svn commit: r216095 - head/sbin/bsdlabel
Author: kevlo
Date: Wed Dec 1 08:07:32 2010
New Revision: 216095
URL: http://svn.freebsd.org/changeset/base/216095
Log:
Closing file descriptors when it's done
Modified:
head/sbin/bsdlabel/bsdlabel.c
Modified: head/sbin/bsdlabel/bsdlabel.c
==
--- head/sbin/bsdlabel/bsdlabel.c Wed Dec 1 05:52:27 2010
(r216094)
+++ head/sbin/bsdlabel/bsdlabel.c Wed Dec 1 08:07:32 2010
(r216095)
@@ -370,10 +370,12 @@ readboot(void)
p[60] = (st.st_size + secsize - 1) / secsize;
p[61] = 1;
p[62] = 0;
+ close(fd);
return;
} else if ((!alphacksum) && st.st_size <= BBSIZE) {
if (read(fd, bootarea, st.st_size) != st.st_size)
err(1, "read error %s", xxboot);
+ close(fd);
return;
}
errx(1, "boot code %s is wrong size", xxboot);
___
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: r216016 - head/sys/sparc64/include
Marius Strobl wrote: On Mon, Nov 29, 2010 at 08:23:08PM +0100, Marius Strobl wrote: On Tue, Nov 30, 2010 at 12:31:31AM +0600, Max Khon wrote: Marius, On Mon, Nov 29, 2010 at 1:45 AM, Marius Strobl wrote: On Sun, Nov 28, 2010 at 07:26:20PM +, Max Khon wrote: Author: fjoe Date: Sun Nov 28 19:26:20 2010 New Revision: 216016 URL: http://svn.freebsd.org/changeset/base/216016 Log: Define VM_KMEM_SIZE_MAX on sparc64. Otherwise kernel built with DEBUG_MEMGUARD panics early in kmeminit() with the message "kmem_suballoc: bad status return of 1" because of zero "size" argument passed to kmem_suballoc() due to "vm_kmem_size_max" being zero. The problem also exists on ia64. Modified: head/sys/sparc64/include/vmparam.h Modified: head/sys/sparc64/include/vmparam.h == --- head/sys/sparc64/include/vmparam.hSun Nov 28 18:59:52 2010 (r216015) +++ head/sys/sparc64/include/vmparam.hSun Nov 28 19:26:20 2010 (r216016) @@ -237,6 +237,14 @@ #endif /* + * Ceiling on amount of kmem_map kva space. + */ +#ifndef VM_KMEM_SIZE_MAX +#define VM_KMEM_SIZE_MAX((VM_MAX_KERNEL_ADDRESS - \ +VM_MIN_KERNEL_ADDRESS + 1) * 3 / 5) +#endif + +/* * Initial pagein size of beginning of executable file. */ #ifndef VM_INITIAL_PAGEIN How was that value determined? I've just copied it from amd64 to be non-zero for now. Do you have a better idea of what it should look like? Well, on sparc64 VM_MAX_KERNEL_ADDRESS already is dynamically adjusted to the maximum appropriate for the specific CPU during the early cycles of the kernel so I'd think one could just use VM_MAX_KERNEL_ADDRESS - VM_MIN_KERNEL_ADDRESS for VM_KMEM_SIZE_MAX there, I'm not sure what the intention of the ceiling provided by that macro actually is though In any case, the commit message of r180210 which changed the amd64 version to the current one talks about limiting the kmem map to 3.6GB and while it also fails to explain where that value comes from it looks rather amd64 specific and the formula used by the macro will result in a different ceiling on sparc64 and thus inappropriate. I've CC'ed alc@ who hopefully can shed some light on this. Apart from this the actual bug here seems to be that memguard_fudge() can't deal with a km_max being zero or that zero is passed to it as kmeminit() allows for VM_KMEM_SIZE_MAX not being defined. Oops, forgot to actually CC a...@. There's nothing particularly amd64-specific about the definition. In general, if you allow the kmem_map, which is basically the kernel's heap, to consume the entire kernel address space as you propose, then you're leaving no room for the buffer cache, thread stacks, pipes, and a few other things. Since the cap on the kmem_map size as defined by r180210 is a fraction of the overall kernel address space size, it scales automatically with the kernel address space size and should be a reasonable cap definition for most architectures. In the specific case of sparc64, I think it's fair to say that the kernel virtual address is sufficiently large and the amount of physical memory in any of the supported machines is small enough in comparison that it hasn't mattered that a kmem_map cap doesn't exist, because most of the aforementioned structures are scaled based on the amount of physical memory. In fact, it probably won't matter any time soon. All of that said, I would suggest fixing memguard_fudge() and reverting r216016 and the follow up change. All I think that is required to fix memguard_fudge() is Index: vm/memguard.c === --- vm/memguard.c (revision 216070) +++ vm/memguard.c (working copy) @@ -186,7 +186,7 @@ memguard_fudge(unsigned long km_size, unsigned lon memguard_mapsize = round_page(memguard_mapsize); if (memguard_mapsize / (2 * PAGE_SIZE) > mem_pgs) memguard_mapsize = mem_pgs * 2 * PAGE_SIZE; - if (km_size + memguard_mapsize > km_max) + if (km_max > 0 && km_size + memguard_mapsize > km_max) return (km_max); return (km_size + memguard_mapsize); } Regards, Alan ___ 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: r216098 - head/sys/geom/label
Author: jh
Date: Wed Dec 1 19:24:07 2010
New Revision: 216098
URL: http://svn.freebsd.org/changeset/base/216098
Log:
- Report an error when a label with invalid name is attempted to be
created with glabel(8).
