Re: svn commit: r256850 - in head: bin/sh etc lib/libc/sys lib/libutil usr.bin/limits usr.bin/procstat
On Sat, Nov 02, 2013 at 07:43:52AM +0100, Stefan Neudorf wrote:
> Konstantin Belousov writes:
>
> > On Wed, Oct 30, 2013 at 10:20:52PM +0100, Jilles Tjoelker wrote:
> >
> >> On Wed, Oct 30, 2013 at 02:01:57PM +0100, Stefan Neudorf wrote:
> >> > Konstantin Belousov writes:
> >> > > Modified: head/usr.bin/limits/limits.c
> >> > > ==
> >> > > --- head/usr.bin/limits/limits.c Mon Oct 21 16:44:53 2013
> >> > > (r256849)
> >> > > +++ head/usr.bin/limits/limits.c Mon Oct 21 16:46:12 2013
> >> > > (r256850)
> >>
> >> > This adds limits -e support only for sh, csh, tcsh. What about other
> >> > shells that already support sbsize and swapuse extensions?
> >>
> >> This can be done later; the kqueue rlimit need not be gated on it.
> >
> > I definitely do not want to go over the {pd,}ksh, zsh, bash, rc and
> > whatever else. Somebody who cares should contact the projects and
> > propose the changes.
>
> Ah, you didn't catch my hint. There's no kqueues rlimit in contrib/tcsh.
> I've submitted one upstream but it'd take some time before the next
> release appears and we can update.
>
> http://mx.gw.com/pipermail/tcsh-bugs/2013-October/000844.html
I can just apply the upstream patch to our contrib/tcsh, without
waiting for import of new version. Can you point me to the way to get
the commit which was referenced in the mail above ?
>
> Same for zsh which also has csh-like rlimits.
>
> http://sourceforge.net/p/zsh/code/ci/790a4fe
pgpx4DJ072aq5.pgp
Description: PGP signature
svn commit: r257541 - head/sys/dev/e1000
Author: kib
Date: Sat Nov 2 09:16:11 2013
New Revision: 257541
URL: http://svnweb.freebsd.org/changeset/base/257541
Log:
Fix several issues with the busdma(9) KPI use in the e1000 drivers.
The problems do not affect bouncing busdma in a visible way, but are
critical for the dmar backend.
- The bus_dmamap_create(9) is not documented to take BUS_DMA_NOWAIT flag.
- Unload descriptor map after receive.
- Do not reset descriptor map to NULL, bus_dmamap_load(9) requires
valid map, and also this leaks the map.
Reported and tested by: pho
Approved by: jfv
Sponsored by: The FreeBSD Foundation
MFC after:2 weeks
Modified:
head/sys/dev/e1000/if_em.c
head/sys/dev/e1000/if_igb.c
head/sys/dev/e1000/if_lem.c
Modified: head/sys/dev/e1000/if_em.c
==
--- head/sys/dev/e1000/if_em.c Sat Nov 2 02:38:32 2013(r257540)
+++ head/sys/dev/e1000/if_em.c Sat Nov 2 09:16:11 2013(r257541)
@@ -4059,8 +4059,7 @@ em_allocate_receive_buffers(struct rx_ri
rxbuf = rxr->rx_buffers;
for (int i = 0; i < adapter->num_rx_desc; i++, rxbuf++) {
rxbuf = &rxr->rx_buffers[i];
- error = bus_dmamap_create(rxr->rxtag, BUS_DMA_NOWAIT,
- &rxbuf->map);
+ error = bus_dmamap_create(rxr->rxtag, 0, &rxbuf->map);
if (error) {
device_printf(dev, "%s: bus_dmamap_create failed: %d\n",
__func__, error);
@@ -4467,6 +4466,7 @@ em_rxeof(struct rx_ring *rxr, int count,
em_rx_discard(rxr, i);
goto next_desc;
}
+ bus_dmamap_unload(rxr->rxtag, rxr->rx_buffers[i].map);
/* Assign correct length to the current fragment */
mp = rxr->rx_buffers[i].m_head;
@@ -4553,6 +4553,8 @@ em_rx_discard(struct rx_ring *rxr, int i
struct em_buffer*rbuf;
rbuf = &rxr->rx_buffers[i];
+ bus_dmamap_unload(rxr->rxtag, rbuf->map);
+
/* Free any previous pieces */
if (rxr->fmp != NULL) {
rxr->fmp->m_flags |= M_PKTHDR;
Modified: head/sys/dev/e1000/if_igb.c
==
--- head/sys/dev/e1000/if_igb.c Sat Nov 2 02:38:32 2013(r257540)
+++ head/sys/dev/e1000/if_igb.c Sat Nov 2 09:16:11 2013(r257541)
