c + atf-c tags=package=atf-dev .. - atf-c++ + atf-c++ tags=package=atf-dev .. - bsm + bsm tags=package=audit-dev .. - bsnmp + bsnmp tags=package=bsnmp-dev .. c++ v1 @@ -118,7 +118,7 @@ scsi .. .. - casper + casper tags=package=libcasper-dev .. crypto .. @@ -190,10 +190,10 @@ wg .. .. - devdctl + devdctl tags=package=utilities-dev .. - edit - readline + edit tags=package=runtime-dev + readline tags=package=runtime-dev .. .. fs @@ -252,39 +252,39 @@ virstor .. .. - gssapi + gssapi tags=package=kerberos-dev .. - gssrpc + gssrpc tags=package=kerberos-dev .. - infiniband - complib + infiniband tags=package=utilities-dev + complib tags=package=utilities-dev .. - iba + iba tags=package=utilities-dev .. - opensm + opensm tags=package=utilities-dev .. - vendor + vendor tags=package=utilities-dev .. .. isofs cd9660 .. .. - kadm5 + kadm5 tags=package=kerberos-dev .. krb5 tags=package=kerberos-dev .. - lib80211 + lib80211 tags=package=runtime-dev .. - lib9p + lib9p tags=package=lib9p-dev .. libipt .. - libmilter + libmilter tags=package=libmilter-dev .. - libxo + libxo tags=package=runtime-dev .. - lzma + lzma tags=package=xz-dev .. machine pc @@ -323,7 +323,7 @@ .. .. netpfil - pf + pf tags=package=pf-dev .. .. netsmb @@ -348,20 +348,20 @@ stm .. .. - openssl + openssl tags=package=openssl-dev .. - pcap + pcap tags=package=utilities-dev .. protocols .. - rdma + rdma tags=package=utilities-dev .. rpc .. rpcsvc .. security - audit + audit tags=package=audit-dev .. mac_biba .. diff --git a/etc/mtree/BSD.usr.dist b/etc/mtree/BSD.usr.dist index 6a8c155e5e73..79db101e74d9 100644 --- a/etc/mtree/BSD.usr.dist +++ b/etc/mtree/BSD.usr.dist @@ -6,35 +6,35 @@ . bin .. - include - private - bsddialog + include tags=package=clibs-dev + private tags=package=clibs-dev + bsddialog tags=package=utilities-dev .. - bsdstat + bsdstat tags=package=libbsdstat-dev .. - event1 + event1 tags=package=libevent1-dev .. - gmock - internal - custom + gmock tags=package=utilities-dev + internal tags=package=utilities-dev + custom tags=package=utilities-dev .. .. .. - gtest - internal - custom + gtest tags=package=utilities-dev + internal tags=package=utilities-dev + custom tags=package=utilities-dev .. .. .. - samplerate + samplerate tags=package=sound-dev .. - sqlite3 + sqlite3 tags=package=libsqlite3-dev .. - ucl + ucl tags=package=libucl-dev .. - yaml + yaml tags=package=libyaml-dev .. - zstd + zstd tags=package=runtime-dev .. .. .. @@ -242,8 +242,8 @@ .. legal .. - llvm - clang + llvm tags=package=clang + clang tags=package=clang .. .. ncurses @@ -857,7 +857,7 @@ scrnmaps .. .. - tabset + tabset tags=package=ncurses-lib .. vi catalog diff --git a/etc/termcap/Makefile b/etc/termcap/Makefile index cd5516f3e04b..ea8fab4e839e 100644 --- a/etc/termcap/Makefile +++ b/etc/termcap/Makefile @@ -1,6 +1,8 @@ .PATH: ${SRCTOP}/share/termcap -PACKAGE= runtime +# Note: This is in ncurses-lib rather than ncurses because without it, ncurses +# doesn't work, and the base ncurses package is optional. +PACKAGE= ncurses-lib CLEANFILES+= termcap.small CONFS= termcap.small diff --git a/krb5/lib/kadm5clnt/Makefile b/krb5/lib/kadm5clnt/Makefile index 52a7187cf9bb..ef01a5f779e3 100644 --- a/krb5/lib/kadm5clnt/Makefile +++ b/krb5/lib/kadm5clnt/Makefile @@ -86,7 +86,8 @@ ${CHPASS_UTIL_STRINGS_ERR_C}: ${CHPASS_UTIL_STRINGS_ERR} rm -f et-c-${.PREFIX}.et et-c-${.PREFIX}.c afterinstall: - ${INSTALL_LIBSYMLINK} ${SHLIB_LINK} ${DESTDIR}${LIBDIR}/libkadm5clnt.so + ${INSTALL_LIBSYMLINK} ${DEV_TAG_ARGS} ${SHLIB_LINK} \ + ${DESTDIR}${LIBDIR}/libkadm5clnt.so .include <bsd.lib.mk> diff --git a/lib/atf/Makefile.inc b/lib/atf/Makefile.inc index bebed0280596..af176036f136 100644 --- a/lib/atf/Makefile.inc +++ b/lib/atf/Makefile.inc @@ -24,7 +24,7 @@ # SUCH DAMAGE. # -PACKAGE= atf +PACKAGE?= atf LIB_PACKAGE= CFLAGS+= -DHAVE_CONFIG_H diff --git a/lib/atf/libatf-c++/tests/Makefile b/lib/atf/libatf-c++/tests/Makefile index 839c6902d6b1..dc052c19df67 100644 --- a/lib/atf/libatf-c++/tests/Makefile +++ b/lib/atf/libatf-c++/tests/Makefile @@ -1,5 +1,7 @@ .include <bsd.init.mk> +PACKAGE= tests + TESTS_SUBDIRS= detail ATF= ${SRCTOP}/contrib/atf diff --git a/lib/atf/libatf-c++/tests/detail/Makefile b/lib/atf/libatf-c++/tests/detail/Makefile index 4b95f8dbd663..55cefe524068 100644 --- a/lib/atf/libatf-c++/tests/detail/Makefile +++ b/lib/atf/libatf-c++/tests/detail/Makefile @@ -1,5 +1,7 @@ .include <bsd.init.mk> +PACKAGE= tests + TESTSDIR= ${TESTSBASE}/lib/atf/libatf-c++/detail ATF= ${SRCTOP}/contrib/atf diff --git a/lib/atf/libatf-c/tests/Makefile b/lib/atf/libatf-c/tests/Makefile index 5647e7b9fcbe..c81c18a91f00 100644 --- a/lib/atf/libatf-c/tests/Makefile +++ b/lib/atf/libatf-c/tests/Makefile @@ -1,5 +1,7 @@ .include <bsd.init.mk> +PACKAGE= tests + TESTS_SUBDIRS= detail ATF= ${SRCTOP}/contrib/atf diff --git a/lib/atf/libatf-c/tests/detail/Makefile b/lib/atf/libatf-c/tests/detail/Makefile index 5123f6f4d796..3fa2919b98b9 100644 --- a/lib/atf/libatf-c/tests/detail/Makefile +++ b/lib/atf/libatf-c/tests/detail/Makefile @@ -1,5 +1,7 @@ .include <bsd.init.mk> +PACKAGE= tests + TESTSDIR= ${TESTSBASE}/lib/atf/libatf-c/detail ATF= ${SRCTOP}/contrib/atf diff --git a/lib/atf/tests/Makefile b/lib/atf/tests/Makefile index 500ff0f20c3b..2609bb593d57 100644 --- a/lib/atf/tests/Makefile +++ b/lib/atf/tests/Makefile @@ -1,4 +1,7 @@ .PATH: ${SRCTOP}/tests + +PACKAGE= tests + KYUAFILE= yes SUBDIR= test-programs diff --git a/lib/googletest/Makefile.inc b/lib/googletest/Makefile.inc index 231d7545f364..43ebace19a15 100644 --- a/lib/googletest/Makefile.inc +++ b/lib/googletest/Makefile.inc @@ -1,5 +1,7 @@ .include <googletest.test.inc.mk> +PACKAGE?= googletest + GTEST_DIR= ${SRCTOP}/contrib/googletest GOOGLEMOCK_SRCROOT= ${GTEST_DIR}/googlemock GOOGLETEST_SRCROOT= ${GTEST_DIR}/googletest diff --git a/lib/googletest/tests/Makefile b/lib/googletest/tests/Makefile index 886b1a2fe49d..350e0fe765fe 100644 --- a/lib/googletest/tests/Makefile +++ b/lib/googletest/tests/Makefile @@ -1,4 +1,7 @@ .PATH: ${SRCTOP}/tests + +PACKAGE= tests + KYUAFILE= yes # Note: we start the gmock_main and gmock tests first since those take up to diff --git a/lib/googletest/tests/Makefile.inc b/lib/googletest/tests/Makefile.inc index 9691aaa93ded..8d19e1fafdea 100644 --- a/lib/googletest/tests/Makefile.inc +++ b/lib/googletest/tests/Makefile.inc @@ -3,6 +3,8 @@ # rather than installing all of them to /usr/tests/lib/googletest TESTSDIR= ${TESTSBASE}/lib/googletest/${.CURDIR:T} +PACKAGE= tests + # Clang's optimizer spends a really long time on these tests at -O2. Changing # -O2 to -O1 reduces the -j32 time for lib/googletest/test from 131s to 71s. # Using -O0 further reduces the time to 29s, and also reduces the disk usage diff --git a/lib/libbsddialog/Makefile b/lib/libbsddialog/Makefile index 2ec633b25147..54390cf87f71 100644 --- a/lib/libbsddialog/Makefile +++ b/lib/libbsddialog/Makefile @@ -13,6 +13,7 @@ SRCS= barbox.c \ libbsddialog.c \ menubox.c \ messagebox.c \ + slider.c \ textbox.c \ theme.c \ timebox.c diff --git a/lib/libcasper/Makefile.inc b/lib/libcasper/Makefile.inc index 00bd221feb27..73a761ba4ce3 100644 --- a/lib/libcasper/Makefile.inc +++ b/lib/libcasper/Makefile.inc @@ -1,5 +1,7 @@ .include <src.opts.mk> +PACKAGE?= libcasper + .if ${MK_CASPER} != "no" CFLAGS+=-DWITH_CASPER .endif diff --git a/lib/libcasper/libcasper/Makefile b/lib/libcasper/libcasper/Makefile index 4db26f665f19..1a794791570f 100644 --- a/lib/libcasper/libcasper/Makefile +++ b/lib/libcasper/libcasper/Makefile @@ -1,5 +1,3 @@ -PACKAGE= runtime - SHLIBDIR?= /lib .include <src.opts.mk> diff --git a/lib/libcasper/services/cap_dns/Makefile b/lib/libcasper/services/cap_dns/Makefile index 4b11c97d29e5..b090c553bd28 100644 --- a/lib/libcasper/services/cap_dns/Makefile +++ b/lib/libcasper/services/cap_dns/Makefile @@ -2,8 +2,6 @@ SHLIBDIR?= /lib .include <src.opts.mk> -PACKAGE= runtime - SHLIB_MAJOR= 2 INCSDIR?= ${INCLUDEDIR}/casper diff --git a/lib/libcasper/services/cap_fileargs/Makefile b/lib/libcasper/services/cap_fileargs/Makefile index 2c52d0887a48..9d70d0ab9237 100644 --- a/lib/libcasper/services/cap_fileargs/Makefile +++ b/lib/libcasper/services/cap_fileargs/Makefile @@ -2,8 +2,6 @@ SHLIBDIR?= /lib .include <src.opts.mk> -PACKAGE= runtime - SHLIB_MAJOR= 1 INCSDIR?= ${INCLUDEDIR}/casper diff --git a/lib/libcasper/services/cap_grp/Makefile b/lib/libcasper/services/cap_grp/Makefile index a921dfa87e7c..13e695813bcf 100644 --- a/lib/libcasper/services/cap_grp/Makefile +++ b/lib/libcasper/services/cap_grp/Makefile @@ -2,8 +2,6 @@ SHLIBDIR?= /lib .include <src.opts.mk> -PACKAGE= runtime - SHLIB_MAJOR= 1 INCSDIR?= ${INCLUDEDIR}/casper diff --git a/lib/libcasper/services/cap_net/Makefile b/lib/libcasper/services/cap_net/Makefile index 1ba35a674a05..4e9814118c41 100644 --- a/lib/libcasper/services/cap_net/Makefile +++ b/lib/libcasper/services/cap_net/Makefile @@ -2,8 +2,6 @@ SHLIBDIR?= /lib .include <src.opts.mk> -PACKAGE=libcasper - SHLIB_MAJOR= 1 INCSDIR?= ${INCLUDEDIR}/casper diff --git a/lib/libcasper/services/cap_netdb/Makefile b/lib/libcasper/services/cap_netdb/Makefile index 853052e78d04..a330eeedeb11 100644 --- a/lib/libcasper/services/cap_netdb/Makefile +++ b/lib/libcasper/services/cap_netdb/Makefile @@ -2,8 +2,6 @@ SHLIBDIR?= /lib .include <src.opts.mk> -PACKAGE= runtime - SHLIB_MAJOR= 1 INCSDIR?= ${INCLUDEDIR}/casper diff --git a/lib/libcasper/services/cap_pwd/Makefile b/lib/libcasper/services/cap_pwd/Makefile index a1e97845c736..ba8df80d5ad7 100644 --- a/lib/libcasper/services/cap_pwd/Makefile +++ b/lib/libcasper/services/cap_pwd/Makefile @@ -2,8 +2,6 @@ SHLIBDIR?= /lib .include <src.opts.mk> -PACKAGE= runtime - SHLIB_MAJOR= 1 INCSDIR?= ${INCLUDEDIR}/casper diff --git a/lib/libcasper/services/cap_sysctl/Makefile b/lib/libcasper/services/cap_sysctl/Makefile index 522313df4ffc..4408bad4efb4 100644 --- a/lib/libcasper/services/cap_sysctl/Makefile +++ b/lib/libcasper/services/cap_sysctl/Makefile @@ -2,8 +2,6 @@ SHLIBDIR?= /lib .include <src.opts.mk> -PACKAGE= runtime - SHLIB_MAJOR= 2 INCSDIR?= ${INCLUDEDIR}/casper diff --git a/lib/libcasper/services/cap_syslog/Makefile b/lib/libcasper/services/cap_syslog/Makefile index 88979d8bed23..d18ad6d76ede 100644 --- a/lib/libcasper/services/cap_syslog/Makefile +++ b/lib/libcasper/services/cap_syslog/Makefile @@ -2,8 +2,6 @@ SHLIBDIR?= /lib .include <src.opts.mk> -PACKAGE= runtime - SHLIB_MAJOR= 1 INCSDIR?= ${INCLUDEDIR}/casper diff --git a/lib/libcasper/services/tests/Makefile b/lib/libcasper/services/tests/Makefile index 29b1b564beca..4b6c72fd86e8 100644 --- a/lib/libcasper/services/tests/Makefile +++ b/lib/libcasper/services/tests/Makefile @@ -1,4 +1,6 @@ .PATH: ${SRCTOP}/tests + +PACKAGE= tests KYUAFILE= yes .include <bsd.test.mk> diff --git a/lib/libcasper/tests/Makefile b/lib/libcasper/tests/Makefile index 29b1b564beca..4b6c72fd86e8 100644 --- a/lib/libcasper/tests/Makefile +++ b/lib/libcasper/tests/Makefile @@ -1,4 +1,6 @@ .PATH: ${SRCTOP}/tests + +PACKAGE= tests KYUAFILE= yes .include <bsd.test.mk> diff --git a/lib/libpam/static_libpam/Makefile b/lib/libpam/static_libpam/Makefile index 8830f09ef0da..703e6a4a2f00 100644 --- a/lib/libpam/static_libpam/Makefile +++ b/lib/libpam/static_libpam/Makefile @@ -45,6 +45,7 @@ MAN= MANNODEV= INCS= MLINKS= +MANNODEVLINKS= MK_TESTS= no # diff --git a/lib/libpfctl/libpfctl.c b/lib/libpfctl/libpfctl.c index e739e55033e2..e747763ae6ef 100644 --- a/lib/libpfctl/libpfctl.c +++ b/lib/libpfctl/libpfctl.c @@ -1491,7 +1491,7 @@ snl_attr_get_pf_rule_labels(struct snl_state *ss, struct nlattr *nla, bool ret; if (l->i >= PF_RULE_MAX_LABEL_COUNT) - return (E2BIG); + return (false); ret = snl_attr_copy_string(ss, nla, (void *)PF_RULE_LABEL_SIZE, l->labels[l->i]); @@ -1561,7 +1561,7 @@ snl_attr_get_pf_timeout(struct snl_state *ss, struct nlattr *nla, bool ret; if (t->i >= PFTM_MAX) - return (E2BIG); + return (false); ret = snl_attr_get_uint32(ss, nla, NULL, &t->timeouts[t->i]); if (ret) @@ -2742,6 +2742,88 @@ int pfctl_table_get_addrs(int dev, struct pfr_table *tbl, struct pfr_addr *addr, return (0); } +struct nl_addrs { + size_t max; + struct pfr_addr *addrs; + size_t count; + size_t total_count; +}; + +#define _OUT(_field) offsetof(struct pfr_addr, _field) +static const struct snl_attr_parser ap_pfr_addr[] = { + { .type = PFR_A_AF, .off = _OUT(pfra_af), .cb = snl_attr_get_uint32 }, + { .type = PFR_A_NET, .off = _OUT(pfra_net), .cb = snl_attr_get_uint8 }, + { .type = PFR_A_NOT, .off = _OUT(pfra_not), .cb = snl_attr_get_bool }, + { .type = PFR_A_ADDR, .off = _OUT(pfra_ip6addr), .cb = snl_attr_get_in6_addr }, +}; +#undef _OUT +SNL_DECLARE_ATTR_PARSER(pfr_addr_parser, ap_pfr_addr); + +static bool +snl_attr_get_pfr_addrs(struct snl_state *ss, struct nlattr *nla, + const void *arg __unused, void *target) +{ + struct nl_addrs *a = (struct nl_addrs *)target; + bool ret; + + if (a->count >= a->max) + return (false); + + ret = snl_parse_header(ss, NLA_DATA(nla), NLA_DATA_LEN(nla), + &pfr_addr_parser, &a->addrs[a->count]); + if (ret) + a->count++; + + return (ret); +} + +#define _OUT(_field) offsetof(struct nl_addrs, _field) +static struct snl_attr_parser ap_table_get_addr[] = { + { .type = PF_TA_ADDR, .off = 0, .cb = snl_attr_get_pfr_addrs }, + { .type = PF_TA_ADDR_COUNT, .off = _OUT(total_count), .cb = snl_attr_get_uint32 }, +}; +#undef _OUT +SNL_DECLARE_PARSER(table_get_addr_parser, struct genlmsghdr, snl_f_p_empty, ap_table_get_addr); +int +pfctl_table_get_addrs_h(struct pfctl_handle *h, struct pfr_table *tbl, + struct pfr_addr *addr, int *size, int flags) +{ + struct nl_addrs addrs = { 0 }; + struct snl_writer nw; + struct snl_errmsg_data e = {}; + struct nlmsghdr *hdr; + uint32_t seq_id; + int family_id; + + family_id = snl_get_genl_family(&h->ss, PFNL_FAMILY_NAME); + if (family_id == 0) + return (ENOTSUP); + + snl_init_writer(&h->ss, &nw); + hdr = snl_create_genl_msg_request(&nw, family_id, PFNL_CMD_TABLE_GET_ADDR); + + snl_add_msg_attr_table(&nw, PF_TA_TABLE, tbl); + snl_add_msg_attr_u32(&nw, PF_TA_FLAGS, flags); + + if ((hdr = snl_finalize_msg(&nw)) == NULL) + return (ENXIO); + + seq_id = hdr->nlmsg_seq; + if (! snl_send_message(&h->ss, hdr)) + return (ENXIO); + + addrs.addrs = addr; + addrs.max = *size; + while ((hdr = snl_read_reply_multi(&h->ss, seq_id, &e)) != NULL) { + if (! snl_parse_nlmsg(&h->ss, hdr, &table_get_addr_parser, &addrs)) + continue; + } + + *size = addrs.total_count; + + return (e.error); +} + int pfctl_set_statusif(struct pfctl_handle *h, const char *ifname) { diff --git a/lib/libpfctl/libpfctl.h b/lib/libpfctl/libpfctl.h index ae4b18dabe75..a5b7e1c23bd0 100644 --- a/lib/libpfctl/libpfctl.h +++ b/lib/libpfctl/libpfctl.h @@ -529,6 +529,8 @@ int pfctl_table_set_addrs_h(struct pfctl_handle *h, struct pfr_table *tbl, int pfctl_table_set_addrs(int dev, struct pfr_table *tbl, struct pfr_addr *addr, int size, int *size2, int *nadd, int *ndel, int *nchange, int flags); +int pfctl_table_get_addrs_h(struct pfctl_handle *h, struct pfr_table *tbl, struct pfr_addr *addr, + int *size, int flags); int pfctl_table_get_addrs(int dev, struct pfr_table *tbl, struct pfr_addr *addr, int *size, int flags); int pfctl_set_statusif(struct pfctl_handle *h, const char *ifname); diff --git a/lib/libsys/Makefile.sys b/lib/libsys/Makefile.sys index bd65b58083c2..1d1a4f1136ce 100644 --- a/lib/libsys/Makefile.sys +++ b/lib/libsys/Makefile.sys @@ -243,6 +243,7 @@ MAN+= abort2.2 \ jail.2 \ kcmp.2 \ kenv.2 \ + kexec_load.2 \ kill.2 \ kldfind.2 \ kldfirstmod.2 \ diff --git a/lib/libsys/kexec_load.2 b/lib/libsys/kexec_load.2 new file mode 100644 index 000000000000..c3d458f34100 --- /dev/null +++ b/lib/libsys/kexec_load.2 @@ -0,0 +1,119 @@ +.\" +.\" SPDX-License-Identifier: BSD-3-Clause +.\" +.\" Copyright (c) 2025 Juniper Networks, Inc. +.\" +.Dd October 29, 2025 +.Dt KEXEC_LOAD 2 +.Os +.Sh NAME +.Nm kexec_load +.Nd prepare new kernel to reboot into +.Sh SYNOPSIS +.Lb libc +.In sys/kexec.h +.Ft int +.Fn kexec_load "uint64_t entry" "unsigned long count" \ + "struct kexec_segment *segments" "unsigned long flags" +.Sh DESCRIPTION +The +.Fn kexec_load +system call loads a new kernel that can be executed later by +.Xr reboot 2 . +Subsequent calls will replace previously loaded images. +.Pp +The +.Fa flags +argument is a bitmask of flags that control the operation of the call. +This argument is present for compatibility with Linux, although it is currently +unused and must be 0. +.Pp +The +.Fa entry +argument is the physical address of the entry point of the new kernel image. +.Pp +The +.Fa count +argument is the number of segments in the image, currently limited to 16. +A value of 0 will unload the currently staged image, if one exists, without +staging a new image. +.Pp +The +.Fa segments +argument is an array of +.Fa count +members of the following structure: +.Bd -literal -offset indent +struct kexec_segment { + void *buf; + size_t bufsz; + vm_paddr_t mem; + vm_size_t memsz; +}; +.Ed +.Pp +The +.Va buf +and +.Va bufsz +members specify a memory region in the caller's address space containing the +source of the segment. +The +.Va mem +and +.Va memsz +members specify the target physical region of the segment. +.Va bufsz +must be less than or equal to +.Va memsz , +and +.Va mem +and +.Va memsz +must be page aligned. +The region covered by +.Va mem +must be in the list covered by the +.Va vm.phys_segs +sysctl. +.Pp +The +.Fn kexec_load +system call stages the kernel image in safe memory along with all +machine-dependent image data until +.Xr reboot 2 +is called with the +.Va RB_KEXEC +flag to load the image and execute the new kernel. +.Sh RETURN VALUES +The +.Fn kexec_load +system call returns 0 on success. +On failure, -1 is returned, and +.Va errno +is set to indicate the error. +On success any previously loaded image is unloaded and replaced with the new +image. +On failure, the previously loaded image is unchanged. +.Sh ERRORS +The following errors may be returned: +.Bl -tag -width Er +.It Bq Er EINVAL +Too many segments in image. +.It Bq Er EINVAL +The value of +.Va bufsz +is larger than +.Va memsz +in one or more segments. +.It Bq Er EINVAL +Machine-dependent load error. +.It Bq Er EBUSY +Another +.Fn kexec_load +call is in progress. +.Sh HISTORY +The +.Nm +system call appeared in +.Fx 16.0 . diff --git a/lib/libsys/pathconf.2 b/lib/libsys/pathconf.2 index 5a983a3a13e2..5348128be706 100644 --- a/lib/libsys/pathconf.2 +++ b/lib/libsys/pathconf.2 @@ -25,7 +25,7 @@ .\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF .\" SUCH DAMAGE. .\" -.Dd August 6, 2025 +.Dd October 31, 2025 .Dt PATHCONF 2 .Os .Sh NAME @@ -180,7 +180,8 @@ Return 1 if named attributes are enabled for the file system, otherwise 0. .It Li _PC_HAS_NAMEDATTR Return 1 if one or more named attributes exist for the file, otherwise 0. .It Li _PC_HAS_HIDDENSYSTEM -Return 1 if both +Return 1 if all of the +.Dv UF_ARCHIVE , .Dv UF_HIDDEN and .Dv UF_SYSTEM @@ -192,6 +193,9 @@ Returns the block size required for block cloning via .Xr copy_file_range 2 for a file system if block cloning is supported, otherwise 0. +.It Li _PC_CASE_INSENSITIVE +Return 1 if the file system performs case insensitive lookups, +otherwise 0. .El .Sh RETURN VALUES If the call to diff --git a/lib/libsys/posix_fallocate.2 b/lib/libsys/posix_fallocate.2 index 8be075b41331..94858c4a0f90 100644 --- a/lib/libsys/posix_fallocate.2 +++ b/lib/libsys/posix_fallocate.2 @@ -25,7 +25,7 @@ .\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF .\" SUCH DAMAGE. .\" -.Dd March 30, 2020 +.Dd November 2, 2025 .Dt POSIX_FALLOCATE 2 .Os .Sh NAME @@ -105,8 +105,7 @@ The .Fa len argument was less than or equal to zero, the .Fa offset -argument was less than zero, -or the operation is not supported by the file system. +argument was less than zero. .It Bq Er EIO An I/O error occurred while reading from or writing to a file system. .It Bq Er EINTEGRITY @@ -123,6 +122,8 @@ media. The file descriptor .Fa fd has insufficient rights. +.It Bq Er EOPNOTSUPP +The operation is not supported by the file system. .It Bq Er ESPIPE The .Fa fd @@ -137,12 +138,29 @@ argument is associated with a pipe or FIFO. The .Fn posix_fallocate system call conforms to -.St -p1003.1-2004 . +.St -p1003.1-2024 . .Sh HISTORY The .Fn posix_fallocate function appeared in .Fx 9.0 . +.Pp +Previous versions of +.Nm +used +.Er EINVAL +to indicate that the operation is not supported by the file system, as specified +in +.St -p1003.1 +Base Specifications, Issue 7. +.St -p1003.1 +Base Specifications, Issue 8 switched to requiring +.Er EOPNOTSUPP +for this error case. +ZFS adopted the latter convention in +.Fx 15.0 , +and the remaining filesystems in base adopted it in +.Fx 15.1 . .Sh AUTHORS .Fn posix_fallocate and this manual page were initially written by diff --git a/lib/libsys/reboot.2 b/lib/libsys/reboot.2 index f6c7bf6c83cc..54fa8b599cd5 100644 --- a/lib/libsys/reboot.2 +++ b/lib/libsys/reboot.2 @@ -25,7 +25,7 @@ .\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF .\" SUCH DAMAGE. .\" -.Dd July 10, 2018 +.Dd October 29, 2025 .Dt REBOOT 2 .Os .Sh NAME @@ -126,6 +126,10 @@ on the console. is actually interpreted by the .Xr init 8 program in the newly booted system. +.It Dv RB_KEXEC +Execute a new kernel loaded via +.Fn kexec_load 2 . +If no kernel was loaded, reboot as normal. .El .Pp When no options are given (i.e., @@ -149,6 +153,7 @@ variable The caller is not the super-user. .El .Sh SEE ALSO +.Xr kexec_load 2 , .Xr crash 8 , .Xr halt 8 , .Xr init 8 , diff --git a/lib/ofed/Makefile.inc b/lib/ofed/Makefile.inc index 1b911c451c01..5a16e0015c07 100644 --- a/lib/ofed/Makefile.inc +++ b/lib/ofed/Makefile.inc @@ -1 +1,4 @@ +PACKAGE?= rdma +LIB_PACKAGE= + WARNS?= 0 diff --git a/libexec/atf/Makefile.inc b/libexec/atf/Makefile.inc index e40827fa0b69..5fd06c35cd09 100644 --- a/libexec/atf/Makefile.inc +++ b/libexec/atf/Makefile.inc @@ -24,7 +24,7 @@ # SUCH DAMAGE. # -PACKAGE= atf +PACKAGE?= atf LIB_PACKAGE= CFLAGS+= -DHAVE_CONFIG_H diff --git a/libexec/atf/atf-check/tests/Makefile b/libexec/atf/atf-check/tests/Makefile index e98f82d941cd..6e21e4ede211 100644 --- a/libexec/atf/atf-check/tests/Makefile +++ b/libexec/atf/atf-check/tests/Makefile @@ -1,6 +1,8 @@ ATF= ${SRCTOP}/contrib/atf .PATH: ${ATF}/atf-sh +PACKAGE= tests + ATF_TESTS_SH= atf-check_test .include <bsd.test.mk> diff --git a/libexec/atf/tests/Makefile b/libexec/atf/tests/Makefile index 29b1b564beca..ad9431e75a63 100644 --- a/libexec/atf/tests/Makefile +++ b/libexec/atf/tests/Makefile @@ -1,4 +1,7 @@ .PATH: ${SRCTOP}/tests + +PACKAGE= tests + KYUAFILE= yes .include <bsd.test.mk> diff --git a/libexec/makewhatis.local/Makefile b/libexec/makewhatis.local/Makefile index 765036623d49..b541dc8e4de1 100644 --- a/libexec/makewhatis.local/Makefile +++ b/libexec/makewhatis.local/Makefile @@ -1,3 +1,4 @@ +PACKAGE= mandoc SCRIPTS= makewhatis.local.sh MAN= makewhatis.local.8 SCRIPTSDIR= ${LIBEXECDIR} diff --git a/release/Makefile.vm b/release/Makefile.vm index a04f779ebebb..142fd6e7bdf5 100644 --- a/release/Makefile.vm +++ b/release/Makefile.vm @@ -129,6 +129,13 @@ ${_CW:tu}${_FS:tu}${_FMT:tu}IMAGE= ${_CW:tl}.${_FS}.${_FMT} cw-${_CW:tl}-${_FS}-${_FMT}: cw-ec2-base-${_FS}-${_FMT} .endif +# Special handling: GCE images ingest src.txz and ports.txz and expect them +# to be in the /ftp/ directory. Note: This will need to be reworked before +# distribution sets go away! +.if ${_CW} == GCE +cw-${_CW:tl}-${_FS}-${_FMT}: ftp +.endif + cw-${_CW:tl}-${_FS}-${_FMT}: ${QEMUTGT} ${PKGBASE_REPO_DIR} mkdir -p ${.OBJDIR}/${.TARGET} env TARGET=${TARGET} TARGET_ARCH=${TARGET_ARCH} SWAPSIZE=${SWAPSIZE} \ diff --git a/release/packages/sets/base-dbg.ucl b/release/packages/sets/base-dbg.ucl index 79e5de22522e..c96b10416dcb 100644 --- a/release/packages/sets/base-dbg.ucl +++ b/release/packages/sets/base-dbg.ucl @@ -28,5 +28,8 @@ deps { }, "set-devel-dbg" { version = "${VERSION}" + }, + "set-optional-dbg" { + version = "${VERSION}" } } diff --git a/release/packages/ucl/googletest-all.ucl b/release/packages/ucl/googletest-all.ucl new file mode 100644 index 000000000000..889e8a65f314 --- /dev/null +++ b/release/packages/ucl/googletest-all.ucl @@ -0,0 +1,33 @@ +/* + * SPDX-License-Identifier: ISC + * + * Copyright (c) 2025 Lexi Winter <ivy@FreeBSD.org> + * + * Permission to use, copy, modify, and distribute this software for any + * purpose with or without fee is hereby granted, provided that the above + * copyright notice and this permission notice appear in all copies. + * + * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES + * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF + * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR + * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES + * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN + * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF + * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE. + */ + +comment = "Unit testing framework" + +desc = <<EOD +Google Test (gtest) is an xUnit-based unit testing framework for C++, +developed by Google LLC. + +This version of Google Test is provided for use by unit tests in the +base system, and is not intended for third-party users. A supported +version of Google Test may be found in the FreeBSD Ports Collection +as devel/googletest. +EOD + +annotations { + set = "optional,optional-jail" +} diff --git a/release/packages/ucl/tests-all.ucl b/release/packages/ucl/tests-all.ucl index 3ad2d0f50e6b..315ac2e8cce0 100644 --- a/release/packages/ucl/tests-all.ucl +++ b/release/packages/ucl/tests-all.ucl @@ -23,24 +23,6 @@ The test suite, installed in /usr/tests, allows the functionality of the installed system to be verified. EOD -deps { - # Nearly all the tests require atf to run. - "atf": { - version = "${VERSION}" - }, - - # The test framework requires Kyua. - "kyua": { - version = "${VERSION}" - }, - - # Since the purpose of the tests is to test the base system, the base - # system must be installed. - "set-base": { - version = "${VERSION}" - } -} - annotations { set = tests } diff --git a/release/packages/ucl/tests.ucl b/release/packages/ucl/tests.ucl index bac72f1534d3..da9eb59295bc 100644 --- a/release/packages/ucl/tests.ucl +++ b/release/packages/ucl/tests.ucl @@ -18,10 +18,29 @@ deps { + # Nearly all the tests require atf to run. + "atf": { + version = "${VERSION}" + }, + # Quite a few tests require flua. "flua" { version = "${VERSION}" }, -} + # Some tests need GoogleTest + "googletest": { + version = "${VERSION}" + }, + # The test framework requires Kyua. + "kyua": { + version = "${VERSION}" + }, + + # Since the purpose of the tests is to test the base system, the base + # system must be installed. + "set-base": { + version = "${VERSION}" + } +} diff --git a/release/tools/gce.conf b/release/tools/gce.conf index a7ccfc27a75c..2fa23f6e54f4 100644 --- a/release/tools/gce.conf +++ b/release/tools/gce.conf @@ -21,7 +21,7 @@ vm_extra_install_base() { echo 'search google.internal' > ${DESTDIR}/etc/resolv.conf echo 'nameserver 169.254.169.254' >> ${DESTDIR}/etc/resolv.conf echo 'nameserver 8.8.8.8' >> ${DESTDIR}/etc/resolv.conf - metalog_add ./etc/resolv.conf + metalog_add_data ./etc/resolv.conf } vm_extra_pre_umount() { @@ -100,11 +100,12 @@ EOF fi if [ -e "${DESTDIR}/../ftp/ports.txz" ]; then tar fxJ ${DESTDIR}/../ftp/ports.txz -C ${DESTDIR} - _INSTALLED_PACKAGES=$(pkg -r ${DESTDIR} info -o -q -a) + _INSTALLED_PACKAGES=$(pkg -r ${DESTDIR} info -o -q -a | grep -v ^base/) for PACKAGE in ${_INSTALLED_PACKAGES}; do make -C ${DESTDIR}/usr/ports/${PACKAGE} fetch \ - DISTDIR=${DESTDIR}/usr/ports/distfiles - + DISTDIR=${DESTDIR}/usr/ports/distfiles \ + DISABLE_VULNERABILITIES=YES \ + I_DONT_CARE_IF_MY_BUILDS_TARGET_THE_WRONG_RELEASE=YES done fi diff --git a/release/tools/vmimage.subr b/release/tools/vmimage.subr index 3bfef585f613..8531e9b8f2d6 100644 --- a/release/tools/vmimage.subr +++ b/release/tools/vmimage.subr @@ -212,6 +212,13 @@ vm_extra_install_packages() { -r ${DESTDIR} \ install -y -r ${PKG_REPO_NAME} $pkg done + INSTALL_AS_USER=yes \ + ${PKG_CMD} \ + -o ABI=${PKG_ABI} \ + -o REPOS_DIR=${PKG_REPOS_DIR} \ + -o PKG_DBDIR=${DESTDIR}/var/db/pkg \ + -r ${DESTDIR} \ + autoremove metalog_add_data ./var/db/pkg/local.sqlite else if [ -n "${WITHOUT_QEMU}" ]; then @@ -224,6 +231,8 @@ vm_extra_install_packages() { chroot ${DESTDIR} ${EMULATOR} env ASSUME_ALWAYS_YES=yes \ /usr/sbin/pkg install -y ${p} done + chroot ${DESTDIR} ${EMULATOR} env ASSUME_ALWAYS_YES=yes \ + /usr/sbin/pkg autoremove fi return 0 diff --git a/sbin/dumpon/dumpon.8 b/sbin/dumpon/dumpon.8 index a57dfef7096d..4e4abb8d4b79 100644 --- a/sbin/dumpon/dumpon.8 +++ b/sbin/dumpon/dumpon.8 @@ -25,7 +25,7 @@ .\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF .\" SUCH DAMAGE. .