Re: svn commit: r356437 - head/lib/libprocstat
In message <20200107064255.18dad9d2@freyja>, "O. Hartmann" writes:
> On Tue, 7 Jan 2020 05:33:33 + (UTC)
> Mateusz Guzik wrote:
>
> > Author: mjg
> > Date: Tue Jan 7 05:33:33 2020
> > New Revision: 356437
> > URL: https://svnweb.freebsd.org/changeset/base/356437
> >
> > Log:
> > procstat: read lo_name instead of now removed v_tag
> >
> > Modified:
> > head/lib/libprocstat/libprocstat.c
> >
> > Modified: head/lib/libprocstat/libprocstat.c
> > ===
> ===
> > --- head/lib/libprocstat/libprocstat.c Tue Jan 7 04:34:29 2020
> > (r356436) +++ head/lib/libprocstat/libprocstat.cTue Jan 7 05:33:33
> > 2020(r356437) @@ -1278,10 +1278,10 @@
> > procstat_get_vnode_info_kvm(kvm_t *kd, struct filestat vn->vn_type =
> > vntype2psfsttype(vnode.v_type); if (vnode.v_type == VNON || vnode.v_type ==
> > VBAD) return (0);
> > - error = kvm_read_all(kd, (unsigned long)vnode.v_tag, tagstr,
> > - sizeof(tagstr));
> > + error = kvm_read_all(kd, (unsigned
> > long)vnode.v_lock.lock_object.lo_name,
> > + tagstr, sizeof(tagstr));
> > if (error == 0) {
> > - warnx("can't read v_tag at %p", (void *)vp);
> > + warnx("can't read lo_name at %p", (void *)vp);
> > goto fail;
> > }
> > tagstr[sizeof(tagstr) - 1] = '\0';
> > ___
> > svn-src-head@freebsd.org mailing list
> > https://lists.freebsd.org/mailman/listinfo/svn-src-head
> > To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
>
> This commit makes buildworld to fail:
>
> [...]
> --- lib/libprocstat__L ---
> /usr/src/lib/libprocstat/libprocstat.c:848:2: warning: variable 'files' is us
> ed
> uninitialized whenever switch default is taken [-Wsometimes-uninitialized]
> default: ^~~
> /usr/src/lib/libprocstat/libprocstat.c:851:6: note: uninitialized use occurs
> here if (files == NULL && errno != EPERM) {
> ^
> /usr/src/lib/libprocstat/libprocstat.c:827:32: note: initialize the variable
> 'files' to silence this warning struct kinfo_file *kif, *files;
> ^
>= NULL
> /usr/src/lib/libprocstat/libprocstat.c:1281:48: error: no member named 'v_tag
> '
> in 'struct vnode'
>
> [...]
>
> Kind regards,
>
> oh
>
>
It would appear you don't have this commit in your tree.
--
Cheers,
Cy Schubert
FreeBSD UNIX: Web: http://www.FreeBSD.org
The need of the many outweighs the greed of the few.
___
svn-src-head@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
Re: svn commit: r356437 - head/lib/libprocstat
On Tue, 7 Jan 2020 05:33:33 + (UTC)
Mateusz Guzik wrote:
> Author: mjg
> Date: Tue Jan 7 05:33:33 2020
> New Revision: 356437
> URL: https://svnweb.freebsd.org/changeset/base/356437
>
> Log:
> procstat: read lo_name instead of now removed v_tag
>
> Modified:
> head/lib/libprocstat/libprocstat.c
>
> Modified: head/lib/libprocstat/libprocstat.c
> ==
> --- head/lib/libprocstat/libprocstat.cTue Jan 7 04:34:29 2020
> (r356436) +++ head/lib/libprocstat/libprocstat.c Tue Jan 7 05:33:33
> 2020 (r356437) @@ -1278,10 +1278,10 @@
> procstat_get_vnode_info_kvm(kvm_t *kd, struct filestat vn->vn_type =
> vntype2psfsttype(vnode.v_type); if (vnode.v_type == VNON || vnode.v_type ==
> VBAD) return (0);
> - error = kvm_read_all(kd, (unsigned long)vnode.v_tag, tagstr,
> - sizeof(tagstr));
> + error = kvm_read_all(kd, (unsigned
> long)vnode.v_lock.lock_object.lo_name,
> + tagstr, sizeof(tagstr));
> if (error == 0) {
> - warnx("can't read v_tag at %p", (void *)vp);
> + warnx("can't read lo_name at %p", (void *)vp);
> goto fail;
> }
> tagstr[sizeof(tagstr) - 1] = '\0';
> ___
> svn-src-head@freebsd.org mailing list
> https://lists.freebsd.org/mailman/listinfo/svn-src-head
> To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
This commit makes buildworld to fail:
[...]
--- lib/libprocstat__L ---
/usr/src/lib/libprocstat/libprocstat.c:848:2: warning: variable 'files' is used
uninitialized whenever switch default is taken [-Wsometimes-uninitialized]
default: ^~~
/usr/src/lib/libprocstat/libprocstat.c:851:6: note: uninitialized use occurs
here if (files == NULL && errno != EPERM) {
^
/usr/src/lib/libprocstat/libprocstat.c:827:32: note: initialize the variable
'files' to silence this warning struct kinfo_file *kif, *files;
^
= NULL
/usr/src/lib/libprocstat/libprocstat.c:1281:48: error: no member named 'v_tag'
in 'struct vnode'
[...]
Kind regards,
oh
___
svn-src-head@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r356437 - head/lib/libprocstat
Author: mjg
Date: Tue Jan 7 05:33:33 2020
New Revision: 356437
URL: https://svnweb.freebsd.org/changeset/base/356437
Log:
procstat: read lo_name instead of now removed v_tag
Modified:
head/lib/libprocstat/libprocstat.c
Modified: head/lib/libprocstat/libprocstat.c
==
--- head/lib/libprocstat/libprocstat.c Tue Jan 7 04:34:29 2020
(r356436)
+++ head/lib/libprocstat/libprocstat.c Tue Jan 7 05:33:33 2020
(r356437)
@@ -1278,10 +1278,10 @@ procstat_get_vnode_info_kvm(kvm_t *kd, struct filestat
vn->vn_type = vntype2psfsttype(vnode.v_type);
if (vnode.v_type == VNON || vnode.v_type == VBAD)
return (0);
- error = kvm_read_all(kd, (unsigned long)vnode.v_tag, tagstr,
- sizeof(tagstr));
+ error = kvm_read_all(kd, (unsigned
long)vnode.v_lock.lock_object.lo_name,
+ tagstr, sizeof(tagstr));
if (error == 0) {
- warnx("can't read v_tag at %p", (void *)vp);
+ warnx("can't read lo_name at %p", (void *)vp);
goto fail;
}
tagstr[sizeof(tagstr) - 1] = '\0';
___
svn-src-head@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r356436 - head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs
Author: mjg
Date: Tue Jan 7 04:34:29 2020
New Revision: 356436
URL: https://svnweb.freebsd.org/changeset/base/356436
Log:
zfs: plug a vnode reserve leak in zfs_make_xattrdir
Modified:
head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/zfs_dir.c
Modified: head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/zfs_dir.c
==
--- head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/zfs_dir.c Tue Jan
7 04:34:03 2020(r356435)
+++ head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/zfs_dir.c Tue Jan
7 04:34:29 2020(r356436)
@@ -829,6 +829,7 @@ zfs_make_xattrdir(znode_t *zp, vattr_t *vap, vnode_t *
if (error) {
zfs_acl_ids_free(_ids);
dmu_tx_abort(tx);
+ getnewvnode_drop_reserve();
return (error);
}
zfs_mknode(zp, vap, tx, cr, IS_XATTR, , _ids);
___
svn-src-head@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r356435 - head/sys/kern
Author: mjg
Date: Tue Jan 7 04:34:03 2020
New Revision: 356435
URL: https://svnweb.freebsd.org/changeset/base/356435
Log:
vfs: prevent numvnodes and freevnodes re-reads when appropriate
Otherwise in code like this:
if (numvnodes > desiredvnodes)
vnlru_free_locked(numvnodes - desiredvnodes, NULL);
numvnodes can drop below desiredvnodes prior to the call and if the
compiler generated another read the subtraction would get a negative
value.
Modified:
head/sys/kern/vfs_subr.c
Modified: head/sys/kern/vfs_subr.c
==
--- head/sys/kern/vfs_subr.cTue Jan 7 04:33:14 2020(r356434)
+++ head/sys/kern/vfs_subr.cTue Jan 7 04:34:03 2020(r356435)
@@ -1211,16 +1211,19 @@ vnlru_free(int count, struct vfsops *mnt_op)
static int
vspace(void)
{
+ u_long rnumvnodes, rfreevnodes;
int space;
gapvnodes = imax(desiredvnodes - wantfreevnodes, 100);
vhiwat = gapvnodes / 11; /* 9% -- just under the 10% in vlrureclaim() */
vlowat = vhiwat / 2;
- if (numvnodes > desiredvnodes)
+ rnumvnodes = atomic_load_long();
+ rfreevnodes = atomic_load_long();
+ if (rnumvnodes > desiredvnodes)
return (0);
- space = desiredvnodes - numvnodes;
+ space = desiredvnodes - rnumvnodes;
if (freevnodes > wantfreevnodes)
- space += freevnodes - wantfreevnodes;
+ space += rfreevnodes - wantfreevnodes;
return (space);
}
@@ -1292,6 +1295,7 @@ static int vnlruproc_sig;
static void
vnlru_proc(void)
{
+ u_long rnumvnodes, rfreevnodes;
struct mount *mp, *nmp;
unsigned long onumvnodes;
int done, force, trigger, usevnodes, vsp;
@@ -1304,13 +1308,14 @@ vnlru_proc(void)
for (;;) {
kproc_suspend_check(vnlruproc);
mtx_lock(_free_list_mtx);
+ rnumvnodes = atomic_load_long();
/*
* If numvnodes is too large (due to desiredvnodes being
* adjusted using its sysctl, or emergency growth), first
* try to reduce it by discarding from the free list.
*/
- if (numvnodes > desiredvnodes)
- vnlru_free_locked(numvnodes - desiredvnodes, NULL);
+ if (rnumvnodes > desiredvnodes)
+ vnlru_free_locked(rnumvnodes - desiredvnodes, NULL);
/*
* Sleep if the vnode cache is in a good state. This is
* when it is not over-full and has space for about a 4%
@@ -1332,7 +1337,10 @@ vnlru_proc(void)
}
mtx_unlock(_free_list_mtx);
done = 0;
- onumvnodes = numvnodes;
+ rnumvnodes = atomic_load_long();
+ rfreevnodes = atomic_load_long();
+
+ onumvnodes = rnumvnodes;
/*
* Calculate parameters for recycling. These are the same
* throughout the loop to give some semblance of fairness.
