Re: [patch] fix sdmmc bug affecting strictly sector-addressable eMMC chips
On Fri, Aug 28, 2015 at 02:13:44AM -0400, kremlin wrote: sdmmc gives up configuring the card after 100 failed CMD1 commands. My patch starts asserting the MMC_OCR_SECTOR_MODE bit after 50 failed attempts. This fixes the bug on my BBB with Micron chip: Nice work! I wonder why you chose to be so conservative (or arbitrary) about the after 50 failed attempts condition. :) Since sector mode is supported by the host software, shouldn't we always set the MMC_OCR_SECTOR_MODE bit and check the returned OCR value to see if the card also supports sector mode addressing? Related thread: http://marc.info/?l=linux-mmcm=129419687808618 Cheers, Uwe
[patch] fix sdmmc bug affecting strictly sector-addressable eMMC chips
Hello,
This patch fixes a bug affecting certain 'BeagleBone Black' devices
which prevents the onboard eMMC chip from initializing. It has been
discussed previously but was misattributed to manufacturers bumping its
size from 2GB - 4GB:
http://marc.info/?l=openbsd-techm=141122894126274w=2
Some Beaglebones (of any revision) are manufactured with Kingston eMMC
chips while others are manufactured with Micron eMMC chips. The Micron
ones fail during the card identification stage as the sdmmc driver
currently does not assert its ability to handle sector-addressed cards.
The Kingston ones don't.
Quoting the JEDEC eMMC standard, page 44:
If there is no indication by a host to a memory that the host is
capable of handling sector type of addressing the higher than 2GB of
density of memory will change its state to Inactive (similarly to a sit-
uation in which there is no common voltage range to work with)
Watch this occur with -DSDHC_DBUG -DSDMMC_DEBUG:
ommmc1: start cmd 1 arg=0x6 data=0x0 dlen=0 flags=0x1030
proc=sdmmc1
ommmc1: wait_state 1 0 1f7(state=1f7_CLEV)
ommmc1: cmd=0x102 blksize=0 blkcount=0
ommmc1: interrupt status=1_CC
ommmc1: intr status 0x1 error 0
resp[0] 0x00ff8080
ommmc1: cmd 1 done (flags=0x1030 error=0)
sdmmc1: can't send memory OCR
ommmc1: software reset reg=0x100
sdmmc1: can't enable card
sdmmc gives up configuring the card after 100 failed CMD1 commands. My
patch starts asserting the MMC_OCR_SECTOR_MODE bit after 50 failed
attempts. This fixes the bug on my BBB with Micron chip:
ommmc1: start cmd 1 arg=0x4006 data=0x0 dlen=0 flags=0x1030
proc=sdmmc1
ommmc1: wait_state 1 0 1f7(state=1f7_CLEV)
ommmc1: cmd=0x102 blksize=0 blkcount=0
ommmc1: interrupt status=1_CC
ommmc1: intr status 0x1 error 0
resp[0] 0xc0ff8080
ommmc1: cmd 1 done (flags=0x1030 error=0)
ommmc1: start cmd 2 arg=0 data=0x0 dlen=0 flags=0x1630 proc=sdmmc1
The card goes on to be interpreted correctly in the SCSI midlayer:
scsibus1 at sdmmc1: 2 targets, initiator 0
sd1 at scsibus1 targ 1 lun 0: SD/MMC, Drive #01, SCSI2 0/direct fixed
sd1: 3744MB, 512 bytes/sector, 7667712 sectors
I have tried mounting, reading, writing to it and everything seems to
work. The sector size is already read from the CSD register considered
during accesses as to prevent misalign errors so no additional work is
needed to support sector mode.