- Fix a typo in an error message.
- Fix comment typos.
Approved by: pjd
Modified:
head/sys/geom/label/g_label.c
Modified: head/sys/geom/label/g_label.c
==
--- head/sys/geom/label/g_label.c Wed Dec 1 15:31:31 2010
(r216097)
+++ head/sys/geom/label/g_label.c Wed Dec 1 19:24:07 2010
(r216098)
@@ -124,13 +124,13 @@ g_label_is_name_ok(const char *label)
{
const char *s;
- /* Check is the label starts from ../ */
+ /* Check if the label starts from ../ */
if (strncmp(label, "../", 3) == 0)
return (0);
- /* Check is the label contains /../ */
+ /* Check if the label contains /../ */
if (strstr(label, "/../") != NULL)
return (0);
- /* Check is the label ends at ../ */
+ /* Check if the label ends at ../ */
if ((s = strstr(label, "/..")) != NULL && s[3] == '\0')
return (0);
return (1);
@@ -151,6 +151,8 @@ g_label_create(struct gctl_req *req, str
G_LABEL_DEBUG(0, "%s contains suspicious label, skipping.",
pp->name);
G_LABEL_DEBUG(1, "%s suspicious label is: %s", pp->name, label);
+ if (req != NULL)
+ gctl_error(req, "Label name %s is invalid.", label);
return (NULL);
}
gp = NULL;
@@ -346,7 +348,7 @@ g_label_ctl_create(struct gctl_req *req,
return;
}
if (*nargs != 2) {
- gctl_error(req, "Invalid number of argument.");
+ gctl_error(req, "Invalid number of arguments.");
return;
}
/*
___
svn-src-head@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r216099 - head/sys/ufs/ffs
Author: kib
Date: Wed Dec 1 21:19:11 2010
New Revision: 216099
URL: http://svn.freebsd.org/changeset/base/216099
Log:
Journal start looks up .sujournal file by doing lookup on the root dvp.
As result, failed softdep_mount() might leave up to two vnodes on the
mp mountlist, preventing mnt_ref from going to zero.
Call ffs_flushfiles() after failed softdep_mount() to clean mountlist.
Initial report by:Garrett Cooper
Reproduced and tested by: pho
Modified:
head/sys/ufs/ffs/ffs_vfsops.c
Modified: head/sys/ufs/ffs/ffs_vfsops.c
==
--- head/sys/ufs/ffs/ffs_vfsops.c Wed Dec 1 19:24:07 2010
(r216098)
+++ head/sys/ufs/ffs/ffs_vfsops.c Wed Dec 1 21:19:11 2010
(r216099)
@@ -928,6 +928,7 @@ ffs_mountfs(devvp, mp, td)
if ((fs->fs_flags & FS_DOSOFTDEP) &&
(error = softdep_mount(devvp, mp, fs, cred)) != 0) {
free(fs->fs_csp, M_UFSMNT);
+ ffs_flushfiles(mp, FORCECLOSE, td);
goto out;
}
if (fs->fs_snapinum[0] != 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: r216100 - head/bin/sh
Author: jilles Date: Wed Dec 1 23:26:32 2010 New Revision: 216100 URL: http://svn.freebsd.org/changeset/base/216100 Log: sh(1): Document that command's -p option also works with -v/-V. This was implemented in r201343. Modified: head/bin/sh/sh.1 Modified: head/bin/sh/sh.1 == --- head/bin/sh/sh.1Wed Dec 1 21:19:11 2010(r216099) +++ head/bin/sh/sh.1Wed Dec 1 23:26:32 2010(r216100) @@ -32,7 +32,7 @@ .\"from: @(#)sh.1 8.6 (Berkeley) 5/4/95 .\" $FreeBSD$ .\" -.Dd November 19, 2010 +.Dd December 1, 2010 .Dt SH 1 .Os .Sh NAME @@ -1712,7 +1712,8 @@ A synonym for the .Ic cd built-in command. .It Ic command Oo Fl p Oc Op Ar utility Op Ar argument ... -.It Ic command Oo Fl v | V Oc Op Ar utility +.It Ic command Oo Fl p Oc Fl v Ar utility +.It Ic command Oo Fl p Oc Fl V Ar utility The first form of invocation executes the specified .Ar utility , ignoring shell functions in the search. ___ 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: r216091 - head/sys/dev/usb
