svn commit: r283515 - head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs
Author: kib Date: Mon May 25 09:08:19 2015 New Revision: 283515 URL: https://svnweb.freebsd.org/changeset/base/283515 Log: Remove excess Giant acquisition around the dounmount() call. Sponsored by: The FreeBSD Foundation MFC after:1 week Modified: head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/zfs_ioctl.c Modified: head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/zfs_ioctl.c == --- head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/zfs_ioctl.c Mon May 25 08:34:55 2015(r283514) +++ head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/zfs_ioctl.c Mon May 25 09:08:19 2015(r283515) @@ -3481,9 +3481,7 @@ zfs_unmount_snap(const char *snapname) #ifdef illumos (void) dounmount(vfsp, MS_FORCE, kcred); #else - mtx_lock(Giant); /* dounmount() */ (void) dounmount(vfsp, MS_FORCE, curthread); - mtx_unlock(Giant); /* dounmount() */ #endif return (0); } ___ 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: r283479 - in head/sys: amd64/amd64 amd64/ia32 arm/arm i386/i386 mips/mips powerpc/powerpc sparc64/sparc64
On Sun, May 24, 2015 at 10:34:48PM +0300, Chagin Dmitry wrote: On Sun, May 24, 2015 at 09:25:15PM +0300, Konstantin Belousov wrote: On Sun, May 24, 2015 at 05:56:03PM +, Dmitry Chagin wrote: Author: dchagin Date: Sun May 24 17:56:02 2015 New Revision: 283479 URL: https://svnweb.freebsd.org/changeset/base/283479 Log: The kernel sends signals to the processes via ABI specific sv_sendsig method. Native ABI do not need signal conversion, only emulators may want this. Usually emulators implements its own sv_sendsig method. For now only ibcs2 emulator does not have own sv_sendsig implementation and depends on native sendsig() method. So, remove any extra attempts to convert signal numbers from native sendsig() methods except from i386 where ibsc2 is living. Modified: head/sys/amd64/amd64/machdep.c head/sys/amd64/ia32/ia32_signal.c head/sys/arm/arm/machdep.c head/sys/i386/i386/machdep.c head/sys/mips/mips/freebsd32_machdep.c head/sys/mips/mips/pm_machdep.c head/sys/powerpc/powerpc/exec_machdep.c head/sys/sparc64/sparc64/machdep.c Modified: head/sys/amd64/amd64/machdep.c == --- head/sys/amd64/amd64/machdep.cSun May 24 17:53:48 2015 (r283478) +++ head/sys/amd64/amd64/machdep.cSun May 24 17:56:02 2015 (r283479) @@ -398,10 +398,6 @@ sendsig(sig_t catcher, ksiginfo_t *ksi, /* Align to 16 bytes. */ sfp = (struct sigframe *)((unsigned long)sp ~0xFul); - /* Translate the signal if appropriate. */ - if (p-p_sysent-sv_sigtbl sig = p-p_sysent-sv_sigsize) - sig = p-p_sysent-sv_sigtbl[_SIG_IDX(sig)]; - Does struct sysent still need sv_sigtbl and sv_sigsize members then ? ubsc2 share sendsig with i386, so this members still used You mean, iBCS2 uses current FreeBSD signal frame layout ? Indeed, and I do not see how this is not broken. The SysV R3 definitely did not knew about things like SSE or AVX, and I am sure that the layout of the signal frame for i386 even in its non-extended part on FreeBSD mutated without looking back to the iBCS2 requirements. But this is a different issue. That said, could iBCS2 use a wrapper around some internal variant of the sendsig() which would take both ksi and translated signal number ? The signal number would be used to set sf_signum. ___ 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: r283514 - in head: share/man/man4 sys/conf sys/dev/sfxge sys/dev/sfxge/common sys/modules/sfxge
Author: arybchik Date: Mon May 25 08:34:55 2015 New Revision: 283514 URL: https://svnweb.freebsd.org/changeset/base/283514 Log: sfxge: add 7xxx NICs family support Support 7xxx adapters including firmware-assisted TSO and VLAN tagging: - Solarflare Flareon Ultra 7000 series 10/40G adapters: - Solarflare SFN7042Q QSFP+ Server Adapter - Solarflare SFN7142Q QSFP+ Server Adapter - Solarflare Flareon Ultra 7000 series 10G adapters: - Solarflare SFN7022F SFP+ Server Adapter - Solarflare SFN7122F SFP+ Server Adapter - Solarflare SFN7322F Precision Time Synchronization Server Adapter - Solarflare Flareon 7000 series 10G adapters: - Solarflare SFN7002F SFP+ Server Adapter Support utilities to configure adapters and update firmware. The work is done by Solarflare developers (Andy Moreton, Andrew Lee and many others), Artem V. Andreev Artem.Andreev at oktetlabs.ru and me. Sponsored by: Solarflare Communications, Inc. MFC after: 2 weeks Causually read by: gnn Differential Revision: https://reviews.freebsd.org/D2618 Added: head/sys/dev/sfxge/common/ef10_tlv_layout.h (contents, props changed) head/sys/dev/sfxge/common/efx_check.h (contents, props changed) head/sys/dev/sfxge/common/efx_crc32.c (contents, props changed) head/sys/dev/sfxge/common/efx_hash.c (contents, props changed) head/sys/dev/sfxge/common/efx_phy_ids.h (contents, props changed) head/sys/dev/sfxge/common/hunt_ev.c (contents, props changed) head/sys/dev/sfxge/common/hunt_filter.c (contents, props changed) head/sys/dev/sfxge/common/hunt_impl.h (contents, props changed) head/sys/dev/sfxge/common/hunt_intr.c (contents, props changed) head/sys/dev/sfxge/common/hunt_mac.c (contents, props changed) head/sys/dev/sfxge/common/hunt_mcdi.c (contents, props changed) head/sys/dev/sfxge/common/hunt_nic.c (contents, props changed) head/sys/dev/sfxge/common/hunt_nvram.c (contents, props changed) head/sys/dev/sfxge/common/hunt_phy.c (contents, props changed) head/sys/dev/sfxge/common/hunt_rx.c (contents, props changed) head/sys/dev/sfxge/common/hunt_sram.c (contents, props changed) head/sys/dev/sfxge/common/hunt_tx.c (contents, props changed) head/sys/dev/sfxge/common/hunt_vpd.c (contents, props changed) head/sys/dev/sfxge/common/mcdi_mon.c (contents, props changed) head/sys/dev/sfxge/common/mcdi_mon.h (contents, props changed) head/sys/dev/sfxge/common/siena_mcdi.c (contents, props changed) head/sys/dev/sfxge/sfxge_ioc.h (contents, props changed) head/sys/dev/sfxge/sfxge_nvram.c (contents, props changed) Deleted: head/sys/dev/sfxge/common/siena_mon.c Modified: head/share/man/man4/sfxge.4 (contents, props changed) head/sys/conf/files head/sys/conf/files.amd64 head/sys/dev/sfxge/common/efsys.h (contents, props changed) head/sys/dev/sfxge/common/efx.h (contents, props changed) head/sys/dev/sfxge/common/efx_bootcfg.c (contents, props changed) head/sys/dev/sfxge/common/efx_ev.c (contents, props changed) head/sys/dev/sfxge/common/efx_filter.c (contents, props changed) head/sys/dev/sfxge/common/efx_impl.h (contents, props changed) head/sys/dev/sfxge/common/efx_intr.c (contents, props changed) head/sys/dev/sfxge/common/efx_mac.c (contents, props changed) head/sys/dev/sfxge/common/efx_mcdi.c (contents, props changed) head/sys/dev/sfxge/common/efx_mcdi.h (contents, props changed) head/sys/dev/sfxge/common/efx_mon.c (contents, props changed) head/sys/dev/sfxge/common/efx_nic.c (contents, props changed) head/sys/dev/sfxge/common/efx_nvram.c (contents, props changed) head/sys/dev/sfxge/common/efx_phy.c (contents, props changed) head/sys/dev/sfxge/common/efx_port.c (contents, props changed) head/sys/dev/sfxge/common/efx_regs.h (contents, props changed) head/sys/dev/sfxge/common/efx_regs_ef10.h (contents, props changed) head/sys/dev/sfxge/common/efx_regs_mcdi.h (contents, props changed) head/sys/dev/sfxge/common/efx_regs_pci.h (contents, props changed) head/sys/dev/sfxge/common/efx_rx.c (contents, props changed) head/sys/dev/sfxge/common/efx_sram.c (contents, props changed) head/sys/dev/sfxge/common/efx_tx.c (contents, props changed) head/sys/dev/sfxge/common/efx_types.h (contents, props changed) head/sys/dev/sfxge/common/efx_vpd.c (contents, props changed) head/sys/dev/sfxge/common/efx_wol.c (contents, props changed) head/sys/dev/sfxge/common/siena_flash.h (contents, props changed) head/sys/dev/sfxge/common/siena_impl.h (contents, props changed) head/sys/dev/sfxge/common/siena_mac.c (contents, props changed) head/sys/dev/sfxge/common/siena_nic.c (contents, props changed) head/sys/dev/sfxge/common/siena_nvram.c (contents, props changed) head/sys/dev/sfxge/common/siena_phy.c (contents, props changed) head/sys/dev/sfxge/common/siena_sram.c (contents, props
svn commit: r283516 - head/sys/x86/xen
Author: royger Date: Mon May 25 09:47:16 2015 New Revision: 283516 URL: https://svnweb.freebsd.org/changeset/base/283516 Log: xen: make sure xenpv bus is the last to attach This is needed so other buses have a chance of attaching a real ISA bus, if none is found xenpv will attach it. Sponsored by: Citrix Systems RD Modified: head/sys/x86/xen/xenpv.c Modified: head/sys/x86/xen/xenpv.c == --- head/sys/x86/xen/xenpv.cMon May 25 09:08:19 2015(r283515) +++ head/sys/x86/xen/xenpv.cMon May 25 09:47:16 2015(r283516) @@ -35,6 +35,7 @@ __FBSDID($FreeBSD$); #include sys/pcpu.h #include sys/rman.h #include sys/smp.h +#include sys/limits.h #include vm/vm.h #include vm/vm_page.h @@ -72,7 +73,11 @@ xenpv_identify(driver_t *driver, device_ if (devclass_get_device(xenpv_devclass, 0)) return; - if (BUS_ADD_CHILD(parent, 0, xenpv, 0) == NULL) + /* +* The xenpv bus should be the last to attach in order +* to properly detect if an ISA bus has already been added. +*/ + if (BUS_ADD_CHILD(parent, UINT_MAX, xenpv, 0) == NULL) panic(Unable to attach xenpv bus.); } ___ 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: r283517 - head/gnu/lib
Author: markm
Date: Mon May 25 10:51:40 2015
New Revision: 283517
URL: https://svnweb.freebsd.org/changeset/base/283517
Log:
Unbreak build where WANT_GDB == no', as libreadline is also used by ntpdc.
Modified:
head/gnu/lib/Makefile
Modified: head/gnu/lib/Makefile
==
--- head/gnu/lib/Makefile Mon May 25 09:47:16 2015(r283516)
+++ head/gnu/lib/Makefile Mon May 25 10:51:40 2015(r283517)
@@ -16,7 +16,7 @@ SUBDIR+= libssp
SUBDIR+= tests
.endif
-.if ${MK_GDB} != no
+.if ${MK_GDB} != no || ${MK_NTP} != no
SUBDIR+= libreadline
.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: r283511 - head/sys/opencrypto
On 25 May 2015, at 01:31 , Mark Johnston ma...@freebsd.org wrote: Author: markj Date: Mon May 25 01:31:39 2015 New Revision: 283511 URL: https://svnweb.freebsd.org/changeset/base/283511 Log: Use the correct number of arguments for the opencrypto:deflate:deflate_global:bad DTrace probe, which is defined to have Once upon a time, these expended to exactly the same if my memory doesn’t fool me. Thanks for cleaning up though! ___ 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: r283469 - head/sys/compat/linux
On Sunday, May 24, 2015 05:40:14 PM Dmitry Chagin wrote: Author: dchagin Date: Sun May 24 17:40:14 2015 New Revision: 283469 URL: https://svnweb.freebsd.org/changeset/base/283469 Log: As fo_fill_kinfo() does not check fo_fill_kinfo to NULL add a fo_fill_kinfo op to eventfdops. Yes, it is mandatory on purpose. :) Can you add a new KF_* file type (you can just call it eventfd in case we decide to grow a native eventfd)? Can you also add any meaningful data to the record that is exported to userland (e.g. old POSIX semaphores export their current count value) and update libprocstat/procstat/fstat to handle this file type? -- John Baldwin ___ 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: r283364 - head/sys/sys
On Sunday, May 24, 2015 11:24:15 AM Alexander V. Chernikov wrote: Author: melifaro Date: Sun May 24 11:24:14 2015 New Revision: 283364 URL: https://svnweb.freebsd.org/changeset/base/283364 Log: Fix SIOCGI2C structure requirement. In reality, SIOCGI2C ioctl requires struct ifreq as many other ioctls. Doing copyin() on (significantly) larger struct ifstat sometimes triggered EFAULT. Reported by:Olivier Cochard-Labbé olivier at cochard.me MFC after: 1 week Note that this breaks the ABI. You may wish to add a SIOCGI2C_OLD that still uses the larger struct so that old programs continue to work. OTOH, they were only working by accident before. If we expect that no code outside of ifconfig is using this yet, then it is probably ok to let this slide. -- John Baldwin ___ 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: r283331 - head/sys/arm/arm
On Saturday, May 23, 2015 10:28:59 PM Andrew Turner wrote:
Author: andrew
Date: Sat May 23 22:28:59 2015
New Revision: 283331
URL: https://svnweb.freebsd.org/changeset/base/283331
Log:
Use the wait-for-event instruction to put the core we have just enabled
to sleep while it waits to start scheduling. The boot core can then use
the send-event instruction to wake the cores when they should enter the
scheduler.
MFC after: 1 week
Modified:
head/sys/arm/arm/mp_machdep.c
Modified: head/sys/arm/arm/mp_machdep.c
==
--- head/sys/arm/arm/mp_machdep.c Sat May 23 21:58:41 2015
(r283330)
+++ head/sys/arm/arm/mp_machdep.c Sat May 23 22:28:59 2015
(r283331)
@@ -185,8 +185,11 @@ init_secondary(int cpu)
atomic_add_rel_32(mp_naps, 1);
/* Spin until the BSP releases the APs */
- while (!aps_ready)
- ;
+ while (!atomic_load_acq_int(aps_ready)) {
+#if __ARM_ARCH = 7
+ __asm __volatile(wfe);
+#endif
+ }
I don't know that this atomic load acquire is really changing
anything here? Since aps_ready is volatile reading it should
already be atomic on each check around the loop.
/* Initialize curthread */
KASSERT(PCPU_GET(idlethread) != NULL, (no idle thread));
@@ -353,6 +356,10 @@ release_aps(void *dummy __unused)
arm_unmask_irq(i);
}
atomic_store_rel_int(aps_ready, 1);
+ /* Wake the other threads up */
+#if __ARM_ARCH = 7
+ armv7_sev();
+#endif
So I'm not at all familiar with these instructions or what they do,
but are the events level triggered? In particular, is there any
sort of race where the sev might arrive in between the check of
aps_ready and the wfe on an AP? (For example, if wfe/sev were
similar to using mwait on x86 for wfe and a memory write for sev,
x86 would require a call to monitor before doing a check of
aps_ready to handle the race like so:
while (!aps_ready) {
monitor(aps_ready);
if (!aps_ready)
mwait();
}
--
John Baldwin
___
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: r283524 - head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs
Author: avg
Date: Mon May 25 11:48:15 2015
New Revision: 283524
URL: https://svnweb.freebsd.org/changeset/base/283524
Log:
dsl_dataset_promote_check: ensure that shared snaps do not become too long
... after they are transfered from the old origin to the new one.
See: https://www.illumos.org/issues/5909
See: https://reviews.csiden.org/r/219/
Reviewed by: mahrens
MFC after:10 days
Sponsored by: ClusterHQ
Modified:
head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/dsl_dataset.c
Modified: head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/dsl_dataset.c
==
--- head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/dsl_dataset.c Mon May
25 11:36:14 2015(r283523)
+++ head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/dsl_dataset.c Mon May
25 11:48:15 2015(r283524)
@@ -2161,12 +2161,14 @@ dsl_dataset_promote_check(void *arg, dmu
int err;
uint64_t unused;
uint64_t ss_mv_cnt;
+ size_t max_snap_len;
err = promote_hold(ddpa, dp, FTAG);
if (err != 0)
return (err);
hds = ddpa-ddpa_clone;
+ max_snap_len = MAXNAMELEN - strlen(ddpa-ddpa_clonename) - 1;
if (dsl_dataset_phys(hds)-ds_flags DS_FLAG_NOPROMOTE) {
promote_rele(ddpa, FTAG);
@@ -2230,6 +2232,10 @@ dsl_dataset_promote_check(void *arg, dmu
/* Check that the snapshot name does not conflict */
VERIFY0(dsl_dataset_get_snapname(ds));
+ if (strlen(ds-ds_snapname) = max_snap_len) {
+ err = SET_ERROR(ENAMETOOLONG);
+ goto out;
+ }
err = dsl_dataset_snap_lookup(hds, ds-ds_snapname, val);
if (err == 0) {
(void) strcpy(ddpa-err_ds, snap-ds-ds_snapname);
___
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: r282240 - in head: cddl/lib/libdtrace sys/netinet
On 24 May 2015, at 22:50, Julian Elischer wrote: On 5/24/15 9:24 PM, George Neville-Neil wrote: On 23 May 2015, at 17:59, hiren panchasara wrote: On 04/29/15 at 05:19P, George V. Neville-Neil wrote: Author: gnn Date: Wed Apr 29 17:19:55 2015 New Revision: 282240 URL: https://svnweb.freebsd.org/changeset/base/282240 Log: Brief demo script showing the various values that can be read via the new SIFTR statically defined tracepoint (SDT). Differential Revision: https://reviews.freebsd.org/D2387 Reviewed by:bz, markj Modified: head/cddl/lib/libdtrace/tcp.d head/sys/netinet/in_kdtrace.c head/sys/netinet/in_kdtrace.h head/sys/netinet/siftr.c George, This seems useful. Can this be MFC'd to 10? I think it can be. I'll mark this and do it when I'm back from holiday. Nag me if I forget :-) Best, George George.. check out this script to make MFCs to 10 an absolute snap.. instead of being a worrying operation, it's a few keystrokes... http://www.freebsd.org/~julian/MFC-10.sh credits to Bapt for the original (ports) idea. Thanks! Best, George ___ 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: r283331 - head/sys/arm/arm
On Mon, 25 May 2015 07:23:28 -0400
John Baldwin j...@freebsd.org wrote:
On Saturday, May 23, 2015 10:28:59 PM Andrew Turner wrote:
Author: andrew
Date: Sat May 23 22:28:59 2015
New Revision: 283331
URL: https://svnweb.freebsd.org/changeset/base/283331
Log:
Use the wait-for-event instruction to put the core we have just
enabled to sleep while it waits to start scheduling. The boot core
can then use the send-event instruction to wake the cores when they
should enter the scheduler.
MFC after:1 week
Modified:
head/sys/arm/arm/mp_machdep.c
Modified: head/sys/arm/arm/mp_machdep.c
==
--- head/sys/arm/arm/mp_machdep.c Sat May 23 21:58:41
2015(r283330) +++ head/sys/arm/arm/mp_machdep.c Sat
May 23 22:28:59 2015(r283331) @@ -185,8 +185,11 @@
init_secondary(int cpu) atomic_add_rel_32(mp_naps, 1);
/* Spin until the BSP releases the APs */
- while (!aps_ready)
- ;
+ while (!atomic_load_acq_int(aps_ready)) {
+#if __ARM_ARCH = 7
+ __asm __volatile(wfe);
+#endif
+ }
I don't know that this atomic load acquire is really changing
anything here? Since aps_ready is volatile reading it should
already be atomic on each check around the loop.
It's also adding acquire semantics to ensure we don't
incorrectly reorder memory operations across the call.
/* Initialize curthread */
KASSERT(PCPU_GET(idlethread) != NULL, (no idle thread));
@@ -353,6 +356,10 @@ release_aps(void *dummy __unused)
arm_unmask_irq(i);
}
atomic_store_rel_int(aps_ready, 1);
+ /* Wake the other threads up */
+#if __ARM_ARCH = 7
+ armv7_sev();
+#endif
So I'm not at all familiar with these instructions or what they do,
but are the events level triggered? In particular, is there any
sort of race where the sev might arrive in between the check of
aps_ready and the wfe on an AP? (For example, if wfe/sev were
similar to using mwait on x86 for wfe and a memory write for sev,
x86 would require a call to monitor before doing a check of
aps_ready to handle the race like so:
while (!aps_ready) {
monitor(aps_ready);
if (!aps_ready)
mwait();
}
The armv7_sev function includes a barrier to ensure any previous memory
operations have been flushed to cache before we send the event. The sev
instruction then sets the event register in every processor.