@@ -3996,7 +3996,6 @@ igb_txeof(struct tx_ring *txr)
buf->map);
m_freem(buf->m_head);
buf->m_head = NULL;
- buf->map = NULL;
}
buf->eop = NULL;
++txr->tx_avail;
@@ -4022,7 +4021,6 @@ igb_txeof(struct tx_ring *txr)
buf->map);
m_freem(buf->m_head);
buf->m_head = NULL;
- buf->map = NULL;
}
++txr->tx_avail;
buf->eop = NULL;
@@ -4230,15 +4228,13 @@ igb_allocate_receive_buffers(struct rx_r
for (i = 0; i < adapter->num_rx_desc; i++) {
rxbuf = &rxr->rx_buffers[i];
- error = bus_dmamap_create(rxr->htag,
- BUS_DMA_NOWAIT, &rxbuf->hmap);
+ error = bus_dmamap_create(rxr->htag, 0, &rxbuf->hmap);
if (error) {
device_printf(dev,
"Unable to create RX head DMA maps\n");
goto fail;
}
- error = bus_dmamap_create(rxr->ptag,
- BUS_DMA_NOWAIT, &rxbuf->pmap);
+ error = bus_dmamap_create(rxr->ptag, 0, &rxbuf->pmap);
if (error) {
device_printf(dev,
"Unable to create RX packet DMA maps\n");
@@ -4758,11 +4754,13 @@ igb_rx_discard(struct rx_ring *rxr, int
if (rbuf->m_head) {
m_free(rbuf->m_head);
rbuf->m_head = NULL;
+ bus_dmamap_unload(rxr->htag, rbuf->hmap);
}
if (rbuf->m_pack) {
m_free(rbuf->m_pack);
rbuf->m_pack = NULL;
+ bus_dmamap_unload(rxr->ptag, rbuf->pmap);
}
return;
@@ -4887,6 +4885,7 @@ igb_rxeof(struct igb_queue *que, int cou
** case only the first header is valid.
*/
if (rxr->hdr_split && rxr->fmp == NULL) {
+ bus_dmamap_unload(rxr->htag, rxbuf->hmap);
hlen = (hdr & E1000_RXDADV_HDRBUFLEN_MASK) >>
E1000_RXDADV_HDRBUFLEN_SHIFT;
if (hlen > IGB_HDR_BUF)
@@ -4919,6 +4918,7 @@ igb_rxeof(struct igb_queue *que, int cou
/* clear buf info for refresh */
Re: svn commit: r257535 - head/sys/netgraph
On 2 Nov 2013, at 05:53, Adrian Chadd wrote: > Hm! A good question! > > On 1 November 2013 22:22, Bruce Evans wrote: > >>>if (harvest.point_to_point) >>> - random_harvest(&(m->m_data), 12, 2, RANDOM_NET_NG); >> >> >> '&(m->m_data)' is not just a pair of style bugs. It gives address of >> the pointer (somewhere in the mbuf header), not the address of pointed- >> to data, so the randomness was almost null. The style bugs are >> excessive parentheses and chumminess with the implementation (non-use >> of the accessor function mtod()). >> >> >>> + random_harvest(mtod(m, const void *), 12, 2, >>> RANDOM_NET_NG); >> >> >> Presumably you really do want to harvest the pointed-to data and there >> are at least 12 bytes of it, so the semantic fix isn't backwards or a >> buffer overrun. >> > > > Mark - did you initially mean the address of the mbuf m_data pointer, > or the data payload itself? As Bruce says - the address of payload data itself. We don’t have 12-byte pointers in FreeBSD. :-) M -- Mark R V Murray signature.asc Description: Message signed with OpenPGP using GPGMail
Re: svn commit: r257535 - head/sys/netgraph
On 2 Nov 2013, at 09:32, Mark R V Murray wrote: >> Mark - did you initially mean the address of the mbuf m_data pointer, >> or the data payload itself? > > As Bruce says - the address of payload data itself. We don’t have 12-byte > pointers in FreeBSD. :-) Cancel that. The address passed must be the address of the m_data field in the mbuf structure. The harvested data is 12 bytes from that address forward, so not the data pointed to by that m_data pointer but the pointer value itself and some following junk too. M -- Mark R V Murray signature.asc Description: Message signed with OpenPGP using GPGMail
svn commit: r257542 - head/sys/ofed/drivers/net/mlx4
Author: alfred
Date: Sat Nov 2 10:49:47 2013
New Revision: 257542
URL: http://svnweb.freebsd.org/changeset/base/257542
Log:
Fix API mismatch exposed by lagg.