\" -.Dd April 23, 2020 +.Dd November 6, 2025 .Dt DUMPON 8 .Os .Sh NAME @@ -228,6 +228,20 @@ total amount of physical memory as reported by the .Va hw.physmem .Xr sysctl 8 variable. +.Sh SYSCTL VARIABLES +The following +.Xr sysctl 8 +variables can be used to modify or monitor the behavior of crash dumps: +.Bl -tag -width "machdep.dump_retry_count" +.It Va debug.minidump +Set the type of kernel crash dump. +Possible values are 0 for a full crash dump or 1 for a minidump. +The default is minidump. +.It Va machdep.dump_retry_count +The maximum number of times dump will retry before giving up. +The default value is 5. +This sysctl is only supported on PowerPC and AMD64. +.El .Sh IMPLEMENTATION NOTES Because the file system layer is already dead by the time a crash dump is taken, it is not possible to send crash dumps directly to a file. @@ -339,7 +353,7 @@ when binds to a server, .Xr dhclient-script 8 can be used to run -.Xr dumpon 8 . +.Nm . For example, to automatically configure .Xr netdump 4 on the vtnet0 interface, add the following to @@ -377,6 +391,7 @@ needed. .Xr loader 8 , .Xr rc 8 , .Xr savecore 8 , +.Xr sysctl 8 , .Xr swapon 8 , .Xr panic 9 .Sh HISTORY diff --git a/sbin/geom/core/geom.c b/sbin/geom/core/geom.c index 2de696ce6a43..51ea52a3e858 100644 --- a/sbin/geom/core/geom.c +++ b/sbin/geom/core/geom.c @@ -897,23 +897,23 @@ list_one_provider(struct gprovider *pp, const char *padding) struct gconfig *conf; char buf[5]; - xo_emit("{Lcw:Name}{:Name}\n", pp->lg_name); + xo_emit("{Lcw:Name}{:name}\n", pp->lg_name); humanize_number(buf, sizeof(buf), (int64_t)pp->lg_mediasize, "", HN_AUTOSCALE, HN_B | HN_NOSPACE | HN_DECIMAL); - xo_emit("{P:/%s}{Lcw:Mediasize}{:Mediasize/%jd} ({N:/%s})\n", + xo_emit("{P:/%s}{Lcw:Mediasize}{:mediasize/%jd} ({N:/%s})\n", padding, (intmax_t)pp->lg_mediasize, buf); - xo_emit("{P:/%s}{Lcw:Sectorsize}{:Sectorsize/%u} \n", + xo_emit("{P:/%s}{Lcw:Sectorsize}{:sectorsize/%u}\n", padding, pp->lg_sectorsize); if (pp->lg_stripesize > 0 || pp->lg_stripeoffset > 0) { - xo_emit("{P:/%s}{Lcw:Stripesize}{Stripesize/%ju}\n", + xo_emit("{P:/%s}{Lcw:Stripesize}{:stripesize/%ju}\n", padding, pp->lg_stripesize); - xo_emit("{P:/%s}{Lcw:Stripeoffset}{Stripeoffset/%ju}\n", + xo_emit("{P:/%s}{Lcw:Stripeoffset}{:stripeoffset/%ju}\n", padding, pp->lg_stripeoffset); } - xo_emit("{P:/%s}{Lcw:Mode}{Mode}\n", padding, pp->lg_mode); + xo_emit("{P:/%s}{Lcw:Mode}{:mode}\n", padding, pp->lg_mode); LIST_FOREACH(conf, &pp->lg_config, lg_config) { xo_emit("{P:/%s}{Lcwa:}{a:}\n", padding, conf->lg_name, - conf->lg_name, conf->lg_val); + conf->lg_name, conf->lg_val ? conf->lg_val : ""); } } @@ -929,24 +929,24 @@ list_one_consumer(struct gconsumer *cp, const char *padding) else { char buf[5]; - xo_emit("{Lcw:Name}{:Name}\n", pp->lg_name); + xo_emit("{Lcw:Name}{:name}\n", pp->lg_name); humanize_number(buf, sizeof(buf), (int64_t)pp->lg_mediasize, "", HN_AUTOSCALE, HN_B | HN_NOSPACE | HN_DECIMAL); - xo_emit("{P:/%s}{Lcw:Mediasize}{:Mediasize/%jd} ({N:/%s})\n", + xo_emit("{P:/%s}{Lcw:Mediasize}{:mediasize/%jd} ({N:/%s})\n", padding, (intmax_t)pp->lg_mediasize, buf); - xo_emit("{P:/%s}{Lcw:Sectorsize}{:Sectorsize/%u}\n", + xo_emit("{P:/%s}{Lcw:Sectorsize}{:sectorsize/%u}\n", padding, pp->lg_sectorsize); if (pp->lg_stripesize > 0 || pp->lg_stripeoffset > 0) { - xo_emit("{P:/%s}{Lcw:Stripesize}{:Stripesize/%ju}\n", + xo_emit("{P:/%s}{Lcw:Stripesize}{:stripesize/%ju}\n", padding, pp->lg_stripesize); - xo_emit("{P:/%s}{Lcw:Stripeoffset}{:Stripesize/%ju}\n", + xo_emit("{P:/%s}{Lcw:Stripeoffset}{:stripeoffset/%ju}\n", padding, pp->lg_stripeoffset); } - xo_emit("{P:/%s}{Lcw:Mode}{:Mode}\n", padding, pp->lg_mode); + xo_emit("{P:/%s}{Lcw:Mode}{:mode}\n", padding, pp->lg_mode); } LIST_FOREACH(conf, &cp->lg_config, lg_config) { xo_emit("{P:/%s}{Lcwa:}{a:}\n", padding, conf->lg_name, - conf->lg_name, conf->lg_val); + conf->lg_name, conf->lg_val ? conf->lg_val : ""); } } @@ -958,13 +958,13 @@ list_one_geom(struct ggeom *gp) struct gconfig *conf; unsigned n; - xo_emit("{Lcw:Geom name}{:Name}\n", gp->lg_name); + xo_emit("{Lcw:Geom name}{:name}\n", gp->lg_name); LIST_FOREACH(conf, &gp->lg_config, lg_config) { xo_emit("{Lcwa:}{a:}\n", conf->lg_name, conf->lg_name, - conf->lg_val); + conf->lg_val ? conf->lg_val : ""); } if (!LIST_EMPTY(&gp->lg_provider)) { - xo_open_list("Providers"); + xo_open_list("providers"); xo_emit("{Tc:Providers}\n"); n = 1; LIST_FOREACH(pp, &gp->lg_provider, lg_provider) { @@ -973,10 +973,10 @@ list_one_geom(struct ggeom *gp) list_one_provider(pp, " "); xo_close_instance("provider"); } - xo_close_list("Providers"); + xo_close_list("providers"); } if (!LIST_EMPTY(&gp->lg_consumer)) { - xo_open_list("Consumers"); + xo_open_list("consumers"); xo_emit("{Tc:Consumers}\n"); n = 1; LIST_FOREACH(cp, &gp->lg_consumer, lg_consumer) { @@ -985,7 +985,7 @@ list_one_geom(struct ggeom *gp) list_one_consumer(cp, " "); xo_close_instance("consumer"); } - xo_close_list("Consumers"); + xo_close_list("consumers"); } xo_emit("\n"); } @@ -1005,10 +1005,10 @@ list_one_geom_by_provider(const char *provider_name) if (gp == NULL) errx(EXIT_FAILURE, "Cannot find provider '%s'.", provider_name); - xo_open_container("Geom"); - xo_emit("{Lwc:Geom class}{:Class}\n", gp->lg_class->lg_name); + xo_open_container(provider_name); + xo_emit("{Lwc:Geom class}{:class}\n", gp->lg_class->lg_name); list_one_geom(gp); - xo_close_container("Geom"); + xo_close_container(provider_name); } static void @@ -1067,12 +1067,12 @@ std_list(struct gctl_req *req, unsigned flags __unused) "an instance named '%s'.", gclass_name, name); } - xo_open_container("Geom"); + xo_open_container(gclass_name); list_one_geom(gp); - xo_close_container("Geom"); + xo_close_container(gclass_name); } } else { - xo_open_list("Geoms"); + xo_open_list(gclass_name); LIST_FOREACH(gp, &classp->lg_geom, lg_geom) { if (LIST_EMPTY(&gp->lg_provider) && !all) continue; @@ -1080,7 +1080,7 @@ std_list(struct gctl_req *req, unsigned flags __unused) list_one_geom(gp); xo_close_instance("geom"); } - xo_close_list("Geoms"); + xo_close_list(gclass_name); } geom_deletetree(&mesh); } @@ -1175,6 +1175,7 @@ status_one_geom(struct ggeom *gp, int script, int name_len, int status_len) { struct gconsumer *cp; struct gconfig *conf; + char fmt[64]; const char *name, *status, *cstate, *csyncr; int gotone, len; @@ -1196,28 +1197,36 @@ status_one_geom(struct ggeom *gp, int script, int name_len, int status_len) csyncr = status_one_consumer(cp, "synchronized"); if (!gotone || script) { if (!gotone) { - xo_emit("{:name/%*s} {:status/%*s} ", + xo_emit("{t:name/%*s} {t:status/%*s} ", name_len, name, status_len, status); + xo_open_list("components"); } else { - xo_emit("{d:name/%*s} {d:status/%*s} ", + /* + * XXX: running the same xo_emit() as above or + * variations of it will cause the XML/JSON to + * produce extra "components" lists in script + * mode + */ + + snprintf(fmt, sizeof(fmt), "%*s %*s ", name_len, name, status_len, status); + xo_emit(fmt); } - xo_open_list("components"); } xo_open_instance("components"); if (cstate != NULL && csyncr != NULL) { xo_emit("{P:/%*s}{:component} ({:state}, {:synchronized})\n", - len, "", cp->lg_provider->lg_name, cstate, csyncr); + len, "", cp->lg_provider->lg_name, cstate, csyncr); } else if (cstate != NULL) { xo_emit("{P:/%*s}{:component} ({:state})\n", - len, "", cp->lg_provider->lg_name, cstate); + len, "", cp->lg_provider->lg_name, cstate); } else if (csyncr != NULL) { xo_emit("{P:/%*s}{:component} ({:synchronized})\n", - len, "", cp->lg_provider->lg_name, csyncr); + len, "", cp->lg_provider->lg_name, csyncr); } else { xo_emit("{P:/%*s}{:component}\n", - len, "", cp->lg_provider->lg_name); + len, "", cp->lg_provider->lg_name); } xo_close_instance("components"); gotone = 1; @@ -1225,13 +1234,11 @@ status_one_geom(struct ggeom *gp, int script, int name_len, int status_len) len = name_len + status_len + 4; } if (!gotone) { - xo_emit("{:name/%*s} {:status/%*s} ", name_len, name, status_len, status); - xo_open_list("components"); - xo_open_instance("components"); - xo_emit("{P:/%*s}{d:component}\n", len, "", "N/A"); - xo_close_instance("components"); + xo_emit("{t:name/%*s} {t:status/%*s} N/A\n", + name_len, name, status_len, status); + } else { + xo_close_list("components"); } - xo_close_list("components"); xo_close_instance("status"); } @@ -1242,6 +1249,7 @@ status_one_geom_prs(struct ggeom *gp, int script, int name_len, int status_len) struct gconsumer *cp; struct gconfig *conf; const char *name, *status, *cstate, *csyncr; + char fmt[64]; int gotone, len; xo_open_instance("status"); @@ -1269,28 +1277,37 @@ status_one_geom_prs(struct ggeom *gp, int script, int name_len, int status_len) csyncr = status_one_consumer(cp, "synchronized"); if (!gotone || script) { if (!gotone) { - xo_emit("{:name/%*s} {:status/%*s} ", + xo_emit("{t:name/%*s} {t:status/%*s} ", name_len, name, status_len, status); + xo_open_list("components"); } else { - xo_emit("{d:name/%*s} {d:status/%*s} ", + /* + * XXX: running the same xo_emit() as + * above or variations of it will + * cause the XML/JSON to produce + * extra "components" lists in + * script mode + */ + + snprintf(fmt, sizeof(fmt), "%*s %*s ", name_len, name, status_len, status); + xo_emit(fmt); } - xo_open_list("components"); } xo_open_instance("component"); if (cstate != NULL && csyncr != NULL) { xo_emit("{P:/%*s}{:component} ({:state}, {:synchronized})\n", - len, "", cp->lg_provider->lg_name, cstate, csyncr); + len, "", cp->lg_provider->lg_name, cstate, csyncr); } else if (cstate != NULL) { xo_emit("{P:/%*s}{:component} ({:state})\n", - len, "", cp->lg_provider->lg_name, cstate); + len, "", cp->lg_provider->lg_name, cstate); } else if (csyncr != NULL) { xo_emit("{P:/%*s}{:component} ({:synchronized})\n", - len, "", cp->lg_provider->lg_name, csyncr); + len, "", cp->lg_provider->lg_name, csyncr); } else { xo_emit("{P:/%*s}{:component}\n", - len, "", cp->lg_provider->lg_name); + len, "", cp->lg_provider->lg_name); } xo_close_instance("component"); gotone = 1; @@ -1298,13 +1315,11 @@ status_one_geom_prs(struct ggeom *gp, int script, int name_len, int status_len) len = name_len + status_len + 4; } if (!gotone) { - xo_emit("{:name/%*s} {:status/%*s} ", name_len, name, status_len, status); - xo_open_list("components"); - xo_open_instance("components"); - xo_emit("{P:/%*s}{d:component}\n", len, "", "N/A"); - xo_close_instance("components"); + xo_emit("{t:name/%*s} {t:status/%*s} N/A\n", + name_len, name, status_len, status); + } else { + xo_close_list("components"); } - xo_close_list("components"); } xo_close_instance("status"); } @@ -1370,7 +1385,7 @@ std_status(struct gctl_req *req, unsigned flags __unused) xo_emit("{T:/%*s} {T:/%*s} {T:Components}\n", name_len, "Name", status_len, "Status"); } - xo_open_list("status"); + xo_open_list(gclass_name); if (nargs > 0) { for (i = 0; i < nargs; i++) { name = gctl_get_ascii(req, "arg%d", i); @@ -1398,7 +1413,7 @@ std_status(struct gctl_req *req, unsigned flags __unused) } } } - xo_close_list("status"); + xo_close_list(gclass_name); end: geom_deletetree(&mesh); } diff --git a/sbin/ifconfig/ifbridge.c b/sbin/ifconfig/ifbridge.c index eff443447c13..8bcf4a638adf 100644 --- a/sbin/ifconfig/ifbridge.c +++ b/sbin/ifconfig/ifbridge.c @@ -811,7 +811,7 @@ unsetbridge_private(if_ctx *ctx, const char *val, int dummy __unused) static int parse_vlans(ifbvlan_set_t *set, const char *str) { - char *s, *token; + char *s, *free_s, *token; /* "none" means the empty vlan set */ if (strcmp(str, "none") == 0) { @@ -829,6 +829,8 @@ parse_vlans(ifbvlan_set_t *set, const char *str) if ((s = strdup(str)) == NULL) return (-1); + /* Keep the original value of s, since strsep() will modify it */ + free_s = s; while ((token = strsep(&s, ",")) != NULL) { unsigned long first, last; @@ -856,11 +858,11 @@ parse_vlans(ifbvlan_set_t *set, const char *str) BRVLAN_SET(set, vlan); } - free(s); + free(free_s); return (0); err: - free(s); + free(free_s); return (-1); } diff --git a/sbin/ipf/libipf/interror.c b/sbin/ipf/libipf/interror.c index 981823ca6bb9..a8dc3be2d5d1 100644 --- a/sbin/ipf/libipf/interror.c +++ b/sbin/ipf/libipf/interror.c @@ -17,7 +17,7 @@ typedef struct { static ipf_error_entry_t *find_error(int); -#define IPF_NUM_ERRORS 477 +#define IPF_NUM_ERRORS sizeof(ipf_errors) / sizeof(ipf_error_entry_t) /* * NO REUSE OF NUMBERS! @@ -25,7 +25,7 @@ static ipf_error_entry_t *find_error(int); * IF YOU WANT TO ADD AN ERROR TO THIS TABLE, _ADD_ A NEW NUMBER. * DO _NOT_ USE AN EMPTY NUMBER OR FILL IN A GAP. */ -static ipf_error_entry_t ipf_errors[IPF_NUM_ERRORS] = { +static ipf_error_entry_t ipf_errors[] = { { 1, "auth table locked/full" }, { 2, "" }, { 3, "copyinptr received bad address" }, @@ -228,6 +228,8 @@ static ipf_error_entry_t ipf_errors[IPF_NUM_ERRORS] = { { 30024, "object size incorrect for hash table" }, { 30025, "hash table size must be at least 1"}, { 30026, "cannot allocate memory for hash table context" }, + { 30027, "hash table larger than maximum allowed" }, + { 30028, "hash table multiplication overflow" }, /* -------------------------------------------------------------------------- */ { 40001, "invalid minor device number for log read" }, { 40002, "read size too small" }, diff --git a/sbin/pfctl/pfctl_radix.c b/sbin/pfctl/pfctl_radix.c index 3b7161420e33..823921953eaf 100644 --- a/sbin/pfctl/pfctl_radix.c +++ b/sbin/pfctl/pfctl_radix.c @@ -182,7 +182,7 @@ pfr_get_addrs(struct pfr_table *tbl, struct pfr_addr *addr, int *size, { int ret; - ret = pfctl_table_get_addrs(dev, tbl, addr, size, flags); + ret = pfctl_table_get_addrs_h(pfh, tbl, addr, size, flags); if (ret) { errno = ret; return (-1); diff --git a/secure/lib/libcrypto/Makefile.inc b/secure/lib/libcrypto/Makefile.inc index 5b281166df61..73c650d590ff 100644 --- a/secure/lib/libcrypto/Makefile.inc +++ b/secure/lib/libcrypto/Makefile.inc @@ -1,6 +1,6 @@ .include <bsd.own.mk> -PACKAGE= openssl +PACKAGE?= openssl LIB_PACKAGE= LCRYPTO_SRC= ${SRCTOP}/crypto/openssl diff --git a/share/doc/llvm/Makefile b/share/doc/llvm/Makefile index 8fde27acf117..35398786a9cb 100644 --- a/share/doc/llvm/Makefile +++ b/share/doc/llvm/Makefile @@ -1,5 +1,7 @@ SUBDIR= clang +PACKAGE= clang + SRCDIR= ${SRCTOP}/contrib/llvm-project/llvm .PATH: ${SRCDIR} ${SRCDIR}/lib/Support diff --git a/share/doc/llvm/Makefile.inc b/share/doc/llvm/Makefile.inc new file mode 100644 index 000000000000..915410fe46c4 --- /dev/null +++ b/share/doc/llvm/Makefile.inc @@ -0,0 +1 @@ +PACKAGE?= clang diff --git a/share/man/man4/Makefile b/share/man/man4/Makefile index fe744776d9b3..34edf6ad455d 100644 --- a/share/man/man4/Makefile +++ b/share/man/man4/Makefile @@ -1005,6 +1005,7 @@ _ccd.4= ccd.4 .if ${MK_CDDL} != "no" _dtrace_provs= dtrace_audit.4 \ + dtrace_callout_execute.4 \ dtrace_dtrace.4 \ dtrace_fbt.4 \ dtrace_io.4 \ @@ -1017,7 +1018,8 @@ _dtrace_provs= dtrace_audit.4 \ dtrace_sctp.4 \ dtrace_tcp.4 \ dtrace_udp.4 \ - dtrace_udplite.4 + dtrace_udplite.4 \ + dtrace_vfs.4 MLINKS+= dtrace_audit.4 dtaudit.4 .endif diff --git a/share/man/man4/ciss.4 b/share/man/man4/ciss.4 index 28d6556ecd85..d731aaddad38 100644 --- a/share/man/man4/ciss.4 +++ b/share/man/man4/ciss.4 @@ -1,7 +1,7 @@ .\" Written by Tom Rhodes .\" This file is in the public domain. .\" -.Dd January 26, 2012 +.Dd November 6, 2025 .Dt CISS 4 .Os .Sh NAME @@ -87,9 +87,10 @@ might be solved by updating the firmware and/or setting the .Va hw.ciss.nop_message_heartbeat tunable to non-zero at boot time. .Sh HARDWARE -Controllers supported by the +The .Nm -driver include: +driver supports controllers implementing +Common Interface for SCSI-3 Support Open Specification v1.04, including: .Pp .Bl -bullet -compact .It @@ -145,6 +146,8 @@ HP Smart Array P430i .It HP Smart Array P431 .It +HP Smart Array P440ar +.It HP Smart Array P530 .It HP Smart Array P531 diff --git a/share/man/man4/ddb.4 b/share/man/man4/ddb.4 index 3c4894c03d62..a882a5204fb2 100644 --- a/share/man/man4/ddb.4 +++ b/share/man/man4/ddb.4 @@ -24,7 +24,7 @@ .\" any improvements or extensions that they make and grant Carnegie Mellon .\" the rights to redistribute these changes. .\" -.Dd June 10, 2025 +.Dd October 31, 2025 .Dt DDB 4 .Os .Sh NAME @@ -604,12 +604,15 @@ The modifier will print command line arguments for each process. .\" .Pp -.It Ic show Cm all tcpcbs Ns Op Li / Ns Cm b Ns Cm l +.It Ic show Cm all tcpcbs Ns Op Li / Ns Cm b Ns Cm i Ns Cm l Show the same output as "show tcpcb" does, but for all TCP control blocks within the system. The .Cm b modifier will request BBLog entries to be printed. +If the +.Cm i +modifier is provided, the corresponding IP control block is also shown. Using the .Cm l modifier will limit the output to TCP control blocks, which are locked. @@ -1106,7 +1109,7 @@ on i386.) Not present on some platforms. .\" .Pp -.It Ic show Cm tcpcb Ns Oo Li / Ns Cm b Oc Ar addr +.It Ic show Cm tcpcb Ns Oo Li / Ns Cm b Ns Cm i Oc Ar addr Print TCP control block .Vt struct tcpcb lying at address @@ -1117,6 +1120,9 @@ header file. The .Cm b modifier will request BBLog entries to be printed. +If the +.Cm i +modifier is provided, the corresponding IP control block is also shown. .\" .Pp .It Ic show Cm thread Op Ar addr | tid diff --git a/share/man/man4/dtrace_callout_execute.4 b/share/man/man4/dtrace_callout_execute.4 new file mode 100644 index 000000000000..1154ed066b97 --- /dev/null +++ b/share/man/man4/dtrace_callout_execute.4 @@ -0,0 +1,68 @@ +.\" +.\" Copyright (c) 2025 Mateusz Piotrowski <0mp@FreeBSD.org> +.\" +.\" SPDX-License-Identifier: BSD-2-Clause +.\" +.Dd November 4, 2025 +.Dt DTRACE_CALLOUT_EXECUTE 4 +.Os +.Sh NAME +.Nm dtrace_callout_execute +.Nd a DTrace provider for the callout API +.Sh SYNOPSIS +.Nm callout_execute Ns Cm :kernel::callout_start +.Nm callout_execute Ns Cm :kernel::callout_end +.Sh DESCRIPTION +The +.Nm callout_execute +provider allows for tracing the +.Xr callout 9 +mechanism. +.Pp +The +.Nm callout_execute Ns Cm :kernel::callout_start +probe fires just before a callout. +.Pp +The +.Nm callout_execute Ns Cm :kernel::callout_end +probe fires right after a callout. +.Pp +The only argument to the +.Nm callout_execute +probes, +.Fa args[0] , +is a callout handler +.Ft struct callout * +of the invoked callout. +.Sh EXAMPLES +.Ss Example 1: Graph of Callout Execution Time +The following +.Xr d 7 +script generates a distribution graph of +.Xr callout 9 +execution times: +.Bd -literal -offset 2n +callout_execute:::callout_start +{ + self->cstart = timestamp; +} + +callout_execute:::callout_end +{ + @length = quantize(timestamp - self->cstart); +} +.Ed +.Sh SEE ALSO +.Xr dtrace 1 , +.Xr tracing 7 , +.Xr callout 9 , +.Xr SDT 9 +.Sh AUTHORS +.An -nosplit +The +.Nm callout_execute +provider was written by +.An Robert N. M. Watson Aq Mt rwatson@FreeBSD.org . +.Pp +This manual page was written by +.An Mateusz Piotrowski Aq Mt 0mp@FreeBSD.org . diff --git a/share/man/man4/dtrace_vfs.4 b/share/man/man4/dtrace_vfs.4 new file mode 100644 index 000000000000..528d5da42f3d --- /dev/null +++ b/share/man/man4/dtrace_vfs.4 @@ -0,0 +1,97 @@ +.\" +.\" Copyright (c) 2025 Mateusz Piotrowski <0mp@FreeBSD.org> +.\" +.\" SPDX-License-Identifier: BSD-2-Clause +.\" +.Dd November 3, 2025 +.Dt DTRACE_VFS 4 +.Os +.Sh NAME +.Nm dtrace_vfs +.Nd a DTrace provider for Virtual File System +.Sh SYNOPSIS +.Sm off +.Nm vfs Cm : fplookup : Ar function Cm : Ar name +.Nm vfs Cm : namecache : Ar function Cm : Ar name +.Nm vfs Cm : namei : Ar function Cm : Ar name +.Nm vfs Cm : vop : Ar function Cm : Ar name +.Sm on +.Sh DESCRIPTION +The DTrace +.Nm vfs +provider allows users to trace events in the +.Xr VFS 9 +layer, the kernel interface for file systems on +.Fx . +.Pp +Run +.Ql dtrace -l -P vfs +to list all +.Nm vfs +probes. +Add +.Fl v +to generate program stability reports, +which contain information about the number of probe arguments and their types. +.Pp +The +.Cm fplookup +module defines a single probe, +.Fn vfs:fplookup:lookup:done "struct nameidata *ndp" "int line" "bool status_code" , +that instruments the fast path lookup code in +.Xr VFS 9 . +.Pp +The +.Cm namecache +module provides probes related to the +.Xr VFS 9 +cache. +Consult the source code in +.Pa src/sys/kern/vfs_cache.c +for more details. +.Pp +The +.Cm namei +module manages probes related to pathname translation and lookup operations. +Refer to +.Xr namei 9 +to learn more. +.Pp +The +.Cm vop +module contains probes related to the functions responsible for +.Xr vnode 9 +operations. +.Sh COMPATIBILITY +This provider is specific to +.Fx . +.Sh EXAMPLES +Check what lookups failed to be handled in a lockless manner: +.Bd -literal -offset 2n +# dtrace -n 'vfs:fplookup:lookup:done { @[arg1, arg2] = count(); }' +.Ed +.Sh SEE ALSO +.Xr dtrace 1 , +.Xr d 7 , +.Xr SDT 9 , +.Xr namei 9 , +.Xr VFS 9 +.Rs +.%A Brendan Gregg +.%A Jim Mauro +.%B DTrace: Dynamic Tracing in Oracle Solaris, Mac OS X and FreeBSD +.%I Prentice Hall +.%P pp. 335\(en351 +.%D 2011 +.%U https://www.brendangregg.com/dtracebook/ +.Re +.Sh AUTHORS +.An -nosplit +The +.Fx +.Nm vfs +provider was written by +.An Robert Watson Aq Mt rwatson@FreeBSD.org . +.Pp +This manual page was written by +.An Mateusz Piotrowski Aq Mt 0mp@FreeBSD.org . diff --git a/share/man/man4/ice.4 b/share/man/man4/ice.4 index c7675e627726..a54a6b3fd6f3 100644 --- a/share/man/man4/ice.4 +++ b/share/man/man4/ice.4 @@ -32,12 +32,12 @@ .\" .\" * Other names and brands may be claimed as the property of others. .\" -.Dd October 3, 2025 +.Dd November 5, 2025 .Dt ICE 4 .Os .Sh NAME .Nm ice -.Nd Intel Ethernet 800 Series Driver +.Nd Intel Ethernet 800 Series 1GbE to 200GbE driver .Sh SYNOPSIS .Cd device iflib .Cd device ice @@ -62,42 +62,75 @@ or .Cd dev.ice.#.pba_number .Cd dev.ice.#.hw.mac.* .Sh DESCRIPTION -.Ss Features The .Nm driver provides support for any PCI Express adapter or LOM -(LAN On Motherboard) -in the Intel\(rg Ethernet 800 Series. -As of this writing, the series includes devices with these model numbers: +.Pq LAN On Motherboard +in the Intel Ethernet 800 Series. +.Pp +The following topics are covered in this manual: .Pp .Bl -bullet -compact .It -Intel\(rg Ethernet Controller E810\-C +.Sx Features +.It +.Sx Dynamic Device Personalization +.It +.Sx Jumbo Frames +.It +.Sx Remote Direct Memory Access +.It +.Sx RDMA Monitoring +.It +.Sx Data Center Bridging +.It +.Sx L3 QoS Mode +.It +.Sx Firmware Link Layer Discovery Protocol Agent +.It +.Sx Link-Level Flow Control +.It +.Sx Forward Error Correction +.It +.Sx Speed and Duplex Configuration +.It +.Sx Disabling physical link when the interface is brought down +.It +.Sx Firmware Logging +.It +.Sx Debug Dump +.It +.Sx Debugging PHY Statistics .It -Intel\(rg Ethernet Controller E810\-XXV +.Sx Transmit Balancing .It -Intel\(rg Ethernet Connection E822\-C +.Sx Thermal Monitoring .It -Intel\(rg Ethernet Connection E822\-L +.Sx Network Memory Buffer Allocation .It -Intel\(rg Ethernet Connection E823\-C +.Sx Additional Utilities .It -Intel\(rg Ethernet Connection E823\-L +.Sx Optics and auto-negotiation .It -Intel\(rg Ethernet Connection E825\-C +.Sx PCI-Express Slot Bandwidth .It -Intel\(rg Ethernet Connection E830\-C +.Sx HARDWARE .It -Intel\(rg Ethernet Connection E830\-CC +.Sx LOADER TUNABLES .It -Intel\(rg Ethernet Connection E830\-L +.Sx SYSCTL VARIABLES .It -Intel\(rg Ethernet Connection E830\-XXV +.Sx INTERRUPT STORMS +.It +.Sx IOVCTL OPTIONS +.It +.Sx SUPPORT +.It +.Sx SEE ALSO +.It +.Sx HISTORY .El -.Pp -For questions related to hardware requirements, refer to the documentation -supplied with the adapter. -.Pp +.Ss Features Support for Jumbo Frames is provided via the interface MTU setting. Selecting an MTU larger than 1500 bytes with the .Xr ifconfig 8 @@ -141,7 +174,7 @@ downloading a new driver or DDP package. Safe Mode only applies to the affected physical function and does not impact any other PFs. See the -.Dq Intel\(rg Ethernet Adapters and Devices User Guide +.Dq Intel Ethernet Adapters and Devices User Guide for more details on DDP and Safe Mode. .Pp If issues are encountered with the DDP package file, an updated driver or @@ -153,8 +186,8 @@ The DDP package cannot be updated if any PF drivers are already loaded. To overwrite a package, unload all PFs and then reload the driver with the new package. .Pp -Only one DDP package can be used per driver, even if more than one -device installed that uses the driver. +Only one DDP package can be used per driver, +even if more than one installed device uses the driver. .Pp Only the first loaded PF per device can download a package for that device. .Ss Jumbo Frames @@ -187,7 +220,7 @@ RoCEv2 (RDMA over Converged Ethernet) protocols. The major difference is that iWARP performs RDMA over TCP, while RoCEv2 uses UDP. .Pp -Devices based on the Intel\(rg Ethernet 800 Series do not support RDMA when +Devices based on the Intel Ethernet 800 Series do not support RDMA when operating in multiport mode with more than 4 ports. .Pp For detailed installation and configuration information for RDMA, see @@ -200,7 +233,7 @@ analysis tools like .Xr tcpdump 1 . This mirroring may impact performance. .Pp -To use RDMA monitoring, more MSI\-X interrupts may need to be reserved. +To use RDMA monitoring, more MSI-X interrupts may need to be reserved. Before the .Nm driver loads, configure the following tunable provided by @@ -209,7 +242,7 @@ driver loads, configure the following tunable provided by dev.ice.<interface #>.iflib.use_extra_msix_vectors=4 .Ed .Pp -The number of extra MSI\-X interrupt vectors may need to be adjusted. +The number of extra MSI-X interrupt vectors may need to be adjusted. .Pp To create/delete the interface: .Bd -literal -offset indent @@ -245,14 +278,15 @@ DCB is normally configured on the network using the DCBX protocol (802.1Qaz), a specialization of LLDP (802.1AB). The .Nm driver supports the following mutually exclusive variants of DCBX support: +.Pp .Bl -bullet -compact .It -Firmware\-based LLDP Agent +Firmware-based LLDP Agent .It -Software\-based LLDP Agent +Software-based LLDP Agent .El .Pp -In firmware\-based mode, firmware intercepts all LLDP traffic and handles DCBX +In firmware-based mode, firmware intercepts all LLDP traffic and handles DCBX negotiation transparently for the user. In this mode, the adapter operates in .Dq willing @@ -262,25 +296,25 @@ The local user can only query the negotiated DCB configuration. For information on configuring DCBX parameters on a switch, please consult the switch manufacturer'ss documentation. .Pp -In software\-based mode, LLDP traffic is forwarded to the network stack and user +In software-based mode, LLDP traffic is forwarded to the network stack and user space, where a software agent can handle it. In this mode, the adapter can operate in .Dq nonwilling DCBX mode and DCB configuration can be both queried and set locally. -This mode requires the FW\-based LLDP Agent to be disabled. +This mode requires the FW-based LLDP Agent to be disabled. .Pp -Firmware\-based mode and software\-based mode are controlled by the +Firmware-based mode and software-based mode are controlled by the .Dq fw_lldp_agent sysctl. Refer to the Firmware Link Layer Discovery Protocol Agent section for more information. .Pp -Link\-level flow control and priority flow control are mutually exclusive. +Link-level flow control and priority flow control are mutually exclusive. The ice driver will disable link flow control when priority flow control is enabled on any traffic class (TC). It will disable priority flow control when link flow control is enabled. .Pp -To enable/disable priority flow control in software\-based DCBX mode: +To enable/disable priority flow control in software-based DCBX mode: .Bd -literal -offset indent sysctl dev.ice.<interface #>.pfc=1 (or 0 to disable) .Ed @@ -307,10 +341,10 @@ For example, to map UP 0 and 1 to TC 0, UP 2 and 3 to TC 1, UP 4 and .Bd -literal -offset indent sysctl dev.ice.<interface #>.up2tc_map=0,0,1,1,2,2,3,3 .Ed -.Ss L3 QoS mode +.Ss L3 QoS Mode The .Nm -driver supports setting DSCP\-based Layer 3 Quality of Service (L3 QoS) +driver supports setting DSCP-based Layer 3 Quality of Service (L3 QoS) in the PF driver. The driver initializes in L2 QoS mode by default; L3 QoS is disabled by default. @@ -319,13 +353,13 @@ Use the following sysctl to enable or disable L3 QoS: sysctl dev.ice.<interface #>.pfc_mode=1 (or 0 to disable) .Ed .Pp -If the L3 QoS mode is disabled, it returns to L2 QoS mode. +If L3 QoS mode is disabled, it returns to L2 QoS mode. .Pp To map a DSCP value to a traffic class, separate the values by commas. -For example, to map DSCPs 0\-3 and DSCP 8 to DCB TCs 0\-3 and 4, respectively: +For example, to map DSCPs 0-3 and DSCP 8 to DCB TCs 0-3 and 4, respectively: .Bd -literal -offset indent -sysctl dev.ice.<interface #>.dscp2tc_map.0\-7=0,1,2,3,0,0,0,0 -sysctl dev.ice.<interface #>.dscp2tc_map.8\-15=4,0,0,0,0,0,0,0 +sysctl dev.ice.<interface #>.dscp2tc_map.0-7=0,1,2,3,0,0,0,0 +sysctl dev.ice.<interface #>.dscp2tc_map.8-15=4,0,0,0,0,0,0,0 .Ed .Pp To change the DSCP mapping back to the default traffic class, set all the @@ -336,25 +370,25 @@ To view the currently configured mappings, use the following: sysctl dev.ice.<interface #>.dscp2tc_map .Ed .Pp -L3 QoS mode is not available when FW\-LLDP is enabled. +L3 QoS mode is not available when FW-LLDP is enabled. .Pp -FW\-LLDP cannot be enabled if L3 QoS mode is active. +FW-LLDP cannot be enabled if L3 QoS mode is active. .Pp -Disable FW\-LLDP before switching to L3 QoS mode. +Disable FW-LLDP before switching to L3 QoS mode. .Pp Refer to the .Sx Firmware Link Layer Discovery Protocol Agent -section in this README for more information on disabling FW\-LLDP. +section in this README for more information on disabling FW-LLDP. .Ss Firmware Link Layer Discovery Protocol Agent -Use sysctl to change FW\-LLDP settings. -The FW\-LLDP setting is per port and persists across boots. +Use sysctl to change FW-LLDP settings. +The FW-LLDP setting is per port and persists across boots. .Pp -To enable the FW\-LLDP Agent: +To enable the FW-LLDP Agent: .Bd -literal -offset indent sysctl dev.ice.<interface #>.fw_lldp_agent=1 .Ed .Pp -To disable the FW\-LLDP Agebt: +To disable the FW-LLDP Agebt: .Bd -literal -offset indent sysctl dev.ice.<interface #>.fw_lldp_agent=0 .Ed @@ -368,11 +402,14 @@ The UEFI HII LLDP Agent attribute must be enabled for this setting to take effect. If the .Dq LLDP AGENT -attribute is set to disabled, the FW\-LLDP Agent cannot be enabled from the +attribute is set to disabled, the FW-LLDP Agent cannot be enabled from the driver. -.Ss Link\-Level Flow Control (LFC) -Ethernet Flow Control (IEEE 802.3x) can be configured with sysctl to enable -receiving and transmitting pause frames for +.Ss Link-Level Flow Control +Ethernet Flow Control +.Pq IEEE 802.3x or LFC +can be configured with +.Xr sysctl 8 +to enable receiving and transmitting pause frames for .Nm . When transmit is enabled, pause frames are generated when the receive packet buffer crosses a predefined threshold. @@ -434,7 +471,7 @@ in case the link partner does not have FEC enabled or is not FEC capable: sysctl dev.ice.<interface #>.allow_no_fec_modules_in_auto=1 .Ed .Pp -NOTE: This flag is currently not supported on the Intel\(rg Ethernet 830 +NOTE: This flag is currently not supported on the Intel Ethernet 830 Series. .Pp To show the current FEC settings that are negotiated on the link: @@ -449,7 +486,7 @@ sysctl dev.ice.<interface #>.requested_fec .Pp To see the valid FEC modes for the link: .Bd -literal -offset indent -sysctl \-d dev.ice.<interface #>.requested_fec +sysctl -d dev.ice.