Ian
Index: sdmmc_mem.c
===
RCS file: /cvs/src/sys/dev/sdmmc/sdmmc_mem.c,v
retrieving revision 1.21
diff -u -p -r1.21 sdmmc_mem.c
--- sdmmc_mem.c 22 Apr 2015 04:02:06 - 1.21
+++ sdmmc_mem.c 28 Aug 2015 05:32:37 -
@@ -521,6 +521,10 @@ sdmmc_mem_send_op_cond(struct sdmmc_soft
cmd.c_arg = ocr;
cmd.c_flags = SCF_CMD_BCR | SCF_RSP_R3;
+ /* Set sector mode bit after 50 failures*/
+ if(i 50) {
+ cmd.c_arg |= MMC_OCR_SECTOR_MODE;
+ }
if (ISSET(sc-sc_flags, SMF_SD_MODE)) {
cmd.c_opcode = SD_APP_OP_COND;
error = sdmmc_app_command(sc, cmd);
Index: sdmmcreg.h
===
RCS file: /cvs/src/sys/dev/sdmmc/sdmmcreg.h,v
retrieving revision 1.6
diff -u -p -r1.6 sdmmcreg.h
--- sdmmcreg.h 23 Sep 2014 12:08:13 - 1.6
+++ sdmmcreg.h 28 Aug 2015 05:32:37 -
@@ -48,6 +48,7 @@
/* OCR bits */
#define MMC_OCR_MEM_READY (131) /* memory power-up status bit */
+#define MMC_OCR_SECTOR_MODE(130) /* Provided in argument
following CMD1 */
#define MMC_OCR_3_5V_3_6V (123)
#define MMC_OCR_3_4V_3_5V (122)
#define MMC_OCR_3_3V_3_4V (121)
readelf on mips64
As Theo and I noticed yesterday:
$ readelf -a /usr/lib/libc.a
readelf: Error: /usr/lib/libc.a: invalid archive string table offset 0
Turns out that on mips64 binutils creates IRIX-style 64-bit archives
which allegedly can be larger than 4GB. The difference is apparently
in the archive map where offsets are 64-bit instead of 32-bit. Since
readelf simply skips over the map and reads the archive members
sequentially all we have to do is recognize /SYM64/ as a valid
header and skip over it.
ok?
Index: readelf.c
===
RCS file: /cvs/src/gnu/usr.bin/binutils-2.17/binutils/readelf.c,v
retrieving revision 1.6
diff -u -p -r1.6 readelf.c
--- readelf.c 25 May 2015 14:56:27 - 1.6
+++ readelf.c 28 Aug 2015 09:55:35 -
@@ -9121,7 +9121,8 @@ process_archive (char *file_name, FILE *
return 1;
}
- if (memcmp (arhdr.ar_name, / , 16) == 0)
+ if (memcmp (arhdr.ar_name, / , 16) == 0
+ || memcmp (arhdr.ar_name, /SYM64/ , 16) == 0)
{
/* This is the archive symbol table. Skip it.
FIXME: We should have an option to dump it. */
Introducing rtvalid(9)
Multiples routing entry issues, including my recent rtalloc(9) rewrite,
led me to believe that we need a simple way to check if a (cached) route
is still valid. So let me introduce rtvalid(9):
The rtvalid() function checks if the route entry rt is still valid and
can be used. Cached entries that are no longer valid should be released
by calling rtfree().
Currently rtvalid(9) prevent you to use a !RTF_UP route or a entry
attached to a stall ifa. In the short term I'd like to extend it
to include the validity of next hop in order to perform ARP or NDP
resolution for RTF_GATEWAY routes.
I did not convert all the tree yet and some conversions might be more
tricky than I though. So I'd like to commit the function first then
commit the conversion on a case-by-case basis. So feel free to ok some
parts of it.
Index: share/man/man9/Makefile
===
RCS file: /cvs/src/share/man/man9/Makefile,v
retrieving revision 1.238
diff -u -p -r1.238 Makefile
--- share/man/man9/Makefile 14 Aug 2015 05:25:29 - 1.238
+++ share/man/man9/Makefile 28 Aug 2015 10:36:02 -
@@ -332,7 +332,7 @@ MLINKS+=rt_timer_add.9 rt_timer_queue_cr
rt_timer_add.9 rt_timer_queue_change.9 \
rt_timer_add.9 rt_timer_queue_destroy.9 \
rt_timer_add.9 rt_timer_remove_all.9
-MLINKS+=rtalloc.9 rtalloc_mpath.9 rtalloc.9 rtfree.9
+MLINKS+=rtalloc.9 rtalloc_mpath.9 rtalloc.9 rtvalid.9 rtalloc.9 rtfree.9
MLINKS+=rtable_add.9 rtable_exists.9 rtable_add.9 rtable_get.9 \
rtable_add.9 rtable_l2.9 rtable_add.9 rtable_l2set.9
MLINKS+=rtlabel_id2name.9 rtlabel_name2id.9 \
Index: share/man/man9/rtalloc.9
===
RCS file: /cvs/src/share/man/man9/rtalloc.9,v
retrieving revision 1.3
diff -u -p -r1.3 rtalloc.9
--- share/man/man9/rtalloc.95 Dec 2014 16:55:53 - 1.3
+++ share/man/man9/rtalloc.928 Aug 2015 10:36:18 -
@@ -1,6 +1,6 @@
.\ $OpenBSD: rtalloc.9,v 1.3 2014/12/05 16:55:53 schwarze Exp $
.\
-.\ Copyright (c) 2014 Martin Pieuchot
+.\ Copyright (c) 2014-2015 Martin Pieuchot
.\
.\ Permission to use, copy, modify, and distribute this software for any
.\ purpose with or without fee is hereby granted, provided that the above
@@ -20,6 +20,7 @@
.Sh NAME
.Nm rtalloc ,
.Nm rtalloc_mpath ,
+.Nm rtvalid,
.Nm rtfree
.Nd routing entry interface
.Sh SYNOPSIS
@@ -30,6 +31,8 @@
.Fn rtalloc struct sockaddr *dst int flags unsigned int rtableid
.Ft struct rtentry *
.Fn rtalloc_mpath struct sockaddr *dst uint32_t *src unsigned int
rtableid
+.Ft int
+.Fn rtvalid struct rtentry *rt
.Ft void
.Fn rtfree struct rtentry *rt
.Sh DESCRIPTION
@@ -65,6 +68,14 @@ but selects a multipath routing entry co
when possible.
