Re: svn commit: r272673 - in head: include lib/libc/string sys/conf sys/libkern sys/sys
Xin LI writes: > Log: > Add explicit_bzero(3) and its kernel counterpart. I would much prefer that we add memset_s(3) from C11, which predates explicit_bzero(3) by either three or five years (depending on whether you count from the publication date or the proposal date). In the longer term, we should also consider adding the rest of annex K. Here's a patch for NetBSD (which was never accepted): https://mail-index.netbsd.org/tech-userlevel/2012/02/24/msg006125.html I realize that these extensions are controversial, but I still believe that having them is better than not having them. As the WG points out, the intention with gets_s(3) and scanf_s(3) is not to encourage the use of those interfaces, but to make it easier to retrofit code that uses gets(3) and scanf(3). DES -- Dag-Erling Smørgrav - d...@des.no ___ 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: r272683 - in head/sys/ofed: drivers/net/mlx4 include/linux/mlx4
Author: hselasky Date: Tue Oct 7 10:04:25 2014 New Revision: 272683 URL: https://svnweb.freebsd.org/changeset/base/272683 Log: Fix compile warning when compiling with GCC. MFC after:3 days Sponsored by: Mellanox Technologies Modified: head/sys/ofed/drivers/net/mlx4/mlx4.h head/sys/ofed/include/linux/mlx4/device.h Modified: head/sys/ofed/drivers/net/mlx4/mlx4.h == --- head/sys/ofed/drivers/net/mlx4/mlx4.h Tue Oct 7 08:23:37 2014 (r272682) +++ head/sys/ofed/drivers/net/mlx4/mlx4.h Tue Oct 7 10:04:25 2014 (r272683) @@ -1234,6 +1234,7 @@ int mlx4_trans_to_dmfs_attach(struct mlx u8 gid[16], u8 port, int block_mcast_loopback, enum mlx4_protocol prot, u64 *reg_id); +int mlx4_SET_MCAST_FLTR(struct mlx4_dev *dev, u8 port, u64 mac, u64 clear, u8 mode); int mlx4_SET_MCAST_FLTR_wrapper(struct mlx4_dev *dev, int slave, struct mlx4_vhcr *vhcr, struct mlx4_cmd_mailbox *inbox, Modified: head/sys/ofed/include/linux/mlx4/device.h == --- head/sys/ofed/include/linux/mlx4/device.h Tue Oct 7 08:23:37 2014 (r272682) +++ head/sys/ofed/include/linux/mlx4/device.h Tue Oct 7 10:04:25 2014 (r272683) @@ -1208,7 +1208,6 @@ int mlx4_multicast_promisc_add(struct ml int mlx4_multicast_promisc_remove(struct mlx4_dev *dev, u32 qpn, u8 port); int mlx4_unicast_promisc_add(struct mlx4_dev *dev, u32 qpn, u8 port); int mlx4_unicast_promisc_remove(struct mlx4_dev *dev, u32 qpn, u8 port); -int mlx4_SET_MCAST_FLTR(struct mlx4_dev *dev, u8 port, u64 mac, u64 clear, u8 mode); int mlx4_register_mac(struct mlx4_dev *dev, u8 port, u64 mac); void mlx4_unregister_mac(struct mlx4_dev *dev, u8 port, u64 mac); ___ 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: r272695 - head/sys/net
Author: ae
Date: Tue Oct 7 13:31:04 2014
New Revision: 272695
URL: https://svnweb.freebsd.org/changeset/base/272695
Log:
Our packet filters use mbuf's rcvif pointer to determine incoming interface.
Change mbuf's rcvif to enc0 and restore it after pfil processing.
PR: 110959
Sponsored by: Yandex LLC
Modified:
head/sys/net/if_enc.c
Modified: head/sys/net/if_enc.c
==
--- head/sys/net/if_enc.c Tue Oct 7 13:30:42 2014(r272694)
+++ head/sys/net/if_enc.c Tue Oct 7 13:31:04 2014(r272695)
@@ -230,6 +230,7 @@ ipsec_filter(struct mbuf **mp, int dir,
{
int error, i;
struct ip *ip;
+ struct ifnet *rcvif;
KASSERT(encif != NULL, ("%s: encif is null", __func__));
KASSERT(flags & (ENC_IN|ENC_OUT),
@@ -268,6 +269,8 @@ ipsec_filter(struct mbuf **mp, int dir,
}
error = 0;
+ rcvif = (*mp)->m_pkthdr.rcvif;
+ (*mp)->m_pkthdr.rcvif = encif;
ip = mtod(*mp, struct ip *);
switch (ip->ip_v) {
#ifdef INET
@@ -298,6 +301,7 @@ ipsec_filter(struct mbuf **mp, int dir,
if (error != 0)
goto bad;
+ (*mp)->m_pkthdr.rcvif = rcvif;
return (error);
bad:
___
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: r272695 - head/sys/net
On Tue, 7 Oct 2014 13:31:05 +
"Andrey V. Elsukov" wrote:
> Author: ae
> Date: Tue Oct 7 13:31:04 2014
> New Revision: 272695
> URL: https://svnweb.freebsd.org/changeset/base/272695
>
> Log:
> Our packet filters use mbuf's rcvif pointer to determine incoming
> interface. Change mbuf's rcvif to enc0 and restore it after pfil
> processing.
Will this work / was this tested with options VIMAGE, where
m_pkthdr.rcvif->if_vnet will no longer match curvnet, except in vnet0?
Marko
> PR: 110959
> Sponsored by: Yandex LLC
>
> Modified:
> head/sys/net/if_enc.c
>
> Modified: head/sys/net/if_enc.c
> ==
> --- head/sys/net/if_enc.c Tue Oct 7 13:30:42 2014
> (r272694) +++ head/sys/net/if_enc.c Tue Oct 7 13:31:04
> 2014 (r272695) @@ -230,6 +230,7 @@ ipsec_filter(struct mbuf
> **mp, int dir, {
> int error, i;
> struct ip *ip;
> + struct ifnet *rcvif;
>
> KASSERT(encif != NULL, ("%s: encif is null", __func__));
> KASSERT(flags & (ENC_IN|ENC_OUT),
> @@ -268,6 +269,8 @@ ipsec_filter(struct mbuf **mp, int dir,
> }
>
> error = 0;
> + rcvif = (*mp)->m_pkthdr.rcvif;
> + (*mp)->m_pkthdr.rcvif = encif;
> ip = mtod(*mp, struct ip *);
> switch (ip->ip_v) {
> #ifdef INET
> @@ -298,6 +301,7 @@ ipsec_filter(struct mbuf **mp, int dir,
> if (error != 0)
> goto bad;
>
> + (*mp)->m_pkthdr.rcvif = rcvif;
> return (error);
>
> bad:
>
___
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: r272701 - head/sys/cddl/compat/opensolaris/sys
Author: avg
Date: Tue Oct 7 14:15:50 2014
New Revision: 272701
URL: https://svnweb.freebsd.org/changeset/base/272701
Log:
make userland __assfail from opensolaris compat honor 'aok' variable
This should allow zdb -A option to actually make difference.
