Re: svn commit: r260486 - head/etc/defaults
We have 2 and 4 hw.ncpu NFS servers with 70+ disks, so there may be cases, where even the default maximum of 256 threads is not enough to feed the -otherwise slow- disks. I guess the real solution here is to change the nfsd worker model to async. (adapting to top poster) On 01/09/14 18:14, Adrian Chadd wrote: If it's one request per nfsd then we should likely scale it separate to ncpu. Otherwise the default for two or core boxes may not be enough. Adrian On Jan 9, 2014 11:29 AM, "Alexander Motin" wrote: On 09.01.2014 18:18, Adrian Chadd wrote: Depends if you're thinking locally or globally. Locally - for nfs? not a big deal. Globally - NFS, ZFS, GELI, geom/cam, NIC, etc.. suddenly your machine could default to having a couple thousand worker threads just for a HBA and a 10GE NIC. That's a little nuts. So, what is your point? Each NFS thread (unlike GEOM or CAM) executes only _one_ request at a time. Would you like your 128-core/many-spindle system executed only 4 synchronous requests at a time? On 9 January 2014 08:00, Alexander Motin wrote: On 09.01.2014 17:57, Adrian Chadd wrote: .. so with say, 128 core boxes showing up, is this really a good default? And what is the price? 16K+ of KVA per thread for thread stack, etc? 4 threads is probably much worse default there. May be nfsd's default could be tuned, but obviously it should not be hardcoded value. On 9 January 2014 07:55, Alexander Motin wrote: Author: mav Date: Thu Jan 9 15:55:55 2014 New Revision: 260486 URL: http://svnweb.freebsd.org/changeset/base/260486 Log: Remove very low default limit of 4 nfsd threads. nfsd's own default is 8 * hw.ncpu, that sounds more appropriate for these SMP/NCQ/... days. Modified: head/etc/defaults/rc.conf Modified: head/etc/defaults/rc.conf == --- head/etc/defaults/rc.conf Thu Jan 9 15:38:28 2014 (r260485) +++ head/etc/defaults/rc.conf Thu Jan 9 15:55:55 2014 (r260486) @@ -311,7 +311,7 @@ nfs_client_enable="NO" # This host is a nfs_access_cache="60" # Client cache timeout in seconds nfs_server_enable="NO" # This host is an NFS server (or NO). oldnfs_server_enable="NO" # Run the old NFS server (YES/NO). -nfs_server_flags="-u -t -n 4" # Flags to nfsd (if enabled). +nfs_server_flags="-u -t" # Flags to nfsd (if enabled). mountd_enable="NO" # Run mountd (or NO). mountd_flags="-r" # Flags to mountd (if NFS server enabled). weak_mountd_authentication="NO"# Allow non-root mount requests to be served. -- Alexander Motin -- Alexander Motin ___ svn-src-...@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-src-head@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
Re: svn commit: r259730 - in head: gnu/lib/csu gnu/lib/libgcc gnu/lib/libstdc++ gnu/lib/libsupc++ lib/atf/libatf-c/tests share/mk sys/conf tools/tools/ath/athstats tools/tools/net80211/wlanstats usr.b
On Thu, 9 Jan 2014, Dimitry Andric wrote: On 09 Jan 2014, at 16:02, Ian Lepore wrote: ... At r259730: cc -c -O -pipe -std=c99 -g -Wall -Wredundant-decls -Wnested-externs -Wstrict-prototypes -Wmissing-prototypes -Wpointer-arith -Winline -Wcast-qual -Wundef -Wno-pointer-sign -fformat-extensions -Wmissing-include-dirs -fdiagnostics-show-option -Wno-error-tautological-compare -Wno-error-empty-body -Wno-error-parentheses-equality -nostdinc -I. -I/local/build/staging/freebsd/imx53/src/sys -I/local/build/staging/freebsd/imx53/src/sys/contrib/altq -I/local/build/staging/freebsd/imx53/src/sys/contrib/libfdt -D_KERNEL -DHAVE_KERNEL_OPTION_HEADERS -include opt_global.h -funwind-tables -ffreestanding -Werror /local/build/staging/freebsd/imx53/src/sys/arm/arm/trap.c The difference is that the -mllvm and -arm-enable-ehabi flags are missing. Aha, this was a nasty mistake, sorry. The main kernel builds do not include bsd.sys.mk, so CFLAGS.clang and CFLAGS.gcc do not work there at the moment. Fixed in r260494. It is a bug for kernel builds to include anything in /usr/share/mk, since this gives a dependency on the host environment. This bug is very old, but not too bad for files that should rarely change, e.g., sys.mk. It it is very bad for critical variations of CFLAGS. The old bug has been expanded exponentially by namespace pollution in environment files. Old versions of conf/*.mk have only the following includes: kern.post.mk:.include "kern.mk" kmod.mk:.include kmod.mk:.include kmod.mk:.include kmod.mk:.include kmod.mk:.include "kern.mk" This shows that the old bug was only in kmod.mk and in nested pollution in sys.mk. It used to be in kern.post.mk, when that included bsd.kern.mk instead of kern.mk. This was fixed for new sources by moving bsd.kern.mk to kern.mk in the kernel source tree, but collaterally enlarged for old sources by removing bsd.kern.mk from the host source tree (thus, old kernel sources no longer compiled with new hosts, even if nothing in *kern.mk changed). The other includes here shouldn't cause any problems since they shouldn't do anything that depends on the host environment including the compiler. conf/*.mk now has the following includes: kern.post.mk:.include "kern.mk" kern.pre.mk:.include kern.pre.mk:.include kmod.mk:.include kmod.mk:.include kmod.mk:.include kmod.mk:.include kmod.mk:.include kmod.mk:.include "kern.mk" bsd.sys.mk still isn't explicitly here. For modules, it is included by bsd.kmod.mk. The layering for *kmod.mk is backwards: - modules Makefiles include bsd.kmod.mk from the host environment - bsd.kmod.mk guesses the system directory and includes conf/kmod.mk from there - conf/kmod.mk then includes other files from the host environment, as above - bsd.kmod.mk then includes bsd.sys.mk from the host environment. This bug is old. I think you just copied the bug from bsd.kmod.mk so that it affects kernel builds too. Bruce ___ svn-src-head@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
Re: svn commit: r260463 - head/sys/dev/usb/wlan
On 2014/01/09 16:55, Gleb Smirnoff wrote: On Thu, Jan 09, 2014 at 01:48:34AM +, Kevin Lo wrote: K> Author: kevlo K> Date: Thu Jan 9 01:48:33 2014 K> New Revision: 260463 K> URL: http://svnweb.freebsd.org/changeset/base/260463 K> K> Log: K> Replace deprecated M_DONTWAIT with M_NOWAIT. K> K> Modified: K> head/sys/dev/usb/wlan/if_rsu.c K> head/sys/dev/usb/wlan/if_urtwn.c K> K> Modified: head/sys/dev/usb/wlan/if_rsu.c K> == K> --- head/sys/dev/usb/wlan/if_rsu.cThu Jan 9 00:59:03 2014 (r260462) K> +++ head/sys/dev/usb/wlan/if_rsu.cThu Jan 9 01:48:33 2014 (r260463) K> @@ -1145,11 +1145,11 @@ rsu_event_survey(struct rsu_softc *sc, u K> pktlen = sizeof(*wh) + le32toh(bss->ieslen); K> if (__predict_false(pktlen > MCLBYTES)) K> return; K> - MGETHDR(m, M_DONTWAIT, MT_DATA); K> + MGETHDR(m, M_NOWAIT, MT_DATA); K> if (__predict_false(m == NULL)) K> return; K> if (pktlen > MHLEN) { K> - MCLGET(m, M_DONTWAIT); K> + MCLGET(m, M_NOWAIT); K> if (!(m->m_flags & M_EXT)) { K> m_free(m); K> return; K> @@ -1358,13 +1358,13 @@ rsu_rx_frame(struct rsu_softc *sc, uint8 K> DPRINTFN(5, "Rx frame len=%d rate=%d infosz=%d rssi=%d\n", K> pktlen, rate, infosz, *rssi); K> K> - MGETHDR(m, M_DONTWAIT, MT_DATA); K> + MGETHDR(m, M_NOWAIT, MT_DATA); K> if (__predict_false(m == NULL)) { K> ifp->if_ierrors++; K> return NULL; K> } K> if (pktlen > MHLEN) { K> - MCLGET(m, M_DONTWAIT); K> + MCLGET(m, M_NOWAIT); K> if (__predict_false(!(m->m_flags & M_EXT))) { K> ifp->if_ierrors++; K> m_freem(m); Thanks! The macros MGETHDR, MCLGET should also be converted to functions. Fixed in r260501. Thanks. Kevin ___ svn-src-head@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r260501 - head/sys/dev/usb/wlan