.Pp
The
+.Fn rtvalid
+function checks if the route entry
+.Fa rt
+is still valid and can be used.
+Cached entries that are no longer valid should be released by calling
+.Fn rtfree .
+.Pp
+The
.Fn rtfree
function releases a reference to the routing entry
.Fa rt ,
@@ -72,9 +83,13 @@ freeing it if the reference count drops
.Sh CONTEXT
.Fn rtalloc ,
.Fn rtalloc_mpath ,
+.Fn rtvalid ,
and
.Fn rtfree
can be called during autoconf, from process context, or from interrupt context.
+.Sh RETURN VALUES
+.Fn rtvalid
+returns 1 if the route entry is valid, otherwise 0.
.Sh SEE ALSO
.Xr route 4 ,
.Xr rtrequest1 9
Index: sys/net/pf.c
===
RCS file: /cvs/src/sys/net/pf.c,v
retrieving revision 1.936
diff -u -p -r1.936 pf.c
--- sys/net/pf.c19 Aug 2015 21:22:41 - 1.936
+++ sys/net/pf.c28 Aug 2015 09:40:40 -
@@ -2949,11 +2949,11 @@ pf_calc_mss(struct pf_addr *addr, sa_fam
#endif /* INET6 */
}
- if (rt rt-rt_ifp) {
+ if (rtvalid(rt)) {
mss = rt-rt_ifp-if_mtu - hlen - sizeof(struct tcphdr);
mss = max(tcp_mssdflt, mss);
- rtfree(rt);
}
+ rtfree(rt);
mss = min(mss, offer);
mss = max(mss, 64); /* sanity - at least max opt space */
return (mss);
Index: sys/net/route.c
===
RCS file: /cvs/src/sys/net/route.c,v
retrieving revision 1.225
diff -u -p -r1.225 route.c
--- sys/net/route.c 24 Aug 2015 22:11:33 - 1.225
+++ sys/net/route.c 28 Aug 2015 10:16:43 -
@@ -297,6 +297,25 @@ rtable_exists(u_int id)/* verify table
return (1);
}
+/*
+ * Returns 1 if the cached ``rt'' entry is still valid, 0 otherwise.
+ */
+int
+rtvalid(struct rtentry *rt)
+{
+ if (rt == NULL)
+ return (0);
+
+ if ((rt-rt_flags RTF_UP) == 0)
+ return (0);
+
+ /* Routes attached to stall ifas should be freed. */
+ if (rt-rt_ifa == NULL || rt-rt_ifa-ifa_ifp == NULL)
+ return (0);
+
+ return (1);
+}
+
struct rtentry *
Re: Using tame() in userland
have a much more stringent one for tsort.