MFC after:2 weeks
Modified:
head/sys/cddl/compat/opensolaris/sys/assfail.h
Modified: head/sys/cddl/compat/opensolaris/sys/assfail.h
==
--- head/sys/cddl/compat/opensolaris/sys/assfail.h Tue Oct 7 14:09:05
2014(r272700)
+++ head/sys/cddl/compat/opensolaris/sys/assfail.h Tue Oct 7 14:15:50
2014(r272701)
@@ -46,20 +46,24 @@ void assfail3(const char *, uintmax_t, c
#else /* !defined(_KERNEL) */
#ifndef HAVE_ASSFAIL
+extern int aok;
+
static __inline int
__assfail(const char *expr, const char *file, int line)
{
(void)fprintf(stderr, "Assertion failed: (%s), file %s, line %d.\n",
expr, file, line);
- abort();
- /* NOTREACHED */
+ if (!aok)
+ abort();
return (0);
}
#define assfail __assfail
#endif
#ifndef HAVE_ASSFAIL3
+extern int aok;
+
static __inline void
__assfail3(const char *expr, uintmax_t lv, const char *op, uintmax_t rv,
const char *file, int line) {
@@ -67,8 +71,8 @@ __assfail3(const char *expr, uintmax_t l
(void)fprintf(stderr,
"Assertion failed: %s (0x%jx %s 0x%jx), file %s, line %d.\n",
expr, lv, op, rv, file, line);
-abort();
-/* NOTREACHED */
+ if (!aok)
+ abort();
}
#define assfail3 __assfail3
#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: r272695 - head/sys/net
On 07.10.2014 18:04, Marko Zec wrote: > On Tue, 7 Oct 2014 13:31:05 + > "Andrey V. Elsukov" wrote: > >> Author: ae >> Date: Tue Oct 7 13:31:04 2014 >> New Revision: 272695 >> URL: https://svnweb.freebsd.org/changeset/base/272695 >> >> Log: >> Our packet filters use mbuf's rcvif pointer to determine incoming >> interface. Change mbuf's rcvif to enc0 and restore it after pfil >> processing. > > Will this work / was this tested with options VIMAGE, where > m_pkthdr.rcvif->if_vnet will no longer match curvnet, except in vnet0? I tested only without VIMAGE. ipfw and pf use if_xname field to compare interfaces. So will this work? -- WBR, Andrey V. Elsukov ___ 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: r272702 - head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs
Author: avg Date: Tue Oct 7 14:30:24 2014 New Revision: 272702 URL: https://svnweb.freebsd.org/changeset/base/272702 Log: reduce L2ARC_WRITE_SIZE on FreeBSD FreeBSD has ARC_BUFC_NUMMETADATALISTS metadata lists and ARC_BUFC_NUMDATALISTS data lists (currently both are 16) while illumos has just a single list of each kind. L2ARC_WRITE_SIZE determines the default value of l2arc_write_max which defines limits on how much data is scanned and written to a cache device during each run of the l2arc feed thread. The limits are applied on the per buffer list basis. Because FreeBSD has more lists we proportionally reduce the limits. Reviewed by: Brendan Gregg (earlier version) MFC after:2 weeks Sponsored by: HybridCluster Modified: head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/arc.c Modified: head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/arc.c == --- head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/arc.c Tue Oct 7 14:15:50 2014(r272701) +++ head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/arc.c Tue Oct 7 14:30:24 2014(r272702) @@ -711,7 +711,9 @@ uint64_t zfs_crc64_table[256]; * Level 2 ARC */ -#defineL2ARC_WRITE_SIZE(8 * 1024 * 1024) /* initial write max */ +/* initial write max */ +#defineL2ARC_WRITE_SIZE\ +(8 * 1024 * 1024 * 2 / (ARC_BUFC_NUMMETADATALISTS + ARC_BUFC_NUMDATALISTS)) #defineL2ARC_HEADROOM 2 /* num of writes */ /* * If we discover during ARC scan any buffers to be compressed, we boost ___ 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: r272695 - head/sys/net
On Tue, 7 Oct 2014 18:17:18 +0400 "Andrey V. Elsukov" wrote: > On 07.10.2014 18:04, Marko Zec wrote: > > On Tue, 7 Oct 2014 13:31:05 + > > "Andrey V. Elsukov" wrote: > > > >> Author: ae > >> Date: Tue Oct 7 13:31:04 2014 > >> New Revision: 272695 > >> URL: https://svnweb.freebsd.org/changeset/base/272695 > >> > >> Log: > >> Our packet filters use mbuf's rcvif pointer to determine incoming > >> interface. Change mbuf's rcvif to enc0 and restore it after pfil > >> processing. > > > > Will this work / was this tested with options VIMAGE, where > > m_pkthdr.rcvif->if_vnet will no longer match curvnet, except in > > vnet0? > > I tested only without VIMAGE. ipfw and pf use if_xname field to > compare interfaces. So will this work? I have no idea whether this would work now, but this change implies that no pfil consumer should reference m_pkthdr.rcvif->if_vnet from now on, ever. Which doesn't seem right to me. If changing m_pkthdr.rcvif to enc0 in ipsec_filter() is really unavoidable, perhaps we could introduce enc0 for each vnet, maybe in a similar manner how hrs@ virtualized gif (271917) and gre (271918) cloners, which (gif) apparently seem to be at the root of the PR 110959 referenced here. Marko ___ 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: r272706 - head/sys/netinet6
Author: tuexen
Date: Tue Oct 7 16:01:17 2014
New Revision: 272706
URL: https://svnweb.freebsd.org/changeset/base/272706
Log:
Fix a bug introduced in
https://svnweb.freebsd.org/base?view=revision&revision=272347
MFC after: 3 days
Modified:
head/sys/netinet6/udp6_usrreq.c
Modified: head/sys/netinet6/udp6_usrreq.c
==
--- head/sys/netinet6/udp6_usrreq.c Tue Oct 7 15:21:20 2014
(r272705)
+++ head/sys/netinet6/udp6_usrreq.c Tue Oct 7 16:01:17 2014
(r272706)
@@ -265,7 +265,7 @@ udp6_input(struct mbuf **mp, int *offp,
if (uh_sum != 0) {
UDPSTAT_INC(udps_badsum);
- /*goto badunlocked;*/
+ goto badunlocked;
}
/*
___
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: r272707 - head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs
Author: avg Date: Tue Oct 7 16:06:10 2014 New Revision: 272707 URL: https://svnweb.freebsd.org/changeset/base/272707 Log: revert r272702: wrong (earlier) change was committed Modified: head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/arc.c Modified: head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/arc.c == --- head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/arc.c Tue Oct 7 16:01:17 2014(r272706) +++ head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/arc.c Tue Oct 7 16:06:10 2014(r272707) @@ -711,9 +711,7 @@ uint64_t zfs_crc64_table[256]; * Level 2 ARC */ -/* initial write max */ -#defineL2ARC_WRITE_SIZE\ -(8 * 1024 * 1024 * 2 / (ARC_BUFC_NUMMETADATALISTS + ARC_BUFC_NUMDATALISTS)) +#defineL2ARC_WRITE_SIZE(8 * 1024 * 1024) /* initial write max */ #defineL2ARC_HEADROOM 2 /* num of writes */ /* * If we discover during ARC scan any buffers to be compressed, we boost ___ 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: r272708 - head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs
Author: avg Date: Tue Oct 7 16:08:21 2014 New Revision: 272708 URL: https://svnweb.freebsd.org/changeset/base/272708 Log: l2arc_write_buffers: reduce headroom value FreeBSD has ARC_BUFC_NUMMETADATALISTS metadata lists and ARC_BUFC_NUMDATALISTS data lists (currently both are 16) while illumos has just a single list of each kind. headroom determines how much data is scanned on a single list during each run of the l2arc feed thread. Because FreeBSD has more lists we proportionally decrease the limit. Reviewed by: Brendan Gregg (earlier version) MFC after:2 weeks Sponsored by: HybridCluster Modified: head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/arc.c Modified: head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/arc.c == --- head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/arc.c Tue Oct 7 16:06:10 2014(r272707) +++ head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/arc.c Tue Oct 7 16:08:21 2014(r272708) @@ -5047,7 +5047,7 @@ l2arc_write_buffers(spa_t *spa, l2arc_de if (ab == NULL) ARCSTAT_BUMP(arcstat_l2_write_buffer_list_null_iter); - headroom = target_sz * l2arc_headroom; + headroom = target_sz * l2arc_headroom * 2 / ARC_BUFC_NUMLISTS; if (do_headroom_boost) headroom = (headroom * l2arc_headroom_boost) / 100; ___ 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: r272596 - head/sys/fs/devfs
On Monday, October 06, 2014 8:52:37 pm Mateusz Guzik wrote:
> On Mon, Oct 06, 2014 at 11:37:32AM -0400, John Baldwin wrote:
> > On Monday, October 06, 2014 06:20:36 AM Mateusz Guzik wrote:
> > > Author: mjg
> > > Date: Mon Oct 6 06:20:35 2014
> > > New Revision: 272596
> > > URL: https://svnweb.freebsd.org/changeset/base/272596
> > >
> > > Log:
> > > devfs: don't take proctree_lock unconditionally in devfs_close
> > >
> > > MFC after: 1 week
> >
> > Just for my sanity:
> >
> > What keeps td->td_proc->p_session static in this case so that it is safe to
> > dereference it? Specifically, if you are preempted after reading p_session
> > but before you then read s_ttyvp, what prevents a race with another thread
> > changing the session of td->td_proc?
> >
>
> Right, it's buggy.
>
> Turns out devfs was quite liberal in relation to that even prior to my
> change.
>
> How about:
>
> diff --git a/sys/fs/devfs/devfs_vnops.c b/sys/fs/devfs/devfs_vnops.c
> index d7009a4..a480e4f 100644
> --- a/sys/fs/devfs/devfs_vnops.c
> +++ b/sys/fs/devfs/devfs_vnops.c
> @@ -499,6 +499,7 @@ devfs_access(struct vop_access_args *ap)
> {
> struct vnode *vp = ap->a_vp;
> struct devfs_dirent *de;
> + struct proc *p;
> int error;
>
> de = vp->v_data;
> @@ -511,11 +512,14 @@ devfs_access(struct vop_access_args *ap)
> return (0);
> if (error != EACCES)
> return (error);
> + p = ap->a_td->td_proc;
> /* We do, however, allow access to the controlling terminal */
> - if (!(ap->a_td->td_proc->p_flag & P_CONTROLT))
> + if (!(p->p_flag & P_CONTROLT))
> return (error);
> - if (ap->a_td->td_proc->p_session->s_ttydp == de->de_cdp)
> - return (0);
> + PROC_LOCK(p);
> + if (p->p_session->s_ttydp == de->de_cdp)
> + error = 0;
> + PROC_UNLOCK(p);
> return (error);
> }
>
> @@ -525,6 +529,7 @@ devfs_close(struct vop_close_args *ap)
> {
> struct vnode *vp = ap->a_vp, *oldvp;
> struct thread *td = ap->a_td;
> + struct proc *p;
> struct cdev *dev = vp->v_rdev;
> struct cdevsw *dsw;
> int vp_locked, error, ref;
> @@ -545,24 +550,30 @@ devfs_close(struct vop_close_args *ap)
>* if the reference count is 2 (this last descriptor
>* plus the session), release the reference from the session.