On Tuesday 30 November 2010 10:51 pm, Weongyo Jeong wrote: > Author: weongyo > Date: Wed Dec 1 03:51:06 2010 > New Revision: 216091 > URL: http://svn.freebsd.org/changeset/base/216091 > > Log: > Explicitly UP and DOWN the usbus interfaces (IFT_USB) when it's > attached or detached. Normally it should be changed through user > land ioctl(2) system calls but it looks there's no apps for USB and > no need. > > With this patch, libpcap would detect the usbus interfaces > correctly and tcpdump(1) could dump the USB packets into PCAP > format with -w option. However it couldn't print the output to > console because there's no printer-routine at tcpdump(1). I have written a very hackish USB packet printer for tcpdump based on Linux version. http://people.freebsd.org/~jkim/tcpdump-usb.diff Note we can only print headers as Linux version does. Are you planning on adopting DLT_USB_LINUX or DLT_USB_LINUX_MMAPPED or applying for our own DLT, BTW? Jung-uk Kim ___ 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: r216101 - head/sys/netinet
Author: lstewart Date: Thu Dec 2 00:47:55 2010 New Revision: 216101 URL: http://svn.freebsd.org/changeset/base/216101 Log: Pass NULL instead of 0 for the th pointer value. NULL != 0 on all platforms. Submitted by: David Hayes MFC after:9 weeks X-MFC with: r215166 Modified: head/sys/netinet/tcp_timer.c Modified: head/sys/netinet/tcp_timer.c == --- head/sys/netinet/tcp_timer.cWed Dec 1 23:26:32 2010 (r216100) +++ head/sys/netinet/tcp_timer.cThu Dec 2 00:47:55 2010 (r216101) @@ -567,7 +567,7 @@ tcp_timer_rexmt(void * xtp) */ tp->t_rtttime = 0; - cc_cong_signal(tp, 0, CC_RTO); + cc_cong_signal(tp, NULL, CC_RTO); (void) tcp_output(tp); ___ 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: r216101 - head/sys/netinet
On 12/02/10 11:47, Lawrence Stewart wrote: > Author: lstewart > Date: Thu Dec 2 00:47:55 2010 > New Revision: 216101 > URL: http://svn.freebsd.org/changeset/base/216101 > > Log: > Pass NULL instead of 0 for the th pointer value. NULL != 0 on all platforms. > Sponsored by: FreeBSD Foundation > Submitted by: David Hayes > MFC after: 9 weeks > X-MFC with: r215166 > > Modified: > head/sys/netinet/tcp_timer.c ___ 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: r216103 - head/sys/netinet
Author: lstewart Date: Thu Dec 2 01:01:37 2010 New Revision: 216103 URL: http://svn.freebsd.org/changeset/base/216103 Log: Set ssthresh appropriately on RTO. This change was accidentally not ported from the pre modular CC stack. Sponsored by: FreeBSD Foundation Submitted by: David Hayes MFC after:9 weeks X-MFC with: r215166 Modified: head/sys/netinet/tcp_input.c Modified: head/sys/netinet/tcp_input.c == --- head/sys/netinet/tcp_input.cThu Dec 2 00:49:02 2010 (r216102) +++ head/sys/netinet/tcp_input.cThu Dec 2 01:01:37 2010 (r216103) @@ -378,6 +378,8 @@ cc_cong_signal(struct tcpcb *tp, struct tp->t_dupacks = 0; tp->t_bytes_acked = 0; EXIT_RECOVERY(tp->t_flags); + tp->snd_ssthresh = max(2, min(tp->snd_wnd, tp->snd_cwnd) / 2 / + tp->t_maxseg) * tp->t_maxseg; tp->snd_cwnd = tp->t_maxseg; break; case CC_RTO_ERR: ___ svn-src-head@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r216104 - head/sys/kern
Author: trasz
Date: Thu Dec 2 01:14:45 2010
New Revision: 216104
URL: http://svn.freebsd.org/changeset/base/216104
Log:
Remove useless NULL checks for M_WAITOK mallocs.