Author: kevlo Date: Fri Jan 10 02:47:20 2014 New Revision: 260501 URL: http://svnweb.freebsd.org/changeset/base/260501 Log: Use m_getcl() instead of MGETHDR/MCLGET macros. Suggested by: glebius Modified: head/sys/dev/usb/wlan/if_rsu.c Modified: head/sys/dev/usb/wlan/if_rsu.c == --- head/sys/dev/usb/wlan/if_rsu.c Fri Jan 10 01:44:34 2014 (r260500) +++ head/sys/dev/usb/wlan/if_rsu.c Fri Jan 10 02:47:20 2014 (r260501) @@ -1145,16 +1145,9 @@ rsu_event_survey(struct rsu_softc *sc, u pktlen = sizeof(*wh) + le32toh(bss->ieslen); if (__predict_false(pktlen > MCLBYTES)) return; - MGETHDR(m, M_NOWAIT, MT_DATA); + m = m_getcl(M_NOWAIT, MT_DATA, M_PKTHDR); if (__predict_false(m == NULL)) return; - if (pktlen > MHLEN) { - MCLGET(m, M_NOWAIT); - if (!(m->m_flags & M_EXT)) { - m_free(m); - return; - } - } wh = mtod(m, struct ieee80211_frame *); wh->i_fc[0] = IEEE80211_FC0_VERSION_0 | IEEE80211_FC0_TYPE_MGT | IEEE80211_FC0_SUBTYPE_BEACON; @@ -1358,19 +1351,11 @@ rsu_rx_frame(struct rsu_softc *sc, uint8 DPRINTFN(5, "Rx frame len=%d rate=%d infosz=%d rssi=%d\n", pktlen, rate, infosz, *rssi); - MGETHDR(m, M_NOWAIT, MT_DATA); + m = m_getcl(M_NOWAIT, MT_DATA, M_PKTHDR); if (__predict_false(m == NULL)) { ifp->if_ierrors++; return NULL; } - if (pktlen > MHLEN) { - MCLGET(m, M_NOWAIT); - if (__predict_false(!(m->m_flags & M_EXT))) { - ifp->if_ierrors++; - m_freem(m); - return NULL; - } - } /* Finalize mbuf. */ m->m_pkthdr.rcvif = ifp; /* Hardware does Rx TCP checksum offload. */ ___ svn-src-head@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
Re: svn commit: r260486 - head/etc/defaults
On Thu, 09 Jan 2014 18:29:11 +0200 Alexander Motin wrote: > On 09.01.2014 18:18, Adrian Chadd wrote: > > Depends if you're thinking locally or globally. > > > > Locally - for nfs? not a big deal. > > > > Globally - NFS, ZFS, GELI, geom/cam, NIC, etc.. suddenly your > > machine could default to having a couple thousand worker threads > > just for a HBA and a 10GE NIC. That's a little nuts. > > So, what is your point? Each NFS thread (unlike GEOM or CAM) executes > only _one_ request at a time. Would you like your > 128-core/many-spindle system executed only 4 synchronous requests at > a time? > I certainly do not want to have 8 * ncpu threads hanging around. While said extra CPUs were all but forced on me by Intel innovators, my NFS workloads hardly did scale at the same pace over the same time period. -- Alexander Kabaev signature.asc Description: PGP signature
Re: svn commit: r260486 - head/etc/defaults
On 10 Jan 2014, at 2:48, Adrian Chadd wrote: > Depends if you're thinking locally or globally. > > Locally - for nfs? not a big deal. > > Globally - NFS, ZFS, GELI, geom/cam, NIC, etc.. suddenly your machine > could default to having a couple thousand worker threads just for a > HBA and a 10GE NIC. That's a little nuts. > Most of those aren't paid unless you actually enable the thing in question. Same with this change, if you aren't using NFS you don't pay the cost. -- Daniel O'Connor software and network engineer for Genesis Software - http://www.gsoft.com.au "The nice thing about standards is that there are so many of them to choose from." -- Andrew Tanenbaum GPG Fingerprint - 5596 B766 97C0 0E94 4347 295E E593 DC20 7B3F CE8C signature.asc Description: Message signed with OpenPGP using GPGMail
Re: svn commit: r260494 - head/sys/conf
On Thu, 2014-01-09 at 22:16 +, Dimitry Andric wrote: > Author: dim > Date: Thu Jan 9 22:16:30 2014 > New Revision: 260494 > URL: http://svnweb.freebsd.org/changeset/base/260494 > > Log: > Fix a braino with r259730: we cannot currently use CFLAGS.gcc or > CFLAGS.clang in sys/conf/Makefile.arm, since the main kernel build does > not use . So revert that particular change for now. > > Pointy hat to: me > Noticed by: zbb > MFC after: 3 days > X-MFC-With: r259730 > > Modified: > head/sys/conf/Makefile.arm > > Modified: head/sys/conf/Makefile.arm > == > --- head/sys/conf/Makefile.armThu Jan 9 20:57:19 2014 > (r260493) > +++ head/sys/conf/Makefile.armThu Jan 9 22:16:30 2014 > (r260494) > @@ -39,16 +39,20 @@ SYSTEM_DEP:= ${SYSTEM_DEP:$S/conf/ldscri > STRIP_FLAGS = -S > .endif > > -CFLAGS.gcc += -mno-thumb-interwork > +.if ${COMPILER_TYPE} != "clang" > +CFLAGS += -mno-thumb-interwork > +.endif > > .if empty(DDB_ENABLED) > -.if defined(WITHOUT_ARM_EABI) > -CFLAGS.gcc += -mno-apcs-frame > +.if defined(WITHOUT_ARM_EABI) && ${COMPILER_TYPE} != "clang" > +CFLAGS += -mno-apcs-frame > .endif > .elif !defined(WITHOUT_ARM_EABI) > CFLAGS += -funwind-tables > +.if ${COMPILER_TYPE} == "clang" > # clang requires us to tell it to emit assembly with unwind information > -CFLAGS.clang += -mllvm -arm-enable-ehabi > +CFLAGS += -mllvm -arm-enable-ehabi > +.endif > .endif > > SYSTEM_LD_ = ${LD} -Bdynamic -T ldscript.$M.noheader ${LDFLAGS} \ That fixed it, now we can get backtraces in DDB again. Thanks. -- Ian ___ svn-src-head@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r260496 - head/sys/netinet6
Author: jmg Date: Thu Jan 9 22:41:18 2014 New Revision: 260496 URL: http://svnweb.freebsd.org/changeset/base/260496 Log: revert part of r260485 which changes how part of the header gets included.. netstat uses -DKERNEL=1 to get these parts and breaks the build w/o it... melifaro@ says that ae@ is probably asleep, and the PR doesn't have this part of the patch... Probably a local change got in by accident.. PR: 185148 Pointy hat to:ae@ Modified: head/sys/netinet6/ip6_mroute.h Modified: head/sys/netinet6/ip6_mroute.h == --- head/sys/netinet6/ip6_mroute.h Thu Jan 9 22:40:51 2014 (r260495) +++ head/sys/netinet6/ip6_mroute.h Thu Jan 9 22:41:18 2014 (r260496) @@ -194,7 +194,7 @@ struct sioc_mif_req6 { u_quad_t obytes;/* Output byte count on mif */ }; -#ifdef _KERNEL +#if defined(_KERNEL) || defined(KERNEL) /* * The kernel's multicast-interface structure. */ ___ svn-src-head@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
Re: svn commit: r259730 - in head: gnu/lib/csu gnu/lib/libgcc gnu/lib/libstdc++ gnu/lib/libsupc++ lib/atf/libatf-c/tests share/mk sys/conf tools/tools/ath/athstats tools/tools/net80211/wlanstats usr.b