@@ -1340,10 +1348,10 @@ vnlru_proc(void)
* of resident pages. We aren't trying to free memory; we
* are trying to recycle or at least free vnodes.
*/
- if (numvnodes <= desiredvnodes)
- usevnodes = numvnodes - freevnodes;
+ if (rnumvnodes <= desiredvnodes)
+ usevnodes = rnumvnodes - rfreevnodes;
else
- usevnodes = numvnodes;
+ usevnodes = rnumvnodes;
if (usevnodes <= 0)
usevnodes = 1;
/*
@@ -1516,14 +1524,17 @@ getnewvnode_wait(int suspended)
void
getnewvnode_reserve(u_int count)
{
+ u_long rnumvnodes, rfreevnodes;
struct thread *td;
/* Pre-adjust like the pre-adjust in getnewvnode(), with any count. */
/* XXX no longer so quick, but this part is not racy. */
mtx_lock(_free_list_mtx);
- if (numvnodes + count > desiredvnodes && freevnodes > wantfreevnodes)
- vnlru_free_locked(ulmin(numvnodes + count - desiredvnodes,
- freevnodes - wantfreevnodes), NULL);
+ rnumvnodes = atomic_load_long();
+ rfreevnodes = atomic_load_long();
+ if (rnumvnodes + count > desiredvnodes && rfreevnodes > wantfreevnodes)
+ vnlru_free_locked(ulmin(rnumvnodes + count - desiredvnodes,
+ rfreevnodes - wantfreevnodes), NULL);
mtx_unlock(_free_list_mtx);
td = curthread;
___
svn-src-head@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r356434 - head/sbin/fsck_msdosfs
Author: delphij Date: Tue Jan 7 04:33:14 2020 New Revision: 356434 URL: https://svnweb.freebsd.org/changeset/base/356434 Log: fsck_msdosfs.8: document -M. Reported by: mckusick Reviewed by: mckusick, emaste, bcr MFC after:28 days X-MFC-with: r356313 Differential Revision:https://reviews.freebsd.org/D23049 Modified: head/sbin/fsck_msdosfs/fsck_msdosfs.8 Modified: head/sbin/fsck_msdosfs/fsck_msdosfs.8 == --- head/sbin/fsck_msdosfs/fsck_msdosfs.8 Tue Jan 7 04:30:49 2020 (r356433) +++ head/sbin/fsck_msdosfs/fsck_msdosfs.8 Tue Jan 7 04:33:14 2020 (r356434) @@ -25,7 +25,7 @@ .\" .\" $FreeBSD$ .\" -.Dd October 3, 2016 +.Dd January 6, 2020 .Dt FSCK_MSDOSFS 8 .Os .Sh NAME @@ -37,7 +37,7 @@ .Op Fl Cf .Ar filesystem ... .Nm -.Op Fl Cny +.Op Fl CMny .Ar filesystem ... .Sh DESCRIPTION The @@ -84,6 +84,21 @@ which seeks to determine whether the file system needs immediately in foreground, or if its cleaning can be deferred to background. FAT (MS-DOS) file systems must always be cleaned in the foreground. A non-zero exit code is always returned for this option. +.It Fl M +Causes +.Nm +to not use +.Xr mmap 2 +when checking a FAT32 file system. +This option is mainly for debugging purposes and is not normally necessary. +The +.Nm +utility will automatically fall back to use a simple LRU cache of 4 MiB +when it failed to perform +.Xr mmap 2 , +or when +.Fl M +is specified. .It Fl f Force .Nm ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r356433 - head/sys/kern
Author: mjg Date: Tue Jan 7 04:30:49 2020 New Revision: 356433 URL: https://svnweb.freebsd.org/changeset/base/356433 Log: vfs: annotate numvnodes and vnode_free_list_mtx with __exclusive_cache_line Modified: head/sys/kern/vfs_subr.c Modified: head/sys/kern/vfs_subr.c == --- head/sys/kern/vfs_subr.cTue Jan 7 04:29:34 2020(r356432) +++ head/sys/kern/vfs_subr.cTue Jan 7 04:30:49 2020(r356433) @@ -139,7 +139,7 @@ static int v_inval_buf_range_locked(struct vnode *vp, * Number of vnodes in existence. Increased whenever getnewvnode() * allocates a new vnode, decreased in vdropl() for VIRF_DOOMED vnode. */ -static unsigned long numvnodes; +static u_long __exclusive_cache_line numvnodes; SYSCTL_ULONG(_vfs, OID_AUTO, numvnodes, CTLFLAG_RD, , 0, "Number of vnodes in existence"); @@ -227,7 +227,7 @@ static struct mtx mntid_mtx; * numvnodes * freevnodes */ -static struct mtx vnode_free_list_mtx; +static struct mtx __exclusive_cache_line vnode_free_list_mtx; /* Publicly exported FS */ struct nfs_public nfs_pub; ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r356432 - in head/sys: kern security/mac_veriexec sys
Author: mjg
Date: Tue Jan 7 04:29:34 2020
New Revision: 356432
URL: https://svnweb.freebsd.org/changeset/base/356432
Log:
vfs: eliminate v_tag from struct vnode
There was only one consumer and it was using it incorrectly.
It is given an equivalent hack.
Reviewed by: jeff
Differential Revision:https://reviews.freebsd.org/D23037
Modified:
head/sys/kern/vfs_subr.c
head/sys/security/mac_veriexec/mac_veriexec.c
head/sys/sys/vnode.h
Modified: head/sys/kern/vfs_subr.c
==
--- head/sys/kern/vfs_subr.cTue Jan 7 04:27:40 2020(r356431)
+++ head/sys/kern/vfs_subr.cTue Jan 7 04:29:34 2020(r356432)
@@ -1657,7 +1657,6 @@ alloc:
KASSERT(vp->v_lockf == NULL, ("stale v_lockf %p", vp));
KASSERT(vp->v_pollinfo == NULL, ("stale v_pollinfo %p", vp));
vp->v_type = VNON;
- vp->v_tag = tag;
vp->v_op = vops;
v_init_counters(vp);
vp->v_bufobj.bo_ops = _ops_bio;
@@ -3679,7 +3678,7 @@ vgonel(struct vnode *vp)
if (mp != NULL)
vn_finished_secondary_write(mp);
VNASSERT(vp->v_object == NULL, vp,
- ("vop_reclaim left v_object vp=%p, tag=%s", vp, vp->v_tag));
+ ("vop_reclaim left v_object vp=%p", vp));
/*
* Clear the advisory locks and wake up waiting threads.
*/
@@ -3697,7 +3696,6 @@ vgonel(struct vnode *vp)
VI_LOCK(vp);
vp->v_vnlock = >v_lock;
vp->v_op = _vnodeops;
- vp->v_tag = "none";
vp->v_type = VBAD;
}
@@ -3733,7 +3731,7 @@ vn_printf(struct vnode *vp, const char *fmt, ...)
vprintf(fmt, ap);
va_end(ap);
printf("%p: ", (void *)vp);
- printf("tag %s, type %s\n", vp->v_tag, typename[vp->v_type]);
+ printf("type %s\n", typename[vp->v_type]);
printf("usecount %d, writecount %d, refcount %d",
vp->v_usecount, vp->v_writecount, vp->v_holdcnt);
switch (vp->v_type) {
Modified: head/sys/security/mac_veriexec/mac_veriexec.c
==
--- head/sys/security/mac_veriexec/mac_veriexec.c Tue Jan 7 04:27:40
2020(r356431)
+++ head/sys/security/mac_veriexec/mac_veriexec.c Tue Jan 7 04:29:34
2020(r356432)
@@ -737,6 +737,22 @@ MAC_POLICY_SET(_veriexec_ops, mac_veriexec, MAC_VE
MPC_LOADTIME_FLAG_NOTLATE, _veriexec_slot);
MODULE_VERSION(mac_veriexec, 1);
+static struct vnode *
+mac_veriexec_bottom_vnode(struct vnode *vp)
+{
+ struct vnode *ldvp = NULL;
+
+ /*
+* XXX This code is bogus. nullfs is not the only stacking
+* filesystem. Less bogus code would add a VOP to reach bottom
+* vnode and would not make assumptions how to get there.
+*/
+ if (vp->v_mount != NULL &&
+ strcmp(vp->v_mount->mnt_vfc->vfc_name, "nullfs") == 0)
+ ldvp = NULLVPTOLOWERVP(vp);
+ return (ldvp);
+}
+
/**
* @brief Get the fingerprint status set on a vnode.
*
@@ -748,6 +764,7 @@ fingerprint_status_t
mac_veriexec_get_fingerprint_status(struct vnode *vp)
{
fingerprint_status_t fps;
+ struct vnode *ldvp;
fps = SLOT(vp->v_label);
switch (fps) {
@@ -757,12 +774,9 @@ mac_veriexec_get_fingerprint_status(struct vnode *vp)
break;
default:
/* we may need to recurse */
- if (strcmp(vp->v_tag, "null") == 0) {
- struct vnode *ldvp;
-
- ldvp = NULLVPTOLOWERVP(vp);
+ ldvp = mac_veriexec_bottom_vnode(vp);
+ if (ldvp != NULL)
return mac_veriexec_get_fingerprint_status(ldvp);
- }
break;
}
return fps;
@@ -808,12 +822,11 @@ void
mac_veriexec_set_fingerprint_status(struct vnode *vp,
fingerprint_status_t fp_status)
{
+ struct vnode *ldvp;
/* recurse until we find the real storage */
- if (strcmp(vp->v_tag, "null") == 0) {
- struct vnode *ldvp;
-
- ldvp = NULLVPTOLOWERVP(vp);
+ ldvp = mac_veriexec_bottom_vnode(vp);
+ if (ldvp != NULL) {
mac_veriexec_set_fingerprint_status(ldvp, fp_status);
return;
}
Modified: head/sys/sys/vnode.h
==
--- head/sys/sys/vnode.hTue Jan 7 04:27:40 2020(r356431)
+++ head/sys/sys/vnode.hTue Jan 7 04:29:34 2020(r356432)
@@ -174,7 +174,6 @@ struct vnode {
int v_writecount; /* I ref count of writers or
(negative) text users */
u_int v_hash;
- const char *v_tag; /* u type of underlying data */
};
#endif /* defined(_KERNEL) || defined(_KVM_VNODE) */
svn commit: r356431 - head/sys/kern
Author: mjg
Date: Tue Jan 7 04:27:40 2020
New Revision: 356431
URL: https://svnweb.freebsd.org/changeset/base/356431
Log:
vfs: add a helper for allocating marker vnodes
Modified:
head/sys/kern/vfs_subr.c
Modified: head/sys/kern/vfs_subr.c
==
--- head/sys/kern/vfs_subr.cTue Jan 7 00:54:45 2020(r356430)
+++ head/sys/kern/vfs_subr.cTue Jan 7 04:27:40 2020(r356431)
@@ -459,6 +459,28 @@ PCTRIE_DEFINE(BUF, buf, b_lblkno, buf_trie_alloc, buf_
#defineMAXVNODES_MAX (512 * 1024 * 1024 / 64)/* 8M */
#endif
+static MALLOC_DEFINE(M_VNODE_MARKER, "vnodemarker", "vnode marker");
+
+static struct vnode *
+vn_alloc_marker(struct mount *mp)
+{
+ struct vnode *vp;
+
+ vp = malloc(sizeof(struct vnode), M_VNODE_MARKER, M_WAITOK | M_ZERO);
+ vp->v_type = VMARKER;
+ vp->v_mount = mp;
+
+ return (vp);
+}
+
+static void
+vn_free_marker(struct vnode *vp)
+{
+
+ MPASS(vp->v_type == VMARKER);
+ free(vp, M_VNODE_MARKER);
+}
+
/*
* Initialize a vnode as it first enters the zone.