Modified:
head/sys/kern/sysv_msg.c
head/sys/kern/sysv_shm.c
Modified: head/sys/kern/sysv_msg.c
==
--- head/sys/kern/sysv_msg.cThu Dec 2 01:01:37 2010(r216103)
+++ head/sys/kern/sysv_msg.cThu Dec 2 01:14:45 2010(r216104)
@@ -200,18 +200,10 @@ msginit()
TUNABLE_INT_FETCH("kern.ipc.msgtql", &msginfo.msgtql);
msgpool = malloc(msginfo.msgmax, M_MSG, M_WAITOK);
- if (msgpool == NULL)
- panic("msgpool is NULL");
msgmaps = malloc(sizeof(struct msgmap) * msginfo.msgseg, M_MSG,
M_WAITOK);
- if (msgmaps == NULL)
- panic("msgmaps is NULL");
msghdrs = malloc(sizeof(struct msg) * msginfo.msgtql, M_MSG, M_WAITOK);
- if (msghdrs == NULL)
- panic("msghdrs is NULL");
msqids = malloc(sizeof(struct msqid_kernel) * msginfo.msgmni, M_MSG,
M_WAITOK);
- if (msqids == NULL)
- panic("msqids is NULL");
/*
* msginfo.msgssz should be a power of two for efficiency reasons.
@@ -233,9 +225,6 @@ msginit()
panic("msginfo.msgseg > 32767");
}
- if (msgmaps == NULL)
- panic("msgmaps is NULL");
-
for (i = 0; i < msginfo.msgseg; i++) {
if (i > 0)
msgmaps[i-1].next = i;
@@ -244,9 +233,6 @@ msginit()
free_msgmaps = 0;
nfree_msgmaps = msginfo.msgseg;
- if (msghdrs == NULL)
- panic("msghdrs is NULL");
-
for (i = 0; i < msginfo.msgtql; i++) {
msghdrs[i].msg_type = 0;
if (i > 0)
@@ -258,9 +244,6 @@ msginit()
}
free_msghdrs = &msghdrs[0];
- if (msqids == NULL)
- panic("msqids is NULL");
-
for (i = 0; i < msginfo.msgmni; i++) {
msqids[i].u.msg_qbytes = 0; /* implies entry is available */
msqids[i].u.msg_perm.seq = 0; /* reset to a known value */
Modified: head/sys/kern/sysv_shm.c
==
--- head/sys/kern/sysv_shm.cThu Dec 2 01:01:37 2010(r216103)
+++ head/sys/kern/sysv_shm.cThu Dec 2 01:14:45 2010(r216104)
@@ -878,8 +878,6 @@ shminit()
shmalloced = shminfo.shmmni;
shmsegs = malloc(shmalloced * sizeof(shmsegs[0]), M_SHM, M_WAITOK);
- if (shmsegs == NULL)
- panic("cannot allocate initial memory for sysvshm");
for (i = 0; i < shmalloced; i++) {
shmsegs[i].u.shm_perm.mode = SHMSEG_FREE;
shmsegs[i].u.shm_perm.seq = 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: r216105 - in head/sys/netinet: . cc
Author: lstewart
Date: Thu Dec 2 01:36:00 2010
New Revision: 216105
URL: http://svn.freebsd.org/changeset/base/216105
Log:
- Reinstantiate the after_idle hook call in tcp_output(), which got lost
somewhere along the way due to mismerging r211464 in our development tree.
- Capture the essence of r211464 in NewReno's after_idle() hook. We don't
use V_ss_fltsz/V_ss_fltsz_local yet which needs to be revisited.
Sponsored by: FreeBSD Foundation
Submitted by: David Hayes
MFC after:9 weeks
X-MFC with: r215166
Modified:
head/sys/netinet/cc/cc_newreno.c
head/sys/netinet/tcp_output.c
Modified: head/sys/netinet/cc/cc_newreno.c
==
--- head/sys/netinet/cc/cc_newreno.cThu Dec 2 01:14:45 2010
(r216104)
+++ head/sys/netinet/cc/cc_newreno.cThu Dec 2 01:36:00 2010
(r216105)
@@ -216,15 +216,28 @@ newreno_post_recovery(struct cc_var *ccv
void
newreno_after_idle(struct cc_var *ccv)
{
+ int rw;
+
/*
-* We have been idle for "a while" and no acks are expected to clock out
-* any data we send -- slow start to get ack "clock" running again.
+* If we've been idle for more than one retransmit timeout the old
+* congestion window is no longer current and we have to reduce it to
+* the restart window before we can transmit again.
+*
+* The restart window is the initial window or the last CWND, whichever
+* is smaller.
+*
+* This is done to prevent us from flooding the path with a full CWND at
+* wirespeed, overloading router and switch buffers along the way.
+*
+* See RFC5681 Section 4.1. "Restarting Idle Connections".
*/
if (V_tcp_do_rfc3390)
- CCV(ccv, snd_cwnd) = min(4 * CCV(ccv, t_maxseg),
+ rw = min(4 * CCV(ccv, t_maxseg),
max(2 * CCV(ccv, t_maxseg), 4380));
else
- CCV(ccv, snd_cwnd) = CCV(ccv, t_maxseg) * 2;
+ rw = CCV(ccv, t_maxseg) * 2;
+
+ CCV(ccv, snd_cwnd) = min(rw, CCV(ccv, snd_cwnd));
}
Modified: head/sys/netinet/tcp_output.c
==
--- head/sys/netinet/tcp_output.c Thu Dec 2 01:14:45 2010
(r216104)
+++ head/sys/netinet/tcp_output.c Thu Dec 2 01:36:00 2010
(r216105)
@@ -148,7 +148,7 @@ tcp_output(struct tcpcb *tp)
{
struct socket *so = tp->t_inpcb->inp_socket;
long len, recwin, sendwin;
- int off, flags, error, rw;
+ int off, flags, error;
struct mbuf *m;
struct ip *ip = NULL;
struct ipovly *ipov = NULL;
@@ -182,37 +182,8 @@ tcp_output(struct tcpcb *tp)
* to send, then transmit; otherwise, investigate further.