The wfe instruction will check this event register and, if it is unset,
it the processor can then enter a low power mode. From my reading of
the documentation, if an event has been signalled before executing the
wfe then the instruction is a nop so will exit the loop as the new
value of aps_ready will be visible on all processors.
Andrew
___
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: r283424 - in head/sys: amd64/linux modules/linux64
On Sunday, May 24, 2015 04:07:12 PM Dmitry Chagin wrote: Author: dchagin Date: Sun May 24 16:07:11 2015 New Revision: 283424 URL: https://svnweb.freebsd.org/changeset/base/283424 Log: Add preliminary support for x86-64 Linux binaries. Differential Revision: https://reviews.freebsd.org/D1076 Added: head/sys/amd64/linux/ head/sys/amd64/linux/Makefile (contents, props changed) head/sys/amd64/linux/linux.h (contents, props changed) head/sys/amd64/linux/linux_dummy.c (contents, props changed) head/sys/amd64/linux/linux_genassym.c (contents, props changed) head/sys/amd64/linux/linux_ipc64.h (contents, props changed) head/sys/amd64/linux/linux_locore.s (contents, props changed) head/sys/amd64/linux/linux_machdep.c (contents, props changed) head/sys/amd64/linux/linux_support.s (contents, props changed) head/sys/amd64/linux/linux_sysvec.c (contents, props changed) head/sys/amd64/linux/linux_vdso.lds.s (contents, props changed) head/sys/amd64/linux/syscalls.conf (contents, props changed) head/sys/amd64/linux/syscalls.master (contents, props changed) head/sys/modules/linux64/ head/sys/modules/linux64/Makefile (contents, props changed) First, I'm really happy to see this land in HEAD! Second, __FreeBSD_version bump perhaps? -- John Baldwin ___ 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: r283525 - head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs
Author: avg
Date: Mon May 25 11:56:57 2015
New Revision: 283525
URL: https://svnweb.freebsd.org/changeset/base/283525
Log:
zfs: fixes for a full stream received into an existing dataset
- this should fail early unless the force flag is set
- if the force flag is set then any local modifications including
snapshots should be undone
See: https://www.illumos.org/issues/5912
See: https://reviews.csiden.org/r/220/
Reviewed by: mahrens, Paul Dagnelie p...@delphix.com
MFC after:15 days
Sponsored by: ClusterHQ
Modified:
head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/dmu_send.c
Modified: head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/dmu_send.c
==
--- head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/dmu_send.c Mon May
25 11:48:15 2015(r283524)
+++ head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/dmu_send.c Mon May
25 11:56:57 2015(r283525)
@@ -983,10 +983,12 @@ recv_begin_check_existing_impl(dmu_recv_
dsl_dataset_rele(snap, FTAG);
} else {
- /* if full, most recent snapshot must be $ORIGIN */
- if (dsl_dataset_phys(ds)-ds_prev_snap_txg = TXG_INITIAL)
- return (SET_ERROR(ENODEV));
- drba-drba_snapobj = dsl_dataset_phys(ds)-ds_prev_snap_obj;
+ /* if full, then must be forced */
+ if (!drba-drba_cookie-drc_force)
+ return (SET_ERROR(EEXIST));
+ /* start from $ORIGIN@$ORIGIN, if supported */
+ drba-drba_snapobj = dp-dp_origin_snap != NULL ?
+ dp-dp_origin_snap-ds_object : 0;
}
return (0);
___
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: r283517 - head/gnu/lib
On 25 May 2015 at 06:51, Mark Murray ma...@freebsd.org wrote: Author: markm Date: Mon May 25 10:51:40 2015 New Revision: 283517 URL: https://svnweb.freebsd.org/changeset/base/283517 Log: Unbreak build where WANT_GDB == no', as libreadline is also used by ntpdc. This is fine as a build fix, but ntpdc links against both libedit and libreadline which seems like a bug. Once that's addressed (dropping the libreadline dependency) we can revert this change. ___ 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: r283526 - in head: . contrib/llvm/include/llvm/Target contrib/llvm/lib/Analysis contrib/llvm/lib/CodeGen contrib/llvm/lib/CodeGen/SelectionDAG contrib/llvm/lib/ExecutionEngine/RuntimeDy...
/R600ISelLowering.cpp head/contrib/llvm/lib/Target/R600/R600Instructions.td head/contrib/llvm/lib/Target/R600/SIAnnotateControlFlow.cpp head/contrib/llvm/lib/Target/R600/SIDefines.h head/contrib/llvm/lib/Target/R600/SIFoldOperands.cpp head/contrib/llvm/lib/Target/R600/SIISelLowering.cpp head/contrib/llvm/lib/Target/R600/SIInsertWaits.cpp head/contrib/llvm/lib/Target/R600/SIInstrFormats.td head/contrib/llvm/lib/Target/R600/SIInstrInfo.cpp head/contrib/llvm/lib/Target/R600/SIInstrInfo.h head/contrib/llvm/lib/Target/R600/SIInstrInfo.td head/contrib/llvm/lib/Target/R600/SIInstructions.td head/contrib/llvm/lib/Target/R600/SILowerControlFlow.cpp head/contrib/llvm/lib/Target/R600/SIRegisterInfo.cpp head/contrib/llvm/lib/Target/R600/SIRegisterInfo.h head/contrib/llvm/lib/Target/R600/SIRegisterInfo.td head/contrib/llvm/lib/Target/R600/VIInstrFormats.td head/contrib/llvm/lib/Target/R600/VIInstructions.td head/contrib/llvm/lib/Target/X86/X86FrameLowering.cpp head/contrib/llvm/lib/Target/X86/X86FrameLowering.h head/contrib/llvm/lib/Target/X86/X86ISelLowering.cpp head/contrib/llvm/lib/Target/X86/X86InstrControl.td head/contrib/llvm/lib/Target/X86/X86IntrinsicsInfo.h head/contrib/llvm/lib/Transforms/Instrumentation/GCOVProfiling.cpp head/contrib/llvm/lib/Transforms/Scalar/GVN.cpp head/contrib/llvm/lib/Transforms/Utils/SimplifyIndVar.cpp head/contrib/llvm/patches/README.TXT head/contrib/llvm/patches/patch-01-freebsd-kprintf.diff head/contrib/llvm/patches/patch-07-llvm-r227752-boot2-shrink.diff head/contrib/llvm/tools/clang/include/clang/Basic/DiagnosticSemaKinds.td head/contrib/llvm/tools/clang/include/clang/Basic/TargetCXXABI.h head/contrib/llvm/tools/clang/include/clang/Basic/TargetInfo.h head/contrib/llvm/tools/clang/include/clang/Sema/Sema.h head/contrib/llvm/tools/clang/include/clang/Sema/Template.h head/contrib/llvm/tools/clang/lib/AST/ASTContext.cpp head/contrib/llvm/tools/clang/lib/AST/ASTDiagnostic.cpp head/contrib/llvm/tools/clang/lib/AST/Decl.cpp head/contrib/llvm/tools/clang/lib/Analysis/ThreadSafety.cpp head/contrib/llvm/tools/clang/lib/Basic/SourceManager.cpp head/contrib/llvm/tools/clang/lib/Basic/TargetInfo.cpp head/contrib/llvm/tools/clang/lib/Basic/Targets.cpp head/contrib/llvm/tools/clang/lib/Basic/Version.cpp head/contrib/llvm/tools/clang/lib/CodeGen/CodeGenModule.cpp head/contrib/llvm/tools/clang/lib/CodeGen/ItaniumCXXABI.cpp head/contrib/llvm/tools/clang/lib/CodeGen/TargetInfo.cpp head/contrib/llvm/tools/clang/lib/Driver/ToolChains.cpp head/contrib/llvm/tools/clang/lib/Frontend/InitPreprocessor.cpp head/contrib/llvm/tools/clang/lib/Sema/DeclSpec.cpp head/contrib/llvm/tools/clang/lib/Sema/SemaChecking.cpp head/contrib/llvm/tools/clang/lib/Sema/SemaExpr.cpp head/contrib/llvm/tools/clang/lib/Sema/SemaExprCXX.cpp head/contrib/llvm/tools/clang/lib/Sema/SemaTemplate.cpp head/etc/mtree/BSD.debug.dist head/etc/mtree/BSD.usr.dist head/lib/clang/include/Makefile head/lib/clang/include/clang/Basic/Version.inc head/lib/clang/include/clang/Config/config.h head/lib/clang/include/llvm/Config/config.h head/lib/clang/include/llvm/Config/llvm-config.h head/lib/libclang_rt/Makefile.inc head/sys/sys/param.h head/tools/build/mk/OptionalObsoleteFiles.inc Directory Properties: head/contrib/llvm/ (props changed) head/contrib/llvm/tools/clang/ (props changed) Modified: head/ObsoleteFiles.inc == --- head/ObsoleteFiles.inc Mon May 25 11:56:57 2015(r283525) +++ head/ObsoleteFiles.inc Mon May 25 13:43:03 2015(r283526) @@ -38,6 +38,67 @@ # xargs -n1 | sort | uniq -d; # done +# 20150525: new clang import which bumps version from 3.6.0 to 3.6.1. +OLD_FILES+=usr/lib/clang/3.6.0/include/__stddef_max_align_t.h +OLD_FILES+=usr/lib/clang/3.6.0/include/__wmmintrin_aes.h +OLD_FILES+=usr/lib/clang/3.6.0/include/__wmmintrin_pclmul.h +OLD_FILES+=usr/lib/clang/3.6.0/include/adxintrin.h +OLD_FILES+=usr/lib/clang/3.6.0/include/altivec.h +OLD_FILES+=usr/lib/clang/3.6.0/include/ammintrin.h +OLD_FILES+=usr/lib/clang/3.6.0/include/arm_acle.h +OLD_FILES+=usr/lib/clang/3.6.0/include/arm_neon.h +OLD_FILES+=usr/lib/clang/3.6.0/include/avx2intrin.h +OLD_FILES+=usr/lib/clang/3.6.0/include/avx512bwintrin.h +OLD_FILES+=usr/lib/clang/3.6.0/include/avx512erintrin.h +OLD_FILES+=usr/lib/clang/3.6.0/include/avx512fintrin.h +OLD_FILES+=usr/lib/clang/3.6.0/include/avx512vlbwintrin.h +OLD_FILES+=usr/lib/clang/3.6.0/include/avx512vlintrin.h +OLD_FILES+=usr/lib/clang/3.6.0/include/avxintrin.h +OLD_FILES+=usr/lib/clang/3.6.0/include/bmi2intrin.h +OLD_FILES+=usr/lib/clang/3.6.0/include/bmiintrin.h +OLD_FILES+=usr/lib/clang/3.6.0/include/cpuid.h +OLD_FILES+=usr/lib/clang/3.6.0/include/emmintrin.h +OLD_FILES+=usr/lib/clang/3.6.0/include/f16cintrin.h +OLD_FILES+=usr/lib/clang/3.6.0/include/fma4intrin.h +OLD_FILES+=usr/lib/clang/3.6.0/include/fmaintrin.h
svn commit: r283527 - in head/sys: dev/ath dev/bwi dev/bwn dev/if_ndis dev/ipw dev/iwi dev/iwn dev/malo dev/mwl dev/ral dev/usb/wlan dev/wi dev/wpi dev/wtap dev/wtap/wtap_hal net80211
Author: glebius Date: Mon May 25 13:51:13 2015 New Revision: 283527 URL: https://svnweb.freebsd.org/changeset/base/283527 Log: Make net80211 drivers supply their device name to the net80211 layer, so that the latter doesn't need to go through struct ifnet to get their name. Sponsored by: Netflix Sponsored by: Nginx, Inc. Modified: head/sys/dev/ath/if_ath.c head/sys/dev/bwi/if_bwi.c head/sys/dev/bwn/if_bwn.c head/sys/dev/if_ndis/if_ndis.c head/sys/dev/ipw/if_ipw.c head/sys/dev/iwi/if_iwi.c head/sys/dev/iwn/if_iwn.c head/sys/dev/malo/if_malo.c head/sys/dev/mwl/if_mwl.c head/sys/dev/ral/rt2560.c head/sys/dev/ral/rt2661.c head/sys/dev/ral/rt2860.c head/sys/dev/usb/wlan/if_rsu.c head/sys/dev/usb/wlan/if_rum.c head/sys/dev/usb/wlan/if_run.c head/sys/dev/usb/wlan/if_uath.c head/sys/dev/usb/wlan/if_upgt.c head/sys/dev/usb/wlan/if_ural.c head/sys/dev/usb/wlan/if_urtw.c head/sys/dev/usb/wlan/if_urtwn.c head/sys/dev/usb/wlan/if_zyd.c head/sys/dev/wi/if_wi.c head/sys/dev/wpi/if_wpi.c head/sys/dev/wtap/if_wtap.c head/sys/dev/wtap/if_wtapvar.h head/sys/dev/wtap/wtap_hal/hal.c head/sys/net80211/ieee80211_var.h Modified: head/sys/dev/ath/if_ath.c == --- head/sys/dev/ath/if_ath.c Mon May 25 13:43:03 2015(r283526) +++ head/sys/dev/ath/if_ath.c Mon May 25 13:51:13 2015(r283527) @@ -592,6 +592,7 @@ ath_attach(u_int16_t devid, struct ath_s goto bad; } ic = ifp-if_l2com; + ic-ic_name = device_get_nameunit(sc-sc_dev); /* set these up early for if_printf use */ if_initname(ifp, device_get_name(sc-sc_dev), Modified: head/sys/dev/bwi/if_bwi.c == --- head/sys/dev/bwi/if_bwi.c Mon May 25 13:43:03 2015(r283526) +++ head/sys/dev/bwi/if_bwi.c Mon May 25 13:51:13 2015(r283527) @@ -507,6 +507,7 @@ bwi_attach(struct bwi_softc *sc) ieee80211_init_channels(ic, NULL, bands); ic-ic_ifp = ifp; + ic-ic_name = device_get_nameunit(dev); ic-ic_caps = IEEE80211_C_STA | IEEE80211_C_SHSLOT | IEEE80211_C_SHPREAMBLE | Modified: head/sys/dev/bwn/if_bwn.c == --- head/sys/dev/bwn/if_bwn.c Mon May 25 13:43:03 2015(r283526) +++ head/sys/dev/bwn/if_bwn.c Mon May 25 13:51:13 2015(r283527) @@ -1058,6 +1058,7 @@ bwn_attach_post(struct bwn_softc *sc) ic = ifp-if_l2com; ic-ic_ifp = ifp; + ic-ic_name = device_get_nameunit(sc-sc_dev); /* XXX not right but it's not used anywhere important */ ic-ic_phytype = IEEE80211_T_OFDM; ic-ic_opmode = IEEE80211_M_STA; Modified: head/sys/dev/if_ndis/if_ndis.c == --- head/sys/dev/if_ndis/if_ndis.c Mon May 25 13:43:03 2015 (r283526) +++ head/sys/dev/if_ndis/if_ndis.c Mon May 25 13:51:13 2015 (r283527) @@ -738,6 +738,7 @@ ndis_attach(dev) ifp-if_ioctl = ndis_ioctl_80211; ic-ic_ifp = ifp; + ic-ic_name = device_get_nameunit(dev); ic-ic_opmode = IEEE80211_M_STA; ic-ic_phytype = IEEE80211_T_DS; ic-ic_caps = IEEE80211_C_8023ENCAP | Modified: head/sys/dev/ipw/if_ipw.c == --- head/sys/dev/ipw/if_ipw.c Mon May 25 13:43:03 2015(r283526) +++ head/sys/dev/ipw/if_ipw.c Mon May 25 13:51:13 2015(r283527) @@ -286,6 +286,7 @@ ipw_attach(device_t dev) IFQ_SET_READY(ifp-if_snd); ic-ic_ifp = ifp; + ic-ic_name = device_get_nameunit(dev); ic-ic_opmode = IEEE80211_M_STA; ic-ic_phytype = IEEE80211_T_DS; Modified: head/sys/dev/iwi/if_iwi.c == --- head/sys/dev/iwi/if_iwi.c Mon May 25 13:43:03 2015(r283526) +++ head/sys/dev/iwi/if_iwi.c Mon May 25 13:51:13 2015(r283527) @@ -364,6 +364,7 @@ iwi_attach(device_t dev) IFQ_SET_READY(ifp-if_snd); ic-ic_ifp = ifp; + ic-ic_name = device_get_nameunit(dev); ic-ic_opmode = IEEE80211_M_STA; ic-ic_phytype = IEEE80211_T_OFDM; /* not only, but not used */ Modified: head/sys/dev/iwn/if_iwn.c == --- head/sys/dev/iwn/if_iwn.c Mon May 25 13:43:03 2015(r283526) +++ head/sys/dev/iwn/if_iwn.c Mon May 25 13:51:13 2015(r283527) @@ -556,6 +556,7 @@ iwn_attach(device_t dev) ic = ifp-if_l2com; ic-ic_ifp = ifp; + ic-ic_name = device_get_nameunit(dev); ic-ic_phytype = IEEE80211_T_OFDM; /* not only, but not
Re: svn commit: r283364 - head/sys/sys
25.05.2015, 14:45, John Baldwin j...@freebsd.org: On Sunday, May 24, 2015 11:24:15 AM Alexander V. Chernikov wrote: Author: melifaro Date: Sun May 24 11:24:14 2015 New Revision: 283364 URL: https://svnweb.freebsd.org/changeset/base/283364 Log: Fix SIOCGI2C structure requirement. In reality, SIOCGI2C ioctl requires struct ifreq as many other ioctls. Doing copyin() on (significantly) larger struct ifstat sometimes triggered EFAULT. Reported by: Olivier Cochard-Labbé olivier at cochard.me MFC after: 1 week Note that this breaks the ABI. You may wish to add a SIOCGI2C_OLD that In general, yes. Since this was not merged to -stable and I'm mostly sure that ifconfig is the only user - this is most likely OK. still uses the larger struct so that old programs continue to work. OTOH, they were only working by accident before. If we expect that no code Well, yes. I'm wondering why this worked at all (copyout tried to get 8XX bytes instead of 32). outside of ifconfig is using this yet, then it is probably ok to let this slide. -- John Baldwin ___ 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: r283529 - head/sys/net80211
Author: glebius
Date: Mon May 25 14:30:44 2015
New Revision: 283529
URL: https://svnweb.freebsd.org/changeset/base/283529
Log:
Use name from ieee80211com instead of parent ifnet, in debugging printfs.
Sponsored by: Netflix
Sponsored by: Nginx, Inc.
Modified:
head/sys/net80211/ieee80211.c
head/sys/net80211/ieee80211_ddb.c
head/sys/net80211/ieee80211_dfs.c
head/sys/net80211/ieee80211_freebsd.c
head/sys/net80211/ieee80211_ht.c
head/sys/net80211/ieee80211_node.c
head/sys/net80211/ieee80211_proto.c
head/sys/net80211/ieee80211_radiotap.c
head/sys/net80211/ieee80211_regdomain.c
head/sys/net80211/ieee80211_var.h
Modified: head/sys/net80211/ieee80211.c
==
--- head/sys/net80211/ieee80211.c Mon May 25 14:12:50 2015
(r283528)
+++ head/sys/net80211/ieee80211.c Mon May 25 14:30:44 2015
(r283529)
@@ -35,9 +35,10 @@ __FBSDID($FreeBSD$);
#include sys/param.h
#include sys/systm.h
#include sys/kernel.h
-
#include sys/socket.h
+#include machine/stdarg.h
+
#include net/if.h
#include net/if_var.h
#include net/if_dl.h
@@ -267,7 +268,20 @@ static void
null_update_chw(struct ieee80211com *ic)
{
- if_printf(ic-ic_ifp, %s: need callback\n, __func__);
+ ic_printf(ic, %s: need callback\n, __func__);
+}
+
+int
+ic_printf(struct ieee80211com *ic, const char * fmt, ...)