*/
@@ -5785,7 +5807,6 @@ vfs_cache_root_set(struct mount *mp, struct vnode *vp)
* This interface replaces MNT_VNODE_FOREACH.
*/
-MALLOC_DEFINE(M_VNODE_MARKER, "vnodemarker", "vnode marker");
struct vnode *
__mnt_vnode_next_all(struct vnode **mvp, struct mount *mp)
@@ -5825,11 +5846,9 @@ __mnt_vnode_first_all(struct vnode **mvp, struct mount
{
struct vnode *vp;
- *mvp = malloc(sizeof(struct vnode), M_VNODE_MARKER, M_WAITOK | M_ZERO);
+ *mvp = vn_alloc_marker(mp);
MNT_ILOCK(mp);
MNT_REF(mp);
- (*mvp)->v_mount = mp;
- (*mvp)->v_type = VMARKER;
TAILQ_FOREACH(vp, >mnt_nvnodelist, v_nmntvnodes) {
/* Allow a racy peek at VIRF_DOOMED to save a lock acquisition.
*/
@@ -5845,7 +5864,7 @@ __mnt_vnode_first_all(struct vnode **mvp, struct mount
if (vp == NULL) {
MNT_REL(mp);
MNT_IUNLOCK(mp);
- free(*mvp, M_VNODE_MARKER);
+ vn_free_marker(*mvp);
*mvp = NULL;
return (NULL);
}
@@ -5869,7 +5888,7 @@ __mnt_vnode_markerfree_all(struct vnode **mvp, struct
TAILQ_REMOVE(>mnt_nvnodelist, *mvp, v_nmntvnodes);
MNT_REL(mp);
MNT_IUNLOCK(mp);
- free(*mvp, M_VNODE_MARKER);
+ vn_free_marker(*mvp);
*mvp = NULL;
}
@@ -5886,7 +5905,7 @@ mnt_vnode_markerfree_active(struct vnode **mvp, struct
MNT_ILOCK(mp);
MNT_REL(mp);
MNT_IUNLOCK(mp);
- free(*mvp, M_VNODE_MARKER);
+ vn_free_marker(*mvp);
*mvp = NULL;
}
@@ -6029,12 +6048,10 @@ __mnt_vnode_first_active(struct vnode **mvp, struct mo
{
struct vnode *vp;
- *mvp = malloc(sizeof(struct vnode), M_VNODE_MARKER, M_WAITOK | M_ZERO);
+ *mvp = vn_alloc_marker(mp);
MNT_ILOCK(mp);
MNT_REF(mp);
MNT_IUNLOCK(mp);
- (*mvp)->v_type = VMARKER;
- (*mvp)->v_mount = mp;
mtx_lock(>mnt_listmtx);
vp = TAILQ_FIRST(>mnt_activevnodelist);
___
svn-src-head@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
Re: svn commit: r356393 - head/sys/vm
Jeff, On Mon, Jan 06, 2020 at 02:51:19AM +, Jeff Roberson wrote: J> Modified: head/sys/vm/vm_page.c J> == J> --- head/sys/vm/vm_page.cMon Jan 6 01:51:23 2020(r356392) J> +++ head/sys/vm/vm_page.cMon Jan 6 02:51:19 2020(r356393) J> @@ -613,10 +613,17 @@ vm_page_startup(vm_offset_t vaddr) J> slab_ipers(sizeof(struct vm_map), UMA_ALIGN_PTR)); J> J> /* J> - * Before going fully functional kmem_init() does allocation J> - * from "KMAP ENTRY" and vmem_create() does allocation from "vmem". J> + * Before we are fully boot strapped we need to account for the J> + * following allocations: J> + * J> + * "KMAP ENTRY" from kmem_init() J> + * "vmem btag" from vmem_startup() J> + * "vmem" from vmem_create() J> + * "KMAP" from vm_map_startup() J> + * J> + * Each needs at least one page per-domain. J> */ J> -boot_pages += 2; J> +boot_pages += 4 * vm_ndomains; J> #endif J> /* J> * CTFLAG_RDTUN doesn't work during the early boot process, so we must I think this chunk brings many extraneous pages per domain. Allocations for "vmem btag" from vmem_startup() are covered by vmem_startup_count(), and allocations for "KMAP" from vm_map_startup() are also counted just a few lines above: /* vmem_startup() calls uma_prealloc(). */ boot_pages += vmem_startup_count(); /* vm_map_startup() calls uma_prealloc(). */ boot_pages += howmany(MAX_KMAP, slab_ipers(sizeof(struct vm_map), UMA_ALIGN_PTR)); The uma_prealloc() allocates as much memory as asked, it may not populate all domains. So, on NUMA it will allocate same amount of pages. Not sure if this is a right decision, but at lease the counting is correct. So, change from 2 pages to 4 isn't needed. The vm_ndomains multiplicator also seems incorrect, since at early stage we run with buckets disabled, and this single allocation will end with calling zone_alloc_item() calling zone_import() with count of 1, which will short-circuit into keg_fetch_slab() before entering domain aware code. Then keg_fetch_slab() will short circuit into startup_alloc(). Do machines with NUMA and lack of UMA_MD_SMALL_ALLOC exist? Interesting to see dmesg with DIAGNOSTIC and this change. -- Gleb Smirnoff ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
Re: svn commit: r356389 - head/sys/vm
On Mon, 6 Jan 2020, Gleb Smirnoff wrote: Jeff, On Sun, Jan 05, 2020 at 10:54:26PM +, Jeff Roberson wrote: J> Author: jeff J> Date: Sun Jan 5 22:54:25 2020 J> New Revision: 356389 J> URL: https://svnweb.freebsd.org/changeset/base/356389 J> J> Log: J> The fix in r356353 was insufficient. Not every architecture returns 0 for J> EARLY_COUNTER. Only amd64 seems to. J> J> Suggested by: markj J> Reported by:lwhsu J> Reviewed by:markj J> PR: 243117 J> J> Modified: J> head/sys/vm/uma_core.c J> J> Modified: head/sys/vm/uma_core.c J> == J> --- head/sys/vm/uma_core.cSun Jan 5 21:35:02 2020(r356388) J> +++ head/sys/vm/uma_core.cSun Jan 5 22:54:25 2020(r356389) J> @@ -4153,8 +4153,10 @@ uma_zone_get_cur(uma_zone_t zone) J> int64_t nitems; J> u_int i; J> J> - nitems = counter_u64_fetch(zone->uz_allocs) - J> - counter_u64_fetch(zone->uz_frees); J> + nitems = 0; J> + if (zone->uz_allocs != EARLY_COUNTER && zone->uz_frees != EARLY_COUNTER) J> + nitems = counter_u64_fetch(zone->uz_allocs) - J> + counter_u64_fetch(zone->uz_frees); J> CPU_FOREACH(i) J> nitems += atomic_load_64(>uz_cpu[i].uc_allocs) - J> atomic_load_64(>uz_cpu[i].uc_frees); J> @@ -4168,7 +4170,9 @@ uma_zone_get_allocs(uma_zone_t zone) J> uint64_t nitems; J> u_int i; J> J> - nitems = counter_u64_fetch(zone->uz_allocs); J> + nitems = 0; J> + if (zone->uz_allocs != EARLY_COUNTER) J> + nitems = counter_u64_fetch(zone->uz_allocs); J> CPU_FOREACH(i) J> nitems += atomic_load_64(>uz_cpu[i].uc_allocs); J> J> @@ -4181,7 +4185,9 @@ uma_zone_get_frees(uma_zone_t zone) J> uint64_t nitems; J> u_int i; J> J> - nitems = counter_u64_fetch(zone->uz_frees); J> + nitems = 0; J> + if (zone->uz_frees != EARLY_COUNTER) J> + nitems = counter_u64_fetch(zone->uz_frees); J> CPU_FOREACH(i) J> nitems += atomic_load_64(>uz_cpu[i].uc_frees); IMHO, tidier code would be not to check the pointers, but check UMA booted status: if (__predict_true(booted == BOOT_RUNNING)) nitems = counter_u64_fetch(zone->uz_frees); That seems fine. I have drained my UMA patch queue for now but if I come back around I will include this. Thanks, Jeff -- Gleb Smirnoff ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
Re: svn commit: r356389 - head/sys/vm
Jeff, On Sun, Jan 05, 2020 at 10:54:26PM +, Jeff Roberson wrote: J> Author: jeff J> Date: Sun Jan 5 22:54:25 2020 J> New Revision: 356389 J> URL: https://svnweb.freebsd.org/changeset/base/356389 J> J> Log: J> The fix in r356353 was insufficient. Not every architecture returns 0 for J> EARLY_COUNTER. Only amd64 seems to. J> J> Suggested by: markj J> Reported by: lwhsu J> Reviewed by: markj J> PR:243117 J> J> Modified: J> head/sys/vm/uma_core.c J> J> Modified: head/sys/vm/uma_core.c J> == J> --- head/sys/vm/uma_core.c Sun Jan 5 21:35:02 2020(r356388) J> +++ head/sys/vm/uma_core.c Sun Jan 5 22:54:25 2020(r356389) J> @@ -4153,8 +4153,10 @@ uma_zone_get_cur(uma_zone_t zone) J> int64_t nitems; J> u_int i; J> J> -nitems = counter_u64_fetch(zone->uz_allocs) - J> -counter_u64_fetch(zone->uz_frees); J> +nitems = 0; J> +if (zone->uz_allocs != EARLY_COUNTER && zone->uz_frees != EARLY_COUNTER) J> +nitems = counter_u64_fetch(zone->uz_allocs) - J> +counter_u64_fetch(zone->uz_frees); J> CPU_FOREACH(i) J> nitems += atomic_load_64(>uz_cpu[i].uc_allocs) - J> atomic_load_64(>uz_cpu[i].uc_frees); J> @@ -4168,7 +4170,9 @@ uma_zone_get_allocs(uma_zone_t zone) J> uint64_t nitems; J> u_int i; J> J> -nitems = counter_u64_fetch(zone->uz_allocs); J> +nitems = 0; J> +if (zone->uz_allocs != EARLY_COUNTER) J> +nitems = counter_u64_fetch(zone->uz_allocs); J> CPU_FOREACH(i) J> nitems += atomic_load_64(>uz_cpu[i].uc_allocs); J> J> @@ -4181,7 +4185,9 @@ uma_zone_get_frees(uma_zone_t zone) J> uint64_t nitems; J> u_int i; J> J> -nitems = counter_u64_fetch(zone->uz_frees); J> +nitems = 0; J> +if (zone->uz_frees != EARLY_COUNTER) J> +nitems = counter_u64_fetch(zone->uz_frees); J> CPU_FOREACH(i) J> nitems += atomic_load_64(>uz_cpu[i].uc_frees); IMHO, tidier code would be not to check the pointers, but check UMA booted status: if (__predict_true(booted == BOOT_RUNNING)) nitems = counter_u64_fetch(zone->uz_frees); -- Gleb Smirnoff ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r356426 - in head/sys/arm64: arm64 include
Author: andrew
Date: Mon Jan 6 20:57:59 2020
New Revision: 356426
URL: https://svnweb.freebsd.org/changeset/base/356426
Log:
Add more Arm arm64 CPU identification values
- Add all the Cortex-A CPU ID register values I can find.