*/
idle = (tp->t_flags & TF_LASTIDLE) || (tp->snd_max == tp->snd_una);
- if (idle && ticks - tp->t_rcvtime >= tp->t_rxtcur) {
- /*
-* If we've been idle for more than one retransmit
-* timeout the old congestion window is no longer
-* current and we have to reduce it to the restart
-* window before we can transmit again.
-*
-* The restart window is the initial window or the last
-* CWND, whichever is smaller.
-*
-* This is done to prevent us from flooding the path with
-* a full CWND at wirespeed, overloading router and switch
-* buffers along the way.
-*
-* See RFC5681 Section 4.1. "Restarting Idle Connections".
-*/
- if (V_tcp_do_rfc3390)
- rw = min(4 * tp->t_maxseg,
-max(2 * tp->t_maxseg, 4380));
-#ifdef INET6
- else if ((isipv6 ? in6_localaddr(&tp->t_inpcb->in6p_faddr) :
- in_localaddr(tp->t_inpcb->inp_faddr)))
-#else
- else if (in_localaddr(tp->t_inpcb->inp_faddr))
-#endif
- rw = V_ss_fltsz_local * tp->t_maxseg;
- else
- rw = V_ss_fltsz * tp->t_maxseg;
-
- tp->snd_cwnd = min(rw, tp->snd_cwnd);
- }
+ if (idle && ticks - tp->t_rcvtime >= tp->t_rxtcur)
+ cc_after_idle(tp);
tp->t_flags &= ~TF_LASTIDLE;
if (idle) {
if (tp->t_flags & TF_MORETOCOME) {
___
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: r216106 - head/usr.bin/find
Author: kevlo Date: Thu Dec 2 01:46:06 2010 New Revision: 216106 URL: http://svn.freebsd.org/changeset/base/216106 Log: Don't error out while searching for empty directories. Submitted by: Bakul Shah Modified: head/usr.bin/find/function.c Modified: head/usr.bin/find/function.c == --- head/usr.bin/find/function.cThu Dec 2 01:36:00 2010 (r216105) +++ head/usr.bin/find/function.cThu Dec 2 01:46:06 2010 (r216106) @@ -559,7 +559,7 @@ f_empty(PLAN *plan __unused, FTSENT *ent empty = 1; dir = opendir(entry->fts_accpath); if (dir == NULL) - err(1, "%s", entry->fts_accpath); + return 0; for (dp = readdir(dir); dp; dp = readdir(dir)) if (dp->d_name[0] != '.' || (dp->d_name[1] != '\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: r216107 - head/sys/netinet/cc
Author: lstewart
Date: Thu Dec 2 02:32:46 2010
New Revision: 216107
URL: http://svn.freebsd.org/changeset/base/216107
Log:
General cleanup of the NewReno CC module (no functional changes):
- Remove superfluous includes and unhelpful comments.
- Alphabetically order functions.
- Make functions static.
Sponsored by: FreeBSD Foundation
MFC after:9 weeks
X-MFC with: r215166
Modified:
head/sys/netinet/cc/cc_newreno.c
Modified: head/sys/netinet/cc/cc_newreno.c
==
--- head/sys/netinet/cc/cc_newreno.cThu Dec 2 01:46:06 2010
(r216106)
+++ head/sys/netinet/cc/cc_newreno.cThu Dec 2 02:32:46 2010
(r216107)
@@ -52,41 +52,35 @@ __FBSDID("$FreeBSD$");
#include
#include
+#include
#include
#include
#include
#include
+#include
-#include
-#include
+#include
#include
-#include
-#include
#include
#include
#include
-void newreno_ack_received(struct cc_var *ccv, uint16_t type);
-void newreno_cong_signal(struct cc_var *ccv, uint32_t type);
-void newreno_post_recovery(struct cc_var *ccv);
-void newreno_after_idle(struct cc_var *ccv);
+static voidnewreno_ack_received(struct cc_var *ccv, uint16_t type);
+static voidnewreno_after_idle(struct cc_var *ccv);
+static voidnewreno_cong_signal(struct cc_var *ccv, uint32_t type);
+static voidnewreno_post_recovery(struct cc_var *ccv);
struct cc_algo newreno_cc_algo = {
.name = "newreno",
.ack_received = newreno_ack_received,
+ .after_idle = newreno_after_idle,
.cong_signal = newreno_cong_signal,
.post_recovery = newreno_post_recovery,
- .after_idle = newreno_after_idle
};
-/*
- * Increase cwnd on receipt of a successful ACK:
- * if cwnd <= ssthresh, increases by 1 MSS per ACK
- * if cwnd > ssthresh, increase by ~1 MSS per RTT
- */
-void
+static void
newreno_ack_received(struct cc_var *ccv, uint16_t type)
{
if (type == CC_ACK && !IN_RECOVERY(CCV(ccv, t_flags)) &&
@@ -153,10 +147,37 @@ newreno_ack_received(struct cc_var *ccv,
}
}
+static void
+newreno_after_idle(struct cc_var *ccv)
+{
+ int rw;
+
+ /*
+* If we've been idle for more than one retransmit timeout the old
+* congestion window is no longer current and we have to reduce it to
+* the restart window before we can transmit again.