When destroying a lagg the driver tries to restore the old mac and
fails due to API mismatch
Modified:
head/sys/ofed/drivers/net/mlx4/en_netdev.c
Modified: head/sys/ofed/drivers/net/mlx4/en_netdev.c
==
--- head/sys/ofed/drivers/net/mlx4/en_netdev.c Sat Nov 2 09:16:11 2013
(r257541)
+++ head/sys/ofed/drivers/net/mlx4/en_netdev.c Sat Nov 2 10:49:47 2013
(r257542)
@@ -633,8 +633,8 @@ int mlx4_en_start_port(struct net_device
en_dbg(DRV, priv, "Setting mac for port %d\n", priv->port);
err = mlx4_register_mac(mdev->dev, priv->port,
mlx4_en_mac_to_u64(IF_LLADDR(dev)));
- if (err) {
- en_err(priv, "Failed setting port mac\n");
+ if (err < 0) {
+ en_err(priv, "Failed setting port mac err=%d\n", err);
goto tx_err;
}
mdev->mac_removed[priv->port] = 0;
___
svn-src-head@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r257543 - head/sys/dev/usb/wlan
Author: alfred Date: Sat Nov 2 11:37:16 2013 New Revision: 257543 URL: http://svnweb.freebsd.org/changeset/base/257543 Log: Add device ID for 'Sanoxy 802.11N' usb Modified: head/sys/dev/usb/wlan/if_urtwn.c Modified: head/sys/dev/usb/wlan/if_urtwn.c == --- head/sys/dev/usb/wlan/if_urtwn.cSat Nov 2 10:49:47 2013 (r257542) +++ head/sys/dev/usb/wlan/if_urtwn.cSat Nov 2 11:37:16 2013 (r257543) @@ -139,6 +139,7 @@ static const STRUCT_USB_HOST_ID urtwn_de URTWN_DEV(REALTEK, RTL8191CU), URTWN_DEV(REALTEK, RTL8192CE), URTWN_DEV(REALTEK, RTL8192CU), + URTWN_DEV(REALTEK, RTL8188CU_0), URTWN_DEV(SITECOMEU,RTL8188CU_1), URTWN_DEV(SITECOMEU,RTL8188CU_2), URTWN_DEV(SITECOMEU,RTL8192CU), ___ 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: r257535 - head/sys/netgraph
Okay. I'll go and re-fix things. Thanks, -a On 2 November 2013 02:44, Mark R V Murray wrote: > > On 2 Nov 2013, at 09:32, Mark R V Murray wrote: >>> Mark - did you initially mean the address of the mbuf m_data pointer, >>> or the data payload itself? >> >> As Bruce says - the address of payload data itself. We don’t have 12-byte >> pointers in FreeBSD. :-) > > Cancel that. > > The address passed must be the address of the m_data field in the mbuf > structure. The harvested data is 12 bytes from that address forward, so not > the data pointed to by that m_data pointer but the pointer value itself and > some following junk too. > > M > -- > Mark R V Murray > ___ 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: r257548 - in head/sys: net netgraph
Author: adrian Date: Sat Nov 2 15:13:02 2013 New Revision: 257548 URL: http://svnweb.freebsd.org/changeset/base/257548 Log: Restore the entropy gathering from the m_data pointer value, not the m_data payload. After talking with markm/bde, this is what markm actually intended. Modified: head/sys/net/if_ethersubr.c head/sys/net/if_tun.c head/sys/netgraph/ng_iface.c Modified: head/sys/net/if_ethersubr.c == --- head/sys/net/if_ethersubr.c Sat Nov 2 14:20:29 2013(r257547) +++ head/sys/net/if_ethersubr.c Sat Nov 2 15:13:02 2013(r257548) @@ -642,7 +642,7 @@ ether_input_internal(struct ifnet *ifp, } if (harvest.ethernet) - random_harvest(mtod(m, const void *), 12, 2, RANDOM_NET_ETHER); + random_harvest(&(m->m_data), 12, 2, RANDOM_NET_ETHER); ether_demux(ifp, m); CURVNET_RESTORE(); Modified: head/sys/net/if_tun.c == --- head/sys/net/if_tun.c Sat Nov 2 14:20:29 2013(r257547) +++ head/sys/net/if_tun.c Sat Nov 2 15:13:02 2013(r257548) @@ -919,7 +919,7 @@ tunwrite(struct cdev *dev, struct uio *u return (EAFNOSUPPORT); } if (harvest.point_to_point) - random_harvest(mtod(m, const void *), 12, 2, RANDOM_NET_TUN); + random_harvest(&(m->m_data), 12, 2, RANDOM_NET_TUN); ifp->if_ibytes += m->m_pkthdr.len; ifp->if_ipackets++; CURVNET_SET(ifp->if_vnet); Modified: head/sys/netgraph/ng_iface.c == --- head/sys/netgraph/ng_iface.cSat Nov 2 14:20:29 2013 (r257547) +++ head/sys/netgraph/ng_iface.cSat Nov 2 15:13:02 2013 (r257548) @@ -776,7 +776,7 @@ ng_iface_rcvdata(hook_p hook, item_p ite return (EAFNOSUPPORT); } if (harvest.point_to_point) - random_harvest(mtod(m, const void *), 12, 2, RANDOM_NET_NG); + random_harvest(&(m->m_data), 12, 2, RANDOM_NET_NG); M_SETFIB(m, ifp->if_fib); netisr_dispatch(isr, m); return (0); ___ svn-src-head@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r257549 - head/sys/arm/include
Author: alc Date: Sat Nov 2 17:08:20 2013 New Revision: 257549 URL: http://svnweb.freebsd.org/changeset/base/257549 Log: Don't create a distinct free page pool for segregating allocations that are accessed through the direct map unless the kernel configuration actually includes a direct map. Only a few configurations do, and for the rest the unnecessary free page pool is a small pessimization. Tested by:zbb MFC after:6 weeks Modified: head/sys/arm/include/vmparam.h Modified: head/sys/arm/include/vmparam.h == --- head/sys/arm/include/vmparam.h Sat Nov 2 15:13:02 2013 (r257548) +++ head/sys/arm/include/vmparam.h Sat Nov 2 17:08:20 2013 (r257549) @@ -82,15 +82,21 @@ #defineVM_PHYSSEG_DENSE /* - * Create three free page pools: VM_FREEPOOL_DEFAULT is the default pool - * from which physical pages are allocated and VM_FREEPOOL_DIRECT is - * the pool from which physical pages for small UMA objects are - * allocated. + * Create two or three free page pools depending on the existence of a direct + * map: VM_FREEPOOL_DEFAULT is the default pool from which physical pages are + * allocated, and VM_FREEPOOL_DIRECT is the pool from which physical pages for + * small UMA objects are allocated. */ +#ifdef ARM_USE_SMALL_ALLOC #defineVM_NFREEPOOL3 #defineVM_FREEPOOL_CACHE 2 -#defineVM_FREEPOOL_DEFAULT 0 #defineVM_FREEPOOL_DIRECT 1 +#else +#defineVM_NFREEPOOL2 +#defineVM_FREEPOOL_CACHE 1 +#defineVM_FREEPOOL_DIRECT 0 +#endif +#defineVM_FREEPOOL_DEFAULT 0 /* * we support 2 free lists: ___ 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: r257550 - head/sys/dev/netmap
Author: luigi
Date: Sat Nov 2 18:03:21 2013
New Revision: 257550
URL: http://svnweb.freebsd.org/changeset/base/257550
Log:
circumvent a couple of warnings:
- on line 2550 intentionally overriding a const qualifier
- on line 3219 intentionally converting uint64_t to a pointer
Modified:
head/sys/dev/netmap/netmap.c
Modified: head/sys/dev/netmap/netmap.c
==
--- head/sys/dev/netmap/netmap.cSat Nov 2 17:08:20 2013
(r257549)
+++ head/sys/dev/netmap/netmap.cSat Nov 2 18:03:21 2013
(r257550)
@@ -2547,7 +2547,7 @@ netmap_ioctl(struct cdev *dev, u_long cm
}
if (memflags & NETMAP_MEM_PRIVATE) {
nmr->nr_ringid |= NETMAP_PRIV_MEM;
- *(uint32_t *)&nifp->ni_flags |= NI_PRIV_MEM;
+ *(uint32_t *)(uintptr_t)&nifp->ni_flags |=
NI_PRIV_MEM;
}
nmr->nr_offset = netmap_mem_if_offset(na->nm_mem, nifp);
} while (0);
@@ -3216,7 +3216,7 @@ nm_bdg_preflush(struct netmap_adapter *n
/* this slot goes into a list so initialize the link field */
ft[ft_i].ft_next = NM_FT_NULL;
buf = ft[ft_i].ft_buf = (slot->flags & NS_INDIRECT) ?
- (void *)slot->ptr : BDG_NMB(na->nm_mem, slot);
+ (void *)(uintptr_t)slot->ptr : BDG_NMB(na->nm_mem,
slot);
prefetch(buf);
++ft_i;
if (slot->flags & NS_MOREFRAG) {
___
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: r257555 - in head/sys: netinet netinet6
Author: tuexen
Date: Sat Nov 2 20:12:19 2013
New Revision: 257555
URL: http://svnweb.freebsd.org/changeset/base/257555
Log:
Changes from upstream to improve compilation when INET or INET6
or none of them is defined.
