svn commit: r228304 - head/sys/dev/usb/template
Author: hselasky
Date: Tue Dec 6 08:08:52 2011
New Revision: 228304
URL: http://svn.freebsd.org/changeset/base/228304
Log:
Correct some bInterval USB template descriptor values.
MFC after:3 days
Modified:
head/sys/dev/usb/template/usb_template.c
head/sys/dev/usb/template/usb_template_kbd.c
head/sys/dev/usb/template/usb_template_modem.c
head/sys/dev/usb/template/usb_template_mouse.c
Modified: head/sys/dev/usb/template/usb_template.c
==
--- head/sys/dev/usb/template/usb_template.cTue Dec 6 07:55:25 2011
(r228303)
+++ head/sys/dev/usb/template/usb_template.cTue Dec 6 08:08:52 2011
(r228304)
@@ -260,7 +260,7 @@ usb_make_endpoint_desc(struct usb_temp_s
ed->bInterval = 1; /* 1 ms */
break;
default:
- ed->bInterval = 8; /* 8*125 us */
+ ed->bInterval = 4; /* 1 ms */
break;
}
break;
Modified: head/sys/dev/usb/template/usb_template_kbd.c
==
--- head/sys/dev/usb/template/usb_template_kbd.cTue Dec 6 07:55:25
2011(r228303)
+++ head/sys/dev/usb/template/usb_template_kbd.cTue Dec 6 08:08:52
2011(r228304)
@@ -89,9 +89,9 @@ static const struct usb_temp_packet_size
};
static const struct usb_temp_interval keyboard_intr_interval = {
- .bInterval[USB_SPEED_LOW] = 2, /* ms */
- .bInterval[USB_SPEED_FULL] = 2,
- .bInterval[USB_SPEED_HIGH] = 2 * 8,
+ .bInterval[USB_SPEED_LOW] = 2, /* 2 ms */
+ .bInterval[USB_SPEED_FULL] = 2, /* 2 ms */
+ .bInterval[USB_SPEED_HIGH] = 5, /* 2 ms */
};
/* The following HID descriptor was dumped from a HP keyboard. */
Modified: head/sys/dev/usb/template/usb_template_modem.c
==
--- head/sys/dev/usb/template/usb_template_modem.c Tue Dec 6 07:55:25
2011(r228303)
+++ head/sys/dev/usb/template/usb_template_modem.c Tue Dec 6 08:08:52
2011(r228304)
@@ -98,9 +98,9 @@ static const struct usb_temp_packet_size
};
static const struct usb_temp_interval modem_intr_interval = {
- .bInterval[USB_SPEED_LOW] = 10,
- .bInterval[USB_SPEED_FULL] = 10,
- .bInterval[USB_SPEED_HIGH] = 10 * 8,
+ .bInterval[USB_SPEED_LOW] = 8, /* 8ms */
+ .bInterval[USB_SPEED_FULL] = 8, /* 8ms */
+ .bInterval[USB_SPEED_HIGH] = 7, /* 8ms */
};
static const struct usb_temp_endpoint_desc modem_ep_0 = {
Modified: head/sys/dev/usb/template/usb_template_mouse.c
==
--- head/sys/dev/usb/template/usb_template_mouse.c Tue Dec 6 07:55:25
2011(r228303)
+++ head/sys/dev/usb/template/usb_template_mouse.c Tue Dec 6 08:08:52
2011(r228304)
@@ -101,9 +101,9 @@ static const struct usb_temp_packet_size
};
static const struct usb_temp_interval mouse_intr_interval = {
- .bInterval[USB_SPEED_LOW] = 2,
- .bInterval[USB_SPEED_FULL] = 2,
- .bInterval[USB_SPEED_HIGH] = 2 * 8,
+ .bInterval[USB_SPEED_LOW] = 2, /* 2ms */
+ .bInterval[USB_SPEED_FULL] = 2, /* 2ms */
+ .bInterval[USB_SPEED_HIGH] = 5, /* 2ms */
};
static const struct usb_temp_endpoint_desc mouse_ep_0 = {
___
svn-src-all@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"
svn commit: r228305 - head/share/man/man4
Author: mav Date: Tue Dec 6 09:12:11 2011 New Revision: 228305 URL: http://svn.freebsd.org/changeset/base/228305 Log: Mention that Cold Presence Detection feature used for hot-plug detection when interface power management is enabled. Modified: head/share/man/man4/ahci.4 Modified: head/share/man/man4/ahci.4 == --- head/share/man/man4/ahci.4 Tue Dec 6 08:08:52 2011(r228304) +++ head/share/man/man4/ahci.4 Tue Dec 6 09:12:11 2011(r228305) @@ -24,7 +24,7 @@ .\" .\" $FreeBSD$ .