>*/
> - if (td && vp == td->td_proc->p_session->s_ttyvp) {
> - oldvp = NULL;
> - sx_xlock(&proctree_lock);
> - if (vp == td->td_proc->p_session->s_ttyvp) {
> - SESS_LOCK(td->td_proc->p_session);
> - VI_LOCK(vp);
> - if (count_dev(dev) == 2 &&
> - (vp->v_iflag & VI_DOOMED) == 0) {
> - td->td_proc->p_session->s_ttyvp = NULL;
> - td->td_proc->p_session->s_ttydp = NULL;
> - oldvp = vp;
> + if (td != NULL) {
> + p = td->td_proc;
> + PROC_LOCK(p);
> + if (vp == p->p_session->s_ttyvp) {
> + PROC_UNLOCK(p);
> + oldvp = NULL;
> + sx_xlock(&proctree_lock);
> + if (vp == p->p_session->s_ttyvp) {
> + SESS_LOCK(p->p_session);
> + VI_LOCK(vp);
> + if (count_dev(dev) == 2 &&
> + (vp->v_iflag & VI_DOOMED) == 0) {
> + p->p_session->s_ttyvp = NULL;
> + p->p_session->s_ttydp = NULL;
> + oldvp = vp;
> + }
> + VI_UNLOCK(vp);
> + SESS_UNLOCK(p->p_session);
> }
> - VI_UNLOCK(vp);
> - SESS_UNLOCK(td->td_proc->p_session);
> - }
> - sx_xunlock(&proctree_lock);
> - if (oldvp != NULL)
> - vrele(oldvp);
> + sx_xunlock(&proctree_lock);
> + if (oldvp != NULL)
> + vrele(oldvp);
> + } else
> + PROC_UNLOCK(p);
> }
> /*
>* We do not want to really close the device if it
> @@ -814,6 +825,7 @@ devfs_prison_check(struct devfs_dirent *de, struct thread
> *td)
> {
> struct cdev_priv *cdp;
> struct ucred *dcr;
> + struct proc *p;
> int error;
>
> cdp = de->de_cdp;
> @@ -827,10 +839,13 @@ devfs_prison_check(struct devfs_dirent *de, struct
> thread *td)
> if (error == 0)
> return (0);
> /* We do, however, allow access to the controlling terminal */
> - if (!(td->td_proc->p_flag & P_CONTROLT))
> + p = td->td_proc;
> + if (!(p->p_flag & P_CONTROLT))
> return
Re: svn commit: r272281 - head/lib/libpam/modules/pam_login_access
Dimitry Andric writes: > And why is 'user' not a char * to begin with? :) man 3 pam_get_item + C rules about casting pointers. DES -- Dag-Erling Smørgrav - d...@des.no ___ 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: r272281 - head/lib/libpam/modules/pam_login_access
Bruce Evans writes: > Other bugs in PAM_VERBOSE_ERROR()'s function [...] PAM_VERBOSE_ERROR should die. The only reason I didn't kill it way back when was to avoid extensive modifications to existing modules. DES -- Dag-Erling Smørgrav - d...@des.no ___ 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: r272710 - head/usr.sbin/bhyve
Author: neel
Date: Tue Oct 7 17:08:53 2014
New Revision: 272710
URL: https://svnweb.freebsd.org/changeset/base/272710
Log:
Implement the FLUSH operation in the virtio-block emulation.
This gets rid of the following error message during FreeBSD guest bootup:
"vtbd0: hard error cmd=flush fsbn 0"
Reported by: rodrigc
Reviewed by: grehan
Modified:
head/usr.sbin/bhyve/pci_virtio_block.c
Modified: head/usr.sbin/bhyve/pci_virtio_block.c
==
--- head/usr.sbin/bhyve/pci_virtio_block.c Tue Oct 7 16:46:11 2014
(r272709)
+++ head/usr.sbin/bhyve/pci_virtio_block.c Tue Oct 7 17:08:53 2014
(r272710)
@@ -94,6 +94,8 @@ struct vtblk_config {
struct virtio_blk_hdr {
#defineVBH_OP_READ 0
#defineVBH_OP_WRITE1
+#defineVBH_OP_FLUSH4
+#defineVBH_OP_FLUSH_OUT5
#defineVBH_OP_IDENT8
#defineVBH_FLAG_BARRIER0x8000 /* OR'ed into vbh_type
*/
uint32_tvbh_type;
@@ -217,6 +219,10 @@ pci_vtblk_proc(struct pci_vtblk_softc *s
MIN(iov[1].iov_len, sizeof(sc->vbsc_ident)));
err = 0;
break;
+ case VBH_OP_FLUSH:
+ case VBH_OP_FLUSH_OUT:
+ err = fsync(sc->vbsc_fd);
+ break;
default:
err = -ENOSYS;
break;
___
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: r272712 - in head/sys: arm/altera/socfpga arm/conf arm/samsung/exynos boot/fdt/dts/arm dev/mmc dev/mmc/host
Author: br
Date: Tue Oct 7 17:39:30 2014
New Revision: 272712
URL: https://svnweb.freebsd.org/changeset/base/272712
Log:
Add driver for Synopsys DesignWare Mobile Storage Host Controller.
Sponsored by: DARPA, AFRL
Added:
head/sys/dev/mmc/host/
head/sys/dev/mmc/host/dwmmc.c (contents, props changed)
head/sys/dev/mmc/host/dwmmc.h (contents, props changed)
Modified:
head/sys/arm/altera/socfpga/files.socfpga
head/sys/arm/altera/socfpga/socfpga_machdep.c
head/sys/arm/conf/EXYNOS5.common
head/sys/arm/conf/SOCKIT
head/sys/arm/samsung/exynos/exynos5_machdep.c
head/sys/arm/samsung/exynos/files.exynos5
head/sys/boot/fdt/dts/arm/exynos5420-arndale-octa.dts
head/sys/boot/fdt/dts/arm/exynos5420-peach-pit.dts
head/sys/boot/fdt/dts/arm/exynos5420.dtsi
head/sys/boot/fdt/dts/arm/socfpga-sockit.dts
head/sys/boot/fdt/dts/arm/socfpga.dtsi
head/sys/dev/mmc/mmc.c
Modified: head/sys/arm/altera/socfpga/files.socfpga
==
--- head/sys/arm/altera/socfpga/files.socfpga Tue Oct 7 17:23:11 2014
(r272711)
+++ head/sys/arm/altera/socfpga/files.socfpga Tue Oct 7 17:39:30 2014
(r272712)
@@ -19,3 +19,4 @@ arm/altera/socfpga/socfpga_manager.c st
arm/altera/socfpga/socfpga_rstmgr.cstandard
dev/dwc/if_dwc.c optional dwc
+dev/mmc/host/dwmmc.c optional dwmmc
Modified: head/sys/arm/altera/socfpga/socfpga_machdep.c
==
--- head/sys/arm/altera/socfpga/socfpga_machdep.c Tue Oct 7 17:23:11
2014(r272711)
+++ head/sys/arm/altera/socfpga/socfpga_machdep.c Tue Oct 7 17:39:30
2014(r272712)
@@ -89,6 +89,9 @@ platform_devmap_init(void)
*/
arm_devmap_add_entry(0xffb0, 0x10);
+ /* dwmmc */
+ arm_devmap_add_entry(0xff70, 0x10);
+
return (0);
}
Modified: head/sys/arm/conf/EXYNOS5.common
==
--- head/sys/arm/conf/EXYNOS5.commonTue Oct 7 17:23:11 2014
(r272711)
+++ head/sys/arm/conf/EXYNOS5.commonTue Oct 7 17:39:30 2014
(r272712)
@@ -80,7 +80,7 @@ options NFS_ROOT# NFS usable as /, re
device mmc # mmc/sd bus
device mmcsd # mmc/sd flash cards
-device sdhci # generic sdhci
+device dwmmc
optionsROOTDEVNAME=\"ufs:/dev/da0\"
Modified: head/sys/arm/conf/SOCKIT
==
--- head/sys/arm/conf/SOCKITTue Oct 7 17:23:11 2014(r272711)
+++ head/sys/arm/conf/SOCKITTue Oct 7 17:39:30 2014(r272712)
@@ -82,7 +82,7 @@ options NFS_ROOT# NFS usable as /, re
device mmc # mmc/sd bus
device mmcsd # mmc/sd flash cards
-device sdhci # generic sdhci
+device dwmmc
optionsROOTDEVNAME=\"ufs:/dev/da0\"
Modified: head/sys/arm/samsung/exynos/exynos5_machdep.c
==
--- head/sys/arm/samsung/exynos/exynos5_machdep.c Tue Oct 7 17:23:11
2014(r272711)
+++ head/sys/arm/samsung/exynos/exynos5_machdep.c Tue Oct 7 17:39:30
2014(r272712)
@@ -78,6 +78,9 @@ platform_devmap_init(void)
/* UART */
arm_devmap_add_entry(0x12C0, 0x10);
+ /* DWMMC */
+ arm_devmap_add_entry(0x1220, 0x10);
+
return (0);
}
Modified: head/sys/arm/samsung/exynos/files.exynos5
==
--- head/sys/arm/samsung/exynos/files.exynos5 Tue Oct 7 17:23:11 2014
(r272711)
+++ head/sys/arm/samsung/exynos/files.exynos5 Tue Oct 7 17:39:30 2014
(r272712)
@@ -33,4 +33,4 @@ arm/samsung/exynos/chrome_ec.coptiona
arm/samsung/exynos/chrome_ec_spi.c optionalchrome_ec_spi
arm/samsung/exynos/chrome_kb.c optionalchrome_kb
-#dev/sdhci/sdhci_fdt.c optionalsdhci
+dev/mmc/host/dwmmc.c optionaldwmmc
Modified: head/sys/boot/fdt/dts/arm/exynos5420-arndale-octa.dts
==
--- head/sys/boot/fdt/dts/arm/exynos5420-arndale-octa.dts Tue Oct 7
17:23:11 2014(r272711)
+++ head/sys/boot/fdt/dts/arm/exynos5420-arndale-octa.dts Tue Oct 7
17:39:30 2014(r272712)
@@ -47,8 +47,19 @@
status = "okay";
};
- sdhci@1222 {
- status = "disabled";
+ mmc2: dwmmc@1222 {
+
svn commit: r272713 - head/sys/dev/hwpmc
Author: bz
Date: Tue Oct 7 18:00:34 2014
New Revision: 272713
URL: https://svnweb.freebsd.org/changeset/base/272713
Log:
Since introducing the extra mapping in r250103 for architectural performance
events we have actually counted 'Branch Instruction Retired' when people
asked for 'Unhalted core cycles' using the 'unhalted-core-cycles' event mask
mnemonic.