MFC after: 3 days
Modified:
head/sys/netinet/sctp_asconf.c
head/sys/netinet/sctp_output.c
head/sys/netinet/sctp_pcb.c
head/sys/netinet/sctp_usrreq.c
head/sys/netinet/sctputil.c
head/sys/netinet6/sctp6_usrreq.c
Modified: head/sys/netinet/sctp_asconf.c
==
--- head/sys/netinet/sctp_asconf.c Sat Nov 2 20:06:17 2013
(r257554)
+++ head/sys/netinet/sctp_asconf.c Sat Nov 2 20:12:19 2013
(r257555)
@@ -150,7 +150,12 @@ sctp_process_asconf_add_ip(struct sockad
struct mbuf *m_reply = NULL;
struct sockaddr_storage sa_store;
struct sctp_paramhdr *ph;
- uint16_t param_type, param_length, aparam_length;
+ uint16_t param_type, aparam_length;
+
+#if defined(INET) || defined(INET6)
+ uint16_t param_length;
+
+#endif
struct sockaddr *sa;
int zero_address = 0;
int bad_address = 0;
@@ -169,8 +174,9 @@ sctp_process_asconf_add_ip(struct sockad
aparam_length = ntohs(aph->ph.param_length);
ph = (struct sctp_paramhdr *)(aph + 1);
param_type = ntohs(ph->param_type);
+#if defined(INET) || defined(INET6)
param_length = ntohs(ph->param_length);
-
+#endif
sa = (struct sockaddr *)&sa_store;
switch (param_type) {
#ifdef INET
@@ -298,7 +304,12 @@ sctp_process_asconf_delete_ip(struct soc
struct mbuf *m_reply = NULL;
struct sockaddr_storage sa_store;
struct sctp_paramhdr *ph;
- uint16_t param_type, param_length, aparam_length;
+ uint16_t param_type, aparam_length;
+
+#if defined(INET) || defined(INET6)
+ uint16_t param_length;
+
+#endif
struct sockaddr *sa;
int zero_address = 0;
int result;
@@ -317,8 +328,9 @@ sctp_process_asconf_delete_ip(struct soc
aparam_length = ntohs(aph->ph.param_length);
ph = (struct sctp_paramhdr *)(aph + 1);
param_type = ntohs(ph->param_type);
+#if defined(INET) || defined(INET6)
param_length = ntohs(ph->param_length);
-
+#endif
sa = (struct sockaddr *)&sa_store;
switch (param_type) {
#ifdef INET
@@ -427,7 +439,12 @@ sctp_process_asconf_set_primary(struct s
struct mbuf *m_reply = NULL;
struct sockaddr_storage sa_store;
struct sctp_paramhdr *ph;
- uint16_t param_type, param_length, aparam_length;
+ uint16_t param_type, aparam_length;
+
+#if defined(INET) || defined(INET6)
+ uint16_t param_length;
+
+#endif
struct sockaddr *sa;
int zero_address = 0;
@@ -445,8 +462,9 @@ sctp_process_asconf_set_primary(struct s
aparam_length = ntohs(aph->ph.param_length);
ph = (struct sctp_paramhdr *)(aph + 1);
param_type = ntohs(ph->param_type);
+#if defined(INET) || defined(INET6)
param_length = ntohs(ph->param_length);
-
+#endif
sa = (struct sockaddr *)&sa_store;
switch (param_type) {
#ifdef INET
@@ -860,10 +878,12 @@ sctp_asconf_addr_match(struct sctp_ascon
static uint32_t
sctp_addr_match(struct sctp_paramhdr *ph, struct sockaddr *sa)
{
+#if defined(INET) || defined(INET6)
uint16_t param_type, param_length;
param_type = ntohs(ph->param_type);
param_length = ntohs(ph->param_length);
+#endif
switch (sa->sa_family) {
#ifdef INET6
case AF_INET6:
@@ -874,7 +894,7 @@ sctp_addr_match(struct sctp_paramhdr *ph
v6addr = (struct sctp_ipv6addr_param *)ph;
if ((param_type == SCTP_IPV6_ADDRESS) &&
- param_length == sizeof(struct sctp_ipv6addr_param)
&&
+ (param_length == sizeof(struct
sctp_ipv6addr_param)) &&
(memcmp(&v6addr->addr, &sin6->sin6_addr,
sizeof(struct in6_addr)) == 0)) {
return (1);
@@ -890,7 +910,7 @@ sctp_addr_match(struct sctp_paramhdr *ph
v4addr = (struct sctp_ipv4addr_param *)ph;
if ((param_type == SCTP_IPV4_ADDRESS) &&
- param_length == sizeof(struct sctp_ipv4addr_param)
&&
+ (param_length == sizeof(struct
sctp_ipv4addr_param)) &&
(memcmp(&v4addr->addr, &sin->sin_addr,
sizeof(struct in_addr)) == 0)) {
return (1);
Modified: head/sys/netinet/sctp_output.c
==
--- head/sys/netinet/sctp_output.c Sat Nov 2 20:06:17 2013
(r257554)
+++ head/sys/netinet/sctp_output.c Sat Nov 2 20:12:19 2013
(r257555)
@@ -1937,10 +1937,13 @@ sctp_is_address_in_
svn commit: r257556 - head/sys/dev/uart
Author: ian
Date: Sat Nov 2 20:14:39 2013
New Revision: 257556
URL: http://svnweb.freebsd.org/changeset/base/257556
Log:
Arrange for uart_cpu_fdt's probe() routine to use the same table of compat
strings as uart_bus_fdt's probe().