\" -.Dd December 2, 2011 +.Dd December 6, 2011 .Dt AHCI 4 .Os .Sh NAME @@ -90,9 +90,9 @@ Some controllers, such as ICH8, do not i Because of artificial entering latency, performance degradation in modes 4 and 5 is much smaller then in modes 2 and 3. .Pp -Note that interface Power Management is not compatible with -device presence detection. -A manual bus reset is needed on device hot-plug. +Note that interface Power Management complicates device presence detection. +A manual bus reset/rescan may be needed after device hot-plug, unless hardware +implements Cold Presence Detection. .It Va hint.ahcich. Ns Ar X Ns Va .sata_rev setting to nonzero value limits maximum SATA revision (speed). Values 1, 2 and 3 are respectively 1.5, 3 and 6Gbps. ___ svn-src-all@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/svn-src-all To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"
Re: svn commit: r228158 - in head: . share/mk sys/conf
On Mon, 5 Dec 2011 21:43:59 +0100
Hans Petter Selasky wrote:
> Hi,
>
> I see regressions when building kernel modules from /usr/ports:
>
> On Monday 05 December 2011 14:39:56 Robert Huff wrote:
> > Hello:
> > When trying to update I get:
> >
> > ===> Building for cuse4bsd-kmod-0.1.21_2
> > make -f
> > /data/port-work/usr/ports/multimedia/cuse4bsd-kmod/work/cuse4bsd-kmod-0.1.
> > 21/Makefile.lib HAVE_DEBUG=YES all Warning: Object directory not changed
> > from original
> > /data/port-work/usr/ports/multimedia/cuse4bsd-kmod/work/cuse4bsd-kmod-0.1.
> > 21 make -f
> > /data/port-work/usr/ports/multimedia/cuse4bsd-kmod/work/cuse4bsd-kmod-0.1.
> > 21/Makefile.kmod all "/sys/conf/kmod.mk", line 204: Malformed conditional
> > (${MK_CTF} != "no") "/sys/conf/kmod.mk", line 206: if-less endif
> > make: fatal errors encountered -- cannot continue
> > *** Error code 1
>
I encountered this problem yesterday. My quick hack was to put
.if defined(MK_CTF) .endif around the offending lines, similar to
EXPORT_SYMS.
--
Gary Jennejohn
___
svn-src-all@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"
svn commit: r228306 - head/sys/kern
Author: kib
Date: Tue Dec 6 11:24:03 2011
New Revision: 228306
URL: http://svn.freebsd.org/changeset/base/228306
Log:
Most users of pipe(2) do not call fstat(2) on the returned pipe descriptors.
Optimize for the case, by lazily allocating the pipe inode number at the
fstat(2) time. If alloc_unr(9) returns failure, do not fail fstat(2), since
uses of inode numbers are even rare then fstat(2), but provide zero inode
forever. Note that alloc_unr() failure is unlikely due to total number
of pipes in the system limited by the number of file descriptors.
Based on the submission by: gianni
MFC after:2 weeks
Modified:
head/sys/kern/sys_pipe.c
Modified: head/sys/kern/sys_pipe.c
==
--- head/sys/kern/sys_pipe.cTue Dec 6 09:12:11 2011(r228305)
+++ head/sys/kern/sys_pipe.cTue Dec 6 11:24:03 2011(r228306)
@@ -569,12 +569,7 @@ pipe_create(pipe, backing)
/* If we're not backing this pipe, no need to do anything. */
error = 0;
}
- if (error == 0) {
- pipe->pipe_ino = alloc_unr(pipeino_unr);
- if (pipe->pipe_ino == -1)
- /* pipeclose will clear allocated kva */
- error = ENOMEM;
- }
+ pipe->pipe_ino = -1;
return (error);
}
@@ -1398,16 +1393,40 @@ pipe_stat(fp, ub, active_cred, td)
struct ucred *active_cred;
struct thread *td;
{
- struct pipe *pipe = fp->f_data;
+ struct pipe *pipe;
+ int new_unr;
#ifdef MAC
int error;
+#endif
+ pipe = fp->f_data;
PIPE_LOCK(pipe);
+#ifdef MAC
error = mac_pipe_check_stat(active_cred, pipe->pipe_pair);
- PIPE_UNLOCK(pipe);
- if (error)
+ if (error) {
+ PIPE_UNLOCK(pipe);
return (error);
+ }
#endif
+ /*
+* Lazily allocate an inode number for the pipe. Most pipe
+* users do not call fstat(2) on the pipe, which means that
+* postponing the inode allocation until it is must be
+* returned to userland is useful. If alloc_unr failed,
+* assign st_ino zero instead of returning an error.