<interface #>.requested_fec .Ed .Ss Speed and Duplex Configuration The speed and duplex settings cannot be hard set. @@ -464,17 +501,17 @@ Supported speeds will vary by device. Depending on the speeds the device supports, valid bits used in a speed mask could include: .Bd -literal -offset indent -0x0 \- Auto -0x2 \- 100 Mbps -0x4 \- 1 Gbps -0x8 \- 2.5 Gbps -0x10 \- 5 Gbps -0x20 \- 10 Gbps -0x80 \- 25 Gbps -0x100 \- 40 Gbps -0x200 \- 50 Gbps -0x400 \- 100 Gbps -0x800 \- 200 Gbps +0x0 - Auto +0x2 - 100 Mbps +0x4 - 1 Gbps +0x8 - 2.5 Gbps +0x10 - 5 Gbps +0x20 - 10 Gbps +0x80 - 25 Gbps +0x100 - 40 Gbps +0x200 - 50 Gbps +0x400 - 100 Gbps +0x800 - 200 Gbps .Ed .Ss Disabling physical link when the interface is brought down When the @@ -494,7 +531,7 @@ The driver allows for the generation of firmware logs for supported categories of events, to help debug issues with Customer Support. Refer to the -.Dq Intel\(rg Ethernet Adapters and Devices User Guide +.Dq Intel Ethernet Adapters and Devices User Guide for an overview of this feature and additional tips. .Pp At a high level, to capture a firmware log: @@ -553,7 +590,7 @@ DCBx (Bit 11) .It Va dcb DCB (Bit 12) .It Va xlr -XLR (function\-level resets; Bit 13) +XLR (function-level resets; Bit 13) .It Va nvm NVM (Bit 14) .It Va auth @@ -561,7 +598,7 @@ Authentication (Bit 15) .It Va vpd Vital Product Data (Bit 16) .It Va iosf -Intel On\-Chip System Fabric (Bit 17) +Intel On-Chip System Fabric (Bit 17) .It Va parser Parser (Bit 18) .It Va sw @@ -649,8 +686,8 @@ dmesg > log_output NOTE: Logging a large number of modules or too high of a verbosity level will add extraneous messages to dmesg and could hinder debug efforts. .Ss Debug Dump -Intel\(rg Ethernet 800 Series devices support debug dump, which allows -gathering of runtime register values from the firmware for +Intel Ethernet 800 Series devices support debug dump, +which allows gathering of runtime register values from the firmware for .Dq clusters of events and then write the results to a single dump file, for debugging complicated issues in the field. @@ -662,7 +699,7 @@ Debug dump captures the current state of the specified cluster(s) and is a stateless snapshot of the whole device. .Pp NOTE: Like with firmware logs, the contents of the debug dump are not -human\-readable. +human-readable. Work with Customer Support to decode the file. .Pp Debug dump is per device, not per PF. @@ -685,7 +722,7 @@ pass the argument. For example: .Bd -literal -offset indent -sysctl \-d dev.ice.0.debug.dump.clusters +sysctl -d dev.ice.0.debug.dump.clusters .Ed .Pp Possible bitmask values for @@ -693,24 +730,24 @@ Possible bitmask values for are: .Bl -bullet -compact .It -0 \- Dump all clusters (only supported on Intel\(rg Ethernet E810 Series and -Intel\(rg Ethernet E830 Series) +0 - Dump all clusters (only supported on Intel Ethernet E810 Series and +Intel Ethernet E830 Series) .It -0x1 \- Switch +0x1 - Switch .It -0x2 \- ACL +0x2 - ACL .It -0x4 \- Tx Scheduler +0x4 - Tx Scheduler .It -0x8 \- Profile Configuration +0x8 - Profile Configuration .It -0x20 \- Link +0x20 - Link .It -0x80 \- DCB +0x80 - DCB .It -0x100 \- L2P +0x100 - L2P .It -0x400000 \- Manageability Transactions (only supported on Intel\(rg Ethernet +0x400000 - Manageability Transactions (only supported on Intel Ethernet E810 Series) .El .Pp @@ -726,11 +763,11 @@ sysctl dev.ice.0.debug.dump.clusters=0 .Pp NOTE: Using 0 will skip Manageability Transactions data. .Pp -If a single cluster is not specified, the driver will dump all clusters to a -single file. +If a single cluster is not specified, +the driver will dump all clusters to a single file. Issue the debug dump command, using the following: .Bd -literal -offset indent -sysctl \-b dev.ice.<interface #>.debug.dump.dump=1 > dump.bin +sysctl -b dev.ice.<interface #>.debug.dump.dump=1 > dump.bin .Ed .Pp NOTE: The driver will not receive the command if the sysctl is not set to @@ -765,13 +802,13 @@ Use the following sysctl to read the PHY registers: sysctl dev.ice.<interface #>.debug.phy_statistics .Ed .Pp -NOTE: The contents of the registers are not human\-readable. +NOTE: The contents of the registers are not human-readable. Like with firmware logs and debug dump, work with Customer Support to decode the file. .Ss Transmit Balancing Some Intel(R) Ethernet 800 Series devices allow for enabling a transmit balancing feature to improve transmit performance under certain conditions. -When enabled, the feature should provide more consistent transmit +When enabled, this feature should provide more consistent transmit performance across queues and/or PFs and VFs. .Pp By default, transmit balancing is disabled in the NVM. @@ -809,7 +846,7 @@ sysctl dev.ice.<interface #>.temp may have a low number of network memory buffers (mbufs) by default. If the number of mbufs available is too low, it may cause the driver to fail to initialize and/or cause the system to become unresponsive. -Check to see if the system is mbuf\-starved by running +Check to see if the system is mbuf-starved by running .Ic netstat Fl m . Increase the number of mbufs by editing the lines below in .Pa /etc/sysctl.conf : @@ -821,8 +858,8 @@ kern.ipc.nmbjumbo16 kern.ipc.nmbufs .Ed .Pp -The amount of memory that should be allocated is system specific, and may require some -trial and error. +The amount of memory that should be allocated is system specific, +and may require some trial and error. Also, increasing the following in .Pa /etc/sysctl.conf could help increase network performance: @@ -847,13 +884,91 @@ To change the behavior of the QSFP28 ports on E810-C adapters, use the Intel To update the firmware on an adapter, use the Intel .Sy Non-Volatile Memory (NVM) Update Utility for Intel Ethernet Network Adapters E810 series - FreeBSD .El +.Ss Optics and auto-negotiation +Modules based on 100GBASE-SR4, +active optical cable (AOC), and active copper cable (ACC) +do not support auto-negotiation per the IEEE specification. +To obtain link with these modules, +auto-negotiation must be turned off on the link partner's switch ports. +.Pp +Note that adapters also support +all passive and active limiting direct attach cables +that comply with SFF-8431 v4.1 and SFF-8472 v10.4 specifications. +.Ss PCI-Express Slot Bandwidth +Some PCIe x8 slots are actually configured as x4 slots. +These slots have insufficient bandwidth +for full line rate with dual port and quad port devices. +In addition, +if a PCIe v4.0 or v3.0-capable adapter is placed into into a PCIe v2.x +slot, full bandwidth will not be possible. +.Pp +The driver detects this situation and +writes the following message in the system log: +.Bd -ragged -offset indent +PCI-Express bandwidth available for this device +may be insufficient for optimal performance. +Please move the device to a different PCI-e link +with more lanes and/or higher transfer rate. +.Ed +.Pp +If this error occurs, +moving the adapter to a true PCIe x8 or x16 slot will resolve the issue. +For best performance, install devices in the following PCI slots: +.Bl -bullet +.It +Any 100Gbps-capable Intel(R) Ethernet 800 Series device: Install in a +PCIe v4.0 x8 or v3.0 x16 slot +.It +A 200Gbps-capable Intel(R) Ethernet 830 Series device: Install in a +PCIe v5.0 x8 or v4.0 x16 slot +.El +.Pp +For questions related to hardware requirements, +refer to the documentation supplied with the adapter. .Sh HARDWARE The .Nm -driver supports the Intel Ethernet 800 series. -Some adapters in this series with SFP28/QSFP28 cages -have firmware that requires that Intel qualified modules are used; these -qualified modules are listed below. +driver supports the following +Intel 800 series 1Gb to 200Gb Ethernet controllers: +.Pp +.Bl -bullet -compact +.It +Intel Ethernet Controller E810-C +.It +Intel Ethernet Controller E810-XXV +.It +Intel Ethernet Connection E822-C +.It +Intel Ethernet Connection E822-L +.It +Intel Ethernet Connection E823-C +.It +Intel Ethernet Connection E823-L +.It +Intel Ethernet Connection E825-C +.It +Intel Ethernet Connection E830-C +.It +Intel Ethernet Connection E830-CC +.It +Intel Ethernet Connection E830-L +.It +Intel Ethernet Connection E830-XXV +.It +Intel Ethernet Connection E835-C +.It +Intel Ethernet Connection E835-CC +.It +Intel Ethernet Connection E835-L +.It +Intel Ethernet Connection E835-XXV +.El +.Pp +The +.Nm +driver supports some adapters in this series with SFP28/QSFP28 cages +which have firmware that requires that Intel qualified modules are used; +these qualified modules are listed below. This qualification check cannot be disabled by the driver. .Pp The @@ -862,13 +977,13 @@ driver supports 100Gb Ethernet adapters with these QSFP28 modules: .Pp .Bl -bullet -compact .It -Intel\(rg 100G QSFP28 100GBASE-SR4 E100GQSFPSR28SRX +Intel 100G QSFP28 100GBASE-SR4 E100GQSFPSR28SRX .It -Intel\(rg 100G QSFP28 100GBASE-SR4 SPTMBP1PMCDF +Intel 100G QSFP28 100GBASE-SR4 SPTMBP1PMCDF .It -Intel\(rg 100G QSFP28 100GBASE-CWDM4 SPTSBP3CLCCO +Intel 100G QSFP28 100GBASE-CWDM4 SPTSBP3CLCCO .It -Intel\(rg 100G QSFP28 100GBASE-DR SPTSLP2SLCDF +Intel 100G QSFP28 100GBASE-DR SPTSLP2SLCDF .El .Pp The @@ -877,11 +992,11 @@ driver supports 25Gb and 10Gb Ethernet adapters with these SFP28 modules: .Pp .Bl -bullet -compact .It -Intel\(rg 10G/25G SFP28 25GBASE-SR E25GSFP28SR +Intel 10G/25G SFP28 25GBASE-SR E25GSFP28SR .It -Intel\(rg 25G SFP28 25GBASE-SR E25GSFP28SRX (Extended Temp) +Intel 25G SFP28 25GBASE-SR E25GSFP28SRX (Extended Temp) .It -Intel\(rg 25G SFP28 25GBASE-LR E25GSFP28LRX (Extended Temp) +Intel 25G SFP28 25GBASE-LR E25GSFP28LRX (Extended Temp) .El .Pp The @@ -890,54 +1005,15 @@ driver supports 10Gb and 1Gb Ethernet adapters with these SFP+ modules: .Pp .Bl -bullet -compact .It -Intel\(rg 1G/10G SFP+ 10GBASE-SR E10GSFPSR +Intel 1G/10G SFP+ 10GBASE-SR E10GSFPSR .It -Intel\(rg 1G/10G SFP+ 10GBASE-SR E10GSFPSRG1P5 +Intel 1G/10G SFP+ 10GBASE-SR E10GSFPSRG1P5 .It -Intel\(rg 1G/10G SFP+ 10GBASE-SR E10GSFPSRG2P5 +Intel 1G/10G SFP+ 10GBASE-SR E10GSFPSRG2P5 .It -Intel\(rg 10G SFP+ 10GBASE-SR E10GSFPSRX (Extended Temp) +Intel 10G SFP+ 10GBASE-SR E10GSFPSRX (Extended Temp) .It -Intel\(rg 1G/10G SFP+ 10GBASE-LR E10GSFPLR -.El -.Pp -Note that adapters also support all passive and active -limiting direct attach cables that comply with SFF-8431 v4.1 and -SFF-8472 v10.4 specifications. -.Pp -This is not an exhaustive list; please consult product documentation for an -up-to-date list of supported media. -.Ss Fiber optics and auto\-negotiation -Modules based on 100GBASE\-SR4, active optical cable (AOC), and active copper -cable (ACC) do not support auto\-negotiation per the IEEE specification. -To obtain link with these modules, auto\-negotiation must be turned off on the -link partner's switch ports. -.Ss PCI-Express Slot Bandwidth -Some PCIe x8 slots are actually configured as x4 slots. -These slots have insufficient bandwidth for full line rate with dual port and -quad port devices. -In addition, if a PCIe v4.0 or v3.0\-capable adapter is placed into a PCIe v2.x -slot, full bandwidth will not be possible. -.Pp -The driver detects this situation and writes the following message in the -system log: -.Bd -literal -offset indent -PCI\-Express bandwidth available for this device may be insufficient for -optimal performance. -Please move the device to a different PCI\-e link with more lanes and/or -higher transfer rate. -.Ed -.Pp -If this error occurs, moving the adapter to a true PCIe x8 or x16 slot will -resolve the issue. -For best performance, install devices in the following PCI slots: -.Bl -bullet -.It -Any 100Gbps\-capable Intel(R) Ethernet 800 Series device: Install in a -PCIe v4.0 x8 or v3.0 x16 slot -.It -A 200Gbps\-capable Intel(R) Ethernet 830 Series device: Install in a -PCIe v5.0 x8 or v4.0 x16 slot +Intel 1G/10G SFP+ 10GBASE-LR E10GSFPLR .El .Sh LOADER TUNABLES Tunables can be set at the @@ -1035,11 +1111,11 @@ on. Disabled by default. .It num-queues Pq uint16_t Specify the number of queues the VF will have. -By default, this is set to the number of MSI\-X vectors supported by the VF +By default, this is set to the number of MSI-X vectors supported by the VF minus one. .It mirror-src-vsi Pq uint16_t Specify which VSI the VF will mirror traffic from by setting this to a value -other than \-1. +other than -1. All traffic from that VSI will be mirrored to this VF. Can be used as an alternative method to mirror RDMA traffic to another interface than the method described in the diff --git a/share/man/man4/mpr.4 b/share/man/man4/mpr.4 index cce21113e5c2..8de46e4f9272 100644 --- a/share/man/man4/mpr.4 +++ b/share/man/man4/mpr.4 @@ -37,7 +37,7 @@ .\" .\" $Id$ .\" -.Dd June 1, 2019 +.Dd September 28, 2025 .Dt MPR 4 .Os .Sh NAME @@ -99,8 +99,12 @@ Broadcom Ltd./Avago Tech (LSI) SAS 3708 (8 Port SAS/PCIe) .It Broadcom Ltd./Avago Tech (LSI) SAS 3716 (16 Port SAS/PCIe) .It +Broadcom Ltd./Avago Tech (LSI) SAS 3808 (8 Port SAS/PCIe) +.It Broadcom Ltd./Avago Tech (LSI) SAS 3816 (16 Port SAS/PCIe) .It +Broadcom Ltd./Avago Tech (LSI) SAS 3908 (8 Port SAS/PCIe) +.It Broadcom Ltd./Avago Tech (LSI) SAS 3916 (16 Port SAS/PCIe) .El .Sh CONFIGURATION diff --git a/share/man/man4/u3g.4 b/share/man/man4/u3g.4 index 51c883b8378e..c7bf1f98880a 100644 --- a/share/man/man4/u3g.4 +++ b/share/man/man4/u3g.4 @@ -94,14 +94,14 @@ Qualcomm Inc. GOBI 1000, 2000 and 3000 devices with MDM1000 or MDM2000 chipsets .It QUECTEL BGX, ECX, EGX, EMX, EPX, RGX series .It -Quectel EM160R +Quectel EM160R, EM060K .Pq see Sx CAVEATS .It Huawei B190, E180v, E220, E3372, E3372v153, E5573Cs322, ('<Huawei Mobile>') .It Novatel U740, MC950D, X950D, etc. .It -Sierra MC875U, MC8775U, etc. +Sierra MC875U, MC8775U, EM7590, etc. .It Panasonic CF-F9 GOBI .El diff --git a/share/man/man5/pf.conf.5 b/share/man/man5/pf.conf.5 index be46b1a47291..c22d983d33e8 100644 --- a/share/man/man5/pf.conf.5 +++ b/share/man/man5/pf.conf.5 @@ -27,7 +27,7 @@ .\" ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE .\" POSSIBILITY OF SUCH DAMAGE. .\" -.Dd October 7, 2025 +.Dd November 3, 2025 .Dt PF.CONF 5 .Os .Sh NAME @@ -3460,6 +3460,12 @@ filteropt = user | group | flags | icmp-type | icmp6-type | "tos" tos | "dnpipe" ( number | "(" number "," number ")" ) | "dnqueue" ( number | "(" number "," number ")" ) | "ridentifier" number | + "binat-to" ( redirhost | "{" redirhost-list "}" ) + [ portspec ] [ pooltype ] | + "rdr-to" ( redirhost | "{" redirhost-list "}" ) + [ portspec ] [ pooltype ] | + "nat-to" ( redirhost | "{" redirhost-list "}" ) + [ portspec ] [ pooltype ] [ "static-port" ] | [ ! ] "received-on" ( interface-name | interface-group ) nat-rule = [ "no" ] "nat" [ "pass" [ "log" [ "(" logopts ")" ] ] ] diff --git a/share/man/man7/d.7 b/share/man/man7/d.7 index 4b00d3d71c79..59b3389b121b 100644 --- a/share/man/man7/d.7 +++ b/share/man/man7/d.7 @@ -198,6 +198,130 @@ The number of nanoseconds since the Epoch Suitable for timestamping logs. .El .Sh BUILT-IN FUNCTIONS +.\" Keep the indentation wide enough for the reader to be able to skim through +.\" function names quickly. +.Bl -tag -width "size_t strlen" +.It Ft string Fn strchr "string s" "char c" +Return a substring of +.Fa s +starting at the first occurance of +.Fa c +in +.Fa s . +Return +.Dv NULL +if +.Fa c +does not occur in +.Fa s . +.Pp +For example, +.Bd -literal -compact -offset indent +strchr("abc", 'b'); +.Ed +returns +.Ql "bc" +and +.Bd -literal -compact -offset indent +strchr("abc", 'd'); +.Ed +returns +.Dv NULL . +.It Ft string Fn strjoin "string s1" "string s2" +Return a string resulting from concatenating +.Fa s1 +and +.Fa s2 . +.Pp +For example, +.Bd -literal -compact -offset indent +strjoin("abc", "def") +.Ed +returns +.Ql abcdef . +.It Ft string Fn strrchr "string s" "char c" +Return a substring of +.Fa s +starting at the last occurance of +.Fa c +in +.Fa s . +Similar to +.Fn strchr . +.It Ft string Fn strstr "string haystack" "string needle" +Return a substring of +.Fa haystack +starting at the first occurrence of +.Fa needle . +Return +.Dv NULL +if +.Fa needle +is not a substring of +.Fa haystack . +.Pp +For example, +.Bd -literal -compact -offset indent +strstr("abc1bc2", "bc") +.Ed +returns +.Ql bc1bc2 +and +.Bd -literal -compact -offset indent +strstr("abc", "xy") +.Ed +returns +.Dv NULL . +.It Ft string Fn strtok "string s" "string separators" +Tokenize +.Fa s +with +.Fa separators . +.Pp +For example, +.Bd -literal -compact -offset indent +strtok("abcdefg", "xyzd") +.Ed +returns +.Ql abc . +.It Ft size_t Fn strlen "string s" +Return the length of string +.Fa s . +.It Ft string Fn substr "string s" "int position" "[int length]" +Return a +substring of string +.Fa s +starting at +.Fa position . +The substring will be at most +.Fa length Ns -long . +If +.Fa length +is not specified, use the rest of the string. +If +.Fa position +is greater than +the size of +.Fa s , +return an empty string. +.Pp +For example, +.Bd -literal -compact -offset indent +substr("abcd", 2) +.Ed +returns +.Ql cd , +.Bd -literal -compact -offset indent +substr("abcd", 2, 1) +.Ed +returns +.Ql c , +and +.Bd -literal -compact -offset indent +substr("abcd", 99) +.Ed +returns an empty string. +.El .Ss Aggregation Functions .Bl -tag -compact -width "llquantize(value, factor, low, high, nsteps)" .It Fn avg value diff --git a/share/man/man9/VFS.9 b/share/man/man9/VFS.9 index a1d0a19bec13..6ea6570bbf6e 100644 --- a/share/man/man9/VFS.9 +++ b/share/man/man9/VFS.9 @@ -26,7 +26,7 @@ .\" (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF .\" THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. .\" -.Dd February 9, 2010 +.Dd November 3, 2025 .Dt VFS 9 .Os .Sh NAME @@ -42,6 +42,7 @@ function from rather than implementing empty functions or casting to .Fa eopnotsupp . .Sh SEE ALSO +.Xr dtrace_vfs 4 , .Xr VFS_CHECKEXP 9 , .Xr VFS_FHTOVP 9 , .Xr VFS_MOUNT 9 , diff --git a/share/man/man9/atomic.9 b/share/man/man9/atomic.9 index c9133c6311a5..b027a0ff0bca 100644 --- a/share/man/man9/atomic.9 +++ b/share/man/man9/atomic.9 @@ -182,35 +182,42 @@ This variant is the default. The second variant has acquire semantics, and the third variant has release semantics. .Pp -When an atomic operation has acquire semantics, the operation must have +An atomic operation can only have +.Em acquire +semantics if it performs a load +from memory. +When an atomic operation has acquire semantics, a load performed as +part of the operation must have completed before any subsequent load or store (by program order) is performed. Conversely, acquire semantics do not require that prior loads or stores have -completed before the atomic operation is performed. -An atomic operation can only have acquire semantics if it performs a load -from memory. +completed before a load from the atomic operation is performed. To denote acquire semantics, the suffix .Dq Li _acq is inserted into the function name immediately prior to the .Dq Li _ Ns Aq Fa type suffix. -For example, to subtract two integers ensuring that the subtraction is +For example, to subtract two integers ensuring that the load of +the value from memory is completed before any subsequent loads and stores are performed, use .Fn atomic_subtract_acq_int . .Pp +An atomic operation can only have +.Em release +semantics if it performs a store to memory. When an atomic operation has release semantics, all prior loads or stores -(by program order) must have completed before the operation is performed. -Conversely, release semantics do not require that the atomic operation must +(by program order) must have completed before a store executed as part of +the operation that is performed. +Conversely, release semantics do not require that a store from the atomic +operation must have completed before any subsequent load or store is performed. -An atomic operation can only have release semantics if it performs a store -to memory. To denote release semantics, the suffix .Dq Li _rel is inserted into the function name immediately prior to the .Dq Li _ Ns Aq Fa type suffix. For example, to add two long integers ensuring that all prior loads and -stores are completed before the addition is performed, use +stores are completed before the store of the result is performed, use .Fn atomic_add_rel_long . .Pp When a release operation by one thread @@ -235,6 +242,33 @@ section. However, they will not prevent the compiler or processor from moving loads or stores into the critical section, which does not violate the semantics of a mutex. +.Ss Architecture-dependent caveats for compare-and-swap +The +.Fn atomic_[f]cmpset_<type> +operations, specifically those without explicitly specified memory +ordering, are defined as relaxed. +Consequently, a thread's accesses to memory locations different from +that of the atomic operation can be reordered in relation to the +atomic operation. +.Pp +However, the implementation on the +.Sy amd64 +and +.Sy i386 +architectures provide sequentially consistent semantics. +In particular, the reordering mentioned above cannot occur. +.Pp +On the +.Sy arm64/aarch64 +architecture, the operation may include either acquire +semantics on the constituent load or release semantics +on the constituent store. +This means that accesses to other locations in program order +before the atomic, might be observed as executed after the load +that is the part of the atomic operation (but not after the store +from the operation due to release). +Similarly, accesses after the atomic might be observed as executed +before the store. .Ss Thread Fence Operations Alternatively, a programmer can use atomic thread fence operations to constrain the reordering of accesses. diff --git a/share/man/man9/buf.9 b/share/man/man9/buf.9 index ecd4a1487735..ff9a1d0d46e0 100644 --- a/share/man/man9/buf.9 +++ b/share/man/man9/buf.9 @@ -36,44 +36,70 @@ The kernel implements a KVM abstraction of the buffer cache which allows it to map potentially disparate vm_page's into contiguous KVM for use by (mainly file system) devices and device I/O. This abstraction supports -block sizes from DEV_BSIZE (usually 512) to upwards of several pages or more. +block sizes from +.Dv DEV_BSIZE +(usually 512) to upwards of several pages or more. It also supports a relatively primitive byte-granular valid range and dirty range currently hardcoded for use by NFS. The code implementing the VM Buffer abstraction is mostly concentrated in -.Pa /usr/src/sys/kern/vfs_bio.c . +.Pa sys/kern/vfs_bio.c +in the +.Fx +source tree. .Pp One of the most important things to remember when dealing with buffer pointers -(struct buf) is that the underlying pages are mapped directly from the buffer +.Pq Vt struct buf +is that the underlying pages are mapped directly from the buffer cache. No data copying occurs in the scheme proper, though some file systems such as UFS do have to copy a little when dealing with file fragments. The second most important thing to remember is that due to the underlying page -mapping, the b_data base pointer in a buf is always *page* aligned, not -*block* aligned. -When you have a VM buffer representing some b_offset and -b_size, the actual start of the buffer is (b_data + (b_offset & PAGE_MASK)) -and not just b_data. +mapping, the +.Va b_data +base pointer in a buf is always +.Em page Ns -aligned , +not +.Em block Ns -aligned . +When you have a VM buffer representing some +.Va b_offset +and +.Va b_size , +the actual start of the buffer is +.Ql b_data + (b_offset & PAGE_MASK) +and not just +.Ql b_data . Finally, the VM system's core buffer cache supports -valid and dirty bits (m->valid, m->dirty) for pages in DEV_BSIZE chunks. +valid and dirty bits +.Pq Va m->valid , m->dirty +for pages in +.Dv DEV_BSIZE +chunks. Thus a platform with a hardware page size of 4096 bytes has 8 valid and 8 dirty bits. These bits are generally set and cleared in groups based on the device block size of the device backing the page. Complete page's worth are often -referred to using the VM_PAGE_BITS_ALL bitmask (i.e., 0xFF if the hardware page +referred to using the +.Dv VM_PAGE_BITS_ALL +bitmask (i.e., 0xFF if the hardware page size is 4096). .Pp VM buffers also keep track of a byte-granular dirty range and valid range. This feature is normally only used by the NFS subsystem. I am not sure why it -is used at all, actually, since we have DEV_BSIZE valid/dirty granularity +is used at all, actually, since we have +.Dv DEV_BSIZE +valid/dirty granularity within the VM buffer. -If a buffer dirty operation creates a 'hole', +If a buffer dirty operation creates a +.Dq hole , the dirty range will extend to cover the hole. If a buffer validation -operation creates a 'hole' the byte-granular valid range is left alone and +operation creates a +.Dq hole +the byte-granular valid range is left alone and will not take into account the new extension. Thus the whole byte-granular abstraction is considered a bad hack and it would be nice if we could get rid @@ -81,16 +107,24 @@ of it completely. .Pp A VM buffer is capable of mapping the underlying VM cache pages into KVM in order to allow the kernel to directly manipulate the data associated with -the (vnode,b_offset,b_size). +the +.Pq Va vnode , b_offset , b_size . The kernel typically unmaps VM buffers the moment -they are no longer needed but often keeps the 'struct buf' structure -instantiated and even bp->b_pages array instantiated despite having unmapped +they are no longer needed but often keeps the +.Vt struct buf +structure +instantiated and even +.Va bp->b_pages +array instantiated despite having unmapped them from KVM. If a page making up a VM buffer is about to undergo I/O, the -system typically unmaps it from KVM and replaces the page in the b_pages[] +system typically unmaps it from KVM and replaces the page in the +.Va b_pages[] array with a place-marker called bogus_page. The place-marker forces any kernel -subsystems referencing the associated struct buf to re-lookup the associated +subsystems referencing the associated +.Vt struct buf +to re-lookup the associated page. I believe the place-marker hack is used to allow sophisticated devices such as file system devices to remap underlying pages in order to deal with, @@ -107,18 +141,29 @@ you wind up with pages marked clean that are actually still dirty. If not treated carefully, these pages could be thrown away! Indeed, a number of -serious bugs related to this hack were not fixed until the 2.2.8/3.0 release. -The kernel uses an instantiated VM buffer (i.e., struct buf) to place-mark pages +serious bugs related to this hack were not fixed until the +.Fx 2.2.8 / +.Fx 3.0 +release. +The kernel uses an instantiated VM buffer (i.e., +.Vt struct buf ) +to place-mark pages in this special state. -The buffer is typically flagged B_DELWRI. +The buffer is typically flagged +.Dv B_DELWRI . When a -device no longer needs a buffer it typically flags it as B_RELBUF. +device no longer needs a buffer it typically flags it as +.Dv B_RELBUF . Due to -the underlying pages being marked clean, the B_DELWRI|B_RELBUF combination must +the underlying pages being marked clean, the +.Ql B_DELWRI|B_RELBUF +combination must be interpreted to mean that the buffer is still actually dirty and must be written to its backing store before it can actually be released. In the case -where B_DELWRI is not set, the underlying dirty pages are still properly +where +.Dv B_DELWRI +is not set, the underlying dirty pages are still properly marked as dirty and the buffer can be completely freed without losing that clean/dirty state information. (XXX do we have to check other flags in @@ -128,7 +173,9 @@ The kernel reserves a portion of its KVM space to hold VM Buffer's data maps. Even though this is virtual space (since the buffers are mapped from the buffer cache), we cannot make it arbitrarily large because -instantiated VM Buffers (struct buf's) prevent their underlying pages in the +instantiated VM Buffers +.Pq Vt struct buf Ap s +prevent their underlying pages in the buffer cache from being freed. This can complicate the life of the paging system. diff --git a/share/man/man9/callout.9 b/share/man/man9/callout.9 index 0e59ef8ab2b1..637049ec1ef5 100644 --- a/share/man/man9/callout.9 +++ b/share/man/man9/callout.9 @@ -27,7 +27,7 @@ .\" ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE .\" POSSIBILITY OF SUCH DAMAGE. .\" -.Dd January 22, 2024 +.Dd November 4, 2025 .Dt CALLOUT 9 .Os .Sh NAME @@ -789,6 +789,8 @@ and functions return a value of one if the callout was still pending when it was called, a zero if the callout could not be stopped and a negative one is it was either not running or has already completed. +.Sh SEE ALSO +.Xr dtrace_callout_execute 4 .Sh HISTORY .Fx initially used the long standing diff --git a/share/man/man9/make_dev.9 b/share/man/man9/make_dev.9 index de56f350faa5..9f2c36fb39a4 100644 --- a/share/man/man9/make_dev.9 +++ b/share/man/man9/make_dev.9 @@ -25,7 +25,7 @@ .\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF .\" SUCH DAMAGE. .