+{
+ va_list ap;
+ int retval;
+
+ retval = printf(%s: , ic-ic_name);
+ va_start(ap, fmt);
+ retval += vprintf(fmt, ap);
+ va_end(ap);
+ return (retval);
}
/*
@@ -284,8 +298,8 @@ ieee80211_ifattach(struct ieee80211com *
KASSERT(ifp-if_type == IFT_IEEE80211, (if_type %d, ifp-if_type));
- IEEE80211_LOCK_INIT(ic, ifp-if_xname);
- IEEE80211_TX_LOCK_INIT(ic, ifp-if_xname);
+ IEEE80211_LOCK_INIT(ic, ic-ic_name);
+ IEEE80211_TX_LOCK_INIT(ic, ic-ic_name);
TAILQ_INIT(ic-ic_vaps);
/* Create a taskqueue for all state changes */
@@ -427,7 +441,7 @@ ieee80211_vap_setup(struct ieee80211com
ifp = if_alloc(IFT_ETHER);
if (ifp == NULL) {
- if_printf(ic-ic_ifp, %s: unable to allocate ifnet\n,
+ ic_printf(ic, %s: unable to allocate ifnet\n,
__func__);
return ENOMEM;
}
@@ -551,7 +565,7 @@ ieee80211_vap_attach(struct ieee80211vap
IEEE80211_DPRINTF(vap, IEEE80211_MSG_STATE,
%s: %s parent %s flags 0x%x flags_ext 0x%x\n,
__func__, ieee80211_opmode_name[vap-iv_opmode],
- ic-ic_ifp-if_xname, vap-iv_flags, vap-iv_flags_ext);
+ ic-ic_name, vap-iv_flags, vap-iv_flags_ext);
/*
* Do late attach work that cannot happen until after
@@ -608,7 +622,7 @@ ieee80211_vap_detach(struct ieee80211vap
IEEE80211_DPRINTF(vap, IEEE80211_MSG_STATE, %s: %s parent %s\n,
__func__, ieee80211_opmode_name[vap-iv_opmode],
- ic-ic_ifp-if_xname);
+ ic-ic_name);
/* NB: bpfdetach is called by ether_ifdetach and claims all taps */
ether_ifdetach(ifp);
@@ -900,7 +914,7 @@ int
ieee80211_chan2ieee(struct ieee80211com *ic, const struct ieee80211_channel *c)
{
if (c == NULL) {
- if_printf(ic-ic_ifp, invalid channel (NULL)\n);
+ ic_printf(ic, invalid channel (NULL)\n);
return 0; /* XXX */
}
return (c == IEEE80211_CHAN_ANYC ? IEEE80211_CHAN_ANY : c-ic_ieee);
@@ -1169,7 +1183,6 @@ ieee80211_get_suprates(struct ieee80211c
void
ieee80211_announce(struct ieee80211com *ic)
{
- struct ifnet *ifp = ic-ic_ifp;
int i, rate, mword;
enum ieee80211_phymode mode;
const struct ieee80211_rateset *rs;
@@ -1178,7 +1191,7 @@ ieee80211_announce(struct ieee80211com *
for (mode = IEEE80211_MODE_AUTO+1; mode IEEE80211_MODE_11NA; mode++) {
if (isclr(ic-ic_modecaps, mode))
continue;
- if_printf(ifp, %s rates: , ieee80211_phymode_name[mode]);
+ ic_printf(ic, %s rates: , ieee80211_phymode_name[mode]);
rs = ic-ic_sup_rates[mode];
for (i = 0; i rs-rs_nrates; i++) {
mword = ieee80211_rate2media(ic, rs-rs_rates[i], mode);
Modified: head/sys/net80211/ieee80211_ddb.c
==
--- head/sys/net80211/ieee80211_ddb.c Mon May 25 14:12:50 2015
(r283528)
+++ head/sys/net80211/ieee80211_ddb.c Mon May 25 14:30:44 2015
(r283529)
@@ -521,6 +521,7 @@ _db_show_com(const struct ieee80211com *
db_printf( %s(%p), vap-iv_ifp-if_xname, vap);
db_printf(\n);
db_printf(\tifp %p(%s), ic-ic_ifp, ic-ic_ifp-if_xname);
+ db_printf(\tname %s, ic-ic_name);
db_printf( comlock %p, ic-ic_comlock);
db_printf(\n);
Re: svn commit: r283424 - in head/sys: amd64/linux modules/linux64
On Mon, May 25, 2015 at 07:48:49AM -0400, John Baldwin wrote: On Sunday, May 24, 2015 04:07:12 PM Dmitry Chagin wrote: Author: dchagin Date: Sun May 24 16:07:11 2015 New Revision: 283424 URL: https://svnweb.freebsd.org/changeset/base/283424 Log: Add preliminary support for x86-64 Linux binaries. Differential Revision:https://reviews.freebsd.org/D1076 Added: head/sys/amd64/linux/ head/sys/amd64/linux/Makefile (contents, props changed) head/sys/amd64/linux/linux.h (contents, props changed) head/sys/amd64/linux/linux_dummy.c (contents, props changed) head/sys/amd64/linux/linux_genassym.c (contents, props changed) head/sys/amd64/linux/linux_ipc64.h (contents, props changed) head/sys/amd64/linux/linux_locore.s (contents, props changed) head/sys/amd64/linux/linux_machdep.c (contents, props changed) head/sys/amd64/linux/linux_support.s (contents, props changed) head/sys/amd64/linux/linux_sysvec.c (contents, props changed) head/sys/amd64/linux/linux_vdso.lds.s (contents, props changed) head/sys/amd64/linux/syscalls.conf (contents, props changed) head/sys/amd64/linux/syscalls.master (contents, props changed) head/sys/modules/linux64/ head/sys/modules/linux64/Makefile (contents, props changed) First, I'm really happy to see this land in HEAD! Thanks! Second, __FreeBSD_version bump perhaps? dim@ bumped, is that enough? -- Have fun! chd ___ 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: r283531 - head/sys/net80211
Author: glebius
Date: Mon May 25 15:08:38 2015
New Revision: 283531
URL: https://svnweb.freebsd.org/changeset/base/283531
Log:
Add void * member to struct ieee80211com, so that drivers can obtain their
softc without going through the struct ifnet.
Sponsored by: Netflix
Sponsored by: Nginx, Inc.
Modified:
head/sys/net80211/ieee80211_var.h
Modified: head/sys/net80211/ieee80211_var.h
==
--- head/sys/net80211/ieee80211_var.h Mon May 25 14:54:10 2015
(r283530)
+++ head/sys/net80211/ieee80211_var.h Mon May 25 15:08:38 2015
(r283531)
@@ -117,6 +117,7 @@ struct ieee80211_frame;
struct ieee80211com {
struct ifnet*ic_ifp;/* associated device */
+ void*ic_softc; /* driver softc */
const char *ic_name; /* usually device name */
ieee80211_com_lock_tic_comlock; /* state update lock */
ieee80211_tx_lock_t ic_txlock; /* ic/vap TX lock */
___
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: r283479 - in head/sys: amd64/amd64 amd64/ia32 arm/arm i386/i386 mips/mips powerpc/powerpc sparc64/sparc64
On Mon, May 25, 2015 at 11:23:02AM +0300, Konstantin Belousov wrote: On Sun, May 24, 2015 at 10:34:48PM +0300, Chagin Dmitry wrote: On Sun, May 24, 2015 at 09:25:15PM +0300, Konstantin Belousov wrote: On Sun, May 24, 2015 at 05:56:03PM +, Dmitry Chagin wrote: Author: dchagin Date: Sun May 24 17:56:02 2015 New Revision: 283479 URL: https://svnweb.freebsd.org/changeset/base/283479 Log: The kernel sends signals to the processes via ABI specific sv_sendsig method. Native ABI do not need signal conversion, only emulators may want this. Usually emulators implements its own sv_sendsig method. For now only ibcs2 emulator does not have own sv_sendsig implementation and depends on native sendsig() method. So, remove any extra attempts to convert signal numbers from native sendsig() methods except from i386 where ibsc2 is living. Modified: head/sys/amd64/amd64/machdep.c head/sys/amd64/ia32/ia32_signal.c head/sys/arm/arm/machdep.c head/sys/i386/i386/machdep.c head/sys/mips/mips/freebsd32_machdep.c head/sys/mips/mips/pm_machdep.c head/sys/powerpc/powerpc/exec_machdep.c head/sys/sparc64/sparc64/machdep.c Modified: head/sys/amd64/amd64/machdep.c == --- head/sys/amd64/amd64/machdep.c Sun May 24 17:53:48 2015 (r283478) +++ head/sys/amd64/amd64/machdep.c Sun May 24 17:56:02 2015 (r283479) @@ -398,10 +398,6 @@ sendsig(sig_t catcher, ksiginfo_t *ksi, /* Align to 16 bytes. */ sfp = (struct sigframe *)((unsigned long)sp ~0xFul); - /* Translate the signal if appropriate. */ - if (p-p_sysent-sv_sigtbl sig = p-p_sysent-sv_sigsize) - sig = p-p_sysent-sv_sigtbl[_SIG_IDX(sig)]; - Does struct sysent still need sv_sigtbl and sv_sigsize members then ? ubsc2 share sendsig with i386, so this members still used You mean, iBCS2 uses current FreeBSD signal frame layout ? no, I mean ibsc2 uses i386 sendsig() method, unfortunatelly I know nothing about ibsc2. Indeed, and I do not see how this is not broken. The SysV R3 definitely did not knew about things like SSE or AVX, and I am sure that the layout of the signal frame for i386 even in its non-extended part on FreeBSD mutated without looking back to the iBCS2 requirements. But this is a different issue. That said, could iBCS2 use a wrapper around some internal variant of the sendsig() which would take both ksi and translated signal number ? The signal number would be used to set sf_signum. hm, ok, I write it -- Have fun! chd ___ 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: r283479 - in head/sys: amd64/amd64 amd64/ia32 arm/arm i386/i386 mips/mips powerpc/powerpc sparc64/sparc64
On Mon, May 25, 2015 at 06:48:17PM +0300, Chagin Dmitry wrote: On Mon, May 25, 2015 at 11:23:02AM +0300, Konstantin Belousov wrote: On Sun, May 24, 2015 at 10:34:48PM +0300, Chagin Dmitry wrote: On Sun, May 24, 2015 at 09:25:15PM +0300, Konstantin Belousov wrote: On Sun, May 24, 2015 at 05:56:03PM +, Dmitry Chagin wrote: Author: dchagin Date: Sun May 24 17:56:02 2015 New Revision: 283479 URL: https://svnweb.freebsd.org/changeset/base/283479 Log: The kernel sends signals to the processes via ABI specific sv_sendsig method. Native ABI do not need signal conversion, only emulators may want this. Usually emulators implements its own sv_sendsig method. For now only ibcs2 emulator does not have own sv_sendsig implementation and depends on native sendsig() method. So, remove any extra attempts to convert signal numbers from native sendsig() methods except from i386 where ibsc2 is living. Modified: head/sys/amd64/amd64/machdep.c head/sys/amd64/ia32/ia32_signal.c head/sys/arm/arm/machdep.c head/sys/i386/i386/machdep.c head/sys/mips/mips/freebsd32_machdep.c head/sys/mips/mips/pm_machdep.c head/sys/powerpc/powerpc/exec_machdep.c head/sys/sparc64/sparc64/machdep.c Modified: head/sys/amd64/amd64/machdep.c == --- head/sys/amd64/amd64/machdep.cSun May 24 17:53:48 2015 (r283478) +++ head/sys/amd64/amd64/machdep.cSun May 24 17:56:02 2015 (r283479) @@ -398,10 +398,6 @@ sendsig(sig_t catcher, ksiginfo_t *ksi, /* Align to 16 bytes. */ sfp = (struct sigframe *)((unsigned long)sp ~0xFul); - /* Translate the signal if appropriate. */ - if (p-p_sysent-sv_sigtbl sig = p-p_sysent-sv_sigsize) - sig = p-p_sysent-sv_sigtbl[_SIG_IDX(sig)]; - Does struct sysent still need sv_sigtbl and sv_sigsize members then ? ubsc2 share sendsig with i386, so this members still used You mean, iBCS2 uses current FreeBSD signal frame layout ? no, I mean ibsc2 uses i386 sendsig() method, unfortunatelly What I stated (use of the modern FreeBSD layout for the signal frame when delivered to iBCS2 process) is the consequence of the usage of sendsig(). I know nothing about ibsc2. Indeed, and I do not see how this is not broken. The SysV R3 definitely did not knew about things like SSE or AVX, and I am sure that the layout of the signal frame for i386 even in its non-extended part on FreeBSD mutated without looking back to the iBCS2 requirements. But this is a different issue. That said, could iBCS2 use a wrapper around some internal variant of the sendsig() which would take both ksi and translated signal number ? The signal number would be used to set sf_signum. hm, ok, I write it I am not sure this is would be a fix of anything. iBCS2 should grow its own sendsig(), like svr4 and other ABIs. I tried to find an information for the frame layout on signal delivery for iBCS2, but was unable to get anything. Basically, I need sys/signal.h and machine/sigframe.h from SCO OpenServer, or whatever the files were called there. ___ 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: r283536 - head/sys/dev/wpi
Author: adrian
Date: Mon May 25 17:06:52 2015
New Revision: 283536
URL: https://svnweb.freebsd.org/changeset/base/283536
Log:
Update wpi(4) to use the new mgmt RX API.
Modified:
head/sys/dev/wpi/if_wpi.c
head/sys/dev/wpi/if_wpivar.h
Modified: head/sys/dev/wpi/if_wpi.c
==
--- head/sys/dev/wpi/if_wpi.c Mon May 25 16:37:41 2015(r283535)
+++ head/sys/dev/wpi/if_wpi.c Mon May 25 17:06:52 2015(r283536)
@@ -177,8 +177,9 @@ static int wpi_add_node_entry_adhoc(stru
static struct ieee80211_node *wpi_node_alloc(struct ieee80211vap *,
const uint8_t mac[IEEE80211_ADDR_LEN]);
static voidwpi_node_free(struct ieee80211_node *);
-static voidwpi_recv_mgmt(struct ieee80211_node *, struct mbuf *, int, int,
- int);
+static voidwpi_recv_mgmt(struct ieee80211_node *, struct mbuf *, int,
+ const struct ieee80211_rx_stats *,
+ int, int);
static voidwpi_restore_node(void *, struct ieee80211_node *);
static voidwpi_restore_node_table(struct wpi_softc *, struct wpi_vap *);
static int wpi_newstate(struct ieee80211vap *, enum ieee80211_state, int);
@@ -1693,15 +1694,16 @@ wpi_check_bss_filter(struct wpi_softc *s
}
static void
-wpi_recv_mgmt(struct ieee80211_node *ni, struct mbuf *m, int subtype, int rssi,
-int nf)
+wpi_recv_mgmt(struct ieee80211_node *ni, struct mbuf *m, int subtype,
+const struct ieee80211_rx_stats *rxs,
+int rssi, int nf)
{
struct ieee80211vap *vap = ni-ni_vap;
struct wpi_softc *sc = vap-iv_ic-ic_ifp-if_softc;
struct wpi_vap *wvp = WPI_VAP(vap);
uint64_t ni_tstamp, rx_tstamp;
- wvp-wv_recv_mgmt(ni, m, subtype, rssi, nf);
+ wvp-wv_recv_mgmt(ni, m, subtype, rxs, rssi, nf);
if (vap-iv_opmode == IEEE80211_M_IBSS
vap-iv_state == IEEE80211_S_RUN
Modified: head/sys/dev/wpi/if_wpivar.h
==
--- head/sys/dev/wpi/if_wpivar.hMon May 25 16:37:41 2015
(r283535)
+++ head/sys/dev/wpi/if_wpivar.hMon May 25 17:06:52 2015
(r283536)
@@ -133,7 +133,9 @@ struct wpi_vap {
int (*wv_newstate)(struct ieee80211vap *,
enum ieee80211_state, int);
void(*wv_recv_mgmt)(struct ieee80211_node *,
- struct mbuf *, int, int, int);
+ struct mbuf *, int,
+ const struct ieee80211_rx_stats *,
+ int, int);
};
#defineWPI_VAP(vap)((struct wpi_vap *)(vap))
___
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: r283331 - head/sys/arm/arm
On Monday, May 25, 2015 09:09:26 AM Ian Lepore wrote: On Mon, 2015-05-25 at 10:31 -0400, John Baldwin wrote: Mmmm, does that mean then that you can (conceivably) lose the race the other way where it sees ap_ready's update before it calls wfe and never calls wfe to harvest the event from sev? (In practice I think this is not possible during boot as AP's can't get preempted and there is typically a long time between AP's being signalled to start and start_aps being set. However, this would be a concern for use of wfe/sev for other use cases such as for the cpu_idle hook perhaps?) That's the you must be prepared to handle spurious wakeups part of the sev/wfe contract. The point of WFE is only power-saving, so if your loop spins one time due to an unharvested prior event flag still set, that's deemed harmless. (Userland is allowed to issue SEV instructions, which always target all cores, so there's no expectation of 1:1 relation between sending and waiting.) Ok, good to know. Thanks! -- John Baldwin ___ 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: r283424 - in head/sys: amd64/linux modules/linux64
On Monday, May 25, 2015 06:13:39 PM Slawa Olhovchenkov wrote: On Mon, May 25, 2015 at 10:40:09AM -0400, John Baldwin wrote: Some other related questions are: can we revive print/acroread now and or use a 64-bit flash plugin after these changes? I am still use print/acroread. As I know removing print/acroread irrelevant to linuxator. This is will by security reasson. I am don't open suspicious pdf from suspicious source and just ignore this removing. I mean more if these updates allow us to update to a newer version of print/acroread that would no longer be vulnerable (if such a thing exists) whether 32- or 64-bit. -- John Baldwin ___ 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: r283528 - head/sys/dev/ipw
Author: glebius
Date: Mon May 25 14:12:50 2015
New Revision: 283528
URL: https://svnweb.freebsd.org/changeset/base/283528
Log:
Don't compare array to NULL.
Found by: clang
Modified:
head/sys/dev/ipw/if_ipw.c
Modified: head/sys/dev/ipw/if_ipw.c
==
--- head/sys/dev/ipw/if_ipw.c Mon May 25 13:51:13 2015(r283527)
+++ head/sys/dev/ipw/if_ipw.c Mon May 25 14:12:50 2015(r283528)
@@ -752,11 +752,8 @@ ipw_release(struct ipw_softc *sc)
}
if (sc-tbd_dmat != NULL) {
- if (sc-stbd_list != NULL) {
- bus_dmamap_unload(sc-tbd_dmat, sc-tbd_map);
- bus_dmamem_free(sc-tbd_dmat, sc-tbd_list,
- sc-tbd_map);
- }
+ bus_dmamap_unload(sc-tbd_dmat, sc-tbd_map);
+ bus_dmamem_free(sc-tbd_dmat, sc-tbd_list, sc-tbd_map);
bus_dma_tag_destroy(sc-tbd_dmat);
}
___
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: r283424 - in head/sys: amd64/linux modules/linux64
On 25 May 2015, at 16:04, Chagin Dmitry dcha...@freebsd.org wrote: On Mon, May 25, 2015 at 07:48:49AM -0400, John Baldwin wrote: On Sunday, May 24, 2015 04:07:12 PM Dmitry Chagin wrote: Author: dchagin Date: Sun May 24 16:07:11 2015 New Revision: 283424 URL: https://svnweb.freebsd.org/changeset/base/283424 Log: Add preliminary support for x86-64 Linux binaries. ... Second, __FreeBSD_version bump perhaps? dim@ bumped, is that enough? I've added notes for both r283424 and r283526 to the porter's handbook: https://svnweb.freebsd.org/doc?view=revisionrevision=46725 -Dimitry signature.asc Description: Message signed with OpenPGP using GPGMail
Re: svn commit: r283364 - head/sys/sys
On Monday, May 25, 2015 04:57:09 PM Alexander V. Chernikov wrote: 25.05.2015, 14:45, John Baldwin j...@freebsd.org: On Sunday, May 24, 2015 11:24:15 AM Alexander V. Chernikov wrote: Author: melifaro Date: Sun May 24 11:24:14 2015 New Revision: 283364 URL: https://svnweb.freebsd.org/changeset/base/283364 Log: Fix SIOCGI2C structure requirement. In reality, SIOCGI2C ioctl requires struct ifreq as many other ioctls. Doing copyin() on (significantly) larger struct ifstat sometimes triggered EFAULT. Reported by: Olivier Cochard-Labbé olivier at cochard.me MFC after: 1 week Note that this breaks the ABI. You may wish to add a SIOCGI2C_OLD that In general, yes. Since this was not merged to -stable and I'm mostly sure that ifconfig is the only user - this is most likely OK. Oh, the MFC after threw me off, I didn't realize it was never merged to stable. In that case compat is certainly not needed. still uses the larger struct so that old programs continue to work. OTOH, they were only working by accident before. If we expect that no code Well, yes. I'm wondering why this worked at all (copyout tried to get 8XX bytes instead of 32). Well, if the ifreq was on the stack there was probably space beyond it, so it was just a buffer overrun (in effect). -- John Baldwin ___ 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: r283424 - in head/sys: amd64/linux modules/linux64
On Monday, May 25, 2015 05:04:21 PM Chagin Dmitry wrote: On Mon, May 25, 2015 at 07:48:49AM -0400, John Baldwin wrote: On Sunday, May 24, 2015 04:07:12 PM Dmitry Chagin wrote: Author: dchagin Date: Sun May 24 16:07:11 2015 New Revision: 283424 URL: https://svnweb.freebsd.org/changeset/base/283424 Log: Add preliminary support for x86-64 Linux binaries. Differential Revision: https://reviews.freebsd.org/D1076 Added: head/sys/amd64/linux/ head/sys/amd64/linux/Makefile (contents, props changed) head/sys/amd64/linux/linux.h (contents, props changed) head/sys/amd64/linux/linux_dummy.c (contents, props changed) head/sys/amd64/linux/linux_genassym.c (contents, props changed) head/sys/amd64/linux/linux_ipc64.h (contents, props changed) head/sys/amd64/linux/linux_locore.s (contents, props changed) head/sys/amd64/linux/linux_machdep.c (contents, props changed) head/sys/amd64/linux/linux_support.s (contents, props changed) head/sys/amd64/linux/linux_sysvec.c (contents, props changed) head/sys/amd64/linux/linux_vdso.lds.s (contents, props changed) head/sys/amd64/linux/syscalls.conf (contents, props changed) head/sys/amd64/linux/syscalls.master (contents, props changed) head/sys/modules/linux64/ head/sys/modules/linux64/Makefile (contents, props changed) First, I'm really happy to see this land in HEAD! Thanks! Second, __FreeBSD_version bump perhaps? dim@ bumped, is that enough? That is fine to reuse, but it should probably be explicitly noted in the comments for that version that it is both for clang 3.6.1 and for x86-64 Linux support. I imagine the ports tree will know which versions include your changes. Also, I don't read ports commits since src is enough of a firehouse, so I have a naive question: is the ports tree updated to supply a 64-bit linux-base? Maybe a mail to current@ on how to use a 64-bit base, etc.? Some other related questions are: can we revive print/acroread now and or use a 64-bit flash plugin after these changes? -- John Baldwin ___ 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: r283331 - head/sys/arm/arm
On Monday, May 25, 2015 01:21:48 PM Andrew Turner wrote:
On Mon, 25 May 2015 07:23:28 -0400
John Baldwin j...@freebsd.org wrote:
On Saturday, May 23, 2015 10:28:59 PM Andrew Turner wrote:
Author: andrew
Date: Sat May 23 22:28:59 2015
New Revision: 283331
URL: https://svnweb.freebsd.org/changeset/base/283331
Log:
Use the wait-for-event instruction to put the core we have just
enabled to sleep while it waits to start scheduling. The boot core
can then use the send-event instruction to wake the cores when they
should enter the scheduler.