Index: tsort.c
===
RCS file: /build/data/openbsd/cvs/src/usr.bin/tsort/tsort.c,v
retrieving revision 1.26
diff -u -p -r1.26 tsort.c
--- tsort.c 29 Jul 2015 10:42:37 - 1.26
+++ tsort.c 28 Aug 2015 08:03:59 -
@@ -28,6 +28,7 @@
#include sysexits.h
#include unistd.h
#include ohash.h
+#include sys/tame.h
/* The complexity of topological sorting is O(e), where e is the
* size of input. While reading input, vertices have to be identified,
@@ -798,6 +799,43 @@ find_longest_cycle(struct array *h, stru
#define plural(n) ((n) 1 ? s : )
+static void
+TAME(int flags, const char *wl[])
+{
+ if (tame(flags, wl) != 0)
+ err(1, untamed program);
+}
+
+static void
+compute_whitelist(int argc, char *argv[], char *args[])
+{
+ int c;
+ int i = 0;
+
+ while ((c = getopt(argc, argv, h:flqrvw)) != -1) {
+ switch(c) {
+ case 'h':
+ args[i++] = optarg;
+ /*FALLTHRU*/
+ case 'f':
+ case 'l':
+ case 'q':
+ case 'r':
+ case 'v':
+ case 'w':
+ break;
+ default:
+ usage();
+ }
+ }
+ argc -= optind;
+ argv += optind;
+ if (argc == 1)
+ args[i++] = argv[0];
+ args[i] = NULL;
+ optind = optreset = 1;
+}
+
int
main(int argc, char *argv[])
{
@@ -806,6 +844,10 @@ main(int argc, char *argv[])
warn_flag, hints_flag, verbose_flag;
unsigned intorder;
+ char *args[3];
+
+ compute_whitelist(argc, argv, args);
+ TAME(TAME_STDIO|TAME_RPATH, args);
order = 0;
reverse_flag = quiet_flag = long_flag =
@@ -863,6 +905,7 @@ main(int argc, char *argv[])
FILE *f;
f = fopen(argv[0], r);
+ TAME(TAME_STDIO, NULL);
if (f == NULL)
err(EX_NOINPUT, Can't open file %s, argv[1]);
order = read_pairs(f, pairs, reverse_flag, argv[1], order,
@@ -871,6 +914,7 @@ main(int argc, char *argv[])
break;
}
case 0:
+ TAME(TAME_STDIO, NULL);
order = read_pairs(stdin, pairs, reverse_flag, stdin,
order, hints_flag == 2);
break;
@@ -878,6 +922,7 @@ main(int argc, char *argv[])
usage();
}
+ TAME(TAME_RW, NULL);
{
struct arrayaux;/* Unrefed nodes/cycle reporting. */
struct arrayremaining;
Re: [patch] fix sdmmc bug affecting strictly sector-addressable eMMC chips
Nice work! I wonder why you chose to be so conservative (or arbitrary) about the after 50 failed attempts condition. :) My reasoning is that after 50 failed CMD1s, either two things are the case: The card will only initialize with the sector bit set, or it is a high capacity device taking a long time to become ready. In either of these cases, the card will be block/sector addressed anyway. Since sector mode is supported by the host software, shouldn't we always set the MMC_OCR_SECTOR_MODE bit and check the returned OCR value to see if the card also supports sector mode addressing? Related thread: http://marc.info/?l=linux-mmcm=129419687808618 That is a good point. I'll fix that.
Re: diff - more satosin/sintosa
3rd update.
Thanks to Kent for testing.
===
RCS file: /cvs/src/sys/net/if_gif.c,v
retrieving revision 1.77
diff -u -p -r1.77 if_gif.c
--- sys/net/if_gif.c17 Jul 2015 18:05:59 - 1.77
+++ sys/net/if_gif.c28 Aug 2015 13:06:21 -
@@ -339,17 +339,17 @@ gif_ioctl(struct ifnet *ifp, u_long cmd,
case SIOCSLIFPHYADDR:
switch (cmd) {
case SIOCSIFPHYADDR:
- src = (struct sockaddr *)
- (((struct in_aliasreq *)data)-ifra_addr);
- dst = (struct sockaddr *)
- (((struct in_aliasreq *)data)-ifra_dstaddr);
+ src = sintosa(
+ (((struct in_aliasreq *)data)-ifra_addr));
+ dst = sintosa(
+ (((struct in_aliasreq *)data)-ifra_dstaddr));
break;
#ifdef INET6
case SIOCSIFPHYADDR_IN6:
- src = (struct sockaddr *)
- (((struct in6_aliasreq *)data)-ifra_addr);
- dst = (struct sockaddr *)
- (((struct in6_aliasreq *)data)-ifra_dstaddr);
+ src = sin6tosa(
+ (((struct in6_aliasreq *)data)-ifra_addr));
+ dst = sin6tosa(
+ (((struct in6_aliasreq *)data)-ifra_dstaddr));
break;
#endif
case SIOCSLIFPHYADDR:
@@ -432,10 +432,10 @@ gif_ioctl(struct ifnet *ifp, u_long cmd,
/* can't configure multiple multi-dest interfaces */
#define multidest(x) \
- (((struct sockaddr_in *)(x))-sin_addr.s_addr == INADDR_ANY)
+ ((satosin((x)))-sin_addr.s_addr == INADDR_ANY)
#ifdef INET6
#define multidest6(x) \
- (IN6_IS_ADDR_UNSPECIFIED(((struct sockaddr_in6 *)(x))-sin6_addr))
+ (IN6_IS_ADDR_UNSPECIFIED((satosin6((x)))-sin6_addr))
#endif
if (dst-sa_family == AF_INET
multidest(dst) multidest(sc2-gif_pdst)) {
@@ -501,8 +501,8 @@ gif_ioctl(struct ifnet *ifp, u_long cmd,
break;
#ifdef INET6
case SIOCGIFPSRCADDR_IN6:
- dst = (struct sockaddr *)
- (((struct in6_ifreq *)data)-ifr_addr);
+ dst = sin6tosa(
+ (((struct in6_ifreq *)data)-ifr_addr));
size = sizeof(((struct in6_ifreq *)data)-ifr_addr);
break;
#endif /* INET6 */
@@ -531,8 +531,7 @@ gif_ioctl(struct ifnet *ifp, u_long cmd,
break;
#ifdef INET6
case SIOCGIFPDSTADDR_IN6:
- dst = (struct sockaddr *)
- (((struct in6_ifreq *)data)-ifr_addr);
+ dst = sin6tosastruct in6_ifreq *)data)-ifr_addr));
size = sizeof(((struct in6_ifreq *)data)-ifr_addr);
break;
#endif /* INET6 */
Index: sys/net/if_tun.c
===
RCS file: /cvs/src/sys/net/if_tun.c,v
retrieving revision 1.151
diff -u -p -r1.151 if_tun.c
--- sys/net/if_tun.c20 Jul 2015 22:54:29 - 1.151
+++ sys/net/if_tun.c28 Aug 2015 13:06:21 -
@@ -446,12 +446,12 @@ tuninit(struct tun_softc *tp)
if (ifa-ifa_addr-sa_family == AF_INET6) {
struct sockaddr_in6 *sin;
- sin = (struct sockaddr_in6 *)ifa-ifa_addr;
+ sin = satosin6(ifa-ifa_addr);
if (!IN6_IS_ADDR_UNSPECIFIED(sin-sin6_addr))
tp-tun_flags |= TUN_IASET;
if (ifp-if_flags IFF_POINTOPOINT) {
- sin = (struct sockaddr_in6 *)ifa-ifa_dstaddr;
+ sin = satosin6(ifa-ifa_dstaddr);
if (sin
!IN6_IS_ADDR_UNSPECIFIED(sin-sin6_addr))
tp-tun_flags |= TUN_DSTADDR;
Index: sys/net/pipex.c
===
RCS file: /cvs/src/sys/net/pipex.c,v
retrieving revision 1.73
diff -u -p -r1.73 pipex.c
--- sys/net/pipex.c 24 Aug 2015 14:00:28 - 1.73
+++ sys/net/pipex.c 28 Aug 2015 13:06:22 -
@@ -2959,10 +2959,9 @@ pipex_sockaddr_hash_key(struct sockaddr
{
switch (sa-sa_family) {
case AF_INET:
- return ntohl(((struct sockaddr_in *)sa)-sin_addr.s_addr);
+ return ntohl(satosin(sa)-sin_addr.s_addr);
case AF_INET6:
- return ntohl(((struct sockaddr_in6 *)sa)-sin6_addr
- .s6_addr32[3]);
+ return
squid 3.5/trunk and SSLv3 removal
Any suggestions as to what might be needed here, can the SSLv3_method
just be replaced with SSLv23_method?