On 09 Jan 2014, at 16:02, Ian Lepore wrote: ... > At r259730: > > cc -c -O -pipe -std=c99 -g -Wall -Wredundant-decls -Wnested-externs > -Wstrict-prototypes -Wmissing-prototypes -Wpointer-arith -Winline > -Wcast-qual -Wundef -Wno-pointer-sign -fformat-extensions > -Wmissing-include-dirs -fdiagnostics-show-option > -Wno-error-tautological-compare -Wno-error-empty-body > -Wno-error-parentheses-equality -nostdinc -I. > -I/local/build/staging/freebsd/imx53/src/sys > -I/local/build/staging/freebsd/imx53/src/sys/contrib/altq > -I/local/build/staging/freebsd/imx53/src/sys/contrib/libfdt -D_KERNEL > -DHAVE_KERNEL_OPTION_HEADERS -include opt_global.h -funwind-tables > -ffreestanding > -Werror /local/build/staging/freebsd/imx53/src/sys/arm/arm/trap.c > > The difference is that the -mllvm and -arm-enable-ehabi flags are > missing. Aha, this was a nasty mistake, sorry. The main kernel builds do not include bsd.sys.mk, so CFLAGS.clang and CFLAGS.gcc do not work there at the moment. Fixed in r260494. -Dimitry signature.asc Description: Message signed with OpenPGP using GPGMail
svn commit: r260494 - head/sys/conf
Author: dim Date: Thu Jan 9 22:16:30 2014 New Revision: 260494 URL: http://svnweb.freebsd.org/changeset/base/260494 Log: Fix a braino with r259730: we cannot currently use CFLAGS.gcc or CFLAGS.clang in sys/conf/Makefile.arm, since the main kernel build does not use . So revert that particular change for now. Pointy hat to:me Noticed by: zbb MFC after:3 days X-MFC-With: r259730 Modified: head/sys/conf/Makefile.arm Modified: head/sys/conf/Makefile.arm == --- head/sys/conf/Makefile.arm Thu Jan 9 20:57:19 2014(r260493) +++ head/sys/conf/Makefile.arm Thu Jan 9 22:16:30 2014(r260494) @@ -39,16 +39,20 @@ SYSTEM_DEP:= ${SYSTEM_DEP:$S/conf/ldscri STRIP_FLAGS = -S .endif -CFLAGS.gcc += -mno-thumb-interwork +.if ${COMPILER_TYPE} != "clang" +CFLAGS += -mno-thumb-interwork +.endif .if empty(DDB_ENABLED) -.if defined(WITHOUT_ARM_EABI) -CFLAGS.gcc += -mno-apcs-frame +.if defined(WITHOUT_ARM_EABI) && ${COMPILER_TYPE} != "clang" +CFLAGS += -mno-apcs-frame .endif .elif !defined(WITHOUT_ARM_EABI) CFLAGS += -funwind-tables +.if ${COMPILER_TYPE} == "clang" # clang requires us to tell it to emit assembly with unwind information -CFLAGS.clang += -mllvm -arm-enable-ehabi +CFLAGS += -mllvm -arm-enable-ehabi +.endif .endif SYSTEM_LD_ = ${LD} -Bdynamic -T ldscript.$M.noheader ${LDFLAGS} \ ___ svn-src-head@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r260493 - head/sys/arm/include
Author: ian Date: Thu Jan 9 20:57:19 2014 New Revision: 260493 URL: http://svnweb.freebsd.org/changeset/base/260493 Log: Add a prototype for the new arm_devmap_print_table(). This should have been part of r260490. Modified: head/sys/arm/include/devmap.h Modified: head/sys/arm/include/devmap.h == --- head/sys/arm/include/devmap.h Thu Jan 9 20:06:14 2014 (r260492) +++ head/sys/arm/include/devmap.h Thu Jan 9 20:57:19 2014 (r260493) @@ -87,4 +87,7 @@ void arm_devmap_bootstrap(vm_offset_t _l void * arm_devmap_ptov(vm_paddr_t _pa, vm_size_t _sz); vm_paddr_t arm_devmap_vtop(void * _va, vm_size_t _sz); +/* Print the static mapping table; used for bootverbose output. */ +void arm_devmap_print_table(void); + #endif ___ svn-src-head@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
Re: svn commit: r260485 - head/sys/netinet6
Andrey V. Elsukov wrote this message on Thu, Jan 09, 2014 at 15:38 +: > Author: ae > Date: Thu Jan 9 15:38:28 2014 > New Revision: 260485 > URL: http://svnweb.freebsd.org/changeset/base/260485 > > Log: > Remove extra nesting from X_ip6_mforward() function. > Also remove disabled definitions from ip6_mroute.h. [...] > Modified: head/sys/netinet6/ip6_mroute.h > == > --- head/sys/netinet6/ip6_mroute.hThu Jan 9 15:35:35 2014 > (r260484) > +++ head/sys/netinet6/ip6_mroute.hThu Jan 9 15:38:28 2014 > (r260485) > @@ -145,11 +145,6 @@ struct mrt6stat { > struct omrt6msg { > u_long unused1; > u_char im6_msgtype;/* what type of message */ > -#if 0 > -#define MRT6MSG_NOCACHE 1 > -#define MRT6MSG_WRONGMIF 2 > -#define MRT6MSG_WHOLEPKT 3 /* used for user level encap*/ > -#endif > u_char im6_mbz;/* must be zero */ > u_char im6_mif;/* mif rec'd on */ > u_char unused2; > @@ -199,7 +194,7 @@ struct sioc_mif_req6 { > u_quad_t obytes;/* Output byte count on mif */ > }; > > -#if defined(_KERNEL) || defined(KERNEL) > +#ifdef _KERNEL > /* > * The kernel's multicast-interface structure. > */ This last change broken netstat... netstat defines KERNEL to get these structures... -- John-Mark Gurney Voice: +1 415 225 5579 "All that I will do, has been done, All that I have, has not." ___ svn-src-head@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r260492 - head/sbin/geom/class/part
Author: lme (ports committer) Date: Thu Jan 9 20:06:14 2014 New Revision: 260492 URL: http://svnweb.freebsd.org/changeset/base/260492 Log: Check if the given argument to 'gpart add' is actually a geom device and give a hint to use 'gpart create' before trying to add a partition. Approved by: pjd Modified: head/sbin/geom/class/part/geom_part.c (contents, props changed) Modified: head/sbin/geom/class/part/geom_part.c == --- head/sbin/geom/class/part/geom_part.c Thu Jan 9 18:53:21 2014 (r260491) +++ head/sbin/geom/class/part/geom_part.c Thu Jan 9 20:06:14 2014 (r260492) @@ -454,8 +454,19 @@ gpart_autofill(struct gctl_req *req) if (s == NULL) abort(); gp = find_geom(cp, s); - if (gp == NULL) - errx(EXIT_FAILURE, "No such geom: %s.", s); + if (gp == NULL) { + if (g_device_path(s) == NULL) { + errx(EXIT_FAILURE, "No such geom %s.", s); + } else { + /* +* We don't free memory allocated by g_device_path() as +* we are about to exit. +*/ + errx(EXIT_FAILURE, + "No partitioning scheme found on geom %s. Create one first using 'gpart create'.", + s); + } + } pp = LIST_FIRST(&gp->lg_consumer)->lg_provider; if (pp == NULL) errx(EXIT_FAILURE, "Provider for geom %s not found.", s); ___ svn-src-head@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r260491 - head/contrib/atf/atf-c++/detail
Author: jmmv Date: Thu Jan 9 18:53:21 2014 New Revision: 260491 URL: http://svnweb.freebsd.org/changeset/base/260491 Log: Use .cpp as the extension for temporary C++ files. Using a .c extension for a C++ file raises the following warning, which breaks our header file tests if the compiler is using -Werror as well: c++: warning: treating 'c' input as 'c++' when in C++ mode, this behavior is deprecated Obtained from:atf (git 3104010c2849330440cc0ce108ff34191349) MFC after:3 days Modified: head/contrib/atf/atf-c++/detail/test_helpers.cpp Modified: head/contrib/atf/atf-c++/detail/test_helpers.cpp == --- head/contrib/atf/atf-c++/detail/test_helpers.cppThu Jan 9 18:51:57 2014(r260490) +++ head/contrib/atf/atf-c++/detail/test_helpers.cppThu Jan 9 18:53:21 2014(r260491) @@ -67,14 +67,14 @@ build_check_cxx_o(const atf::tests::tc& void header_check(const char *hdrname) { -std::ofstream srcfile("test.c"); +std::ofstream srcfile("test.cpp"); ATF_REQUIRE(srcfile); srcfile << "#include <" << hdrname << ">\n"; srcfile.close(); const std::string failmsg = std::string("Header check failed; ") + hdrname + " is not self-contained"; -build_check_cxx_o_aux(atf::fs::path("test.c"), failmsg.c_str(), true); +build_check_cxx_o_aux(atf::fs::path("test.cpp"), failmsg.c_str(), true); } atf::fs::path ___ svn-src-head@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r260490 - head/sys/arm/arm