MFC after: 1 week
Modified:
head/sys/arm/arm/mp_machdep.c
Modified: head/sys/arm/arm/mp_machdep.c
==
--- head/sys/arm/arm/mp_machdep.c Sat May 23 21:58:41
2015 (r283330) +++ head/sys/arm/arm/mp_machdep.c Sat
May 23 22:28:59 2015 (r283331) @@ -185,8 +185,11 @@
init_secondary(int cpu) atomic_add_rel_32(mp_naps, 1);
/* Spin until the BSP releases the APs */
- while (!aps_ready)
- ;
+ while (!atomic_load_acq_int(aps_ready)) {
+#if __ARM_ARCH = 7
+ __asm __volatile(wfe);
+#endif
+ }
I don't know that this atomic load acquire is really changing
anything here? Since aps_ready is volatile reading it should
already be atomic on each check around the loop.
It's also adding acquire semantics to ensure we don't
incorrectly reorder memory operations across the call.
I think the _rel barrier on the update to mp_naps above probably
already does that, but ok.
/* Initialize curthread */
KASSERT(PCPU_GET(idlethread) != NULL, (no idle thread));
@@ -353,6 +356,10 @@ release_aps(void *dummy __unused)
arm_unmask_irq(i);
}
atomic_store_rel_int(aps_ready, 1);
+ /* Wake the other threads up */
+#if __ARM_ARCH = 7
+ armv7_sev();
+#endif
So I'm not at all familiar with these instructions or what they do,
but are the events level triggered? In particular, is there any
sort of race where the sev might arrive in between the check of
aps_ready and the wfe on an AP? (For example, if wfe/sev were
similar to using mwait on x86 for wfe and a memory write for sev,
x86 would require a call to monitor before doing a check of
aps_ready to handle the race like so:
while (!aps_ready) {
monitor(aps_ready);
if (!aps_ready)
mwait();
}
The armv7_sev function includes a barrier to ensure any previous memory
operations have been flushed to cache before we send the event. The sev
instruction then sets the event register in every processor.
The wfe instruction will check this event register and, if it is unset,
it the processor can then enter a low power mode. From my reading of
the documentation, if an event has been signalled before executing the
wfe then the instruction is a nop so will exit the loop as the new
value of aps_ready will be visible on all processors.
Mmmm, does that mean then that you can (conceivably) lose the race the other
way where it sees ap_ready's update before it calls wfe and never calls
wfe to harvest the event from sev? (In practice I think this is not
possible during boot as AP's can't get preempted and there is typically
a long time between AP's being signalled to start and start_aps being
set. However, this would be a concern for use of wfe/sev for other use
cases such as for the cpu_idle hook perhaps?)
--
John Baldwin
___
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: r283533 - head/sys/dev/iwn
Author: glebius
Date: Mon May 25 15:12:23 2015
New Revision: 283533
URL: https://svnweb.freebsd.org/changeset/base/283533
Log:
Use ic_printf() instead of if_printf().
Modified:
head/sys/dev/iwn/if_iwn.c
Modified: head/sys/dev/iwn/if_iwn.c
==
--- head/sys/dev/iwn/if_iwn.c Mon May 25 15:09:17 2015(r283532)
+++ head/sys/dev/iwn/if_iwn.c Mon May 25 15:12:23 2015(r283533)
@@ -2536,8 +2536,7 @@ iwn_setregdomain(struct ieee80211com *ic
channel = iwn_find_eeprom_channel(sc, c);
if (channel == NULL) {
- if_printf(ic-ic_ifp,
- %s: invalid channel %u freq %u/0x%x\n,
+ ic_printf(ic, %s: invalid channel %u freq %u/0x%x\n,
__func__, c-ic_ieee, c-ic_freq, c-ic_flags);
return EINVAL;
}
@@ -4905,7 +4904,7 @@ iwn_watchdog(void *arg)
if (sc-sc_tx_timer 0) {
if (--sc-sc_tx_timer == 0) {
- if_printf(ifp, device timeout\n);
+ ic_printf(ic, device timeout\n);
ieee80211_runtask(ic, sc-sc_reinit_task);
return;
}
___
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: r283331 - head/sys/arm/arm
On Mon, 2015-05-25 at 10:31 -0400, John Baldwin wrote:
On Monday, May 25, 2015 01:21:48 PM Andrew Turner wrote:
On Mon, 25 May 2015 07:23:28 -0400
John Baldwin j...@freebsd.org wrote:
On Saturday, May 23, 2015 10:28:59 PM Andrew Turner wrote:
Author: andrew
Date: Sat May 23 22:28:59 2015
New Revision: 283331
URL: https://svnweb.freebsd.org/changeset/base/283331
Log:
Use the wait-for-event instruction to put the core we have just
enabled to sleep while it waits to start scheduling. The boot core
can then use the send-event instruction to wake the cores when they
should enter the scheduler.
MFC after:1 week
Modified:
head/sys/arm/arm/mp_machdep.c
Modified: head/sys/arm/arm/mp_machdep.c
==
--- head/sys/arm/arm/mp_machdep.c Sat May 23 21:58:41
2015(r283330) +++ head/sys/arm/arm/mp_machdep.c Sat
May 23 22:28:59 2015(r283331) @@ -185,8 +185,11 @@
init_secondary(int cpu) atomic_add_rel_32(mp_naps, 1);
/* Spin until the BSP releases the APs */
- while (!aps_ready)
- ;
+ while (!atomic_load_acq_int(aps_ready)) {
+#if __ARM_ARCH = 7
+ __asm __volatile(wfe);
+#endif
+ }
I don't know that this atomic load acquire is really changing
anything here? Since aps_ready is volatile reading it should
already be atomic on each check around the loop.
It's also adding acquire semantics to ensure we don't
incorrectly reorder memory operations across the call.
I think the _rel barrier on the update to mp_naps above probably
already does that, but ok.
/* Initialize curthread */
KASSERT(PCPU_GET(idlethread) != NULL, (no idle thread));
@@ -353,6 +356,10 @@ release_aps(void *dummy __unused)
arm_unmask_irq(i);
}
atomic_store_rel_int(aps_ready, 1);
+ /* Wake the other threads up */
+#if __ARM_ARCH = 7
+ armv7_sev();
+#endif
So I'm not at all familiar with these instructions or what they do,
but are the events level triggered? In particular, is there any
sort of race where the sev might arrive in between the check of
aps_ready and the wfe on an AP? (For example, if wfe/sev were
similar to using mwait on x86 for wfe and a memory write for sev,
x86 would require a call to monitor before doing a check of
aps_ready to handle the race like so:
while (!aps_ready) {
monitor(aps_ready);
if (!aps_ready)
mwait();
}
The armv7_sev function includes a barrier to ensure any previous memory
operations have been flushed to cache before we send the event. The sev
instruction then sets the event register in every processor.
The wfe instruction will check this event register and, if it is unset,
it the processor can then enter a low power mode. From my reading of
the documentation, if an event has been signalled before executing the
wfe then the instruction is a nop so will exit the loop as the new
value of aps_ready will be visible on all processors.
Mmmm, does that mean then that you can (conceivably) lose the race the other
way where it sees ap_ready's update before it calls wfe and never calls
wfe to harvest the event from sev? (In practice I think this is not
possible during boot as AP's can't get preempted and there is typically
a long time between AP's being signalled to start and start_aps being
set. However, this would be a concern for use of wfe/sev for other use
cases such as for the cpu_idle hook perhaps?)
That's the you must be prepared to handle spurious wakeups part of the
sev/wfe contract. The point of WFE is only power-saving, so if your
loop spins one time due to an unharvested prior event flag still set,
that's deemed harmless. (Userland is allowed to issue SEV instructions,
which always target all cores, so there's no expectation of 1:1 relation
between sending and waiting.)
-- 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: r283535 - in head/sys: dev/ath dev/wi net80211
Author: adrian
Date: Mon May 25 16:37:41 2015
New Revision: 283535
URL: https://svnweb.freebsd.org/changeset/base/283535
Log:
Begin plumbing ieee80211_rx_stats through the receive path.
Smart NICs with firmware (eg wpi, iwn, the new atheros parts, the intel 7260
series, etc) support doing a lot of things in firmware. This includes but
isn't limited to things like scanning, sending probe requests and receiving
probe responses. However, net80211 doesn't know about any of this - it still
drives the whole scan/probe infrastructure itself.
In order to move towards suppoting smart NICs, the receive path needs to
know about the channel/details for each received packet. In at least
the iwn and 7260 firmware (and I believe wpi, but I haven't tried it yet)
it will do the scanning, power-save and off-channel buffering for you -
all you need to do is handle receiving beacons and probe responses on
channels that aren't what you're currently on. However the whole receive
path is peppered with ic-ic_curchan and manual scan/powersave handling.
The beacon parsing code also checks ic-ic_curchan to determine if the
received beacon is on the correct channel or not.[1]
So:
* add freq/ieee values to ieee80211_rx_stats;
* change ieee80211_parse_beacon() to accept the 'current' channel
as an argument;
* modify the iv_input() and iv_recv_mgmt() methods to include the rx_stats;
* add a new method - ieee80211_lookup_channel_rxstats() - that looks up
a channel based on the contents of ieee80211_rx_stats;
* if it exists, use it in the mgmt path to switch the current channel
(which still defaults to ic-ic_curchan) over to something determined
by rx_stats.
This is enough to kick-start scan offload support in the Intel 7260
driver that Rui/I are working on. It also is a good start for scan
offload support for a handful of existing NICs (wpi, iwn, some USB
parts) and it'll very likely dramatically improve stability/performance
there. It's not the whole thing - notably, we don't need to do powersave,
we should not scan all channels, and we should leave probe request sending
to the firmware and not do it ourselves. But, this allows for continued
development on the above features whilst actually having a somewhat
working NIC.
TODO:
* Finish tidying up how the net80211 input path works.
Right now ieee80211_input / ieee80211_input_all act as the top-level
that everything feeds into; it should change so the MIMO input routines
are those and the legacy routines are phased out.
* The band selection should be done by the driver, not by the net80211
layer.
* ieee80211_lookup_channel_rxstats() only determines 11b or 11g channels
for now - this is enough for scanning, but not 100% true in all cases.
If we ever need to handle off-channel scan support for things like
static-40MHz or static-80MHz, or turbo-G, or half/quarter rates,
then we should extend this.
[1] This is a side effect of frequency-hopping and CCK modes - you
can receive beacons when you think you're on a different channel.
In particular, CCK (which is used by the low 11b rates, eg beacons!)
is decodable from adjacent channels - just at a low SNR.
FH is a side effect of having the hardware/firmware do the frequency
hopping - it may pick up beacons transmitted from other FH networks
that are in a different phase of hopping frequencies.
Modified:
head/sys/dev/ath/if_ath_rx.c
head/sys/dev/ath/if_ath_rx.h
head/sys/dev/ath/if_athvar.h
head/sys/dev/wi/if_wi.c
head/sys/dev/wi/if_wivar.h
head/sys/net80211/ieee80211.c
head/sys/net80211/ieee80211_adhoc.c
head/sys/net80211/ieee80211_hostap.c
head/sys/net80211/ieee80211_input.c
head/sys/net80211/ieee80211_input.h
head/sys/net80211/ieee80211_mesh.c
head/sys/net80211/ieee80211_monitor.c
head/sys/net80211/ieee80211_proto.h
head/sys/net80211/ieee80211_sta.c
head/sys/net80211/ieee80211_tdma.c
head/sys/net80211/ieee80211_tdma.h
head/sys/net80211/ieee80211_var.h
head/sys/net80211/ieee80211_wds.c
Modified: head/sys/dev/ath/if_ath_rx.c
==
--- head/sys/dev/ath/if_ath_rx.cMon May 25 15:18:32 2015
(r283534)
+++ head/sys/dev/ath/if_ath_rx.cMon May 25 16:37:41 2015
(r283535)
@@ -327,7 +327,7 @@ ath_legacy_rxbuf_init(struct ath_softc *
*/
void
ath_recv_mgmt(struct ieee80211_node *ni, struct mbuf *m,
- int subtype, int rssi, int nf)
+ int subtype, const struct ieee80211_rx_stats *rxs, int rssi, int nf)
{
struct ieee80211vap *vap = ni-ni_vap;
struct ath_softc *sc = vap-iv_ic-ic_ifp-if_softc;
@@ -353,7 +353,7 @@ ath_recv_mgmt(struct ieee80211_node *ni,
* Call up first so subsequent work can use information
* potentially stored in the node (e.g. for ibss merge).
*/
-
Re: svn commit: r283331 - head/sys/arm/arm
On Mon, 2015-05-25 at 07:23 -0400, John Baldwin wrote:
On Saturday, May 23, 2015 10:28:59 PM Andrew Turner wrote:
Author: andrew
Date: Sat May 23 22:28:59 2015
New Revision: 283331
URL: https://svnweb.freebsd.org/changeset/base/283331
Log:
Use the wait-for-event instruction to put the core we have just enabled
to sleep while it waits to start scheduling. The boot core can then use
the send-event instruction to wake the cores when they should enter the
scheduler.
MFC after:1 week
Modified:
head/sys/arm/arm/mp_machdep.c
Modified: head/sys/arm/arm/mp_machdep.c
==
--- head/sys/arm/arm/mp_machdep.c Sat May 23 21:58:41 2015
(r283330)
+++ head/sys/arm/arm/mp_machdep.c Sat May 23 22:28:59 2015
(r283331)
@@ -185,8 +185,11 @@ init_secondary(int cpu)
atomic_add_rel_32(mp_naps, 1);
/* Spin until the BSP releases the APs */
- while (!aps_ready)
- ;
+ while (!atomic_load_acq_int(aps_ready)) {
+#if __ARM_ARCH = 7
+ __asm __volatile(wfe);
+#endif
+ }
I don't know that this atomic load acquire is really changing
anything here? Since aps_ready is volatile reading it should
already be atomic on each check around the loop.
/* Initialize curthread */
KASSERT(PCPU_GET(idlethread) != NULL, (no idle thread));
@@ -353,6 +356,10 @@ release_aps(void *dummy __unused)
arm_unmask_irq(i);
}
atomic_store_rel_int(aps_ready, 1);
+ /* Wake the other threads up */
+#if __ARM_ARCH = 7
+ armv7_sev();
+#endif
So I'm not at all familiar with these instructions or what they do,
but are the events level triggered? In particular, is there any
sort of race where the sev might arrive in between the check of
aps_ready and the wfe on an AP? (For example, if wfe/sev were
similar to using mwait on x86 for wfe and a memory write for sev,
x86 would require a call to monitor before doing a check of
aps_ready to handle the race like so:
while (!aps_ready) {
monitor(aps_ready);
if (!aps_ready)
mwait();
}
The arm send-event/wait-for-event system includes a 1-bit event latch
per core that indicates whether an event arrived since the prior wait.
The latch is checked atomically by the hardware as part of going into
low-power mode, so you can't get stuck waiting by an event-arrival race,
you just need to be prepared to handle spurious wakeups.
-- 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: r283532 - head/sys/dev/iwn
Author: glebius
Date: Mon May 25 15:09:17 2015
New Revision: 283532
URL: https://svnweb.freebsd.org/changeset/base/283532
Log:
Store softc in ic_softc and access it without using struct ifnet.
Sponsored by: Netflix
Sponsored by: Nginx, Inc.
Modified:
head/sys/dev/iwn/if_iwn.c
Modified: head/sys/dev/iwn/if_iwn.c
==
--- head/sys/dev/iwn/if_iwn.c Mon May 25 15:08:38 2015(r283531)
+++ head/sys/dev/iwn/if_iwn.c Mon May 25 15:09:17 2015(r283532)
@@ -556,6 +556,7 @@ iwn_attach(device_t dev)
ic = ifp-if_l2com;
ic-ic_ifp = ifp;
+ ic-ic_softc = sc;
ic-ic_name = device_get_nameunit(dev);
ic-ic_phytype = IEEE80211_T_OFDM; /* not only, but not used */
ic-ic_opmode = IEEE80211_M_STA;/* default to BSS mode */
@@ -1316,10 +1317,10 @@ iwn_vap_create(struct ieee80211com *ic,
const uint8_t bssid[IEEE80211_ADDR_LEN],
const uint8_t mac[IEEE80211_ADDR_LEN])
{
+ struct iwn_softc *sc = ic-ic_softc;
struct iwn_vap *ivp;
struct ieee80211vap *vap;
uint8_t mac1[IEEE80211_ADDR_LEN];
- struct iwn_softc *sc = ic-ic_ifp-if_softc;
if (!TAILQ_EMPTY(ic-ic_vaps)) /* only one at a time */
return NULL;
@@ -2526,7 +2527,7 @@ static int
iwn_setregdomain(struct ieee80211com *ic, struct ieee80211_regdomain *rd,
int nchan, struct ieee80211_channel chans[])
{
- struct iwn_softc *sc = ic-ic_ifp-if_softc;
+ struct iwn_softc *sc = ic-ic_softc;
int i;
for (i = 0; i nchan; i++) {
@@ -2798,7 +2799,7 @@ iwn_newstate(struct ieee80211vap *vap, e
{
struct iwn_vap *ivp = IWN_VAP(vap);
struct ieee80211com *ic = vap-iv_ic;
- struct iwn_softc *sc = ic-ic_ifp-if_softc;
+ struct iwn_softc *sc = ic-ic_softc;
int error = 0;
DPRINTF(sc, IWN_DEBUG_TRACE, -%s begin\n, __func__);
@@ -5248,7 +5249,7 @@ static int
iwn_updateedca(struct ieee80211com *ic)
{
#define IWN_EXP2(x)((1 (x)) - 1)/* CWmin = 2^ECWmin - 1 */
- struct iwn_softc *sc = ic-ic_ifp-if_softc;
+ struct iwn_softc *sc = ic-ic_softc;
struct iwn_edca_params cmd;
int aci;
@@ -7082,7 +7083,7 @@ iwn_ampdu_rx_start(struct ieee80211_node
int baparamset, int batimeout, int baseqctl)
{
#define MS(_v, _f) (((_v) _f) _f##_S)
- struct iwn_softc *sc = ni-ni_ic-ic_ifp-if_softc;
+ struct iwn_softc *sc = ni-ni_ic-ic_softc;
struct iwn_ops *ops = sc-ops;
struct iwn_node *wn = (void *)ni;
struct iwn_node_info node;
@@ -7118,7 +7119,7 @@ static void
iwn_ampdu_rx_stop(struct ieee80211_node *ni, struct ieee80211_rx_ampdu *rap)
{
struct ieee80211com *ic = ni-ni_ic;
- struct iwn_softc *sc = ic-ic_ifp-if_softc;
+ struct iwn_softc *sc = ic-ic_softc;
struct iwn_ops *ops = sc-ops;
struct iwn_node *wn = (void *)ni;
struct iwn_node_info node;
@@ -7146,7 +7147,7 @@ static int
iwn_addba_request(struct ieee80211_node *ni, struct ieee80211_tx_ampdu *tap,
int dialogtoken, int baparamset, int batimeout)
{
- struct iwn_softc *sc = ni-ni_ic-ic_ifp-if_softc;
+ struct iwn_softc *sc = ni-ni_ic-ic_softc;
int qid;
DPRINTF(sc, IWN_DEBUG_TRACE, -Doing %s\n, __func__);
@@ -7176,7 +7177,7 @@ static int
iwn_addba_response(struct ieee80211_node *ni, struct ieee80211_tx_ampdu *tap,
int code, int baparamset, int batimeout)
{
- struct iwn_softc *sc = ni-ni_ic-ic_ifp-if_softc;
+ struct iwn_softc *sc = ni-ni_ic-ic_softc;
int qid = *(int *)tap-txa_private;
uint8_t tid = tap-txa_tid;
int ret;
@@ -7205,7 +7206,7 @@ iwn_ampdu_tx_start(struct ieee80211com *
uint8_t tid)
{
struct ieee80211_tx_ampdu *tap = ni-ni_tx_ampdu[tid];
- struct iwn_softc *sc = ni-ni_ic-ic_ifp-if_softc;
+ struct iwn_softc *sc = ni-ni_ic-ic_softc;
struct iwn_ops *ops = sc-ops;
struct iwn_node *wn = (void *)ni;
struct iwn_node_info node;
@@ -7239,7 +7240,7 @@ iwn_ampdu_tx_start(struct ieee80211com *
static void
iwn_ampdu_tx_stop(struct ieee80211_node *ni, struct ieee80211_tx_ampdu *tap)
{
- struct iwn_softc *sc = ni-ni_ic-ic_ifp-if_softc;
+ struct iwn_softc *sc = ni-ni_ic-ic_softc;
struct iwn_ops *ops = sc-ops;
uint8_t tid = tap-txa_tid;
int qid;
@@ -8774,8 +8775,8 @@ static void
iwn_scan_curchan(struct ieee80211_scan_state *ss, unsigned long maxdwell)
{
struct ieee80211vap *vap = ss-ss_vap;
- struct iwn_softc *sc = vap-iv_ic-ic_ifp-if_softc;
struct ieee80211com *ic = vap-iv_ic;
+ struct iwn_softc *sc = ic-ic_softc;
int error;
IWN_LOCK(sc);
___
svn-src-head@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to
svn commit: r283530 - head/sys/net80211
Author: glebius Date: Mon May 25 14:54:10 2015 New Revision: 283530 URL: https://svnweb.freebsd.org/changeset/base/283530 Log: Remove unused include. Modified: head/sys/net80211/ieee80211_radiotap.c Modified: head/sys/net80211/ieee80211_radiotap.c == --- head/sys/net80211/ieee80211_radiotap.c Mon May 25 14:30:44 2015 (r283529) +++ head/sys/net80211/ieee80211_radiotap.c Mon May 25 14:54:10 2015 (r283530) @@ -43,7 +43,6 @@ __FBSDID($FreeBSD$); #include net/bpf.h #include net/if.h #include net/if_var.h -#include net/if_llc.h #include net/if_media.h #include net/ethernet.h ___ 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: r283424 - in head/sys: amd64/linux modules/linux64
On Monday, May 25, 2015 10:40:09 AM John Baldwin wrote:
On Monday, May 25, 2015 05:04:21 PM Chagin Dmitry wrote:
That is fine to reuse, but it should probably be explicitly noted in the
comments for that version that it is both for clang 3.6.1 and for x86-64
Linux support. I imagine the ports tree will know which versions include
want to
your changes.
