git: 110113bc086f - main - sysctl(9): Enable vnet sysctl variables to be loader tunable
- Go to: [ bottom of page ] [ top of archives ] [ this month ]
Date: Sat, 09 Sep 2023 08:21:15 UTC
The branch main has been updated by zlei:
URL: https://cgit.FreeBSD.org/src/commit/?id=110113bc086f5df1a9b6547edb1ab0cec698c55c
commit 110113bc086f5df1a9b6547edb1ab0cec698c55c
Author: Zhenlei Huang <zlei@FreeBSD.org>
AuthorDate: 2023-09-09 08:06:23 +0000
Commit: Zhenlei Huang <zlei@FreeBSD.org>
CommitDate: 2023-09-09 08:06:23 +0000
sysctl(9): Enable vnet sysctl variables to be loader tunable
Complete phase two of 3da1cf1e88f8.
In 3da1cf1e88f8, the meaning of the flag CTLFLAG_TUN is extended to
automatically check if there is a kernel environment variable which
shall initialize the SYSCTL during early boot. It works for all SYSCTL
types both statically and dynamically created ones, except for the
SYSCTLs which belong to VNETs.
This change extends the meaning further, to allow it also works for
the SYSCTLs which belong to VNETs. A typical usage is
```
VNET_DEFINE_STATIC(int, foo) = 0;
SYSCTL_INT(_net, OID_AUTO, foo, CTLFLAG_RWTUN | CTLFLAG_VNET,
&VNET_NAME(foo), 0, "Description of the foo loader tunable");
```
Note that the implementation has a limitation. It behaves the same way
as that of non-vnet loader tunables. That is, after the kernel or modules
being initialized, any changes (e.g. via kenv) to kernel environment
variable will not affect the corresponding vnet variable of subsequently
created VNETs. To overcome it, we can use TUNABLE_XXX_FETCH to fetch
the kernel environment variable into those vnet variables during vnet
constructing.
This change will fix the following SYSCTLs those belong to VNETs and
have CTLFLAG_TUN flag:
```
net.add_addr_allfibs
net.bpf.optimize_writers
net.inet.tcp.fastopen.ccache_buckets
net.link.bridge.inherit_mac
net.link.bridge.ipfw_arp
net.link.bridge.log_stp
net.link.bridge.pfil_bridge
net.link.bridge.pfil_local_phys
net.link.bridge.pfil_member
net.link.bridge.pfil_onlyip
net.link.lagg.default_use_flowid
net.link.lagg.default_use_numa
net.link.lagg.default_flowid_shift
net.link.lagg.lacp.debug
net.link.lagg.lacp.default_strict_mode
```
Although the following vnet SYSCTLs have CTLFLAG_TUN flag, theirs
values are re-fetched via TUNABLE_XXX_FETCH, thus are not affected
by this change.
```
net.inet.ip.reass_hashsize
net.inet.tcp.hostcache.cachelimit
net.inet.tcp.hostcache.hashsize
net.inet.tcp.hostcache.bucketlimit
net.inet.tcp.syncache.bucketlimit
net.inet.tcp.syncache.cachelimit
net.inet.tcp.syncache.hashsize
net.key.spdcache.maxentries
net.key.spdcache.threshold
```
In memoriam: hselasky
Discussed with: hselasky, glebius
Fixes: 3da1cf1e88f8 Extend the meaning of the CTLFLAG_TUN flag ...
MFC after: 2 weeks
Relnotes: yes
Differential Revision: https://reviews.freebsd.org/D39638
---
sys/kern/kern_linker.c | 3 +++
sys/kern/kern_sysctl.c | 3 ---
sys/kern/link_elf.c | 20 ++++++++++++++++++++
sys/kern/link_elf_obj.c | 30 +++++++++++++++++++++++++++++-
sys/kern/linker_if.m | 9 +++++++++
5 files changed, 61 insertions(+), 4 deletions(-)
diff --git a/sys/kern/kern_linker.c b/sys/kern/kern_linker.c
index a1073512e856..2367bf35da76 100644
--- a/sys/kern/kern_linker.c
+++ b/sys/kern/kern_linker.c
@@ -471,6 +471,9 @@ linker_load_file(const char *filename, linker_file_t *result)
}
modules = !TAILQ_EMPTY(&lf->modules);
linker_file_register_sysctls(lf, false);
+#ifdef VIMAGE
+ LINKER_PROPAGATE_VNETS(lf);
+#endif
linker_file_sysinit(lf);
lf->flags |= LINKER_FILE_LINKED;
diff --git a/sys/kern/kern_sysctl.c b/sys/kern/kern_sysctl.c
index 814579a80f5a..a1d502d58bff 100644
--- a/sys/kern/kern_sysctl.c
+++ b/sys/kern/kern_sysctl.c
@@ -510,9 +510,6 @@ sysctl_register_oid(struct sysctl_oid *oidp)