Reviewed by: jimharris
Discussed with: gnn, rwatson
MFC after:3 days
Sponsored by: DARPA/AFRL
Modified:
head/sys/dev/hwpmc/hwpmc_core.c
Modified: head/sys/dev/hwpmc/hwpmc_core.c
==
--- head/sys/dev/hwpmc/hwpmc_core.c Tue Oct 7 17:39:30 2014
(r272712)
+++ head/sys/dev/hwpmc/hwpmc_core.c Tue Oct 7 18:00:34 2014
(r272713)
@@ -1796,7 +1796,7 @@ iap_is_event_architectural(enum pmc_even
switch (pe) {
case PMC_EV_IAP_ARCH_UNH_COR_CYC:
ae = CORE_AE_UNHALTED_CORE_CYCLES;
- *map = PMC_EV_IAP_EVENT_C4H_00H;
+ *map = PMC_EV_IAP_EVENT_3CH_00H;
break;
case PMC_EV_IAP_ARCH_INS_RET:
ae = CORE_AE_INSTRUCTION_RETIRED;
___
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: r272715 - head/sys/dev/vt
Author: ray
Date: Tue Oct 7 18:47:53 2014
New Revision: 272715
URL: https://svnweb.freebsd.org/changeset/base/272715
Log:
Allow vt(4) to disable terminal bell with `sysctl kern.vt.bell_enable=0`,
similar as syscons(4) do.
Submitted by: Tiwei Bie
MFC after:2 weeks
Sponsored by: The FreeBSD Foundation
Modified:
head/sys/dev/vt/vt_core.c
Modified: head/sys/dev/vt/vt_core.c
==
--- head/sys/dev/vt/vt_core.c Tue Oct 7 18:22:05 2014(r272714)
+++ head/sys/dev/vt/vt_core.c Tue Oct 7 18:47:53 2014(r272715)
@@ -120,6 +120,7 @@ const struct terminal_class vt_termclass
static SYSCTL_NODE(_kern, OID_AUTO, vt, CTLFLAG_RD, 0, "vt(9) parameters");
VT_SYSCTL_INT(enable_altgr, 1, "Enable AltGr key (Do not assume R.Alt as
Alt)");
+VT_SYSCTL_INT(enable_bell, 1, "Enable bell");
VT_SYSCTL_INT(debug, 0, "vt(9) debug level");
VT_SYSCTL_INT(deadtimer, 15, "Time to wait busy process in VT_PROCESS mode");
VT_SYSCTL_INT(suspendswitch, 1, "Switch to VT0 before suspend");
@@ -904,6 +905,9 @@ vtterm_bell(struct terminal *tm)
struct vt_window *vw = tm->tm_softc;
struct vt_device *vd = vw->vw_device;
+ if (!vt_enable_bell)
+ return;
+
if (vd->vd_flags & VDF_QUIET_BELL)
return;
@@ -915,6 +919,9 @@ vtterm_beep(struct terminal *tm, u_int p
{
u_int freq, period;
+ if (!vt_enable_bell)
+ return;
+
if ((param == 0) || ((param & 0x) == 0)) {
vtterm_bell(tm);
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"
svn commit: r272716 - in head/sys/dev/usb: . net
Author: gavin
Date: Tue Oct 7 19:07:50 2014
New Revision: 272716
URL: https://svnweb.freebsd.org/changeset/base/272716
Log:
Support the Vodafone R215 LET USB dongle, which is apparently a rebadged
E5372 with different product IDs.
Interestingly, the standard E5372 IDs (12d1:1506) are currently listed in
u3g.c and are the same as the E3131. However, the R215/E5372 is an NCM
device and works well with cdce(4) whereas the E3131 isn't. More work
may be needed to better identify the other device IDs.
MFC after:1 week
Modified:
head/sys/dev/usb/net/if_cdce.c
head/sys/dev/usb/usbdevs
Modified: head/sys/dev/usb/net/if_cdce.c
==
--- head/sys/dev/usb/net/if_cdce.c Tue Oct 7 18:47:53 2014
(r272715)
+++ head/sys/dev/usb/net/if_cdce.c Tue Oct 7 19:07:50 2014
(r272716)
@@ -273,6 +273,7 @@ static const struct usb_ether_methods cd
static const STRUCT_USB_HOST_ID cdce_switch_devs[] = {
{USB_VPI(USB_VENDOR_HUAWEI, USB_PRODUCT_HUAWEI_E3272_INIT,
MSC_EJECT_HUAWEI2)},
+ {USB_VPI(USB_VENDOR_HUAWEI, USB_PRODUCT_HUAWEI_R215_INIT,
MSC_EJECT_HUAWEI2)},
};
static const STRUCT_USB_HOST_ID cdce_host_devs[] = {
Modified: head/sys/dev/usb/usbdevs
==
--- head/sys/dev/usb/usbdevsTue Oct 7 18:47:53 2014(r272715)
+++ head/sys/dev/usb/usbdevsTue Oct 7 19:07:50 2014(r272716)
@@ -2353,6 +2353,8 @@ product HUAWEI E3131 0x1506 3G modem
product HUAWEI K3765_INIT 0x1520 K3765 Initial
product HUAWEI K4505_INIT 0x1521 K4505 Initial
product HUAWEI E3272_INIT 0x155b LTE modem initial
+product HUAWEI R215_INIT 0x1582 LTE modem initial
+product HUAWEI R2150x1588 LTE modem
product HUAWEI ETS2055 0x1803 CDMA modem
product HUAWEI E1730x1c05 3G modem
product HUAWEI E173_INIT 0x1c0b 3G modem initial
___
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: r272717 - head/usr.sbin/autofs
Author: trasz
Date: Tue Oct 7 19:08:02 2014
New Revision: 272717
URL: https://svnweb.freebsd.org/changeset/base/272717
Log:
Remove call to access(2) which didn't serve any purpose, and make it more
tolerant to errors.
MFC after:1 month
Sponsored by: The FreeBSD Foundation
Modified:
head/usr.sbin/autofs/common.c
Modified: head/usr.sbin/autofs/common.c
==
--- head/usr.sbin/autofs/common.c Tue Oct 7 19:07:50 2014
(r272716)
+++ head/usr.sbin/autofs/common.c Tue Oct 7 19:08:02 2014
(r272717)
@@ -169,17 +169,12 @@ create_directory(const char *path)
if (component == NULL)
break;
concat(&partial, &component);
- //log_debugx("checking \"%s\" for existence", partial);
- error = access(partial, F_OK);
- if (error == 0)
- continue;
- if (errno != ENOENT)
- log_err(1, "cannot access %s", partial);
- log_debugx("directory %s does not exist, creating",
- partial);
+ //log_debugx("creating \"%s\"", partial);
error = mkdir(partial, 0755);
- if (error != 0)
- log_err(1, "cannot create %s", partial);
+ if (error != 0 && errno != EEXIST) {
+ log_warn("cannot create %s", partial);
+ return;
+ }
}
free(tofree);
___
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: r272718 - head/sys/kern
Author: jkim
Date: Tue Oct 7 20:13:47 2014
New Revision: 272718
URL: https://svnweb.freebsd.org/changeset/base/272718
Log:
Make kern.nswbuf tunable from loader.