--
John Baldwin
___
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: r283424 - in head/sys: amd64/linux modules/linux64
On Mon, May 25, 2015 at 10:40:09AM -0400, John Baldwin wrote: Some other related questions are: can we revive print/acroread now and or use a 64-bit flash plugin after these changes? I am still use print/acroread. As I know removing print/acroread irrelevant to linuxator. This is will by security reasson. I am don't open suspicious pdf from suspicious source and just ignore this removing. ___ 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: r283424 - in head/sys: amd64/linux modules/linux64
On Mon, May 25, 2015 at 12:32:41PM -0400, John Baldwin wrote: On Monday, May 25, 2015 06:13:39 PM Slawa Olhovchenkov wrote: On Mon, May 25, 2015 at 10:40:09AM -0400, John Baldwin wrote: Some other related questions are: can we revive print/acroread now and or use a 64-bit flash plugin after these changes? I am still use print/acroread. As I know removing print/acroread irrelevant to linuxator. This is will by security reasson. I am don't open suspicious pdf from suspicious source and just ignore this removing. I mean more if these updates allow us to update to a newer version of print/acroread that would no longer be vulnerable (if such a thing exists) whether 32- or 64-bit. Currenly I don't see any linux in Acrobat Reader support OS: https://get.adobe.com/reader/otherversions/ ___ 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: r283537 - in head/sys/dev: ath bwi bwn if_ndis ipw iwi malo mwl ral usb/wlan wi wpi wtap
Author: glebius Date: Mon May 25 18:50:26 2015 New Revision: 283537 URL: https://svnweb.freebsd.org/changeset/base/283537 Log: Set ic_softc in all 802.11 drivers. Not required right now, but will be used quite soon. Sponsored by: Netflix Sponsored by: Nginx, Inc. Modified: head/sys/dev/ath/if_ath.c head/sys/dev/bwi/if_bwi.c head/sys/dev/bwn/if_bwn.c head/sys/dev/if_ndis/if_ndis.c head/sys/dev/ipw/if_ipw.c head/sys/dev/iwi/if_iwi.c head/sys/dev/malo/if_malo.c head/sys/dev/mwl/if_mwl.c head/sys/dev/ral/rt2560.c head/sys/dev/ral/rt2661.c head/sys/dev/ral/rt2860.c head/sys/dev/usb/wlan/if_rsu.c head/sys/dev/usb/wlan/if_rum.c head/sys/dev/usb/wlan/if_run.c head/sys/dev/usb/wlan/if_uath.c head/sys/dev/usb/wlan/if_upgt.c head/sys/dev/usb/wlan/if_ural.c head/sys/dev/usb/wlan/if_urtw.c head/sys/dev/usb/wlan/if_urtwn.c head/sys/dev/usb/wlan/if_zyd.c head/sys/dev/wi/if_wi.c head/sys/dev/wpi/if_wpi.c head/sys/dev/wtap/if_wtap.c Modified: head/sys/dev/ath/if_ath.c == --- head/sys/dev/ath/if_ath.c Mon May 25 17:06:52 2015(r283536) +++ head/sys/dev/ath/if_ath.c Mon May 25 18:50:26 2015(r283537) @@ -592,6 +592,7 @@ ath_attach(u_int16_t devid, struct ath_s goto bad; } ic = ifp-if_l2com; + ic-ic_softc = sc; ic-ic_name = device_get_nameunit(sc-sc_dev); /* set these up early for if_printf use */ Modified: head/sys/dev/bwi/if_bwi.c == --- head/sys/dev/bwi/if_bwi.c Mon May 25 17:06:52 2015(r283536) +++ head/sys/dev/bwi/if_bwi.c Mon May 25 18:50:26 2015(r283537) @@ -507,6 +507,7 @@ bwi_attach(struct bwi_softc *sc) ieee80211_init_channels(ic, NULL, bands); ic-ic_ifp = ifp; + ic-ic_softc = sc; ic-ic_name = device_get_nameunit(dev); ic-ic_caps = IEEE80211_C_STA | IEEE80211_C_SHSLOT | Modified: head/sys/dev/bwn/if_bwn.c == --- head/sys/dev/bwn/if_bwn.c Mon May 25 17:06:52 2015(r283536) +++ head/sys/dev/bwn/if_bwn.c Mon May 25 18:50:26 2015(r283537) @@ -1058,6 +1058,7 @@ bwn_attach_post(struct bwn_softc *sc) ic = ifp-if_l2com; ic-ic_ifp = ifp; + ic-ic_softc = sc; ic-ic_name = device_get_nameunit(sc-sc_dev); /* XXX not right but it's not used anywhere important */ ic-ic_phytype = IEEE80211_T_OFDM; Modified: head/sys/dev/if_ndis/if_ndis.c == --- head/sys/dev/if_ndis/if_ndis.c Mon May 25 17:06:52 2015 (r283536) +++ head/sys/dev/if_ndis/if_ndis.c Mon May 25 18:50:26 2015 (r283537) @@ -738,6 +738,7 @@ ndis_attach(dev) ifp-if_ioctl = ndis_ioctl_80211; ic-ic_ifp = ifp; + ic-ic_softc = sc; ic-ic_name = device_get_nameunit(dev); ic-ic_opmode = IEEE80211_M_STA; ic-ic_phytype = IEEE80211_T_DS; Modified: head/sys/dev/ipw/if_ipw.c == --- head/sys/dev/ipw/if_ipw.c Mon May 25 17:06:52 2015(r283536) +++ head/sys/dev/ipw/if_ipw.c Mon May 25 18:50:26 2015(r283537) @@ -286,6 +286,7 @@ ipw_attach(device_t dev) IFQ_SET_READY(ifp-if_snd); ic-ic_ifp = ifp; + ic-ic_softc = sc; ic-ic_name = device_get_nameunit(dev); ic-ic_opmode = IEEE80211_M_STA; ic-ic_phytype = IEEE80211_T_DS; Modified: head/sys/dev/iwi/if_iwi.c == --- head/sys/dev/iwi/if_iwi.c Mon May 25 17:06:52 2015(r283536) +++ head/sys/dev/iwi/if_iwi.c Mon May 25 18:50:26 2015(r283537) @@ -364,6 +364,7 @@ iwi_attach(device_t dev) IFQ_SET_READY(ifp-if_snd); ic-ic_ifp = ifp; + ic-ic_softc = sc; ic-ic_name = device_get_nameunit(dev); ic-ic_opmode = IEEE80211_M_STA; ic-ic_phytype = IEEE80211_T_OFDM; /* not only, but not used */ Modified: head/sys/dev/malo/if_malo.c == --- head/sys/dev/malo/if_malo.c Mon May 25 17:06:52 2015(r283536) +++ head/sys/dev/malo/if_malo.c Mon May 25 18:50:26 2015(r283537) @@ -276,6 +276,7 @@ malo_attach(uint16_t devid, struct malo_ IFQ_SET_READY(ifp-if_snd); ic-ic_ifp = ifp; + ic-ic_softc = sc; ic-ic_name = device_get_nameunit(sc-malo_dev); /* XXX not right but it's not used anywhere important */ ic-ic_phytype = IEEE80211_T_OFDM; Modified: head/sys/dev/mwl/if_mwl.c == ---
svn commit: r283540 - in head/sys: dev/ath dev/bwi dev/bwn dev/if_ndis dev/iwn dev/malo dev/mwl dev/ral dev/usb/wlan dev/wi dev/wpi dev/wtap net80211
Author: glebius
Date: Mon May 25 19:53:29 2015
New Revision: 283540
URL: https://svnweb.freebsd.org/changeset/base/283540
Log:
Change three methods in struct ieee80211com, namely ic_updateslot,
ic_update_mcast and ic_update_promisc, to pass pointer to the ieee80211com,
not to the ifnet.
Sponsored by: Netflix
Sponsored by: Nginx, Inc.
Modified:
head/sys/dev/ath/if_ath.c
head/sys/dev/bwi/if_bwi.c
head/sys/dev/bwn/if_bwn.c
head/sys/dev/if_ndis/if_ndis.c
head/sys/dev/iwn/if_iwn.c
head/sys/dev/malo/if_malo.c
head/sys/dev/mwl/if_mwl.c
head/sys/dev/ral/rt2560.c
head/sys/dev/ral/rt2661.c
head/sys/dev/ral/rt2860.c
head/sys/dev/usb/wlan/if_rsu.c
head/sys/dev/usb/wlan/if_rum.c
head/sys/dev/usb/wlan/if_run.c
head/sys/dev/usb/wlan/if_uath.c
head/sys/dev/usb/wlan/if_upgt.c
head/sys/dev/usb/wlan/if_ural.c
head/sys/dev/usb/wlan/if_urtw.c
head/sys/dev/usb/wlan/if_urtwn.c
head/sys/dev/usb/wlan/if_zyd.c
head/sys/dev/wi/if_wi.c
head/sys/dev/wpi/if_wpi.c
head/sys/dev/wtap/if_wtap.c
head/sys/net80211/ieee80211.c
head/sys/net80211/ieee80211_proto.c
head/sys/net80211/ieee80211_var.h
Modified: head/sys/dev/ath/if_ath.c
==
--- head/sys/dev/ath/if_ath.c Mon May 25 19:48:48 2015(r283539)
+++ head/sys/dev/ath/if_ath.c Mon May 25 19:53:29 2015(r283540)
@@ -166,9 +166,9 @@ static void ath_bmiss_proc(void *, int);
static voidath_key_update_begin(struct ieee80211vap *);
static voidath_key_update_end(struct ieee80211vap *);
static voidath_update_mcast_hw(struct ath_softc *);
-static voidath_update_mcast(struct ifnet *);
-static voidath_update_promisc(struct ifnet *);
-static voidath_updateslot(struct ifnet *);
+static voidath_update_mcast(struct ieee80211com *);
+static voidath_update_promisc(struct ieee80211com *);
+static voidath_updateslot(struct ieee80211com *);
static voidath_bstuck_proc(void *, int);
static voidath_reset_proc(void *, int);
static int ath_desc_alloc(struct ath_softc *);
@@ -3547,9 +3547,9 @@ ath_key_update_end(struct ieee80211vap *
}
static void
-ath_update_promisc(struct ifnet *ifp)
+ath_update_promisc(struct ieee80211com *ic)
{
- struct ath_softc *sc = ifp-if_softc;
+ struct ath_softc *sc = ic-ic_softc;
u_int32_t rfilt;
/* configure rx filter */
@@ -3611,9 +3611,9 @@ ath_update_mcast_hw(struct ath_softc *sc
* awake before operating.
*/
static void
-ath_update_mcast(struct ifnet *ifp)
+ath_update_mcast(struct ieee80211com *ic)
{
- struct ath_softc *sc = ifp-if_softc;
+ struct ath_softc *sc = ic-ic_softc;
ATH_LOCK(sc);
ath_power_set_power_state(sc, HAL_PM_AWAKE);
@@ -3697,10 +3697,9 @@ ath_setslottime(struct ath_softc *sc)
* slot time based on the current setting.
*/
static void
-ath_updateslot(struct ifnet *ifp)
+ath_updateslot(struct ieee80211com *ic)
{
- struct ath_softc *sc = ifp-if_softc;
- struct ieee80211com *ic = ifp-if_l2com;
+ struct ath_softc *sc = ic-ic_softc;
/*
* When not coordinating the BSS, change the hardware
Modified: head/sys/dev/bwi/if_bwi.c
==
--- head/sys/dev/bwi/if_bwi.c Mon May 25 19:48:48 2015(r283539)
+++ head/sys/dev/bwi/if_bwi.c Mon May 25 19:53:29 2015(r283540)
@@ -113,7 +113,7 @@ static void bwi_scan_start(struct ieee80
static voidbwi_set_channel(struct ieee80211com *);
static voidbwi_scan_end(struct ieee80211com *);
static int bwi_newstate(struct ieee80211vap *, enum ieee80211_state, int);
-static voidbwi_updateslot(struct ifnet *);
+static voidbwi_updateslot(struct ieee80211com *);
static int bwi_media_change(struct ifnet *);
static voidbwi_calibrate(void *);
@@ -3735,14 +3735,13 @@ bwi_set_bssid(struct bwi_softc *sc, cons
}
static void
-bwi_updateslot(struct ifnet *ifp)
+bwi_updateslot(struct ieee80211com *ic)
{
- struct bwi_softc *sc = ifp-if_softc;
- struct ieee80211com *ic = ifp-if_l2com;
+ struct bwi_softc *sc = ic-ic_softc;
struct bwi_mac *mac;
BWI_LOCK(sc);
- if (ifp-if_drv_flags IFF_DRV_RUNNING) {
+ if (ic-ic_ifp-if_drv_flags IFF_DRV_RUNNING) {
DPRINTF(sc, BWI_DBG_80211, %s\n, __func__);
KASSERT(sc-sc_cur_regwin-rw_type == BWI_REGWIN_T_MAC,
Modified: head/sys/dev/bwn/if_bwn.c
==
--- head/sys/dev/bwn/if_bwn.c Mon May 25 19:48:48 2015(r283539)
+++ head/sys/dev/bwn/if_bwn.c Mon May 25 19:53:29 2015(r283540)
@@ -181,8 +181,8 @@ static void bwn_addchannels(struct ieee8
const struct bwn_channelinfo *, int);
static int bwn_raw_xmit(struct ieee80211_node *, struct mbuf *,
const struct ieee80211_bpf_params
RE: svn commit: r280849 - in head: contrib/ntp contrib/ntp/adjtimedcontrib/ntp/arlib contrib/ntp/clockstuff contrib/ntp/conf contrib/ntp/htmlcontrib/ntp/html/build contrib/ntp/html/drivers contrib/nt.
Thanks bapt@. I will take a look after work tonight. ~Cy -Original Message- From: Baptiste Daroussin Sent: 25/05/2015 12:45 To: Cy Schubert Cc: src-committ...@freebsd.org; svn-src-...@freebsd.org; svn-src-head@freebsd.org Subject: Re: svn commit: r280849 - in head: contrib/ntp contrib/ntp/adjtimedcontrib/ntp/arlib contrib/ntp/clockstuff contrib/ntp/conf contrib/ntp/htmlcontrib/ntp/html/build contrib/ntp/html/drivers contrib/nt... On Mon, Mar 30, 2015 at 01:30:17PM +, Cy Schubert wrote: Author: cy Date: Mon Mar 30 13:30:15 2015 New Revision: 280849 URL: https://svnweb.freebsd.org/changeset/base/280849 Log: MFV ntp 4.2.8p1 (r258945, r275970, r276091, r276092, r276093, r278284) Thanks to roberto for providing pointers to wedge this into HEAD. Approved by:roberto Thanks for the update, but you have reintroduced overlinking of useless libraries all over ntp. for example with ntpdc: The binary does not need at all to link with ncurses explicitly The binary does not need to link with both libedit and readline (libedit is enough and prefered) The binary does not need to link to libmd as it does not use it. The binary needs libcrypto but absolutly not to libssl Also ntpdc does not build if WITHOUT_OPENSSL because in the config.h it is hardcoded that it needs openssl, for example: #define USE_OPENSSL_CRYPTO_RAND 1 Before the update, libmd was used for the digests functions. gGven now it is hardcoded that openssl is required libmd should be replaced by libcrypto but you should not have both. I'm working on fixing the overlink, but given I know nothing about libntp's code I will let others have a look on what to do with the MK_OPENSSL. Best regards, Bapt ___ 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: r283539 - head/sys/net80211
Author: glebius
Date: Mon May 25 19:48:48 2015
New Revision: 283539
URL: https://svnweb.freebsd.org/changeset/base/283539
Log:
When sending RTM_IEEE80211_CSA, RTM_IEEE80211_RADAR, RTM_IEEE80211_CAC and
RTM_IEEE80211_RADIO routing messages, broadcast them on all vap interfaces
instead of sending them on parent.
Reviewed by: adrian
Sponsored by: Netflix
Sponsored by: Nginx, Inc.