RB_INSERT(sysctl_oid_list, parent, oidp);
if ((oidp->oid_kind & CTLTYPE) != CTLTYPE_NODE &&
-#ifdef VIMAGE
- (oidp->oid_kind & CTLFLAG_VNET) == 0 &&
-#endif
(oidp->oid_kind & CTLFLAG_TUN) != 0 &&
(oidp->oid_kind & CTLFLAG_NOFETCH) == 0) {
/* only fetch value once */
diff --git a/sys/kern/link_elf.c b/sys/kern/link_elf.c
index 5be2db9d32bf..3e9998f27baa 100644
--- a/sys/kern/link_elf.c
+++ b/sys/kern/link_elf.c
@@ -160,6 +160,9 @@ static int link_elf_each_function_nameval(linker_file_t,
static void link_elf_reloc_local(linker_file_t);
static long link_elf_symtab_get(linker_file_t, const Elf_Sym **);
static long link_elf_strtab_get(linker_file_t, caddr_t *);
+#ifdef VIMAGE
+static void link_elf_propagate_vnets(linker_file_t);
+#endif
static int elf_lookup(linker_file_t, Elf_Size, int, Elf_Addr *);
static kobj_method_t link_elf_methods[] = {
@@ -178,6 +181,9 @@ static kobj_method_t link_elf_methods[] = {
KOBJMETHOD(linker_ctf_get, link_elf_ctf_get),
KOBJMETHOD(linker_symtab_get, link_elf_symtab_get),
KOBJMETHOD(linker_strtab_get, link_elf_strtab_get),
+#ifdef VIMAGE
+ KOBJMETHOD(linker_propagate_vnets, link_elf_propagate_vnets),
+#endif
KOBJMETHOD_END
};
@@ -1923,6 +1929,20 @@ link_elf_strtab_get(linker_file_t lf, caddr_t *strtab)
return (ef->ddbstrcnt);
}
+#ifdef VIMAGE
+static void
+link_elf_propagate_vnets(linker_file_t lf)
+{
+ elf_file_t ef = (elf_file_t)lf;
+ int size;
+
+ if (ef->vnet_base != 0) {
+ size = (uintptr_t)ef->vnet_stop - (uintptr_t)ef->vnet_start;
+ vnet_data_copy((void *)ef->vnet_base, size);
+ }
+}
+#endif
+
#if defined(__i386__) || defined(__amd64__) || defined(__aarch64__) || defined(__powerpc__)
/*
* Use this lookup routine when performing relocations early during boot.
diff --git a/sys/kern/link_elf_obj.c b/sys/kern/link_elf_obj.c
index b853ca284f7d..768808d2102e 100644
--- a/sys/kern/link_elf_obj.c
+++ b/sys/kern/link_elf_obj.c
@@ -150,6 +150,9 @@ static int link_elf_each_function_nameval(linker_file_t,
static int link_elf_reloc_local(linker_file_t, bool);
static long link_elf_symtab_get(linker_file_t, const Elf_Sym **);
static long link_elf_strtab_get(linker_file_t, caddr_t *);
+#ifdef VIMAGE
+static void link_elf_propagate_vnets(linker_file_t);
+#endif
static int elf_obj_lookup(linker_file_t lf, Elf_Size symidx, int deps,
Elf_Addr *);
@@ -170,6 +173,9 @@ static kobj_method_t link_elf_methods[] = {
KOBJMETHOD(linker_ctf_get, link_elf_ctf_get),
KOBJMETHOD(linker_symtab_get, link_elf_symtab_get),
KOBJMETHOD(linker_strtab_get, link_elf_strtab_get),
+#ifdef VIMAGE
+ KOBJMETHOD(linker_propagate_vnets, link_elf_propagate_vnets),
+#endif
KOBJMETHOD_END
};
@@ -1848,7 +1854,7 @@ link_elf_symtab_get(linker_file_t lf, const Elf_Sym **symtab)
return (0);
return (ef->ddbsymcnt);
}
-
+
static long
link_elf_strtab_get(linker_file_t lf, caddr_t *strtab)
{
@@ -1859,3 +1865,25 @@ link_elf_strtab_get(linker_file_t lf, caddr_t *strtab)
return (0);
return (ef->ddbstrcnt);
}
+
+#ifdef VIMAGE
+static void
+link_elf_propagate_vnets(linker_file_t lf)
+{
+ elf_file_t ef = (elf_file_t) lf;
+
+ if (ef->progtab) {
+ for (int i = 0; i < ef->nprogtab; i++) {
+ if (ef->progtab[i].size == 0)
+ continue;
+ if (ef->progtab[i].name == NULL)
+ continue;
+ if (strcmp(ef->progtab[i].name, VNET_SETNAME) == 0) {
+ vnet_data_copy(ef->progtab[i].addr,
+ ef->progtab[i].size);
+ break;
+ }
+ }
+ }
+}
+#endif
diff --git a/sys/kern/linker_if.m b/sys/kern/linker_if.m
index 0722390d4e20..a50ed1ea84a3 100644
--- a/sys/kern/linker_if.m
+++ b/sys/kern/linker_if.m
@@ -154,3 +154,12 @@ STATICMETHOD int link_preload {
METHOD int link_preload_finish {
linker_file_t file;
};
+
+#ifdef VIMAGE
+#
+# Propagate system tunable values to all vnets.
+#
+METHOD void propagate_vnets {
+ linker_file_t file;
+};
+#endif