- Add the Neoverse-N1 ID regiser value [1]
- Sort macros by register value.
PR: 243065
Submitted by: Ali Saidi [1]
Sponsored by: DARPA, AFRL (other than [1])
Modified:
head/sys/arm64/arm64/identcpu.c
head/sys/arm64/include/cpu.h
Modified: head/sys/arm64/arm64/identcpu.c
==
--- head/sys/arm64/arm64/identcpu.c Mon Jan 6 19:47:59 2020
(r356425)
+++ head/sys/arm64/arm64/identcpu.c Mon Jan 6 20:57:59 2020
(r356426)
@@ -156,9 +156,14 @@ static const struct cpu_parts cpu_parts_arm[] = {
{ CPU_PART_CORTEX_A53, "Cortex-A53" },
{ CPU_PART_CORTEX_A55, "Cortex-A55" },
{ CPU_PART_CORTEX_A57, "Cortex-A57" },
+ { CPU_PART_CORTEX_A65, "Cortex-A65" },
{ CPU_PART_CORTEX_A72, "Cortex-A72" },
{ CPU_PART_CORTEX_A73, "Cortex-A73" },
{ CPU_PART_CORTEX_A75, "Cortex-A75" },
+ { CPU_PART_CORTEX_A76, "Cortex-A76" },
+ { CPU_PART_CORTEX_A76AE, "Cortex-A76AE" },
+ { CPU_PART_CORTEX_A77, "Cortex-A77" },
+ { CPU_PART_NEOVERSE_N1, "Neoverse-N1" },
CPU_PART_NONE,
};
/* Cavium */
Modified: head/sys/arm64/include/cpu.h
==
--- head/sys/arm64/include/cpu.hMon Jan 6 19:47:59 2020
(r356425)
+++ head/sys/arm64/include/cpu.hMon Jan 6 20:57:59 2020
(r356426)
@@ -81,13 +81,18 @@
/* ARM Part numbers */
#defineCPU_PART_FOUNDATION 0xD00
-#defineCPU_PART_CORTEX_A35 0xD04
#defineCPU_PART_CORTEX_A53 0xD03
+#defineCPU_PART_CORTEX_A35 0xD04
#defineCPU_PART_CORTEX_A55 0xD05
+#defineCPU_PART_CORTEX_A65 0xD06
#defineCPU_PART_CORTEX_A57 0xD07
#defineCPU_PART_CORTEX_A72 0xD08
#defineCPU_PART_CORTEX_A73 0xD09
#defineCPU_PART_CORTEX_A75 0xD0A
+#defineCPU_PART_CORTEX_A76 0xD0B
+#defineCPU_PART_NEOVERSE_N10xD0C
+#defineCPU_PART_CORTEX_A77 0xD0D
+#defineCPU_PART_CORTEX_A76AE 0xD0E
/* Cavium Part numbers */
#defineCPU_PART_THUNDERX 0x0A1
___
svn-src-head@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r356425 - head/sys/kern
Author: kaktus
Date: Mon Jan 6 19:47:59 2020
New Revision: 356425
URL: https://svnweb.freebsd.org/changeset/base/356425
Log:
kern_sysctl: make sysctl.debug work as intended
r136999 introduced SYSTCL_DEBUG but apparently "opt_sysctl.h" was never
included making the option ignored.
r322954 introduced sysctl.reuse_test with OID number equal to 0, effectively
shadowing the very special sysctl.debug one. Use OID_AUTO as it doesn't need
any special treatment.
Reviewed by: kib (mentor)
Approved by: kib (mentor)
Differential Revision:https://reviews.freebsd.org/D23056
Modified:
head/sys/kern/kern_sysctl.c
Modified: head/sys/kern/kern_sysctl.c
==
--- head/sys/kern/kern_sysctl.c Mon Jan 6 19:35:22 2020(r356424)
+++ head/sys/kern/kern_sysctl.c Mon Jan 6 19:47:59 2020(r356425)
@@ -43,6 +43,7 @@ __FBSDID("$FreeBSD$");
#include "opt_capsicum.h"
#include "opt_ddb.h"
#include "opt_ktrace.h"
+#include "opt_sysctl.h"
#include
#include
@@ -404,8 +405,9 @@ sysctl_reuse_test(SYSCTL_HANDLER_ARGS)
SYSCTL_RUNLOCK();
return (0);
}
-SYSCTL_PROC(_sysctl, 0, reuse_test, CTLTYPE_STRING|CTLFLAG_RD|CTLFLAG_MPSAFE,
- 0, 0, sysctl_reuse_test, "-", "");
+SYSCTL_PROC(_sysctl, OID_AUTO, reuse_test,
+CTLTYPE_STRING | CTLFLAG_RD | CTLFLAG_MPSAFE, 0, 0, sysctl_reuse_test, "-",
+"");
#endif
void
___
svn-src-head@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r356424 - head/stand/libsa/zfs
Author: tsoome
Date: Mon Jan 6 19:35:22 2020
New Revision: 356424
URL: https://svnweb.freebsd.org/changeset/base/356424
Log:
loader: update zfsimpl.c from illumos review suggestions
Add extra comments and update error handling.
MFH: 1 week
Modified:
head/stand/libsa/zfs/zfsimpl.c
Modified: head/stand/libsa/zfs/zfsimpl.c
==
--- head/stand/libsa/zfs/zfsimpl.c Mon Jan 6 18:25:58 2020
(r356423)
+++ head/stand/libsa/zfs/zfsimpl.c Mon Jan 6 19:35:22 2020
(r356424)
@@ -1097,7 +1097,8 @@ vdev_create(uint64_t guid, vdev_read_t *_read)
vdev->v_read = _read;
/*
-* root vdev has no read function.
+* root vdev has no read function, we use this fact to
+* skip setting up data we do not need for root vdev.
* We only point root vdev from spa.
*/
if (_read != NULL) {
@@ -1142,7 +1143,7 @@ vdev_set_initial_state(vdev_t *vdev, const unsigned ch
else if (isnt_present != 0)
vdev->v_state = VDEV_STATE_CANT_OPEN;
- vdev->v_islog = is_log == 1;
+ vdev->v_islog = is_log != 0;
}
static int
@@ -1234,9 +1235,8 @@ vdev_init(uint64_t guid, const unsigned char *nvlist,
if (strcmp(type, "raidz") == 0) {
if (vdev->v_nparity < 1 ||
vdev->v_nparity > 3) {
- printf("ZFS: can only boot from disk, "
- "mirror, raidz1, raidz2 and raidz3 "
- "vdevs\n");
+ printf("ZFS: invalid raidz parity: %d\n",
+ vdev->v_nparity);
return (EIO);
}
(void) asprintf(, "%s%d-%" PRIu64, type,
@@ -1310,16 +1310,17 @@ vdev_insert(vdev_t *top_vdev, vdev_t *vdev)
if (previous == NULL) {
STAILQ_INSERT_HEAD(_vdev->v_children, vdev, v_childlink);
- count = vdev_child_count(top_vdev);
- if (top_vdev->v_nchildren < count)
- top_vdev->v_nchildren = count;
+ } else if (previous->v_id == vdev->v_id) {
+ /*
+* This vdev was configured from label config,
+* do not insert duplicate.
+*/
return;
+ } else {
+ STAILQ_INSERT_AFTER(_vdev->v_children, previous, vdev,
+ v_childlink);
}
- if (previous->v_id == vdev->v_id)
- return;
-
- STAILQ_INSERT_AFTER(_vdev->v_children, previous, vdev, v_childlink);
count = vdev_child_count(top_vdev);
if (top_vdev->v_nchildren < count)
top_vdev->v_nchildren = count;
@@ -1365,6 +1366,10 @@ vdev_from_nvlist(spa_t *spa, uint64_t top_guid, const
kids = nvlist_next(kids);
}
} else {
+ /*
+* When there are no children, nvlist_find() does return
+* error, reset it because leaf devices have no children.