MFC after:1 week
Modified:
head/sys/kern/vfs_bio.c
Modified: head/sys/kern/vfs_bio.c
==
--- head/sys/kern/vfs_bio.c Tue Oct 7 19:08:02 2014(r272717)
+++ head/sys/kern/vfs_bio.c Tue Oct 7 20:13:47 2014(r272718)
@@ -667,6 +667,10 @@ bd_speedup(void)
mtx_unlock(&bdlock);
}
+#ifndef NSWBUF_MIN
+#defineNSWBUF_MIN 16
+#endif
+
#ifdef __i386__
#defineTRANSIENT_DENOM 5
#else
@@ -778,11 +782,10 @@ kern_vfs_bio_buffer_alloc(caddr_t v, lon
* swbufs are used as temporary holders for I/O, such as paging I/O.
* We have no less then 16 and no more then 256.
*/
- nswbuf = max(min(nbuf/4, 256), 16);
-#ifdef NSWBUF_MIN
+ nswbuf = min(nbuf / 4, 256);
+ TUNABLE_INT_FETCH("kern.nswbuf", &nswbuf);
if (nswbuf < NSWBUF_MIN)
nswbuf = NSWBUF_MIN;
-#endif
/*
* Reserve space for the buffer cache buffers
___
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: r272649 - head/contrib/byacc
On Mon, Oct 6, 2014 at 7:49 AM, Baptiste Daroussin wrote: > > Why a direct commit instead of a proper import of 20141005 version? it > would > have imho been better, simpler and easier to maintain. > Because according to Thomas Dickey, he did not fix all the memory related problems in snapshot 20141005. Specifically there are still some issues not fixed when realloc() is called. I was going to wait for the next snapshot. However, I wanted to get that one fix into FreeBSD to fix the kyua tests, which have been failing under Jenkins. See: https://jenkins.freebsd.org/jenkins/job/FreeBSD_HEAD-tests2/ (see runs #10 and #11). -- Craig ___ 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: r272719 - in head/sys/dev/cxgbe: common tom
Author: np
Date: Tue Oct 7 21:26:22 2014
New Revision: 272719
URL: https://svnweb.freebsd.org/changeset/base/272719
Log:
cxgbe/tom: don't leak resources tied to an active open request that
cannot be sent to the chip because a prerequisite L2 resolution
failed.
Submitted by: Hariprasad at chelsio dot com (original version)
MFC after:2 weeks.
Modified:
head/sys/dev/cxgbe/common/t4_msg.h
head/sys/dev/cxgbe/tom/t4_connect.c
head/sys/dev/cxgbe/tom/t4_tom.c
head/sys/dev/cxgbe/tom/t4_tom.h
head/sys/dev/cxgbe/tom/t4_tom_l2t.c
Modified: head/sys/dev/cxgbe/common/t4_msg.h
==
--- head/sys/dev/cxgbe/common/t4_msg.h Tue Oct 7 20:13:47 2014
(r272718)
+++ head/sys/dev/cxgbe/common/t4_msg.h Tue Oct 7 21:26:22 2014
(r272719)
@@ -273,6 +273,7 @@ union opcode_tid {
/* extract the TID from a CPL command */
#define GET_TID(cmd) (G_TID(ntohl(OPCODE_TID(cmd
+#define GET_OPCODE(cmd) ((cmd)->ot.opcode)
/* partitioning of TID fields that also carry a queue id */
#define S_TID_TID0
Modified: head/sys/dev/cxgbe/tom/t4_connect.c
==
--- head/sys/dev/cxgbe/tom/t4_connect.c Tue Oct 7 20:13:47 2014
(r272718)
+++ head/sys/dev/cxgbe/tom/t4_connect.c Tue Oct 7 21:26:22 2014
(r272719)
@@ -115,8 +115,8 @@ do_act_establish(struct sge_iq *iq, cons
{
struct adapter *sc = iq->adapter;
const struct cpl_act_establish *cpl = (const void *)(rss + 1);
- unsigned int tid = GET_TID(cpl);
- unsigned int atid = G_TID_TID(ntohl(cpl->tos_atid));
+ u_int tid = GET_TID(cpl);
+ u_int atid = G_TID_TID(ntohl(cpl->tos_atid));
struct toepcb *toep = lookup_atid(sc, atid);
struct inpcb *inp = toep->inp;
@@ -178,17 +178,34 @@ act_open_rpl_status_to_errno(int status)
}
}
+void
+act_open_failure_cleanup(struct adapter *sc, u_int atid, u_int status)
+{
+ struct toepcb *toep = lookup_atid(sc, atid);
+ struct inpcb *inp = toep->inp;
+ struct toedev *tod = &toep->td->tod;
+
+ free_atid(sc, atid);
+ toep->tid = -1;
+
+ if (status != EAGAIN)
+ INP_INFO_WLOCK(&V_tcbinfo);
+ INP_WLOCK(inp);
+ toe_connect_failed(tod, inp, status);
+ final_cpl_received(toep); /* unlocks inp */
+ if (status != EAGAIN)
+ INP_INFO_WUNLOCK(&V_tcbinfo);
+}
+
static int
do_act_open_rpl(struct sge_iq *iq, const struct rss_header *rss,
struct mbuf *m)
{
struct adapter *sc = iq->adapter;
const struct cpl_act_open_rpl *cpl = (const void *)(rss + 1);
- unsigned int atid = G_TID_TID(G_AOPEN_ATID(be32toh(cpl->atid_status)));
- unsigned int status = G_AOPEN_STATUS(be32toh(cpl->atid_status));
+ u_int atid = G_TID_TID(G_AOPEN_ATID(be32toh(cpl->atid_status)));
+ u_int status = G_AOPEN_STATUS(be32toh(cpl->atid_status));
struct toepcb *toep = lookup_atid(sc, atid);
- struct inpcb *inp = toep->inp;
- struct toedev *tod = &toep->td->tod;
int rc;
KASSERT(m == NULL, ("%s: wasn't expecting payload", __func__));
@@ -200,20 +217,11 @@ do_act_open_rpl(struct sge_iq *iq, const
if (negative_advice(status))
return (0);
- free_atid(sc, atid);
- toep->tid = -1;
-
if (status && act_open_has_tid(status))
release_tid(sc, GET_TID(cpl), toep->ctrlq);
rc = act_open_rpl_status_to_errno(status);
- if (rc != EAGAIN)
- INP_INFO_WLOCK(&V_tcbinfo);
- INP_WLOCK(inp);
- toe_connect_failed(tod, inp, rc);
- final_cpl_received(toep); /* unlocks inp */
- if (rc != EAGAIN)
- INP_INFO_WUNLOCK(&V_tcbinfo);
+ act_open_failure_cleanup(sc, atid, rc);
return (0);
}
Modified: head/sys/dev/cxgbe/tom/t4_tom.c
==
--- head/sys/dev/cxgbe/tom/t4_tom.c Tue Oct 7 20:13:47 2014
(r272718)
+++ head/sys/dev/cxgbe/tom/t4_tom.c Tue Oct 7 21:26:22 2014
(r272719)
@@ -98,6 +98,7 @@ static void t4_clip_task(void *, int);
static void update_clip_table(struct adapter *, struct tom_data *);
static void destroy_clip_table(struct adapter *, struct tom_data *);
static void free_tom_data(struct adapter *, struct tom_data *);
+static void reclaim_wr_resources(void *, int);
static int in6_ifaddr_gen;
static eventhandler_tag ifaddr_evhandler;
@@ -903,6 +904,8 @@ free_tom_data(struct adapter *sc, struct
if (td->listen_mask != 0)
hashdestroy(td->listen_hash, M_CXGBE, td->listen_mask);
+ if (mtx_initialized(&td->unsent_wr_lock))
+ mtx_destroy(&td->unsent_wr_lock);
if (mtx_initialized(&td->lctx_hash_lock))
mtx_destroy(&td->lctx_hash_lock);
if (mtx_initialized(&td->toep_li
svn commit: r272720 - head/sys/netinet
Author: sbruno
Date: Tue Oct 7 21:50:28 2014
New Revision: 272720
URL: https://svnweb.freebsd.org/changeset/base/272720
Log:
Implement PLPMTUD blackhole detection (RFC 4821), inspired by code
from xnu sources. If we encounter a network where ICMP is blocked
the Needs Frag indicator may not propagate back to us. Attempt to
downshift the mss once to a preconfigured value.
Default this feature to off for now while we do not have a full PLPMTUD
implementation in our stack.