+* Special pipe_ino values:
+* -1 - not yet initialized;
+* 0 - alloc_unr failed, return 0 as st_ino forever.
+*/
+ if (pipe->pipe_ino == (ino_t)-1) {
+ new_unr = alloc_unr(pipeino_unr);
+ if (new_unr != -1)
+ pipe->pipe_ino = new_unr;
+ else
+ pipe->pipe_ino = 0;
+ }
+ PIPE_UNLOCK(pipe);
+
bzero(ub, sizeof(*ub));
ub->st_mode = S_IFIFO;
ub->st_blksize = PAGE_SIZE;
___
svn-src-all@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"
svn commit: r228307 - in head: secure/lib/libcrypto share/mk
Author: kib
Date: Tue Dec 6 11:28:17 2011
New Revision: 228307
URL: http://svn.freebsd.org/changeset/base/228307
Log:
Force linker error when created shared library contains a relocation
against text. Provide the override switch to turn off the strict
behaviour. Apparently, openssl libcrypto needs it due to assembler
code not being PIC.
Discussed with: bf
MFC after:2 weeks
Modified:
head/secure/lib/libcrypto/Makefile
head/share/mk/bsd.lib.mk
Modified: head/secure/lib/libcrypto/Makefile
==
--- head/secure/lib/libcrypto/Makefile Tue Dec 6 11:24:03 2011
(r228306)
+++ head/secure/lib/libcrypto/Makefile Tue Dec 6 11:28:17 2011
(r228307)
@@ -7,6 +7,7 @@ SUBDIR= engines
LIB= crypto
SHLIB_MAJOR= 6
+ALLOW_SHARED_TEXTREL=
NO_LINT=
Modified: head/share/mk/bsd.lib.mk
==
--- head/share/mk/bsd.lib.mkTue Dec 6 11:24:03 2011(r228306)
+++ head/share/mk/bsd.lib.mkTue Dec 6 11:28:17 2011(r228307)
@@ -167,6 +167,11 @@ SOBJS+=${OBJS:.o=.So}
.if defined(SHLIB_NAME)
_LIBS+=${SHLIB_NAME}
+SOLINKOPTS=-shared -Wl,-x
+.if !defined(ALLOW_SHARED_TEXTREL)
+SOLINKOPTS+= -Wl,--fatal-warnings -Wl,--warn-shared-textrel
+.endif
+
.if target(beforelinking)
${SHLIB_NAME}: ${SOBJS} beforelinking
.else
@@ -178,11 +183,11 @@ ${SHLIB_NAME}: ${SOBJS}
@ln -fs ${.TARGET} ${SHLIB_LINK}
.endif
.if !defined(NM)
- @${CC} ${LDFLAGS} ${SSP_CFLAGS} -shared -Wl,-x \
+ @${CC} ${LDFLAGS} ${SSP_CFLAGS} ${SOLINKOPTS} \
-o ${.TARGET} -Wl,-soname,${SONAME} \
`lorder ${SOBJS} | tsort -q` ${LDADD}
.else
- @${CC} ${LDFLAGS} ${SSP_CFLAGS} -shared -Wl,-x \
+ @${CC} ${LDFLAGS} ${SSP_CFLAGS} ${SOLINKOPTS} \
-o ${.TARGET} -Wl,-soname,${SONAME} \
`NM='${NM}' lorder ${SOBJS} | tsort -q` ${LDADD}
.endif
___
svn-src-all@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"
svn commit: r228308 - head/kerberos5
Author: ume
Date: Tue Dec 6 12:03:01 2011
New Revision: 228308
URL: http://svn.freebsd.org/changeset/base/228308
Log:
Don't support OpenLDAP during lib32 build.