Modified:
head/sys/net80211/ieee80211_freebsd.c
Modified: head/sys/net80211/ieee80211_freebsd.c
==
--- head/sys/net80211/ieee80211_freebsd.c Mon May 25 19:18:16 2015
(r283538)
+++ head/sys/net80211/ieee80211_freebsd.c Mon May 25 19:48:48 2015
(r283539)
@@ -694,8 +694,9 @@ void
ieee80211_notify_csa(struct ieee80211com *ic,
const struct ieee80211_channel *c, int mode, int count)
{
- struct ifnet *ifp = ic-ic_ifp;
struct ieee80211_csa_event iev;
+ struct ieee80211vap *vap;
+ struct ifnet *ifp;
memset(iev, 0, sizeof(iev));
iev.iev_flags = c-ic_flags;
@@ -703,42 +704,53 @@ ieee80211_notify_csa(struct ieee80211com
iev.iev_ieee = c-ic_ieee;
iev.iev_mode = mode;
iev.iev_count = count;
- CURVNET_SET(ifp-if_vnet);
- rt_ieee80211msg(ifp, RTM_IEEE80211_CSA, iev, sizeof(iev));
- CURVNET_RESTORE();
+ TAILQ_FOREACH(vap, ic-ic_vaps, iv_next) {
+ ifp = vap-iv_ifp;
+ CURVNET_SET(ifp-if_vnet);
+ rt_ieee80211msg(ifp, RTM_IEEE80211_CSA, iev, sizeof(iev));
+ CURVNET_RESTORE();
+ }
}
void
ieee80211_notify_radar(struct ieee80211com *ic,
const struct ieee80211_channel *c)
{
- struct ifnet *ifp = ic-ic_ifp;
struct ieee80211_radar_event iev;
+ struct ieee80211vap *vap;
+ struct ifnet *ifp;
memset(iev, 0, sizeof(iev));
iev.iev_flags = c-ic_flags;
iev.iev_freq = c-ic_freq;
iev.iev_ieee = c-ic_ieee;
- CURVNET_SET(ifp-if_vnet);
- rt_ieee80211msg(ifp, RTM_IEEE80211_RADAR, iev, sizeof(iev));
- CURVNET_RESTORE();
+ TAILQ_FOREACH(vap, ic-ic_vaps, iv_next) {
+ ifp = vap-iv_ifp;
+ CURVNET_SET(ifp-if_vnet);
+ rt_ieee80211msg(ifp, RTM_IEEE80211_RADAR, iev, sizeof(iev));
+ CURVNET_RESTORE();
+ }
}
void
ieee80211_notify_cac(struct ieee80211com *ic,
const struct ieee80211_channel *c, enum ieee80211_notify_cac_event type)
{
- struct ifnet *ifp = ic-ic_ifp;
struct ieee80211_cac_event iev;
+ struct ieee80211vap *vap;
+ struct ifnet *ifp;
memset(iev, 0, sizeof(iev));
iev.iev_flags = c-ic_flags;
iev.iev_freq = c-ic_freq;
iev.iev_ieee = c-ic_ieee;
iev.iev_type = type;
- CURVNET_SET(ifp-if_vnet);
- rt_ieee80211msg(ifp, RTM_IEEE80211_CAC, iev, sizeof(iev));
- CURVNET_RESTORE();
+ TAILQ_FOREACH(vap, ic-ic_vaps, iv_next) {
+ ifp = vap-iv_ifp;
+ CURVNET_SET(ifp-if_vnet);
+ rt_ieee80211msg(ifp, RTM_IEEE80211_CAC, iev, sizeof(iev));
+ CURVNET_RESTORE();
+ }
}
void
@@ -782,14 +794,18 @@ ieee80211_notify_country(struct ieee8021
void
ieee80211_notify_radio(struct ieee80211com *ic, int state)
{
- struct ifnet *ifp = ic-ic_ifp;
struct ieee80211_radio_event iev;
+ struct ieee80211vap *vap;
+ struct ifnet *ifp;
memset(iev, 0, sizeof(iev));
iev.iev_state = state;
- CURVNET_SET(ifp-if_vnet);
- rt_ieee80211msg(ifp, RTM_IEEE80211_RADIO, iev, sizeof(iev));
- CURVNET_RESTORE();
+ TAILQ_FOREACH(vap, ic-ic_vaps, iv_next) {
+ ifp = vap-iv_ifp;
+ CURVNET_SET(ifp-if_vnet);
+ rt_ieee80211msg(ifp, RTM_IEEE80211_RADIO, iev, sizeof(iev));
+ CURVNET_RESTORE();
+ }
}
void
___
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: r280849 - in head: contrib/ntp contrib/ntp/adjtimedcontrib/ntp/arlib contrib/ntp/clockstuff contrib/ntp/conf contrib/ntp/htmlcontrib/ntp/html/build contrib/ntp/html/drivers contrib/nt.
On Mon, May 25, 2015 at 01:19:52PM -0700, Cy schubert wrote: Thanks bapt@. I will take a look after work tonight. ~Cy Actually I have been able to fix all the issues, but double check never hurts :) I have figured out that building WITHOUT_OPENSSL disables the AUTOKEY protocol support so I bet keeping the switch in ntp makes sense (this is only used in ntp-keygen) which is why in this makefile there is no alternative to libcrypto with libmd, so when one will build WITHOUT_OPENSSL it will be built without the autokey feature. Best regards, Bapt pgp5FHtKGMaCy.pgp Description: PGP signature
svn commit: r283544 - in head/sys: amd64/linux32 compat/linux i386/linux
Author: dchagin
Date: Mon May 25 20:44:46 2015
New Revision: 283544
URL: https://svnweb.freebsd.org/changeset/base/283544
Log:
When I merged the lemul branch I missied kib@'s r282708 commit.
This is not the final fix as I need properly cleanup thread resources
before other threads suicide.
Tested by:Ruslan Makhmatkhanov
Modified:
head/sys/amd64/linux32/linux32_machdep.c
head/sys/compat/linux/linux_emul.c
head/sys/i386/linux/linux_machdep.c
Modified: head/sys/amd64/linux32/linux32_machdep.c
==
--- head/sys/amd64/linux32/linux32_machdep.cMon May 25 20:29:35 2015
(r283543)
+++ head/sys/amd64/linux32/linux32_machdep.cMon May 25 20:44:46 2015
(r283544)
@@ -126,7 +126,6 @@ int
linux_execve(struct thread *td, struct linux_execve_args *args)
{
struct image_args eargs;
- struct vmspace *oldvmspace;
char *path;
int error;
@@ -137,19 +136,11 @@ linux_execve(struct thread *td, struct l
printf(ARGS(execve, %s), path);
#endif
- error = pre_execve(td, oldvmspace);
- if (error != 0) {
- free(path, M_TEMP);
- return (error);
- }
error = freebsd32_exec_copyin_args(eargs, path, UIO_SYSSPACE,
args-argp, args-envp);
free(path, M_TEMP);
if (error == 0)
- error = kern_execve(td, eargs, NULL);
- if (error == 0)
error = linux_common_execve(td, eargs);
- post_execve(td, error, oldvmspace);
return (error);
}
Modified: head/sys/compat/linux/linux_emul.c
==
--- head/sys/compat/linux/linux_emul.c Mon May 25 20:29:35 2015
(r283543)
+++ head/sys/compat/linux/linux_emul.c Mon May 25 20:44:46 2015
(r283544)
@@ -172,27 +172,19 @@ linux_common_execve(struct thread *td, s
{
struct linux_pemuldata *pem;
struct epoll_emuldata *emd;
+ struct vmspace *oldvmspace;
struct linux_emuldata *em;
struct proc *p;
int error;
p = td-td_proc;
- /*
-* Unlike FreeBSD abort all other threads before
-* proceeding exec.
-*/
- PROC_LOCK(p);
- /* See exit1() comments. */
- thread_suspend_check(0);
- while (p-p_flag P_HADTHREADS) {
- if (!thread_single(p, SINGLE_EXIT))
- break;
- thread_suspend_check(0);
- }
- PROC_UNLOCK(p);
+ error = pre_execve(td, oldvmspace);
+ if (error != 0)
+ return (error);
error = kern_execve(td, eargs, NULL);
+ post_execve(td, error, oldvmspace);
if (error != 0)
return (error);
Modified: head/sys/i386/linux/linux_machdep.c
==
--- head/sys/i386/linux/linux_machdep.c Mon May 25 20:29:35 2015
(r283543)
+++ head/sys/i386/linux/linux_machdep.c Mon May 25 20:44:46 2015
(r283544)
@@ -104,7 +104,6 @@ int
linux_execve(struct thread *td, struct linux_execve_args *args)
{
struct image_args eargs;
- struct vmspace *oldvmspace;
char *newpath;
int error;
@@ -115,19 +114,11 @@ linux_execve(struct thread *td, struct l
printf(ARGS(execve, %s), newpath);
#endif
- error = pre_execve(td, oldvmspace);
- if (error != 0) {
- free(newpath, M_TEMP);
- return (error);
- }
error = exec_copyin_args(eargs, newpath, UIO_SYSSPACE,
args-argp, args-envp);
free(newpath, M_TEMP);
if (error == 0)
- error = kern_execve(td, eargs, NULL);
- if (error == 0)
error = linux_common_execve(td, eargs);
- post_execve(td, error, oldvmspace);
return (error);
}
___
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: r280849 - in head: contrib/ntpcontrib/ntp/adjtimedcontrib/ntp/arlib contrib/ntp/clockstuff contrib/ntp/confcontrib/ntp/htmlcontrib/ntp/html/build contrib/ntp/html/drivers contrib/nt...
I don't have internet access at the moment, just a cell phone with really crappy data performance, so I'm not in the know right now. But if you don't mind I can make openssl a conditional. ~Cy -Original Message- From: Baptiste Daroussin Sent: 25/05/2015 13:36 To: Cy schubert Cc: Cy Schubert; src-committ...@freebsd.org; svn-src-...@freebsd.org; svn-src-head@freebsd.org Subject: Re: svn commit: r280849 - in head: contrib/ntpcontrib/ntp/adjtimedcontrib/ntp/arlib contrib/ntp/clockstuff contrib/ntp/confcontrib/ntp/htmlcontrib/ntp/html/build contrib/ntp/html/drivers contrib/nt... On Mon, May 25, 2015 at 01:19:52PM -0700, Cy schubert wrote: Thanks bapt@. I will take a look after work tonight. ~Cy Actually I have been able to fix all the issues, but double check never hurts :) I have figured out that building WITHOUT_OPENSSL disables the AUTOKEY protocol support so I bet keeping the switch in ntp makes sense (this is only used in ntp-keygen) which is why in this makefile there is no alternative to libcrypto with libmd, so when one will build WITHOUT_OPENSSL it will be built without the autokey feature. Best regards, Bapt ___ 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: r280849 - in head: contrib/ntpcontrib/ntp/adjtimedcontrib/ntp/arlib contrib/ntp/clockstuff contrib/ntp/confcontrib/ntp/htmlcontrib/ntp/html/build contrib/ntp/html/drivers contrib/nt...
On Mon, May 25, 2015 at 02:02:07PM -0700, Cy schubert wrote: I don't have internet access at the moment, just a cell phone with really crappy data performance, so I'm not in the know right now. But if you don't mind I can make openssl a conditional. ~Cy This is what I did, no rush, I have fixed everthing in r283542 Best regards, Bapt pgpEHiK2WYGcP.pgp Description: PGP signature
svn commit: r283538 - head/sys/net80211
Author: adrian
Date: Mon May 25 19:18:16 2015
New Revision: 283538
URL: https://svnweb.freebsd.org/changeset/base/283538
Log:
Convert malloc/free back to #define's, as part of OS portability work.
DragonflyBSD uses the FreeBSD wireless stack and drivers. Their malloc()
API is named differently, so they don't have userland/kernel symbol
clashes like we do (think libuinet.)
So, to make it easier for them and to port to other BSDs/other operating
systems, start hiding the malloc specific bits behind defines in
ieee80211_freebsd.h.
DragonflyBSD can now put these portability defines in their local
ieee80211_dragonflybsd.h.
This should be a great big no-op for everyone running wifi.
TODO:
* kill M_WAITOK - some platforms just don't want you to use it
* .. and/or handle it returning NULL rather than waiting forever.
* MALLOC_DEFINE() ?
* Migrate the well-known malloc names (eg M_TEMP) to net80211
namespace defines.
Modified:
head/sys/net80211/ieee80211_acl.c
head/sys/net80211/ieee80211_amrr.c
head/sys/net80211/ieee80211_crypto_ccmp.c
head/sys/net80211/ieee80211_crypto_tkip.c
head/sys/net80211/ieee80211_crypto_wep.c
head/sys/net80211/ieee80211_freebsd.c
head/sys/net80211/ieee80211_freebsd.h
head/sys/net80211/ieee80211_hostap.c
head/sys/net80211/ieee80211_hwmp.c
head/sys/net80211/ieee80211_input.c
head/sys/net80211/ieee80211_ioctl.c
head/sys/net80211/ieee80211_mesh.c
head/sys/net80211/ieee80211_node.c
head/sys/net80211/ieee80211_output.c
head/sys/net80211/ieee80211_power.c
head/sys/net80211/ieee80211_proto.c
head/sys/net80211/ieee80211_ratectl_none.c
head/sys/net80211/ieee80211_regdomain.c
head/sys/net80211/ieee80211_rssadapt.c
head/sys/net80211/ieee80211_scan_sta.c
head/sys/net80211/ieee80211_scan_sw.c
head/sys/net80211/ieee80211_sta.c
head/sys/net80211/ieee80211_superg.c
head/sys/net80211/ieee80211_tdma.c
Modified: head/sys/net80211/ieee80211_acl.c
==
--- head/sys/net80211/ieee80211_acl.c Mon May 25 18:50:26 2015
(r283537)
+++ head/sys/net80211/ieee80211_acl.c Mon May 25 19:18:16 2015
(r283538)
@@ -99,8 +99,8 @@ acl_attach(struct ieee80211vap *vap)
{
struct aclstate *as;
- as = (struct aclstate *) malloc(sizeof(struct aclstate),
- M_80211_ACL, M_NOWAIT | M_ZERO);
+ as = (struct aclstate *) IEEE80211_MALLOC(sizeof(struct aclstate),
+ M_80211_ACL, IEEE80211_M_NOWAIT | IEEE80211_M_ZERO);
if (as == NULL)
return 0;
ACL_LOCK_INIT(as, acl);
@@ -123,7 +123,7 @@ acl_detach(struct ieee80211vap *vap)
acl_free_all(vap);
vap-iv_as = NULL;
ACL_LOCK_DESTROY(as);
- free(as, M_80211_ACL);
+ IEEE80211_FREE(as, M_80211_ACL);
}
static __inline struct acl *
@@ -147,7 +147,7 @@ _acl_free(struct aclstate *as, struct ac
TAILQ_REMOVE(as-as_list, acl, acl_list);
LIST_REMOVE(acl, acl_hash);
- free(acl, M_80211_ACL);
+ IEEE80211_FREE(acl, M_80211_ACL);
as-as_nacls--;
}
@@ -175,7 +175,8 @@ acl_add(struct ieee80211vap *vap, const
struct acl *acl, *new;
int hash;
- new = (struct acl *) malloc(sizeof(struct acl), M_80211_ACL, M_NOWAIT |
M_ZERO);
+ new = (struct acl *) IEEE80211_MALLOC(sizeof(struct acl),
+ M_80211_ACL, IEEE80211_M_NOWAIT | IEEE80211_M_ZERO);
if (new == NULL) {
IEEE80211_DPRINTF(vap, IEEE80211_MSG_ACL,
ACL: add %s failed, no memory\n, ether_sprintf(mac));
@@ -188,7 +189,7 @@ acl_add(struct ieee80211vap *vap, const
LIST_FOREACH(acl, as-as_hash[hash], acl_hash) {
if (IEEE80211_ADDR_EQ(acl-acl_macaddr, mac)) {
ACL_UNLOCK(as);
- free(new, M_80211_ACL);
+ IEEE80211_FREE(new, M_80211_ACL);
IEEE80211_DPRINTF(vap, IEEE80211_MSG_ACL,
ACL: add %s failed, already present\n,
ether_sprintf(mac));
@@ -302,8 +303,8 @@ acl_getioctl(struct ieee80211vap *vap, s
ireq-i_len = space;/* return required space */
return 0; /* NB: must not error */
}
- ap = (struct ieee80211req_maclist *) malloc(space,
- M_TEMP, M_NOWAIT);
+ ap = (struct ieee80211req_maclist *) IEEE80211_MALLOC(space,
+ M_TEMP, IEEE80211_M_NOWAIT);
if (ap == NULL)
return ENOMEM;
i = 0;
@@ -318,7 +319,7 @@ acl_getioctl(struct ieee80211vap *vap, s
ireq-i_len = space;
} else
error = copyout(ap, ireq-i_data, ireq-i_len);
- free(ap, M_TEMP);
+ IEEE80211_FREE(ap, M_TEMP);
Re: svn commit: r283538 - head/sys/net80211
On Mon, May 25, 2015 at 07:18:16PM +, Adrian Chadd wrote: A * kill M_WAITOK - some platforms just don't want you to use it Actually all functions called in device attach context or in syscall context should use M_WAITOK. This makes API more robust and simplifies code a lot. Can we achieve portability without losing this nice feature of FreeBSD? -- 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
Re: svn commit: r280849 - in head: contrib/ntp contrib/ntp/adjtimed contrib/ntp/arlib contrib/ntp/clockstuff contrib/ntp/conf contrib/ntp/html contrib/ntp/html/build contrib/ntp/html/drivers contrib/n
On Mon, Mar 30, 2015 at 01:30:17PM +, Cy Schubert wrote: Author: cy Date: Mon Mar 30 13:30:15 2015 New Revision: 280849 URL: https://svnweb.freebsd.org/changeset/base/280849 Log: MFV ntp 4.2.8p1 (r258945, r275970, r276091, r276092, r276093, r278284) Thanks to roberto for providing pointers to wedge this into HEAD. Approved by:roberto Thanks for the update, but you have reintroduced overlinking of useless libraries all over ntp. for example with ntpdc: The binary does not need at all to link with ncurses explicitly The binary does not need to link with both libedit and readline (libedit is enough and prefered) The binary does not need to link to libmd as it does not use it. The binary needs libcrypto but absolutly not to libssl Also ntpdc does not build if WITHOUT_OPENSSL because in the config.h it is hardcoded that it needs openssl, for example: #define USE_OPENSSL_CRYPTO_RAND 1 Before the update, libmd was used for the digests functions. gGven now it is hardcoded that openssl is required libmd should be replaced by libcrypto but you should not have both. I'm working on fixing the overlink, but given I know nothing about libntp's code I will let others have a look on what to do with the MK_OPENSSL. Best regards, Bapt pgpZzLZHSo_T3.pgp Description: PGP signature
svn commit: r283542 - in head/usr.sbin/ntp: . ntp-keygen ntpd ntpdate ntpdc ntpq sntp
Author: bapt
Date: Mon May 25 20:27:46 2015
New Revision: 283542
URL: https://svnweb.freebsd.org/changeset/base/283542
Log:
Fix overlinking again after recent ntp updates
Fix building WITHOUT_OPENSSL
Modified:
head/usr.sbin/ntp/Makefile.inc
head/usr.sbin/ntp/config.h
head/usr.sbin/ntp/ntp-keygen/Makefile
head/usr.sbin/ntp/ntpd/Makefile
head/usr.sbin/ntp/ntpdate/Makefile
head/usr.sbin/ntp/ntpdc/Makefile
head/usr.sbin/ntp/ntpq/Makefile
head/usr.sbin/ntp/sntp/Makefile
Modified: head/usr.sbin/ntp/Makefile.inc
==
--- head/usr.sbin/ntp/Makefile.inc Mon May 25 20:06:49 2015
(r283541)
+++ head/usr.sbin/ntp/Makefile.inc Mon May 25 20:27:46 2015
(r283542)
@@ -11,7 +11,7 @@ NTPDEFS= -DSYS_FREEBSD
CFLAGS+= ${NTPDEFS} ${DEFS_LOCAL} ${CLOCKDEFS}
.if ${MK_OPENSSL} != no !defined(RELEASE_CRUNCH)
-CFLAGS+= -DOPENSSL
+CFLAGS+= -DOPENSSL -DUSE_OPENSSL_CRYPTO_RAND -DAUTOKEY
.endif
WARNS?=0
Modified: head/usr.sbin/ntp/config.h
==
--- head/usr.sbin/ntp/config.h Mon May 25 20:06:49 2015(r283541)
+++ head/usr.sbin/ntp/config.h Mon May 25 20:27:46 2015(r283542)
@@ -9,7 +9,7 @@
/* #undef ADJTIME_IS_ACCURATE */
/* Support NTP Autokey protocol? */
-#define AUTOKEY 1
+/* #define AUTOKEY 1 */
/* why not HAVE_P_S? */
/* #undef CALL_PTHREAD_SETCONCURRENCY */
@@ -1578,7 +1578,7 @@ typedef unsigned int uintptr_t;
#define USE_FSETOWNCTTY 1
/* Use OpenSSL's crypto random functions */
-#define USE_OPENSSL_CRYPTO_RAND 1
+/* #define USE_OPENSSL_CRYPTO_RAND 1 */
/* OK to use snprintb()? */
/* #undef USE_SNPRINTB */
Modified: head/usr.sbin/ntp/ntp-keygen/Makefile
==
--- head/usr.sbin/ntp/ntp-keygen/Makefile Mon May 25 20:06:49 2015
(r283541)
+++ head/usr.sbin/ntp/ntp-keygen/Makefile Mon May 25 20:27:46 2015
(r283542)
@@ -23,7 +23,7 @@ CFLAGS+= -I${.CURDIR}/../../../contrib/n
LIBADD+= ntp opts pthread
.if ${MK_OPENSSL} != no
-LIBADD+= md crypto
+LIBADD+= crypto
.endif
.include bsd.prog.mk
Modified: head/usr.sbin/ntp/ntpd/Makefile
==
--- head/usr.sbin/ntp/ntpd/Makefile Mon May 25 20:06:49 2015
(r283541)
+++ head/usr.sbin/ntp/ntpd/Makefile Mon May 25 20:27:46 2015
(r283542)
@@ -35,10 +35,12 @@ CFLAGS+= -I${.CURDIR}/../../../contrib/n
-I${.CURDIR}/../ \
-I${.CURDIR}
-LIBADD=parse ntp m rt opts md pthread
+LIBADD=parse ntp m opts pthread
.if ${MK_OPENSSL} != no
LIBADD+= crypto
+.else
+LIBADD+= md
.endif
CLEANFILES+= .version version.c
Modified: head/usr.sbin/ntp/ntpdate/Makefile
==
--- head/usr.sbin/ntp/ntpdate/Makefile Mon May 25 20:06:49 2015
(r283541)
+++ head/usr.sbin/ntp/ntpdate/Makefile Mon May 25 20:27:46 2015
(r283542)
@@ -14,10 +14,12 @@ CFLAGS+= -I${.CURDIR}/../../../contrib/n
-I${.CURDIR}/../../../contrib/ntp/lib/isc/pthreads/include \
-I${.CURDIR}/../
-LIBADD=rt ntp m pthread
+LIBADD=ntp m pthread
.if ${MK_OPENSSL} != no
-LIBADD+= md ssl crypto
+LIBADD+= crypto
+.else
+LIBADD+= md
.endif
CLEANFILES+= .version version.c
Modified: head/usr.sbin/ntp/ntpdc/Makefile
==
--- head/usr.sbin/ntp/ntpdc/MakefileMon May 25 20:06:49 2015
(r283541)
+++ head/usr.sbin/ntp/ntpdc/MakefileMon May 25 20:27:46 2015
(r283542)
@@ -18,12 +18,14 @@ CFLAGS+= -I${.CURDIR}/../../../contrib/n
-I${.CURDIR}/../../../lib/libc/${MACHINE_ARCH} \
-I${.CURDIR}/../ -I${.CURDIR}
-LIBADD=edit md ntp m readline opts ncurses pthread
+LIBADD=edit ntp m opts pthread
CFLAGS+= -DHAVE_LIBEDIT -DHAVE_READLINE_READLINE_H \
-I${DESTDIR}/${INCLUDEDIR}/edit
.if ${MK_OPENSSL} != no
-LIBADD+= ssl crypto
+LIBADD+= crypto
+.else
+LIBADD+= md
.endif
CLEANFILES+= .version version.c
Modified: head/usr.sbin/ntp/ntpq/Makefile
==
--- head/usr.sbin/ntp/ntpq/Makefile Mon May 25 20:06:49 2015
(r283541)
+++ head/usr.sbin/ntp/ntpq/Makefile Mon May 25 20:27:46 2015
(r283542)
@@ -21,10 +21,12 @@ CFLAGS+= -I${.CURDIR}/../../../contrib/n
-I${.CURDIR}/../../../contrib/ntp/sntp/libopts \
-I${.CURDIR}/../
-LIBADD+= edit md ntp opts m pthread
+LIBADD+= edit ntp opts m pthread
.if ${MK_OPENSSL} != no
-LIBADD+= ssl crypto
+LIBADD+= crypto
+.else
+LIBADD+= md
.endif
CFLAGS+= -DHAVE_LIBEDIT
svn commit: r283543 - head/gnu/lib
Author: bapt
Date: Mon May 25 20:29:35 2015
New Revision: 283543
URL: https://svnweb.freebsd.org/changeset/base/283543
Log:
ntp is now again libreadline free, so only build libreadline for gdb
Modified:
head/gnu/lib/Makefile
Modified: head/gnu/lib/Makefile
==
--- head/gnu/lib/Makefile Mon May 25 20:27:46 2015(r283542)
+++ head/gnu/lib/Makefile Mon May 25 20:29:35 2015(r283543)
@@ -16,7 +16,7 @@ SUBDIR+= libssp
SUBDIR+= tests
.endif
-.if ${MK_GDB} != no || ${MK_NTP} != no
+.if ${MK_GDB} != no
SUBDIR+= libreadline
.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: r283511 - head/sys/opencrypto
On Mon, May 25, 2015 at 11:26:01AM +, Bjoern A. Zeeb wrote: On 25 May 2015, at 01:31 , Mark Johnston ma...@freebsd.org wrote: Author: markj Date: Mon May 25 01:31:39 2015 New Revision: 283511 URL: https://svnweb.freebsd.org/changeset/base/283511 Log: Use the correct number of arguments for the opencrypto:deflate:deflate_global:bad DTrace probe, which is defined to have Once upon a time, these expended to exactly the same if my memory doesn’t fool me. Thanks for cleaning up though! They still do I think. Some upcoming work on SDT will turn this sort of discrepancy into a compile error though; this was the one occurrence of it that I ran into. -Mark ___ 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: r283541 - head/sys/net80211
Author: glebius
Date: Mon May 25 20:06:49 2015
New Revision: 283541
URL: https://svnweb.freebsd.org/changeset/base/283541
Log:
Cleanup compat shims for FreeBSD versions that predate 10.0-RELEASE.