The bus code uses ofw_bus_search_compatible() and that's not an option in
cpu (console) code -- it runs way before the ofw routines are usable. So
the console probe has its own loop to search the table, but now at least
there's only one table to be maintained when new devices are added.
Modified:
head/sys/dev/uart/uart.h
head/sys/dev/uart/uart_bus_fdt.c
head/sys/dev/uart/uart_cpu_fdt.c
Modified: head/sys/dev/uart/uart.h
==
--- head/sys/dev/uart/uart.hSat Nov 2 20:12:19 2013(r257555)
+++ head/sys/dev/uart/uart.hSat Nov 2 20:14:39 2013(r257556)
@@ -76,6 +76,11 @@ extern struct uart_class uart_pl011_clas
extern struct uart_class uart_cdnc_class __attribute__((weak));
extern struct uart_class uart_ti8250_class __attribute__((weak));
+#ifdef FDT
+struct ofw_compat_data;
+extern const struct ofw_compat_data *uart_fdt_compat_data;
+#endif
+
#ifdef PC98
struct uart_class *uart_pc98_getdev(u_long port);
#endif
Modified: head/sys/dev/uart/uart_bus_fdt.c
==
--- head/sys/dev/uart/uart_bus_fdt.cSat Nov 2 20:12:19 2013
(r257555)
+++ head/sys/dev/uart/uart_bus_fdt.cSat Nov 2 20:14:39 2013
(r257556)
@@ -30,6 +30,8 @@
#include
__FBSDID("$FreeBSD$");
+#include "opt_platform.h"
+
#include
#include
#include
@@ -85,6 +87,9 @@ static struct ofw_compat_data compat_dat
{NULL, (uintptr_t)NULL},
};
+/* Export the compat_data table for use by the uart_cpu_fdt.c probe routine. */
+const struct ofw_compat_data *uart_fdt_compat_data = compat_data;
+
static int
uart_fdt_get_clock(phandle_t node, pcell_t *cell)
{
Modified: head/sys/dev/uart/uart_cpu_fdt.c
==
--- head/sys/dev/uart/uart_cpu_fdt.cSat Nov 2 20:12:19 2013
(r257555)
+++ head/sys/dev/uart/uart_cpu_fdt.cSat Nov 2 20:14:39 2013
(r257556)
@@ -30,6 +30,8 @@
#include
__FBSDID("$FreeBSD$");
+#include "opt_platform.h"
+
#include
#include
#include
@@ -118,6 +120,7 @@ uart_cpu_getdev(int devtype, struct uart
const char *propnames[] = {"stdout-path", "linux,stdout-path", "stdout",
"stdin-path", "stdin", NULL};
const char **name;
+ const struct ofw_compat_data *cd;
struct uart_class *class;
phandle_t node, chosen;
pcell_t shift, br, rclk;
@@ -166,22 +169,13 @@ uart_cpu_getdev(int devtype, struct uart
/*
* Finalize configuration.
*/
- if (fdt_is_compatible(node, "fsl,imx-uart"))
- class = &uart_imx_class;
- else if (fdt_is_compatible(node, "quicc"))
- class = &uart_quicc_class;
- else if (fdt_is_compatible(node, "lpc"))
- class = &uart_lpc_class;
- else if (fdt_is_compatible(node, "arm,pl011"))
- class = &uart_pl011_class;
- else if (fdt_is_compatible(node, "exynos"))
- class = &uart_s3c2410_class;
- else if (fdt_is_compatible(node, "cadence,uart"))
- class = &uart_cdnc_class;
- else if (fdt_is_compatible(node, "ti,ns16550"))
- class = &uart_ti8250_class;
- else if (fdt_is_compatible(node, "ns16550"))
- class = &uart_ns8250_class;
+ for (cd = uart_fdt_compat_data; cd->ocd_str != NULL; ++cd) {
+ if (fdt_is_compatible(node, cd->ocd_str))
+ break;
+ }
+ if (cd->ocd_str == NULL)
+ return (ENXIO);
+ class = (struct uart_class *)cd->ocd_data;
di->bas.chan = 0;
di->bas.regshft = (u_int)shift;
___
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: r257557 - head/sys/arm/freescale/imx