Author: ian Date: Thu Jan 9 18:51:57 2014 New Revision: 260490 URL: http://svnweb.freebsd.org/changeset/base/260490 Log: Add a function to print the contents of the static device mapping table, and invoke it for bootverbose logging, and also from a new DDB command, "show devmap". Also tweak the format string for the bootverbose output of physical memory chunks to get the leading zeros in the hex values. Modified: head/sys/arm/arm/devmap.c head/sys/arm/arm/machdep.c Modified: head/sys/arm/arm/devmap.c == --- head/sys/arm/arm/devmap.c Thu Jan 9 18:28:58 2014(r260489) +++ head/sys/arm/arm/devmap.c Thu Jan 9 18:51:57 2014(r260490) @@ -31,6 +31,8 @@ __FBSDID("$FreeBSD$"); * Routines for mapping device memory. */ +#include "opt_ddb.h" + #include #include #include @@ -54,6 +56,36 @@ static u_int akva_devmap_idx; static vm_offset_t akva_devmap_vaddr = ARM_VECTORS_HIGH; /* + * Print the contents of the static mapping table using the provided printf-like + * output function (which will be either printf or db_printf). + */ +static void +devmap_dump_table(int (*prfunc)(const char *, ...)) +{ + const struct arm_devmap_entry *pd; + + if (devmap_table == NULL || devmap_table[0].pd_size == 0) { + prfunc("No static device mappings.\n"); + return; + } + + prfunc("Static device mappings:\n"); + for (pd = devmap_table; pd->pd_size != 0; ++pd) { + prfunc(" 0x%08x - 0x%08x mapped at VA 0x%08x\n", + pd->pd_pa, pd->pd_pa + pd->pd_size - 1, pd->pd_va); + } +} + +/* + * Print the contents of the static mapping table. Used for bootverbose. + */ +void +arm_devmap_print_table() +{ + devmap_dump_table(printf); +} + +/* * Return the "last" kva address used by the registered devmap table. It's * actually the lowest address used by the static mappings, i.e., the address of * the first unusable byte of KVA. @@ -266,3 +298,13 @@ pmap_unmapdev(vm_offset_t va, vm_size_t kva_free(va, origsize); } +#ifdef DDB +#include + +DB_SHOW_COMMAND(devmap, db_show_devmap) +{ + devmap_dump_table(db_printf); +} + +#endif /* DDB */ + Modified: head/sys/arm/arm/machdep.c == --- head/sys/arm/arm/machdep.c Thu Jan 9 18:28:58 2014(r260489) +++ head/sys/arm/arm/machdep.c Thu Jan 9 18:51:57 2014(r260490) @@ -379,10 +379,10 @@ cpu_startup(void *dummy) vm_paddr_t size; size = phys_avail[indx + 1] - phys_avail[indx]; - printf("%#08jx - %#08jx, %ju bytes (%ju pages)\n", + printf(" 0x%08jx - 0x%08jx, %ju KBytes (%ju pages)\n", (uintmax_t)phys_avail[indx], (uintmax_t)phys_avail[indx + 1] - 1, - (uintmax_t)size, (uintmax_t)size / PAGE_SIZE); + (uintmax_t)size / 1024, (uintmax_t)size / PAGE_SIZE); } } @@ -392,6 +392,9 @@ cpu_startup(void *dummy) (uintmax_t)ptoa(cnt.v_free_count), (uintmax_t)ptoa(cnt.v_free_count) / 1048576); + if (bootverbose) + arm_devmap_print_table(); + bufinit(); vm_pager_bufferinit(); pcb->un_32.pcb32_und_sp = (u_int)thread0.td_kstack + ___ svn-src-head@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r260488 - head/sys/net
Author: melifaro Date: Thu Jan 9 18:13:25 2014 New Revision: 260488 URL: http://svnweb.freebsd.org/changeset/base/260488 Log: Split rt_newaddrmsg_fib() into two different functions. Adding/deleting interface addresses involves access to 3 different subsystems, int different parts of code. Each call can fail, so reporting successful operation by rtsock in the middle of the process error-prone. Further split routing notification API and actual rtsock calls via creating public-available rt_addrmsg() / rt_routemsg() functions with "private" rtsock_* backend. MFC after:2 weeks Modified: head/sys/net/route.c head/sys/net/route.h head/sys/net/rtsock.c Modified: head/sys/net/route.c == --- head/sys/net/route.cThu Jan 9 17:27:00 2014(r260487) +++ head/sys/net/route.cThu Jan 9 18:13:25 2014(r260488) @@ -37,6 +37,7 @@ #include "opt_inet.h" #include "opt_inet6.h" #include "opt_route.h" +#include "opt_sctp.h" #include "opt_mrouting.h" #include "opt_mpath.h" @@ -52,6 +53,7 @@ #include #include #include +#include #include #include @@ -86,6 +88,13 @@ #defineRT_NUMFIBS 1 #endif +#if defined(INET) || defined(INET6) +#ifdef SCTP +extern void sctp_addr_change(struct ifaddr *ifa, int cmd); +#endif /* SCTP */ +#endif + + /* This is read-only.. */ u_int rt_numfibs = RT_NUMFIBS; SYSCTL_UINT(_net, OID_AUTO, fibs, CTLFLAG_RD, &rt_numfibs, 0, ""); @@ -118,7 +127,8 @@ VNET_DEFINE(int, rttrash); /* routes no /* compare two sockaddr structures */ -#definesa_equal(a1, a2) (bcmp((a1), (a2), (a1)->sa_len) == 0) +#definesa_equal(a1, a2) (((a1)->sa_len == (a2)->sa_len) && \ +(bcmp((a1), (a2), (a1)->sa_len) == 0)) /* * Convert a 'struct radix_node *' to a 'struct rtentry *'. @@ -1731,3 +1741,99 @@ rtinit(struct ifaddr *ifa, int cmd, int } return (rtinit1(ifa, cmd, flags, fib)); } + +/* + * Announce interface address arrival/withdraw + * Returns 0 on success. + */ +int +rt_addrmsg(int cmd, struct ifaddr *ifa, int fibnum) +{ + + KASSERT(cmd == RTM_ADD || cmd == RTM_DELETE, + ("unexpected cmd %u", cmd)); + + if (fibnum != RT_ALL_FIBS) { + KASSERT(fibnum >= 0 && fibnum < rt_numfibs, ("%s: " + "fibnum out of range 0 <= %d < %d", __func__, +fibnum, rt_numfibs)); + } + + return (rtsock_addrmsg(cmd, ifa, fibnum)); +} + + +/* + * Announce route addition/removal + * Users of this function MUST validate input data BEFORE calling. + * However we have to be able to handle invalid data: + * if some userland app sends us "invalid" route message (invalid mask, + * no dst, wrokg address families, etc...) we need to pass it back + * to app (and any other rtsock consumers) with rtm_errno field set to + * non-zero value. + * Returns 0 on success. + */ +int +rt_routemsg(int cmd, struct ifnet *ifp, int error, struct rtentry *rt, +int fibnum) +{ + + KASSERT(cmd == RTM_ADD || cmd == RTM_DELETE, + ("unexpected cmd %u", cmd)); + + if (fibnum != RT_ALL_FIBS) { + KASSERT(fibnum >= 0 && fibnum < rt_numfibs, ("%s: " + "fibnum out of range 0 <= %d < %d", __func__, +fibnum, rt_numfibs)); + } + + KASSERT(rt_key(rt) != NULL, (":%s: rt_key must be supplied", __func__)); + + return (rtsock_routemsg(cmd, ifp, error, rt, fibnum)); +} + +void +rt_newaddrmsg(int cmd, struct ifaddr *ifa, int error, struct rtentry *rt) +{ + + rt_newaddrmsg_fib(cmd, ifa, error, rt, RT_ALL_FIBS); +} + +/* + * This is called to generate messages from the routing socket + * indicating a network interface has had addresses associated with it. + */ +void +rt_newaddrmsg_fib(int cmd, struct ifaddr *ifa, int error, struct rtentry *rt, +int fibnum) +{ + + KASSERT(cmd == RTM_ADD || cmd == RTM_DELETE, + ("unexpected cmd %u", cmd)); + if (fibnum != RT_ALL_FIBS) { + KASSERT(fibnum >= 0 && fibnum < rt_numfibs, ("%s: " + "fibnum out of range 0 <= %d < %d", __func__, +fibnum, rt_numfibs)); + } + +#if defined(INET) || defined(INET6) +#ifdef SCTP + /* +* notify the SCTP stack +* this will only get called when an address is added/deleted +* XXX pass the ifaddr struct instead if ifa->ifa_addr... +*/ + sctp_addr_change(ifa, cmd); +#endif /* SCTP */ +#endif + if (cmd == RTM_ADD) { + rt_addrmsg(cmd, ifa, fibnum); + if (rt != NULL) + rt_routemsg(cmd, ifa->ifa_ifp, error, rt, fibnum); + } else { + if (rt != NULL) + rt_routemsg(cmd, ifa->ifa_ifp, error, rt, fibnum); + rt_addrmsg(cmd, ifa, fibnum); + } +} + Modified: head/sys/net/route.