+*
+* The restart window is the initial window or the last CWND, whichever
+* is smaller.
+*
+* This is done to prevent us from flooding the path with a full CWND at
+* wirespeed, overloading router and switch buffers along the way.
+*
+* See RFC5681 Section 4.1. "Restarting Idle Connections".
+*/
+ if (V_tcp_do_rfc3390)
+ rw = min(4 * CCV(ccv, t_maxseg),
+ max(2 * CCV(ccv, t_maxseg), 4380));
+ else
+ rw = CCV(ccv, t_maxseg) * 2;
+
+ CCV(ccv, snd_cwnd) = min(rw, CCV(ccv, snd_cwnd));
+}
+
/*
- * manage congestion signals
+ * Perform any necessary tasks before we enter congestion recovery.
*/
-void
+static void
newreno_cong_signal(struct cc_var *ccv, uint32_t type)
{
u_int win;
@@ -183,11 +204,9 @@ newreno_cong_signal(struct cc_var *ccv,
}
/*
- * decrease the cwnd in response to packet loss or a transmit timeout.
- * th can be null, in which case cwnd will be set according to reno instead
- * of new reno.
+ * Perform any necessary tasks before we exit congestion recovery.
*/
-void
+static void
newreno_post_recovery(struct cc_var *ccv)
{
if (IN_FASTRECOVERY(CCV(ccv, t_flags))) {
@@ -209,36 +228,5 @@ newreno_post_recovery(struct cc_var *ccv
}
}
-/*
- * if a connection has been idle for a while and more data is ready to be sent,
- * reset cwnd
- */
-void
-newreno_after_idle(struct cc_var *ccv)
-{
- int rw;
-
- /*
-* If we've been idle for more than one retransmit timeout the old
-* congestion window is no longer current and we have to reduce it to
-* the restart window before we can transmit again.
-*
-* The restart window is the initial window or the last CWND, whichever
-* is smaller.
-*
-* This is done to prevent us from flooding the path with a full CWND at
-* wirespeed, overloading router and switch buffers along the way.
-*
-* See RFC5681 Section 4.1. "Restarting Idle Connections".
-*/
- if (V_tcp_do_rfc3390)
- rw = min(4 * CCV(ccv, t_maxseg),
- max(2 * CCV(ccv, t_maxseg), 4380));
- else
- rw = CCV(ccv, t_maxseg) * 2;
-
- CCV(ccv, snd_cwnd) = min(rw, CCV(ccv, snd_cwnd));
-}
-
DECLARE_CC_MODULE(newreno, &newreno_cc_algo);
___
svn-src-head@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-head
To u
svn commit: r216109 - in head/sys/modules: . cc
Author: lstewart
Date: Thu Dec 2 03:10:35 2010
New Revision: 216109
URL: http://svn.freebsd.org/changeset/base/216109
Log:
Add build infrastructure for the forthcoming CC algorithm modules.