+*/
rc = 0;
}
@@ -1491,7 +1496,7 @@ vdev_init_from_nvlist(spa_t *spa, const unsigned char
/* Wrong guid?! */
if (spa->spa_guid != pool_guid)
- return (EIO);
+ return (EINVAL);
spa->spa_root_vdev->v_nchildren = vdev_children;
@@ -1502,7 +1507,7 @@ vdev_init_from_nvlist(spa_t *spa, const unsigned char
* MOS config has at least one child for root vdev.
*/
if (rc != 0)
- return (EIO);
+ return (rc);
for (int i = 0; i < nkids; i++) {
uint64_t guid;
@@ -3266,7 +3271,10 @@ zfs_spa_init(spa_t *spa)
if (rc != 0)
return (rc);
- /* Update vdevs from MOS config. */
+ /*
+* Update vdevs from MOS config. Note, we do skip encoding bytes
+* here. See also vdev_label_read_config().
+*/
rc = vdev_init_from_nvlist(spa, nvlist + 4);
free(nvlist);
return (rc);
___
svn-src-head@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r356423 - in head: share/mk tests/sys/fs
Author: kevans
Date: Mon Jan 6 18:25:58 2020
New Revision: 356423
URL: https://svnweb.freebsd.org/changeset/base/356423
Log:
Refine fusefs test workaround for mips+llvm
This re-enables building the googletest suite by default on mips and instead
specifically doesn't build fusefs tests for mips+clang builds. clang will
easily spent >= 1.5 hours compiling a single file due to a bug in
optimization (see LLVM PR 43263), so turn these off for now while that's
hashed out.
GCC builds are unaffected and build the fusefs tests as-is. Clang builds
only happen by early adopters attempting to hash out the remaining issues.
The comment has been updated to reflect its new position and use less strong
wording about imposing on people.
Discussed with: ngie, asomers
Reviewed by: ngie
Modified:
head/share/mk/src.opts.mk
head/tests/sys/fs/Makefile
Modified: head/share/mk/src.opts.mk
==
--- head/share/mk/src.opts.mk Mon Jan 6 18:15:55 2020(r356422)
+++ head/share/mk/src.opts.mk Mon Jan 6 18:25:58 2020(r356423)
@@ -108,6 +108,7 @@ __DEFAULT_YES_OPTIONS = \
GDB \
GNU_DIFF \
GNU_GREP \
+GOOGLETEST \
GPIO \
HAST \
HTML \
@@ -263,15 +264,6 @@ __T=${MACHINE_ARCH}
__TT=${TARGET}
.else
__TT=${MACHINE}
-.endif
-
-# Default GOOGLETEST to off for MIPS while LLVM PR 43263 is active. Part
-# of the fusefs tests trigger excessively long compile times. It does
-# eventually succeed, but this shouldn't be forced on those building by
default.
-.if ${__TT} == "mips"
-__DEFAULT_NO_OPTIONS+= GOOGLETEST
-.else
-__DEFAULT_YES_OPTIONS+=GOOGLETEST
.endif
# All supported backends for LLVM_TARGET_XXX
Modified: head/tests/sys/fs/Makefile
==
--- head/tests/sys/fs/Makefile Mon Jan 6 18:15:55 2020(r356422)
+++ head/tests/sys/fs/Makefile Mon Jan 6 18:25:58 2020(r356423)
@@ -12,7 +12,12 @@ TESTSRC= ${SRCTOP}/contrib/netbsd-tests/fs
#TESTS_SUBDIRS+= nullfs # XXX: needs rump
# fusefs tests cannot be compiled/used without the googletest infrastructure.
.if ${COMPILER_FEATURES:Mc++14} && ${MK_GOOGLETEST} != "no"
+# Don't build fusefs tests for MIPS while LLVM PR 43263 is active, as part
+# of the fusefs tests trigger excessively long compile time. The build does
+# eventually succeed, but it's perhaps better to not do so by default for now.
+.if ${MACHINE_CPUARCH} != "mips" || ${COMPILER_TYPE} != "clang"
TESTS_SUBDIRS+=fusefs
+.endif
.endif
TESTS_SUBDIRS+=tmpfs
___
svn-src-head@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r356422 - in head: lib/libc lib/libsysdecode targets/pseudo/userland/gnu
Author: kevans Date: Mon Jan 6 18:15:55 2020 New Revision: 356422 URL: https://svnweb.freebsd.org/changeset/base/356422 Log: Update libssp paths in various Makefile.depend* files I've been advised that the model that uses these are fairly resilient, but we do know the proper path to use (or remove, in the case of ^/targets/...), so go ahead and update them to reflect that. Modified: head/lib/libc/Makefile.depend.options head/lib/libsysdecode/Makefile.depend head/targets/pseudo/userland/gnu/Makefile.depend Modified: head/lib/libc/Makefile.depend.options == --- head/lib/libc/Makefile.depend.options Mon Jan 6 18:07:54 2020 (r356421) +++ head/lib/libc/Makefile.depend.options Mon Jan 6 18:15:55 2020 (r356422) @@ -3,6 +3,6 @@ DIRDEPS_OPTIONS= SSP -DIRDEPS.SSP.yes= gnu/lib/libssp/libssp_nonshared +DIRDEPS.SSP.yes= lib/libssp_nonshared .include Modified: head/lib/libsysdecode/Makefile.depend == --- head/lib/libsysdecode/Makefile.depend Mon Jan 6 18:07:54 2020 (r356421) +++ head/lib/libsysdecode/Makefile.depend Mon Jan 6 18:15:55 2020 (r356422) @@ -7,7 +7,6 @@ DIRDEPS = \ gnu/lib/libdialog \ gnu/lib/libgomp \ gnu/lib/libregex \ - gnu/lib/libssp \ gnu/lib/libstdc++ \ gnu/lib/libsupc++ \ include \ @@ -100,6 +99,7 @@ DIRDEPS = \ lib/libsdp \ lib/libsecureboot \ lib/libsqlite3 \ + lib/libssp \ lib/libstand \ lib/libstdthreads \ lib/libtacplus \ Modified: head/targets/pseudo/userland/gnu/Makefile.depend == --- head/targets/pseudo/userland/gnu/Makefile.dependMon Jan 6 18:07:54 2020(r356421) +++ head/targets/pseudo/userland/gnu/Makefile.dependMon Jan 6 18:15:55 2020(r356422) @@ -12,8 +12,6 @@ DIRDEPS = \ gnu/lib/libgcov \ gnu/lib/libgomp \ gnu/lib/libregex/doc \ - gnu/lib/libssp \ - gnu/lib/libssp/libssp_nonshared \ gnu/lib/libstdc++ \ gnu/lib/libsupc++ \ gnu/usr.bin/binutils/ar \ ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r356421 - head/share/man/man5
Author: emaste Date: Mon Jan 6 18:07:54 2020 New Revision: 356421 URL: https://svnweb.freebsd.org/changeset/base/356421 Log: src.conf.5: regen after r356419, reenabling powerpc in universe Modified: head/share/man/man5/src.conf.5 Modified: head/share/man/man5/src.conf.5 == --- head/share/man/man5/src.conf.5 Mon Jan 6 18:02:02 2020 (r356420) +++ head/share/man/man5/src.conf.5 Mon Jan 6 18:07:54 2020 (r356421) @@ -1,6 +1,6 @@ .\" DO NOT EDIT-- this file is @generated by tools/build/options/makeman. .\" $FreeBSD$ -.Dd January 5, 2020 +.Dd January 6, 2020 .Dt SRC.CONF 5 .Os .Sh NAME @@ -197,7 +197,7 @@ as part of the normal system build. .Pp This is a default setting on -amd64/amd64, arm/armv6, arm/armv7, i386/i386, mips/mipsel, mips/mips, mips/mips64el, mips/mips64, mips/mipsn32, mips/mipselhf, mips/mipshf, mips/mips64elhf, mips/mips64hf and sparc64/sparc64. +amd64/amd64, arm/armv6, arm/armv7, i386/i386, powerpc/powerpc, powerpc/powerpc64, mips/mipsel, mips/mips, mips/mips64el, mips/mips64, mips/mipsn32, mips/mipselhf, mips/mipshf, mips/mips64elhf, mips/mips64hf and sparc64/sparc64. .It Va WITHOUT_BINUTILS_BOOTSTRAP Set to not build binutils (as, ld, and objdump) as part of the bootstrap process. @@ -213,7 +213,7 @@ Set build binutils (as, ld, and objdump) as part of the bootstrap process. .Pp This is a default setting on -amd64/amd64, arm/armv6, arm/armv7 and i386/i386. +amd64/amd64, arm/armv6, arm/armv7, i386/i386, powerpc/powerpc and powerpc/powerpc64. .It Va WITHOUT_BLACKLIST Set this if you do not want to build .Xr blacklistd 8 @@ -268,7 +268,7 @@ and .Pa crtend.o . .Pp This is a default setting on -amd64/amd64, arm/armv6, arm/armv7, arm64/aarch64, i386/i386, riscv/riscv64, mips/mipsel, mips/mips, mips/mips64el, mips/mips64, mips/mipsn32, mips/mipselhf, mips/mipshf, mips/mips64elhf and mips/mips64hf. +amd64/amd64, arm/armv6, arm/armv7, arm64/aarch64, i386/i386, powerpc/powerpc, powerpc/powerpc64, riscv/riscv64, mips/mipsel, mips/mips, mips/mips64el, mips/mips64, mips/mipsn32, mips/mipselhf, mips/mipshf, mips/mips64elhf and mips/mips64hf. .It Va WITH_BSD_GREP Install BSD-licensed grep as '[ef]grep' instead of GNU grep. .It Va WITHOUT_BSNMP @@ -380,7 +380,7 @@ When set, it enforces these options: Set to build the Clang C/C++ compiler during the normal phase of the build. .Pp This is a default setting on -amd64/amd64, arm/armv6, arm/armv7, arm64/aarch64, i386/i386, riscv/riscv64, mips/mipsel, mips/mips, mips/mips64el, mips/mips64, mips/mipsn32, mips/mipselhf, mips/mipshf, mips/mips64elhf and mips/mips64hf. +amd64/amd64, arm/armv6, arm/armv7, arm64/aarch64, i386/i386, powerpc/powerpc, powerpc/powerpc64, riscv/riscv64, mips/mipsel, mips/mips, mips/mips64el, mips/mips64, mips/mipsn32, mips/mipselhf, mips/mipshf, mips/mips64elhf and mips/mips64hf. .It Va WITHOUT_CLANG_BOOTSTRAP Set to not build the Clang C/C++ compiler during the bootstrap phase of the build. @@ -393,7 +393,7 @@ riscv/riscv64, mips/mipsel, mips/mips, mips/mips64el, Set to build the Clang C/C++ compiler during the bootstrap phase of the build. .Pp This is a default setting on -amd64/amd64, arm/armv6, arm/armv7, arm64/aarch64 and i386/i386. +amd64/amd64, arm/armv6, arm/armv7, arm64/aarch64, i386/i386, powerpc/powerpc and powerpc/powerpc64. .It Va WITH_CLANG_EXTRAS Set to build additional clang and llvm tools, such as bugpoint and clang-format. @@ -408,7 +408,7 @@ Set to build the ARCMigrate, Rewriter and StaticAnalyz Clang C/C++ compiler. .Pp This is a default setting on -amd64/amd64, arm/armv6, arm/armv7, arm64/aarch64, i386/i386, riscv/riscv64, mips/mipsel, mips/mips, mips/mips64el, mips/mips64, mips/mipsn32, mips/mipselhf, mips/mipshf, mips/mips64elhf and mips/mips64hf. +amd64/amd64, arm/armv6, arm/armv7, arm64/aarch64, i386/i386, powerpc/powerpc, powerpc/powerpc64, riscv/riscv64, mips/mipsel, mips/mips, mips/mips64el, mips/mips64, mips/mipsn32, mips/mipselhf, mips/mipshf, mips/mips64elhf and mips/mips64hf. .It Va WITHOUT_CLANG_IS_CC Set to install the GCC compiler as .Pa /usr/bin/cc , @@ -426,7 +426,7 @@ and .Pa /usr/bin/cpp . .Pp This is a default setting on -amd64/amd64, arm/armv6, arm/armv7, arm64/aarch64 and i386/i386. +amd64/amd64, arm/armv6, arm/armv7, arm64/aarch64, i386/i386, powerpc/powerpc and powerpc/powerpc64. .It Va WITHOUT_CPP Set to not build .Xr cpp 1 . @@ -495,13 +495,13 @@ Set to not build .Xr cxgbetool 8 .Pp This is a default setting on -arm/armv6, arm/armv7, riscv/riscv64, mips/mipsel, mips/mips, mips/mips64el, mips/mips64, mips/mipsn32, mips/mipselhf, mips/mipshf, mips/mips64elhf and mips/mips64hf. +arm/armv6, arm/armv7, powerpc/powerpc, riscv/riscv64, mips/mipsel, mips/mips, mips/mips64el, mips/mips64, mips/mipsn32, mips/mipselhf, mips/mipshf, mips/mips64elhf and mips/mips64hf. .It Va WITH_CXGBETOOL Set to build .Xr cxgbetool 8
svn commit: r356420 - head/sys/mips/mips
Author: jhb
Date: Mon Jan 6 18:02:02 2020
New Revision: 356420
URL: https://svnweb.freebsd.org/changeset/base/356420
Log:
Simplify arguments to signal handlers on mips.