Re: svn commit: r260486 - head/etc/defaults
If it's one request per nfsd then we should likely scale it separate to ncpu. Otherwise the default for two or core boxes may not be enough. Adrian On Jan 9, 2014 11:29 AM, "Alexander Motin" wrote: > On 09.01.2014 18:18, Adrian Chadd wrote: > >> Depends if you're thinking locally or globally. >> >> Locally - for nfs? not a big deal. >> >> Globally - NFS, ZFS, GELI, geom/cam, NIC, etc.. suddenly your machine >> could default to having a couple thousand worker threads just for a >> HBA and a 10GE NIC. That's a little nuts. >> > > So, what is your point? Each NFS thread (unlike GEOM or CAM) executes only > _one_ request at a time. Would you like your 128-core/many-spindle system > executed only 4 synchronous requests at a time? > > On 9 January 2014 08:00, Alexander Motin wrote: >> >>> On 09.01.2014 17:57, Adrian Chadd wrote: >>> .. so with say, 128 core boxes showing up, is this really a good default? >>> >>> >>> And what is the price? 16K+ of KVA per thread for thread stack, etc? 4 >>> threads is probably much worse default there. May be nfsd's default >>> could be >>> tuned, but obviously it should not be hardcoded value. >>> >>> >>> On 9 January 2014 07:55, Alexander Motin wrote: > > Author: mav > Date: Thu Jan 9 15:55:55 2014 > New Revision: 260486 > URL: http://svnweb.freebsd.org/changeset/base/260486 > > Log: > Remove very low default limit of 4 nfsd threads. nfsd's own > default > is > 8 * hw.ncpu, that sounds more appropriate for these SMP/NCQ/... > days. > > Modified: > head/etc/defaults/rc.conf > > Modified: head/etc/defaults/rc.conf > > > == > --- head/etc/defaults/rc.conf Thu Jan 9 15:38:28 2014 > (r260485) > +++ head/etc/defaults/rc.conf Thu Jan 9 15:55:55 2014 > (r260486) > @@ -311,7 +311,7 @@ nfs_client_enable="NO" # This host is > a >nfs_access_cache="60" # Client cache timeout in seconds >nfs_server_enable="NO" # This host is an NFS server (or NO). >oldnfs_server_enable="NO" # Run the old NFS server (YES/NO). > -nfs_server_flags="-u -t -n 4" # Flags to nfsd (if enabled). > +nfs_server_flags="-u -t" # Flags to nfsd (if enabled). >mountd_enable="NO" # Run mountd (or NO). >mountd_flags="-r" # Flags to mountd (if NFS server > enabled). >weak_mountd_authentication="NO"# Allow non-root mount > requests > to be served. > >>> >>> >>> -- >>> Alexander Motin >>> >> > > -- > Alexander Motin > ___ svn-src-head@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
Re: svn commit: r260486 - head/etc/defaults
On 09.01.2014 18:18, Adrian Chadd wrote: Depends if you're thinking locally or globally. Locally - for nfs? not a big deal. Globally - NFS, ZFS, GELI, geom/cam, NIC, etc.. suddenly your machine could default to having a couple thousand worker threads just for a HBA and a 10GE NIC. That's a little nuts. So, what is your point? Each NFS thread (unlike GEOM or CAM) executes only _one_ request at a time. Would you like your 128-core/many-spindle system executed only 4 synchronous requests at a time? On 9 January 2014 08:00, Alexander Motin wrote: On 09.01.2014 17:57, Adrian Chadd wrote: .. so with say, 128 core boxes showing up, is this really a good default? And what is the price? 16K+ of KVA per thread for thread stack, etc? 4 threads is probably much worse default there. May be nfsd's default could be tuned, but obviously it should not be hardcoded value. On 9 January 2014 07:55, Alexander Motin wrote: Author: mav Date: Thu Jan 9 15:55:55 2014 New Revision: 260486 URL: http://svnweb.freebsd.org/changeset/base/260486 Log: Remove very low default limit of 4 nfsd threads. nfsd's own default is 8 * hw.ncpu, that sounds more appropriate for these SMP/NCQ/... days. Modified: head/etc/defaults/rc.conf Modified: head/etc/defaults/rc.conf == --- head/etc/defaults/rc.conf Thu Jan 9 15:38:28 2014(r260485) +++ head/etc/defaults/rc.conf Thu Jan 9 15:55:55 2014(r260486) @@ -311,7 +311,7 @@ nfs_client_enable="NO" # This host is a nfs_access_cache="60" # Client cache timeout in seconds nfs_server_enable="NO" # This host is an NFS server (or NO). oldnfs_server_enable="NO" # Run the old NFS server (YES/NO). -nfs_server_flags="-u -t -n 4" # Flags to nfsd (if enabled). +nfs_server_flags="-u -t" # Flags to nfsd (if enabled). mountd_enable="NO" # Run mountd (or NO). mountd_flags="-r" # Flags to mountd (if NFS server enabled). weak_mountd_authentication="NO"# Allow non-root mount requests to be served. -- Alexander Motin -- Alexander Motin ___ svn-src-head@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
Re: svn commit: r260486 - head/etc/defaults
Hi, Depends if you're thinking locally or globally. Locally - for nfs? not a big deal. Globally - NFS, ZFS, GELI, geom/cam, NIC, etc.. suddenly your machine could default to having a couple thousand worker threads just for a HBA and a 10GE NIC. That's a little nuts. -a On 9 January 2014 08:00, Alexander Motin wrote: > On 09.01.2014 17:57, Adrian Chadd wrote: >> >> .. so with say, 128 core boxes showing up, is this really a good default? > > > And what is the price? 16K+ of KVA per thread for thread stack, etc? 4 > threads is probably much worse default there. May be nfsd's default could be > tuned, but obviously it should not be hardcoded value. > > >> On 9 January 2014 07:55, Alexander Motin wrote: >>> >>> Author: mav >>> Date: Thu Jan 9 15:55:55 2014 >>> New Revision: 260486 >>> URL: http://svnweb.freebsd.org/changeset/base/260486 >>> >>> Log: >>>Remove very low default limit of 4 nfsd threads. nfsd's own default >>> is >>>8 * hw.ncpu, that sounds more appropriate for these SMP/NCQ/... days. >>> >>> Modified: >>>head/etc/defaults/rc.conf >>> >>> Modified: head/etc/defaults/rc.conf >>> >>> == >>> --- head/etc/defaults/rc.conf Thu Jan 9 15:38:28 2014(r260485) >>> +++ head/etc/defaults/rc.conf Thu Jan 9 15:55:55 2014(r260486) >>> @@ -311,7 +311,7 @@ nfs_client_enable="NO" # This host is a >>> nfs_access_cache="60" # Client cache timeout in seconds >>> nfs_server_enable="NO" # This host is an NFS server (or NO). >>> oldnfs_server_enable="NO" # Run the old NFS server (YES/NO). >>> -nfs_server_flags="-u -t -n 4" # Flags to nfsd (if enabled). >>> +nfs_server_flags="-u -t" # Flags to nfsd (if enabled). >>> mountd_enable="NO" # Run mountd (or NO). >>> mountd_flags="-r" # Flags to mountd (if NFS server >>> enabled). >>> weak_mountd_authentication="NO"# Allow non-root mount requests >>> to be served. > > > > -- > Alexander Motin ___ svn-src-head@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
Re: svn commit: r260486 - head/etc/defaults
On 09.01.2014 17:57, Adrian Chadd wrote: .. so with say, 128 core boxes showing up, is this really a good default? And what is the price? 16K+ of KVA per thread for thread stack, etc? 4 threads is probably much worse default there. May be nfsd's default could be tuned, but obviously it should not be hardcoded value. On 9 January 2014 07:55, Alexander Motin wrote: Author: mav Date: Thu Jan 9 15:55:55 2014 New Revision: 260486 URL: http://svnweb.freebsd.org/changeset/base/260486 Log: Remove very low default limit of 4 nfsd threads. nfsd's own default is 8 * hw.ncpu, that sounds more appropriate for these SMP/NCQ/... days. Modified: head/etc/defaults/rc.conf Modified: head/etc/defaults/rc.conf == --- head/etc/defaults/rc.conf Thu Jan 9 15:38:28 2014(r260485) +++ head/etc/defaults/rc.conf Thu Jan 9 15:55:55 2014(r260486) @@ -311,7 +311,7 @@ nfs_client_enable="NO" # This host is a nfs_access_cache="60" # Client cache timeout in seconds nfs_server_enable="NO" # This host is an NFS server (or NO). oldnfs_server_enable="NO" # Run the old NFS server (YES/NO). -nfs_server_flags="-u -t -n 4" # Flags to nfsd (if enabled). +nfs_server_flags="-u -t" # Flags to nfsd (if enabled). mountd_enable="NO" # Run mountd (or NO). mountd_flags="-r" # Flags to mountd (if NFS server enabled). weak_mountd_authentication="NO"# Allow non-root mount requests to be served. -- Alexander Motin ___ svn-src-head@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