There are no plans to merge anything save a trivial bugfix to stable/9.
Discussed with: adrian
Modified:
head/sys/net80211/ieee80211.c
head/sys/net80211/ieee80211_freebsd.c
head/sys/net80211/ieee80211_freebsd.h
head/sys/net80211/ieee80211_hostap.c
head/sys/net80211/ieee80211_input.c
head/sys/net80211/ieee80211_output.c
head/sys/net80211/ieee80211_proto.h
head/sys/net80211/ieee80211_var.h
Modified: head/sys/net80211/ieee80211.c
==
--- head/sys/net80211/ieee80211.c Mon May 25 19:53:29 2015
(r283540)
+++ head/sys/net80211/ieee80211.c Mon May 25 20:06:49 2015
(r283541)
@@ -245,15 +245,9 @@ null_transmit(struct ifnet *ifp, struct
return EACCES; /* XXX EIO/EPERM? */
}
-#if __FreeBSD_version = 131
static int
null_output(struct ifnet *ifp, struct mbuf *m,
const struct sockaddr *dst, struct route *ro)
-#else
-static int
-null_output(struct ifnet *ifp, struct mbuf *m,
- struct sockaddr *dst, struct route *ro)
-#endif
{
if_printf(ifp, discard raw packet\n);
return null_transmit(ifp, m);
Modified: head/sys/net80211/ieee80211_freebsd.c
==
--- head/sys/net80211/ieee80211_freebsd.c Mon May 25 19:53:29 2015
(r283540)
+++ head/sys/net80211/ieee80211_freebsd.c Mon May 25 20:06:49 2015
(r283541)
@@ -66,10 +66,8 @@ SYSCTL_INT(_net_wlan, OID_AUTO, debug, C
static MALLOC_DEFINE(M_80211_COM, 80211com, 802.11 com state);
-#if __FreeBSD_version = 120
static const char wlanname[] = wlan;
static struct if_clone *wlan_cloner;
-#endif
/*
* Allocate/free com structure in conjunction with ifnet;
@@ -136,18 +134,10 @@ wlan_clone_create(struct if_clone *ifc,
if_printf(ifp, TDMA not supported\n);
return EOPNOTSUPP;
}
-#if __FreeBSD_version = 120
vap = ic-ic_vap_create(ic, wlanname, unit,
cp.icp_opmode, cp.icp_flags, cp.icp_bssid,
cp.icp_flags IEEE80211_CLONE_MACADDR ?
cp.icp_macaddr : (const uint8_t *)IF_LLADDR(ifp));
-#else
- vap = ic-ic_vap_create(ic, ifc-ifc_name, unit,
- cp.icp_opmode, cp.icp_flags, cp.icp_bssid,
- cp.icp_flags IEEE80211_CLONE_MACADDR ?
- cp.icp_macaddr : (const uint8_t *)IF_LLADDR(ifp));
-
-#endif
return (vap == NULL ? EIO : 0);
}
@@ -161,19 +151,11 @@ wlan_clone_destroy(struct ifnet *ifp)
ic-ic_vap_delete(vap);
}
-#if __FreeBSD_version 120
-IFC_SIMPLE_DECLARE(wlan, 0);
-#endif
-
void
ieee80211_vap_destroy(struct ieee80211vap *vap)
{
CURVNET_SET(vap-iv_ifp-if_vnet);
-#if __FreeBSD_version = 120
if_clone_destroyif(wlan_cloner, vap-iv_ifp);
-#else
- if_clone_destroyif(wlan_cloner, vap-iv_ifp);
-#endif
CURVNET_RESTORE();
}
@@ -891,21 +873,13 @@ wlan_modevent(module_t mod, int type, vo
bpf_track, 0, EVENTHANDLER_PRI_ANY);
wlan_ifllevent = EVENTHANDLER_REGISTER(iflladdr_event,
wlan_iflladdr, NULL, EVENTHANDLER_PRI_ANY);
-#if __FreeBSD_version = 120
wlan_cloner = if_clone_simple(wlanname, wlan_clone_create,
wlan_clone_destroy, 0);
-#else
- if_clone_attach(wlan_cloner);
-#endif
if_register_com_alloc(IFT_IEEE80211, wlan_alloc, wlan_free);
return 0;
case MOD_UNLOAD:
if_deregister_com_alloc(IFT_IEEE80211);
-#if __FreeBSD_version = 120
if_clone_detach(wlan_cloner);
-#else
- if_clone_detach(wlan_cloner);
-#endif
EVENTHANDLER_DEREGISTER(bpf_track, wlan_bpfevent);
EVENTHANDLER_DEREGISTER(iflladdr_event, wlan_ifllevent);
return 0;
@@ -914,11 +888,7 @@ wlan_modevent(module_t mod, int type, vo
}
static moduledata_t wlan_mod = {
-#if __FreeBSD_version = 120
wlanname,
-#else
- wlan,
-#endif
wlan_modevent,
0
};
Modified: head/sys/net80211/ieee80211_freebsd.h
==
--- head/sys/net80211/ieee80211_freebsd.h Mon May 25 19:53:29 2015
(r283540)
+++ head/sys/net80211/ieee80211_freebsd.h Mon May 25 20:06:49 2015
(r283541)
@@ -234,21 +234,9 @@ struct mbuf *ieee80211_getmgtframe(uint8
#defineM_FFM_PROTO6/* fast frame */
#defineM_TXCB M_PROTO7/* do tx complete
callback */
#defineM_AMPDU_MPDUM_PROTO8
svn commit: r283546 - in head/sys: kern sys
Author: jhb
Date: Mon May 25 22:13:22 2015
New Revision: 283546
URL: https://svnweb.freebsd.org/changeset/base/283546
Log:
Add KTR tracing for some MI ptrace events.
Differential Revision:https://reviews.freebsd.org/D2643
Reviewed by: kib
Modified:
head/sys/kern/kern_exit.c
head/sys/kern/kern_fork.c
head/sys/kern/kern_sig.c
head/sys/kern/sys_process.c
head/sys/sys/ktr_class.h
Modified: head/sys/kern/kern_exit.c
==
--- head/sys/kern/kern_exit.c Mon May 25 22:12:30 2015(r283545)
+++ head/sys/kern/kern_exit.c Mon May 25 22:13:22 2015(r283546)
@@ -525,6 +525,8 @@ exit1(struct thread *td, int rv)
*/
while ((q = LIST_FIRST(p-p_orphans)) != NULL) {
PROC_LOCK(q);
+ CTR2(KTR_PTRACE, exit: pid %d, clearing orphan %d, p-p_pid,
+ q-p_pid);
clear_orphan(q);
PROC_UNLOCK(q);
}
@@ -857,6 +859,9 @@ proc_reap(struct thread *td, struct proc
t = proc_realparent(p);
PROC_LOCK(t);
PROC_LOCK(p);
+ CTR2(KTR_PTRACE,
+ wait: traced child %d moved back to parent %d, p-p_pid,
+ t-p_pid);
proc_reparent(p, t);
p-p_oppid = 0;
PROC_UNLOCK(p);
@@ -1216,6 +1221,10 @@ loop:
PROC_UNLOCK(q);
}
+ CTR4(KTR_PTRACE,
+ wait: returning trapped pid %d status %#x (xstat %d) xthread %d,
+ p-p_pid, W_STOPCODE(p-p_xstat), p-p_xstat,
+ p-p_xthread != NULL ? p-p_xthread-td_tid : -1);
PROC_UNLOCK(p);
return (0);
}
Modified: head/sys/kern/kern_fork.c
==
--- head/sys/kern/kern_fork.c Mon May 25 22:12:30 2015(r283545)
+++ head/sys/kern/kern_fork.c Mon May 25 22:13:22 2015(r283546)
@@ -1035,6 +1035,9 @@ fork_return(struct thread *td, struct tr
dbg = p-p_pptr-p_pptr;
p-p_flag |= P_TRACED;
p-p_oppid = p-p_pptr-p_pid;
+ CTR2(KTR_PTRACE,
+ fork_return: attaching to new child pid %d: oppid %d,
+ p-p_pid, p-p_oppid);
proc_reparent(p, dbg);
sx_xunlock(proctree_lock);
td-td_dbgflags |= TDB_CHILD;
Modified: head/sys/kern/kern_sig.c
==
--- head/sys/kern/kern_sig.cMon May 25 22:12:30 2015(r283545)
+++ head/sys/kern/kern_sig.cMon May 25 22:13:22 2015(r283546)
@@ -2478,6 +2478,8 @@ ptracestop(struct thread *td, int sig)
td-td_dbgflags |= TDB_XSIG;
td-td_xsig = sig;
+ CTR4(KTR_PTRACE, ptracestop: tid %d (pid %d) flags %#x sig %d,
+ td-td_tid, p-p_pid, td-td_dbgflags, sig);
PROC_SLOCK(p);
while ((p-p_flag P_TRACED) (td-td_dbgflags TDB_XSIG)) {
if (p-p_flag P_SINGLE_EXIT) {
Modified: head/sys/kern/sys_process.c
==
--- head/sys/kern/sys_process.c Mon May 25 22:12:30 2015(r283545)
+++ head/sys/kern/sys_process.c Mon May 25 22:13:22 2015(r283546)
@@ -432,6 +432,9 @@ ptrace_vm_entry(struct thread *td, struc
free(freepath, M_TEMP);
}
}
+ if (error == 0)
+ CTR3(KTR_PTRACE, PT_VM_ENTRY: pid %d, entry %d, start %p,
+ p-p_pid, pve-pve_entry, pve-pve_start);
return (error);
}
@@ -826,6 +829,7 @@ kern_ptrace(struct thread *td, int req,
if (p-p_flag P_PPWAIT)
p-p_flag |= P_PPTRACE;
p-p_oppid = p-p_pptr-p_pid;
+ CTR1(KTR_PTRACE, PT_TRACE_ME: pid %d, p-p_pid);
break;
case PT_ATTACH:
@@ -845,17 +849,25 @@ kern_ptrace(struct thread *td, int req,
proc_reparent(p, td-td_proc);
}
data = SIGSTOP;
+ CTR2(KTR_PTRACE, PT_ATTACH: pid %d, oppid %d, p-p_pid,
+ p-p_oppid);
goto sendsig; /* in PT_CONTINUE below */
case PT_CLEARSTEP:
+ CTR2(KTR_PTRACE, PT_CLEARSTEP: tid %d (pid %d), td2-td_tid,
+ p-p_pid);
error = ptrace_clear_single_step(td2);
break;
case PT_SETSTEP:
+ CTR2(KTR_PTRACE, PT_SETSTEP: tid %d (pid %d), td2-td_tid,
+ p-p_pid);
error = ptrace_single_step(td2);
break;
case PT_SUSPEND:
+
svn commit: r283550 - head/sys/arm/broadcom/bcm2835
Author: loos
Date: Tue May 26 01:30:09 2015
New Revision: 283550
URL: https://svnweb.freebsd.org/changeset/base/283550
Log:
Remove unused mutex and softc variables.
Modified:
head/sys/arm/broadcom/bcm2835/bcm2835_sdhci.c
Modified: head/sys/arm/broadcom/bcm2835/bcm2835_sdhci.c
==
--- head/sys/arm/broadcom/bcm2835/bcm2835_sdhci.c Tue May 26 01:09:56
2015(r283549)
+++ head/sys/arm/broadcom/bcm2835/bcm2835_sdhci.c Tue May 26 01:30:09
2015(r283550)
@@ -76,23 +76,13 @@ TUNABLE_INT(hw.bcm2835.sdhci.pio_mode,
struct bcm_sdhci_softc {
device_tsc_dev;
- struct mtx sc_mtx;
struct resource * sc_mem_res;
struct resource * sc_irq_res;
bus_space_tag_t sc_bst;
bus_space_handle_t sc_bsh;
void * sc_intrhand;
struct mmc_request *sc_req;
- struct mmc_data * sc_data;
- uint32_tsc_flags;
-#defineLPC_SD_FLAGS_IGNORECRC (1 0)
- int sc_xfer_direction;
-#defineDIRECTION_READ 0
-#defineDIRECTION_WRITE 1
- int sc_xfer_done;
- int sc_bus_busy;
struct sdhci_slot sc_slot;
- int sc_dma_inuse;
int sc_dma_ch;
bus_dma_tag_t sc_dma_tag;
bus_dmamap_tsc_dma_map;
@@ -113,11 +103,6 @@ static void bcm_sdhci_intr(void *);
static int bcm_sdhci_get_ro(device_t, device_t);
static void bcm_sdhci_dma_intr(int ch, void *arg);
-#definebcm_sdhci_lock(_sc)
\
-mtx_lock(_sc-sc_mtx);
-#definebcm_sdhci_unlock(_sc)
\
-mtx_unlock(_sc-sc_mtx);
-
static void
bcm_sdhci_dmacb(void *arg, bus_dma_segment_t *segs, int nseg, int err)
{
@@ -187,8 +172,6 @@ bcm_sdhci_attach(device_t dev)
if (bootverbose)
device_printf(dev, SDHCI frequency: %dMHz\n, default_freq);
- mtx_init(sc-sc_mtx, bcm sdhci, sdhci, MTX_DEF);
-
rid = 0;
sc-sc_mem_res = bus_alloc_resource_any(dev, SYS_RES_MEMORY, rid,
RF_ACTIVE);
@@ -277,7 +260,6 @@ fail:
bus_release_resource(dev, SYS_RES_IRQ, 0, sc-sc_irq_res);
if (sc-sc_mem_res)
bus_release_resource(dev, SYS_RES_MEMORY, 0, sc-sc_mem_res);
- mtx_destroy(sc-sc_mtx);
return (err);
}
___
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: r282240 - in head: cddl/lib/libdtrace sys/netinet
On 5/25/15 8:27 PM, George Neville-Neil wrote: On 24 May 2015, at 22:50, Julian Elischer wrote: On 5/24/15 9:24 PM, George Neville-Neil wrote: On 23 May 2015, at 17:59, hiren panchasara wrote: On 04/29/15 at 05:19P, George V. Neville-Neil wrote: Author: gnn Date: Wed Apr 29 17:19:55 2015 New Revision: 282240 URL: https://svnweb.freebsd.org/changeset/base/282240 Log: Brief demo script showing the various values that can be read via the new SIFTR statically defined tracepoint (SDT). Differential Revision: https://reviews.freebsd.org/D2387 Reviewed by:bz, markj Modified: head/cddl/lib/libdtrace/tcp.d head/sys/netinet/in_kdtrace.c head/sys/netinet/in_kdtrace.h head/sys/netinet/siftr.c George, This seems useful. Can this be MFC'd to 10? I think it can be. I'll mark this and do it when I'm back from holiday. Nag me if I forget :-) Best, George George.. check out this script to make MFCs to 10 an absolute snap.. instead of being a worrying operation, it's a few keystrokes... http://www.freebsd.org/~julian/MFC-10.sh credits to Bapt for the original (ports) idea. you don't even need to have a tree checked out.. it does all that.. Thanks! Best, George ___ 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: r280849 - in head: contrib/ntpcontrib/ntp/adjtimedcontrib/ntp/arlib contrib/ntp/clockstuff contrib/ntp/confcontrib/ntp/htmlcontrib/ntp/html/build contrib/ntp/html/drivers contrib/nt...