\" -.Dd January 19, 2025 +.Dd November 4, 2025 .Dt MAKE_DEV 9 .Os .Sh NAME @@ -387,14 +387,18 @@ function is the same as: destroy_dev_sched_cb(cdev, NULL, NULL); .Ed .Pp -The +Neither the .Fn d_close -driver method cannot call +driver method, nor a +.Xr devfs_cdevpriv 9 +.Fa dtr +method can .Fn destroy_dev directly. Doing so causes deadlock when .Fn destroy_dev -waits for all threads to leave the driver methods. +waits for all threads to leave the driver methods and finish executing any +per-open destructors. Also, because .Fn destroy_dev sleeps, no non-sleepable locks may be held over the call. diff --git a/share/misc/bsd-family-tree b/share/misc/bsd-family-tree index b0502dfc7925..82e9ac45c495 100644 --- a/share/misc/bsd-family-tree +++ b/share/misc/bsd-family-tree @@ -479,7 +479,10 @@ FreeBSD 5.2 | | | | | | | | | DragonFly 6.4.1 | | | | | DragonFly 6.4.2 | FreeBSD | | | | - | 14.3 | | OpenBSD 7.8 | + | 14.3 | | | | + | macOS | | | + | 26 | | | + | | | OpenBSD 7.8 | | | | | | FreeBSD 16 -current | NetBSD -current OpenBSD -current DragonFly -current | | | | | @@ -925,6 +928,7 @@ OpenBSD 7.7 2025-04-28 [OBD] DragonFly 6.4.1 2025-04-30 [DFB] DragonFly 6.4.2 2025-05-09 [DFB] FreeBSD 14.3 2025-06-10 [FBD] +macOS 26 2025-09-15 [APL] OpenBSD 7.8 2025-10-22 [OBD] Bibliography diff --git a/share/mk/bsd.prog.mk b/share/mk/bsd.prog.mk index 10e1c177e2b2..5697845079e2 100644 --- a/share/mk/bsd.prog.mk +++ b/share/mk/bsd.prog.mk @@ -49,9 +49,11 @@ CXXFLAGS+= -mretpoline LDFLAGS+= -Wl,-zretpolineplt .endif .else +.if !defined(_NO_INCLUDE_COMPILERMK) .warning Retpoline requested but not supported by compiler or linker .endif .endif +.endif # LLD sensibly defaults to -znoexecstack, so do the same for BFD LDFLAGS.bfd+= -Wl,-znoexecstack .if ${MK_BRANCH_PROTECTION} != "no" @@ -71,9 +73,11 @@ CFLAGS+= -enable-trivial-auto-var-init-zero-knowing-it-will-be-removed-from-clan CXXFLAGS+= -enable-trivial-auto-var-init-zero-knowing-it-will-be-removed-from-clang .endif .else +.if !defined(_NO_INCLUDE_COMPILERMK) .warning INIT_ALL (${OPT_INIT_ALL}) requested but not supported by compiler .endif .endif +.endif # Zero used registers on return (mitigate some ROP) .if ${MK_ZEROREGS} != "no" diff --git a/share/tabset/Makefile b/share/tabset/Makefile index fe7519084716..4da1a3650888 100644 --- a/share/tabset/Makefile +++ b/share/tabset/Makefile @@ -1,3 +1,5 @@ +PACKAGE= ncurses-lib + FILES= 3101 9837 aa aed512 beehive diablo dtc382 hp700-wy ibm3101 std \ stdcrt tandem653 teleray vt100 vt100-w wyse-adds xerox1720 xerox1730 \ xerox1730-lm zenith29 diff --git a/share/termcap/Makefile b/share/termcap/Makefile index 34ad41fd520a..603b098e4de1 100644 --- a/share/termcap/Makefile +++ b/share/termcap/Makefile @@ -1,10 +1,13 @@ # reorder gives an editor command for most common terminals # (in reverse order from n'th to 1'st most commonly used) # to move them to the front of termcap -# -MAN= termcap.5 -PACKAGE= runtime +MAN= termcap.5 +MANPACKAGE= ncurses + +# Note: This is in ncurses-lib rather than ncurses because without it, ncurses +# doesn't work, and the base ncurses package is optional. +PACKAGE= ncurses-lib FILES= termcap termcap.db FILESDIR= ${BINDIR}/misc CLEANFILES+= termcap.db @@ -16,7 +19,7 @@ termcap.db: termcap ${CAP_MKDB_CMD} ${CAP_MKDB_ENDIAN} -f ${.TARGET:R} ${.ALLSRC} etc-termcap: - ${INSTALL_SYMLINK} -T "package=runtime" \ + ${INSTALL_SYMLINK} -T "package=ncurses-lib" \ ${BINDIR}/misc/termcap ${DESTDIR}/etc/termcap .include <bsd.prog.mk> diff --git a/stand/libsa/zfs/zfsimpl.c b/stand/libsa/zfs/zfsimpl.c index f15d9b016068..e5920004bd9d 100644 --- a/stand/libsa/zfs/zfsimpl.c +++ b/stand/libsa/zfs/zfsimpl.c @@ -128,6 +128,7 @@ static const char *features_for_read[] = { "org.open-zfs:large_blocks", "org.openzfs:blake3", "org.zfsonlinux:large_dnode", + "com.klarasystems:dynamic_gang_header", NULL }; @@ -141,6 +142,8 @@ static uint64_t dnode_cache_bn; static char *dnode_cache_buf; static int zio_read(const spa_t *spa, const blkptr_t *bp, void *buf); +static int zio_read_impl(const spa_t *spa, const blkptr_t *bp, void *buf, + bool print); static int zfs_get_root(const spa_t *spa, uint64_t *objid); static int zfs_rlookup(const spa_t *spa, uint64_t objnum, char *result); static int zap_lookup(const spa_t *spa, const dnode_phys_t *dnode, @@ -530,7 +533,7 @@ vdev_indirect_mapping_duplicate_adjacent_entries(vdev_t *vd, uint64_t offset, } static vdev_t * -vdev_lookup_top(spa_t *spa, uint64_t vdev) +vdev_lookup_top(const spa_t *spa, uint64_t vdev) { vdev_t *rvd; vdev_list_t *vlist; @@ -2270,45 +2273,77 @@ ilog2(int n) return (-1); } +static inline uint64_t +gbh_nblkptrs(uint64_t size) +{ + ASSERT(IS_P2ALIGNED(size, sizeof(blkptr_t))); + return ((size - sizeof(zio_eck_t)) / sizeof(blkptr_t)); +} + static int zio_read_gang(const spa_t *spa, const blkptr_t *bp, void *buf) { blkptr_t gbh_bp; - zio_gbh_phys_t zio_gb; + void *gbuf; char *pbuf; - int i; + uint64_t gangblocksize; + int err, i; + + gangblocksize = UINT64_MAX; + for (int dva = 0; dva < BP_GET_NDVAS(bp); dva++) { + vdev_t *vd = vdev_lookup_top(spa, + DVA_GET_VDEV(&bp->blk_dva[dva])); + gangblocksize = MIN(gangblocksize, 1ULL << vd->v_ashift); + } /* Artificial BP for gang block header. */ gbh_bp = *bp; - BP_SET_PSIZE(&gbh_bp, SPA_GANGBLOCKSIZE); - BP_SET_LSIZE(&gbh_bp, SPA_GANGBLOCKSIZE); + BP_SET_PSIZE(&gbh_bp, gangblocksize); + BP_SET_LSIZE(&gbh_bp, gangblocksize); BP_SET_CHECKSUM(&gbh_bp, ZIO_CHECKSUM_GANG_HEADER); BP_SET_COMPRESS(&gbh_bp, ZIO_COMPRESS_OFF); for (i = 0; i < SPA_DVAS_PER_BP; i++) DVA_SET_GANG(&gbh_bp.blk_dva[i], 0); + gbuf = malloc(gangblocksize); + if (gbuf == NULL) + return (ENOMEM); /* Read gang header block using the artificial BP. */ - if (zio_read(spa, &gbh_bp, &zio_gb)) + err = zio_read_impl(spa, &gbh_bp, gbuf, false); + if ((err == EIO || err == ECKSUM) && + gangblocksize > SPA_OLD_GANGBLOCKSIZE) { + /* This might be a legacy gang block header, try again. */ + gangblocksize = SPA_OLD_GANGBLOCKSIZE; + BP_SET_PSIZE(&gbh_bp, gangblocksize); + BP_SET_LSIZE(&gbh_bp, gangblocksize); + err = zio_read(spa, &gbh_bp, gbuf); + } + if (err != 0) { + free(gbuf); return (EIO); + } pbuf = buf; - for (i = 0; i < SPA_GBH_NBLKPTRS; i++) { - blkptr_t *gbp = &zio_gb.zg_blkptr[i]; + for (i = 0; i < gbh_nblkptrs(gangblocksize); i++) { + blkptr_t *gbp = &((blkptr_t *)gbuf)[i]; if (BP_IS_HOLE(gbp)) continue; - if (zio_read(spa, gbp, pbuf)) + if (zio_read(spa, gbp, pbuf)) { + free(gbuf); return (EIO); + } pbuf += BP_GET_PSIZE(gbp); } + free(gbuf); if (zio_checksum_verify(spa, bp, buf)) return (EIO); return (0); } static int -zio_read(const spa_t *spa, const blkptr_t *bp, void *buf) +zio_read_impl(const spa_t *spa, const blkptr_t *bp, void *buf, bool print) { int cpfunc = BP_GET_COMPRESS(bp); uint64_t align, size; @@ -2340,7 +2375,7 @@ zio_read(const spa_t *spa, const blkptr_t *bp, void *buf) size, buf, BP_GET_LSIZE(bp)); free(pbuf); } - if (error != 0) + if (error != 0 && print) printf("ZFS: i/o error - unable to decompress " "block pointer data, error %d\n", error); return (error); @@ -2394,7 +2429,7 @@ zio_read(const spa_t *spa, const blkptr_t *bp, void *buf) BP_GET_PSIZE(bp), buf, BP_GET_LSIZE(bp)); else if (size != BP_GET_PSIZE(bp)) bcopy(pbuf, buf, BP_GET_PSIZE(bp)); - } else { + } else if (print) { printf("zio_read error: %d\n", error); } if (buf != pbuf) @@ -2402,13 +2437,19 @@ zio_read(const spa_t *spa, const blkptr_t *bp, void *buf) if (error == 0) break; } - if (error != 0) + if (error != 0 && print) printf("ZFS: i/o error - all block copies unavailable\n"); return (error); } static int +zio_read(const spa_t *spa, const blkptr_t *bp, void *buf) +{ + return (zio_read_impl(spa, bp, buf, true)); +} + +static int dnode_read(const spa_t *spa, const dnode_phys_t *dnode, off_t offset, void *buf, size_t buflen) { diff --git a/sys/amd64/include/vmm.h b/sys/amd64/include/vmm.h index ad67510fecf3..5cf1ae2d769c 100644 --- a/sys/amd64/include/vmm.h +++ b/sys/amd64/include/vmm.h @@ -122,33 +122,7 @@ enum x2apic_state { #define VM_INTINFO_HWEXCEPTION (3 << 8) #define VM_INTINFO_SWINTR (4 << 8) -/* - * The VM name has to fit into the pathname length constraints of devfs, - * governed primarily by SPECNAMELEN. The length is the total number of - * characters in the full path, relative to the mount point and not - * including any leading '/' characters. - * A prefix and a suffix are added to the name specified by the user. - * The prefix is usually "vmm/" or "vmm.io/", but can be a few characters - * longer for future use. - * The suffix is a string that identifies a bootrom image or some similar - * image that is attached to the VM. A separator character gets added to - * the suffix automatically when generating the full path, so it must be - * accounted for, reducing the effective length by 1. - * The effective length of a VM name is 229 bytes for FreeBSD 13 and 37 - * bytes for FreeBSD 12. A minimum length is set for safety and supports - * a SPECNAMELEN as small as 32 on old systems. - */ -#define VM_MAX_PREFIXLEN 10 -#define VM_MAX_SUFFIXLEN 15 -#define VM_MIN_NAMELEN 6 -#define VM_MAX_NAMELEN \ - (SPECNAMELEN - VM_MAX_PREFIXLEN - VM_MAX_SUFFIXLEN - 1) - #ifdef _KERNEL -#include <sys/kassert.h> - -CTASSERT(VM_MAX_NAMELEN >= VM_MIN_NAMELEN); - struct vm; struct vm_exception; struct vm_mem; @@ -232,8 +206,6 @@ struct vmm_ops { extern const struct vmm_ops vmm_ops_intel; extern const struct vmm_ops vmm_ops_amd; -extern u_int vm_maxcpu; /* maximum virtual cpus */ - int vm_create(const char *name, struct vm **retvm); struct vcpu *vm_alloc_vcpu(struct vm *vm, int vcpuid); void vm_disable_vcpu_creation(struct vm *vm); @@ -383,7 +355,8 @@ vcpu_should_yield(struct vcpu *vcpu) #endif void *vcpu_stats(struct vcpu *vcpu); -void vcpu_notify_event(struct vcpu *vcpu, bool lapic_intr); +void vcpu_notify_event(struct vcpu *vcpu); +void vcpu_notify_lapic(struct vcpu *vcpu); struct vm_mem *vm_mem(struct vm *vm); struct vatpic *vm_atpic(struct vm *vm); struct vatpit *vm_atpit(struct vm *vm); diff --git a/sys/amd64/include/vmm_dev.h b/sys/amd64/include/vmm_dev.h index 441330fd57b8..f1c07a983a4b 100644 --- a/sys/amd64/include/vmm_dev.h +++ b/sys/amd64/include/vmm_dev.h @@ -34,6 +34,8 @@ #include <machine/vmm.h> #include <machine/vmm_snapshot.h> +#include <dev/vmm/vmm_param.h> + struct vm_memmap { vm_paddr_t gpa; int segid; /* memory segment */ diff --git a/sys/amd64/vmm/intel/vmx.c b/sys/amd64/vmm/intel/vmx.c index 842281ab862e..4189c1214b40 100644 --- a/sys/amd64/vmm/intel/vmx.c +++ b/sys/amd64/vmm/intel/vmx.c @@ -27,7 +27,6 @@ * SUCH DAMAGE. */ -#include <sys/cdefs.h> #include "opt_bhyve_snapshot.h" #include <sys/param.h> @@ -58,6 +57,7 @@ #include <machine/vmm_instruction_emul.h> #include <machine/vmm_snapshot.h> +#include <dev/vmm/vmm_dev.h> #include <dev/vmm/vmm_ktr.h> #include <dev/vmm/vmm_mem.h> diff --git a/sys/amd64/vmm/io/ppt.c b/sys/amd64/vmm/io/ppt.c index 2cb459fb848f..6feac5dcbbed 100644 --- a/sys/amd64/vmm/io/ppt.c +++ b/sys/amd64/vmm/io/ppt.c @@ -336,13 +336,6 @@ ppt_teardown_msix(struct pptdev *ppt) } int -ppt_avail_devices(void) -{ - - return (num_pptdevs); -} - -int ppt_assigned_devices(struct vm *vm) { struct pptdev *ppt; diff --git a/sys/amd64/vmm/io/ppt.h b/sys/amd64/vmm/io/ppt.h index f97c399564d7..9377f34d50e6 100644 --- a/sys/amd64/vmm/io/ppt.h +++ b/sys/amd64/vmm/io/ppt.h @@ -43,12 +43,6 @@ int ppt_assigned_devices(struct vm *vm); bool ppt_is_mmio(struct vm *vm, vm_paddr_t gpa); /* - * Returns the number of devices sequestered by the ppt driver for assignment - * to virtual machines. - */ -int ppt_avail_devices(void); - -/* * The following functions should never be called directly. * Use 'vm_assign_pptdev()' and 'vm_unassign_pptdev()' instead. */ diff --git a/sys/amd64/vmm/io/vlapic.c b/sys/amd64/vmm/io/vlapic.c index 9879dfa164a4..afd5045de574 100644 --- a/sys/amd64/vmm/io/vlapic.c +++ b/sys/amd64/vmm/io/vlapic.c @@ -456,7 +456,7 @@ vlapic_fire_lvt(struct vlapic *vlapic, u_int lvt) return (0); } if (vlapic_set_intr_ready(vlapic, vec, false)) - vcpu_notify_event(vlapic->vcpu, true); + vcpu_notify_lapic(vlapic->vcpu); break; case APIC_LVT_DM_NMI: vm_inject_nmi(vlapic->vcpu); diff --git a/sys/amd64/vmm/vmm.c b/sys/amd64/vmm/vmm.c index f2bea0d82b5c..2890e990633d 100644 --- a/sys/amd64/vmm/vmm.c +++ b/sys/amd64/vmm/vmm.c @@ -31,7 +31,6 @@ #include <sys/param.h> #include <sys/systm.h> #include <sys/kernel.h> -#include <sys/module.h> #include <sys/sysctl.h> #include <sys/malloc.h> #include <sys/pcpu.h> @@ -189,8 +188,6 @@ struct vm { #define VMM_CTR4(vcpu, format, p1, p2, p3, p4) \ VCPU_CTR4((vcpu)->vm, (vcpu)->vcpuid, format, p1, p2, p3, p4) -static int vmm_initialized; - static void vmmops_panic(void); static void @@ -270,11 +267,7 @@ static int trap_wbinvd; SYSCTL_INT(_hw_vmm, OID_AUTO, trap_wbinvd, CTLFLAG_RDTUN, &trap_wbinvd, 0, "WBINVD triggers a VM-exit"); -u_int vm_maxcpu; -SYSCTL_UINT(_hw_vmm, OID_AUTO, maxcpu, CTLFLAG_RDTUN | CTLFLAG_NOFETCH, - &vm_maxcpu, 0, "Maximum number of vCPUs"); - -static void vcpu_notify_event_locked(struct vcpu *vcpu, bool lapic_intr); +static void vcpu_notify_event_locked(struct vcpu *vcpu); /* global statistics */ VMM_STAT(VCPU_MIGRATIONS, "vcpu migration across host cpus"); @@ -299,14 +292,6 @@ VMM_STAT(VMEXIT_USERSPACE, "number of vm exits handled in userspace"); VMM_STAT(VMEXIT_RENDEZVOUS, "number of times rendezvous pending at exit"); VMM_STAT(VMEXIT_EXCEPTION, "number of vm exits due to exceptions"); -/* - * Upper limit on vm_maxcpu. Limited by use of uint16_t types for CPU - * counts as well as range of vpid values for VT-x and by the capacity - * of cpuset_t masks. The call to new_unrhdr() in vpid_init() in - * vmx.c requires 'vm_maxcpu + 1 <= 0xffff', hence the '- 1' below. - */ -#define VM_MAXCPU MIN(0xffff - 1, CPU_SETSIZE) - #ifdef KTR static const char * vcpu_state2str(enum vcpu_state state) @@ -402,22 +387,12 @@ vm_exitinfo_cpuset(struct vcpu *vcpu) return (&vcpu->exitinfo_cpuset); } -static int -vmm_init(void) +int +vmm_modinit(void) { if (!vmm_is_hw_supported()) return (ENXIO); - vm_maxcpu = mp_ncpus; - TUNABLE_INT_FETCH("hw.vmm.maxcpu", &vm_maxcpu); - - if (vm_maxcpu > VM_MAXCPU) { - printf("vmm: vm_maxcpu clamped to %u\n", VM_MAXCPU); - vm_maxcpu = VM_MAXCPU; - } - if (vm_maxcpu == 0) - vm_maxcpu = 1; - vmm_host_state_init(); vmm_ipinum = lapic_ipi_alloc(pti ? &IDTVEC(justreturn1_pti) : @@ -431,70 +406,17 @@ vmm_init(void) return (vmmops_modinit(vmm_ipinum)); } -static int -vmm_handler(module_t mod, int what, void *arg) +int +vmm_modcleanup(void) { - int error; - - switch (what) { - case MOD_LOAD: - if (vmm_is_hw_supported()) { - error = vmmdev_init(); - if (error != 0) - break; - error = vmm_init(); - if (error == 0) - vmm_initialized = 1; - else - (void)vmmdev_cleanup(); - } else { - error = ENXIO; - } - break; - case MOD_UNLOAD: - if (vmm_is_hw_supported()) { - error = vmmdev_cleanup(); - if (error == 0) { - vmm_suspend_p = NULL; - vmm_resume_p = NULL; - iommu_cleanup(); - if (vmm_ipinum != IPI_AST) - lapic_ipi_free(vmm_ipinum); - error = vmmops_modcleanup(); - /* - * Something bad happened - prevent new - * VMs from being created - */ - if (error) - vmm_initialized = 0; - } - } else { - error = 0; - } - break; - default: - error = 0; - break; - } - return (error); + vmm_suspend_p = NULL; + vmm_resume_p = NULL; + iommu_cleanup(); + if (vmm_ipinum != IPI_AST) + lapic_ipi_free(vmm_ipinum); + return (vmmops_modcleanup()); } -static moduledata_t vmm_kmod = { - "vmm", - vmm_handler, - NULL -}; - -/* - * vmm initialization has the following dependencies: - * - * - VT-x initialization requires smp_rendezvous() and therefore must happen - * after SMP is fully functional (after SI_SUB_SMP). - * - vmm device initialization requires an initialized devfs. - */ -DECLARE_MODULE(vmm, vmm_kmod, MAX(SI_SUB_SMP, SI_SUB_DEVFS) + 1, SI_ORDER_ANY); -MODULE_VERSION(vmm, 1); - static void vm_init(struct vm *vm, bool create) { @@ -573,29 +495,12 @@ vm_unlock_vcpus(struct vm *vm) sx_unlock(&vm->vcpus_init_lock); } -/* - * The default CPU topology is a single thread per package. - */ -u_int cores_per_package = 1; -u_int threads_per_core = 1; - int vm_create(const char *name, struct vm **retvm) { struct vm *vm; int error; - /* - * If vmm.ko could not be successfully initialized then don't attempt - * to create the virtual machine. - */ - if (!vmm_initialized) - return (ENXIO); - - if (name == NULL || strnlen(name, VM_MAX_NAMELEN + 1) == - VM_MAX_NAMELEN + 1) - return (EINVAL); - vm = malloc(sizeof(struct vm), M_VM, M_WAITOK | M_ZERO); error = vm_mem_init(&vm->mem, 0, VM_MAXUSER_ADDRESS_LA48); if (error != 0) { @@ -609,8 +514,8 @@ vm_create(const char *name, struct vm **retvm) M_ZERO); vm->sockets = 1; - vm->cores = cores_per_package; /* XXX backwards compatibility */ - vm->threads = threads_per_core; /* XXX backwards compatibility */ + vm->cores = 1; /* XXX backwards compatibility */ + vm->threads = 1; /* XXX backwards compatibility */ vm->maxcpus = vm_maxcpu; vm_init(vm, true); @@ -1028,7 +933,7 @@ vcpu_wait_idle(struct vcpu *vcpu) KASSERT(vcpu->state != VCPU_IDLE, ("vcpu already idle")); vcpu->reqidle = 1; - vcpu_notify_event_locked(vcpu, false); + vcpu_notify_event_locked(vcpu); VMM_CTR1(vcpu, "vcpu state change from %s to " "idle requested", vcpu_state2str(vcpu->state)); msleep_spin(&vcpu->state, &vcpu->mtx, "vmstat", hz); @@ -1509,7 +1414,7 @@ vm_handle_suspend(struct vcpu *vcpu, bool *retu) */ for (i = 0; i < vm->maxcpus; i++) { if (CPU_ISSET(i, &vm->suspended_cpus)) { - vcpu_notify_event(vm_vcpu(vm, i), false); + vcpu_notify_event(vm_vcpu(vm, i)); } } @@ -1583,7 +1488,7 @@ vm_suspend(struct vm *vm, enum vm_suspend_how how) */ for (i = 0; i < vm->maxcpus; i++) { if (CPU_ISSET(i, &vm->active_cpus)) - vcpu_notify_event(vm_vcpu(vm, i), false); + vcpu_notify_event(vm_vcpu(vm, i)); } return (0); @@ -2063,7 +1968,7 @@ vm_inject_nmi(struct vcpu *vcpu) { vcpu->nmi_pending = 1; - vcpu_notify_event(vcpu, false); + vcpu_notify_event(vcpu); return (0); } @@ -2090,7 +1995,7 @@ vm_inject_extint(struct vcpu *vcpu) { vcpu->extint_pending = 1; - vcpu_notify_event(vcpu, false); + vcpu_notify_event(vcpu); return (0); } @@ -2261,14 +2166,14 @@ vm_suspend_cpu(struct vm *vm, struct vcpu *vcpu) vm->debug_cpus = vm->active_cpus; for (int i = 0; i < vm->maxcpus; i++) { if (CPU_ISSET(i, &vm->active_cpus)) - vcpu_notify_event(vm_vcpu(vm, i), false); + vcpu_notify_event(vm_vcpu(vm, i)); } } else { if (!CPU_ISSET(vcpu->vcpuid, &vm->active_cpus)) return (EINVAL); CPU_SET_ATOMIC(vcpu->vcpuid, &vm->debug_cpus); - vcpu_notify_event(vcpu, false); + vcpu_notify_event(vcpu); } return (0); } @@ -2376,7 +2281,7 @@ vm_set_x2apic_state(struct vcpu *vcpu, enum x2apic_state state) * to the host_cpu to cause the vcpu to trap into the hypervisor. */ static void -vcpu_notify_event_locked(struct vcpu *vcpu, bool lapic_intr) +vcpu_notify_event_locked(struct vcpu *vcpu) { int hostcpu; @@ -2384,12 +2289,7 @@ vcpu_notify_event_locked(struct vcpu *vcpu, bool lapic_intr) if (vcpu->state == VCPU_RUNNING) { KASSERT(hostcpu != NOCPU, ("vcpu running on invalid hostcpu")); if (hostcpu != curcpu) { - if (lapic_intr) { - vlapic_post_intr(vcpu->vlapic, hostcpu, - vmm_ipinum); - } else { - ipi_cpu(hostcpu, vmm_ipinum); - } + ipi_cpu(hostcpu, vmm_ipinum); } else { /* * If the 'vcpu' is running on 'curcpu' then it must @@ -2407,10 +2307,21 @@ vcpu_notify_event_locked(struct vcpu *vcpu, bool lapic_intr) } void -vcpu_notify_event(struct vcpu *vcpu, bool lapic_intr) +vcpu_notify_event(struct vcpu *vcpu) { vcpu_lock(vcpu); - vcpu_notify_event_locked(vcpu, lapic_intr); + vcpu_notify_event_locked(vcpu); + vcpu_unlock(vcpu); +} + +void +vcpu_notify_lapic(struct vcpu *vcpu) +{ + vcpu_lock(vcpu); + if (vcpu->state == VCPU_RUNNING && vcpu->hostcpu != curcpu) + vlapic_post_intr(vcpu->vlapic, vcpu->hostcpu, vmm_ipinum); + else + vcpu_notify_event_locked(vcpu); vcpu_unlock(vcpu); } @@ -2472,7 +2383,7 @@ restart: */ for (i = 0; i < vm->maxcpus; i++) { if (CPU_ISSET(i, &dest)) - vcpu_notify_event(vm_vcpu(vm, i), false); + vcpu_notify_event(vm_vcpu(vm, i)); } return (vm_handle_rendezvous(vcpu)); diff --git a/sys/amd64/vmm/vmm_lapic.c b/sys/amd64/vmm/vmm_lapic.c index 0cae01f172ec..63bdee69bb59 100644 --- a/sys/amd64/vmm/vmm_lapic.c +++ b/sys/amd64/vmm/vmm_lapic.c @@ -61,7 +61,7 @@ lapic_set_intr(struct vcpu *vcpu, int vector, bool level) vlapic = vm_lapic(vcpu); if (vlapic_set_intr_ready(vlapic, vector, level)) - vcpu_notify_event(vcpu, true); + vcpu_notify_lapic(vcpu); return (0); } diff --git a/sys/arm64/include/vmm.h b/sys/arm64/include/vmm.h index 696a69669a2a..e67540eac66d 100644 --- a/sys/arm64/include/vmm.h +++ b/sys/arm64/include/vmm.h @@ -106,27 +106,6 @@ enum vm_reg_name { #define VM_GUEST_BASE_IPA 0x80000000UL /* Guest kernel start ipa */ -/* - * The VM name has to fit into the pathname length constraints of devfs, - * governed primarily by SPECNAMELEN. The length is the total number of - * characters in the full path, relative to the mount point and not - * including any leading '/' characters. - * A prefix and a suffix are added to the name specified by the user. - * The prefix is usually "vmm/" or "vmm.io/", but can be a few characters - * longer for future use. - * The suffix is a string that identifies a bootrom image or some similar - * image that is attached to the VM. A separator character gets added to - * the suffix automatically when generating the full path, so it must be - * accounted for, reducing the effective length by 1. - * The effective length of a VM name is 229 bytes for FreeBSD 13 and 37 - * bytes for FreeBSD 12. A minimum length is set for safety and supports - * a SPECNAMELEN as small as 32 on old systems. - */ -#define VM_MAX_PREFIXLEN 10 -#define VM_MAX_SUFFIXLEN 15 -#define VM_MAX_NAMELEN \ - (SPECNAMELEN - VM_MAX_PREFIXLEN - VM_MAX_SUFFIXLEN - 1) - #ifdef _KERNEL struct vm; struct vm_exception; diff --git a/sys/arm64/include/vmm_dev.h b/sys/arm64/include/vmm_dev.h index 219f1116c728..289ff0fe1fc9 100644 --- a/sys/arm64/include/vmm_dev.h +++ b/sys/arm64/include/vmm_dev.h @@ -31,6 +31,8 @@ #include <machine/vmm.h> +#include <dev/vmm/vmm_param.h> + struct vm_memmap { vm_paddr_t gpa; int segid; /* memory segment */ diff --git a/sys/arm64/vmm/vmm.c b/sys/arm64/vmm/vmm.c index e7b2b5d8c360..31d2fb3f516b 100644 --- a/sys/arm64/vmm/vmm.c +++ b/sys/arm64/vmm/vmm.c @@ -33,7 +33,6 @@ #include <sys/linker.h> #include <sys/lock.h> #include <sys/malloc.h> -#include <sys/module.h> #include <sys/mutex.h> #include <sys/pcpu.h> #include <sys/proc.h> @@ -125,7 +124,7 @@ struct vm { volatile cpuset_t suspended_cpus; /* (i) suspended vcpus */ volatile cpuset_t halted_cpus; /* (x) cpus in a hard halt */ struct vm_mem mem; /* (i) guest memory */ - char name[VM_MAX_NAMELEN]; /* (o) virtual machine name */ + char name[VM_MAX_NAMELEN + 1]; /* (o) virtual machine name */ struct vcpu **vcpu; /* (i) guest vcpus */ struct vmm_mmio_region mmio_region[VM_MAX_MMIO_REGIONS]; /* (o) guest MMIO regions */ @@ -138,8 +137,6 @@ struct vm { struct sx vcpus_init_lock; /* (o) */ }; -static bool vmm_initialized = false; - static int vm_handle_wfi(struct vcpu *vcpu, struct vm_exit *vme, bool *retu); @@ -208,10 +205,6 @@ static const struct vmm_regs vmm_arch_regs_masks = { /* Host registers masked by vmm_arch_regs_masks. */ static struct vmm_regs vmm_arch_regs; -u_int vm_maxcpu; -SYSCTL_UINT(_hw_vmm, OID_AUTO, maxcpu, CTLFLAG_RDTUN | CTLFLAG_NOFETCH, - &vm_maxcpu, 0, "Maximum number of vCPUs"); - static void vcpu_notify_event_locked(struct vcpu *vcpu); /* global statistics */ @@ -231,12 +224,6 @@ VMM_STAT(VMEXIT_SS, "number of vmexits for a single-step exception"); VMM_STAT(VMEXIT_UNHANDLED_EL2, "number of vmexits for an unhandled EL2 exception"); VMM_STAT(VMEXIT_UNHANDLED, "number of vmexits for an unhandled exception"); -/* - * Upper limit on vm_maxcpu. We could increase this to 28 bits, but this - * is a safe value for now. - */ -#define VM_MAXCPU MIN(0xffff - 1, CPU_SETSIZE) - static int vmm_regs_init(struct vmm_regs *regs, const struct vmm_regs *masks) { @@ -323,20 +310,14 @@ vmm_unsupported_quirk(void) return (0); } -static int -vmm_init(void) +int +vmm_modinit(void) { int error; - vm_maxcpu = mp_ncpus; - TUNABLE_INT_FETCH("hw.vmm.maxcpu", &vm_maxcpu); - - if (vm_maxcpu > VM_MAXCPU) { - printf("vmm: vm_maxcpu clamped to %u\n", VM_MAXCPU); - vm_maxcpu = VM_MAXCPU; - } - if (vm_maxcpu == 0) - vm_maxcpu = 1; + error = vmm_unsupported_quirk(); + if (error != 0) + return (error); error = vmm_regs_init(&vmm_arch_regs, &vmm_arch_regs_masks); if (error != 0) @@ -345,61 +326,12 @@ vmm_init(void) return (vmmops_modinit(0)); } -static int -vmm_handler(module_t mod, int what, void *arg) +int +vmm_modcleanup(void) { - int error; - - switch (what) { - case MOD_LOAD: - error = vmm_unsupported_quirk(); - if (error != 0) - break; - error = vmmdev_init(); - if (error != 0) - break; - error = vmm_init(); - if (error == 0) - vmm_initialized = true; - else - (void)vmmdev_cleanup(); - break; - case MOD_UNLOAD: - error = vmmdev_cleanup(); - if (error == 0 && vmm_initialized) { - error = vmmops_modcleanup(); - if (error) { - /* - * Something bad happened - prevent new - * VMs from being created - */ - vmm_initialized = false; - } - } - break; - default: - error = 0; - break; - } - return (error); + return (vmmops_modcleanup()); } -static moduledata_t vmm_kmod = { - "vmm", - vmm_handler, - NULL -}; - -/* - * vmm initialization has the following dependencies: - * - * - HYP initialization requires smp_rendezvous() and therefore must happen - * after SMP is fully functional (after SI_SUB_SMP). - * - vmm device initialization requires an initialized devfs. - */ -DECLARE_MODULE(vmm, vmm_kmod, MAX(SI_SUB_SMP, SI_SUB_DEVFS) + 1, SI_ORDER_ANY); -MODULE_VERSION(vmm, 1); - static void vm_init(struct vm *vm, bool create) { @@ -441,10 +373,6 @@ vm_alloc_vcpu(struct vm *vm, int vcpuid) if (vcpuid < 0 || vcpuid >= vm_get_maxcpus(vm)) return (NULL); - /* Some interrupt controllers may have a CPU limit */ - if (vcpuid >= vgic_max_cpu_count(vm->cookie)) - return (NULL); - vcpu = (struct vcpu *) atomic_load_acq_ptr((uintptr_t *)&vm->vcpu[vcpuid]); if (__predict_true(vcpu != NULL)) @@ -453,6 +381,12 @@ vm_alloc_vcpu(struct vm *vm, int vcpuid) sx_xlock(&vm->vcpus_init_lock); vcpu = vm->vcpu[vcpuid]; if (vcpu == NULL && !vm->dying) { + /* Some interrupt controllers may have a CPU limit */ + if (vcpuid >= vgic_max_cpu_count(vm->cookie)) { + sx_xunlock(&vm->vcpus_init_lock); + return (NULL); + } + vcpu = vcpu_alloc(vm, vcpuid); vcpu_init(vcpu); @@ -485,16 +419,6 @@ vm_create(const char *name, struct vm **retvm) struct vm *vm; int error; - /* - * If vmm.ko could not be successfully initialized then don't attempt - * to create the virtual machine. - */ - if (!vmm_initialized) - return (ENXIO); - - if (name == NULL || strlen(name) >= VM_MAX_NAMELEN) - return (EINVAL); - vm = malloc(sizeof(struct vm), M_VMM, M_WAITOK | M_ZERO); error = vm_mem_init(&vm->mem, 0, 1ul << 39); if (error != 0) { diff --git a/sys/cam/ata/ata_da.c b/sys/cam/ata/ata_da.c index 08747cd59131..9434756b87f9 100644 --- a/sys/cam/ata/ata_da.c +++ b/sys/cam/ata/ata_da.c @@ -2328,15 +2328,38 @@ adastart(struct cam_periph *periph, union ccb *start_ccb) { struct ada_softc *softc = (struct ada_softc *)periph->softc; struct ccb_ataio *ataio = &start_ccb->ataio; + uint32_t priority = start_ccb->ccb_h.pinfo.priority; CAM_DEBUG(periph->path, CAM_DEBUG_TRACE, ("adastart\n")); + /* + * When we're running the state machine, we should only accept DEV CCBs. + * When we're doing normal I/O we should only accept NORMAL CCBs. + * + * While in the state machine, we carefully single step the queue, but + * there's no protection for 'extra' calls to xpt_schedule() at the + * wrong priority. Guard against that so that we filter any CCBs that + * are offered at the wrong priority. This avoids generating requests + * that are at normal priority. +` */ + if ((softc->state != ADA_STATE_NORMAL && priority != CAM_PRIORITY_DEV) || + (softc->state == ADA_STATE_NORMAL && priority != CAM_PRIORITY_NORMAL)) { + xpt_print(periph->path, "Bad priority for state %d prio %d\n", + softc->state, priority); + xpt_release_ccb(start_ccb); + return; + } + switch (softc->state) { case ADA_STATE_NORMAL: { struct bio *bp; uint8_t tag_code; + KASSERT(priority == CAM_PRIORITY_NORMAL, + ("Expected priority %d, found %d in state normal", + CAM_PRIORITY_NORMAL, priority)); + bp = cam_iosched_next_bio(softc->cam_iosched); if (bp == NULL) { xpt_release_ccb(start_ccb); @@ -2555,6 +2578,11 @@ out: case ADA_STATE_RAHEAD: case ADA_STATE_WCACHE: { + KASSERT(priority == CAM_PRIORITY_DEV, + ("Expected priority %d, found %d in state %s", + CAM_PRIORITY_DEV, priority, + softc->state == ADA_STATE_RAHEAD ? "rahead" : "wcache")); + cam_fill_ataio(ataio, 1, adadone, @@ -2581,6 +2609,10 @@ out: { struct ata_gp_log_dir *log_dir; + KASSERT(priority == CAM_PRIORITY_DEV, + ("Expected priority %d, found %d in state logdir", + CAM_PRIORITY_DEV, priority)); + if ((softc->flags & ADA_FLAG_CAN_LOG) == 0) { adaprobedone(periph, start_ccb); break; @@ -2615,6 +2647,10 @@ out: { struct ata_identify_log_pages *id_dir; + KASSERT(priority == CAM_PRIORITY_DEV, + ("Expected priority %d, found %d in state iddir", + CAM_PRIORITY_DEV, priority)); + id_dir = malloc(sizeof(*id_dir), M_ATADA, M_NOWAIT | M_ZERO); if (id_dir == NULL) { xpt_print(periph->path, "Couldn't malloc id_dir " @@ -2643,6 +2679,10 @@ out: { struct ata_identify_log_sup_cap *sup_cap; + KASSERT(priority == CAM_PRIORITY_DEV, + ("Expected priority %d, found %d in state sup_cap", + CAM_PRIORITY_DEV, priority)); + sup_cap = malloc(sizeof(*sup_cap), M_ATADA, M_NOWAIT|M_ZERO); if (sup_cap == NULL) { xpt_print(periph->path, "Couldn't malloc sup_cap " @@ -2671,6 +2711,10 @@ out: { struct ata_zoned_info_log *ata_zone; + KASSERT(priority == CAM_PRIORITY_DEV, + ("Expected priority %d, found %d in state zone", + CAM_PRIORITY_DEV, priority)); + ata_zone = malloc(sizeof(*ata_zone), M_ATADA, M_NOWAIT|M_ZERO); if (ata_zone == NULL) { xpt_print(periph->path, "Couldn't malloc ata_zone " @@ -2896,6 +2940,10 @@ adadone(struct cam_periph *periph, union ccb *done_ccb) struct bio *bp; int error; + KASSERT(priority == CAM_PRIORITY_NORMAL, + ("Expected priority %d, found %d for normal I/O", + CAM_PRIORITY_NORMAL, priority)); + cam_periph_lock(periph); bp = (struct bio *)done_ccb->ccb_h.ccb_bp; if ((done_ccb->ccb_h.status & CAM_STATUS_MASK) != CAM_REQ_CMP) { @@ -3000,6 +3048,10 @@ adadone(struct cam_periph *periph, union ccb *done_ccb) } case ADA_CCB_RAHEAD: { + KASSERT(priority == CAM_PRIORITY_DEV, + ("Expected priority %d, found %d in ccb state rahead", + CAM_PRIORITY_DEV, priority)); + if ((done_ccb->ccb_h.status & CAM_STATUS_MASK) != CAM_REQ_CMP) { if (adaerror(done_ccb, 0, 0) == ERESTART) { /* Drop freeze taken due to CAM_DEV_QFREEZE */ @@ -3023,6 +3075,10 @@ adadone(struct cam_periph *periph, union ccb *done_ccb) } case ADA_CCB_WCACHE: { + KASSERT(priority == CAM_PRIORITY_DEV, + ("Expected priority %d, found %d in ccb state wcache", + CAM_PRIORITY_DEV, priority)); + if ((done_ccb->ccb_h.status & CAM_STATUS_MASK) != CAM_REQ_CMP) { if (adaerror(done_ccb, 0, 0) == ERESTART) { /* Drop freeze taken due to CAM_DEV_QFREEZE */ @@ -3054,6 +3110,10 @@ adadone(struct cam_periph *periph, union ccb *done_ccb) { int error; + KASSERT(priority == CAM_PRIORITY_DEV, + ("Expected priority %d, found %d in ccb state logdir", + CAM_PRIORITY_DEV, priority)); + if ((done_ccb->ccb_h.status & CAM_STATUS_MASK) == CAM_REQ_CMP) { error = 0; softc->valid_logdir_len = 0; @@ -3123,6 +3183,10 @@ adadone(struct cam_periph *periph, union ccb *done_ccb) case ADA_CCB_IDDIR: { int error; + KASSERT(priority == CAM_PRIORITY_DEV, + ("Expected priority %d, found %d in ccb state iddir", + CAM_PRIORITY_DEV, priority)); + if ((ataio->ccb_h.status & CAM_STATUS_MASK) == CAM_REQ_CMP) { off_t entries_offset, max_entries; error = 0; @@ -3208,6 +3272,10 @@ adadone(struct cam_periph *periph, union ccb *done_ccb) case ADA_CCB_SUP_CAP: { int error; + KASSERT(priority == CAM_PRIORITY_DEV, + ("Expected priority %d, found %d in ccb state sup_cap", + CAM_PRIORITY_DEV, priority)); + if ((ataio->ccb_h.status & CAM_STATUS_MASK) == CAM_REQ_CMP) { uint32_t valid_len; size_t needed_size; @@ -3312,6 +3380,10 @@ adadone(struct cam_periph *periph, union ccb *done_ccb) case ADA_CCB_ZONE: { int error; + KASSERT(priority == CAM_PRIORITY_DEV, + ("Expected priority %d, found %d in ccb state zone", + CAM_PRIORITY_DEV, priority)); + if ((ataio->ccb_h.status & CAM_STATUS_MASK) == CAM_REQ_CMP) { struct ata_zoned_info_log *zi_log; uint32_t valid_len; diff --git a/sys/cam/scsi/scsi_da.c b/sys/cam/scsi/scsi_da.c index c0c0be12856b..773a786d08f7 100644 --- a/sys/cam/scsi/scsi_da.c +++ b/sys/cam/scsi/scsi_da.c @@ -3369,12 +3369,33 @@ static void dastart(struct cam_periph *periph, union ccb *start_ccb) { struct da_softc *softc; + uint32_t priority = start_ccb->ccb_h.pinfo.priority; cam_periph_assert(periph, MA_OWNED); softc = (struct