http://bazaar.launchpad.net/~squid/squid/3.5/view/head:/src/ssl/bio.cc#L1009
| ciphers += 2;
| if (ciphersLen) {
| const SSL_METHOD *method = SSLv3_method();
| const int cs = method-put_cipher_by_char(NULL, NULL);
| assert(cs 0);
| for (size_t i = 0; i ciphersLen; i += cs) {
| const SSL_CIPHER *c = method-get_cipher_by_char((ciphers + i));
| if (c != NULL) {
| if (!clientRequestedCiphers.empty())
| clientRequestedCiphers.append(:);
| clientRequestedCiphers.append(c-name);
| } else
| unknownCiphers = true;
| }
| }
| debugs(83, 7, Ciphers requested by client:
clientRequestedCiphers);
Re: readelf on mips64
On Fri, Aug 28, 2015 at 12:00:41PM +0200, Mark Kettenis wrote:
As Theo and I noticed yesterday:
$ readelf -a /usr/lib/libc.a
readelf: Error: /usr/lib/libc.a: invalid archive string table offset 0
Turns out that on mips64 binutils creates IRIX-style 64-bit archives
which allegedly can be larger than 4GB. The difference is apparently
in the archive map where offsets are 64-bit instead of 32-bit. Since
readelf simply skips over the map and reads the archive members
sequentially all we have to do is recognize /SYM64/ as a valid
header and skip over it.
ok?
ok visa@
Index: readelf.c
===
RCS file: /cvs/src/gnu/usr.bin/binutils-2.17/binutils/readelf.c,v
retrieving revision 1.6
diff -u -p -r1.6 readelf.c
--- readelf.c 25 May 2015 14:56:27 - 1.6
+++ readelf.c 28 Aug 2015 09:55:35 -
@@ -9121,7 +9121,8 @@ process_archive (char *file_name, FILE *
return 1;
}
- if (memcmp (arhdr.ar_name, / , 16) == 0)
+ if (memcmp (arhdr.ar_name, / , 16) == 0
+ || memcmp (arhdr.ar_name, /SYM64/ , 16) == 0)
{
/* This is the archive symbol table. Skip it.
FIXME: We should have an option to dump it. */
Should PIE be mentioned in security.html?
Not sure about its location in the list. Index: security.html === RCS file: /cvs/www/security.html,v retrieving revision 1.422 diff -u -p -r1.422 security.html --- security.html 2 Jul 2015 05:49:04 - 1.422 +++ security.html 29 Aug 2015 03:59:35 - @@ -158,6 +158,7 @@ written somewhere, but perhaps not taken liChroot jailing liNew uids liProPolice + liPosition-independent executables (PIE) li... and others /ul p
RFC 5071 DHCP options
I use dhcpd(8) to boot some boxes with PXELINUX. The numbered options work
but dhcpd.conf(5) is easier to maintain with names. These options are
defined in RFC 5071.
Nathanael
Index: sbin/dhclient/tables.c
===
RCS file: /cvs/src/sbin/dhclient/tables.c,v
retrieving revision 1.18
diff -u -p -u -p -r1.18 tables.c
--- sbin/dhclient/tables.c 21 Jan 2014 03:07:50 - 1.18
+++ sbin/dhclient/tables.c 27 Aug 2015 09:13:26 -
@@ -269,9 +269,9 @@ const struct option dhcp_options[256] =
/* 206 */ { option-206, X },
/* 207 */ { option-207, X },
/* 208 */ { option-208, X },
- /* 209 */ { option-209, X },
- /* 210 */ { option-210, X },
- /* 211 */ { option-211, X },
+ /* 209 */ { config-file, t },
+ /* 210 */ { path-prefix, t },
+ /* 211 */ { reboot-time, L },
/* 212 */ { option-212, X },
/* 213 */ { option-213, X },
/* 214 */ { option-214, X },
Index: usr.sbin/dhcpd/dhcp-options.5
===
RCS file: /cvs/src/usr.sbin/dhcpd/dhcp-options.5,v
retrieving revision 1.21
diff -u -p -u -p -r1.21 dhcp-options.5
--- usr.sbin/dhcpd/dhcp-options.5 2 Jun 2015 16:02:45 - 1.21
+++ usr.sbin/dhcpd/dhcp-options.5 27 Aug 2015 09:13:29 -
@@ -176,6 +176,10 @@ This option does the same as
.Ic classless-static-routes ,
but uses option code 249 instead of 121,
since Windows XP and Windows Server 2003 ignore option 121.
+.It Ic option config-file Ar string ;
+This option specifies the configuration file for the PXELINUX boot loader.
+This is typically used in conjuction with the path-prefix option.
+See RFC 5071 for details.
.It Ic option cookie-servers Ar ip-address Oo , Ar ip-address ... Oc ;
The
.Ic cookie-servers
@@ -453,6 +457,11 @@ Path MTU Discovery as defined in RFC 119
The table is formatted as a list of 16-bit unsigned integers,
ordered from smallest to largest.