Re: svn commit: r260486 - head/etc/defaults
.. so with say, 128 core boxes showing up, is this really a good default? -a On 9 January 2014 07:55, Alexander Motin wrote: > Author: mav > Date: Thu Jan 9 15:55:55 2014 > New Revision: 260486 > URL: http://svnweb.freebsd.org/changeset/base/260486 > > Log: > Remove very low default limit of 4 nfsd threads. nfsd's own default is > 8 * hw.ncpu, that sounds more appropriate for these SMP/NCQ/... days. > > Modified: > head/etc/defaults/rc.conf > > Modified: head/etc/defaults/rc.conf > == > --- head/etc/defaults/rc.conf Thu Jan 9 15:38:28 2014(r260485) > +++ head/etc/defaults/rc.conf Thu Jan 9 15:55:55 2014(r260486) > @@ -311,7 +311,7 @@ nfs_client_enable="NO" # This host is a > nfs_access_cache="60" # Client cache timeout in seconds > nfs_server_enable="NO" # This host is an NFS server (or NO). > oldnfs_server_enable="NO" # Run the old NFS server (YES/NO). > -nfs_server_flags="-u -t -n 4" # Flags to nfsd (if enabled). > +nfs_server_flags="-u -t" # Flags to nfsd (if enabled). > mountd_enable="NO" # Run mountd (or NO). > mountd_flags="-r" # Flags to mountd (if NFS server enabled). > weak_mountd_authentication="NO"# Allow non-root mount requests to be > served. ___ svn-src-head@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r260486 - head/etc/defaults
Author: mav Date: Thu Jan 9 15:55:55 2014 New Revision: 260486 URL: http://svnweb.freebsd.org/changeset/base/260486 Log: Remove very low default limit of 4 nfsd threads. nfsd's own default is 8 * hw.ncpu, that sounds more appropriate for these SMP/NCQ/... days. Modified: head/etc/defaults/rc.conf Modified: head/etc/defaults/rc.conf == --- head/etc/defaults/rc.conf Thu Jan 9 15:38:28 2014(r260485) +++ head/etc/defaults/rc.conf Thu Jan 9 15:55:55 2014(r260486) @@ -311,7 +311,7 @@ nfs_client_enable="NO" # This host is a nfs_access_cache="60" # Client cache timeout in seconds nfs_server_enable="NO" # This host is an NFS server (or NO). oldnfs_server_enable="NO" # Run the old NFS server (YES/NO). -nfs_server_flags="-u -t -n 4" # Flags to nfsd (if enabled). +nfs_server_flags="-u -t" # Flags to nfsd (if enabled). mountd_enable="NO" # Run mountd (or NO). mountd_flags="-r" # Flags to mountd (if NFS server enabled). weak_mountd_authentication="NO"# Allow non-root mount requests to be served. ___ svn-src-head@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r260484 - head/sbin/kldload
Author: bapt Date: Thu Jan 9 15:35:35 2014 New Revision: 260484 URL: http://svnweb.freebsd.org/changeset/base/260484 Log: Fix error counting Modified: head/sbin/kldload/kldload.c Modified: head/sbin/kldload/kldload.c == --- head/sbin/kldload/kldload.c Thu Jan 9 15:34:23 2014(r260483) +++ head/sbin/kldload/kldload.c Thu Jan 9 15:35:35 2014(r260484) @@ -181,14 +181,13 @@ main(int argc, char** argv) printf("%s is already " "loaded\n", argv[0]); } else { - if (errno == EEXIST) { + if (errno == EEXIST) warnx("can't load %s: module " "already loaded or " "in kernel", argv[0]); - } else { + else warn("can't load %s", argv[0]); - errors++; - } + errors++; } } else { if (verbose) ___ svn-src-head@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r260485 - head/sys/netinet6
Author: ae Date: Thu Jan 9 15:38:28 2014 New Revision: 260485 URL: http://svnweb.freebsd.org/changeset/base/260485 Log: Remove extra nesting from X_ip6_mforward() function. Also remove disabled definitions from ip6_mroute.h. PR: 185148 Sponsored by: Yandex LLC Modified: head/sys/netinet6/ip6_mroute.c head/sys/netinet6/ip6_mroute.h Modified: head/sys/netinet6/ip6_mroute.c == --- head/sys/netinet6/ip6_mroute.c Thu Jan 9 15:35:35 2014 (r260484) +++ head/sys/netinet6/ip6_mroute.c Thu Jan 9 15:38:28 2014 (r260485) @@ -1044,11 +1044,19 @@ socket_send(struct socket *s, struct mbu int X_ip6_mforward(struct ip6_hdr *ip6, struct ifnet *ifp, struct mbuf *m) { + struct rtdetq *rte; + struct mbuf *mb0; struct mf6c *rt; struct mif6 *mifp; struct mbuf *mm; + u_long hash; mifi_t mifi; char ip6bufs[INET6_ADDRSTRLEN], ip6bufd[INET6_ADDRSTRLEN]; +#ifdef UPCALL_TIMING + struct timeval tp; + + GET_TIME(tp); +#endif /* UPCALL_TIMING */ MRT6_DLOG(DEBUG_FORWARD, "src %s, dst %s, ifindex %d", ip6_sprintf(ip6bufs, &ip6->ip6_src), @@ -1096,200 +1104,178 @@ X_ip6_mforward(struct ip6_hdr *ip6, stru if (rt) { MFC6_UNLOCK(); return (ip6_mdq(m, ifp, rt)); - } else { - /* -* If we don't have a route for packet's origin, -* Make a copy of the packet & -* send message to routing daemon -*/ + } - struct mbuf *mb0; - struct rtdetq *rte; - u_long hash; -/* int i, npkts;*/ -#ifdef UPCALL_TIMING - struct timeval tp; + /* +* If we don't have a route for packet's origin, +* Make a copy of the packet & send message to routing daemon. +*/ + MRT6STAT_INC(mrt6s_no_route); + MRT6_DLOG(DEBUG_FORWARD | DEBUG_MFC, "no rte s %s g %s", + ip6_sprintf(ip6bufs, &ip6->ip6_src), + ip6_sprintf(ip6bufd, &ip6->ip6_dst)); - GET_TIME(tp); -#endif /* UPCALL_TIMING */ + /* +* Allocate mbufs early so that we don't do extra work if we +* are just going to fail anyway. +*/ + rte = (struct rtdetq *)malloc(sizeof(*rte), M_MRTABLE6, M_NOWAIT); + if (rte == NULL) { + MFC6_UNLOCK(); + return (ENOBUFS); + } + mb0 = m_copy(m, 0, M_COPYALL); + /* +* Pullup packet header if needed before storing it, +* as other references may modify it in the meantime. +*/ + if (mb0 && (M_HASCL(mb0) || mb0->m_len < sizeof(struct ip6_hdr))) + mb0 = m_pullup(mb0, sizeof(struct ip6_hdr)); + if (mb0 == NULL) { + free(rte, M_MRTABLE6); + MFC6_UNLOCK(); + return (ENOBUFS); + } - MRT6STAT_INC(mrt6s_no_route); - MRT6_DLOG(DEBUG_FORWARD | DEBUG_MFC, "no rte s %s g %s", - ip6_sprintf(ip6bufs, &ip6->ip6_src), - ip6_sprintf(ip6bufd, &ip6->ip6_dst)); + /* is there an upcall waiting for this packet? */ + hash = MF6CHASH(ip6->ip6_src, ip6->ip6_dst); + for (rt = mf6ctable[hash]; rt; rt = rt->mf6c_next) { + if (IN6_ARE_ADDR_EQUAL(&ip6->ip6_src, + &rt->mf6c_origin.sin6_addr) && + IN6_ARE_ADDR_EQUAL(&ip6->ip6_dst, + &rt->mf6c_mcastgrp.sin6_addr) && (rt->mf6c_stall != NULL)) + break; + } - /* -* Allocate mbufs early so that we don't do extra work if we -* are just going to fail anyway. -*/ - rte = (struct rtdetq *)malloc(sizeof(*rte), M_MRTABLE6, - M_NOWAIT); - if (rte == NULL) { + if (rt == NULL) { + struct mrt6msg *im; +#ifdef MRT6_OINIT + struct omrt6msg *oim; +#endif + /* no upcall, so make a new entry */ + rt = (struct mf6c *)malloc(sizeof(*rt), M_MRTABLE6, M_NOWAIT); + if (rt == NULL) { + free(rte, M_MRTABLE6); + m_freem(mb0); MFC6_UNLOCK(); return (ENOBUFS); } - mb0 = m_copy(m, 0, M_COPYALL); /* -* Pullup packet header if needed before storing it, -* as other references may modify it in the meantime. +* Make a copy of the header to send to the user +* level process */ - if (mb0 && - (M_HASCL(mb0) || mb0->m_len < sizeof(struct ip6_hdr))) - mb0 = m_pullup(mb0, sizeof(struct ip6_hdr
svn commit: r260483 - head/sbin/kldload