Author: ian Date: Sat Nov 2 21:07:39 2013 New Revision: 257557 URL: http://svnweb.freebsd.org/changeset/base/257557 Log: Add a missing register definition. Modified: head/sys/arm/freescale/imx/imx6_anatopreg.h Modified: head/sys/arm/freescale/imx/imx6_anatopreg.h == --- head/sys/arm/freescale/imx/imx6_anatopreg.h Sat Nov 2 20:14:39 2013 (r257556) +++ head/sys/arm/freescale/imx/imx6_anatopreg.h Sat Nov 2 21:07:39 2013 (r257557) @@ -81,6 +81,17 @@ #defineIMX6_ANALOG_CCM_PFD_528_SET 0x104 #defineIMX6_ANALOG_CCM_PFD_528_CLR 0x108 #defineIMX6_ANALOG_CCM_PFD_528_TOG 0x10C +#defineIMX6_ANALOG_PMU_REG_CORE0x140 +#define IMX6_ANALOG_PMU_REG2_TARG_SHIFT 18 +#define IMX6_ANALOG_PMU_REG2_TARG_MASK \ +(0x1f << IMX6_ANALOG_PMU_REG2_TARG_SHIFT) +#define IMX6_ANALOG_PMU_REG1_TARG_SHIFT 9 +#define IMX6_ANALOG_PMU_REG1_TARG_MASK \ +(0x1f << IMX6_ANALOG_PMU_REG1_TARG_SHIFT) +#define IMX6_ANALOG_PMU_REG0_TARG_SHIFT 0 +#define IMX6_ANALOG_PMU_REG0_TARG_MASK \ +(0x1f << IMX6_ANALOG_PMU_REG0_TARG_SHIFT) + #defineIMX6_ANALOG_CCM_MISC0 0x150 #defineIMX6_ANALOG_CCM_MISC0_SET 0x154 #defineIMX6_ANALOG_CCM_MISC0_CLR 0x158 ___ 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: r256849 - in head/sys: kern sys
On Mon, Oct 21, 2013 at 04:44:53PM +, Konstantin Belousov wrote:
> Author: kib
> Date: Mon Oct 21 16:44:53 2013
> New Revision: 256849
> URL: http://svnweb.freebsd.org/changeset/base/256849
> Log:
> Add a resource limit for the total number of kqueues available to the
> user. Kqueue now saves the ucred of the allocating thread, to
> correctly decrement the counter on close.
> Under some specific and not real-world use scenario for kqueue, it is
> possible for the kqueues to consume memory proportional to the square
> of the number of the filedescriptors available to the process. Limit
> allows administrator to prevent the abuse.
> This is kernel-mode side of the change, with the user-mode enabling
> commit following.
> Reported and tested by: pho
> Discussed with: jmg
> Sponsored by: The FreeBSD Foundation
> MFC after: 2 weeks
> [snip]
> + if (!chgkqcnt(cred->cr_ruidinfo, 1, lim_cur(td->td_proc,
> + RLIMIT_KQUEUES))) {
> + PROC_UNLOCK(p);
> + crfree(cred);
> + return (EMFILE);
> + }
Perhaps this error should be [ENOMEM] instead of [EMFILE] as it is
unrelated to the number of file descriptors in the process. This error
is already listed in the kqueue(2) man page but appears to be not
generated. The rlimit can then be mentioned in the man page.
--
Jilles Tjoelker
___
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: r256849 - in head/sys: kern sys
On Sat, Nov 02, 2013 at 10:17:27PM +0100, Jilles Tjoelker wrote:
> On Mon, Oct 21, 2013 at 04:44:53PM +, Konstantin Belousov wrote:
> > Author: kib
> > Date: Mon Oct 21 16:44:53 2013
> > New Revision: 256849
> > URL: http://svnweb.freebsd.org/changeset/base/256849
>
> > Log:
> > Add a resource limit for the total number of kqueues available to the
> > user. Kqueue now saves the ucred of the allocating thread, to
> > correctly decrement the counter on close.
>
> > Under some specific and not real-world use scenario for kqueue, it is
> > possible for the kqueues to consume memory proportional to the square
> > of the number of the filedescriptors available to the process. Limit
> > allows administrator to prevent the abuse.
>
> > This is kernel-mode side of the change, with the user-mode enabling
> > commit following.
>
> > Reported and tested by: pho
> > Discussed with: jmg
> > Sponsored by: The FreeBSD Foundation
> > MFC after:2 weeks
>
> > [snip]
> > + if (!chgkqcnt(cred->cr_ruidinfo, 1, lim_cur(td->td_proc,
> > + RLIMIT_KQUEUES))) {
> > + PROC_UNLOCK(p);
> > + crfree(cred);
> > + return (EMFILE);
> > + }
>
> Perhaps this error should be [ENOMEM] instead of [EMFILE] as it is
> unrelated to the number of file descriptors in the process. This error
> is already listed in the kqueue(2) man page but appears to be not
> generated. The rlimit can then be mentioned in the man page.