da_softc *)periph->softc; CAM_DEBUG(periph->path, CAM_DEBUG_TRACE, ("dastart\n")); + /* + * When we're running the state machine, we should only accept DEV CCBs. + * When we're doing normal I/O we should only accept NORMAL CCBs. + * + * While in the state machine, we carefully single step the queue, but + * there's no protection for 'extra' calls to xpt_schedule() at the + * wrong priority. Guard against that so that we filter any CCBs that + * are offered at the wrong priority. This avoids generating requests + * that are at normal priority. In addition, though we can't easily + * enforce it, one must not transition to the NORMAL state via the + * skipstate mechanism. +` */ + if ((softc->state != DA_STATE_NORMAL && priority != CAM_PRIORITY_DEV) || + (softc->state == DA_STATE_NORMAL && priority != CAM_PRIORITY_NORMAL)) { + xpt_print(periph->path, "Bad priority for state %d prio %d\n", + softc->state, priority); + xpt_release_ccb(start_ccb); + return; + } + skipstate: switch (softc->state) { case DA_STATE_NORMAL: diff --git a/sys/cddl/boot/zfs/zfsimpl.h b/sys/cddl/boot/zfs/zfsimpl.h index c9de1fe4c391..d3ae3c32635d 100644 --- a/sys/cddl/boot/zfs/zfsimpl.h +++ b/sys/cddl/boot/zfs/zfsimpl.h @@ -94,6 +94,7 @@ typedef enum { B_FALSE, B_TRUE } boolean_t; #define P2END(x, align) (-(~(x) & -(align))) #define P2PHASEUP(x, align, phase) ((phase) - (((phase) - (x)) & -(align))) #define P2BOUNDARY(off, len, align) (((off) ^ ((off) + (len) - 1)) > (align) - 1) +#define IS_P2ALIGNED(v, a) ((((uintptr_t)(v)) & ((uintptr_t)(a) - 1)) == 0) /* * General-purpose 32-bit and 64-bit bitfield encodings. @@ -498,19 +499,7 @@ typedef struct zio_eck { * Gang block headers are self-checksumming and contain an array * of block pointers. */ -#define SPA_GANGBLOCKSIZE SPA_MINBLOCKSIZE -#define SPA_GBH_NBLKPTRS ((SPA_GANGBLOCKSIZE - \ - sizeof (zio_eck_t)) / sizeof (blkptr_t)) -#define SPA_GBH_FILLER ((SPA_GANGBLOCKSIZE - \ - sizeof (zio_eck_t) - \ - (SPA_GBH_NBLKPTRS * sizeof (blkptr_t))) /\ - sizeof (uint64_t)) - -typedef struct zio_gbh { - blkptr_t zg_blkptr[SPA_GBH_NBLKPTRS]; - uint64_t zg_filler[SPA_GBH_FILLER]; - zio_eck_t zg_tail; -} zio_gbh_phys_t; +#define SPA_OLD_GANGBLOCKSIZE SPA_MINBLOCKSIZE #define VDEV_RAIDZ_MAXPARITY 3 diff --git a/sys/contrib/vchiq/interface/vchiq_arm/vchiq_2835_arm.c b/sys/contrib/vchiq/interface/vchiq_arm/vchiq_2835_arm.c index f4aee12dec53..185e81e71bdc 100644 --- a/sys/contrib/vchiq/interface/vchiq_arm/vchiq_2835_arm.c +++ b/sys/contrib/vchiq/interface/vchiq_arm/vchiq_2835_arm.c @@ -546,7 +546,9 @@ free_pagelist(BULKINFO_T *bi, int actual) pagelist = bi->pagelist; vchiq_log_trace(vchiq_arm_log_level, - "free_pagelist - %zx, %d (%u bytes @%p)", (size_t)pagelist, actual, pagelist->length, bi->buf); + "free_pagelist - %zx, %d (%lu bytes @%p)", + (size_t)pagelist, (int)actual, (unsigned long)pagelist->length, + bi->buf); num_pages = (pagelist->length + pagelist->offset + PAGE_SIZE - 1) / diff --git a/sys/dev/amdtemp/amdtemp.c b/sys/dev/amdtemp/amdtemp.c index 79ccdc8c79fb..b1ecb014a2b0 100644 --- a/sys/dev/amdtemp/amdtemp.c +++ b/sys/dev/amdtemp/amdtemp.c @@ -642,7 +642,7 @@ amdtemp_intrhook(void *arg) OID_AUTO, "temperature", CTLTYPE_INT | CTLFLAG_RD | CTLFLAG_MPSAFE, dev, sensor, amdtemp_sysctl, "IK", - "Current temparature"); + "Current temperature"); } } if (sc->sc_ich.ich_arg != NULL) diff --git a/sys/dev/ath/ath_hal/ar5416/ar5416_cal_iq.c b/sys/dev/ath/ath_hal/ar5416/ar5416_cal_iq.c index a0c27828bbc1..50d7cc0aa7a8 100644 --- a/sys/dev/ath/ath_hal/ar5416/ar5416_cal_iq.c +++ b/sys/dev/ath/ath_hal/ar5416/ar5416_cal_iq.c @@ -76,7 +76,7 @@ ar5416IQCalibration(struct ath_hal *ah, uint8_t numChains) HALDEBUG(ah, HAL_DEBUG_PERCAL, "Start IQ Cal and Correction for Chain %d\n", i); HALDEBUG(ah, HAL_DEBUG_PERCAL, - "Orignal: iq_corr_meas = 0x%08x\n", iqCorrMeas); + "Original: iq_corr_meas = 0x%08x\n", iqCorrMeas); iqCorrNeg = 0; /* iqCorrMeas is always negative. */ diff --git a/sys/dev/bwi/if_bwi.c b/sys/dev/bwi/if_bwi.c index 85146d4c4010..80fc5e9e47af 100644 --- a/sys/dev/bwi/if_bwi.c +++ b/sys/dev/bwi/if_bwi.c @@ -2900,7 +2900,7 @@ bwi_plcp_header(const struct ieee80211_rate_table *rt, else if (modtype == IEEE80211_T_DS) bwi_ds_plcp_header(plcp, pkt_len, rate); else - panic("unsupport modulation type %u\n", modtype); + panic("unsupported modulation type %u\n", modtype); } static int diff --git a/sys/dev/cesa/cesa.c b/sys/dev/cesa/cesa.c index 405b619d6e5b..7439dcdbc1ee 100644 --- a/sys/dev/cesa/cesa.c +++ b/sys/dev/cesa/cesa.c @@ -286,7 +286,7 @@ cesa_alloc_tdesc(struct cesa_softc *sc) CESA_GENERIC_ALLOC_LOCKED(sc, ctd, tdesc); if (!ctd) - device_printf(sc->sc_dev, "TDMA descriptors pool exhaused. " + device_printf(sc->sc_dev, "TDMA descriptors pool exhausted. " "Consider increasing CESA_TDMA_DESCRIPTORS.\n"); return (ctd); @@ -299,7 +299,7 @@ cesa_alloc_sdesc(struct cesa_softc *sc, struct cesa_request *cr) CESA_GENERIC_ALLOC_LOCKED(sc, csd, sdesc); if (!csd) { - device_printf(sc->sc_dev, "SA descriptors pool exhaused. " + device_printf(sc->sc_dev, "SA descriptors pool exhausted. " "Consider increasing CESA_SA_DESCRIPTORS.\n"); return (NULL); } diff --git a/sys/dev/ciss/ciss.c b/sys/dev/ciss/ciss.c index d4ede91f6b35..dd3df631119a 100644 --- a/sys/dev/ciss/ciss.c +++ b/sys/dev/ciss/ciss.c @@ -3140,7 +3140,7 @@ ciss_cam_action(struct cam_sim *sim, union ccb *ccb) } default: /* we can't do this */ - debug(1, "unspported func_code = 0x%x", ccb->ccb_h.func_code); + debug(1, "unsupported func_code = 0x%x", ccb->ccb_h.func_code); ccb->ccb_h.status = CAM_REQ_INVALID; break; } diff --git a/sys/dev/igc/if_igc.c b/sys/dev/igc/if_igc.c index f199a128c783..d6c06803990f 100644 --- a/sys/dev/igc/if_igc.c +++ b/sys/dev/igc/if_igc.c @@ -2816,7 +2816,7 @@ igc_add_hw_stats(struct igc_softc *sc) "Oversized Packets Received"); SYSCTL_ADD_UQUAD(ctx, stat_list, OID_AUTO, "recv_jabber", CTLFLAG_RD, &sc->stats.rjc, - "Recevied Jabber"); + "Received Jabber"); SYSCTL_ADD_UQUAD(ctx, stat_list, OID_AUTO, "recv_errs", CTLFLAG_RD, &sc->stats.rxerrc, "Receive Errors"); diff --git a/sys/dev/isci/scil/scic_sds_remote_node_context.c b/sys/dev/isci/scil/scic_sds_remote_node_context.c index aa1e8d840282..42dd81aa1874 100644 --- a/sys/dev/isci/scil/scic_sds_remote_node_context.c +++ b/sys/dev/isci/scil/scic_sds_remote_node_context.c @@ -663,7 +663,7 @@ SCI_STATUS scic_sds_remote_node_context_invalidating_state_event_handler( SCIC_LOG_OBJECT_SSP_REMOTE_TARGET | SCIC_LOG_OBJECT_SMP_REMOTE_TARGET | SCIC_LOG_OBJECT_STP_REMOTE_TARGET, - "SCIC Remote Node Context 0x%x was suspeneded by hardware while being invalidated.\n", + "SCIC Remote Node Context 0x%x was suspended by hardware while being invalidated.\n", this_rnc )); status = SCI_SUCCESS; @@ -718,7 +718,7 @@ SCI_STATUS scic_sds_remote_node_context_resuming_state_event_handler( SCIC_LOG_OBJECT_SSP_REMOTE_TARGET | SCIC_LOG_OBJECT_SMP_REMOTE_TARGET | SCIC_LOG_OBJECT_STP_REMOTE_TARGET, - "SCIC Remote Node Context 0x%x was suspeneded by hardware while being resumed.\n", + "SCIC Remote Node Context 0x%x was suspended by hardware while being resumed.\n", this_rnc )); status = SCI_SUCCESS; diff --git a/sys/dev/mfi/mfi.c b/sys/dev/mfi/mfi.c index 13e5dfc84fd1..a7d98f06aea3 100644 --- a/sys/dev/mfi/mfi.c +++ b/sys/dev/mfi/mfi.c @@ -2829,7 +2829,7 @@ mfi_check_command_post(struct mfi_softc *sc, struct mfi_command *cm) if (ld->ld_id == cm->cm_frame->dcmd.mbox[0]) break; } - KASSERT(ld != NULL, ("volume dissappeared")); + KASSERT(ld != NULL, ("volume disappeared")); if (cm->cm_frame->header.cmd_status == MFI_STAT_OK) { mtx_unlock(&sc->mfi_io_lock); bus_topo_lock(); diff --git a/sys/dev/mlx5/mlx5_en/en_hw_tls.h b/sys/dev/mlx5/mlx5_en/en_hw_tls.h index d637314e040e..cd57d2ac5f72 100644 --- a/sys/dev/mlx5/mlx5_en/en_hw_tls.h +++ b/sys/dev/mlx5/mlx5_en/en_hw_tls.h @@ -82,6 +82,8 @@ struct mlx5e_tls { struct sysctl_ctx_list ctx; struct mlx5e_tls_stats stats; struct workqueue_struct *wq; + struct workqueue_struct *prealloc_wq; + struct work_struct prealloc_work; uma_zone_t zone; uint32_t max_resources; /* max number of resources */ int zone_max; @@ -92,6 +94,7 @@ struct mlx5e_tls { int mlx5e_tls_init(struct mlx5e_priv *); void mlx5e_tls_cleanup(struct mlx5e_priv *); int mlx5e_sq_tls_xmit(struct mlx5e_sq *, struct mlx5e_xmit_args *, struct mbuf **); +void mlx5e_tls_prealloc_tags(struct mlx5e_priv *priv); if_snd_tag_alloc_t mlx5e_tls_snd_tag_alloc; diff --git a/sys/dev/mlx5/mlx5_en/mlx5_en_hw_tls.c b/sys/dev/mlx5/mlx5_en/mlx5_en_hw_tls.c index 6c83de5f3580..851316ccfcd7 100644 --- a/sys/dev/mlx5/mlx5_en/mlx5_en_hw_tls.c +++ b/sys/dev/mlx5/mlx5_en/mlx5_en_hw_tls.c @@ -80,23 +80,39 @@ static const char *mlx5e_tls_stats_desc[] = { }; static void mlx5e_tls_work(struct work_struct *); +static void mlx5e_tls_prealloc_work(struct work_struct *); /* - * Expand the tls tag UMA zone in a sleepable context + * Expand the tls tag UMA zone in an async context */ static void -mlx5e_prealloc_tags(struct mlx5e_priv *priv, int nitems) +mlx5e_tls_prealloc_work(struct work_struct *work) { + struct mlx5e_priv *priv; + struct mlx5e_tls *ptls; struct mlx5e_tls_tag **tags; - int i; + int i, nitems; + + ptls = container_of(work, struct mlx5e_tls, prealloc_work); + priv = container_of(ptls, struct mlx5e_priv, tls); + nitems = ptls->zone_max; tags = malloc(sizeof(tags[0]) * nitems, - M_MLX5E_TLS, M_WAITOK); - for (i = 0; i < nitems; i++) - tags[i] = uma_zalloc(priv->tls.zone, M_WAITOK); + M_MLX5E_TLS, M_WAITOK | M_ZERO); + for (i = 0; i < nitems; i++) { + tags[i] = uma_zalloc(priv->tls.zone, M_NOWAIT); + /* + * If the allocation fails, its likely we are competing + * with real consumers of tags and the zone is full, + * so exit the loop, and release the tags like we would + * if we allocated all "nitems" + */ + if (tags[i] == NULL) + break; + } __compiler_membar(); - for (i = 0; i < nitems; i++) + for (i = 0; i < nitems && tags[i] != NULL; i++) uma_zfree(priv->tls.zone, tags[i]); free(tags, M_MLX5E_TLS); } @@ -244,8 +260,6 @@ mlx5e_tls_init(struct mlx5e_priv *priv) } uma_zone_set_max(ptls->zone, ptls->zone_max); - if (prealloc_tags != 0) - mlx5e_prealloc_tags(priv, ptls->zone_max); for (x = 0; x != MLX5E_TLS_STATS_NUM; x++) ptls->stats.arg[x] = counter_u64_alloc(M_WAITOK); @@ -271,6 +285,23 @@ mlx5e_tls_init(struct mlx5e_priv *priv) } void +mlx5e_tls_prealloc_tags(struct mlx5e_priv *priv) +{ + struct mlx5e_tls *ptls = &priv->tls; + int prealloc_tags = 0; + + if (ptls->prealloc_wq != NULL) + return; + + TUNABLE_INT_FETCH("hw.mlx5.tls_prealloc_tags", &prealloc_tags); + if (prealloc_tags == 0) + return; + ptls->prealloc_wq = create_singlethread_workqueue("mlx5-tls-prealloc_wq"); + INIT_WORK(&ptls->prealloc_work, mlx5e_tls_prealloc_work); + queue_work(ptls->prealloc_wq, &ptls->prealloc_work); +} + +void mlx5e_tls_cleanup(struct mlx5e_priv *priv) { struct mlx5e_tls *ptls = &priv->tls; @@ -280,6 +311,10 @@ mlx5e_tls_cleanup(struct mlx5e_priv *priv) return; ptls->init = 0; + if (ptls->prealloc_wq != NULL) { + flush_workqueue(ptls->prealloc_wq); + destroy_workqueue(ptls->prealloc_wq); + } flush_workqueue(ptls->wq); sysctl_ctx_free(&ptls->ctx); uma_zdestroy(ptls->zone); diff --git a/sys/dev/mlx5/mlx5_en/mlx5_en_main.c b/sys/dev/mlx5/mlx5_en/mlx5_en_main.c index f83506bda1aa..ee9c53bb0a60 100644 --- a/sys/dev/mlx5/mlx5_en/mlx5_en_main.c +++ b/sys/dev/mlx5/mlx5_en/mlx5_en_main.c @@ -3335,6 +3335,9 @@ mlx5e_open_locked(if_t ifp) mlx5e_update_carrier(priv); + if ((if_getcapenable(ifp) & (IFCAP_TXTLS4 | IFCAP_TXTLS6)) != 0) + mlx5e_tls_prealloc_tags(priv); + return (0); err_close_channels: diff --git a/sys/dev/mmc/mmc_fdt_helpers.c b/sys/dev/mmc/mmc_fdt_helpers.c index aed85dab55f4..980785464a00 100644 --- a/sys/dev/mmc/mmc_fdt_helpers.c +++ b/sys/dev/mmc/mmc_fdt_helpers.c @@ -160,6 +160,17 @@ cd_setup(struct mmc_helper *helper, phandle_t node) } /* + * If the device has no card-detection, treat it as non-removable. + * This could be improved by polling for detection. + */ + if (helper->props & MMC_PROP_BROKEN_CD) { + helper->cd_disabled = true; + if (bootverbose) + device_printf(dev, "Broken card-detect\n"); + return; + } + + /* * If there is no cd-gpios property, then presumably the hardware * PRESENT_STATE register and interrupts will reflect card state * properly, and there's nothing more for us to do. Our get_present() diff --git a/sys/dev/mmc/mmcsd.c b/sys/dev/mmc/mmcsd.c index 5b9cb93c7b31..f2965048b285 100644 --- a/sys/dev/mmc/mmcsd.c +++ b/sys/dev/mmc/mmcsd.c @@ -1422,7 +1422,7 @@ mmcsd_task(void *arg) struct mmcsd_softc *sc; struct bio *bp; device_t dev, mmcbus; - int bio_error, err, sz; + int abio_error, err, sz; part = arg; sc = part->sc; @@ -1430,7 +1430,7 @@ mmcsd_task(void *arg) mmcbus = sc->mmcbus; while (1) { - bio_error = 0; + abio_error = 0; MMCSD_DISK_LOCK(part); do { if (part->running == 0) @@ -1475,11 +1475,11 @@ mmcsd_task(void *arg) } else if (bp->bio_cmd == BIO_DELETE) block = mmcsd_delete(part, bp); else - bio_error = EOPNOTSUPP; + abio_error = EOPNOTSUPP; release: MMCBUS_RELEASE_BUS(mmcbus, dev); if (block < end) { - bp->bio_error = (bio_error == 0) ? EIO : bio_error; + bp->bio_error = (abio_error == 0) ? EIO : abio_error; bp->bio_resid = (end - block) * sz; bp->bio_flags |= BIO_ERROR; } else diff --git a/sys/dev/nvme/nvme_ns.c b/sys/dev/nvme/nvme_ns.c index f4a588373c98..17684cc14ba2 100644 --- a/sys/dev/nvme/nvme_ns.c +++ b/sys/dev/nvme/nvme_ns.c @@ -45,7 +45,7 @@ #include "nvme_private.h" #include "nvme_linux.h" -static void nvme_bio_child_inbed(struct bio *parent, int bio_error); +static void nvme_bio_child_inbed(struct bio *parent, int abio_error); static void nvme_bio_child_done(void *arg, const struct nvme_completion *cpl); static uint32_t nvme_get_num_segments(uint64_t addr, uint64_t size, @@ -275,14 +275,14 @@ nvme_ns_bio_done(void *arg, const struct nvme_completion *status) } static void -nvme_bio_child_inbed(struct bio *parent, int bio_error) +nvme_bio_child_inbed(struct bio *parent, int abio_error) { struct nvme_completion parent_cpl; int children, inbed; - if (bio_error != 0) { + if (abio_error != 0) { parent->bio_flags |= BIO_ERROR; - parent->bio_error = bio_error; + parent->bio_error = abio_error; } /* @@ -309,12 +309,12 @@ nvme_bio_child_done(void *arg, const struct nvme_completion *cpl) { struct bio *child = arg; struct bio *parent; - int bio_error; + int abio_error; parent = child->bio_parent; g_destroy_bio(child); - bio_error = nvme_completion_is_error(cpl) ? EIO : 0; - nvme_bio_child_inbed(parent, bio_error); + abio_error = nvme_completion_is_error(cpl) ? EIO : 0; + nvme_bio_child_inbed(parent, abio_error); } static uint32_t diff --git a/sys/dev/oce/oce_sysctl.c b/sys/dev/oce/oce_sysctl.c index 1b903d8d027a..544bec9438c4 100644 --- a/sys/dev/oce/oce_sysctl.c +++ b/sys/dev/oce/oce_sysctl.c @@ -716,7 +716,7 @@ oce_add_stats_sysctls_be3(POCE_SOFTC sc, "Total Received Bytes"); SYSCTL_ADD_UINT(ctx, rx_stat_list, OID_AUTO, "total_frags", CTLFLAG_RD, &stats->rx.t_rx_frags, 0, - "Total Received Fragements"); + "Total Received Fragments"); SYSCTL_ADD_UINT(ctx, rx_stat_list, OID_AUTO, "total_mcast_pkts", CTLFLAG_RD, &stats->rx.t_rx_mcast_pkts, 0, "Total Received Multicast Packets"); @@ -748,7 +748,7 @@ oce_add_stats_sysctls_be3(POCE_SOFTC sc, "Receive Packets"); SYSCTL_ADD_QUAD(ctx, queue_stats_list, OID_AUTO, "rx_bytes", CTLFLAG_RD, &sc->rq[i]->rx_stats.rx_bytes, - "Recived Bytes"); + "Received Bytes"); SYSCTL_ADD_UINT(ctx, queue_stats_list, OID_AUTO, "rx_frags", CTLFLAG_RD, &sc->rq[i]->rx_stats.rx_frags, 0, "Received Fragments"); @@ -786,7 +786,7 @@ oce_add_stats_sysctls_be3(POCE_SOFTC sc, "ERX Errors"); SYSCTL_ADD_UINT(ctx, rx_stat_list, OID_AUTO, "alignment_errors", CTLFLAG_RD, &stats->u0.be.rx_drops_too_many_frags, 0, - "RX Alignmnet Errors"); + "RX Alignment Errors"); SYSCTL_ADD_UINT(ctx, rx_stat_list, OID_AUTO, "in_range_errors", CTLFLAG_RD, &stats->u0.be.rx_in_range_errors, 0, "In Range Errors"); @@ -932,7 +932,7 @@ oce_add_stats_sysctls_xe201(POCE_SOFTC sc, "Total Received Bytes"); SYSCTL_ADD_UINT(ctx, rx_stat_list, OID_AUTO, "total_frags", CTLFLAG_RD, &stats->rx.t_rx_frags, 0, - "Total Received Fragements"); + "Total Received Fragments"); SYSCTL_ADD_UINT(ctx, rx_stat_list, OID_AUTO, "total_mcast_pkts", CTLFLAG_RD, &stats->rx.t_rx_mcast_pkts, 0, "Total Received Multicast Packets"); @@ -961,7 +961,7 @@ oce_add_stats_sysctls_xe201(POCE_SOFTC sc, "Receive Packets"); SYSCTL_ADD_QUAD(ctx, queue_stats_list, OID_AUTO, "rx_bytes", CTLFLAG_RD, &sc->rq[i]->rx_stats.rx_bytes, - "Recived Bytes"); + "Received Bytes"); SYSCTL_ADD_UINT(ctx, queue_stats_list, OID_AUTO, "rx_frags", CTLFLAG_RD, &sc->rq[i]->rx_stats.rx_frags, 0, "Received Fragments"); @@ -989,7 +989,7 @@ oce_add_stats_sysctls_xe201(POCE_SOFTC sc, "CRC Errors"); SYSCTL_ADD_UQUAD(ctx, rx_stat_list, OID_AUTO, "alignment_errors", CTLFLAG_RD, &stats->u0.xe201.rx_alignment_errors, - "RX Alignmnet Errors"); + "RX Alignment Errors"); SYSCTL_ADD_UINT(ctx, rx_stat_list, OID_AUTO, "in_range_errors", CTLFLAG_RD, &stats->u0.xe201.rx_in_range_errors, 0, "In Range Errors"); diff --git a/sys/dev/random/fenestrasX/fx_pool.c b/sys/dev/random/fenestrasX/fx_pool.c index 8e63b345a1bd..b6ffc202769e 100644 --- a/sys/dev/random/fenestrasX/fx_pool.c +++ b/sys/dev/random/fenestrasX/fx_pool.c @@ -127,7 +127,7 @@ static const struct fxrng_ent_cls fxrng_garbage = { */ static const struct fxrng_ent_char { const struct fxrng_ent_cls *entc_cls; -} fxrng_ent_char[ENTROPYSOURCE] = { +} fxrng_ent_char[/*ENTROPYSOURCE*/] = { [RANDOM_CACHED] = { .entc_cls = &fxrng_hi_push, }, @@ -213,6 +213,7 @@ static const struct fxrng_ent_char { .entc_cls = &fxrng_hi_pull, }, }; +CTASSERT(nitems(fxrng_ent_char) == ENTROPYSOURCE); /* Useful for single-bit-per-source state. */ BITSET_DEFINE(fxrng_bits, ENTROPYSOURCE); diff --git a/sys/dev/random/random_harvestq.c b/sys/dev/random/random_harvestq.c index 643dbac1fc8b..b591ffd3b544 100644 --- a/sys/dev/random/random_harvestq.c +++ b/sys/dev/random/random_harvestq.c @@ -88,7 +88,7 @@ static void random_sources_feed(void); static __read_mostly bool epoch_inited; static __read_mostly epoch_t rs_epoch; -static const char *random_source_descr[ENTROPYSOURCE]; +static const char *random_source_descr[]; /* * How many events to queue up. We create this many items in @@ -109,6 +109,7 @@ volatile int random_kthread_control; * Updates are synchronized by the harvest mutex. */ __read_frequently u_int hc_source_mask; +CTASSERT(ENTROPYSOURCE <= sizeof(hc_source_mask) * NBBY); struct random_sources { CK_LIST_ENTRY(random_sources) rrs_entries; @@ -647,7 +648,7 @@ SYSCTL_PROC(_kern_random_harvest, OID_AUTO, mask_bin, random_print_harvestmask, "A", "Entropy harvesting mask (printable)"); -static const char *random_source_descr[ENTROPYSOURCE] = { +static const char *random_source_descr[/*ENTROPYSOURCE*/] = { [RANDOM_CACHED] = "CACHED", [RANDOM_ATTACH] = "ATTACH", [RANDOM_KEYBOARD] = "KEYBOARD", @@ -678,6 +679,7 @@ static const char *random_source_descr[ENTROPYSOURCE] = { [RANDOM_PURE_ARM_TRNG] = "PURE_ARM_TRNG", /* "ENTROPYSOURCE" */ }; +CTASSERT(nitems(random_source_descr) == ENTROPYSOURCE); static int random_print_harvestmask_symbolic(SYSCTL_HANDLER_ARGS) diff --git a/sys/dev/thunderbolt/tb_pcib.c b/sys/dev/thunderbolt/tb_pcib.c index 00738984ad1c..bc4fc1ce00ec 100644 --- a/sys/dev/thunderbolt/tb_pcib.c +++ b/sys/dev/thunderbolt/tb_pcib.c @@ -557,8 +557,20 @@ static int tb_pci_probe(device_t dev) { struct tb_pcib_ident *n; + device_t parent; + devclass_t dc; - if ((n = tb_pcib_find_ident(device_get_parent(dev))) != NULL) { + /* + * This driver is only valid if the parent device is a PCI-PCI + * bridge. To determine that, check if the grandparent is a + * PCI bus. + */ + parent = device_get_parent(dev); + dc = device_get_devclass(device_get_parent(parent)); + if (strcmp(devclass_get_name(dc), "pci") != 0) + return (ENXIO); + + if ((n = tb_pcib_find_ident(parent)) != NULL) { switch (n->flags & TB_GEN_MASK) { case TB_GEN_TB1: device_set_desc(dev, "Thunderbolt 1 Link"); diff --git a/sys/dev/virtio/virtqueue.c b/sys/dev/virtio/virtqueue.c index cc7a233d60ee..41e01549c8b2 100644 --- a/sys/dev/virtio/virtqueue.c +++ b/sys/dev/virtio/virtqueue.c @@ -580,7 +580,8 @@ virtqueue_dequeue(struct virtqueue *vq, uint32_t *len) void *cookie; uint16_t used_idx, desc_idx; - if (vq->vq_used_cons_idx == vq_htog16(vq, vq->vq_ring.used->idx)) + if (vq->vq_used_cons_idx == + vq_htog16(vq, atomic_load_16(&vq->vq_ring.used->idx))) return (NULL); used_idx = vq->vq_used_cons_idx++ & (vq->vq_nentries - 1); diff --git a/sys/dev/vmm/vmm_dev.c b/sys/dev/vmm/vmm_dev.c index ebbceb25b69e..d6543bf6534e 100644 --- a/sys/dev/vmm/vmm_dev.c +++ b/sys/dev/vmm/vmm_dev.c @@ -14,9 +14,11 @@ #include <sys/kernel.h> #include <sys/malloc.h> #include <sys/mman.h> +#include <sys/module.h> #include <sys/priv.h> #include <sys/proc.h> #include <sys/queue.h> +#include <sys/smp.h> #include <sys/sx.h> #include <sys/sysctl.h> #include <sys/ucred.h> @@ -78,6 +80,8 @@ struct vmmdev_softc { int flags; }; +static bool vmm_initialized = false; + static SLIST_HEAD(, vmmdev_softc) head; static unsigned pr_allow_flag; @@ -88,6 +92,10 @@ static MALLOC_DEFINE(M_VMMDEV, "vmmdev", "vmmdev"); SYSCTL_DECL(_hw_vmm); +u_int vm_maxcpu; +SYSCTL_UINT(_hw_vmm, OID_AUTO, maxcpu, CTLFLAG_RDTUN | CTLFLAG_NOFETCH, + &vm_maxcpu, 0, "Maximum number of vCPUs"); + static void devmem_destroy(void *arg); static int devmem_create_cdev(struct vmmdev_softc *sc, int id, char *devmem); @@ -619,20 +627,16 @@ vmmdev_ioctl(struct cdev *cdev, u_long cmd, caddr_t data, int fflag, } error = domainset_populate(&domain, mask, mseg->ds_policy, mseg->ds_mask_size); - if (error) { - free(mask, M_VMMDEV); + free(mask, M_VMMDEV); + if (error) break; - } domainset = domainset_create(&domain); if (domainset == NULL) { error = EINVAL; - free(mask, M_VMMDEV); break; } - free(mask, M_VMMDEV); } error = alloc_memseg(sc, mseg, sizeof(mseg->name), domainset); - break; } case VM_GET_MEMSEG: @@ -985,6 +989,9 @@ vmmdev_create(const char *name, struct ucred *cred) struct vm *vm; int error; + if (name == NULL || strlen(name) > VM_MAX_NAMELEN) + return (EINVAL); + sx_xlock(&vmmdev_mtx); sc = vmmdev_lookup(name, cred); if (sc != NULL) { @@ -1025,6 +1032,9 @@ sysctl_vmm_create(SYSCTL_HANDLER_ARGS) char *buf; int error, buflen; + if (!vmm_initialized) + return (ENXIO); + error = vmm_priv_check(req->td->td_ucred); if (error != 0) return (error); @@ -1110,7 +1120,7 @@ static struct cdevsw vmmctlsw = { .d_ioctl = vmmctl_ioctl, }; -int +static int vmmdev_init(void) { int error; @@ -1126,7 +1136,7 @@ vmmdev_init(void) return (error); } -int +static int vmmdev_cleanup(void) { sx_xlock(&vmmdev_mtx); @@ -1144,6 +1154,71 @@ vmmdev_cleanup(void) } static int +vmm_handler(module_t mod, int what, void *arg) +{ + int error; + + switch (what) { + case MOD_LOAD: + error = vmmdev_init(); + if (error != 0) + break; + + vm_maxcpu = mp_ncpus; + TUNABLE_INT_FETCH("hw.vmm.maxcpu", &vm_maxcpu); + if (vm_maxcpu > VM_MAXCPU) { + printf("vmm: vm_maxcpu clamped to %u\n", VM_MAXCPU); + vm_maxcpu = VM_MAXCPU; + } + if (vm_maxcpu == 0) + vm_maxcpu = 1; + + error = vmm_modinit(); + if (error == 0) + vmm_initialized = true; + else { + error = vmmdev_cleanup(); + KASSERT(error == 0, + ("%s: vmmdev_cleanup failed: %d", __func__, error)); + } + break; + case MOD_UNLOAD: + error = vmmdev_cleanup(); + if (error == 0 && vmm_initialized) { + error = vmm_modcleanup(); + if (error) { + /* + * Something bad happened - prevent new + * VMs from being created + */ + vmm_initialized = false; + } + } + break; + default: + error = 0; + break; + } + return (error); +} + +static moduledata_t vmm_kmod = { + "vmm", + vmm_handler, + NULL +}; + +/* + * vmm initialization has the following dependencies: + * + * - Initialization requires smp_rendezvous() and therefore must happen + * after SMP is fully functional (after SI_SUB_SMP). + * - vmm device initialization requires an initialized devfs. + */ +DECLARE_MODULE(vmm, vmm_kmod, MAX(SI_SUB_SMP, SI_SUB_DEVFS) + 1, SI_ORDER_ANY); +MODULE_VERSION(vmm, 1); + +static int devmem_mmap_single(struct cdev *cdev, vm_ooffset_t *offset, vm_size_t len, struct vm_object **objp, int nprot) { diff --git a/sys/dev/vmm/vmm_dev.h b/sys/dev/vmm/vmm_dev.h index 2881a7063565..f14176c8afad 100644 --- a/sys/dev/vmm/vmm_dev.h +++ b/sys/dev/vmm/vmm_dev.h @@ -11,15 +11,19 @@ #include <sys/types.h> #include <sys/ioccom.h> + #include <machine/vmm_dev.h> +#include <dev/vmm/vmm_param.h> + #ifdef _KERNEL struct thread; struct vm; struct vcpu; -int vmmdev_init(void); -int vmmdev_cleanup(void); +int vmm_modinit(void); +int vmm_modcleanup(void); + int vmmdev_machdep_ioctl(struct vm *vm, struct vcpu *vcpu, u_long cmd, caddr_t data, int fflag, struct thread *td); @@ -53,6 +57,17 @@ struct vmmdev_ioctl { extern const struct vmmdev_ioctl vmmdev_machdep_ioctls[]; extern const size_t vmmdev_machdep_ioctl_count; +/* + * Upper limit on vm_maxcpu. Limited by use of uint16_t types for CPU counts as + * well as range of vpid values for VT-x on amd64 and by the capacity of + * cpuset_t masks. The call to new_unrhdr() in vpid_init() in vmx.c requires + * 'vm_maxcpu + 1 <= 0xffff', hence the '- 1' below. + */ +#define VM_MAXCPU MIN(0xffff - 1, CPU_SETSIZE) + +/* Maximum number of vCPUs in a single VM. */ +extern u_int vm_maxcpu; + #endif /* _KERNEL */ struct vmmctl_vm_create { diff --git a/sys/dev/vmm/vmm_param.h b/sys/dev/vmm/vmm_param.h new file mode 100644 index 000000000000..a5040eb0f58c --- /dev/null +++ b/sys/dev/vmm/vmm_param.h @@ -0,0 +1,33 @@ +/*- + * SPDX-License-Identifier: BSD-2-Clause + * + * Copyright (c) 2011 NetApp, Inc. + * All rights reserved. + */ + +#ifndef _DEV_VMM_PARAM_H_ +#define _DEV_VMM_PARAM_H_ + +/* + * The VM name has to fit into the pathname length constraints of devfs, + * governed primarily by SPECNAMELEN. The length is the total number of + * characters in the full path, relative to the mount point and not + * including any leading '/' characters. + * A prefix and a suffix are added to the name specified by the user. + * The prefix is usually "vmm/" or "vmm.io/", but can be a few characters + * longer for future use. + * The suffix is a string that identifies a bootrom image or some similar + * image that is attached to the VM. A separator character gets added to + * the suffix automatically when generating the full path, so it must be + * accounted for, reducing the effective length by 1. + * The effective length of a VM name is 229 bytes for FreeBSD 13 and 37 + * bytes for FreeBSD 12. A minimum length is set for safety and supports + * a SPECNAMELEN as small as 32 on old systems. + */ +#define VM_MAX_PREFIXLEN 10 +#define VM_MAX_SUFFIXLEN 15 +#define VM_MIN_NAMELEN 6 +#define VM_MAX_NAMELEN \ + (SPECNAMELEN - VM_MAX_PREFIXLEN - VM_MAX_SUFFIXLEN - 1) + +#endif /* !_DEV_VMM_PARAM_H_ */ diff --git a/sys/dev/vnic/nicvf_main.c b/sys/dev/vnic/nicvf_main.c index dd44e420c78f..59f7abeacdd5 100644 --- a/sys/dev/vnic/nicvf_main.c +++ b/sys/dev/vnic/nicvf_main.c @@ -1402,7 +1402,7 @@ nicvf_allocate_net_interrupts(struct nicvf *nic) /* MSI-X must be configured by now */ if (!nic->msix_enabled) { - device_printf(nic->dev, "Cannot alloacte queue interrups. " + device_printf(nic->dev, "Cannot alloacte queue interrupts. " "MSI-X interrupts disabled.\n"); return (ENXIO); } diff --git a/sys/fs/fuse/fuse_internal.c b/sys/fs/fuse/fuse_internal.c index 61fe2ed032f6..eba0a8a79ff3 100644 --- a/sys/fs/fuse/fuse_internal.c +++ b/sys/fs/fuse/fuse_internal.c @@ -1063,6 +1063,8 @@ fuse_internal_init_callback(struct fuse_ticket *tick, struct uio *uio) if (!fuse_libabi_geq(data, 7, 28)) fsess_set_notimpl(data->mp, FUSE_COPY_FILE_RANGE); + if (fuse_libabi_geq(data, 7, 33) && (fiio->flags & FUSE_SETXATTR_EXT)) + data->dataflags |= FSESS_SETXATTR_EXT; out: if (err) { fdata_set_dead(data); @@ -1115,7 +1117,8 @@ fuse_internal_send_init(struct fuse_data *data, struct thread *td) */ fiii->flags = FUSE_ASYNC_READ | FUSE_POSIX_LOCKS | FUSE_EXPORT_SUPPORT | FUSE_BIG_WRITES | FUSE_WRITEBACK_CACHE - | FUSE_NO_OPEN_SUPPORT | FUSE_NO_OPENDIR_SUPPORT; + | FUSE_NO_OPEN_SUPPORT | FUSE_NO_OPENDIR_SUPPORT + | FUSE_SETXATTR_EXT; fuse_insert_callback(fdi.tick, fuse_internal_init_callback); fuse_insert_message(fdi.tick, false); diff --git a/sys/fs/fuse/fuse_ipc.h b/sys/fs/fuse/fuse_ipc.h index 3bfc859dbac9..d9d79f38c269 100644 --- a/sys/fs/fuse/fuse_ipc.h +++ b/sys/fs/fuse/fuse_ipc.h @@ -243,6 +243,7 @@ struct fuse_data { #define FSESS_MNTOPTS_MASK ( \ FSESS_DAEMON_CAN_SPY | FSESS_PUSH_SYMLINKS_IN | \ FSESS_DEFAULT_PERMISSIONS | FSESS_INTR) +#define FSESS_SETXATTR_EXT 0x8000000 /* extended fuse_setxattr_in */ extern int fuse_data_cache_mode; diff --git a/sys/fs/fuse/fuse_vnops.c b/sys/fs/fuse/fuse_vnops.c index 97aa23bfb0b0..ef5aee5de34c 100644 --- a/sys/fs/fuse/fuse_vnops.c +++ b/sys/fs/fuse/fuse_vnops.c @@ -625,7 +625,7 @@ fuse_vnop_allocate(struct vop_allocate_args *ap) return (EROFS); if (fsess_not_impl(mp, FUSE_FALLOCATE)) - return (EXTERROR(EINVAL, "This server does not implement " + return (EXTERROR(EOPNOTSUPP, "This server does not implement " "FUSE_FALLOCATE")); io.uio_offset = *offset; @@ -656,14 +656,14 @@ fuse_vnop_allocate(struct vop_allocate_args *ap) if (err == ENOSYS) { fsess_set_notimpl(mp, FUSE_FALLOCATE); - err = EXTERROR(EINVAL, "This server does not implement " + err = EXTERROR(EOPNOTSUPP, "This server does not implement " "FUSE_ALLOCATE"); } else if (err == EOPNOTSUPP) { /* * The file system server does not support FUSE_FALLOCATE with * the supplied mode for this particular file. */ - err = EXTERROR(EINVAL, "This file can't be pre-allocated"); + err = EXTERROR(EOPNOTSUPP, "This file can't be pre-allocated"); } else if (!err) { *offset += *len; *len = 0; @@ -2777,7 +2777,7 @@ fuse_vnop_setextattr(struct vop_setextattr_args *ap) strlen(ap->a_name) + 1; /* older FUSE servers use a smaller fuse_setxattr_in struct*/ - if (fuse_libabi_geq(fuse_get_mpdata(mp), 7, 33)) + if (fuse_get_mpdata(mp)->dataflags & FSESS_SETXATTR_EXT) struct_size = sizeof(*set_xattr_in); fdisp_init(&fdi, len + struct_size + uio->uio_resid); @@ -2786,7 +2786,7 @@ fuse_vnop_setextattr(struct vop_setextattr_args *ap) set_xattr_in = fdi.indata; set_xattr_in->size = uio->uio_resid; - if (fuse_libabi_geq(fuse_get_mpdata(mp), 7, 33)) { + if (fuse_get_mpdata(mp)->dataflags & FSESS_SETXATTR_EXT) { set_xattr_in->setxattr_flags = 0; set_xattr_in->padding = 0; } diff --git a/sys/fs/nfs/nfs_commonsubs.c b/sys/fs/nfs/nfs_commonsubs.c index f580a394a735..707ad5749ab2 100644 --- a/sys/fs/nfs/nfs_commonsubs.c +++ b/sys/fs/nfs/nfs_commonsubs.c @@ -658,7 +658,7 @@ nfscl_fillsattr(struct nfsrv_descript *nd, struct vattr *vap, NFSSETBIT_ATTRBIT(&attrbits, NFSATTRBIT_TIMECREATE); (void) nfsv4_fillattr(nd, vp->v_mount, vp, NULL, vap, NULL, 0, &attrbits, NULL, NULL, 0, 0, 0, 0, (uint64_t)0, NULL, - false, false, false, 0); + false, false, false, 0, NULL, false); break; } } @@ -1706,11 +1706,18 @@ nfsv4_loadattr(struct nfsrv_descript *nd, vnode_t vp, attrsum += NFSX_UNSIGNED; break; case NFSATTRBIT_CASEINSENSITIVE: - NFSM_DISSECT(tl, u_int32_t *, NFSX_UNSIGNED); + NFSM_DISSECT(tl, uint32_t *, NFSX_UNSIGNED); if (compare) { if (!