Adds the following new sysctl's for control:
net.inet.tcp.pmtud_blackhole_detection -- turns on/off this feature
net.inet.tcp.pmtud_blackhole_mss -- mss to try for ipv4
net.inet.tcp.v6pmtud_blackhole_mss -- mss to try for ipv6
Adds the following new sysctl's for monitoring:
-- Number of times the code was activated to attempt a mss downshift
net.inet.tcp.pmtud_blackhole_activated
-- Number of times the blackhole mss was used in an attempt to downshift
net.inet.tcp.pmtud_blackhole_min_activated
-- Number of times that we failed to connect after we downshifted the mss
net.inet.tcp.pmtud_blackhole_failed
Phabricator: https://reviews.freebsd.org/D506
Reviewed by: rpaulo bz
MFC after:2 weeks
Relnotes: yes
Sponsored by: Limelight Networks
Modified:
head/sys/netinet/tcp_output.c
head/sys/netinet/tcp_timer.c
head/sys/netinet/tcp_var.h
Modified: head/sys/netinet/tcp_output.c
==
--- head/sys/netinet/tcp_output.c Tue Oct 7 21:26:22 2014
(r272719)
+++ head/sys/netinet/tcp_output.c Tue Oct 7 21:50:28 2014
(r272720)
@@ -675,6 +675,12 @@ just_return:
send:
SOCKBUF_LOCK_ASSERT(&so->so_snd);
+ if (len > 0) {
+ if (len >= tp->t_maxseg)
+ tp->t_flags2 |= TF2_PLPMTU_MAXSEGSNT;
+ else
+ tp->t_flags2 &= ~TF2_PLPMTU_MAXSEGSNT;
+ }
/*
* Before ESTABLISHED, force sending of initial options
* unless TCP set not to do any options.
@@ -1303,8 +1309,12 @@ send:
*
* NB: Don't set DF on small MTU/MSS to have a safe fallback.
*/
- if (V_path_mtu_discovery && tp->t_maxopd > V_tcp_minmss)
+ if (V_path_mtu_discovery && tp->t_maxopd > V_tcp_minmss) {
ip->ip_off |= htons(IP_DF);
+ tp->t_flags2 |= TF2_PLPMTU_PMTUD;
+ } else {
+ tp->t_flags2 &= ~TF2_PLPMTU_PMTUD;
+ }
if (tp->t_state == TCPS_SYN_SENT)
TCP_PROBE5(connect__request, NULL, tp, ip, tp, th);
Modified: head/sys/netinet/tcp_timer.c
==
--- head/sys/netinet/tcp_timer.cTue Oct 7 21:26:22 2014
(r272719)
+++ head/sys/netinet/tcp_timer.cTue Oct 7 21:50:28 2014
(r272720)
@@ -66,6 +66,9 @@ __FBSDID("$FreeBSD$");
#include
#include
#include
+#ifdef INET6
+#include
+#endif
#include
#ifdef TCPDEBUG
#include
@@ -127,6 +130,54 @@ SYSCTL_INT(_net_inet_tcp, OID_AUTO, rexm
&tcp_rexmit_drop_options, 0,
"Drop TCP options from 3rd and later retransmitted SYN");
+static VNET_DEFINE(int, tcp_pmtud_blackhole_detect);
+#defineV_tcp_pmtud_blackhole_detectVNET(tcp_pmtud_blackhole_detect)
+SYSCTL_INT(_net_inet_tcp, OID_AUTO, pmtud_blackhole_detection,
+CTLTYPE_INT|CTLFLAG_RW|CTLFLAG_VNET,
+&VNET_NAME(tcp_pmtud_blackhole_detect), 0,
+"Path MTU Discovery Black Hole Detection Enabled");
+
+static VNET_DEFINE(int, tcp_pmtud_blackhole_activated);
+#defineV_tcp_pmtud_blackhole_activated \
+VNET(tcp_pmtud_blackhole_activated)
+SYSCTL_INT(_net_inet_tcp, OID_AUTO, pmtud_blackhole_activated,
+CTLTYPE_INT|CTLFLAG_RD|CTLFLAG_VNET,
+&VNET_NAME(tcp_pmtud_blackhole_activated), 0,
+"Path MTU Discovery Black Hole Detection, Activation Count");
+
+static VNET_DEFINE(int, tcp_pmtud_blackhole_activated_min_mss);
+#defineV_tcp_pmtud_blackhole_activated_min_mss \
+VNET(tcp_pmtud_blackhole_activated_min_mss)
+SYSCTL_INT(_net_inet_tcp, OID_AUTO, pmtud_blackhole_activated_min_mss,
+CTLTYPE_INT|CTLFLAG_RD|CTLFLAG_VNET,
+&VNET_NAME(tcp_pmtud_blackhole_activated_min_mss), 0,
+"Path MTU Discovery Black Hole Detection, Activation Count at min MSS");
+
+static VNET_DEFINE(int, tcp_pmtud_blackhole_failed);
+#defineV_tcp_pmtud_blackhole_failedVNET(tcp_pmtud_blackhole_failed)
+SYSCTL_INT(_net_inet_tcp, OID_AUTO, pmtud_blackhole_failed,
+CTLTYPE_INT|CTLFLAG_RD|CTLFLAG_VNET,
+&VNET_NAME(tcp_pmtud_blackhole_failed), 0,
+"Path MTU Discovery Black Hole Detection, Failure Count");
+
+#ifdef INET
+static VNET_DEFINE(int, tcp_pmtud_blackhole_mss) = 1200;
+#defineV_tcp_pmtud_blackhole_mss VNET(tcp_pmtud_blackhole_mss)
+SYSCTL_INT(_net_inet_tcp, OID_AUTO, pmtud_blackhole_mss,
+CTLTYPE_INT|CTLFLAG_RW|CTLFLAG_VNET,
+&
Re: svn commit: r272649 - head/contrib/byacc
On Tue, Oct 07, 2014 at 02:02:34PM -0700, Craig Rodrigues wrote: > On Mon, Oct 6, 2014 at 7:49 AM, Baptiste Daroussin wrote: > > > > > Why a direct commit instead of a proper import of 20141005 version? it > > would > > have imho been better, simpler and easier to maintain. > > > > Because according to Thomas Dickey, he did not fix all > the memory related problems in snapshot 20141005. Specifically there are > still > some issues not fixed when realloc() is called. > I was going to wait for the next snapshot. > However, I wanted to get that one fix into FreeBSD to fix the kyua tests, > which have been failing under Jenkins. See: > https://jenkins.freebsd.org/jenkins/job/FreeBSD_HEAD-tests2/ > (see runs #10 and #11). In that case I will import the new one which has the fix (do you confirm?) Best regards, Bapt pgpVhsW6mvrbq.pgp Description: PGP signature
Re: svn commit: r272384 - head/usr.bin/mkimg
On Fri, 3 Oct 2014, Ed Maste wrote: > On 2 October 2014 10:43, Ed Maste wrote: > > > > I've been using brooks' NO_ROOT support along with makefs / mkimg to > > build and test changes by creating an image to boot in QEMU. This > > change provides a noticeable improvement in the cycle time. > > I've had a couple of inquiries about the workflow I've been using, so > I've added a brief set of steps to my Wiki page at > https://wiki.freebsd.org/EdMaste/BuildVM . > > With -DNO_ROOT for the install targets an mtree file named METALOG > file is created at the top of DESTDIR. Files are installed owned by > the user, without special flags. Makefs reads the METALOG file and > applies the desired ownership, permissions and flags in the generated > file system. > > Then mkimg creates an image with a GPT partition table, the UFS > filesystem created by makefs, and the various boot loader bits for > legacy and UEFI boot. Wouldn't it be useful for automating backing up config directories? I'd think about copying, say, /etc and /usr/local/etc to sometemporary place, changing owner to non-privileged user, and then commit changes (removals should be treated specially, of course) to some kind of SCM? Or, does such project exist already? I failed to find it, but maybe my goole-fu is lacking necessary components ;) -- Sincerely, D.Marck [DM5020, MCK-RIPE, DM3-RIPN] [ FreeBSD committer: ma...@freebsd.org ] *** Dmitry Morozovsky --- D.Marck --- Wild Woozle --- ma...@rinet.ru *** ___ 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: r272721 - in head/sys/dev: alc ale
Author: yongari
Date: Wed Oct 8 01:03:32 2014
New Revision: 272721
URL: https://svnweb.freebsd.org/changeset/base/272721
Log:
Fix a long standing bug in MAC statistics register access. One
additional register was erroneously added in the MAC register set
such that 7 TX statistics counters were wrong.
Modified:
head/sys/dev/alc/if_alc.c
head/sys/dev/alc/if_alcreg.h
head/sys/dev/ale/if_ale.c
head/sys/dev/ale/if_alereg.h
Modified: head/sys/dev/alc/if_alc.c
==
--- head/sys/dev/alc/if_alc.c Tue Oct 7 21:50:28 2014(r272720)
+++ head/sys/dev/alc/if_alc.c Wed Oct 8 01:03:32 2014(r272721)
@@ -1287,8 +1287,6 @@ alc_sysctl_node(struct alc_softc *sc)
&stats->tx_late_colls, "Late collisions");
ALC_SYSCTL_STAT_ADD32(ctx, child, "excess_colls",
&stats->tx_excess_colls, "Excessive collisions");
- ALC_SYSCTL_STAT_ADD32(ctx, child, "abort",
- &stats->tx_abort, "Aborted frames due to Excessive collisions");
ALC_SYSCTL_STAT_ADD32(ctx, child, "underruns",
&stats->tx_underrun, "FIFO underruns");
ALC_SYSCTL_STAT_ADD32(ctx, child, "desc_underruns",
@@ -2599,7 +2597,6 @@ alc_stats_update(struct alc_softc *sc)
stat->tx_multi_colls += smb->tx_multi_colls;
stat->tx_late_colls += smb->tx_late_colls;
stat->tx_excess_colls += smb->tx_excess_colls;
- stat->tx_abort += smb->tx_abort;
stat->tx_underrun += smb->tx_underrun;
stat->tx_desc_underrun += smb->tx_desc_underrun;
stat->tx_lenerrs += smb->tx_lenerrs;
@@ -2612,17 +2609,10 @@ alc_stats_update(struct alc_softc *sc)
if_inc_counter(ifp, IFCOUNTER_COLLISIONS, smb->tx_single_colls +
smb->tx_multi_colls * 2 + smb->tx_late_colls +
- smb->tx_abort * HDPX_CFG_RETRY_DEFAULT);
+ smb->tx_excess_colls * HDPX_CFG_RETRY_DEFAULT);
- /*
-* XXX
-* tx_pkts_truncated counter looks suspicious. It constantly
-* increments with no sign of Tx errors. This may indicate
-* the counter name is not correct one so I've removed the
-* counter in output errors.