Sponsored by: FreeBSD Foundation
Submitted by: David Hayes
MFC after:3 months
Added:
head/sys/modules/cc/
head/sys/modules/cc/Makefile (contents, props changed)
Modified:
head/sys/modules/Makefile
Modified: head/sys/modules/Makefile
==
--- head/sys/modules/Makefile Thu Dec 2 02:34:39 2010(r216108)
+++ head/sys/modules/Makefile Thu Dec 2 03:10:35 2010(r216109)
@@ -51,6 +51,7 @@ SUBDIR= ${_3dfx} \
${_cardbus} \
cas \
${_cbb} \
+ cc \
cd9660 \
cd9660_iconv \
${_ce} \
Added: head/sys/modules/cc/Makefile
==
--- /dev/null 00:00:00 1970 (empty, because file is newly added)
+++ head/sys/modules/cc/MakefileThu Dec 2 03:10:35 2010
(r216109)
@@ -0,0 +1,5 @@
+# $FreeBSD$
+
+SUBDIR=
+
+.include
___
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: r216101 - head/sys/netinet
On Thu, 2 Dec 2010, Lawrence Stewart wrote: Log: Pass NULL instead of 0 for the th pointer value. NULL != 0 on all platforms. Submitted by: David Hayes MFC after: 9 weeks X-MFC with:r215166 Modified: head/sys/netinet/tcp_timer.c Modified: head/sys/netinet/tcp_timer.c == --- head/sys/netinet/tcp_timer.cWed Dec 1 23:26:32 2010 (r216100) +++ head/sys/netinet/tcp_timer.cThu Dec 2 00:47:55 2010 (r216101) @@ -567,7 +567,7 @@ tcp_timer_rexmt(void * xtp) */ tp->t_rtttime = 0; - cc_cong_signal(tp, 0, CC_RTO); + cc_cong_signal(tp, NULL, CC_RTO); (void) tcp_output(tp); Er this never passed 0, (provided a prototype is in scope), and can neverpass either 0 or NULL, since 0 and NULL are null pointer constants and functions cannot pass constants. It used to pass a null pointer, after converting the null pointer constant 0 according to the prototype. Now it passes the same null pointer after converting the possibly-different null pointer constant NULL according to the prototype. NULL == 0 on all platform, since both are null pointer constants, and one of the following cases applies: - if NULL is an integer constant with value 0, then of course it equals plain int 0. Moreover, even if the type of NULL is different from the type of plain int 0, the types become identical before the comparison is done, since both operands are converted to a common type. - if NULL is an integer constant with value 0 cast to (void *), then it certainly has a different type than plain int 0, and may have a different representation. However, when compared with plain 0, both operands are converted to a common type, which is necessarily that of NULL. This is a pointer type, so so both operands are null pointers. These null pointers may have different representations, but any two null pointers to the same type are specified to compare equal. So NULL == 0 in this case too. The last case is essentially what applies in function calls. Now the conversion of either NULL or 0 or any other null pointer constant to a null pointer is done according to the prototype. The results are not necessarily the same like I said above, but they compare the same. Thus spelling the null pointer constant as 0 made no difference to the higher-level results on any platform (unless you do something like memcmp of null pointers, and use an exotic platform where null pointers have different representations). It was just a style bug. Bruce ___ svn-src-head@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r216111 - head/sys/dev/fxp
Author: yongari
Date: Thu Dec 2 03:53:29 2010
New Revision: 216111
URL: http://svn.freebsd.org/changeset/base/216111
Log:
If RX lockup workaround is enabled, fxp(4) will periodically reset
the controller to workaround silicon bug of i82557. Each reset will
re-establish link which in turn triggers MII status change
callback. The callback will try to reconfigure controller if the
controller is not i82557 to enable flow-control. This caused
endless link UP/DOWN when the workaround was enabled on non-i82557
controller.
To fix the issue, apply RX lockup workaround only for i82557.
Previously it blindly checked undocumented EEPROM location such
that it sometimes enabled the workaround for other controllers. At
this time, only i82557 is known to have the silicon bug.
This fixes a regression introduced in r215906 which enabled flow
control support for all controllers except i82557.
Reported by: Karl Denninger (karl <> denninger dot net)
Tested by:Karl Denninger (karl <> denninger dot net)
MFC after:3 days
Modified:
head/sys/dev/fxp/if_fxp.c
Modified: head/sys/dev/fxp/if_fxp.c
==
--- head/sys/dev/fxp/if_fxp.c Thu Dec 2 03:34:27 2010(r216110)
+++ head/sys/dev/fxp/if_fxp.c Thu Dec 2 03:53:29 2010(r216111)
@@ -526,10 +526,12 @@ fxp_attach(device_t dev)
}
/* Receiver lock-up workaround detection. */
- fxp_read_eeprom(sc, &data, 3, 1);
- if ((data & 0x03) != 0x03) {
- sc->flags |= FXP_FLAG_RXBUG;
- device_printf(dev, "Enabling Rx lock-up workaround\n");
+ if (sc->revision < FXP_REV_82558_A4) {
+ fxp_read_eeprom(sc, &data, 3, 1);
+ if ((data & 0x03) != 0x03) {
+ sc->flags |= FXP_FLAG_RXBUG;
+ device_printf(dev, "Enabling Rx lock-up workaround\n");
+ }
}
/*
___
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: r216112 - head/sys/kern
Author: imp
Date: Thu Dec 2 04:28:01 2010
New Revision: 216112
URL: http://svn.freebsd.org/changeset/base/216112
Log:
removed tag is '-', not '+'.
remove extra return.