(*retcmpp)) { - if (*tl != newnfs_false) - *retcmpp = NFSERR_NOTSAME; + if (vp == NULL || VOP_PATHCONF(vp, + _PC_CASE_INSENSITIVE, + &has_pathconf) != 0) + has_pathconf = 0; + if ((has_pathconf != 0 && + *tl != newnfs_true) || + (has_pathconf == 0 && + *tl != newnfs_false)) + *retcmpp = NFSERR_NOTSAME; } } else if (pc != NULL) { pc->pc_caseinsensitive = @@ -2690,7 +2697,8 @@ nfsv4_fillattr(struct nfsrv_descript *nd, struct mount *mp, vnode_t vp, nfsattrbit_t *attrbitp, struct ucred *cred, NFSPROC_T *p, int isdgram, int reterr, int supports_nfsv4acls, int at_root, uint64_t mounted_on_fileno, struct statfs *pnfssf, bool xattrsupp, bool has_hiddensystem, - bool has_namedattr, uint32_t clone_blksize) + bool has_namedattr, uint32_t clone_blksize, fsid_t *fsidp, + bool has_caseinsensitive) { int bitpos, retnum = 0; u_int32_t *tl; @@ -2865,10 +2873,12 @@ nfsv4_fillattr(struct nfsrv_descript *nd, struct mount *mp, vnode_t vp, break; case NFSATTRBIT_FSID: NFSM_BUILD(tl, u_int32_t *, NFSX_V4FSID); + if (fsidp == NULL) + fsidp = &mp->mnt_stat.f_fsid; *tl++ = 0; - *tl++ = txdr_unsigned(mp->mnt_stat.f_fsid.val[0]); + *tl++ = txdr_unsigned(fsidp->val[0]); *tl++ = 0; - *tl = txdr_unsigned(mp->mnt_stat.f_fsid.val[1]); + *tl = txdr_unsigned(fsidp->val[1]); retnum += NFSX_V4FSID; break; case NFSATTRBIT_UNIQUEHANDLES: @@ -2914,8 +2924,11 @@ nfsv4_fillattr(struct nfsrv_descript *nd, struct mount *mp, vnode_t vp, retnum += NFSX_UNSIGNED; break; case NFSATTRBIT_CASEINSENSITIVE: - NFSM_BUILD(tl, u_int32_t *, NFSX_UNSIGNED); - *tl = newnfs_false; + NFSM_BUILD(tl, uint32_t *, NFSX_UNSIGNED); + if (has_caseinsensitive) + *tl = newnfs_true; + else + *tl = newnfs_false; retnum += NFSX_UNSIGNED; break; case NFSATTRBIT_CASEPRESERVING: diff --git a/sys/fs/nfs/nfs_var.h b/sys/fs/nfs/nfs_var.h index 16a76c060e78..7db3952ecf5c 100644 --- a/sys/fs/nfs/nfs_var.h +++ b/sys/fs/nfs/nfs_var.h @@ -398,7 +398,7 @@ void nfsrv_wcc(struct nfsrv_descript *, int, struct nfsvattr *, int, int nfsv4_fillattr(struct nfsrv_descript *, struct mount *, vnode_t, NFSACL_T *, struct vattr *, fhandle_t *, int, nfsattrbit_t *, struct ucred *, NFSPROC_T *, int, int, int, int, uint64_t, struct statfs *, bool, bool, - bool, uint32_t); + bool, uint32_t, fsid_t *, bool); void nfsrv_fillattr(struct nfsrv_descript *, struct nfsvattr *); struct mbuf *nfsrv_adj(struct mbuf *, int, int); void nfsrv_postopattr(struct nfsrv_descript *, int, struct nfsvattr *); @@ -740,7 +740,7 @@ int nfsvno_updfilerev(vnode_t, struct nfsvattr *, struct nfsrv_descript *, int nfsvno_fillattr(struct nfsrv_descript *, struct mount *, vnode_t, struct nfsvattr *, fhandle_t *, int, nfsattrbit_t *, struct ucred *, NFSPROC_T *, int, int, int, int, uint64_t, bool, bool, - bool, uint32_t); + bool, uint32_t, bool); int nfsrv_sattr(struct nfsrv_descript *, vnode_t, struct nfsvattr *, nfsattrbit_t *, NFSACL_T *, NFSPROC_T *); int nfsv4_sattr(struct nfsrv_descript *, vnode_t, struct nfsvattr *, nfsattrbit_t *, diff --git a/sys/fs/nfsclient/nfs_clrpcops.c b/sys/fs/nfsclient/nfs_clrpcops.c index b61218958550..f5deef183efb 100644 --- a/sys/fs/nfsclient/nfs_clrpcops.c +++ b/sys/fs/nfsclient/nfs_clrpcops.c @@ -5452,7 +5452,7 @@ nfsrpc_setaclrpc(vnode_t vp, struct ucred *cred, NFSPROC_T *p, NFSSETBIT_ATTRBIT(&attrbits, NFSATTRBIT_ACL); (void) nfsv4_fillattr(nd, vp->v_mount, vp, aclp, NULL, NULL, 0, &attrbits, NULL, NULL, 0, 0, 0, 0, (uint64_t)0, NULL, false, false, - false, 0); + false, 0, NULL, false); error = nfscl_request(nd, vp, p, cred); if (error) return (error); diff --git a/sys/fs/nfsclient/nfs_clstate.c b/sys/fs/nfsclient/nfs_clstate.c index aa9d01fc4632..712d49c7160c 100644 --- a/sys/fs/nfsclient/nfs_clstate.c +++ b/sys/fs/nfsclient/nfs_clstate.c @@ -3701,7 +3701,8 @@ nfscl_docb(struct nfsrv_descript *nd, NFSPROC_T *p) if (!error) (void) nfsv4_fillattr(nd, NULL, NULL, NULL, &va, NULL, 0, &rattrbits, NULL, p, 0, 0, 0, 0, - (uint64_t)0, NULL, false, false, false, 0); + (uint64_t)0, NULL, false, false, false, 0, + NULL, false); break; case NFSV4OP_CBRECALL: NFSCL_DEBUG(4, "cbrecall\n"); diff --git a/sys/fs/nfsclient/nfs_clvnops.c b/sys/fs/nfsclient/nfs_clvnops.c index f80cf30669ca..193d8b6cd5eb 100644 --- a/sys/fs/nfsclient/nfs_clvnops.c +++ b/sys/fs/nfsclient/nfs_clvnops.c @@ -3896,11 +3896,15 @@ nfs_allocate(struct vop_allocate_args *ap) mtx_lock(&nmp->nm_mtx); nmp->nm_privflag |= NFSMNTP_NOALLOCATE; mtx_unlock(&nmp->nm_mtx); - error = EINVAL; + error = EOPNOTSUPP; } } else { + /* + * Pre-v4.2 NFS server that doesn't support it, or a newer + * NFS server that has indicated that it doesn't support it. + */ mtx_unlock(&nmp->nm_mtx); - error = EINVAL; + error = EOPNOTSUPP; } if (attrflag != 0) { ret = nfscl_loadattrcache(&vp, &nfsva, NULL, 0, 1); @@ -4677,12 +4681,13 @@ nfs_pathconf(struct vop_pathconf_args *ap) clone_blksize = 0; if ((NFS_ISV34(vp) && (ap->a_name == _PC_LINK_MAX || ap->a_name == _PC_NAME_MAX || ap->a_name == _PC_CHOWN_RESTRICTED || - ap->a_name == _PC_NO_TRUNC)) || + ap->a_name == _PC_NO_TRUNC || + ap->a_name == _PC_CASE_INSENSITIVE)) || (NFS_ISV4(vp) && (ap->a_name == _PC_ACL_NFS4 || ap->a_name == _PC_HAS_NAMEDATTR || ap->a_name == _PC_CLONE_BLKSIZE))) { /* - * Since only the above 4 a_names are returned by the NFSv3 + * Since only the above 5 a_names are returned by the NFSv3 * Pathconf RPC, there is no point in doing it for others. * For NFSv4, the Pathconf RPC (actually a Getattr Op.) can * be used for _PC_ACL_NFS4, _PC_HAS_NAMEDATTR and @@ -4849,6 +4854,9 @@ nfs_pathconf(struct vop_pathconf_args *ap) case _PC_CLONE_BLKSIZE: *ap->a_retval = clone_blksize; break; + case _PC_CASE_INSENSITIVE: + *ap->a_retval = pc.pc_caseinsensitive; + break; default: error = vop_stdpathconf(ap); diff --git a/sys/fs/nfsserver/nfs_nfsdport.c b/sys/fs/nfsserver/nfs_nfsdport.c index 9fe3f4426124..841ec2315f1c 100644 --- a/sys/fs/nfsserver/nfs_nfsdport.c +++ b/sys/fs/nfsserver/nfs_nfsdport.c @@ -2114,7 +2114,7 @@ nfsvno_fillattr(struct nfsrv_descript *nd, struct mount *mp, struct vnode *vp, struct ucred *cred, struct thread *p, int isdgram, int reterr, int supports_nfsv4acls, int at_root, uint64_t mounted_on_fileno, bool xattrsupp, bool has_hiddensystem, bool has_namedattr, - uint32_t clone_blksize) + uint32_t clone_blksize, bool has_caseinsensitive) { struct statfs *sf; int error; @@ -2135,7 +2135,7 @@ nfsvno_fillattr(struct nfsrv_descript *nd, struct mount *mp, struct vnode *vp, error = nfsv4_fillattr(nd, mp, vp, NULL, &nvap->na_vattr, fhp, rderror, attrbitp, cred, p, isdgram, reterr, supports_nfsv4acls, at_root, mounted_on_fileno, sf, xattrsupp, has_hiddensystem, has_namedattr, - clone_blksize); + clone_blksize, NULL, has_caseinsensitive); free(sf, M_TEMP); NFSEXITCODE2(0, nd); return (error); @@ -2468,7 +2468,7 @@ nfsrvd_readdirplus(struct nfsrv_descript *nd, int isdgram, int bextpg0, bextpg1, bextpgsiz0, bextpgsiz1; size_t atsiz; long pathval; - bool has_hiddensystem, has_namedattr, xattrsupp; + bool has_caseinsensitive, has_hiddensystem, has_namedattr, xattrsupp; if (nd->nd_repstat) { nfsrv_postopattr(nd, getret, &at); @@ -2949,6 +2949,7 @@ ateof: xattrsupp = false; has_hiddensystem = false; has_namedattr = false; + has_caseinsensitive = false; clone_blksize = 0; if (nvp != NULL) { supports_nfsv4acls = @@ -2978,6 +2979,11 @@ ateof: &pathval) != 0) pathval = 0; clone_blksize = pathval; + if (VOP_PATHCONF(nvp, + _PC_CASE_INSENSITIVE, + &pathval) != 0) + pathval = 0; + has_caseinsensitive = pathval > 0; NFSVOPUNLOCK(nvp); } else supports_nfsv4acls = 0; @@ -2999,7 +3005,7 @@ ateof: supports_nfsv4acls, at_root, mounted_on_fileno, xattrsupp, has_hiddensystem, has_namedattr, - clone_blksize); + clone_blksize, has_caseinsensitive); } else { dirlen += nfsvno_fillattr(nd, new_mp, nvp, nvap, &nfh, r, &attrbits, @@ -3007,7 +3013,7 @@ ateof: supports_nfsv4acls, at_root, mounted_on_fileno, xattrsupp, has_hiddensystem, has_namedattr, - clone_blksize); + clone_blksize, has_caseinsensitive); } if (nvp != NULL) vrele(nvp); @@ -6405,7 +6411,7 @@ nfsrv_setacldsdorpc(fhandle_t *fhp, struct ucred *cred, NFSPROC_T *p, * the same type (VREG). */ nfsv4_fillattr(nd, NULL, vp, aclp, NULL, NULL, 0, &attrbits, NULL, - NULL, 0, 0, 0, 0, 0, NULL, false, false, false, 0); + NULL, 0, 0, 0, 0, 0, NULL, false, false, false, 0, NULL, false); error = newnfs_request(nd, nmp, NULL, &nmp->nm_sockreq, NULL, p, cred, NFS_PROG, NFS_VER4, NULL, 1, NULL, NULL); if (error != 0) { diff --git a/sys/fs/nfsserver/nfs_nfsdserv.c b/sys/fs/nfsserver/nfs_nfsdserv.c index 67af0cf71175..394b63c2ab07 100644 --- a/sys/fs/nfsserver/nfs_nfsdserv.c +++ b/sys/fs/nfsserver/nfs_nfsdserv.c @@ -252,7 +252,7 @@ nfsrvd_getattr(struct nfsrv_descript *nd, int isdgram, struct thread *p = curthread; size_t atsiz; long pathval; - bool has_hiddensystem, has_namedattr, xattrsupp; + bool has_caseinsensitive, has_hiddensystem, has_namedattr, xattrsupp; uint32_t clone_blksize; if (nd->nd_repstat) @@ -336,6 +336,10 @@ nfsrvd_getattr(struct nfsrv_descript *nd, int isdgram, &pathval) != 0) pathval = 0; clone_blksize = pathval; + if (VOP_PATHCONF(vp, _PC_CASE_INSENSITIVE, + &pathval) != 0) + pathval = 0; + has_caseinsensitive = pathval > 0; mp = vp->v_mount; if (nfsrv_enable_crossmntpt != 0 && vp->v_type == VDIR && @@ -371,7 +375,8 @@ nfsrvd_getattr(struct nfsrv_descript *nd, int isdgram, isdgram, 1, supports_nfsv4acls, at_root, mounted_on_fileno, xattrsupp, has_hiddensystem, - has_namedattr, clone_blksize); + has_namedattr, clone_blksize, + has_caseinsensitive); vfs_unbusy(mp); } vrele(vp); diff --git a/sys/geom/geom_dev.c b/sys/geom/geom_dev.c index db0bc77a752f..a723d06334a0 100644 --- a/sys/geom/geom_dev.c +++ b/sys/geom/geom_dev.c @@ -734,6 +734,10 @@ g_dev_done(struct bio *bp2) g_trace(G_T_BIO, "g_dev_done(%p) had error %d", bp2, bp2->bio_error); bp->bio_flags |= BIO_ERROR; + if ((bp2->bio_flags & BIO_EXTERR) != 0) { + bp->bio_flags |= BIO_EXTERR; + bp->bio_exterr = bp2->bio_exterr; + } } else { if (bp->bio_cmd == BIO_READ) KNOTE_UNLOCKED(&sc->sc_selinfo.si_note, NOTE_READ); diff --git a/sys/geom/geom_disk.c b/sys/geom/geom_disk.c index 9dbf00371dba..b267130d1e0c 100644 --- a/sys/geom/geom_disk.c +++ b/sys/geom/geom_disk.c @@ -235,8 +235,14 @@ g_disk_done(struct bio *bp) bp2 = bp->bio_parent; binuptime(&now); mtx_lock(&sc->done_mtx); - if (bp2->bio_error == 0) - bp2->bio_error = bp->bio_error; + if (bp2->bio_error == 0) { + if ((bp->bio_flags & BIO_EXTERR) != 0) { + bp2->bio_flags |= BIO_EXTERR; + bp2->bio_exterr = bp->bio_exterr; + } else { + bp2->bio_error = bp->bio_error; + } + } bp2->bio_completed += bp->bio_length - bp->bio_resid; if (bp->bio_cmd == BIO_READ) diff --git a/sys/geom/geom_subr.c b/sys/geom/geom_subr.c index 2a6ce1ab6486..c5dce730da79 100644 --- a/sys/geom/geom_subr.c +++ b/sys/geom/geom_subr.c @@ -38,9 +38,11 @@ #include <sys/cdefs.h> #include "opt_ddb.h" +#define EXTERR_CATEGORY EXTERR_CAT_GEOM #include <sys/param.h> #include <sys/systm.h> #include <sys/devicestat.h> +#include <sys/exterrvar.h> #include <sys/kernel.h> #include <sys/malloc.h> #include <sys/bio.h> @@ -1162,8 +1164,14 @@ g_std_done(struct bio *bp) struct bio *bp2; bp2 = bp->bio_parent; - if (bp2->bio_error == 0) - bp2->bio_error = bp->bio_error; + if (bp2->bio_error == 0) { + if ((bp->bio_flags & BIO_EXTERR) != 0) { + bp2->bio_flags |= BIO_EXTERR; + bp2->bio_exterr = bp->bio_exterr; + } else { + bp2->bio_error = bp->bio_error; + } + } bp2->bio_completed += bp->bio_completed; g_destroy_bio(bp); bp2->bio_inbed++; @@ -1668,6 +1676,8 @@ DB_SHOW_COMMAND(bio, db_show_bio) db_printf(" caller2: %p\n", bp->bio_caller2); db_printf(" bio_from: %p\n", bp->bio_from); db_printf(" bio_to: %p\n", bp->bio_to); + if ((bp->bio_flags & BIO_EXTERR) != 0) + exterr_db_print(&bp->bio_exterr); #if defined(BUF_TRACKING) || defined(FULL_BUF_TRACKING) db_printf(" bio_track_bp: %p\n", bp->bio_track_bp); diff --git a/sys/geom/geom_vfs.c b/sys/geom/geom_vfs.c index 9b5e5a84191f..122e2f6a02ec 100644 --- a/sys/geom/geom_vfs.c +++ b/sys/geom/geom_vfs.c @@ -26,9 +26,11 @@ * SUCH DAMAGE. */ +#define EXTERR_CATEGORY EXTERR_CAT_GEOMVFS #include <sys/param.h> #include <sys/systm.h> #include <sys/bio.h> +#include <sys/exterrvar.h> #include <sys/kernel.h> #include <sys/lock.h> #include <sys/malloc.h> @@ -156,10 +158,13 @@ g_vfs_done(struct bio *bip) " suppressing further ENXIO"); } } - bp->b_error = bip->bio_error; bp->b_ioflags = bip->bio_flags; if (bip->bio_error) bp->b_ioflags |= BIO_ERROR; + if ((bp->b_ioflags & BIO_EXTERR) != 0) + bp->b_exterr = bip->bio_exterr; + else + bp->b_error = bip->bio_error; bp->b_resid = bp->b_bcount - bip->bio_completed; g_destroy_bio(bip); @@ -195,6 +200,8 @@ g_vfs_strategy(struct bufobj *bo, struct buf *bp) mtx_unlock(&sc->sc_mtx); bp->b_error = ENXIO; bp->b_ioflags |= BIO_ERROR; + EXTERROR_KE(&bp->b_exterr, ENXIO, + "orphaned or enxio active"); bufdone(bp); return; } diff --git a/sys/kern/kern_jail.c b/sys/kern/kern_jail.c index 267b60ffb5bc..523b7e314a10 100644 --- a/sys/kern/kern_jail.c +++ b/sys/kern/kern_jail.c @@ -3047,6 +3047,8 @@ do_jail_attach(struct thread *td, struct prison *pr, int drflags) setsugid(p); #ifdef RACCT racct_proc_ucred_changed(p, oldcred, newcred); +#endif +#ifdef RCTL crhold(newcred); #endif PROC_UNLOCK(p); diff --git a/sys/kern/kern_loginclass.c b/sys/kern/kern_loginclass.c index 55db6c28a1db..0c111c4f78d8 100644 --- a/sys/kern/kern_loginclass.c +++ b/sys/kern/kern_loginclass.c @@ -225,6 +225,8 @@ sys_setloginclass(struct thread *td, struct setloginclass_args *uap) proc_set_cred(p, newcred); #ifdef RACCT racct_proc_ucred_changed(p, oldcred, newcred); +#endif +#ifdef RCTL crhold(newcred); #endif PROC_UNLOCK(p); diff --git a/sys/kern/kern_prot.c b/sys/kern/kern_prot.c index a4c5bcc52529..3c145851b683 100644 --- a/sys/kern/kern_prot.c +++ b/sys/kern/kern_prot.c @@ -696,7 +696,7 @@ kern_setcred(struct thread *const td, const u_int flags, gid_t *groups = NULL; gid_t smallgroups[CRED_SMALLGROUPS_NB]; int error; - bool cred_set; + bool cred_set = false; /* Bail out on unrecognized flags. */ if (flags & ~SETCREDF_MASK) @@ -839,17 +839,32 @@ kern_setcred(struct thread *const td, const u_int flags, if (cred_set) { setsugid(p); to_free_cred = old_cred; +#ifdef RACCT + racct_proc_ucred_changed(p, old_cred, new_cred); +#endif +#ifdef RCTL + crhold(new_cred); +#endif MPASS(error == 0); } else error = EAGAIN; unlock_finish: PROC_UNLOCK(p); + /* * Part 3: After releasing the process lock, we perform cleanups and * finishing operations. */ +#ifdef RCTL + if (cred_set) { + rctl_proc_ucred_changed(p, new_cred); + /* Paired with the crhold() just above. */ + crfree(new_cred); + } +#endif + #ifdef MAC if (mac_set_proc_data != NULL) mac_set_proc_finish(td, proc_label_set, mac_set_proc_data); @@ -982,6 +997,8 @@ sys_setuid(struct thread *td, struct setuid_args *uap) proc_set_cred(p, newcred); #ifdef RACCT racct_proc_ucred_changed(p, oldcred, newcred); +#endif +#ifdef RCTL crhold(newcred); #endif PROC_UNLOCK(p); @@ -1390,6 +1407,8 @@ sys_setreuid(struct thread *td, struct setreuid_args *uap) proc_set_cred(p, newcred); #ifdef RACCT racct_proc_ucred_changed(p, oldcred, newcred); +#endif +#ifdef RCTL crhold(newcred); #endif PROC_UNLOCK(p); @@ -1536,6 +1555,8 @@ sys_setresuid(struct thread *td, struct setresuid_args *uap) proc_set_cred(p, newcred); #ifdef RACCT racct_proc_ucred_changed(p, oldcred, newcred); +#endif +#ifdef RCTL crhold(newcred); #endif PROC_UNLOCK(p); diff --git a/sys/kern/kern_racct.c b/sys/kern/kern_racct.c index 2aab151aba08..17b64ad00bb5 100644 --- a/sys/kern/kern_racct.c +++ b/sys/kern/kern_racct.c @@ -1236,16 +1236,20 @@ racct_updatepcpu_containers(void) racct_updatepcpu_post, NULL, NULL); } +static bool +racct_proc_to_skip(const struct proc *p) +{ + PROC_LOCK_ASSERT(p, MA_OWNED); + return (p->p_state != PRS_NORMAL || (p->p_flag & P_IDLEPROC) != 0); +} + static void racctd(void) { struct proc *p; - struct proc *idle; ASSERT_RACCT_ENABLED(); - idle = STAILQ_FIRST(&cpuhead)->pc_idlethread->td_proc; - for (;;) { racct_decay(); @@ -1253,12 +1257,7 @@ racctd(void) FOREACH_PROC_IN_SYSTEM(p) { PROC_LOCK(p); - if (p == idle) { - PROC_UNLOCK(p); - continue; - } - if (p->p_state != PRS_NORMAL || - (p->p_flag & P_IDLEPROC) != 0) { + if (racct_proc_to_skip(p)) { PROC_UNLOCK(p); continue; } @@ -1284,7 +1283,7 @@ racctd(void) */ FOREACH_PROC_IN_SYSTEM(p) { PROC_LOCK(p); - if (p->p_state != PRS_NORMAL) { + if (racct_proc_to_skip(p)) { PROC_UNLOCK(p); continue; } diff --git a/sys/kern/subr_syscall.c b/sys/kern/subr_syscall.c index d5b3b62f0821..48896529f685 100644 --- a/sys/kern/subr_syscall.c +++ b/sys/kern/subr_syscall.c @@ -55,8 +55,8 @@ syscallenter(struct thread *td) struct proc *p; struct syscall_args *sa; struct sysent *se; - int error, traced; - bool sy_thr_static; + int error; + bool sy_thr_static, traced; VM_CNT_INC(v_syscall); p = td->td_proc; @@ -219,7 +219,7 @@ syscallret(struct thread *td) struct proc *p; struct syscall_args *sa; ksiginfo_t ksi; - int traced; + bool traced; KASSERT(td->td_errno != ERELOOKUP, ("ERELOOKUP not consumed syscall %d", td->td_sa.code)); @@ -250,9 +250,9 @@ syscallret(struct thread *td) } #endif - traced = 0; + traced = false; if (__predict_false(p->p_flag & P_TRACED)) { - traced = 1; + traced = true; PROC_LOCK(p); td->td_dbgflags |= TDB_SCX; PROC_UNLOCK(p); diff --git a/sys/kern/sys_generic.c b/sys/kern/sys_generic.c index 7d666da9f88b..b84f675d1dcb 100644 --- a/sys/kern/sys_generic.c +++ b/sys/kern/sys_generic.c @@ -2345,3 +2345,35 @@ exterr_set(int eerror, int category, const char *mmsg, uintptr_t pp1, } return (eerror); } + +int +exterr_set_from(const struct kexterr *ke) +{ + struct thread *td; + + td = curthread; + if ((td->td_pflags2 & TDP2_UEXTERR) != 0) { + td->td_pflags2 |= TDP2_EXTERR; + td->td_kexterr = *ke; + } + return (td->td_kexterr.error); +} + +void +exterr_clear(struct kexterr *ke) +{ + memset(ke, 0, sizeof(*ke)); +} + +#include "opt_ddb.h" +#ifdef DDB +#include <ddb/ddb.h> + +void +exterr_db_print(struct kexterr *ke) +{ + db_printf("errno %d cat %d msg %s p1 %#jx p2 %#jx line %d\n", + ke->error, ke->cat, ke->msg == NULL ? "<none>" : ke->msg, + (uintmax_t)ke->p1, (uintmax_t)ke->p2, ke->src_line); +} +#endif diff --git a/sys/kern/vfs_aio.c b/sys/kern/vfs_aio.c index 60916a9fbd32..02d4b8426757 100644 --- a/sys/kern/vfs_aio.c +++ b/sys/kern/vfs_aio.c @@ -2487,7 +2487,7 @@ aio_biowakeup(struct bio *bp) long bcount = bp->bio_bcount; long resid = bp->bio_resid; int opcode, nblks; - int bio_error = bp->bio_error; + int abio_error = bp->bio_error; uint16_t flags = bp->bio_flags; opcode = job->uaiocb.aio_lio_opcode; @@ -2503,16 +2503,16 @@ aio_biowakeup(struct bio *bp) * error of whichever failed bio completed last. */ if (flags & BIO_ERROR) - atomic_store_int(&job->error, bio_error); + atomic_store_int(&job->error, abio_error); if (opcode & LIO_WRITE) atomic_add_int(&job->outblock, nblks); else atomic_add_int(&job->inblock, nblks); if (refcount_release(&job->nbio)) { - bio_error = atomic_load_int(&job->error); - if (bio_error != 0) - aio_complete(job, -1, bio_error); + abio_error = atomic_load_int(&job->error); + if (abio_error != 0) + aio_complete(job, -1, abio_error); else aio_complete(job, atomic_load_long(&job->nbytes), 0); } diff --git a/sys/kern/vfs_bio.c b/sys/kern/vfs_bio.c index 19c39e42bafa..880cc6b99951 100644 --- a/sys/kern/vfs_bio.c +++ b/sys/kern/vfs_bio.c @@ -44,6 +44,7 @@ * see man buf(9) for more info. */ +#define EXTERR_CATEGORY EXTERR_CAT_VFSBIO #include <sys/param.h> #include <sys/systm.h> #include <sys/asan.h> @@ -55,6 +56,7 @@ #include <sys/counter.h> #include <sys/devicestat.h> #include <sys/eventhandler.h> +#include <sys/exterrvar.h> #include <sys/fail.h> #include <sys/ktr.h> #include <sys/limits.h> @@ -1775,7 +1777,6 @@ buf_alloc(struct bufdomain *bd) bp->b_blkno = bp->b_lblkno = 0; bp->b_offset = NOOFFSET; bp->b_iodone = 0; - bp->b_error = 0; bp->b_resid = 0; bp->b_bcount = 0; bp->b_npages = 0; @@ -1785,6 +1786,7 @@ buf_alloc(struct bufdomain *bd) bp->b_fsprivate1 = NULL; bp->b_fsprivate2 = NULL; bp->b_fsprivate3 = NULL; + exterr_clear(&bp->b_exterr); LIST_INIT(&bp->b_dep); return (bp); @@ -2276,7 +2278,7 @@ breadn_flags(struct vnode *vp, daddr_t blkno, daddr_t dblkno, int size, } if ((flags & GB_CVTENXIO) != 0) bp->b_xflags |= BX_CVTENXIO; - bp->b_ioflags &= ~BIO_ERROR; + bp->b_ioflags &= ~(BIO_ERROR | BIO_EXTERR); if (bp->b_rcred == NOCRED && cred != NOCRED) bp->b_rcred = crhold(cred); vfs_busy_pages(bp, 0); @@ -2353,7 +2355,7 @@ bufwrite(struct buf *bp) bundirty(bp); bp->b_flags &= ~B_DONE; - bp->b_ioflags &= ~BIO_ERROR; + bp->b_ioflags &= ~(BIO_ERROR | BIO_EXTERR); bp->b_flags |= B_CACHE; bp->b_iocmd = BIO_WRITE; @@ -4520,8 +4522,11 @@ biowait(struct bio *bp, const char *wmesg) while ((bp->bio_flags & BIO_DONE) == 0) msleep(bp, mtxp, PRIBIO, wmesg, 0); mtx_unlock(mtxp); - if (bp->bio_error != 0) + if (bp->bio_error != 0) { + if ((bp->bio_flags & BIO_EXTERR) != 0) + return (exterr_set_from(&bp->bio_exterr)); return (bp->bio_error); + } if (!(bp->bio_flags & BIO_ERROR)) return (0); return (EIO); @@ -4568,6 +4573,8 @@ bufwait(struct buf *bp) return (EINTR); } if (bp->b_ioflags & BIO_ERROR) { + if ((bp->b_ioflags & BIO_EXTERR) != 0) + exterr_set_from(&bp->b_exterr); return (bp->b_error ? bp->b_error : EIO); } else { return (0); @@ -5522,6 +5529,8 @@ DB_SHOW_COMMAND(buffer, db_show_buffer) db_printf("\n"); } BUF_LOCKPRINTINFO(bp); + if ((bp->b_ioflags & BIO_EXTERR) != 0) + exterr_db_print(&bp->b_exterr); #if defined(FULL_BUF_TRACKING) db_printf("b_io_tracking: b_io_tcnt = %u\n", bp->b_io_tcnt); diff --git a/sys/net/if_tuntap.c b/sys/net/if_tuntap.c index 56bb90cce9bc..0dc3a58f6ae6 100644 --- a/sys/net/if_tuntap.c +++ b/sys/net/if_tuntap.c @@ -138,6 +138,7 @@ struct tuntap_softc { #define TUN_READY (TUN_OPEN | TUN_INITED) pid_t tun_pid; /* owning pid */ + struct epoch_context tun_epoch_ctx; struct ifnet *tun_ifp; /* the interface */ struct sigio *tun_sigio; /* async I/O info */ struct tuntap_driver *tun_drv; /* appropriate driver */ @@ -630,6 +631,18 @@ out: CURVNET_RESTORE(); } +static void +tunfree(struct epoch_context *ctx) +{ + struct tuntap_softc *tp; + + tp = __containerof(ctx, struct tuntap_softc, tun_epoch_ctx); + + /* Any remaining resources that would be needed by a concurrent open. */ + mtx_destroy(&tp->tun_mtx); + free(tp, M_TUN); +} + static int tun_destroy(struct tuntap_softc *tp, bool may_intr) { @@ -649,7 +662,7 @@ tun_destroy(struct tuntap_softc *tp, bool may_intr) error = cv_wait_sig(&tp->tun_cv, &tp->tun_mtx); else cv_wait(&tp->tun_cv, &tp->tun_mtx); - if (error != 0) { + if (error != 0 && tp->tun_busy != 0) { tp->tun_flags &= ~TUN_DYING; TUN_UNLOCK(tp); return (error); @@ -663,8 +676,18 @@ tun_destroy(struct tuntap_softc *tp, bool may_intr) TAILQ_REMOVE(&tunhead, tp, tun_list); mtx_unlock(&tunmtx); - /* destroy_dev will take care of any alias. */ - destroy_dev(tp->tun_dev); + /* + * destroy_dev will take care of any alias. For transient tunnels, + * we're being called from close(2) so we can't destroy it ourselves + * without deadlocking, but we already know that we can cleanup + * everything else and just continue to prevent it from being reopened. + */ + if ((tp->tun_flags & TUN_TRANSIENT) != 0) { + atomic_store_ptr(&tp->tun_dev->si_drv1, tp->tun_dev); + destroy_dev_sched(tp->tun_dev); + } else { + destroy_dev(tp->tun_dev); + } seldrain(&tp->tun_rsel); knlist_clear(&tp->tun_rsel.si_note, 0); knlist_destroy(&tp->tun_rsel.si_note); @@ -679,9 +702,8 @@ tun_destroy(struct tuntap_softc *tp, bool may_intr) sx_xunlock(&tun_ioctl_sx); free_unr(tp->tun_drv->unrhdr, TUN2IFP(tp)->if_dunit); if_free(TUN2IFP(tp)); - mtx_destroy(&tp->tun_mtx); cv_destroy(&tp->tun_cv); - free(tp, M_TUN); + NET_EPOCH_CALL(tunfree, &tp->tun_epoch_ctx); CURVNET_RESTORE(); return (0); @@ -742,9 +764,11 @@ tun_uninit(const void *unused __unused) mtx_unlock(&tunmtx); for (i = 0; i < nitems(tuntap_drivers); ++i) { drv = &tuntap_drivers[i]; + destroy_dev_drain(&drv->cdevsw); delete_unrhdr(drv->unrhdr); clone_cleanup(&drv->clones); } + NET_EPOCH_DRAIN_CALLBACKS(); mtx_destroy(&tunmtx); } SYSUNINIT(tun_uninit, SI_SUB_PROTO_IF, SI_ORDER_ANY, tun_uninit, NULL); @@ -1104,19 +1128,43 @@ out: static int tunopen(struct cdev *dev, int flag, int mode, struct thread *td) { + struct epoch_tracker et; struct ifnet *ifp; struct tuntap_softc *tp; + void *p; int error __diagused, tunflags; + /* + * Transient tunnels do deferred destroy of the tun device but want + * to immediately cleanup state, so they clobber si_drv1 to avoid a + * use-after-free in case someone does happen to open it in the interim. + * We avoid using NULL to be able to distinguish from an uninitialized + * cdev. + * + * We use the net epoch here to let a concurrent tun_destroy() schedule + * freeing our tuntap_softc, in case we entered here and loaded si_drv1 + * before it was swapped out. If we managed to load this while it was + * still a softc, then the concurrent tun_destroy() hasn't yet scheduled + * it to be free- that will take place sometime after the epoch we just + * entered, so we can safely use it. + */ + NET_EPOCH_ENTER(et); + p = atomic_load_ptr(&dev->si_drv1); + if (p == dev) { + NET_EPOCH_EXIT(et); + return (ENXIO); + } + tunflags = 0; CURVNET_SET(TD_TO_VNET(td)); error = tuntap_name2info(dev->si_name, NULL, &tunflags); if (error != 0) { CURVNET_RESTORE(); + NET_EPOCH_EXIT(et); return (error); /* Shouldn't happen */ } - tp = dev->si_drv1; + tp = p; KASSERT(tp != NULL, ("si_drv1 should have been initialized at creation")); @@ -1124,14 +1172,17 @@ tunopen(struct cdev *dev, int flag, int mode, struct thread *td) if ((tp->tun_flags & TUN_INITED) == 0) { TUN_UNLOCK(tp); CURVNET_RESTORE(); + NET_EPOCH_EXIT(et); return (ENXIO); } if ((tp->tun_flags & (TUN_OPEN | TUN_DYING)) != 0) { TUN_UNLOCK(tp); CURVNET_RESTORE(); + NET_EPOCH_EXIT(et); return (EBUSY); } + NET_EPOCH_EXIT(et); error = tun_busy_locked(tp); KASSERT(error == 0, ("Must be able to busy an unopen tunnel")); ifp = TUN2IFP(tp); diff --git a/sys/netinet/in_pcb.c b/sys/netinet/in_pcb.c index 712ff28768dc..b7dae78fb2c2 100644 --- a/sys/netinet/in_pcb.c +++ b/sys/netinet/in_pcb.c @@ -3058,143 +3058,7 @@ db_print_inconninfo(struct in_conninfo *inc, const char *name, int indent) ntohs(inc->inc_fport)); } -static void -db_print_inpflags(int inp_flags) -{ - int comma; - - comma = 0; - if (inp_flags & INP_RECVOPTS) { - db_printf("%sINP_RECVOPTS", comma ? ", " : ""); - comma = 1; - } - if (inp_flags & INP_RECVRETOPTS) { - db_printf("%sINP_RECVRETOPTS", comma ? ", " : ""); - comma = 1; - } - if (inp_flags & INP_RECVDSTADDR) { - db_printf("%sINP_RECVDSTADDR", comma ? ", " : ""); - comma = 1; - } - if (inp_flags & INP_ORIGDSTADDR) { - db_printf("%sINP_ORIGDSTADDR", comma ? ", " : ""); - comma = 1; - } - if (inp_flags & INP_HDRINCL) { - db_printf("%sINP_HDRINCL", comma ? ", " : ""); - comma = 1; - } - if (inp_flags & INP_HIGHPORT) { - db_printf("%sINP_HIGHPORT", comma ? ", " : ""); - comma = 1; - } - if (inp_flags & INP_LOWPORT) { - db_printf("%sINP_LOWPORT", comma ? ", " : ""); - comma = 1; - } - if (inp_flags & INP_ANONPORT) { - db_printf("%sINP_ANONPORT", comma ? ", " : ""); - comma = 1; - } - if (inp_flags & INP_RECVIF) { - db_printf("%sINP_RECVIF", comma ? ", " : ""); - comma = 1; - } - if (inp_flags & INP_MTUDISC) { - db_printf("%sINP_MTUDISC", comma ? ", " : ""); - comma = 1; - } - if (inp_flags & INP_RECVTTL) { - db_printf("%sINP_RECVTTL", comma ? ", " : ""); - comma = 1; - } - if (inp_flags & INP_DONTFRAG) { - db_printf("%sINP_DONTFRAG", comma ? ", " : ""); - comma = 1; - } - if (inp_flags & INP_RECVTOS) { - db_printf("%sINP_RECVTOS", comma ? ", " : ""); - comma = 1; - } - if (inp_flags & IN6P_IPV6_V6ONLY) { - db_printf("%sIN6P_IPV6_V6ONLY", comma ? ", " : ""); - comma = 1; - } - if (inp_flags & IN6P_PKTINFO) { - db_printf("%sIN6P_PKTINFO", comma ? ", " : ""); - comma = 1; - } - if (inp_flags & IN6P_HOPLIMIT) { - db_printf("%sIN6P_HOPLIMIT", comma ? ", " : ""); - comma = 1; - } - if (inp_flags & IN6P_HOPOPTS) { - db_printf("%sIN6P_HOPOPTS", comma ? ", " : ""); - comma = 1; - } - if (inp_flags & IN6P_DSTOPTS) { - db_printf("%sIN6P_DSTOPTS", comma ? ", " : ""); - comma = 1; - } - if (inp_flags & IN6P_RTHDR) { - db_printf("%sIN6P_RTHDR", comma ? ", " : ""); - comma = 1; - } - if (inp_flags & IN6P_RTHDRDSTOPTS) { - db_printf("%sIN6P_RTHDRDSTOPTS", comma ? ", " : ""); - comma = 1; - } - if (inp_flags & IN6P_TCLASS) { - db_printf("%sIN6P_TCLASS", comma ? ", " : ""); - comma = 1; - } - if (inp_flags & IN6P_AUTOFLOWLABEL) { - db_printf("%sIN6P_AUTOFLOWLABEL", comma ? ", " : ""); - comma = 1; - } - if (inp_flags & INP_ONESBCAST) { - db_printf("%sINP_ONESBCAST", comma ? ", " : ""); - comma = 1; - } - if (inp_flags & INP_DROPPED) { - db_printf("%sINP_DROPPED", comma ? ", " : ""); - comma = 1; - } - if (inp_flags & INP_SOCKREF) { - db_printf("%sINP_SOCKREF", comma ? ", " : ""); - comma = 1; - } - if (inp_flags & IN6P_RFC2292) { - db_printf("%sIN6P_RFC2292", comma ? ", " : ""); - comma = 1; - } - if (inp_flags & IN6P_MTU) { - db_printf("IN6P_MTU%s", comma ? ", " : ""); - comma = 1; - } -} - -static void -db_print_inpvflag(u_char inp_vflag) -{ - int comma; - - comma = 0; - if (inp_vflag & INP_IPV4) { - db_printf("%sINP_IPV4", comma ? ", " : ""); - comma = 1; - } - if (inp_vflag & INP_IPV6) { - db_printf("%sINP_IPV6", comma ? ", " : ""); - comma = 1; - } - if (inp_vflag & INP_IPV6PROTO) { - db_printf("%sINP_IPV6PROTO", comma ? ", " : ""); - comma = 1; - } -} - -static void +void db_print_inpcb(struct inpcb *inp, const char *name, int indent) { @@ -3204,38 +3068,39 @@ db_print_inpcb(struct inpcb *inp, const char *name, int indent) indent += 2; db_print_indent(indent); - db_printf("inp_flow: 0x%x\n", inp->inp_flow); + db_printf("inp_flow: 0x%x inp_label: %p\n", inp->inp_flow, + inp->inp_label); db_print_inconninfo(&inp->inp_inc, "inp_conninfo", indent); db_print_indent(indent); - db_printf("inp_label: %p inp_flags: 0x%x (", - inp->inp_label, inp->inp_flags); - db_print_inpflags(inp->inp_flags); - db_printf(")\n"); + db_printf("inp_flags: 0x%b\n", inp->inp_flags, INP_FLAGS_BITS); db_print_indent(indent); - db_printf("inp_sp: %p inp_vflag: 0x%x (", inp->inp_sp, - inp->inp_vflag); - db_print_inpvflag(inp->inp_vflag); - db_printf(")\n"); + db_printf("inp_flags2: 0x%b\n", inp->inp_flags2, INP_FLAGS2_BITS); + + db_print_indent(indent); + db_printf("inp_sp: %p inp_vflag: 0x%b\n", inp->inp_sp, + inp->inp_vflag, INP_VFLAGS_BITS); db_print_indent(indent); db_printf("inp_ip_ttl: %d inp_ip_p: %d inp_ip_minttl: %d\n", inp->inp_ip_ttl, inp->inp_ip_p, inp->inp_ip_minttl); - db_print_indent(indent); #ifdef INET6 if (inp->inp_vflag & INP_IPV6) { + db_print_indent(indent); db_printf("in6p_options: %p in6p_outputopts: %p " "in6p_moptions: %p\n", inp->in6p_options, inp->in6p_outputopts, inp->in6p_moptions); + db_print_indent(indent); db_printf("in6p_icmp6filt: %p in6p_cksum %d " "in6p_hops %u\n", inp->in6p_icmp6filt, inp->in6p_cksum, inp->in6p_hops); } else #endif { + db_print_indent(indent); db_printf("inp_ip_tos: %d inp_ip_options: %p " "inp_ip_moptions: %p\n", inp->inp_ip_tos, inp->inp_options, inp->inp_moptions); diff --git a/sys/netinet/in_pcb.h b/sys/netinet/in_pcb.h index 9e0618e87601..975b8129c70d 100644 --- a/sys/netinet/in_pcb.h +++ b/sys/netinet/in_pcb.h @@ -539,6 +539,9 @@ void inp_4tuple_get(struct inpcb *inp, uint32_t *laddr, uint16_t *lp, #define INP_IPV6 0x2 #define INP_IPV6PROTO 0x4 /* opened under IPv6 protocol */ +/* inp_vflags description for use with printf(9) %b identifier. */ +#define INP_VFLAGS_BITS "\20\1INP_IPV4\2INP_IPV6\3INP_IPV6PROTO" + /* * Flags for inp_flags. */ @@ -582,6 +585,17 @@ void inp_4tuple_get(struct inpcb *inp, uint32_t *laddr, uint16_t *lp, IN6P_TCLASS|IN6P_AUTOFLOWLABEL|IN6P_RFC2292|\ IN6P_MTU) +/* inp_flags description for use with printf(9) %b identifier. */ +#define INP_FLAGS_BITS "\20" \ + "\1INP_RECVOPTS\2INP_RECVRETOPTS\3INP_RECVDSTADDR\4INP_HDRINCL" \ + "\5INP_HIGHPORT\6INP_LOWPORT\7INP_ANONPORT\10INP_RECVIF" \ + "\11INP_MTUDISC\12INP_FREED\13INP_RECVTTL\14INP_DONTFRAG" \ + "\15INP_BINDANY\16INP_INHASHLIST\17INP_RECVTOS\20IN6P_IPV6_V6ONLY" \ + "\21IN6P_PKTINFO\22IN6P_HOPLIMIT\23IN6P_HOPOPTS\24IN6P_DSTOPTS" \ + "\25IN6P_RTHDR\26IN6P_RTHDRDSTOPTS\27IN6P_TCLASS\30IN6P_AUTOFLOWLABEL" \ + "\31INP_INLBGROUP\32INP_ONESBCAST\33INP_DROPPED\34INP_SOCKREF" \ + "\35INP_RESERVED_0\36INP_BOUNDFIB\37IN6P_RFC2292\40IN6P_MTU" + /* * Flags for inp_flags2. */ @@ -610,6 +624,13 @@ void inp_4tuple_get(struct inpcb *inp, uint32_t *laddr, uint16_t *lp, #define INP_2PCP_MASK (INP_2PCP_BIT0 | INP_2PCP_BIT1 | INP_2PCP_BIT2) #define INP_2PCP_SHIFT 18 /* shift PCP field in/out of inp_flags2 */ +/* inp_flags2 description for use with printf(9) %b identifier. */ +#define INP_FLAGS2_BITS "\20" \ + "\11INP_RECVFLOWID\12INP_RECVRSSBUCKETID" \ + "\13INP_RATE_LIMIT_CHANGED\14INP_ORIGDSTADDR" \ + "\22INP_2PCP_SET\23INP_2PCP_BIT0\24INP_2PCP_BIT1" \ + "\25INP_2PCP_BIT2" + /* * Flags passed to in_pcblookup*(), inp_smr_lock() and inp_next(). */ @@ -730,6 +751,9 @@ int in_pcbquery_txrlevel(struct inpcb *, uint32_t *); void in_pcboutput_txrtlmt(struct inpcb *, struct ifnet *, struct mbuf *); void in_pcboutput_eagain(struct inpcb *); #endif +#ifdef DDB +void db_print_inpcb(struct inpcb *, const char *, int); +#endif #endif /* _KERNEL */ #endif /* !