Author: bapt Date: Thu Jan 9 15:34:23 2014 New Revision: 260483 URL: http://svnweb.freebsd.org/changeset/base/260483 Log: Import error message shown to the user when trying to load a module that is already loaded or compiled withing the kernel MFC after:1 week Modified: head/sbin/kldload/kldload.c Modified: head/sbin/kldload/kldload.c == --- head/sbin/kldload/kldload.c Thu Jan 9 15:31:44 2014(r260482) +++ head/sbin/kldload/kldload.c Thu Jan 9 15:34:23 2014(r260483) @@ -181,8 +181,14 @@ main(int argc, char** argv) printf("%s is already " "loaded\n", argv[0]); } else { - warn("can't load %s", argv[0]); - errors++; + if (errno == EEXIST) { + warnx("can't load %s: module " + "already loaded or " + "in kernel", argv[0]); + } else { + warn("can't load %s", argv[0]); + errors++; + } } } else { if (verbose) ___ svn-src-head@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r260482 - head/sys/netgraph/bluetooth/drivers/ubt
Author: adrian Date: Thu Jan 9 15:31:44 2014 New Revision: 260482 URL: http://svnweb.freebsd.org/changeset/base/260482 Log: Be much more specific (and correct) about the device id matching. These device IDs have an AR3012 bluetooth device that shows up with bcdDevice=1 when it doesn't have the firmware loaded, and bcdDevice=2 when it's ready to speak full HCI. Tested: * AR5B225 PCIe - AR9485 + AR3012 Modified: head/sys/netgraph/bluetooth/drivers/ubt/ng_ubt.c Modified: head/sys/netgraph/bluetooth/drivers/ubt/ng_ubt.c == --- head/sys/netgraph/bluetooth/drivers/ubt/ng_ubt.cThu Jan 9 14:58:06 2014(r260481) +++ head/sys/netgraph/bluetooth/drivers/ubt/ng_ubt.cThu Jan 9 15:31:44 2014(r260482) @@ -399,26 +399,26 @@ static const STRUCT_USB_HOST_ID ubt_igno { USB_VPI(0x03f0, 0x311d, 0) }, /* Atheros 3012 with sflash firmware */ - { USB_VPI(0x0cf3, 0x3004, 0) }, - { USB_VPI(0x0cf3, 0x311d, 0) }, - { USB_VPI(0x13d3, 0x3375, 0) }, - { USB_VPI(0x04ca, 0x3005, 0) }, - { USB_VPI(0x04ca, 0x3006, 0) }, - { USB_VPI(0x04ca, 0x3008, 0) }, - { USB_VPI(0x13d3, 0x3362, 0) }, - { USB_VPI(0x0cf3, 0xe004, 0) }, - { USB_VPI(0x0930, 0x0219, 0) }, - { USB_VPI(0x0489, 0xe057, 0) }, - { USB_VPI(0x13d3, 0x3393, 0) }, - { USB_VPI(0x0489, 0xe04e, 0) }, - { USB_VPI(0x0489, 0xe056, 0) }, + { USB_VPI(0x0cf3, 0x3004, 0), USB_DEV_BCD_LTEQ(1) }, + { USB_VPI(0x0cf3, 0x311d, 0), USB_DEV_BCD_LTEQ(1) }, + { USB_VPI(0x13d3, 0x3375, 0), USB_DEV_BCD_LTEQ(1) }, + { USB_VPI(0x04ca, 0x3005, 0), USB_DEV_BCD_LTEQ(1) }, + { USB_VPI(0x04ca, 0x3006, 0), USB_DEV_BCD_LTEQ(1) }, + { USB_VPI(0x04ca, 0x3008, 0), USB_DEV_BCD_LTEQ(1) }, + { USB_VPI(0x13d3, 0x3362, 0), USB_DEV_BCD_LTEQ(1) }, + { USB_VPI(0x0cf3, 0xe004, 0), USB_DEV_BCD_LTEQ(1) }, + { USB_VPI(0x0930, 0x0219, 0), USB_DEV_BCD_LTEQ(1) }, + { USB_VPI(0x0489, 0xe057, 0), USB_DEV_BCD_LTEQ(1) }, + { USB_VPI(0x13d3, 0x3393, 0), USB_DEV_BCD_LTEQ(1) }, + { USB_VPI(0x0489, 0xe04e, 0), USB_DEV_BCD_LTEQ(1) }, + { USB_VPI(0x0489, 0xe056, 0), USB_DEV_BCD_LTEQ(1) }, /* Atheros AR5BBU12 with sflash firmware */ - { USB_VPI(0x0489, 0xe02c, 0) }, + { USB_VPI(0x0489, 0xe02c, 0), USB_DEV_BCD_LTEQ(1) }, /* Atheros AR5BBU12 with sflash firmware */ - { USB_VPI(0x0489, 0xe03c, 0) }, - { USB_VPI(0x0489, 0xe036, 0) }, + { USB_VPI(0x0489, 0xe03c, 0), USB_DEV_BCD_LTEQ(1) }, + { USB_VPI(0x0489, 0xe036, 0), USB_DEV_BCD_LTEQ(1) }, }; /* List of supported bluetooth devices */ ___ svn-src-head@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
Re: svn commit: r259730 - in head: gnu/lib/csu gnu/lib/libgcc gnu/lib/libstdc++ gnu/lib/libsupc++ lib/atf/libatf-c/tests share/mk sys/conf tools/tools/ath/athstats tools/tools/net80211/wlanstats usr.b
On Thu, 2014-01-09 at 12:58 +0100, Dimitry Andric wrote: > On 09 Jan 2014, at 02:19, Zbigniew Bodek wrote: > > 2013/12/22 Dimitry Andric : > ... > >> Modified: > >> head/gnu/lib/csu/Makefile > >> head/gnu/lib/libgcc/Makefile > >> head/gnu/lib/libstdc++/Makefile > >> head/gnu/lib/libsupc++/Makefile > >> head/lib/atf/libatf-c/tests/Makefile > >> head/share/mk/bsd.sys.mk > >> head/sys/conf/Makefile.arm > >> head/tools/tools/ath/athstats/Makefile > >> head/tools/tools/net80211/wlanstats/Makefile > >> head/usr.bin/mkcsmapper/Makefile.inc > > > > It appears that this change broke backtrace on ARM. > > Now only last stack frame can be displayed in bt. > > However I don't have any fix or suggestion. > > Can you please try reverting each of these Makefiles individually, and > check when the problem disappears? The only effective change caused by > this commit is that some CFLAGS might appear in a different order. > > It might be possible that one of the Makefiles involved is overly > sensitive to this. My first guess would be gnu/lib/csu, and after that > gnu/lib/libgcc. > > -Dimitry > The difference is the compile flags. For example, when building the kernel at r259729: cc -c -O -pipe -std=c99 -g -Wall -Wredundant-decls -Wnested-externs -Wstrict-prototypes -Wmissing-prototypes -Wpointer-arith -Winline -Wcast-qual -Wundef -Wno-pointer-sign -fformat-extensions -Wmissing-include-dirs -fdiagnostics-show-option -Wno-error-tautological-compare -Wno-error-empty-body -Wno-error-parentheses-equality -nostdinc -I. -I/local/build/staging/freebsd/imx53/src/sys -I/local/build/staging/freebsd/imx53/src/sys/contrib/altq -I/local/build/staging/freebsd/imx53/src/sys/contrib/libfdt -D_KERNEL -DHAVE_KERNEL_OPTION_HEADERS -include opt_global.h -funwind-tables -mllvm -arm-enable-ehabi -ffreestanding -Werror /local/build/staging/freebsd/imx53/src/sys/arm/arm/trap.c At r259730: cc -c -O -pipe -std=c99 -g -Wall -Wredundant-decls -Wnested-externs -Wstrict-prototypes -Wmissing-prototypes -Wpointer-arith -Winline -Wcast-qual -Wundef -Wno-pointer-sign -fformat-extensions -Wmissing-include-dirs -fdiagnostics-show-option -Wno-error-tautological-compare -Wno-error-empty-body -Wno-error-parentheses-equality -nostdinc -I. -I/local/build/staging/freebsd/imx53/src/sys -I/local/build/staging/freebsd/imx53/src/sys/contrib/altq -I/local/build/staging/freebsd/imx53/src/sys/contrib/libfdt -D_KERNEL -DHAVE_KERNEL_OPTION_HEADERS -include opt_global.h -funwind-tables -ffreestanding -Werror /local/build/staging/freebsd/imx53/src/sys/arm/arm/trap.c The difference is that the -mllvm and -arm-enable-ehabi flags are missing. -- Ian ___ svn-src-head@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r260481 - head/sys/netinet6
Author: ae Date: Thu Jan 9 14:58:06 2014 New Revision: 260481 URL: http://svnweb.freebsd.org/changeset/base/260481 Log: Add MRT6_DLOG() macro for debugging. Reduce number of MRT6DEBUG ifdefs and fix some broken format strings. MFC after:1 week Sponsored by: Yandex LLC Modified: head/sys/netinet6/ip6_mroute.c Modified: head/sys/netinet6/ip6_mroute.c == --- head/sys/netinet6/ip6_mroute.c Thu Jan 9 11:15:05 2014 (r260480) +++ head/sys/netinet6/ip6_mroute.c Thu Jan 9 14:58:06 2014 (r260481) @@ -221,6 +221,14 @@ static VNET_DEFINE(u_int, mrt6debug) = 0 #define DEBUG_XMIT 0x10 #define DEBUG_REG 0x20 #define DEBUG_PIM 0x40 +#defineDEBUG_ERR 0x80 +#defineDEBUG_ANY 0x7f +#defineMRT6_DLOG(m, fmt, ...) \ + if (V_mrt6debug & (m)) \ + log(((m) & DEBUG_ERR) ? LOG_ERR: LOG_DEBUG, \ + "%s: " fmt "\n", __func__, ##__VA_ARGS__) +#else +#defineMRT6_DLOG(m, fmt, ...) #endif static voidexpire_upcalls(void *); @@ -526,12 +534,8 @@ static int ip6_mrouter_init(struct socket *so, int v, int cmd) { -#ifdef MRT6DEBUG - if (V_mrt6debug) - log(LOG_DEBUG, - "ip6_mrouter_init: so_type = %d, pr_protocol = %d\n", - so->so_type, so->so_proto->pr_protocol); -#endif + MRT6_DLOG(DEBUG_ANY, "so_type = %d, pr_protocol = %d", + so->so_type, so->so_proto->pr_protocol); if (so->so_type != SOCK_RAW || so->so_proto->pr_protocol != IPPROTO_ICMPV6) @@ -560,11 +564,7 @@ ip6_mrouter_init(struct socket *so, int expire_upcalls, NULL); MROUTER6_UNLOCK(); - -#ifdef MRT6DEBUG - if (V_mrt6debug) - log(LOG_DEBUG, "ip6_mrouter_init\n"); -#endif + MRT6_DLOG(DEBUG_ANY, "finished"); return (0); } @@ -642,11 +642,7 @@ X_ip6_mrouter_done(void) V_ip6_mrouter_ver = 0; MROUTER6_UNLOCK(); - -#ifdef MRT6DEBUG - if (V_mrt6debug) - log(LOG_DEBUG, "ip6_mrouter_done\n"); -#endif + MRT6_DLOG(DEBUG_ANY, "finished"); return (0); } @@ -727,14 +723,8 @@ add_m6if(struct mif6ctl *mifcp) nummifs = mifcp->mif6c_mifi + 1; MIF6_UNLOCK(); - -#ifdef MRT6DEBUG - if (V_mrt6debug) - log(LOG_DEBUG, - "add_mif #%d, phyint %s\n", - mifcp->mif6c_mifi, - ifp->if_xname); -#endif + MRT6_DLOG(DEBUG_ANY, "mif #%d, phyint %s", mifcp->mif6c_mifi, + if_name(ifp)); return (0); } @@ -777,11 +767,7 @@ del_m6if_locked(mifi_t *mifip) if (mif6table[mifi - 1].m6_ifp) break; nummifs = mifi; - -#ifdef MRT6DEBUG - if (V_mrt6debug) - log(LOG_DEBUG, "del_m6if %d, nummifs %d\n", *mifip, nummifs); -#endif + MRT6_DLOG(DEBUG_ANY, "mif %d, nummifs %d", *mifip, nummifs); return (0); } @@ -817,15 +803,10 @@ add_m6fc(struct mf6cctl *mfccp) /* If an entry already exists, just update the fields */ if (rt) { -#ifdef MRT6DEBUG - if (V_mrt6debug & DEBUG_MFC) { - log(LOG_DEBUG, - "add_m6fc no upcall h %d o %s g %s p %x\n", - ip6_sprintf(ip6bufo, &mfccp->mf6cc_origin.sin6_addr), - ip6_sprintf(ip6bufg, &mfccp->mf6cc_mcastgrp.sin6_addr), - mfccp->mf6cc_parent); - } -#endif + MRT6_DLOG(DEBUG_MFC, "no upcall o %s g %s p %x", + ip6_sprintf(ip6bufo, &mfccp->mf6cc_origin.sin6_addr), + ip6_sprintf(ip6bufg, &mfccp->mf6cc_mcastgrp.sin6_addr), + mfccp->mf6cc_parent); rt->mf6c_parent = mfccp->mf6cc_parent; rt->mf6c_ifset = mfccp->mf6cc_ifset; @@ -856,16 +837,12 @@ add_m6fc(struct mf6cctl *mfccp) &mfccp->mf6cc_mcastgrp.sin6_addr), mfccp->mf6cc_parent, rt->mf6c_stall); -#ifdef MRT6DEBUG - if (V_mrt6debug & DEBUG_MFC) - log(LOG_DEBUG, - "add_m6fc o %s g %s p %x dbg %x\n", - ip6_sprintf(ip6bufo, - &mfccp->mf6cc_origin.sin6_addr), - ip6_sprintf(ip6bufg, - &mfccp->mf6cc_mcastgrp.sin6_addr), - mfccp->mf6cc_parent, rt->mf6c_stall); -#endif + MRT6_DLOG(DEBUG_MFC, "o %s g %s p %x dbg %p", + ip6_sprintf(ip6bufo, + &mfccp->mf6cc_origin.sin6_addr), + ip6_sprintf(ip6bufg, + &mfccp->mf
Re: svn commit: r259730 - in head: gnu/lib/csu gnu/lib/libgcc gnu/lib/libstdc++ gnu/lib/libsupc++ lib/atf/libatf-c/tests share/mk sys/conf tools/tools/ath/athstats tools/tools/net80211/wlanstats usr.b
On 09 Jan 2014, at 02:19, Zbigniew Bodek wrote: > 2013/12/22 Dimitry Andric : ... >> Modified: >> head/gnu/lib/csu/Makefile >> head/gnu/lib/libgcc/Makefile >> head/gnu/lib/libstdc++/Makefile >> head/gnu/lib/libsupc++/Makefile >> head/lib/atf/libatf-c/tests/Makefile >> head/share/mk/bsd.sys.mk >> head/sys/conf/Makefile.arm >> head/tools/tools/ath/athstats/Makefile >> head/tools/tools/net80211/wlanstats/Makefile >> head/usr.bin/mkcsmapper/Makefile.inc > > It appears that this change broke backtrace on ARM. > Now only last stack frame can be displayed in bt. > However I don't have any fix or suggestion. Can you please try reverting each of these Makefiles individually, and check when the problem disappears? The only effective change caused by this commit is that some CFLAGS might appear in a different order. It might be possible that one of the Makefiles involved is overly sensitive to this. My first guess would be gnu/lib/csu, and after that gnu/lib/libgcc. -Dimitry signature.asc Description: Message signed with OpenPGP using GPGMail
svn commit: r260472 - head/sbin/route
Author: melifaro Date: Thu Jan 9 10:40:36 2014 New Revision: 260472 URL: http://svnweb.freebsd.org/changeset/base/260472 Log: Do some more sanity checks in route(8): require netmask to have the same address family as destination. Found by: jmg MFC after:2 weeks Modified: head/sbin/route/route.c Modified: head/sbin/route/route.c == --- head/sbin/route/route.c Thu Jan 9 09:19:59 2014(r260471) +++ head/sbin/route/route.c Thu Jan 9 10:40:36 2014(r260472) @@ -958,11 +958,18 @@ newroute(int argc, char **argv) } } + /* Do some sanity checks on resulting request */ if (so[RTAX_DST].ss_len == 0) { warnx("destination parameter required"); usage(NULL); } + if (so[RTAX_NETMASK].ss_len != 0 && + so[RTAX_DST].ss_family != so[RTAX_NETMASK].ss_family) { + warnx("destination and netmask family need to be the same"); + usage(NULL); + } + if (nrflags & F_FORCEHOST) { nrflags |= F_ISHOST; #ifdef INET6 ___ svn-src-head@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r260471 - head/sbin/casperd
Author: pjd Date: Thu Jan 9 09:19:59 2014 New Revision: 260471 URL: http://svnweb.freebsd.org/changeset/base/260471 Log: Always create /var/run/casper with correct permissions and don't depend on the calling process' umask. Submitted by: Mikhail Modified: head/sbin/casperd/casperd.c Modified: head/sbin/casperd/casperd.c == --- head/sbin/casperd/casperd.c Thu Jan 9 09:16:35 2014(r260470) +++ head/sbin/casperd/casperd.c Thu Jan 9 09:19:59 2014(r260471) @@ -541,6 +541,7 @@ main_loop(const char *sockpath, struct p struct casper_service *casserv; struct service_connection *sconn, *sconntmp; int lsock, sock, maxfd, ret; + mode_t oldumask; lsock = socket(AF_UNIX, SOCK_STREAM, 0); if (lsock == -1) @@ -554,8 +555,10 @@ main_loop(const char *sockpath, struct p sizeof(sun.sun_path)); sun.sun_len = SUN_LEN(&sun); + oldumask = umask(S_IXUSR | S_IXGRP | S_IXOTH); if (bind(lsock, (struct sockaddr *)&sun, sizeof(sun)) == -1) pjdlog_exit(1, "Unable to bind to %s", sockpath); + (void)umask(oldumask); if (listen(lsock, 8) == -1) pjdlog_exit(1, "Unable to listen on %s", sockpath); ___ svn-src-head@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
Re: svn commit: r260463 - head/sys/dev/usb/wlan
On Thu, Jan 09, 2014 at 01:48:34AM +, Kevin Lo wrote: K> Author: kevlo K> Date: Thu Jan 9 01:48:33 2014 K> New Revision: 260463 K> URL: http://svnweb.freebsd.org/changeset/base/260463 K> K> Log: K> Replace deprecated M_DONTWAIT with M_NOWAIT. K> K> Modified: K> head/sys/dev/usb/wlan/if_rsu.c K> head/sys/dev/usb/wlan/if_urtwn.c K> K> Modified: head/sys/dev/usb/wlan/if_rsu.c K> == K> --- head/sys/dev/usb/wlan/if_rsu.c Thu Jan 9 00:59:03 2014 (r260462) K> +++ head/sys/dev/usb/wlan/if_rsu.c Thu Jan 9 01:48:33 2014 (r260463) K> @@ -1145,11 +1145,11 @@ rsu_event_survey(struct rsu_softc *sc, u K> pktlen = sizeof(*wh) + le32toh(bss->ieslen); K> if (__predict_false(pktlen > MCLBYTES)) K> return; K> -MGETHDR(m, M_DONTWAIT, MT_DATA); K> +MGETHDR(m, M_NOWAIT, MT_DATA); K> if (__predict_false(m == NULL)) K> return; K> if (pktlen > MHLEN) { K> -MCLGET(m, M_DONTWAIT); K> +MCLGET(m, M_NOWAIT); K> if (!(m->m_flags & M_EXT)) { K> m_free(m); K> return; K> @@ -1358,13 +1358,13 @@ rsu_rx_frame(struct rsu_softc *sc, uint8 K> DPRINTFN(5, "Rx frame len=%d rate=%d infosz=%d rssi=%d\n", K> pktlen, rate, infosz, *rssi); K> K> -MGETHDR(m, M_DONTWAIT, MT_DATA); K> +MGETHDR(m, M_NOWAIT, MT_DATA); K> if (__predict_false(m == NULL)) { K> ifp->if_ierrors++; K> return NULL; K> } K> if (pktlen > MHLEN) { K> -MCLGET(m, M_DONTWAIT); K> +MCLGET(m, M_NOWAIT); K> if (__predict_false(!(m->m_flags & M_EXT))) { K> ifp->if_ierrors++; K> m_freem(m); Thanks! The macros MGETHDR, MCLGET should also be converted to functions. -- Totus tuus, Glebius. ___ svn-src-head@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"