In message 20150525210441.ge39...@ivaldir.etoilebsd.net, Baptiste Daroussin w rites: --LSp5EJdfMPwZcMS1 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Mon, May 25, 2015 at 02:02:07PM -0700, Cy schubert wrote: I don't have internet access at the moment, just a cell phone with really= crappy data performance, so I'm not in the know right now. But if you don'= t mind I can make openssl a conditional. =20 ~Cy This is what I did, no rush, I have fixed everthing in r283542 Ah yes. I see. Thank you. -- Cheers, Cy Schubert cy.schub...@komquats.com or cy.schub...@cschubert.com FreeBSD UNIX: c...@freebsd.org Web: http://www.FreeBSD.org The need of the many outweighs the greed of the few. ___ 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: r283547 - head/sys/arm/conf
Ian Lepore wrote this message on Mon, May 25, 2015 at 23:27 +: Author: ian Date: Mon May 25 23:27:13 2015 New Revision: 283547 URL: https://svnweb.freebsd.org/changeset/base/283547 Log: Ensure that all arm kernel configs contain ALT_BREAK_TO_DEBUGGER and not BREAK_TO_DEBUGGER if they have a serial console (most do). A burst of serial line noise (such as unplugging a usb serial adapter) can look like a break and drop a working system into the debugger. The alt break sequence (CR~^B) works fine on both serial and non-serial consoles. Just so you know, this removes support for video console to break into the debugger via keyboard... This mean ctrl-alt-esc and other key sequences won't work anymore... -- 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: r283549 - head/share/mk
Author: bapt
Date: Tue May 26 01:09:56 2015
New Revision: 283549
URL: https://svnweb.freebsd.org/changeset/base/283549
Log:
Considers cases when NO_SHARED?=[no|NO] as dynamically linking
This reduces overlinking for parts of the build system where NO_SHARED is
set to no/NO
Modified:
head/share/mk/src.libnames.mk
Modified: head/share/mk/src.libnames.mk
==
--- head/share/mk/src.libnames.mk Tue May 26 00:44:29 2015
(r283548)
+++ head/share/mk/src.libnames.mk Tue May 26 01:09:56 2015
(r283549)
@@ -255,7 +255,7 @@ LDADD_${_l}?= -lprivate${_l}
.else
LDADD_${_l}?= ${LDADD_${_l}_L} -l${_l}
.endif
-.if defined(_DP_${_l}) defined(NO_SHARED)
+.if defined(_DP_${_l}) defined(NO_SHARED) (${NO_SHARED} != no
${NO_SHARED} != NO)
.for _d in ${_DP_${_l}}
DPADD_${_l}+= ${DPADD_${_d}}
LDADD_${_l}+= ${LDADD_${_d}}
___
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: r283547 - head/sys/arm/conf
Author: ian Date: Mon May 25 23:27:13 2015 New Revision: 283547 URL: https://svnweb.freebsd.org/changeset/base/283547 Log: Ensure that all arm kernel configs contain ALT_BREAK_TO_DEBUGGER and not BREAK_TO_DEBUGGER if they have a serial console (most do). A burst of serial line noise (such as unplugging a usb serial adapter) can look like a break and drop a working system into the debugger. The alt break sequence (CR~^B) works fine on both serial and non-serial consoles. Modified: head/sys/arm/conf/ALPINE head/sys/arm/conf/BEAGLEBONE head/sys/arm/conf/CUBIEBOARD head/sys/arm/conf/CUBIEBOARD2 head/sys/arm/conf/DOCKSTAR head/sys/arm/conf/DREAMPLUG-1001 head/sys/arm/conf/EFIKA_MX head/sys/arm/conf/EXYNOS5.common head/sys/arm/conf/PANDABOARD head/sys/arm/conf/RK3188 head/sys/arm/conf/RPI-B head/sys/arm/conf/RPI2 head/sys/arm/conf/SOCKIT.common head/sys/arm/conf/VYBRID head/sys/arm/conf/ZEDBOARD Modified: head/sys/arm/conf/ALPINE == --- head/sys/arm/conf/ALPINEMon May 25 22:13:22 2015(r283546) +++ head/sys/arm/conf/ALPINEMon May 25 23:27:13 2015(r283547) @@ -31,7 +31,7 @@ options SMP # Enable multiple cores # Debugging makeoptionsDEBUG=-g#Build kernel with gdb(1) debug symbols -optionsBREAK_TO_DEBUGGER +optionsALT_BREAK_TO_DEBUGGER optionsKDB optionsDDB #Enable the kernel debugger Modified: head/sys/arm/conf/BEAGLEBONE == --- head/sys/arm/conf/BEAGLEBONEMon May 25 22:13:22 2015 (r283546) +++ head/sys/arm/conf/BEAGLEBONEMon May 25 23:27:13 2015 (r283547) @@ -40,7 +40,7 @@ options PLATFORM # Debugging for use in -current makeoptionsDEBUG=-g# Build kernel with gdb(1) debug symbols -optionsBREAK_TO_DEBUGGER +optionsALT_BREAK_TO_DEBUGGER #options VERBOSE_SYSINIT # Enable verbose sysinit messages optionsKDB # Enable kernel debugger support # For minimum debugger support (stable branch) use: Modified: head/sys/arm/conf/CUBIEBOARD == --- head/sys/arm/conf/CUBIEBOARDMon May 25 22:13:22 2015 (r283546) +++ head/sys/arm/conf/CUBIEBOARDMon May 25 23:27:13 2015 (r283547) @@ -29,7 +29,7 @@ options SCHED_4BSD # 4BSD scheduler # Debugging for use in -current makeoptionsDEBUG=-g# Build kernel with gdb(1) debug symbols -optionsBREAK_TO_DEBUGGER +optionsALT_BREAK_TO_DEBUGGER #options VERBOSE_SYSINIT # Enable verbose sysinit messages optionsKDB # Enable kernel debugger support # For minimum debugger support (stable branch) use: Modified: head/sys/arm/conf/CUBIEBOARD2 == --- head/sys/arm/conf/CUBIEBOARD2 Mon May 25 22:13:22 2015 (r283546) +++ head/sys/arm/conf/CUBIEBOARD2 Mon May 25 23:27:13 2015 (r283547) @@ -30,7 +30,7 @@ options SMP # Enable multiple cores # Debugging for use in -current makeoptionsDEBUG=-g# Build kernel with gdb(1) debug symbols -optionsBREAK_TO_DEBUGGER +optionsALT_BREAK_TO_DEBUGGER #options VERBOSE_SYSINIT # Enable verbose sysinit messages optionsKDB # Enable kernel debugger support # For minimum debugger support (stable branch) use: Modified: head/sys/arm/conf/DOCKSTAR == --- head/sys/arm/conf/DOCKSTAR Mon May 25 22:13:22 2015(r283546) +++ head/sys/arm/conf/DOCKSTAR Mon May 25 23:27:13 2015(r283547) @@ -50,7 +50,6 @@ options _KPOSIX_PRIORITY_SCHEDULING # P # Debugging for use in -current makeoptionsDEBUG=-g# Build kernel with gdb(1) debug symbols -optionsBREAK_TO_DEBUGGER optionsALT_BREAK_TO_DEBUGGER optionsKDB # Enable kernel debugger support optionsDDB # Enable the kernel debugger Modified: head/sys/arm/conf/DREAMPLUG-1001 == --- head/sys/arm/conf/DREAMPLUG-1001Mon May 25 22:13:22 2015 (r283546) +++ head/sys/arm/conf/DREAMPLUG-1001Mon May 25 23:27:13 2015 (r283547) @@ -53,7 +53,6 @@ options _KPOSIX_PRIORITY_SCHEDULING # P # Debugging for use in -current makeoptionsDEBUG=-g# Build kernel with gdb(1) debug symbols -optionsBREAK_TO_DEBUGGER optionsALT_BREAK_TO_DEBUGGER optionsKDB
Re: svn commit: r283424 - in head/sys: amd64/linux modules/linux64
On Monday, May 25, 2015 08:27:34 PM Slawa Olhovchenkov wrote: On Mon, May 25, 2015 at 12:32:41PM -0400, John Baldwin wrote: On Monday, May 25, 2015 06:13:39 PM Slawa Olhovchenkov wrote: On Mon, May 25, 2015 at 10:40:09AM -0400, John Baldwin wrote: Some other related questions are: can we revive print/acroread now and or use a 64-bit flash plugin after these changes? I am still use print/acroread. As I know removing print/acroread irrelevant to linuxator. This is will by security reasson. I am don't open suspicious pdf from suspicious source and just ignore this removing. I mean more if these updates allow us to update to a newer version of print/acroread that would no longer be vulnerable (if such a thing exists) whether 32- or 64-bit. Currenly I don't see any linux in Acrobat Reader support OS: https://get.adobe.com/reader/otherversions/ Humm, it seems it was pulled last year. :-/ -- John Baldwin ___ 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: r283551 - head/sys/kern
Author: delphij Date: Tue May 26 01:40:33 2015 New Revision: 283551 URL: https://svnweb.freebsd.org/changeset/base/283551 Log: MFuser/delphij/zfs-arc-rebase@r281754: In r256613, taskqueue_enqueue_locked() have been modified to release the task queue lock before returning. In r276665, taskqueue_drain_all() will call taskqueue_enqueue_locked() to insert the barrier task into the queue, but did not reacquire the lock after it but later code expects the lock still being held (e.g. TQ_SLEEP()). The barrier task is special and if we release then reacquire the lock, there would be a small race window where a high priority task could sneak into the queue. Looking more closely, the race seems to be tolerable but is undesirable from semantics standpoint. To solve this, in taskqueue_drain_tq_queue(), instead of directly calling taskqueue_enqueue_locked(), insert the barrier task directly without releasing the lock. Modified: head/sys/kern/subr_taskqueue.c Directory Properties: head/ (props changed) head/sys/ (props changed) Modified: head/sys/kern/subr_taskqueue.c == --- head/sys/kern/subr_taskqueue.c Tue May 26 01:30:09 2015 (r283550) +++ head/sys/kern/subr_taskqueue.c Tue May 26 01:40:33 2015 (r283551) @@ -323,17 +323,16 @@ taskqueue_drain_tq_queue(struct taskqueu return; /* -* Enqueue our barrier with the lowest possible priority -* so we are inserted after all current tasks. +* Enqueue our barrier after all current tasks, but with +* the highest priority so that newly queued tasks cannot +* pass it. Because of the high priority, we can not use +* taskqueue_enqueue_locked directly (which drops the lock +* anyway) so just insert it at tail while we have the +* queue lock. */ - TASK_INIT(t_barrier, 0, taskqueue_task_nop_fn, t_barrier); - taskqueue_enqueue_locked(queue, t_barrier); - - /* -* Raise the barrier's priority so newly queued tasks cannot -* pass it. -*/ - t_barrier.ta_priority = USHRT_MAX; + TASK_INIT(t_barrier, USHRT_MAX, taskqueue_task_nop_fn, t_barrier); + STAILQ_INSERT_TAIL(queue-tq_queue, t_barrier, ta_link); + t_barrier.ta_pending = 1; /* * Once the barrier has executed, all previously queued tasks ___ 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: r283424 - in head/sys: amd64/linux modules/linux64
All, The enhancements to the Linux runtime are probably of more interest to folk porting server applications; I am particularly happy to see recvmmsg() and sendmmsg() go in. It might also be wise to emulate the getrandom() API, even if this is only in terms of wrapping the relevant sysctl for now. On 25/05/2015 18:27, Slawa Olhovchenkov wrote: Currenly I don't see any linux in Acrobat Reader support OS: https://get.adobe.com/reader/otherversions/ It got pulled. Of course, support for PDF's forms varies greatly outside of official Adobe product. A better alternative might be the MuPDF backend for KDE Okular, or for Zathura, neither of which are packaged anywhere in the Linux world yet. Of course, this could be built natively, without needing the Linux runtime support. I've noticed that rendering performance seems disappointing in the open source Linux PDF readers, as compared to OS X Preview.app on similar hardware. 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
svn commit: r283555 - head/sys/net80211
Author: adrian
Date: Tue May 26 04:25:11 2015
New Revision: 283555
URL: https://svnweb.freebsd.org/changeset/base/283555
Log:
Migrate the 802.11s locks out into ieee80211_freebsd.h, so they can be
defined as platform specific bits.
This is to help make OS portability easier.
Modified:
head/sys/net80211/ieee80211_freebsd.h
head/sys/net80211/ieee80211_mesh.c
head/sys/net80211/ieee80211_mesh.h
Modified: head/sys/net80211/ieee80211_freebsd.h
==
--- head/sys/net80211/ieee80211_freebsd.h Tue May 26 03:58:18 2015
(r283554)
+++ head/sys/net80211/ieee80211_freebsd.h Tue May 26 04:25:11 2015
(r283555)
@@ -184,6 +184,27 @@ typedef struct mtx ieee80211_scan_table_
#defineIEEE80211_SCAN_TABLE_UNLOCK(_st)
mtx_unlock((_st)-st_lock)
/*
+ * Mesh node/routing definitions.
+ */
+typedef struct mtx ieee80211_rte_lock_t;
+#defineMESH_RT_ENTRY_LOCK_INIT(_rt, _name) \
+ mtx_init((rt)-rt_lock, _name, 802.11s route entry, MTX_DEF)
+#defineMESH_RT_ENTRY_LOCK_DESTROY(_rt) \
+ mtx_destroy((_rt)-rt_lock)
+#defineMESH_RT_ENTRY_LOCK(rt) mtx_lock((rt)-rt_lock)
+#defineMESH_RT_ENTRY_LOCK_ASSERT(rt) mtx_assert((rt)-rt_lock,
MA_OWNED)
+#defineMESH_RT_ENTRY_UNLOCK(rt)mtx_unlock((rt)-rt_lock)
+
+typedef struct mtx ieee80211_rt_lock_t;
+#defineMESH_RT_LOCK(ms)mtx_lock((ms)-ms_rt_lock)
+#defineMESH_RT_LOCK_ASSERT(ms) mtx_assert((ms)-ms_rt_lock, MA_OWNED)
+#defineMESH_RT_UNLOCK(ms) mtx_unlock((ms)-ms_rt_lock)
+#defineMESH_RT_LOCK_INIT(ms, name) \
+ mtx_init((ms)-ms_rt_lock, name, 802.11s routing table, MTX_DEF)
+#defineMESH_RT_LOCK_DESTROY(ms) \
+ mtx_destroy((ms)-ms_rt_lock)
+
+/*
* Node reference counting definitions.
*
* ieee80211_node_initref initialize the reference count to 1
Modified: head/sys/net80211/ieee80211_mesh.c
==
--- head/sys/net80211/ieee80211_mesh.c Tue May 26 03:58:18 2015
(r283554)
+++ head/sys/net80211/ieee80211_mesh.c Tue May 26 04:25:11 2015
(r283555)
@@ -162,14 +162,6 @@ static const struct ieee80211_mesh_proto
static struct ieee80211_mesh_proto_pathmesh_proto_paths[4];
static struct ieee80211_mesh_proto_metric mesh_proto_metrics[4];
-#defineRT_ENTRY_LOCK(rt) mtx_lock((rt)-rt_lock)
-#defineRT_ENTRY_LOCK_ASSERT(rt) mtx_assert((rt)-rt_lock, MA_OWNED)
-#defineRT_ENTRY_UNLOCK(rt) mtx_unlock((rt)-rt_lock)
-
-#defineMESH_RT_LOCK(ms)mtx_lock((ms)-ms_rt_lock)
-#defineMESH_RT_LOCK_ASSERT(ms) mtx_assert((ms)-ms_rt_lock, MA_OWNED)
-#defineMESH_RT_UNLOCK(ms) mtx_unlock((ms)-ms_rt_lock)
-
MALLOC_DEFINE(M_80211_MESH_PREQ, 80211preq, 802.11 MESH Path Request
frame);
MALLOC_DEFINE(M_80211_MESH_PREP, 80211prep, 802.11 MESH Path Reply frame);
MALLOC_DEFINE(M_80211_MESH_PERR, 80211perr, 802.11 MESH Path Error frame);
@@ -218,7 +210,7 @@ mesh_rt_add_locked(struct ieee80211vap *
rt-rt_vap = vap;
IEEE80211_ADDR_COPY(rt-rt_dest, dest);
rt-rt_priv = (void *)ALIGN(rt[1]);
- mtx_init(rt-rt_lock, MBSS_RT, 802.11s route entry,
MTX_DEF);
+ MESH_RT_ENTRY_LOCK_INIT(rt, MBSS_RT);
callout_init(rt-rt_discovery, 1);
rt-rt_updtime = ticks; /* create time */
TAILQ_INSERT_TAIL(ms-ms_routes, rt, rt_next);
@@ -271,11 +263,11 @@ ieee80211_mesh_rt_update(struct ieee8021
KASSERT(rt != NULL, (route is NULL));
now = ticks;
- RT_ENTRY_LOCK(rt);
+ MESH_RT_ENTRY_LOCK(rt);
/* dont clobber a proxy entry gated by us */
if (rt-rt_flags IEEE80211_MESHRT_FLAGS_PROXY rt-rt_nhops == 0) {
- RT_ENTRY_UNLOCK(rt);
+ MESH_RT_ENTRY_UNLOCK(rt);
return rt-rt_lifetime;
}
@@ -296,7 +288,7 @@ ieee80211_mesh_rt_update(struct ieee8021
new_lifetime, rt-rt_lifetime);
}
lifetime = rt-rt_lifetime;
- RT_ENTRY_UNLOCK(rt);
+ MESH_RT_ENTRY_UNLOCK(rt);
return lifetime;
}
@@ -358,9 +350,9 @@ mesh_rt_del(struct ieee80211_mesh_state
* Grab the lock before destroying it, to be sure no one else
* is holding the route.
*/
- RT_ENTRY_LOCK(rt);
+ MESH_RT_ENTRY_LOCK(rt);
callout_drain(rt-rt_discovery);
- mtx_destroy(rt-rt_lock);
+ MESH_RT_ENTRY_LOCK_DESTROY(rt);
IEEE80211_FREE(rt, M_80211_MESH_RT);
}
@@ -653,7 +645,7 @@ mesh_vdetach(struct ieee80211vap *vap)
ieee80211_iterate_nodes(vap-iv_ic-ic_sta, mesh_vdetach_peers,
NULL);
ieee80211_mesh_rt_flush(vap);
- mtx_destroy(ms-ms_rt_lock);
+ MESH_RT_LOCK_DESTROY(ms);
svn commit: r283556 - head/sys/net80211
Author: adrian
Date: Tue May 26 04:37:59 2015
New Revision: 283556
URL: https://svnweb.freebsd.org/changeset/base/283556
Log:
Migrate the scan iterator lock typedefs out into ieee80211_freebsd.h.
This removes the final piece of freebsd-specific locking that snuck
into the tree - it's now all inside ieee80211_freebsd.h.
Modified:
head/sys/net80211/ieee80211_freebsd.h
head/sys/net80211/ieee80211_scan_sta.c
Modified: head/sys/net80211/ieee80211_freebsd.h
==
--- head/sys/net80211/ieee80211_freebsd.h Tue May 26 04:25:11 2015
(r283555)
+++ head/sys/net80211/ieee80211_freebsd.h Tue May 26 04:37:59 2015
(r283556)
@@ -183,6 +183,13 @@ typedef struct mtx ieee80211_scan_table_
#defineIEEE80211_SCAN_TABLE_LOCK(_st)
mtx_lock((_st)-st_lock)
#defineIEEE80211_SCAN_TABLE_UNLOCK(_st)
mtx_unlock((_st)-st_lock)
+typedef struct mtx ieee80211_scan_iter_lock_t;
+#defineIEEE80211_SCAN_ITER_LOCK_INIT(_st, _name) \
+ mtx_init((_st)-st_scanlock, _name, 802.11 scangen, MTX_DEF)
+#defineIEEE80211_SCAN_ITER_LOCK_DESTROY(_st)
mtx_destroy((_st)-st_scanlock)
+#defineIEEE80211_SCAN_ITER_LOCK(_st)
mtx_lock((_st)-st_scanlock)
+#defineIEEE80211_SCAN_ITER_UNLOCK(_st) mtx_unlock((_st)-st_scanlock)
+
/*
* Mesh node/routing definitions.
*/
Modified: head/sys/net80211/ieee80211_scan_sta.c
==
--- head/sys/net80211/ieee80211_scan_sta.c Tue May 26 04:25:11 2015
(r283555)
+++ head/sys/net80211/ieee80211_scan_sta.c Tue May 26 04:37:59 2015
(r283556)
@@ -102,7 +102,7 @@ struct sta_table {
ieee80211_scan_table_lock_t st_lock;/* on scan table */
TAILQ_HEAD(, sta_entry) st_entry; /* all entries */
LIST_HEAD(, sta_entry) st_hash[STA_HASHSIZE];
- struct mtx st_scanlock;/* on st_scaniter */
+ ieee80211_scan_iter_lock_t st_scanlock; /* on st_scaniter */
u_int st_scaniter;/* gen# for iterator */
u_int st_scangen; /* scan generation # */
int st_newscan;
@@ -165,7 +165,7 @@ sta_attach(struct ieee80211_scan_state *
if (st == NULL)
return 0;
IEEE80211_SCAN_TABLE_LOCK_INIT(st, scantable);
- mtx_init(st-st_scanlock, scangen, 802.11 scangen, MTX_DEF);
+ IEEE80211_SCAN_ITER_LOCK_INIT(st, scangen);
TAILQ_INIT(st-st_entry);
ss-ss_priv = st;
nrefs++;/* NB: we assume caller locking */
@@ -183,7 +183,7 @@ sta_detach(struct ieee80211_scan_state *
if (st != NULL) {
sta_flush_table(st);
IEEE80211_SCAN_TABLE_LOCK_DESTROY(st);
- mtx_destroy(st-st_scanlock);
+ IEEE80211_SCAN_ITER_LOCK_DESTROY(st);
IEEE80211_FREE(st, M_80211_SCAN);
KASSERT(nrefs 0, (imbalanced attach/detach));
nrefs--;/* NB: we assume caller locking */
@@ -1403,7 +1403,7 @@ sta_iterate(struct ieee80211_scan_state
struct sta_entry *se;
u_int gen;
- mtx_lock(st-st_scanlock);
+ IEEE80211_SCAN_ITER_LOCK(st);
gen = st-st_scaniter++;
restart:
IEEE80211_SCAN_TABLE_LOCK(st);
@@ -1419,7 +1419,7 @@ restart:
}
IEEE80211_SCAN_TABLE_UNLOCK(st);
- mtx_unlock(st-st_scanlock);
+ IEEE80211_SCAN_ITER_UNLOCK(st);
}
static void
___
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