-*/
- if_inc_counter(ifp, IFCOUNTER_OERRORS,
- smb->tx_abort + smb->tx_late_colls + smb->tx_underrun);
+ if_inc_counter(ifp, IFCOUNTER_OERRORS, smb->tx_late_colls +
+ smb->tx_excess_colls + smb->tx_underrun + smb->tx_pkts_truncated);
if_inc_counter(ifp, IFCOUNTER_IPACKETS, smb->rx_frames);
Modified: head/sys/dev/alc/if_alcreg.h
==
--- head/sys/dev/alc/if_alcreg.hTue Oct 7 21:50:28 2014
(r272720)
+++ head/sys/dev/alc/if_alcreg.hWed Oct 8 01:03:32 2014
(r272721)
@@ -860,7 +860,6 @@ struct smb {
uint32_t tx_multi_colls;
uint32_t tx_late_colls;
uint32_t tx_excess_colls;
- uint32_t tx_abort;
uint32_t tx_underrun;
uint32_t tx_desc_underrun;
uint32_t tx_lenerrs;
Modified: head/sys/dev/ale/if_ale.c
==
--- head/sys/dev/ale/if_ale.c Tue Oct 7 21:50:28 2014(r272720)
+++ head/sys/dev/ale/if_ale.c Wed Oct 8 01:03:32 2014(r272721)
@@ -946,8 +946,6 @@ ale_sysctl_node(struct ale_softc *sc)
&stats->tx_late_colls, "Late collisions");
ALE_SYSCTL_STAT_ADD32(ctx, child, "excess_colls",
&stats->tx_excess_colls, "Excessive collisions");
- ALE_SYSCTL_STAT_ADD32(ctx, child, "abort",
- &stats->tx_abort, "Aborted frames due to Excessive collisions");
ALE_SYSCTL_STAT_ADD32(ctx, child, "underruns",
&stats->tx_underrun, "FIFO underruns");
ALE_SYSCTL_STAT_ADD32(ctx, child, "desc_underruns",
@@ -2197,7 +2195,6 @@ ale_stats_update(struct ale_softc *sc)
stat->tx_multi_colls += smb->tx_multi_colls;
stat->tx_late_colls += smb->tx_late_colls;
stat->tx_excess_colls += smb->tx_excess_colls;
- stat->tx_abort += smb->tx_abort;
stat->tx_underrun += smb->tx_underrun;
stat->tx_desc_underrun += smb->tx_desc_underrun;
stat->tx_lenerrs += smb->tx_lenerrs;
@@ -2210,17 +2207,10 @@ ale_stats_update(struct ale_softc *sc)
if_inc_counter(ifp, IFCOUNTER_COLLISIONS, smb->tx_single_colls +
smb->tx_multi_colls * 2 + smb->tx_late_colls +
- smb->tx_abort * HDPX_CFG_RETRY_DEFAULT);
+ smb->tx_excess_colls * HDPX_CFG_RETRY_DEFAULT);
- /*
-* XXX
-* tx_pkts_truncated counter looks suspicious. It constantly
-* increments with no sign of Tx errors. This may indicate
-* the counter name is not correct one so I've removed the
-* counter in output errors.
-*/
- if_inc_counter(ifp, IFCOUNTER_OERRORS,
- smb->tx_abort + smb->tx_late_colls + smb->tx_
svn commit: r272728 - head/usr.bin/w
Author: eadler
Date: Wed Oct 8 05:04:31 2014
New Revision: 272728
URL: https://svnweb.freebsd.org/changeset/base/272728
Log:
don't reinvent the wheel: rely on basename(3)
Reviewed by: nwhitehorn
Modified:
head/usr.bin/w/w.c
Modified: head/usr.bin/w/w.c
==
--- head/usr.bin/w/w.c Wed Oct 8 04:42:56 2014(r272727)
+++ head/usr.bin/w/w.c Wed Oct 8 05:04:31 2014(r272728)
@@ -68,6 +68,7 @@ static const char sccsid[] = "@(#)w.c 8.
#include
#include
#include
+#include
#include
#include
#include
@@ -121,7 +122,6 @@ static struct entry {
static void pr_header(time_t *, int);
static struct stat *ttystat(char *);
static void usage(int);
-static int this_is_uptime(const char *s);
char *fmt_argv(char **, char *, char *, size_t); /* ../../bin/ps/fmt.c */
@@ -144,7 +144,7 @@ main(int argc, char *argv[])
use_comma = (*nl_langinfo(RADIXCHAR) != ',');
/* Are we w(1) or uptime(1)? */
- if (this_is_uptime(argv[0]) == 0) {
+ if (strcmp(basename(argv[0]), "uptime") == 0) {
wcmd = 0;
p = "";
} else {
@@ -512,17 +512,3 @@ usage(int wcmd)
(void)fprintf(stderr, "usage: uptime\n");
exit(1);
}
-
-static int
-this_is_uptime(const char *s)
-{
- const char *u;
-
- if ((u = strrchr(s, '/')) != NULL)
- ++u;
- else
- u = s;
- if (strcmp(u, "uptime") == 0)
- return (0);
- return (-1);
-}
___
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: r272729 - head/sys/dev/pci
Author: yongari
Date: Wed Oct 8 05:34:39 2014
New Revision: 272729
URL: https://svnweb.freebsd.org/changeset/base/272729
Log:
Add new quirk PCI_QUIRK_MSI_INTX_BUG to pci(4).
QAC AR816x/E2200 controller has a silicon bug that MSI interrupt
does not assert if PCIM_CMD_INTxDIS bit of command register is set.
Reviewed by: jhb
Modified:
head/sys/dev/pci/pci.c
Modified: head/sys/dev/pci/pci.c
==
--- head/sys/dev/pci/pci.c Wed Oct 8 05:04:31 2014(r272728)
+++ head/sys/dev/pci/pci.c Wed Oct 8 05:34:39 2014(r272729)
@@ -207,6 +207,7 @@ struct pci_quirk {
#definePCI_QUIRK_ENABLE_MSI_VM 3 /* Older chipset in VM where MSI
works */
#definePCI_QUIRK_UNMAP_REG 4 /* Ignore PCI map register */
#definePCI_QUIRK_DISABLE_MSIX 5 /* MSI-X doesn't work */
+#definePCI_QUIRK_MSI_INTX_BUG 5 /* PCIM_CMD_INTxDIS disables MSI */
int arg1;
int arg2;
};
@@ -266,6 +267,15 @@ static const struct pci_quirk pci_quirks
*/
{ 0x43851002, PCI_QUIRK_UNMAP_REG, 0x14, 0 },
+ /*
+* Atheros AR8161/AR8162/E2200 ethernet controller has a bug that
+* MSI interrupt does not assert if PCIM_CMD_INTxDIS bit of the
+* command register is set.
+*/
+ { 0x10911969, PCI_QUIRK_MSI_INTX_BUG, 0, 0 },
+ { 0xE0911969, PCI_QUIRK_MSI_INTX_BUG, 0, 0 },
+ { 0x10901969, PCI_QUIRK_MSI_INTX_BUG, 0, 0 },
+
{ 0 }
};
@@ -3856,8 +3866,14 @@ pci_setup_intr(device_t dev, device_t ch
mte->mte_handlers++;
}
- /* Make sure that INTx is disabled if we are using MSI/MSIX */
- pci_set_command_bit(dev, child, PCIM_CMD_INTxDIS);
+ if (!pci_has_quirk(pci_get_devid(dev),
+ PCI_QUIRK_MSI_INTX_BUG)) {
+ /*
+* Make sure that INTx is disabled if we are
+* using MSI/MSIX
+*/
+ pci_set_command_bit(dev, child, PCIM_CMD_INTxDIS);
+ }
bad:
if (error) {
(void)bus_generic_teardown_intr(dev, child, irq,
___
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: r272730 - head/sys/dev/alc
Author: yongari
Date: Wed Oct 8 05:47:01 2014
New Revision: 272730
URL: https://svnweb.freebsd.org/changeset/base/272730
Log:
Add support for QAC AR816x/AR817x Gigabit/Fast Ethernet controllers.
These controllers seem to have the same feature of AR813x/AR815x and
improved RSS support(4 TX queues and 8 RX queues). alc(4) supports
all hardware features except RSS. I didn't implement RX checksum
offloading for AR816x/AR817x just because I couldn't get
confirmation from the Vendor whether AR816x/AR817x corrected its
predecessor's RX checksum offloading bug on fragmented packets.