_NETINET_IN_PCB_H_ */ diff --git a/sys/netinet/tcp_input.c b/sys/netinet/tcp_input.c index 2146b0cac48f..9c58c2815d13 100644 --- a/sys/netinet/tcp_input.c +++ b/sys/netinet/tcp_input.c @@ -1192,11 +1192,10 @@ tfo_socket_result: if (thflags & TH_ACK) { if ((s = tcp_log_addrs(&inc, th, NULL, NULL))) log(LOG_DEBUG, "%s; %s: Listen socket: " - "SYN|ACK invalid, segment rejected\n", + "SYN|ACK invalid, segment ignored\n", s, __func__); - syncache_badack(&inc, port); /* XXX: Not needed! */ TCPSTAT_INC(tcps_badsyn); - goto dropwithreset; + goto dropunlock; } /* * If the drop_synfin option is enabled, drop all diff --git a/sys/netinet/tcp_syncache.c b/sys/netinet/tcp_syncache.c index 3a7755e9f09e..fa7035771714 100644 --- a/sys/netinet/tcp_syncache.c +++ b/sys/netinet/tcp_syncache.c @@ -713,23 +713,6 @@ done: } void -syncache_badack(struct in_conninfo *inc, uint16_t port) -{ - struct syncache *sc; - struct syncache_head *sch; - - if (syncache_cookiesonly()) - return; - sc = syncache_lookup(inc, &sch); /* returns locked sch */ - SCH_LOCK_ASSERT(sch); - if ((sc != NULL) && (sc->sc_port == port)) { - syncache_drop(sc, sch); - TCPSTAT_INC(tcps_sc_badack); - } - SCH_UNLOCK(sch); -} - -void syncache_unreach(struct in_conninfo *inc, tcp_seq th_seq, uint16_t port) { struct syncache *sc; diff --git a/sys/netinet/tcp_syncache.h b/sys/netinet/tcp_syncache.h index 37f6ff3d6ca9..c916b4de6ae0 100644 --- a/sys/netinet/tcp_syncache.h +++ b/sys/netinet/tcp_syncache.h @@ -45,7 +45,6 @@ struct socket * syncache_add(struct in_conninfo *, struct tcpopt *, void *, void *, uint8_t, uint16_t); void syncache_chkrst(struct in_conninfo *, struct tcphdr *, struct mbuf *, uint16_t); -void syncache_badack(struct in_conninfo *, uint16_t); int syncache_pcblist(struct sysctl_req *); struct syncache { diff --git a/sys/netinet/tcp_usrreq.c b/sys/netinet/tcp_usrreq.c index 98c934955121..4d1a6455d09e 100644 --- a/sys/netinet/tcp_usrreq.c +++ b/sys/netinet/tcp_usrreq.c @@ -2799,258 +2799,6 @@ db_print_tstate(int t_state) } static void -db_print_tflags(u_int t_flags) -{ - int comma; - - comma = 0; - if (t_flags & TF_ACKNOW) { - db_printf("%sTF_ACKNOW", comma ? ", " : ""); - comma = 1; - } - if (t_flags & TF_DELACK) { - db_printf("%sTF_DELACK", comma ? ", " : ""); - comma = 1; - } - if (t_flags & TF_NODELAY) { - db_printf("%sTF_NODELAY", comma ? ", " : ""); - comma = 1; - } - if (t_flags & TF_NOOPT) { - db_printf("%sTF_NOOPT", comma ? ", " : ""); - comma = 1; - } - if (t_flags & TF_SENTFIN) { - db_printf("%sTF_SENTFIN", comma ? ", " : ""); - comma = 1; - } - if (t_flags & TF_REQ_SCALE) { - db_printf("%sTF_REQ_SCALE", comma ? ", " : ""); - comma = 1; - } - if (t_flags & TF_RCVD_SCALE) { - db_printf("%sTF_RECVD_SCALE", comma ? ", " : ""); - comma = 1; - } - if (t_flags & TF_REQ_TSTMP) { - db_printf("%sTF_REQ_TSTMP", comma ? ", " : ""); - comma = 1; - } - if (t_flags & TF_RCVD_TSTMP) { - db_printf("%sTF_RCVD_TSTMP", comma ? ", " : ""); - comma = 1; - } - if (t_flags & TF_SACK_PERMIT) { - db_printf("%sTF_SACK_PERMIT", comma ? ", " : ""); - comma = 1; - } - if (t_flags & TF_NEEDSYN) { - db_printf("%sTF_NEEDSYN", comma ? ", " : ""); - comma = 1; - } - if (t_flags & TF_NEEDFIN) { - db_printf("%sTF_NEEDFIN", comma ? ", " : ""); - comma = 1; - } - if (t_flags & TF_NOPUSH) { - db_printf("%sTF_NOPUSH", comma ? ", " : ""); - comma = 1; - } - if (t_flags & TF_PREVVALID) { - db_printf("%sTF_PREVVALID", comma ? ", " : ""); - comma = 1; - } - if (t_flags & TF_WAKESOR) { - db_printf("%sTF_WAKESOR", comma ? ", " : ""); - comma = 1; - } - if (t_flags & TF_GPUTINPROG) { - db_printf("%sTF_GPUTINPROG", comma ? ", " : ""); - comma = 1; - } - if (t_flags & TF_MORETOCOME) { - db_printf("%sTF_MORETOCOME", comma ? ", " : ""); - comma = 1; - } - if (t_flags & TF_SONOTCONN) { - db_printf("%sTF_SONOTCONN", comma ? ", " : ""); - comma = 1; - } - if (t_flags & TF_LASTIDLE) { - db_printf("%sTF_LASTIDLE", comma ? ", " : ""); - comma = 1; - } - if (t_flags & TF_RXWIN0SENT) { - db_printf("%sTF_RXWIN0SENT", comma ? ", " : ""); - comma = 1; - } - if (t_flags & TF_FASTRECOVERY) { - db_printf("%sTF_FASTRECOVERY", comma ? ", " : ""); - comma = 1; - } - if (t_flags & TF_WASFRECOVERY) { - db_printf("%sTF_WASFRECOVERY", comma ? ", " : ""); - comma = 1; - } - if (t_flags & TF_SIGNATURE) { - db_printf("%sTF_SIGNATURE", comma ? ", " : ""); - comma = 1; - } - if (t_flags & TF_FORCEDATA) { - db_printf("%sTF_FORCEDATA", comma ? ", " : ""); - comma = 1; - } - if (t_flags & TF_TSO) { - db_printf("%sTF_TSO", comma ? ", " : ""); - comma = 1; - } - if (t_flags & TF_TOE) { - db_printf("%sTF_TOE", comma ? ", " : ""); - comma = 1; - } - if (t_flags & TF_CLOSED) { - db_printf("%sTF_CLOSED", comma ? ", " : ""); - comma = 1; - } - if (t_flags & TF_SENTSYN) { - db_printf("%sTF_SENTSYN", comma ? ", " : ""); - comma = 1; - } - if (t_flags & TF_LRD) { - db_printf("%sTF_LRD", comma ? ", " : ""); - comma = 1; - } - if (t_flags & TF_CONGRECOVERY) { - db_printf("%sTF_CONGRECOVERY", comma ? ", " : ""); - comma = 1; - } - if (t_flags & TF_WASCRECOVERY) { - db_printf("%sTF_WASCRECOVERY", comma ? ", " : ""); - comma = 1; - } - if (t_flags & TF_FASTOPEN) { - db_printf("%sTF_FASTOPEN", comma ? ", " : ""); - comma = 1; - } -} - -static void -db_print_tflags2(u_int t_flags2) -{ - int comma; - - comma = 0; - if (t_flags2 & TF2_PLPMTU_BLACKHOLE) { - db_printf("%sTF2_PLPMTU_BLACKHOLE", comma ? ", " : ""); - comma = 1; - } - if (t_flags2 & TF2_PLPMTU_PMTUD) { - db_printf("%sTF2_PLPMTU_PMTUD", comma ? ", " : ""); - comma = 1; - } - if (t_flags2 & TF2_PLPMTU_MAXSEGSNT) { - db_printf("%sTF2_PLPMTU_MAXSEGSNT", comma ? ", " : ""); - comma = 1; - } - if (t_flags2 & TF2_LOG_AUTO) { - db_printf("%sTF2_LOG_AUTO", comma ? ", " : ""); - comma = 1; - } - if (t_flags2 & TF2_DROP_AF_DATA) { - db_printf("%sTF2_DROP_AF_DATA", comma ? ", " : ""); - comma = 1; - } - if (t_flags2 & TF2_ECN_PERMIT) { - db_printf("%sTF2_ECN_PERMIT", comma ? ", " : ""); - comma = 1; - } - if (t_flags2 & TF2_ECN_SND_CWR) { - db_printf("%sTF2_ECN_SND_CWR", comma ? ", " : ""); - comma = 1; - } - if (t_flags2 & TF2_ECN_SND_ECE) { - db_printf("%sTF2_ECN_SND_ECE", comma ? ", " : ""); - comma = 1; - } - if (t_flags2 & TF2_ACE_PERMIT) { - db_printf("%sTF2_ACE_PERMIT", comma ? ", " : ""); - comma = 1; - } - if (t_flags2 & TF2_HPTS_CPU_SET) { - db_printf("%sTF2_HPTS_CPU_SET", comma ? ", " : ""); - comma = 1; - } - if (t_flags2 & TF2_FBYTES_COMPLETE) { - db_printf("%sTF2_FBYTES_COMPLETE", comma ? ", " : ""); - comma = 1; - } - if (t_flags2 & TF2_ECN_USE_ECT1) { - db_printf("%sTF2_ECN_USE_ECT1", comma ? ", " : ""); - comma = 1; - } - if (t_flags2 & TF2_TCP_ACCOUNTING) { - db_printf("%sTF2_TCP_ACCOUNTING", comma ? ", " : ""); - comma = 1; - } - if (t_flags2 & TF2_HPTS_CALLS) { - db_printf("%sTF2_HPTS_CALLS", comma ? ", " : ""); - comma = 1; - } - if (t_flags2 & TF2_MBUF_L_ACKS) { - db_printf("%sTF2_MBUF_L_ACKS", comma ? ", " : ""); - comma = 1; - } - if (t_flags2 & TF2_MBUF_ACKCMP) { - db_printf("%sTF2_MBUF_ACKCMP", comma ? ", " : ""); - comma = 1; - } - if (t_flags2 & TF2_SUPPORTS_MBUFQ) { - db_printf("%sTF2_SUPPORTS_MBUFQ", comma ? ", " : ""); - comma = 1; - } - if (t_flags2 & TF2_MBUF_QUEUE_READY) { - db_printf("%sTF2_MBUF_QUEUE_READY", comma ? ", " : ""); - comma = 1; - } - if (t_flags2 & TF2_DONT_SACK_QUEUE) { - db_printf("%sTF2_DONT_SACK_QUEUE", comma ? ", " : ""); - comma = 1; - } - if (t_flags2 & TF2_CANNOT_DO_ECN) { - db_printf("%sTF2_CANNOT_DO_ECN", comma ? ", " : ""); - comma = 1; - } - if (t_flags2 & TF2_PROC_SACK_PROHIBIT) { - db_printf("%sTF2_PROC_SACK_PROHIBIT", comma ? ", " : ""); - comma = 1; - } - if (t_flags2 & TF2_IPSEC_TSO) { - db_printf("%sTF2_IPSEC_TSO", comma ? ", " : ""); - comma = 1; - } - if (t_flags2 & TF2_NO_ISS_CHECK) { - db_printf("%sTF2_NO_ISS_CHECK", comma ? ", " : ""); - comma = 1; - } -} - -static void -db_print_toobflags(char t_oobflags) -{ - int comma; - - comma = 0; - if (t_oobflags & TCPOOB_HAVEDATA) { - db_printf("%sTCPOOB_HAVEDATA", comma ? ", " : ""); - comma = 1; - } - if (t_oobflags & TCPOOB_HADDATA) { - db_printf("%sTCPOOB_HADDATA", comma ? ", " : ""); - comma = 1; - } -} - -static void db_print_bblog_state(int state) { switch (state) { @@ -3088,7 +2836,8 @@ db_print_bblog_state(int state) } static void -db_print_tcpcb(struct tcpcb *tp, const char *name, int indent, bool show_bblog) +db_print_tcpcb(struct tcpcb *tp, const char *name, int indent, bool show_bblog, + bool show_inpcb) { db_print_indent(indent); @@ -3096,6 +2845,9 @@ db_print_tcpcb(struct tcpcb *tp, const char *name, int indent, bool show_bblog) indent += 2; + if (show_inpcb) + db_print_inpcb(tptoinpcb(tp), "t_inpcb", indent); + db_print_indent(indent); db_printf("t_segq first: %p t_segqlen: %d t_dupacks: %d\n", TAILQ_FIRST(&tp->t_segq), tp->t_segqlen, tp->t_dupacks); @@ -3110,14 +2862,10 @@ db_print_tcpcb(struct tcpcb *tp, const char *name, int indent, bool show_bblog) db_printf(")\n"); db_print_indent(indent); - db_printf("t_flags: 0x%x (", tp->t_flags); - db_print_tflags(tp->t_flags); - db_printf(")\n"); + db_printf("t_flags: 0x%b\n", tp->t_flags, TF_BITS); db_print_indent(indent); - db_printf("t_flags2: 0x%x (", tp->t_flags2); - db_print_tflags2(tp->t_flags2); - db_printf(")\n"); + db_printf("t_flags2: 0x%b\n", tp->t_flags2, TF2_BITS); db_print_indent(indent); db_printf("snd_una: 0x%08x snd_max: 0x%08x snd_nxt: 0x%08x\n", @@ -3164,9 +2912,8 @@ db_print_tcpcb(struct tcpcb *tp, const char *name, int indent, bool show_bblog) tp->t_rttupdated, tp->max_sndwnd, tp->t_softerror); db_print_indent(indent); - db_printf("t_oobflags: 0x%x (", tp->t_oobflags); - db_print_toobflags(tp->t_oobflags); - db_printf(") t_iobc: 0x%02x\n", tp->t_iobc); + db_printf("t_oobflags: 0x%b t_iobc: 0x%02x\n", tp->t_oobflags, + TCPOOB_BITS, tp->t_iobc); db_print_indent(indent); db_printf("snd_scale: %u rcv_scale: %u request_r_scale: %u\n", @@ -3227,33 +2974,36 @@ db_print_tcpcb(struct tcpcb *tp, const char *name, int indent, bool show_bblog) DB_SHOW_COMMAND(tcpcb, db_show_tcpcb) { struct tcpcb *tp; - bool show_bblog; + bool show_bblog, show_inpcb; if (!have_addr) { - db_printf("usage: show tcpcb <addr>\n"); + db_printf("usage: show tcpcb[/bi] <addr>\n"); return; } show_bblog = strchr(modif, 'b') != NULL; + show_inpcb = strchr(modif, 'i') != NULL; tp = (struct tcpcb *)addr; - - db_print_tcpcb(tp, "tcpcb", 0, show_bblog); + db_print_tcpcb(tp, "tcpcb", 0, show_bblog, show_inpcb); } DB_SHOW_ALL_COMMAND(tcpcbs, db_show_all_tcpcbs) { VNET_ITERATOR_DECL(vnet_iter); struct inpcb *inp; - bool only_locked, show_bblog; + struct tcpcb *tp; + bool only_locked, show_bblog, show_inpcb; only_locked = strchr(modif, 'l') != NULL; show_bblog = strchr(modif, 'b') != NULL; + show_inpcb = strchr(modif, 'i') != NULL; VNET_FOREACH(vnet_iter) { CURVNET_SET(vnet_iter); CK_LIST_FOREACH(inp, &V_tcbinfo.ipi_listhead, inp_list) { if (only_locked && inp->inp_lock.rw_lock == RW_UNLOCKED) continue; - db_print_tcpcb(intotcpcb(inp), "tcpcb", 0, show_bblog); + tp = intotcpcb(inp); + db_print_tcpcb(tp, "tcpcb", 0, show_bblog, show_inpcb); if (db_pager_quit) break; } diff --git a/sys/netinet/tcp_var.h b/sys/netinet/tcp_var.h index c3be95c80798..f9297be46af7 100644 --- a/sys/netinet/tcp_var.h +++ b/sys/netinet/tcp_var.h @@ -795,6 +795,17 @@ tcp_packets_this_ack(struct tcpcb *tp, tcp_seq ack) #define TF_WASCRECOVERY 0x40000000 /* was in congestion recovery */ #define TF_FASTOPEN 0x80000000 /* TCP Fast Open indication */ +/* t_flags description for use with printf(9) %b identifier. */ +#define TF_BITS "\20" \ + "\1TF_ACKNOW\2TF_DELACK\3TF_NODELAY\4TF_NOOPT" \ + "\5TF_SENTFIN\6TF_REQ_SCALE\7TF_RCVD_SCALE\10TF_REQ_TSTMP" \ + "\11TF_RCVD_TSTMP\12TF_SACK_PERMIT\13TF_NEEDSYN\14TF_NEEDFIN" \ + "\15TF_NOPUSH\16TF_PREVVALID\17TF_WAKESOR\20TF_GPUTINPROG" \ + "\21TF_MORETOCOME\22TF_SONOTCONN\23TF_LASTIDLE\24TF_RXWIN0SENT" \ + "\25TF_FASTRECOVERY\26TF_WASFRECOVERY\27TF_SIGNATURE\30TF_FORCEDATA" \ + "\31TF_TSO\32TF_TOE\33TF_CLOSED\34TF_SENTSYN" \ + "\35TF_LRD\36TF_CONGRECOVERY\37TF_WASCRECOVERY\40TF_FASTOPEN" + #define IN_FASTRECOVERY(t_flags) (t_flags & TF_FASTRECOVERY) #define ENTER_FASTRECOVERY(t_flags) t_flags |= TF_FASTRECOVERY #define EXIT_FASTRECOVERY(t_flags) t_flags &= ~TF_FASTRECOVERY @@ -815,6 +826,9 @@ tcp_packets_this_ack(struct tcpcb *tp, tcp_seq ack) #define TCPOOB_HAVEDATA 0x01 #define TCPOOB_HADDATA 0x02 +/* t_oobflags description for use with printf(9) %b identifier. */ +#define TCPOOB_BITS "\20\1TCPOOB_HAVEDATA\2TCPOOB_HADDATA" + /* * Flags for the extended TCP flags field, t_flags2 */ @@ -842,6 +856,21 @@ tcp_packets_this_ack(struct tcpcb *tp, tcp_seq ack) #define TF2_IPSEC_TSO 0x00200000 /* IPSEC + TSO supported */ #define TF2_NO_ISS_CHECK 0x00400000 /* Don't check SEG.ACK against ISS */ +/* t_flags2 description for use with printf(9) %b identifier. */ +#define TF2_BITS "\20" \ + "\1TF2_PLPMTU_BLACKHOLE\2TF2_PLPMTU_PMTUD" \ + "\3TF2_PLPMTU_MAXSEGSNT\4TF2_LOG_AUTO" \ + "\5TF2_DROP_AF_DATA\6TF2_ECN_PERMIT" \ + "\7TF2_ECN_SND_CWR\10TF2_ECN_SND_ECE" \ + "\11TF2_ACE_PERMIT\12TF2_HPTS_CPU_SET" \ + "\13TF2_FBYTES_COMPLETE\14TF2_ECN_USE_ECT1" \ + "\15TF2_TCP_ACCOUNTING\16TF2_HPTS_CALLS" \ + "\17TF2_MBUF_L_ACKS\20TF2_MBUF_ACKCMP" \ + "\21TF2_SUPPORTS_MBUFQ\22TF2_MBUF_QUEUE_READY" \ + "\23TF2_DONT_SACK_QUEUE\24TF2_CANNOT_DO_ECN" \ + "\25TF2_PROC_SACK_PROHIBIT\26TF2_IPSEC_TSO" \ + "\27TF2_NO_ISS_CHECK" + /* * Structure to hold TCP options that are only used during segment * processing (in tcp_input), but not held in the tcpcb. diff --git a/sys/netinet6/mld6.c b/sys/netinet6/mld6.c index a825658bd9ee..5dfe48908a4f 100644 --- a/sys/netinet6/mld6.c +++ b/sys/netinet6/mld6.c @@ -3267,6 +3267,7 @@ mld_init(void *unused __unused) mld_po.ip6po_hbh = &mld_ra.hbh; mld_po.ip6po_prefer_tempaddr = IP6PO_TEMPADDR_NOTPREFER; mld_po.ip6po_flags = IP6PO_DONTFRAG; + mld_po.ip6po_valid = IP6PO_VALID_HLIM | IP6PO_VALID_HBH; callout_init(&mldslow_callout, 1); callout_reset(&mldslow_callout, hz / MLD_SLOWHZ, mld_slowtimo, NULL); diff --git a/sys/netipsec/ipsec_offload.c b/sys/netipsec/ipsec_offload.c index 59a107881676..3583fc50f51b 100644 --- a/sys/netipsec/ipsec_offload.c +++ b/sys/netipsec/ipsec_offload.c @@ -289,19 +289,18 @@ ipsec_accel_sa_newkey_cb(if_t ifp, void *arg) be32toh(tq->sav->spi), tq->sav->flags, tq->sav->seq); priv = NULL; drv_spi = alloc_unr(drv_spi_unr); - if (tq->sav->accel_ifname != NULL && - strcmp(tq->sav->accel_ifname, if_name(ifp)) != 0) { - error = ipsec_accel_handle_sav(tq->sav, - ifp, drv_spi, priv, IFP_HS_REJECTED, NULL); - goto out; - } if (drv_spi == -1) { - /* XXXKIB */ dprintf("ipsec_accel_sa_install_newkey: cannot alloc " "drv_spi if %s spi %#x\n", if_name(ifp), be32toh(tq->sav->spi)); return (0); } + if (tq->sav->accel_ifname != NULL && + strcmp(tq->sav->accel_ifname, if_name(ifp)) != 0) { + error = ipsec_accel_handle_sav(tq->sav, + ifp, drv_spi, priv, IFP_HS_REJECTED, NULL); + goto out; + } error = ifp->if_ipsec_accel_m->if_sa_newkey(ifp, tq->sav, drv_spi, &priv); if (error != 0) { diff --git a/sys/netpfil/ipfilter/netinet/ip_htable.c b/sys/netpfil/ipfilter/netinet/ip_htable.c index 3f765cfab947..5f5c04732d69 100644 --- a/sys/netpfil/ipfilter/netinet/ip_htable.c +++ b/sys/netpfil/ipfilter/netinet/ip_htable.c @@ -96,6 +96,8 @@ typedef struct ipf_htable_softc_s { u_long ipf_nhtnodes[LOOKUP_POOL_SZ]; iphtable_t *ipf_htables[LOOKUP_POOL_SZ]; iphtent_t *ipf_node_explist; + ipftuneable_t *ipf_htable_tune; + u_int ipf_htable_size_max; } ipf_htable_softc_t; ipf_lookup_t ipf_htable_backend = { @@ -122,6 +124,18 @@ ipf_lookup_t ipf_htable_backend = { }; +static ipftuneable_t ipf_htable_tuneables[] = { + { { (void *)offsetof(ipf_htable_softc_t, ipf_htable_size_max) }, + "htable_size_max", 1, 0x7fffffff, + stsizeof(ipf_htable_softc_t, ipf_htable_size_max), + 0, NULL, NULL }, + { { NULL }, + NULL, 0, 0, + 0, + 0, NULL, NULL } +}; + + /* ------------------------------------------------------------------------ */ /* Function: ipf_htable_soft_create */ /* Returns: void * - NULL = failure, else pointer to local context */ @@ -142,6 +156,18 @@ ipf_htable_soft_create(ipf_main_softc_t *softc) bzero((char *)softh, sizeof(*softh)); + softh->ipf_htable_tune = ipf_tune_array_copy(softh, + sizeof(ipf_htable_tuneables), + ipf_htable_tuneables); + if (softh->ipf_htable_tune == NULL) { + ipf_htable_soft_destroy(softc, softh); + return (NULL); + } + if (ipf_tune_array_link(softc, softh->ipf_htable_tune) == -1) { + ipf_htable_soft_destroy(softc, softh); + return (NULL); + } + return (softh); } @@ -160,6 +186,12 @@ ipf_htable_soft_destroy(ipf_main_softc_t *softc, void *arg) { ipf_htable_softc_t *softh = arg; + if (softh->ipf_htable_tune != NULL) { + ipf_tune_array_unlink(softc, softh->ipf_htable_tune); + KFREES(softh->ipf_htable_tune, sizeof(ipf_htable_tuneables)); + softh->ipf_htable_tune = NULL; + } + KFREE(softh); } @@ -179,6 +211,8 @@ ipf_htable_soft_init(ipf_main_softc_t *softc, void *arg) bzero((char *)softh, sizeof(*softh)); + softh->ipf_htable_size_max = IPHTABLE_MAX_SIZE; + return (0); } @@ -327,6 +361,15 @@ ipf_htable_create(ipf_main_softc_t *softc, void *arg, iplookupop_t *op) iph->iph_name[sizeof(iph->iph_name) - 1] = '\0'; } + if ((iph->iph_size == 0) || + (iph->iph_size > softh->ipf_htable_size_max)) { + IPFERROR(30027); + return (EINVAL); + } + if (iph->iph_size > ( SIZE_MAX / sizeof(*iph->iph_table))) { + IPFERROR(30028); + return (EINVAL); + } KMALLOCS(iph->iph_table, iphtent_t **, iph->iph_size * sizeof(*iph->iph_table)); if (iph->iph_table == NULL) { diff --git a/sys/netpfil/ipfilter/netinet/ip_htable.h b/sys/netpfil/ipfilter/netinet/ip_htable.h index 55c289e57ff6..3a8782ccd4b2 100644 --- a/sys/netpfil/ipfilter/netinet/ip_htable.h +++ b/sys/netpfil/ipfilter/netinet/ip_htable.h @@ -55,6 +55,8 @@ typedef struct iphtable_s { char iph_name[FR_GROUPLEN]; /* hash table number */ } iphtable_t; +#define IPHTABLE_MAX_SIZE 1024 + /* iph_type */ #define IPHASH_LOOKUP 0 #define IPHASH_GROUPMAP 1 diff --git a/sys/netpfil/ipfw/pmod/tcpmod.c b/sys/netpfil/ipfw/pmod/tcpmod.c index 0338dc792c64..50074ee98cca 100644 --- a/sys/netpfil/ipfw/pmod/tcpmod.c +++ b/sys/netpfil/ipfw/pmod/tcpmod.c @@ -57,7 +57,8 @@ VNET_DEFINE_STATIC(uint32_t, tcpmod_setmss_eid) = 0; #define V_tcpmod_setmss_eid VNET(tcpmod_setmss_eid) static int -tcpmod_setmss(struct mbuf **mp, struct tcphdr *tcp, int tlen, uint16_t mss) +tcpmod_setmss(struct mbuf **mp, struct tcphdr *tcp, int tlen, uint16_t mss, + int *done) { struct mbuf *m; u_char *cp; @@ -72,8 +73,10 @@ tcpmod_setmss(struct mbuf **mp, struct tcphdr *tcp, int tlen, uint16_t mss) * TCP header with options. */ *mp = m = m_pullup(m, m->m_pkthdr.len); - if (m == NULL) + if (m == NULL) { + *done = 1; return (ret); + } } /* Parse TCP options. */ for (tlen -= sizeof(struct tcphdr), cp = (u_char *)(tcp + 1); @@ -114,7 +117,7 @@ tcpmod_setmss(struct mbuf **mp, struct tcphdr *tcp, int tlen, uint16_t mss) #ifdef INET6 static int -tcpmod_ipv6_setmss(struct mbuf **mp, uint16_t mss) +tcpmod_ipv6_setmss(struct mbuf **mp, uint16_t mss, int *done) { struct ip6_hdr *ip6; struct ip6_hbh *hbh; @@ -142,13 +145,13 @@ tcpmod_ipv6_setmss(struct mbuf **mp, uint16_t mss) /* We must have TCP options and enough data in a packet. */ if (hlen <= sizeof(struct tcphdr) || hlen > plen) return (IP_FW_DENY); - return (tcpmod_setmss(mp, tcp, hlen, mss)); + return (tcpmod_setmss(mp, tcp, hlen, mss, done)); } #endif /* INET6 */ #ifdef INET static int -tcpmod_ipv4_setmss(struct mbuf **mp, uint16_t mss) +tcpmod_ipv4_setmss(struct mbuf **mp, uint16_t mss, int *done) { struct tcphdr *tcp; struct ip *ip; @@ -162,7 +165,7 @@ tcpmod_ipv4_setmss(struct mbuf **mp, uint16_t mss) /* We must have TCP options and enough data in a packet. */ if (hlen <= sizeof(struct tcphdr) || hlen > plen) return (IP_FW_DENY); - return (tcpmod_setmss(mp, tcp, hlen, mss)); + return (tcpmod_setmss(mp, tcp, hlen, mss, done)); } #endif /* INET */ @@ -206,19 +209,23 @@ ipfw_tcpmod(struct ip_fw_chain *chain, struct ip_fw_args *args, switch (args->f_id.addr_type) { #ifdef INET case 4: - ret = tcpmod_ipv4_setmss(&args->m, htons(icmd->arg1)); + ret = tcpmod_ipv4_setmss(&args->m, htons(icmd->arg1), + done); break; #endif #ifdef INET6 case 6: - ret = tcpmod_ipv6_setmss(&args->m, htons(icmd->arg1)); + ret = tcpmod_ipv6_setmss(&args->m, htons(icmd->arg1), + done); break; #endif } /* * We return zero in both @ret and @done on success, and ipfw_chk() * will update rule counters. Otherwise a packet will not be matched - * by rule. + * by rule. We passed @done around above in case we hit a fatal error + * somewhere, we'll return non-zero but signal that rule processing + * cannot succeed. */ return (ret); } diff --git a/sys/netpfil/pf/pf_nl.c b/sys/netpfil/pf/pf_nl.c index 21d4db1b8478..993981a9c0de 100644 --- a/sys/netpfil/pf/pf_nl.c +++ b/sys/netpfil/pf/pf_nl.c @@ -2246,6 +2246,87 @@ pf_handle_table_set_addrs(struct nlmsghdr *hdr, struct nl_pstate *npt) return (error); } +static int +nlattr_add_pfr_addr(struct nl_writer *nw, int attr, const struct pfr_addr *a) +{ + int off = nlattr_add_nested(nw, attr); + if (off == 0) + return (false); + + nlattr_add_u32(nw, PFR_A_AF, a->pfra_af); + nlattr_add_u8(nw, PFR_A_NET, a->pfra_net); + nlattr_add_bool(nw, PFR_A_NOT, a->pfra_not); + nlattr_add_in6_addr(nw, PFR_A_ADDR, &a->pfra_u._pfra_ip6addr); + + nlattr_set_len(nw, off); + + return (true); +} + +static int +pf_handle_table_get_addrs(struct nlmsghdr *hdr, struct nl_pstate *npt) +{ + struct pfioc_table attrs = { 0 }; + struct pfr_addr *pfras; + struct nl_writer *nw = npt->nw; + struct genlmsghdr *ghdr_new; + int size = 0; + int error; + + PF_RULES_RLOCK_TRACKER; + + error = nl_parse_nlmsg(hdr, &table_addr_parser, npt, &attrs); + if (error != 0) + return (error); + + PF_RULES_RLOCK(); + /* Get required size. */ + error = pfr_get_addrs(&attrs.pfrio_table, NULL, + &size, attrs.pfrio_flags | PFR_FLAG_USERIOCTL); + if (error != 0) { + PF_RULES_RUNLOCK(); + return (error); + } + pfras = mallocarray(size, sizeof(struct pfr_addr), M_PF, + M_NOWAIT | M_ZERO); + if (pfras == NULL) { + PF_RULES_RUNLOCK(); + return (ENOMEM); + } + /* Now get the addresses. */ + error = pfr_get_addrs(&attrs.pfrio_table, pfras, + &size, attrs.pfrio_flags | PFR_FLAG_USERIOCTL); + PF_RULES_RUNLOCK(); + if (error != 0) + goto out; + + for (int i = 0; i < size; i++) { + if (!nlmsg_reply(nw, hdr, sizeof(struct genlmsghdr))) { + nlmsg_abort(nw); + error = ENOMEM; + goto out; + } + ghdr_new = nlmsg_reserve_object(nw, struct genlmsghdr); + ghdr_new->cmd = PFNL_CMD_TABLE_GET_ADDR; + ghdr_new->version = 0; + ghdr_new->reserved = 0; + + if (i == 0) + nlattr_add_u32(nw, PF_TA_ADDR_COUNT, size); + + nlattr_add_pfr_addr(nw, PF_TA_ADDR, &pfras[i]); + if (!nlmsg_end(nw)) { + nlmsg_abort(nw); + error = ENOMEM; + goto out; + } + } + +out: + free(pfras, M_PF); + return (error); +} + static const struct nlhdr_parser *all_parsers[] = { &state_parser, &addrule_parser, @@ -2504,6 +2585,13 @@ static const struct genl_cmd pf_cmds[] = { .cmd_flags = GENL_CMD_CAP_DO | GENL_CMD_CAP_HASPOL, .cmd_priv = PRIV_NETINET_PF, }, + { + .cmd_num = PFNL_CMD_TABLE_GET_ADDR, + .cmd_name = "TABLE_GET_ADDRS", + .cmd_cb = pf_handle_table_get_addrs, + .cmd_flags = GENL_CMD_CAP_DUMP | GENL_CMD_CAP_HASPOL, + .cmd_priv = PRIV_NETINET_PF, + }, }; void diff --git a/sys/netpfil/pf/pf_nl.h b/sys/netpfil/pf/pf_nl.h index d1538ab4ff5b..e1eb3e628df5 100644 --- a/sys/netpfil/pf/pf_nl.h +++ b/sys/netpfil/pf/pf_nl.h @@ -70,6 +70,7 @@ enum { PFNL_CMD_TABLE_ADD_ADDR = 32, PFNL_CMD_TABLE_DEL_ADDR = 33, PFNL_CMD_TABLE_SET_ADDR = 34, + PFNL_CMD_TABLE_GET_ADDR = 35, __PFNL_CMD_MAX, }; #define PFNL_CMD_MAX (__PFNL_CMD_MAX -1) @@ -485,6 +486,7 @@ enum pf_table_addrs_t { PF_TA_NBR_ADDED = 4, /* u32 */ PF_TA_NBR_DELETED = 5, /* u32 */ PF_TA_NBR_CHANGED = 6, /* u32 */ + PF_TA_ADDR_COUNT = 7, /* u32 */ }; #ifdef _KERNEL diff --git a/sys/riscv/include/vmm.h b/sys/riscv/include/vmm.h index e227dd825966..361140834805 100644 --- a/sys/riscv/include/vmm.h +++ b/sys/riscv/include/vmm.h @@ -103,9 +103,6 @@ enum vm_reg_name { #define VM_INTINFO_HWEXCEPTION (3 << 8) #define VM_INTINFO_SWINTR (4 << 8) -#define VM_MAX_NAMELEN 32 -#define VM_MAX_SUFFIXLEN 15 - #ifdef _KERNEL struct vm; diff --git a/sys/riscv/include/vmm_dev.h b/sys/riscv/include/vmm_dev.h index 4d30d5a1c35b..a60e545b8f52 100644 --- a/sys/riscv/include/vmm_dev.h +++ b/sys/riscv/include/vmm_dev.h @@ -38,6 +38,8 @@ #include <machine/vmm.h> +#include <dev/vmm/vmm_param.h> + struct vm_memmap { vm_paddr_t gpa; int segid; /* memory segment */ diff --git a/sys/riscv/vmm/vmm.c b/sys/riscv/vmm/vmm.c index a9eb9d144336..23b57ad3b7aa 100644 --- a/sys/riscv/vmm/vmm.c +++ b/sys/riscv/vmm/vmm.c @@ -38,7 +38,6 @@ #include <sys/linker.h> #include <sys/lock.h> #include <sys/malloc.h> -#include <sys/module.h> #include <sys/mutex.h> #include <sys/pcpu.h> #include <sys/proc.h> @@ -121,7 +120,7 @@ struct vm { volatile cpuset_t suspended_cpus; /* (i) suspended vcpus */ volatile cpuset_t halted_cpus; /* (x) cpus in a hard halt */ struct vm_mem mem; /* (i) [m+v] guest memory */ - char name[VM_MAX_NAMELEN]; /* (o) virtual machine name */ + char name[VM_MAX_NAMELEN + 1]; /* (o) virtual machine name */ struct vcpu **vcpu; /* (i) guest vcpus */ struct vmm_mmio_region mmio_region[VM_MAX_MMIO_REGIONS]; /* (o) guest MMIO regions */ @@ -133,8 +132,6 @@ struct vm { struct sx vcpus_init_lock; /* (o) */ }; -static bool vmm_initialized = false; - static MALLOC_DEFINE(M_VMM, "vmm", "vmm"); /* statistics */ @@ -146,10 +143,6 @@ static int vmm_ipinum; SYSCTL_INT(_hw_vmm, OID_AUTO, ipinum, CTLFLAG_RD, &vmm_ipinum, 0, "IPI vector used for vcpu notifications"); -u_int vm_maxcpu; -SYSCTL_UINT(_hw_vmm, OID_AUTO, maxcpu, CTLFLAG_RDTUN | CTLFLAG_NOFETCH, - &vm_maxcpu, 0, "Maximum number of vCPUs"); - static void vcpu_notify_event_locked(struct vcpu *vcpu); /* global statistics */ @@ -157,12 +150,6 @@ VMM_STAT(VMEXIT_COUNT, "total number of vm exits"); VMM_STAT(VMEXIT_IRQ, "number of vmexits for an irq"); VMM_STAT(VMEXIT_UNHANDLED, "number of vmexits for an unhandled exception"); -/* - * Upper limit on vm_maxcpu. We could increase this to 28 bits, but this - * is a safe value for now. - */ -#define VM_MAXCPU MIN(0xffff - 1, CPU_SETSIZE) - static void vcpu_cleanup(struct vcpu *vcpu, bool destroy) { @@ -210,75 +197,18 @@ vm_exitinfo(struct vcpu *vcpu) return (&vcpu->exitinfo); } -static int -vmm_init(void) +int +vmm_modinit(void) { - - vm_maxcpu = mp_ncpus; - - TUNABLE_INT_FETCH("hw.vmm.maxcpu", &vm_maxcpu); - - if (vm_maxcpu > VM_MAXCPU) { - printf("vmm: vm_maxcpu clamped to %u\n", VM_MAXCPU); - vm_maxcpu = VM_MAXCPU; - } - - if (vm_maxcpu == 0) - vm_maxcpu = 1; - return (vmmops_modinit()); } -static int -vmm_handler(module_t mod, int what, void *arg) +int +vmm_modcleanup(void) { - int error; - - switch (what) { - case MOD_LOAD: - error = vmmdev_init(); - if (error != 0) - break; - error = vmm_init(); - if (error == 0) - vmm_initialized = true; - else - (void)vmmdev_cleanup(); - break; - case MOD_UNLOAD: - error = vmmdev_cleanup(); - if (error == 0 && vmm_initialized) { - error = vmmops_modcleanup(); - if (error) { - /* - * Something bad happened - prevent new - * VMs from being created - */ - vmm_initialized = false; - } - } - break; - default: - error = 0; - break; - } - return (error); + return (vmmops_modcleanup()); } -static