- Use ksi_addr directly as si_addr in the siginfo instead of the
'badvaddr' register.
- Remove a duplicate assignment of si_code.
- Use ksi_addr as the 4th argument to the old-style handler instead of
'badvaddr'.
Reviewed by: brooks, kevans
Sponsored by: DARPA
Differential Revision:https://reviews.freebsd.org/D23013
Modified:
head/sys/mips/mips/pm_machdep.c
Modified: head/sys/mips/mips/pm_machdep.c
==
--- head/sys/mips/mips/pm_machdep.c Mon Jan 6 17:36:28 2020
(r356419)
+++ head/sys/mips/mips/pm_machdep.c Mon Jan 6 18:02:02 2020
(r356420)
@@ -147,12 +147,10 @@ sendsig(sig_t catcher, ksiginfo_t *ksi, sigset_t *mask
/* fill siginfo structure */
sf.sf_si = ksi->ksi_info;
sf.sf_si.si_signo = sig;
- sf.sf_si.si_code = ksi->ksi_code;
- sf.sf_si.si_addr = (void*)(intptr_t)regs->badvaddr;
} else {
/* Old FreeBSD-style arguments. */
regs->a1 = ksi->ksi_code;
- regs->a3 = regs->badvaddr;
+ regs->a3 = (uintptr_t)ksi->ksi_addr;
/* sf.sf_ahu.sf_handler = catcher; */
}
___
svn-src-head@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r356419 - head
Author: jhb
Date: Mon Jan 6 17:36:28 2020
New Revision: 356419
URL: https://svnweb.freebsd.org/changeset/base/356419
Log:
Enable powerpc in make tinderbox using clang instead of GCC.
powerpcspe is disabled for now until clang/llvm issues with spe
have been fixed.
Reviewed by: imp
Differential Revision:https://reviews.freebsd.org/D23031
Modified:
head/Makefile
Modified: head/Makefile
==
--- head/Makefile Mon Jan 6 17:34:17 2020(r356418)
+++ head/Makefile Mon Jan 6 17:36:28 2020(r356419)
@@ -491,15 +491,13 @@ worlds: .PHONY
# honor that most of all.
#
_OBSOLETE_GCC_TARGETS=mips sparc64
-.if defined(MAKE_OBSOLETE_GCC)
-_OBSOLETE_GCC_TARGETS+=powerpc
-.endif
-TARGETS?=amd64 arm arm64 i386 riscv ${_OBSOLETE_GCC_TARGETS}
+TARGETS?=amd64 arm arm64 i386 powerpc riscv ${_OBSOLETE_GCC_TARGETS}
_UNIVERSE_TARGETS= ${TARGETS}
TARGET_ARCHES_arm?=armv6 armv7
TARGET_ARCHES_arm64?= aarch64
TARGET_ARCHES_mips?= mipsel mips mips64el mips64 mipsn32 mipselhf mipshf
mips64elhf mips64hf
-TARGET_ARCHES_powerpc?=powerpc powerpc64 powerpcspe
+# powerpcspe excluded until clang fixed
+TARGET_ARCHES_powerpc?=powerpc powerpc64
# riscv64sf excluded due to PR 232085
TARGET_ARCHES_riscv?= riscv64
.for target in ${TARGETS}
@@ -510,12 +508,10 @@ MAKE_PARAMS_riscv?= CROSS_TOOLCHAIN=riscv64-gcc
.if !defined(MAKE_OBSOLETE_GCC)
OBSOLETE_GCC_TARGETS=${_OBSOLETE_GCC_TARGETS}
MAKE_PARAMS_mips?= CROSS_TOOLCHAIN=mips-gcc6
-MAKE_PARAMS_powerpc?= CROSS_TOOLCHAIN=powerpc64-gcc6
MAKE_PARAMS_sparc64?= CROSS_TOOLCHAIN=sparc64-gcc6
.endif
TOOLCHAINS_mips= mips-gcc6
-TOOLCHAINS_powerpc=powerpc64-gcc6
TOOLCHAINS_riscv= riscv64-gcc
TOOLCHAINS_sparc64=sparc64-gcc6
___
svn-src-head@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r356418 - head
Author: jhb
Date: Mon Jan 6 17:34:17 2020
New Revision: 356418
URL: https://svnweb.freebsd.org/changeset/base/356418
Log:
Only build kernels for enabled TARGET_ARCHes in make universe/tinderbox.
Previously, all of the kernels for a given TARGET were built if that
target was enabled. This was implemented by having each kernel built
via a universe_kernconf_ target that was depended on by a
universe_kernconfs target. However, this meant that if one did a
build with a limited set of TARGET_ARCH values for a given TARGET,
kernels could be built for which we hadn't built a world or toolchain.
For example, 'make TARGETS=mips TARGET_ARCHES_mips=mips64' would build
mips32 kernels.
Fix this by adding an extra layer of indirection in the kernel make
targets. universe_kernconf_ is now a dependency of a new
universe_kernconfs_. universe_kernconfs in turn depends
on a list of universe_kernconfs_ values, but only the
values enabled in TARGET_ARCHES_.
Reviewed by: imp
MFC after:1 month
Differential Revision:https://reviews.freebsd.org/D23031
Modified:
head/Makefile
Modified: head/Makefile
==
--- head/Makefile Mon Jan 6 15:29:14 2020(r356417)
+++ head/Makefile Mon Jan 6 17:34:17 2020(r356418)
@@ -724,7 +724,7 @@ TARGET_ARCH_${kernel}!= cd ${KERNSRCDIR}/${TARGET}/con
.if empty(TARGET_ARCH_${kernel})
.error "Target architecture for ${TARGET}/conf/${kernel} unknown. config(8)
likely too old."
.endif
-universe_kernconfs: universe_kernconf_${TARGET}_${kernel}
+universe_kernconfs_${TARGET_ARCH_${kernel}}:
universe_kernconf_${TARGET}_${kernel}
universe_kernconf_${TARGET}_${kernel}: .MAKE
@echo ">> ${TARGET}.${TARGET_ARCH_${kernel}} ${kernel} kernel started
on `LC_ALL=C date`"
@(cd ${.CURDIR} && env __MAKE_CONF=/dev/null \
@@ -737,6 +737,9 @@ universe_kernconf_${TARGET}_${kernel}: .MAKE
(echo "${TARGET} ${kernel} kernel failed," \
"check _.${TARGET}.${kernel} for details"| ${MAKEFAIL}))
@echo ">> ${TARGET}.${TARGET_ARCH_${kernel}} ${kernel} kernel completed
on `LC_ALL=C date`"
+.endfor
+.for target_arch in ${TARGET_ARCHES_${TARGET}}
+universe_kernconfs: universe_kernconfs_${target_arch} .PHONY
.endfor
.endif # make(universe_kernels)
universe: universe_epilogue
___
svn-src-head@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r356417 - in head/sys/netinet: . tcp_stacks
Author: rrs
Date: Mon Jan 6 15:29:14 2020
New Revision: 356417
URL: https://svnweb.freebsd.org/changeset/base/356417
Log:
This catches rack up in the recent changes to ECN and
also commonizes the functions that both the freebsd and
rack stack uses.