This change adds supports for the following controllers.
o AR8161 PCIe Gigabit Ethernet controller
o AR8162 PCIe Fast Ethernet controller
o AR8171 PCIe Gigabit Ethernet controller
o AR8172 PCIe Fast Ethernet controller
o Killer E2200 Gigabit Ethernet controller
Tested by:Many
Relnotes: yes
MFC after:2 weeks
HW donated by:Qualcomm Atheros Communications, Inc.
Modified:
head/sys/dev/alc/if_alc.c
head/sys/dev/alc/if_alcreg.h
head/sys/dev/alc/if_alcvar.h
Modified: head/sys/dev/alc/if_alc.c
==
--- head/sys/dev/alc/if_alc.c Wed Oct 8 05:34:39 2014(r272729)
+++ head/sys/dev/alc/if_alc.c Wed Oct 8 05:47:01 2014(r272730)
@@ -111,17 +111,31 @@ static struct alc_ident alc_ident_table[
"Atheros AR8152 v1.1 PCIe Fast Ethernet" },
{ VENDORID_ATHEROS, DEVICEID_ATHEROS_AR8152_B2, 6 * 1024,
"Atheros AR8152 v2.0 PCIe Fast Ethernet" },
+ { VENDORID_ATHEROS, DEVICEID_ATHEROS_AR8161, 9 * 1024,
+ "Atheros AR8161 PCIe Gigabit Ethernet" },
+ { VENDORID_ATHEROS, DEVICEID_ATHEROS_AR8162, 9 * 1024,
+ "Atheros AR8161 PCIe Fast Ethernet" },
+ { VENDORID_ATHEROS, DEVICEID_ATHEROS_AR8171, 9 * 1024,
+ "Atheros AR8161 PCIe Gigabit Ethernet" },
+ { VENDORID_ATHEROS, DEVICEID_ATHEROS_AR8172, 9 * 1024,
+ "Atheros AR8161 PCIe Fast Ethernet" },
+ { VENDORID_ATHEROS, DEVICEID_ATHEROS_E2200, 9 * 1024,
+ "Killer E2200 Gigabit Ethernet" },
{ 0, 0, 0, NULL}
};
-static voidalc_aspm(struct alc_softc *, int);
+static voidalc_aspm(struct alc_softc *, int, int);
+static voidalc_aspm_813x(struct alc_softc *, int);
+static voidalc_aspm_816x(struct alc_softc *, int);
static int alc_attach(device_t);
static int alc_check_boundary(struct alc_softc *);
+static voidalc_config_msi(struct alc_softc *);
static int alc_detach(device_t);
static voidalc_disable_l0s_l1(struct alc_softc *);
static int alc_dma_alloc(struct alc_softc *);
static voidalc_dma_free(struct alc_softc *);
static voidalc_dmamap_cb(void *, bus_dma_segment_t *, int, int);
+static voidalc_dsp_fixup(struct alc_softc *, int);
static int alc_encap(struct alc_softc *, struct mbuf **);
static struct alc_ident *
alc_find_ident(device_t);
@@ -130,6 +144,9 @@ static struct mbuf *
alc_fixup_rx(struct ifnet *, struct mbuf *);
#endif
static voidalc_get_macaddr(struct alc_softc *);
+static voidalc_get_macaddr_813x(struct alc_softc *);
+static voidalc_get_macaddr_816x(struct alc_softc *);
+static voidalc_get_macaddr_par(struct alc_softc *);
static voidalc_init(void *);
static voidalc_init_cmb(struct alc_softc *);
static voidalc_init_locked(struct alc_softc *);
@@ -141,14 +158,26 @@ static void alc_int_task(void *, int);
static int alc_intr(void *);
static int alc_ioctl(struct ifnet *, u_long, caddr_t);
static voidalc_mac_config(struct alc_softc *);
+static uint32_talc_mii_readreg_813x(struct alc_softc *, int, int);
+static uint32_talc_mii_readreg_816x(struct alc_softc *, int, int);
+static uint32_talc_mii_writereg_813x(struct alc_softc *, int, int,
int);
+static uint32_talc_mii_writereg_816x(struct alc_softc *, int, int,
int);
static int alc_miibus_readreg(device_t, int, int);
static voidalc_miibus_statchg(device_t);
static int alc_miibus_writereg(device_t, int, int, int);
+static uint32_talc_miidbg_readreg(struct alc_softc *, int);
+static uint32_talc_miidbg_writereg(struct alc_softc *, int, int);
+static uint32_talc_miiext_readreg(struct alc_softc *, int, int);
+static uint32_talc_miiext_writereg(struct alc_softc *, int, int, int);
static int alc_mediachange(struct ifnet *);
+static int alc_mediachange_locked(struct alc_softc *);
static voidalc_mediastatus(struct ifnet *, struct ifmediareq *);
static int alc_newbuf(struct alc_softc *, struct alc_rxdesc *);
+static voidalc_osc_reset(struct alc_softc *);
static voidalc_phy_down(struct alc_softc *);
static voidalc_phy_reset(struct alc_softc *);
+static voidalc_phy_reset_813x(struct alc_softc *);
+static void
svn commit: r272731 - head/share/man/man4
Author: yongari Date: Wed Oct 8 05:49:10 2014 New Revision: 272731 URL: https://svnweb.freebsd.org/changeset/base/272731 Log: Document newly added controller AR816x/AR817x. Modified: head/share/man/man4/alc.4 Modified: head/share/man/man4/alc.4 == --- head/share/man/man4/alc.4 Wed Oct 8 05:47:01 2014(r272730) +++ head/share/man/man4/alc.4 Wed Oct 8 05:49:10 2014(r272731) @@ -24,12 +24,12 @@ .\" .\" $FreeBSD$ .\" -.Dd August 9, 2010 +.Dd October 8, 2014 .Dt ALC 4 .Os .Sh NAME .Nm alc -.Nd Atheros AR813x/AR815x Gigabit/Fast Ethernet driver +.Nd Atheros AR813x/AR815x/AR816x/AR817x Gigabit/Fast Ethernet driver .Sh SYNOPSIS To compile this driver into the kernel, place the following lines in your @@ -48,8 +48,8 @@ if_alc_load="YES" .Sh DESCRIPTION The .Nm -device driver provides support for Atheros AR813x and AR815x PCI -Express Gigabit/Fast Ethernet controllers. +device driver provides support for Atheros AR813x, AR815x, AR816x +and AR817x PCI Express Gigabit/Fast Ethernet controllers. .Pp All LOMs supported by the .Nm @@ -58,9 +58,9 @@ segmentation offload (TSO), hardware VLA features, Wake On Lan (WOL) and an interrupt moderation mechanism as well as a 64-bit multicast hash filter. .Pp -The AR813x and AR815x supports Jumbo Frames (up to 9216 and 6144 -bytes, respectively), which can be configured via the interface -MTU setting. +The AR813x, AR815x, AR816x and AR817x supports Jumbo Frames (up to +9216, 6144, 9216 and 9216 bytes, respectively), which can be +configured via the interface MTU setting. Selecting an MTU larger than 1500 bytes with the .Xr ifconfig 8 utility configures the adapter to receive and transmit Jumbo Frames. @@ -112,6 +112,16 @@ Atheros AR8151 v2.0 PCI Express Gigabit Atheros AR8152 v1.1 PCI Express Fast Ethernet controller .It Atheros AR8152 v2.0 PCI Express Fast Ethernet controller +.It +Atheros AR8161 PCI Express Gigabit Ethernet controller +.It +Atheros AR8162 PCI Express Fast Ethernet controller +.It +Atheros AR8171 PCI Express Gigabit Ethernet controller +.It +Atheros AR8172 PCI Express Fast Ethernet controller +.It +Killer E2200 Gigabit Ethernet controller .El .Sh LOADER TUNABLES Tunables can be set at the ___ 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: r272732 - head/sys/dev/pci
Author: yongari
Date: Wed Oct 8 05:53:04 2014
New Revision: 272732
URL: https://svnweb.freebsd.org/changeset/base/272732
Log:
Oops, fix typo made in r272729.
Modified:
head/sys/dev/pci/pci.c
Modified: head/sys/dev/pci/pci.c
==
--- head/sys/dev/pci/pci.c Wed Oct 8 05:49:10 2014(r272731)
+++ head/sys/dev/pci/pci.c Wed Oct 8 05:53:04 2014(r272732)
@@ -207,7 +207,7 @@ struct pci_quirk {
#definePCI_QUIRK_ENABLE_MSI_VM 3 /* Older chipset in VM where MSI
works */
#definePCI_QUIRK_UNMAP_REG 4 /* Ignore PCI map register */
#definePCI_QUIRK_DISABLE_MSIX 5 /* MSI-X doesn't work */
-#definePCI_QUIRK_MSI_INTX_BUG 5 /* PCIM_CMD_INTxDIS disables MSI */
+#definePCI_QUIRK_MSI_INTX_BUG 6 /* PCIM_CMD_INTxDIS disables MSI */
int arg1;
int arg2;
};
___
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"