Modified:
head/kerberos5/Makefile.inc
Modified: head/kerberos5/Makefile.inc
==
--- head/kerberos5/Makefile.inc Tue Dec 6 11:28:17 2011(r228307)
+++ head/kerberos5/Makefile.inc Tue Dec 6 12:03:01 2011(r228308)
@@ -6,7 +6,7 @@ KRB5DIR=${.CURDIR}/../../../crypto/heim
CFLAGS+= -DHAVE_CONFIG_H -I${.CURDIR}/../../include
-.if defined(WITH_OPENLDAP)
+.if defined(WITH_OPENLDAP) && !defined(COMPAT_32BIT)
OPENLDAPBASE?= /usr/local
LDAPLDADD= -lldap -llber
LDAPDPADD= ${LDAPLDADD:C;^-l(.*)$;${OPENLDAPBASE}/lib/lib\1.a;}
___
svn-src-all@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"
Re: svn commit: r228158 - in head: . share/mk sys/conf
On (05/12/2011 21:43), Hans Petter Selasky wrote:
> Hi,
>
> I see regressions when building kernel modules from /usr/ports:
make buildworld && make installworld should fix it
>
> On Monday 05 December 2011 14:39:56 Robert Huff wrote:
> > Hello:
> > When trying to update I get:
> >
> > ===> Building for cuse4bsd-kmod-0.1.21_2
> > make -f
> > /data/port-work/usr/ports/multimedia/cuse4bsd-kmod/work/cuse4bsd-kmod-0.1.
> > 21/Makefile.lib HAVE_DEBUG=YES all Warning: Object directory not changed
> > from original
> > /data/port-work/usr/ports/multimedia/cuse4bsd-kmod/work/cuse4bsd-kmod-0.1.
> > 21 make -f
> > /data/port-work/usr/ports/multimedia/cuse4bsd-kmod/work/cuse4bsd-kmod-0.1.
> > 21/Makefile.kmod all "/sys/conf/kmod.mk", line 204: Malformed conditional
> > (${MK_CTF} != "no") "/sys/conf/kmod.mk", line 206: if-less endif
> > make: fatal errors encountered -- cannot continue
> > *** Error code 1
>
>
> --HPS
___
svn-src-all@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"
Re: svn commit: r228207 - head/sys/kern
On Saturday, December 03, 2011 5:51:42 am Ivan Klymenko wrote: > В Fri, 2 Dec 2011 19:59:46 + (UTC) > John Baldwin пишет: > > > Author: jhb > > Date: Fri Dec 2 19:59:46 2011 > > New Revision: 228207 > > URL: http://svn.freebsd.org/changeset/base/228207 > > > > Log: > > When changing the user priority of a thread, change the real > > priority in addition to the user priority for threads whose current > > real priority is equal to the previous user priority or if the new > > priority is a real-time priority. This allows priority changes of > > other threads to have an immediate effect. > > > Thank you! > It's a little corrected the situation with interactivity with using ULE > scheduler... > But quite a bit... :) Hmm, I would not have expected this to make much of a difference unless you are explicitly running the 'rtprio' command. -- John Baldwin ___ svn-src-all@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/svn-src-all To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"
Re: svn commit: r227549 - stable/7/sys/nfsclient
On Thursday, December 01, 2011 5:23:32 pm Andriy Gapon wrote: > on 01/12/2011 23:42 John Baldwin said the following: > > On Thursday, December 01, 2011 10:59:54 am Mark Saad wrote: > >> On Wed, Nov 30, 2011 at 6:10 PM, Doug Barton wrote: > >>> On 11/30/2011 14:52, Rick Macklem wrote: > I'm relatively new to FreeBSD, but I've never heard of an MFC to a > Release branch. (As far as I know, if it ever happens, it's for a > serious errata or security issue.) > >>> > >>> You are correct sir. :) > >>> > >> So how would I ever see this in an official release as an errata fix ? > > > > At this point you likely won't. The vast majority of patches to releases > > are > > only for security advisories. Very few errata are merged back to releases > > currently. It may be that we are too hesitant to do so, but the manpower > > overhead of actually getting the changes merged back and getting appropriate > > folks to sign off on everything is extremely high (and the exact process for > > getting an EN approved is not very clear or really even documented to my > > knowledge). > > > > Just to clarify - the fix will be in the future (official) releases from 8, 9 > and so on branches. Yes, but without EN's there is no currently supported release a user can use to get this bug fix. I still think we are too hesitant to merge EN's. -- John Baldwin ___ svn-src-all@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/svn-src-all To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"
svn commit: r228311 - head/sys/conf
Author: fjoe
Date: Tue Dec 6 18:01:09 2011
New Revision: 228311
URL: http://svn.freebsd.org/changeset/base/228311
Log:
MK_CTF is not defined when kmod.mk is used with old bsd.own.mk.