This is reasonable. Would you make the change yourself, please ?
pgpFVUtUxKY6f.pgp
Description: PGP signature
Re: svn commit: r257535 - head/sys/netgraph
On Sat, 2 Nov 2013, Mark R V Murray wrote: On 2 Nov 2013, at 09:32, Mark R V Murray wrote: Mark - did you initially mean the address of the mbuf m_data pointer, or the data payload itself? As Bruce says - the address of payload data itself. We don?t have 12-byte pointers in FreeBSD. :-) Cancel that. The address passed must be the address of the m_data field in the mbuf structure. The harvested data is 12 bytes from that address forward, so not the data pointed to by that m_data pointer but the pointer value itself and some following junk too. Is that really worth using? The mbuf data pointer is only slightly more random than the mbuf pointer (equally non-random if the data is in the mbuf). The bytes following it are even less random. I think the mbuf pointer usually points to an IP header. 12 bytes gives everything except the source and destination addresses. 20 would cover those too. There can't be buffer overruns in practice, even with a much larger size since small buffers go in the mbuf. In the mbuf header starting at the data pointer, there are always 16 bytes in the header alone, and many more after the header. On 32-bit systems, the header has 2 more bytes in mh_type after the 12 followed by 2 bytes of unnamed padding. The padding might be random but is more likely to be always 0. 16 altogether. On 64-bit systems, the header has 4 more bytes in mh_flags after the 12, then 2+2 for mh_type and padding as above, then 4 more of unnamed padding. 24 altogether. So why 12? Bruce___ svn-src-head@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
Re: svn commit: r257535 - head/sys/netgraph
On 2 Nov 2013, at 22:28, Bruce Evans wrote: > On Sat, 2 Nov 2013, Mark R V Murray wrote: > >> On 2 Nov 2013, at 09:32, Mark R V Murray wrote: Mark - did you initially mean the address of the mbuf m_data pointer, or the data payload itself? >>> >>> As Bruce says - the address of payload data itself. We don’t have 12-byte >>> pointers in FreeBSD. :-) >> >> Cancel that. >> >> The address passed must be the address of the m_data field in the mbuf >> structure. The harvested data is 12 bytes from that address forward, so not >> the data pointed to by that m_data pointer but the pointer value itself and >> some following junk too. > > Is that really worth using? Yes. Most of the value comes from the timing. The 12 bytes we snatch on the way past are opportunistic, and hopefully cheap due to being in-cache. M -- -- Mark R V Murray signature.asc Description: Message signed with OpenPGP using GPGMail
svn commit: r257561 - head/sys/arm/freescale/imx
Author: ian
Date: Sat Nov 2 22:44:35 2013
New Revision: 257561
URL: http://svnweb.freebsd.org/changeset/base/257561
Log:
Bugfix: the attach routine needs to use the same table of fdt compat
strings that the probe routine used.
Modified:
head/sys/arm/freescale/imx/imx_sdhci.c
Modified: head/sys/arm/freescale/imx/imx_sdhci.c
==
--- head/sys/arm/freescale/imx/imx_sdhci.c Sat Nov 2 22:18:41 2013
(r257560)
+++ head/sys/arm/freescale/imx/imx_sdhci.c Sat Nov 2 22:44:35 2013
(r257561)
@@ -558,13 +558,9 @@ imx_sdhci_attach(device_t dev)
sc->dev = dev;
- if (ofw_bus_is_compatible(dev, "fsl,imx51-esdhc")) {
- sc->hwtype = HWTYPE_ESDHC;
- } else if (ofw_bus_is_compatible(dev, "fsl,imx-usdhc")) {
- sc->hwtype = HWTYPE_USDHC;
- } else {
+ sc->hwtype = ofw_bus_search_compatible(dev, compat_data)->ocd_data;
+ if (sc->hwtype == HWTYPE_NONE)
panic("Impossible: not compatible in imx_sdhci_attach()");
- }
rid = 0;
sc->mem_res = bus_alloc_resource_any(dev, SYS_RES_MEMORY, &rid,
___
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: this breaks builds of head on stable/9 (Re: svn commit: r257268 - head/gnu/usr.bin/binutils/ld
On Fri, 1 Nov 2013 17:15:32 -0500, Brooks Davis writes: >(In practice WITHOUT_BMAKE is broken and corrupts your object tree with a >make that fails until manually deleted.) That's new? it worked last time I checked - which was around the time I made sure the make thus built was named for the flavor (fmake or bmake) desired so wouldn't accidentally be used when you change your mind. >We should decide if it should be fixed or removed before 10 goes out. AFAIK I committed workarounds for all the functional differences that worried people, before 10 was branched. So perhaps WITHOUT_BMAKE can be deprecated. ___ 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: this breaks builds of head on stable/9 (Re: svn commit: r257268 - head/gnu/usr.bin/binutils/ld
>told me what was wrong. The issue is that the :U modifer I used is a >bmake feature and that the toolchain target doesn't bootstrap bmake (or >perhaps doesn't do it early enough). I plan to commit a workaround for Hmmm toolchain is in the same TGTS list as buildworld, so should depend on upgrade_checks which is how the needed make gets built. Oh, that only applies if ALWAYS_CHECK_MAKE is defined. Might be a good idea to set ALWAYS_CHECK_MAKE is host isn't head? ___ 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"