Sponsored by:Netflix Inc
Differential Revision:https://reviews.freebsd.org/D23052
Modified:
head/sys/netinet/tcp_input.c
head/sys/netinet/tcp_stacks/rack.c
head/sys/netinet/tcp_var.h
Modified: head/sys/netinet/tcp_input.c
==
--- head/sys/netinet/tcp_input.cMon Jan 6 13:21:10 2020
(r356416)
+++ head/sys/netinet/tcp_input.cMon Jan 6 15:29:14 2020
(r356417)
@@ -514,7 +514,7 @@ cc_post_recovery(struct tcpcb *tp, struct tcphdr *th)
(tlen <= tp->t_maxseg) && \
(V_tcp_delack_enabled || (tp->t_flags & TF_NEEDSYN)))
-static void inline
+void inline
cc_ecnpkt_handler(struct tcpcb *tp, struct tcphdr *th, uint8_t iptos)
{
INP_WLOCK_ASSERT(tp->t_inpcb);
Modified: head/sys/netinet/tcp_stacks/rack.c
==
--- head/sys/netinet/tcp_stacks/rack.c Mon Jan 6 13:21:10 2020
(r356416)
+++ head/sys/netinet/tcp_stacks/rack.c Mon Jan 6 15:29:14 2020
(r356417)
@@ -7715,6 +7715,10 @@ rack_do_segment_nounlock(struct mbuf *m, struct tcphdr
TCPSTAT_INC(tcps_ecn_ect1);
break;
}
+
+ /* Process a packet differently from RFC3168. */
+ cc_ecnpkt_handler(tp, th, iptos);
+
/* Congestion experienced. */
if (thflags & TH_ECE) {
rack_cong_signal(tp, th, CC_ECN);
Modified: head/sys/netinet/tcp_var.h
==
--- head/sys/netinet/tcp_var.h Mon Jan 6 13:21:10 2020(r356416)
+++ head/sys/netinet/tcp_var.h Mon Jan 6 15:29:14 2020(r356417)
@@ -891,6 +891,7 @@ voidcc_ack_received(struct tcpcb *tp, struct tcphdr
*
uint16_t nsegs, uint16_t type);
void cc_conn_init(struct tcpcb *tp);
void cc_post_recovery(struct tcpcb *tp, struct tcphdr *th);
+voidcc_ecnpkt_handler(struct tcpcb *tp, struct tcphdr *th, uint8_t iptos);
void cc_cong_signal(struct tcpcb *tp, struct tcphdr *th, uint32_t type);
#ifdef TCP_HHOOK
void hhook_run_tcp_est_in(struct tcpcb *tp,
___
svn-src-head@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
Re: svn commit: r356097 - in head: share/man/man9 sys/compat/ndis sys/dev/oce sys/kern sys/libkern sys/sys
On 2019-12-26 20:41, Conrad Meyer wrote: Author: cem Date: Thu Dec 26 19:41:09 2019 New Revision: 356097 URL: https://svnweb.freebsd.org/changeset/base/356097 Log: random(9): Deprecate random(9), remove meaningless srandom(9) srandom(9) is meaningless on SMP systems or any system with, say, interrupts. One could never rely on random(9) to produce a reproducible sequence of outputs on the basis of a specific srandom() seed because the global state was shared by all kernel contexts. As such, removing it is literally indistinguishable to random(9) consumers (as compared with retaining it). Mark random(9) as deprecated and slated for quick removal. This is not to say we intend to remove all fast, non-cryptographic PRNG(s) in the kernel. It/they just won't be random(9), as it exists today, in either name or implementation. Before random(9) is removed, a replacement will be provided and in-tree consumers will be converted. Note that despite the name, the random(9) interface does not bear any resemblance to random(3). Instead, it is the same crummy 1988 Park-Miller LCG used in libc rand(3). Hi, Having some kind of replacment for this using atomics(), might be useful for system testing. Sometimes you need a white-noise function for initializing test parameters and you also need srandom to reproducue failing tests. --HPS ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r356416 - head/contrib/libarchive/libarchive
Author: mm
Date: Mon Jan 6 13:21:10 2020
New Revision: 356416
URL: https://svnweb.freebsd.org/changeset/base/356416
Log:
MFV r356415
Sync libarchive with vendor
Relevant vendor changes:
Issue #1302: Re-do fix for archive_write_client_open()
X-MFC-With: r356212,r356365
MFC after:1 week
Modified:
head/contrib/libarchive/libarchive/archive_cryptor.c
head/contrib/libarchive/libarchive/archive_hmac.c
head/contrib/libarchive/libarchive/archive_write.c
Directory Properties:
head/contrib/libarchive/ (props changed)
Modified: head/contrib/libarchive/libarchive/archive_cryptor.c
==
--- head/contrib/libarchive/libarchive/archive_cryptor.cMon Jan 6
13:13:58 2020(r356415)
+++ head/contrib/libarchive/libarchive/archive_cryptor.cMon Jan 6
13:21:10 2020(r356416)
@@ -299,6 +299,7 @@ aes_ctr_release(archive_crypto_ctx *ctx)
}
#elif defined(HAVE_LIBMBEDCRYPTO) && defined(HAVE_MBEDTLS_AES_H)
+
static int
aes_ctr_init(archive_crypto_ctx *ctx, const uint8_t *key, size_t key_len)
{
Modified: head/contrib/libarchive/libarchive/archive_hmac.c
==
--- head/contrib/libarchive/libarchive/archive_hmac.c Mon Jan 6 13:13:58
2020(r356415)
+++ head/contrib/libarchive/libarchive/archive_hmac.c Mon Jan 6 13:21:10
2020(r356416)
@@ -151,6 +151,7 @@ __hmac_sha1_cleanup(archive_hmac_sha1_ctx *ctx)
}
#elif defined(HAVE_LIBMBEDCRYPTO) && defined(HAVE_MBEDTLS_MD_H)
+
static int
__hmac_sha1_init(archive_hmac_sha1_ctx *ctx, const uint8_t *key, size_t
key_len)
{
Modified: head/contrib/libarchive/libarchive/archive_write.c
==
--- head/contrib/libarchive/libarchive/archive_write.c Mon Jan 6 13:13:58
2020(r356415)
+++ head/contrib/libarchive/libarchive/archive_write.c Mon Jan 6 13:21:10
2020(r356416)
@@ -339,6 +339,7 @@ archive_write_client_open(struct archive_write_filter
struct archive_none *state;
void *buffer;
size_t buffer_size;
+ int ret;
f->bytes_per_block = archive_write_get_bytes_per_block(f->archive);
f->bytes_in_last_block =
@@ -363,7 +364,13 @@ archive_write_client_open(struct archive_write_filter
if (a->client_opener == NULL)
return (ARCHIVE_OK);
- return (a->client_opener(f->archive, a->client_data));
+ ret = a->client_opener(f->archive, a->client_data);
+ if (ret != ARCHIVE_OK) {
+ free(state->buffer);
+ free(state);
+ f->data = NULL;
+ }
+ return (ret);
}
static int
@@ -449,30 +456,6 @@ archive_write_client_write(struct archive_write_filter
}
static int
-archive_write_client_free(struct archive_write_filter *f)
-{
- struct archive_write *a = (struct archive_write *)f->archive;
- struct archive_none *state = (struct archive_none *)f->data;
-
- if (state != NULL) {
- free(state->buffer);
- free(state);
- state = NULL;
- }
-
- a->client_data = NULL;
- /* Clear passphrase. */
- if (a->passphrase != NULL) {
- memset(a->passphrase, 0, strlen(a->passphrase));
- free(a->passphrase);
- a->passphrase = NULL;
- }
-
- return (ARCHIVE_OK);
-}
-
-
-static int
archive_write_client_close(struct archive_write_filter *f)
{
struct archive_write *a = (struct archive_write *)f->archive;
@@ -508,7 +491,15 @@ archive_write_client_close(struct archive_write_filter
}
if (a->client_closer)
(*a->client_closer)(>archive, a->client_data);
-
+ free(state->buffer);
+ free(state);
+ a->client_data = NULL;
+ /* Clear passphrase. */
+ if (a->passphrase != NULL) {
+ memset(a->passphrase, 0, strlen(a->passphrase));
+ free(a->passphrase);
+ a->passphrase = NULL;
+ }
/* Clear the close handler myself not to be called again. */
f->state = ARCHIVE_WRITE_FILTER_STATE_CLOSED;
return (ret);
@@ -539,7 +530,6 @@ archive_write_open(struct archive *_a, void *client_da
client_filter->open = archive_write_client_open;
client_filter->write = archive_write_client_write;
client_filter->close = archive_write_client_close;
- client_filter->free = archive_write_client_free;
ret = __archive_write_filters_open(a);
if (ret < ARCHIVE_WARN) {
___
svn-src-head@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r356414 - head/sys/netinet
Author: rrs
Date: Mon Jan 6 12:48:06 2020
New Revision: 356414
URL: https://svnweb.freebsd.org/changeset/base/356414
Log:
This change adds a small feature to the tcp logging code. Basically
a connection can now have a separate tag added to the id.
Obtained from:Lawrence Stewart
Sponsored by: Netflix Inc
Differential Revision:https://reviews.freebsd.org/D22866
Modified:
head/sys/netinet/tcp_log_buf.c
head/sys/netinet/tcp_log_buf.h
head/sys/netinet/tcp_var.h
Modified: head/sys/netinet/tcp_log_buf.c
==
--- head/sys/netinet/tcp_log_buf.c Mon Jan 6 10:52:13 2020
(r356413)
+++ head/sys/netinet/tcp_log_buf.c Mon Jan 6 12:48:06 2020
(r356414)
@@ -43,7 +43,7 @@ __FBSDID("$FreeBSD$");
#include
#include
#include
-#include
+#include /* Must come after qmath.h and tree.h */
#include
#include
@@ -78,6 +78,7 @@ static u_long tcp_log_auto_ratio = 0;
static volatile u_long tcp_log_auto_ratio_cur = 0;
static uint32_t tcp_log_auto_mode = TCP_LOG_STATE_TAIL;
static bool tcp_log_auto_all = false;
+static uint32_t tcp_disable_all_bb_logs = 0;
RB_PROTOTYPE_STATIC(tcp_log_id_tree, tcp_log_id_bucket, tlb_rb, tcp_log_id_cmp)
@@ -111,6 +112,10 @@ SYSCTL_UMA_CUR(_net_inet_tcp_bb, OID_AUTO, log_id_tcpc
SYSCTL_U32(_net_inet_tcp_bb, OID_AUTO, log_version, CTLFLAG_RD,
_log_version,
0, "Version of log formats exported");
+SYSCTL_U32(_net_inet_tcp_bb, OID_AUTO, disable_all, CTLFLAG_RW,
+_disable_all_bb_logs, TCP_LOG_STATE_HEAD_AUTO,
+"Disable all BB logging for all connections");
+
SYSCTL_ULONG(_net_inet_tcp_bb, OID_AUTO, log_auto_ratio, CTLFLAG_RW,
_log_auto_ratio, 0, "Do auto capturing for 1 out of N sessions");
@@ -156,7 +161,18 @@ SYSCTL_COUNTER_U64(_net_inet_tcp_bb, OID_AUTO, freed,
#ifdef INVARIANTS
#defineTCPLOG_DEBUG_RINGBUF
#endif
+/* Number of requests to consider a PBCID "active". */
+#defineACTIVE_REQUEST_COUNT10
+/* Statistic tracking for "active" PBCIDs. */
+static counter_u64_t tcp_log_pcb_ids_cur;
+static counter_u64_t tcp_log_pcb_ids_tot;
+
+SYSCTL_COUNTER_U64(_net_inet_tcp_bb, OID_AUTO, pcb_ids_cur, CTLFLAG_RD,
+_log_pcb_ids_cur, "Number of pcb IDs allocated in the system");
+SYSCTL_COUNTER_U64(_net_inet_tcp_bb, OID_AUTO, pcb_ids_tot, CTLFLAG_RD,
+_log_pcb_ids_tot, "Total number of pcb IDs that have been allocated");
+
struct tcp_log_mem
{
STAILQ_ENTRY(tcp_log_mem) tlm_queue;
@@ -240,10 +256,14 @@ struct tcp_log_id_bucket
* (struct tcp_log_id_bucket *) and (char *) interchangeably.