Modified:
head/sys/conf/kmod.mk
Modified: head/sys/conf/kmod.mk
==
--- head/sys/conf/kmod.mk Tue Dec 6 17:54:59 2011(r228310)
+++ head/sys/conf/kmod.mk Tue Dec 6 18:01:09 2011(r228311)
@@ -201,7 +201,7 @@ ${KMOD}.kld: ${OBJS}
${FULLPROG}: ${OBJS}
.endif
${LD} ${LDFLAGS} -r -d -o ${.TARGET} ${OBJS}
-.if ${MK_CTF} != "no"
+.if defined(MK_CTF) && ${MK_CTF} != "no"
${CTFMERGE} ${CTFFLAGS} -o ${.TARGET} ${OBJS}
.endif
.if defined(EXPORT_SYMS)
___
svn-src-all@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"
Re: svn commit: r228158 - in head: . share/mk sys/conf
Gary, Hans,
On Tue, Dec 6, 2011 at 4:06 PM, Gary Jennejohn wrote:
> I see regressions when building kernel modules from /usr/ports:
> >
> > On Monday 05 December 2011 14:39:56 Robert Huff wrote:
> > > Hello:
> > > When trying to update I get:
> > >
> > > ===> Building for cuse4bsd-kmod-0.1.21_2
> > > make -f
> > >
> /data/port-work/usr/ports/multimedia/cuse4bsd-kmod/work/cuse4bsd-kmod-0.1.
> > > 21/Makefile.lib HAVE_DEBUG=YES all Warning: Object directory not
> changed
> > > from original
> > >
> /data/port-work/usr/ports/multimedia/cuse4bsd-kmod/work/cuse4bsd-kmod-0.1.
> > > 21 make -f
> > >
> /data/port-work/usr/ports/multimedia/cuse4bsd-kmod/work/cuse4bsd-kmod-0.1.
> > > 21/Makefile.kmod all "/sys/conf/kmod.mk", line 204: Malformed
> conditional
> > > (${MK_CTF} != "no") "/sys/conf/kmod.mk", line 206: if-less endif
> > > make: fatal errors encountered -- cannot continue
> > > *** Error code 1
> >
>
> I encountered this problem yesterday. My quick hack was to put
> .if defined(MK_CTF) .endif around the offending lines, similar to
> EXPORT_SYMS.
>
This should be fixed now (rev. 228311)
Max
___
svn-src-all@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"
svn commit: r228313 - head/sys/netinet
Author: glebius
Date: Tue Dec 6 20:55:20 2011
New Revision: 228313
URL: http://svn.freebsd.org/changeset/base/228313
Log:
Fix a very special case when SIOCAIFADDR supplies mask of 0.0.0.0,
don't overwrite the mask with autoguessing based on classes.