The minimum MTU value cannot be smaller than 68.
+.It Ic option path-prefix Ar string ;
+This option specifies the path prefix for a PXELINUX boot loader
+configuration file.
+This is typically used in conjuction with the config-file option.
+See RFC 5071 for details.
.It Ic option perform-mask-discovery Ar flag ;
This option specifies whether or not the client should perform subnet mask
discovery using ICMP.
@@ -472,6 +481,10 @@ The
.Ic pop-server
option specifies a list of POP3 servers available to the client.
Servers should be listed in order of preference.
+.It Ic option reboot-time Ar uint32 ;
+This option determines how long the PXELINUX boot loader waits to connect
+to a TFTP server before rebooting.
+See RFC 5071 for details.
.It Ic option relay-agent-information Ar string ;
This is a container option for specific agent-supplied sub-options.
See RFC 3046 for details.
Index: usr.sbin/dhcpd/tables.c
===
RCS file: /cvs/src/usr.sbin/dhcpd/tables.c,v
retrieving revision 1.11
diff -u -p -u -p -r1.11 tables.c
--- usr.sbin/dhcpd/tables.c 27 Jun 2015 14:29:39 - 1.11
+++ usr.sbin/dhcpd/tables.c 27 Aug 2015 09:13:29 -
@@ -271,9 +271,9 @@ struct option dhcp_options[256] = {
{ option-206, X,dhcp_universe, 206 },
{ option-207, X,dhcp_universe, 207 },
{ option-208, X,dhcp_universe, 208 },
- { option-209, X,dhcp_universe, 209 },
- { option-210, X,dhcp_universe, 210 },
- { option-211, X,dhcp_universe, 211 },
+ { config-file, t, dhcp_universe, 209 },
+ { path-prefix, t, dhcp_universe, 210 },
+ { reboot-time, L, dhcp_universe, 211 },
{ option-212, X,dhcp_universe, 212 },
{ option-213, X,dhcp_universe, 213 },
{ option-214, X,dhcp_universe, 214 },
syslogd host matches ip
Hi,
The +host feature allows to select log messages from a specific
host. Normally syslogd does a reverse lookup on the sender's
address. If that fails or if syslogd has been started with the -n
option, the +host is matched with the IP. Unfortunatelty IP addresses
consist of characters that are not allowed in syslog.conf. So take
the list of valid characters from FreeBSD.
ok?
bluhm
Index: usr.sbin/syslogd/syslogd.c
===
RCS file: /data/mirror/openbsd/cvs/src/usr.sbin/syslogd/syslogd.c,v
retrieving revision 1.179
diff -u -p -r1.179 syslogd.c
--- usr.sbin/syslogd/syslogd.c 27 Aug 2015 17:53:35 - 1.179
+++ usr.sbin/syslogd/syslogd.c 28 Aug 2015 16:07:43 -
@@ -2053,12 +2053,13 @@ init(void)
continue;
}
for (i = 0; i NAME_MAX; i++) {
- if (!isalnum((unsigned char)p[i])
- p[i] != '-' p[i] != '+')
+ if (!isalnum((unsigned char)*p)
+ *p != '-' *p != '+' *p != '.'
+ *p != ',' *p != ':' *p != '%')
break;
- hostblock[i] = p[i];
+ hostblock[i] = *p++;
}
- hostblock[i] = 0;
+ hostblock[i] = '\0';
continue;
}
LibreSSL 2.2.3 Prerelease available for testing
We have put together a prerelease of LibreSSL 2.2.3. This release primarily targets a bug that causes connections with some SSL clients to fail during the initial handshake. Preliminary release notes and a tarball are available at the following link: https://github.com/libressl-portable/portable/releases/tag/v2.2.3
synaptics touchpads: w mode and resolution
Some weeks ago a change was made in pms to support Synaptics touchpads
that don't provide W mode. I assume that only fairly old models are
concerned, and that the variant in the patch below is more accurate. It
only fakes W values where necessary.
The patch contains a second change, a check whether the resolution query
was successful. According to the Synaptics PS/2 Interfacing Guide, bit
7 of the second response byte will be set in case of success. It seems
that if it isn't set, the other bytes are garbage or have an unknown
meaning.
Unfortunately I only have that old model - with firmware 4.1 - for
testing. Could someone confirm that newer models aren't affected by
these changes?