*/
chartlb_id[TCP_LOG_ID_LEN];
+ chartlb_tag[TCP_LOG_TAG_LEN];
RB_ENTRY(tcp_log_id_bucket) tlb_rb;
struct tcp_log_id_head tlb_head;
struct mtx tlb_mtx;
volatile u_int tlb_refcnt;
+ volatile u_int tlb_reqcnt;
+ uint32_ttlb_loglimit;
+ uint8_t tlb_logstate;
};
struct tcp_log_id_node
@@ -285,6 +305,7 @@ tcp_log_selectauto(void)
* this session.
*/
if (tcp_log_auto_ratio &&
+ (tcp_disable_all_bb_logs == 0) &&
(atomic_fetchadd_long(_log_auto_ratio_cur, 1) %
tcp_log_auto_ratio) == 0)
return (true);
@@ -337,6 +358,7 @@ tcp_log_remove_bucket(struct tcp_log_id_bucket *tlb)
#endif
}
TCPID_BUCKET_LOCK_DESTROY(tlb);
+ counter_u64_add(tcp_log_pcb_ids_cur, (int64_t)-1);
uma_zfree(tcp_log_bucket_zone, tlb);
}
@@ -484,7 +506,53 @@ tcp_log_grow_tlb(char *tlb_id, struct tcpcb *tp)
#endif
}
+static void
+tcp_log_increment_reqcnt(struct tcp_log_id_bucket *tlb)
+{
+
+ atomic_fetchadd_int(>tlb_reqcnt, 1);
+}
+
/*
+ * Associate the specified tag with a particular TCP log ID.
+ * Called with INPCB locked. Returns with it unlocked.
+ * Returns 0 on success or EOPNOTSUPP if the connection has no TCP log ID.
+ */
+int
+tcp_log_set_tag(struct tcpcb *tp, char *tag)
+{
+ struct tcp_log_id_bucket *tlb;
+ int tree_locked;
+
+ INP_WLOCK_ASSERT(tp->t_inpcb);
+
+ tree_locked = TREE_UNLOCKED;
+ tlb = tp->t_lib;
+ if (tlb == NULL) {
+ INP_WUNLOCK(tp->t_inpcb);
+ return (EOPNOTSUPP);
+ }
+
+ TCPID_BUCKET_REF(tlb);
+ INP_WUNLOCK(tp->t_inpcb);
+ TCPID_BUCKET_LOCK(tlb);
+ strlcpy(tlb->tlb_tag, tag, TCP_LOG_TAG_LEN);
+ if (!tcp_log_unref_bucket(tlb, _locked, NULL))
+ TCPID_BUCKET_UNLOCK(tlb);
+
+ if (tree_locked == TREE_WLOCKED) {
+ TCPID_TREE_WLOCK_ASSERT();
+ TCPID_TREE_WUNLOCK();
+ } else if (tree_locked == TREE_RLOCKED) {
+ TCPID_TREE_RLOCK_ASSERT();
+ TCPID_TREE_RUNLOCK();
+ } else
+
svn commit: r356413 - in head/sys: amd64/amd64 arm64/arm64
Author: kaktus Date: Mon Jan 6 10:52:13 2020 New Revision: 356413 URL: https://svnweb.freebsd.org/changeset/base/356413 Log: sysctl: mark more nodes as MPSAFE vm.kvm_size and vm.kvm_free are read only and marked as MPSAFE on i386 already. Mark them as that on amd64 and arm64 too to avoid locking Giant. Reviewed by: kib (mentor) Approved by: kib (mentor) Differential Revision:https://reviews.freebsd.org/D23039 Modified: head/sys/amd64/amd64/pmap.c head/sys/arm64/arm64/pmap.c Modified: head/sys/amd64/amd64/pmap.c == --- head/sys/amd64/amd64/pmap.c Mon Jan 6 09:51:54 2020(r356412) +++ head/sys/amd64/amd64/pmap.c Mon Jan 6 10:52:13 2020(r356413) @@ -4072,8 +4072,9 @@ kvm_size(SYSCTL_HANDLER_ARGS) return sysctl_handle_long(oidp, , 0, req); } -SYSCTL_PROC(_vm, OID_AUTO, kvm_size, CTLTYPE_LONG|CTLFLAG_RD, -0, 0, kvm_size, "LU", "Size of KVM"); +SYSCTL_PROC(_vm, OID_AUTO, kvm_size, CTLTYPE_LONG | CTLFLAG_RD | CTLFLAG_MPSAFE, +0, 0, kvm_size, "LU", +"Size of KVM"); static int kvm_free(SYSCTL_HANDLER_ARGS) @@ -4082,8 +4083,9 @@ kvm_free(SYSCTL_HANDLER_ARGS) return sysctl_handle_long(oidp, , 0, req); } -SYSCTL_PROC(_vm, OID_AUTO, kvm_free, CTLTYPE_LONG|CTLFLAG_RD, -0, 0, kvm_free, "LU", "Amount of KVM free"); +SYSCTL_PROC(_vm, OID_AUTO, kvm_free, CTLTYPE_LONG | CTLFLAG_RD | CTLFLAG_MPSAFE, +0, 0, kvm_free, "LU", +"Amount of KVM free"); /* * Allocate physical memory for the vm_page array and map it into KVA, Modified: head/sys/arm64/arm64/pmap.c == --- head/sys/arm64/arm64/pmap.c Mon Jan 6 09:51:54 2020(r356412) +++ head/sys/arm64/arm64/pmap.c Mon Jan 6 10:52:13 2020(r356413) @@ -1845,8 +1845,9 @@ kvm_size(SYSCTL_HANDLER_ARGS) return sysctl_handle_long(oidp, , 0, req); } -SYSCTL_PROC(_vm, OID_AUTO, kvm_size, CTLTYPE_LONG|CTLFLAG_RD, -0, 0, kvm_size, "LU", "Size of KVM"); +SYSCTL_PROC(_vm, OID_AUTO, kvm_size, CTLTYPE_LONG | CTLFLAG_RD | CTLFLAG_MPSAFE, +0, 0, kvm_size, "LU", +"Size of KVM"); static int kvm_free(SYSCTL_HANDLER_ARGS) @@ -1855,8 +1856,9 @@ kvm_free(SYSCTL_HANDLER_ARGS) return sysctl_handle_long(oidp, , 0, req); } -SYSCTL_PROC(_vm, OID_AUTO, kvm_free, CTLTYPE_LONG|CTLFLAG_RD, -0, 0, kvm_free, "LU", "Amount of KVM free"); +SYSCTL_PROC(_vm, OID_AUTO, kvm_free, CTLTYPE_LONG | CTLFLAG_RD | CTLFLAG_MPSAFE, +0, 0, kvm_free, "LU", +"Amount of KVM free"); /* * grow the number of kernel page table entries, if needed ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r356409 - in head/sys: dev/usb sys
Author: hselasky
Date: Mon Jan 6 09:49:20 2020
New Revision: 356409
URL: https://svnweb.freebsd.org/changeset/base/356409
Log:
Add own counter for cancelled USB transfers.
Do not count these as errors.
MFC after:1 week
Sponsored by: Mellanox Technologies
Modified:
head/sys/dev/usb/usb_device.h
head/sys/dev/usb/usb_transfer.c
head/sys/sys/param.h
Modified: head/sys/dev/usb/usb_device.h
==
--- head/sys/dev/usb/usb_device.h Mon Jan 6 09:40:59 2020
(r356408)
+++ head/sys/dev/usb/usb_device.h Mon Jan 6 09:49:20 2020
(r356409)
@@ -191,6 +191,7 @@ struct usb_device {
/* statistics */
struct usb_device_statistics stats_err;
struct usb_device_statistics stats_ok;
+ struct usb_device_statistics stats_cancelled;
/* generic clear stall message */
struct usb_udev_msg cs_msg[2];
Modified: head/sys/dev/usb/usb_transfer.c
==
--- head/sys/dev/usb/usb_transfer.c Mon Jan 6 09:40:59 2020
(r356408)
+++ head/sys/dev/usb/usb_transfer.c Mon Jan 6 09:49:20 2020
(r356409)
@@ -2593,7 +2593,10 @@ usbd_transfer_done(struct usb_xfer *xfer, usb_error_t
}
#endif
/* keep some statistics */
- if (xfer->error) {
+ if (xfer->error == USB_ERR_CANCELLED) {
+ info->udev->stats_cancelled.uds_requests
+ [xfer->endpoint->edesc->bmAttributes & UE_XFERTYPE]++;
+ } else if (xfer->error != USB_ERR_NORMAL_COMPLETION) {
info->udev->stats_err.uds_requests
[xfer->endpoint->edesc->bmAttributes & UE_XFERTYPE]++;
} else {
Modified: head/sys/sys/param.h
==
--- head/sys/sys/param.hMon Jan 6 09:40:59 2020(r356408)
+++ head/sys/sys/param.hMon Jan 6 09:49:20 2020(r356409)
@@ -60,7 +60,7 @@
* in the range 5 to 9.
*/
#undef __FreeBSD_version
-#define __FreeBSD_version 1300074 /* Master, propagated to newvers */
+#define __FreeBSD_version 1300075 /* Master, propagated to newvers */
/*
* __FreeBSD_kernel__ indicates that this system uses the kernel of FreeBSD,
___
svn-src-head@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