Modified:
head/sys/netinet/in.c
Modified: head/sys/netinet/in.c
==
--- head/sys/netinet/in.c Tue Dec 6 19:04:45 2011(r228312)
+++ head/sys/netinet/in.c Tue Dec 6 20:55:20 2011(r228313)
@@ -73,7 +73,7 @@ static intin_addprefix(struct in_ifaddr
static int in_scrubprefix(struct in_ifaddr *, u_int);
static voidin_socktrim(struct sockaddr_in *);
static int in_ifinit(struct ifnet *,
- struct in_ifaddr *, struct sockaddr_in *, int);
+ struct in_ifaddr *, struct sockaddr_in *, int, int);
static voidin_purgemaddrs(struct ifnet *);
static VNET_DEFINE(int, sameprefixcarponly);
@@ -517,7 +517,7 @@ in_control(struct socket *so, u_long cmd
case SIOCSIFADDR:
error = in_ifinit(ifp, ia,
- (struct sockaddr_in *) &ifr->ifr_addr, 1);
+ (struct sockaddr_in *) &ifr->ifr_addr, 1, 0);
if (error != 0 && iaIsNew)
break;
if (error == 0) {
@@ -569,7 +569,8 @@ in_control(struct socket *so, u_long cmd
maskIsNew = 1; /* We lie; but the effect's the same */
}
if (hostIsNew || maskIsNew)
- error = in_ifinit(ifp, ia, &ifra->ifra_addr, 0);
+ error = in_ifinit(ifp, ia, &ifra->ifra_addr, 0,
+ maskIsNew);
if (error != 0 && iaIsNew)
break;
@@ -842,7 +843,7 @@ in_ifscrub(struct ifnet *ifp, struct in_
*/
static int
in_ifinit(struct ifnet *ifp, struct in_ifaddr *ia, struct sockaddr_in *sin,
-int scrub)
+int scrub, int masksupplied)
{
register u_long i = ntohl(sin->sin_addr.s_addr);
int flags = RTF_UP, error = 0;
@@ -872,7 +873,7 @@ in_ifinit(struct ifnet *ifp, struct in_i
* Be compatible with network classes, if netmask isn't supplied,
* guess it based on classes.
*/
- if (ia->ia_subnetmask == 0) {
+ if (!masksupplied) {
if (IN_CLASSA(i))
ia->ia_subnetmask = IN_CLASSA_NET;
else if (IN_CLASSB(i))
___
svn-src-all@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"
Re: svn commit: r227997 - vendor/libcxxrt/8931d9e5180830a5433d16ae6b3ad8dd9e629512
On Sat, 2011-11-26 at 14:20:34 +, David Chisnall wrote: > Author: theraven > Date: Sat Nov 26 14:20:34 2011 > New Revision: 227997 > URL: http://svn.freebsd.org/changeset/base/227997 > > Log: > Versioned snapshot for libcxxrt > > Approved by:dim (mentor) > > Added: > vendor/libcxxrt/8931d9e5180830a5433d16ae6b3ad8dd9e629512/ > - copied from r227996, vendor/libcxxrt/dist/ Ho humm, would you really call this a versioned directory? It's kinda silly, IMHO, as it provides no clues as to the evolution of the code. It's just a content identifier and cannot even be sorted. Please consider using dated snapshots instead, e.g. -MM-DD, etc. Thanks Uli ___ svn-src-all@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/svn-src-all To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"
svn commit: r228317 - head/sys/kern
Author: alc Date: Wed Dec 7 07:03:14 2011 New Revision: 228317 URL: http://svn.freebsd.org/changeset/base/228317 Log: Eliminate the possibility of 32-bit arithmetic overflow in the calculation of vm_kmem_size that may occur if the system administrator has specified a vm.vm_kmem_size tunable value that exceeds the hard cap. PR: 162741 Submitted by: Adam McDougall Reviewed by: bde@ MFC after:3 weeks Modified: head/sys/kern/kern_malloc.c Modified: head/sys/kern/kern_malloc.c == --- head/sys/kern/kern_malloc.c Wed Dec 7 00:22:34 2011(r228316) +++ head/sys/kern/kern_malloc.c Wed Dec 7 07:03:14 2011(r228317) @@ -740,11 +740,11 @@ kmeminit(void *dummy) /* * Limit kmem virtual size to twice the physical memory. * This allows for kmem map sparseness, but limits the size -* to something sane. Be careful to not overflow the 32bit -* ints while doing the check. +* to something sane. Be careful to not overflow the 32bit +* ints while doing the check or the adjustment. */ - if (((vm_kmem_size / 2) / PAGE_SIZE) > cnt.v_page_count) - vm_kmem_size = 2 * cnt.v_page_count * PAGE_SIZE; + if (vm_kmem_size / 2 / PAGE_SIZE > mem_size) + vm_kmem_size = 2 * mem_size * PAGE_SIZE; #ifdef DEBUG_MEMGUARD tmp = memguard_fudge(vm_kmem_size, vm_kmem_size_max); ___ svn-src-all@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/svn-src-all To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"