Index: dev/pckbc/pms.c
===
RCS file: /cvs/src/sys/dev/pckbc/pms.c,v
retrieving revision 1.65
diff -u -p -r1.65 pms.c
--- dev/pckbc/pms.c 23 Aug 2015 04:45:23 - 1.65
+++ dev/pckbc/pms.c 28 Aug 2015 18:04:45 -
@@ -949,8 +949,10 @@ synaptics_get_hwinfo(struct pms_softc *s
return (-1);
}
- syn-res_x = SYNAPTICS_RESOLUTION_X(syn-resolution);
- syn-res_y = SYNAPTICS_RESOLUTION_Y(syn-resolution);
+ if (syn-resolution 0x8000) {
+ syn-res_x = SYNAPTICS_RESOLUTION_X(syn-resolution);
+ syn-res_y = SYNAPTICS_RESOLUTION_Y(syn-resolution);
+ }
syn-min_x = SYNAPTICS_XMIN_BEZEL;
syn-min_y = SYNAPTICS_YMIN_BEZEL;
syn-max_x = (syn-dimension) ?
@@ -1163,30 +1165,21 @@ pms_proc_synaptics(struct pms_softc *sc)
w = ((sc-packet[0] 0x30) 2) | ((sc-packet[0] 0x04) 1) |
((sc-packet[3] 0x04) 2);
+ z = sc-packet[2];
- /*
-* Conform to the encoding understood by
-* /usr/xenocara/driver/xf86-input-synaptics/src/wsconscomm.c
-*/
- switch (w) {
- case 0:
- /* fingerwidth 5, numfingers 2 */
- break;
- case 1:
- /* fingerwidth 5, numfingers 3 */
- break;
- case 5:
- /* fingerwidth 5, numfingers 1 */
- break;
- case 4:
- case 8:
- /* fingerwidth 4, numfingers 1 */
- w = 4;
- break;
- default:
- break;
+ if ((syn-capabilities SYNAPTICS_CAP_EXTENDED) == 0) {
+ /*
+* Emulate W mode for models that don't provide it. Bit 3
+* of the w-input signals a touch (finger), Bit 2 and
+* the gesture bits 1-0 can be ignored.
+*/
+ if (w 8)
+ w = 4;
+ else
+ z = w = 0;
}
+
if ((syn-capabilities SYNAPTICS_CAP_PASSTHROUGH) w == 3) {
synaptics_sec_proc(sc);
return;
@@ -1203,7 +1196,6 @@ pms_proc_synaptics(struct pms_softc *sc)
sc-packet[4];
y = ((sc-packet[3] 0x20) 7) | ((sc-packet[1] 0xf0) 4) |
sc-packet[5];
- z = sc-packet[2];
buttons = ((sc-packet[0] sc-packet[3]) 0x01) ?
WSMOUSE_BUTTON(1) : 0;
Re: syslogd host matches ip
On Fri, Aug 28, 2015 at 06:11:02PM +0200, Alexander Bluhm wrote:
The +host feature allows to select log messages from a specific
host. Normally syslogd does a reverse lookup on the sender's
address. If that fails or if syslogd has been started with the -n
option, the +host is matched with the IP. Unfortunatelty IP addresses
consist of characters that are not allowed in syslog.conf. So take
the list of valid characters from FreeBSD.
Add _ to the list to help people with broken hostnames.
Suggested by sven falempin.
ok?
bluhm
Index: usr.sbin/syslogd/syslogd.c
===
RCS file: /data/mirror/openbsd/cvs/src/usr.sbin/syslogd/syslogd.c,v
retrieving revision 1.179
diff -u -p -r1.179 syslogd.c
--- usr.sbin/syslogd/syslogd.c 27 Aug 2015 17:53:35 - 1.179
+++ usr.sbin/syslogd/syslogd.c 28 Aug 2015 17:48:16 -
@@ -2053,12 +2053,14 @@ init(void)
continue;
}
for (i = 0; i NAME_MAX; i++) {
- if (!isalnum((unsigned char)p[i])
- p[i] != '-' p[i] != '+')
+ if (!isalnum((unsigned char)*p)
+ *p != '-' *p != '+' *p != '.'
+ *p != ',' *p != ':' *p != '%'
+ *p != '_')
break;
- hostblock[i] = p[i];
+ hostblock[i] = *p++;
}
- hostblock[i] = 0;
+ hostblock[i] = '\0';
continue;
}