moduledata_t vmm_kmod = { - "vmm", - vmm_handler, - NULL -}; - -/* - * vmm initialization has the following dependencies: - * - * - vmm device initialization requires an initialized devfs. - */ -DECLARE_MODULE(vmm, vmm_kmod, SI_SUB_DEVFS + 1, SI_ORDER_ANY); -MODULE_VERSION(vmm, 1); - static void vm_init(struct vm *vm, bool create) { @@ -359,16 +289,6 @@ vm_create(const char *name, struct vm **retvm) struct vm *vm; int error; - /* - * If vmm.ko could not be successfully initialized then don't attempt - * to create the virtual machine. - */ - if (!vmm_initialized) - return (ENXIO); - - if (name == NULL || strlen(name) >= VM_MAX_NAMELEN) - return (EINVAL); - vm = malloc(sizeof(struct vm), M_VMM, M_WAITOK | M_ZERO); error = vm_mem_init(&vm->mem, 0, 1ul << 39); if (error != 0) { diff --git a/sys/sys/bio.h b/sys/sys/bio.h index 74d2b03bd180..5c12c858f3e5 100644 --- a/sys/sys/bio.h +++ b/sys/sys/bio.h @@ -37,6 +37,7 @@ #ifndef _SYS_BIO_H_ #define _SYS_BIO_H_ +#include <sys/_exterr.h> #include <sys/queue.h> #include <sys/disk_zone.h> @@ -65,10 +66,12 @@ #define BIO_TRANSIENT_MAPPING 0x20 #define BIO_VLIST 0x40 #define BIO_SWAP 0x200 /* Swap-related I/O */ +#define BIO_EXTERR 0x2000 #define BIO_SPEEDUP_WRITE 0x4000 /* Resource shortage at upper layers */ #define BIO_SPEEDUP_TRIM 0x8000 /* Resource shortage at upper layers */ -#define PRINT_BIO_FLAGS "\20\20speedup_trim\17speedup_write\12swap\7vlist\6transient_mapping\5unmapped" \ +#define PRINT_BIO_FLAGS "\20\20speedup_trim\17speedup_write\16exterr" \ + "\12swap\7vlist\6transient_mapping\5unmapped" \ "\4ordered\3onqueue\2done\1error" @@ -94,7 +97,6 @@ struct bio { struct vm_page **bio_ma; /* Or unmapped. */ int bio_ma_offset; /* Offset in the first page of bio_ma. */ int bio_ma_n; /* Number of pages in bio_ma. */ - int bio_error; /* Errno for BIO_ERROR. */ long bio_resid; /* Remaining I/O in bytes. */ void (*bio_done)(struct bio *); void *bio_driver1; /* Private use by the provider. */ @@ -130,8 +132,12 @@ struct bio { /* XXX: these go away when bio chaining is introduced */ daddr_t bio_pblkno; /* physical block number */ + struct kexterr bio_exterr; }; +/* Errno for BIO_ERROR. */ +#define bio_error bio_exterr.error + struct uio; struct devstat; diff --git a/sys/sys/buf.h b/sys/sys/buf.h index 064d5cb05214..f08f05e6d50f 100644 --- a/sys/sys/buf.h +++ b/sys/sys/buf.h @@ -37,6 +37,7 @@ #ifndef _SYS_BUF_H_ #define _SYS_BUF_H_ +#include <sys/_exterr.h> #include <sys/bufobj.h> #include <sys/queue.h> #include <sys/lock.h> @@ -98,7 +99,6 @@ struct buf { long b_bcount; void *b_caller1; caddr_t b_data; - int b_error; uint16_t b_iocmd; /* BIO_* bio_cmd from bio.h */ uint16_t b_ioflags; /* BIO_* bio_flags from bio.h */ off_t b_iooffset; @@ -153,10 +153,12 @@ struct buf { #elif defined(BUF_TRACKING) const char *b_io_tracking; #endif + struct kexterr b_exterr; struct vm_page *b_pages[]; }; #define b_object b_bufobj->bo_object +#define b_error b_exterr.error /* * These flags are kept in b_flags. @@ -390,6 +392,12 @@ struct buf { _lockmgr_disown(&(bp)->b_lock, LOCK_FILE, LOCK_LINE) #endif +#define BUF_EXTERR_FROM_CURTHR(bp) \ + bp->b_exterr = curthread->td_kexterr + +#define BUF_EXTERR_TO_CURTHR(bp) \ + curthread->td_kexterr = bp->b_exterr + #endif /* _KERNEL */ struct buf_queue_head { diff --git a/sys/sys/exterr_cat.h b/sys/sys/exterr_cat.h index 43f31e1d5dd6..318e774542ca 100644 --- a/sys/sys/exterr_cat.h +++ b/sys/sys/exterr_cat.h @@ -21,6 +21,9 @@ #define EXTERR_CAT_BRIDGE 7 #define EXTERR_CAT_SWAP 8 #define EXTERR_CAT_VFSSYSCALL 9 +#define EXTERR_CAT_VFSBIO 10 +#define EXTERR_CAT_GEOMVFS 11 +#define EXTERR_CAT_GEOM 12 #endif diff --git a/sys/sys/exterrvar.h b/sys/sys/exterrvar.h index 6783a0d2d84f..8e2961356a1e 100644 --- a/sys/sys/exterrvar.h +++ b/sys/sys/exterrvar.h @@ -37,6 +37,26 @@ #define SET_ERROR_MSG(mmsg) NULL #endif +#define _SET_ERROR2_KE(kep, eerror, mmsg, pp1, pp2) ({ \ + (kep)->error = (eerror); \ + (kep)->cat = EXTERR_CATEGORY; \ + (kep)->msg = SET_ERROR_MSG(mmsg); \ + (kep)->p1 = (pp1); \ + (kep)->p2 = (pp2); \ + (kep)->src_line = __LINE__; \ + (kep)->error; \ +}) +#define _SET_ERROR0_KE(kep, eerror, mmsg) \ + _SET_ERROR2_KE(kep, eerror, mmsg, 0, 0) +#define _SET_ERROR1_KE(kep, eerror, mmsg, pp1) \ + _SET_ERROR2_KE(kep, eerror, mmsg, pp1, 0) + +#define _EXTERROR_MACRO_KE(kep, eerror, mmsg, _1, _2, NAME, ...) \ + NAME +#define EXTERROR_KE(...) \ + _EXTERROR_MACRO_KE(__VA_ARGS__, _SET_ERROR2_KE, _SET_ERROR1_KE, \ + _SET_ERROR0_KE)(__VA_ARGS__) + #define _SET_ERROR2(eerror, mmsg, pp1, pp2) \ exterr_set(eerror, EXTERR_CATEGORY, SET_ERROR_MSG(mmsg), \ (uintptr_t)(pp1), (uintptr_t)(pp2), __LINE__) @@ -49,6 +69,9 @@ _EXTERROR_MACRO(__VA_ARGS__, _SET_ERROR2, _SET_ERROR1, \ _SET_ERROR0)(__VA_ARGS__) +void exterr_clear(struct kexterr *ke); +void exterr_db_print(struct kexterr *ke); +int exterr_set_from(const struct kexterr *ke); int exterr_set(int eerror, int category, const char *mmsg, uintptr_t pp1, uintptr_t pp2, int line); int exterr_to_ue(struct thread *td, struct uexterror *ue); diff --git a/sys/sys/param.h b/sys/sys/param.h index 957f1762a17c..bdfe4a1cfde3 100644 --- a/sys/sys/param.h +++ b/sys/sys/param.h @@ -74,7 +74,7 @@ * cannot include sys/param.h and should only be updated here. */ #undef __FreeBSD_version -#define __FreeBSD_version 1600003 +#define __FreeBSD_version 1600004 /* * __FreeBSD_kernel__ indicates that this system uses the kernel of FreeBSD, diff --git a/tests/sys/fs/fusefs/fallocate.cc b/tests/sys/fs/fusefs/fallocate.cc index 4e5b047b78b7..1a3a0af36236 100644 --- a/tests/sys/fs/fusefs/fallocate.cc +++ b/tests/sys/fs/fusefs/fallocate.cc @@ -205,7 +205,7 @@ TEST_F(Fspacectl, enosys) EXPECT_EQ(0, fspacectl(fd, SPACECTL_DEALLOC, &rqsr, 0, NULL)); /* Neither should posix_fallocate query the daemon */ - EXPECT_EQ(EINVAL, posix_fallocate(fd, off1, len1)); + EXPECT_EQ(EOPNOTSUPP, posix_fallocate(fd, off1, len1)); leak(fd); } @@ -548,7 +548,7 @@ INSTANTIATE_TEST_SUITE_P(FspacectlCache, FspacectlCache, /* * If the server returns ENOSYS, it indicates that the server does not support - * FUSE_FALLOCATE. This and future calls should return EINVAL. + * FUSE_FALLOCATE. This and future calls should return EOPNOTSUPP. */ TEST_F(PosixFallocate, enosys) { @@ -570,10 +570,10 @@ TEST_F(PosixFallocate, enosys) fd = open(FULLPATH, O_RDWR); ASSERT_LE(0, fd) << strerror(errno); - EXPECT_EQ(EINVAL, posix_fallocate(fd, off0, len0)); + EXPECT_EQ(EOPNOTSUPP, posix_fallocate(fd, off0, len0)); /* Subsequent calls shouldn't query the daemon*/ - EXPECT_EQ(EINVAL, posix_fallocate(fd, off0, len0)); + EXPECT_EQ(EOPNOTSUPP, posix_fallocate(fd, off0, len0)); /* Neither should VOP_DEALLOCATE query the daemon */ EXPECT_EQ(0, fspacectl(fd, SPACECTL_DEALLOC, &rqsr, 0, NULL)); @@ -607,10 +607,10 @@ TEST_F(PosixFallocate, eopnotsupp) fd = open(FULLPATH, O_RDWR); ASSERT_LE(0, fd) << strerror(errno); - EXPECT_EQ(EINVAL, posix_fallocate(fd, fsize, length)); + EXPECT_EQ(EOPNOTSUPP, posix_fallocate(fd, fsize, length)); /* Subsequent calls should still query the daemon*/ - EXPECT_EQ(EINVAL, posix_fallocate(fd, offset, length)); + EXPECT_EQ(EOPNOTSUPP, posix_fallocate(fd, offset, length)); /* And subsequent VOP_DEALLOCATE calls should also query the daemon */ rqsr.r_len = length; @@ -759,7 +759,7 @@ TEST_F(PosixFallocate, rlimit_fsize) } /* With older servers, no FUSE_FALLOCATE should be attempted */ -TEST_F(PosixFallocate_7_18, einval) +TEST_F(PosixFallocate_7_18, eopnotsupp) { const char FULLPATH[] = "mountpoint/some_file.txt"; const char RELPATH[] = "some_file.txt"; @@ -773,7 +773,7 @@ TEST_F(PosixFallocate_7_18, einval) fd = open(FULLPATH, O_RDWR); ASSERT_LE(0, fd) << strerror(errno); - EXPECT_EQ(EINVAL, posix_fallocate(fd, offset, length)); + EXPECT_EQ(EOPNOTSUPP, posix_fallocate(fd, offset, length)); leak(fd); } diff --git a/tests/sys/fs/fusefs/xattr.cc b/tests/sys/fs/fusefs/xattr.cc index 0ab203c96254..afeacd4a249e 100644 --- a/tests/sys/fs/fusefs/xattr.cc +++ b/tests/sys/fs/fusefs/xattr.cc @@ -100,7 +100,11 @@ void expect_removexattr(uint64_t ino, const char *attr, int error) ).WillOnce(Invoke(ReturnErrno(error))); } -void expect_setxattr(uint64_t ino, const char *attr, const char *value, +/* + * Expect a FUSE_SETXATTR request in the format used by protocol 7.33 and + * later, with the FUSE_SETXATTR_EXT bit set. + */ +void expect_setxattr_ext(uint64_t ino, const char *attr, const char *value, ProcessMockerT r) { EXPECT_CALL(*m_mock, process( @@ -119,16 +123,10 @@ void expect_setxattr(uint64_t ino, const char *attr, const char *value, ).WillOnce(Invoke(r)); } -}; - -class Xattr_7_32:public FuseTest { -public: -virtual void SetUp() -{ - m_kernel_minor_version = 32; - FuseTest::SetUp(); -} - +/* + * Expect a FUSE_SETXATTR request in the format used by protocol 7.32 and + * earlier. + */ void expect_setxattr_7_32(uint64_t ino, const char *attr, const char *value, ProcessMockerT r) { @@ -148,6 +146,15 @@ void expect_setxattr_7_32(uint64_t ino, const char *attr, const char *value, } }; +class Xattr_7_32: public Xattr { +public: +virtual void SetUp() +{ + m_kernel_minor_version = 32; + Xattr::SetUp(); +} +}; + class Getxattr: public Xattr {}; class Listxattr: public Xattr {}; @@ -182,6 +189,13 @@ void TearDown() { class Removexattr: public Xattr {}; class Setxattr: public Xattr {}; +class SetxattrExt: public Setxattr { +public: +virtual void SetUp() { + m_init_flags |= FUSE_SETXATTR_EXT; + Setxattr::SetUp(); +} +}; class Setxattr_7_32:public Xattr_7_32 {}; class RofsXattr: public Xattr { public: @@ -773,7 +787,7 @@ TEST_F(Setxattr, enosys) ssize_t r; expect_lookup(RELPATH, ino, S_IFREG | 0644, 0, 2); - expect_setxattr(ino, "user.foo", value, ReturnErrno(ENOSYS)); + expect_setxattr_7_32(ino, "user.foo", value, ReturnErrno(ENOSYS)); r = extattr_set_file(FULLPATH, ns, "foo", (const void*)value, value_len); @@ -800,7 +814,7 @@ TEST_F(Setxattr, enotsup) ssize_t r; expect_lookup(RELPATH, ino, S_IFREG | 0644, 0, 1); - expect_setxattr(ino, "user.foo", value, ReturnErrno(ENOTSUP)); + expect_setxattr_7_32(ino, "user.foo", value, ReturnErrno(ENOTSUP)); r = extattr_set_file(FULLPATH, ns, "foo", (const void*)value, value_len); @@ -820,7 +834,7 @@ TEST_F(Setxattr, user) ssize_t r; expect_lookup(RELPATH, ino, S_IFREG | 0644, 0, 1); - expect_setxattr(ino, "user.foo", value, ReturnErrno(0)); + expect_setxattr_7_32(ino, "user.foo", value, ReturnErrno(0)); r = extattr_set_file(FULLPATH, ns, "foo", (const void*)value, value_len); @@ -839,7 +853,7 @@ TEST_F(Setxattr, system) ssize_t r; expect_lookup(RELPATH, ino, S_IFREG | 0644, 0, 1); - expect_setxattr(ino, "system.foo", value, ReturnErrno(0)); + expect_setxattr_7_32(ino, "system.foo", value, ReturnErrno(0)); r = extattr_set_file(FULLPATH, ns, "foo", (const void*)value, value_len); @@ -847,6 +861,10 @@ TEST_F(Setxattr, system) } +/* + * For servers using protocol 7.32 and older, the kernel should use the older + * FUSE_SETXATTR format. + */ TEST_F(Setxattr_7_32, ok) { uint64_t ino = 42; @@ -863,6 +881,25 @@ TEST_F(Setxattr_7_32, ok) ASSERT_EQ(value_len, r) << strerror(errno); } +/* + * Successfully set a user attribute using the extended format + */ +TEST_F(SetxattrExt, user) +{ + uint64_t ino = 42; + const char value[] = "whatever"; + ssize_t value_len = strlen(value) + 1; + int ns = EXTATTR_NAMESPACE_USER; + ssize_t r; + + expect_lookup(RELPATH, ino, S_IFREG | 0644, 0, 1); + expect_setxattr_ext(ino, "user.foo", value, ReturnErrno(0)); + + r = extattr_set_file(FULLPATH, ns, "foo", (const void*)value, + value_len); + ASSERT_EQ(value_len, r) << strerror(errno); +} + TEST_F(RofsXattr, deleteextattr_erofs) { uint64_t ino = 42; diff --git a/usr.bin/cut/cut.1 b/usr.bin/cut/cut.1 index 4f5b7b9bb9a5..42a1726e6e24 100644 --- a/usr.bin/cut/cut.1 +++ b/usr.bin/cut/cut.1 @@ -1,3 +1,6 @@ +.\" +.\" SPDX-License-Identifier: BSD-3-Clause +.\" .\" Copyright (c) 1989, 1990, 1993 .\" The Regents of the University of California. All rights reserved. .\" @@ -86,7 +89,7 @@ It is not an error to select columns or fields not present in the input line. .Pp The options are as follows: -.Bl -tag -width indent +.Bl -tag -width "-d delim" .It Fl b Ar list The .Ar list diff --git a/usr.bin/id/id.c b/usr.bin/id/id.c index 5f9d2670caa3..7ba07daad11e 100644 --- a/usr.bin/id/id.c +++ b/usr.bin/id/id.c @@ -67,19 +67,13 @@ main(int argc, char *argv[]) { struct group *gr; struct passwd *pw; -#ifdef USE_BSM_AUDIT - bool Aflag; -#endif - bool Gflag, Mflag, Pflag; + bool Aflag, Gflag, Mflag, Pflag; bool cflag, dflag, gflag, nflag, pflag, rflag, sflag, uflag; int ch, combo, error, id; const char *myname, *optstr; char loginclass[MAXLOGNAME]; -#ifdef USE_BSM_AUDIT - Aflag = false; -#endif - Gflag = Mflag = Pflag = false; + Aflag = Gflag = Mflag = Pflag = false; cflag = dflag = gflag = nflag = pflag = rflag = sflag = uflag = false; myname = getprogname(); diff --git a/usr.bin/ktrace/subr.c b/usr.bin/ktrace/subr.c index 422a37bb413d..fac335948f46 100644 --- a/usr.bin/ktrace/subr.c +++ b/usr.bin/ktrace/subr.c @@ -89,6 +89,7 @@ getpoints(char *s) break; case 'x': facs |= KTRFAC_EXTERR; + break; case '+': facs |= DEF_POINTS; break; diff --git a/usr.bin/sockstat/main.c b/usr.bin/sockstat/main.c index 07663e54534d..1f174d827e1a 100644 --- a/usr.bin/sockstat/main.c +++ b/usr.bin/sockstat/main.c @@ -88,6 +88,7 @@ static bool opt_A; /* Show kernel address of pcb */ static bool opt_b; /* Show BBLog state */ static bool opt_C; /* Show congestion control */ static bool opt_c; /* Show connected sockets */ +static bool opt_F; /* Show sockets for selected user only */ static bool opt_f; /* Show FIB numbers */ static bool opt_I; /* Show spliced socket addresses */ static bool opt_i; /* Show inp_gencnt */ @@ -115,6 +116,12 @@ static size_t default_numprotos = nitems(default_protos); static int *protos; /* protocols to use */ static size_t numprotos; /* allocated size of protos[] */ +/* + * Show sockets for user username or UID specified + */ +static char *filter_user_optarg = NULL; /* saved optarg for username/UID resolving */ +static uid_t filter_user_uid; /* UID to show sockets for */ + struct addr { union { struct sockaddr_storage address; @@ -217,6 +224,18 @@ _enforce_ksize(size_t received_size, size_t expected_size, const char *struct_na } #define enforce_ksize(_sz, _struct) (_enforce_ksize(_sz, sizeof(_struct), #_struct)) +static inline bool +filtered_uid(uid_t i_uid) +{ + return ((i_uid) == filter_user_uid); +} + +static inline bool +need_nosocks(void) +{ + return !(opt_F || (opt_j >= 0)); +} + static int get_proto_type(const char *proto) { @@ -758,7 +777,8 @@ gather_inet(int proto) if (sock->socket != 0) RB_INSERT(socks_t, &socks, sock); else - SLIST_INSERT_HEAD(&nosocks, sock, socket_list); + if (need_nosocks()) + SLIST_INSERT_HEAD(&nosocks, sock, socket_list); } out: free(buf); @@ -862,6 +882,8 @@ getfiles(void) struct xfile *xfiles; size_t len, olen; + int filenum = 0; + olen = len = sizeof(*xfiles); if ((xfiles = malloc(len)) == NULL) xo_err(1, "malloc()"); @@ -880,14 +902,23 @@ getfiles(void) if ((files = malloc(nfiles * sizeof(struct file))) == NULL) xo_err(1, "malloc()"); + /* Fill files structure, optionally for specified user */ for (int i = 0; i < nfiles; i++) { - files[i].xf_data = xfiles[i].xf_data; - files[i].xf_pid = xfiles[i].xf_pid; - files[i].xf_uid = xfiles[i].xf_uid; - files[i].xf_fd = xfiles[i].xf_fd; - RB_INSERT(files_t, &ftree, &files[i]); + if (opt_F && !filtered_uid(xfiles[i].xf_uid)) + continue; + files[filenum].xf_data = xfiles[i].xf_data; + files[filenum].xf_pid = xfiles[i].xf_pid; + files[filenum].xf_uid = xfiles[i].xf_uid; + files[filenum].xf_fd = xfiles[i].xf_fd; + RB_INSERT(files_t, &ftree, &files[filenum]); + filenum++; } + /* Adjust global nfiles to match the number of files we + * actually filled into files[] array + */ + nfiles = filenum; + free(xfiles); } @@ -1584,6 +1615,24 @@ display_sock(struct sock *s, struct col_widths *cw, char *buf, size_t bufsize) static void display(void) { + static const char *__HDR_USER="USER", + *__HDR_COMMAND="COMMAND", + *__HDR_PID="PID", + *__HDR_FD="FD", + *__HDR_PROTO="PROTO", + *__HDR_LOCAL_ADDRESS="LOCAL ADDRESS", + *__HDR_FOREIGN_ADDRESS="FOREIGN ADDRESS", + *__HDR_PCB_KVA="PCB KVA", + *__HDR_FIB="FIB", + *__HDR_SPLICE_ADDRESS="SPLICE ADDRESS", + *__HDR_ID="ID", + *__HDR_ENCAPS="ENCAPS", + *__HDR_PATH_STATE="PATH STATE", + *__HDR_CONN_STATE="CONN STATE", + *__HDR_BBLOG_STATE="BBLOG STATE", + *__HDR_STACK="STACK", + *__HDR_CC="CC"; + struct passwd *pwd; struct file *xf; struct sock *s; @@ -1598,23 +1647,23 @@ display(void) if (!is_xo_style_encoding) { cw = (struct col_widths) { - .user = strlen("USER"), + .user = strlen(__HDR_USER), .command = 10, - .pid = strlen("PID"), - .fd = strlen("FD"), - .proto = strlen("PROTO"), - .local_addr = opt_w ? strlen("LOCAL ADDRESS") : 21, - .foreign_addr = opt_w ? strlen("FOREIGN ADDRESS") : 21, + .pid = strlen(__HDR_PID), + .fd = strlen(__HDR_FD), + .proto = strlen(__HDR_PROTO), + .local_addr = opt_w ? strlen(__HDR_LOCAL_ADDRESS) : 21, + .foreign_addr = opt_w ? strlen(__HDR_FOREIGN_ADDRESS) : 21, .pcb_kva = 18, - .fib = strlen("FIB"), - .splice_address = strlen("SPLICE ADDRESS"), - .inp_gencnt = strlen("ID"), - .encaps = strlen("ENCAPS"), - .path_state = strlen("PATH STATE"), - .conn_state = strlen("CONN STATE"), - .bblog_state = strlen("BBLOG STATE"), - .stack = strlen("STACK"), - .cc = strlen("CC"), + .fib = strlen(__HDR_FIB), + .splice_address = strlen(__HDR_SPLICE_ADDRESS), + .inp_gencnt = strlen(__HDR_ID), + .encaps = strlen(__HDR_ENCAPS), + .path_state = strlen(__HDR_PATH_STATE), + .conn_state = strlen(__HDR_CONN_STATE), + .bblog_state = strlen(__HDR_BBLOG_STATE), + .stack = strlen(__HDR_STACK), + .cc = strlen(__HDR_CC), }; calculate_column_widths(&cw); } else @@ -1625,34 +1674,34 @@ display(void) xo_open_list("socket"); if (!opt_q) { xo_emit("{T:/%-*s} {T:/%-*s} {T:/%*s} {T:/%*s} {T:/%-*s} " - "{T:/%-*s} {T:/%-*s}", cw.user, "USER", cw.command, - "COMMAND", cw.pid, "PID", cw.fd, "FD", cw.proto, - "PROTO", cw.local_addr, "LOCAL ADDRESS", - cw.foreign_addr, "FOREIGN ADDRESS"); + "{T:/%-*s} {T:/%-*s}", cw.user, __HDR_USER, cw.command, + __HDR_COMMAND, cw.pid, __HDR_PID, cw.fd, __HDR_FD, cw.proto, + __HDR_PROTO, cw.local_addr, __HDR_LOCAL_ADDRESS, + cw.foreign_addr, __HDR_FOREIGN_ADDRESS); if (opt_A) - xo_emit(" {T:/%-*s}", cw.pcb_kva, "PCB KVA"); + xo_emit(" {T:/%-*s}", cw.pcb_kva, __HDR_PCB_KVA); if (opt_f) /* RT_MAXFIBS is 65535. */ - xo_emit(" {T:/%*s}", cw.fib, "FIB"); + xo_emit(" {T:/%*s}", cw.fib, __HDR_FIB); if (opt_I) xo_emit(" {T:/%-*s}", cw.splice_address, - "SPLICE ADDRESS"); + __HDR_SPLICE_ADDRESS); if (opt_i) - xo_emit(" {T:/%*s}", cw.inp_gencnt, "ID"); + xo_emit(" {T:/%*s}", cw.inp_gencnt, __HDR_ID); if (opt_U) - xo_emit(" {T:/%*s}", cw.encaps, "ENCAPS"); + xo_emit(" {T:/%*s}", cw.encaps, __HDR_ENCAPS); if (opt_s) { if (show_path_state) xo_emit(" {T:/%-*s}", cw.path_state, - "PATH STATE"); - xo_emit(" {T:/%-*s}", cw.conn_state, "CONN STATE"); + __HDR_PATH_STATE); + xo_emit(" {T:/%-*s}", cw.conn_state, __HDR_CONN_STATE); } if (opt_b) - xo_emit(" {T:/%-*s}", cw.bblog_state, "BBLOG STATE"); + xo_emit(" {T:/%-*s}", cw.bblog_state, __HDR_BBLOG_STATE); if (opt_S) - xo_emit(" {T:/%-*s}", cw.stack, "STACK"); + xo_emit(" {T:/%-*s}", cw.stack, __HDR_STACK); if (opt_C) - xo_emit(" {T:/%-*s}", cw.cc, "CC"); + xo_emit(" {T:/%-*s}", cw.cc, __HDR_CC); xo_emit("\n"); } cap_setpassent(cappwd, 1); @@ -1684,7 +1733,7 @@ display(void) xo_close_instance("socket"); } } - if (opt_j >= 0) + if (!need_nosocks()) goto out; SLIST_FOREACH(s, &nosocks, socket_list) { if (!check_ports(s)) @@ -1775,11 +1824,44 @@ jail_getvnet(int jid) return (vnet); } +/* + * Parse username and/or UID + */ +static bool +parse_filter_user(void) +{ + struct passwd *pwd; + char *ep; + uid_t uid; + bool rv = false; + + uid = (uid_t)strtol(filter_user_optarg, &ep, 10); + + /* Open and/or rewind capsicumized password file */ + cap_setpassent(cappwd, 1); + + if (*ep == '\0') { + /* We have an UID specified, check if it's valid */ + if ((pwd = cap_getpwuid(cappwd, uid)) == NULL) + goto out; + filter_user_uid = uid; + } else { + /* Check if we have a valid username */ + if ((pwd = cap_getpwnam(cappwd, filter_user_optarg)) == NULL) + goto out; + filter_user_uid = pwd->pw_uid; + } + + rv = true; +out: + return (rv); +} + static void usage(void) { xo_error( -"usage: sockstat [--libxo ...] [-46AbCcfIiLlnqSsUuvw] [-j jid] [-p ports]\n" +"usage: sockstat [--libxo ...] [-46AbCcfIiLlnqSsUuvw] [-F uid/username] [-j jid] [-p ports]\n" " [-P protocols]\n"); exit(1); } @@ -1789,8 +1871,8 @@ main(int argc, char *argv[]) { cap_channel_t *capcas; cap_net_limit_t *limit; - const char *pwdcmds[] = { "setpassent", "getpwuid" }; - const char *pwdfields[] = { "pw_name" }; + const char *pwdcmds[] = { "setpassent", "getpwuid", "getpwnam" }; + const char *pwdfields[] = { "pw_name", "pw_uid" }; int protos_defined = -1; int o, i, err; @@ -1803,7 +1885,7 @@ main(int argc, char *argv[]) is_xo_style_encoding = true; } opt_j = -1; - while ((o = getopt(argc, argv, "46AbCcfIij:Llnp:P:qSsUuvw")) != -1) + while ((o = getopt(argc, argv, "46AbCcF:fIij:Llnp:P:qSsUuvw")) != -1) switch (o) { case '4': opt_4 = true; @@ -1823,6 +1905,11 @@ main(int argc, char *argv[]) case 'c': opt_c = true; break; + case 'F': + /* Save optarg for later use when we enter capabilities mode */ + filter_user_optarg = optarg; + opt_F = true; + break; case 'f': opt_f = true; break; @@ -1934,6 +2021,9 @@ main(int argc, char *argv[]) if (cap_pwd_limit_fields(cappwd, pwdfields, nitems(pwdfields)) < 0) xo_err(1, "Unable to apply pwd commands limits"); + if (opt_F && !parse_filter_user()) + xo_errx(1, "Invalid username or UID specified"); + if ((!opt_4 && !opt_6) && protos_defined != -1) opt_4 = opt_6 = true; if (!opt_4 && !opt_6 && !opt_u) diff --git a/usr.bin/sockstat/sockstat.1 b/usr.bin/sockstat/sockstat.1 index 1498fb1d88f7..b0fae81ee566 100644 --- a/usr.bin/sockstat/sockstat.1 +++ b/usr.bin/sockstat/sockstat.1 @@ -25,7 +25,7 @@ .\" (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF .\" THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. .\" -.Dd October 14, 2025 +.Dd October 29, 2025 .Dt SOCKSTAT 1 .Os .Sh NAME @@ -35,6 +35,7 @@ .Nm .Op Fl -libxo .Op Fl 46AbCcfIiLlnqSsUuvw +.Op Fl F Ar user .Op Fl j Ar jail .Op Fl p Ar ports .Op Fl P Ar protocols @@ -73,6 +74,10 @@ Display the congestion control module, if applicable. This is currently only implemented for TCP. .It Fl c Show connected sockets. +.It Fl F Ar user +Show sockets for specified +.Ar user +(user name or UID) only. .It Fl f Show the FIB number of each socket. .It Fl I diff --git a/usr.sbin/cron/lib/env.c b/usr.sbin/cron/lib/env.c index 287dd8636293..5a2d7ad60756 100644 --- a/usr.sbin/cron/lib/env.c +++ b/usr.sbin/cron/lib/env.c @@ -55,7 +55,7 @@ env_copy(char **envp) for (count = 0; envp[count] != NULL; count++) ; - p = (char **) malloc((count+1) * sizeof(char *)); /* 1 for the NULL */ + p = (char **) reallocarray(NULL, count+1, sizeof(char *)); /* 1 for the NULL */ if (p == NULL) { errno = ENOMEM; return NULL; @@ -112,8 +112,7 @@ env_set(char **envp, char *envstr) * one, save our string over the old null pointer, and return resized * array. */ - p = (char **) realloc((void *) envp, - (unsigned) ((count+1) * sizeof(char *))); + p = (char **) reallocarray(envp, count+1, sizeof(char *)); if (p == NULL) { /* XXX env_free(envp); */ errno = ENOMEM; diff --git a/usr.sbin/freebsd-update/freebsd-update.sh b/usr.sbin/freebsd-update/freebsd-update.sh index 143d93a6dcc0..2a07bc1fb7bc 100644 --- a/usr.sbin/freebsd-update/freebsd-update.sh +++ b/usr.sbin/freebsd-update/freebsd-update.sh @@ -1103,7 +1103,7 @@ IDS_check_params () { check_pkgbase() { # Packaged base requires that pkg is bootstrapped. - if ! pkg -r ${BASEDIR} -N >/dev/null 2>/dev/null; then + if ! pkg -N -r ${BASEDIR} >/dev/null 2>/dev/null; then return 1 fi # uname(1) is used by pkg to determine ABI, so it should exist. diff --git a/usr.sbin/inetd/inetd.8 b/usr.sbin/inetd/inetd.8 index d2a4331bb79c..189415caa711 100644 --- a/usr.sbin/inetd/inetd.8 +++ b/usr.sbin/inetd/inetd.8 @@ -25,7 +25,7 @@ .\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF .\" SUCH DAMAGE. .\" -.Dd September 25, 2025 +.Dd November 5, 2025 .Dt INETD 8 .Os .Sh NAME @@ -787,44 +787,6 @@ the pid of the currently running .Sh "EXAMPLES" Examples for a variety of services are available in .Pa /etc/inetd.conf . -.Pp -It includes examples for -.Nm bootpd , -.Nm comsat , -.Nm cvs , -.Nm date , -.Nm fingerd , -.Nm ftpd , -.Nm imapd , -.Nm nc , -.Nm nmbd , -.Nm nntpd , -.Nm rlogind , -.Nm rpc.rquotad , -.Nm rpc.rusersd , -.Nm rpc.rwalld , -.Nm rpc.statd , -.Nm rpc.sprayd , -.Nm rshd , -.Nm prometheus_sysctl_exporter , -.Nm smtpd , -.Nm smbd , -.Nm swat -.Nm talkd , -.Nm telnetd , -.Nm tftpd , -.Nm uucpd . -.Pp -The internal services provided by -.Nm -for daytime, time, echo, discard and chargen are also -included, as well as chargen for -.Nm ipsec -Authentication Headers -.Pp -Examples for handling auth requests via -.Nm identd , -are similarly included. .Sh "ERROR MESSAGES" The .Nm diff --git a/usr.sbin/mixer/mixer.8 b/usr.sbin/mixer/mixer.8 index 819d8ae73ab1..d7de675bceee 100644 --- a/usr.sbin/mixer/mixer.8 +++ b/usr.sbin/mixer/mixer.8 @@ -19,7 +19,7 @@ .\" OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN .\" THE SOFTWARE. .\" -.Dd August 14, 2024 +.Dd October 31, 2025 .Dt MIXER 8 .Os .Sh NAME @@ -28,7 +28,7 @@ .Sh SYNOPSIS .Nm .Op Fl f Ar device -.Op Fl d Ar pcmN | N Op Fl V Ar voss_device:mode +.Op Fl d Ar pcmX | X Op Fl V Ar voss_device:mode .Op Fl os .Op Ar dev Ns Op Cm \&. Ns Ar control Ns Op Cm \&= Ns Ar value .Ar ... @@ -47,10 +47,10 @@ The options are as follows: .It Fl a Print the values for all mixer devices available in the system .Pq see Sx FILES . -.It Fl d Ar pcmN | N +.It Fl d Ar pcmX | X Change the default audio card to -.Ar pcmN , -where N is the unit number (e.g for pcm0, the unit number is 0). +.Ar pcmX , +where X is the device's unit number (e.g for pcm0, the unit number is 0). See .Sx EXAMPLES on how to list all available audio devices in the system. @@ -246,30 +246,22 @@ makes the only recording device. .El .Sh FILES -.Bl -tag -width /dev/mixerN -compact -.It Pa /dev/mixerN -The mixer device, where -.Ar N -is the number of that device, for example -.Ar /dev/mixer0 . -PCM cards and mixers have a 1:1 relationship, which means that +.Bl -tag -width "/dev/mixerX" -compact +.It Pa /dev/mixerX +The mixer device, where X is the unit number of that device, +.Pa /dev/dsp* +devices and +.Pa /dev/mixer* +devices have a 1:1 relationship, which means that, for instance, .Pa /dev/mixer0 -is the mixer for -.Pa /dev/pcm0 -and so on. -By default, +is the mixer device for +.Pa /dev/dsp0 . +.It /dev/mixer +Alias to the default device's mixer device. .Nm -prints both the audio card's number and the mixer associated with it -in the form of -.Ar pcmN:mixer . -The -.Pa /dev/mixer -file, although it does not exist in the filesystem, points to the default -mixer device and is the file -.Nm -opens when the +opens this when the .Fl f Ar device -option has not been specified. +option is not specified. .El .Sh EXAMPLES List all available audio devices in the system: diff --git a/usr.sbin/periodic/etc/weekly/Makefile b/usr.sbin/periodic/etc/weekly/Makefile index d194a988acf0..a5483534c029 100644 --- a/usr.sbin/periodic/etc/weekly/Makefile +++ b/usr.sbin/periodic/etc/weekly/Makefile @@ -8,12 +8,11 @@ CONFS= 340.noid \ # NB: keep these sorted by MK_* knobs -.if ${MK_LOCATE} != "no" -CONFS+= 310.locate -.endif +CONFGROUPS.${MK_LOCATE}+= LOCATE +LOCATE= 310.locate -.if ${MK_MAN_UTILS} != "no" -CONFS+= 320.whatis -.endif +CONFGROUPS.${MK_MAN_UTILS}+= WHATIS +WHATISPACKAGE= mandoc +WHATIS= 320.whatis .include <bsd.prog.mk> diff --git a/usr.sbin/sysrc/sysrc.8 b/usr.sbin/sysrc/sysrc.8 index bdf3353c2cf9..cb32f72ea587 100644 --- a/usr.sbin/sysrc/sysrc.8 +++ b/usr.sbin/sysrc/sysrc.8 @@ -1,3 +1,6 @@ +.\" +.\" SPDX-License-Identifier: BSD-2-Clause +.\" .\" Copyright (c) 2011-2016 Devin Teske .\" All rights reserved. .\" @@ -408,62 +411,52 @@ and .It Pa /usr/local/etc/rc.conf.d/name/* .El .Sh EXAMPLES -Below are some simple examples of how -.Nm -can be used to query certain values from the -.Xr rc.conf 5 -collection of system configuration files: -.Pp -.Nm -sshd_enable -.Dl returns the value of $sshd_enable, usually YES or NO . +.Ss Working with rc.conf files +Ask the value of +.Cm sshd_enable , +usually YES or NO: +.Dl sysrc sshd_enable .Pp -.Nm -defaultrouter -.Dl returns IP address of default router Pq if configured . -.Pp -Working on other files, such as +Return the IP address of default router +.Pq if configured : +.Dl sysrc defaultrouter +.Ss Working with other files +Return the value of the MAILTO setting, if configured, from .Xr crontab 5 : +.Dl sysrc -f /etc/crontab MAILTO .Pp -.Nm --f /etc/crontab MAILTO -.Dl returns the value of the MAILTO setting Pq if configured . -.Pp -Appending to existing values: -.Pp -.Nm -\&cloned_interfaces+=gif0 -.Dl appends Qo gif0 Qc to $cloned_interfaces Pq see APPENDING VALUES . -.Pp -.Nm -\&cloned_interfaces-=gif0 -.Dl removes Qo gif0 Qc from $cloned_interfaces Pq see SUBTRACTING VALUES . -.Pp -In addition to the above syntax, -.Nm -also supports inline -.Xr sh 1 -PARAMETER expansion for changing the way values are reported, shown below: +Append +.Dq gif0 +to $cloned_interfaces +.Pq see Sx APPENDING VALUES : +.Dl sysrc cloned_interfaces+=gif0 .Pp -.Nm -\&'hostname%%.*' -.Dl returns $hostname up to (but not including) first `.' . +Remove +.Dq gif0 +from $cloned_interfaces +.Pq see Sx SUBTRACTING VALUES : +.Dl sysrc cloned_interfaces-=gif0 +.Ss Inline shell parameter expansion +Return $hostname up to, but not including, first +.Ql \&. : +.Dl sysrc 'hostname%%.*' .Pp -.Nm -\&'network_interfaces%%[$IFS]*' -.Dl returns first word of $network_interfaces . +Return first word of $network_interfaces: +.Dl sysrc 'network_interfaces%%[$IFS]*' .Pp -.Nm -\&'ntpdate_flags##*[$IFS]' -.Dl returns last word of $ntpdate_flags (time server address) . +Return last word of $ntpdate_flags +.Pq time server address : +.Dl sysrc 'ntpdate_flags##*[$IFS]' .Pp -.Nm -usbd_flags-"default" -.Dl returns $usbd_flags or "default" if unset or NULL . +Return $usbd_flags or +.Dq default +if unset or NULL: +.Dl sysrc usbd_flags-"default" .Pp -.Nm -cloned_interfaces+"alternate" -.Dl returns "alternate" if $cloned_interfaces is set . +Return +.Dq alternate +if $cloned_interfaces is set: +.Dl sysrc cloned_interfaces+"alternate" .Sh SEE ALSO .Xr rc.conf 5 , .Xr jail 8 , |