Modified:
head/sys/kern/subr_bus.c
Modified: head/sys/kern/subr_bus.c
==
--- head/sys/kern/subr_bus.cThu Dec 2 03:53:29 2010(r216111)
+++ head/sys/kern/subr_bus.cThu Dec 2 04:28:01 2010(r216112)
@@ -718,8 +718,7 @@ devadded(device_t dev)
static void
devremoved(device_t dev)
{
- devaddq("+", device_get_nameunit(dev), dev);
- return;
+ devaddq("-", device_get_nameunit(dev), dev);
}
/*
___
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: r216101 - head/sys/netinet
On 12/02/10 14:45, Bruce Evans wrote: > On Thu, 2 Dec 2010, Lawrence Stewart wrote: > >> Log: >> Pass NULL instead of 0 for the th pointer value. NULL != 0 on all >> platforms. >> >> Submitted by:David Hayes >> MFC after:9 weeks >> X-MFC with:r215166 >> >> Modified: >> head/sys/netinet/tcp_timer.c >> >> Modified: head/sys/netinet/tcp_timer.c >> == >> >> --- head/sys/netinet/tcp_timer.cWed Dec 1 23:26:32 2010(r216100) >> +++ head/sys/netinet/tcp_timer.cThu Dec 2 00:47:55 2010(r216101) >> @@ -567,7 +567,7 @@ tcp_timer_rexmt(void * xtp) >> */ >> tp->t_rtttime = 0; >> >> -cc_cong_signal(tp, 0, CC_RTO); >> +cc_cong_signal(tp, NULL, CC_RTO); >> >> (void) tcp_output(tp); >> > > Er this never passed 0, (provided a prototype is in scope), and can > neverpass > either 0 or NULL, since 0 and NULL are null pointer constants and functions > cannot pass constants. > > It used to pass a null pointer, after converting the null pointer > constant 0 according to the prototype. Now it passes the same null > pointer after converting the possibly-different null pointer constant > NULL according to the prototype. > > NULL == 0 on all platform, since both are null pointer constants, and > one of the following cases applies: > - if NULL is an integer constant with value 0, then of course it equals > plain int 0. Moreover, even if the type of NULL is different from > the type of plain int 0, the types become identical before the > comparison is done, since both operands are converted to a common > type. > - if NULL is an integer constant with value 0 cast to (void *), then it > certainly has a different type than plain int 0, and may have a > different representation. However, when compared with plain 0, both > operands are converted to a common type, which is necessarily that > of NULL. This is a pointer type, so so both operands are null > pointers. These null pointers may have different representations, > but any two null pointers to the same type are specified to compare > equal. So NULL == 0 in this case too. > > The last case is essentially what applies in function calls. Now the > conversion of either NULL or 0 or any other null pointer constant to a > null pointer is done according to the prototype. The results are not > necessarily the same like I said above, but they compare the same. > Thus spelling the null pointer constant as 0 made no difference to > the higher-level results on any platform (unless you do something like > memcmp of null pointers, and use an exotic platform where null pointers > have different representations). It was just a style bug. hmm, thanks for the informative response. I'm not sure where I got it in my head that NULL != 0 on all platforms, but you've now corrected that bit of misunderstanding. For the record, the now known to be bogus commit log was my own creation based on my (mis)understanding and did not come from David. When it comes time to MFC, should I use the bogus log message or correct it to something like: "Fix a minor style nit by passing NULL instead of 0 for the th pointer value." Cheers, Lawrence ___ 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: r209110 - in head/lib/msun: . src
On Tue, Jun 15, 2010, David Schultz wrote: > On Tue, Jun 15, 2010, Kostik Belousov wrote: > > On Sat, Jun 12, 2010 at 05:32:05PM +, David Schultz wrote: > > > Author: das > > > Date: Sat Jun 12 17:32:05 2010 > > > New Revision: 209110 > > > URL: http://svn.freebsd.org/changeset/base/209110 > > > > > > Log: > > > Introduce __isnanf() as an alias for isnanf(), and make the isnan() > > > macro expand to __isnanf() instead of isnanf() for float arguments. > > > This change is needed because isnanf() isn't declared in strict POSIX > > > or C99 mode. > > > > > > Compatibility note: Apps using isnan(float) that are compiled after > > > this change won't link against an older libm. > > > > > > Reported by:Florian Forster > > > > May be, it makes sense to remove the default version for the isnan symbol ? > > Wouldn't this mean apps that use isnanf() directly will no longer > compile? isnanf() is a historical BSD interface, and although > it's been deprecated for many years, it's still declared (if > __BSD_VISIBLE). > > Oops, to complicate matters further, I just noticed that we > already have isnanf and __isnanf symbols in libc, so maybe the new > symbol isn't needed. (isnan() and isnanf() are in libc because > that's where they were historically.) The second version in > libm looks like a mistake (wrong scope of the #if 0 in s_isnan.c.) > Perhaps we could just remove the duplicate symbols from libm. Any thoughts on removing the isnanf and __isnanf symbols from libm? Both symbols are already in libc for historical reasons, so the duplication isn't needed. Although we've had the duplicate isnanf symbol in libm for several releases, I don't believe removing it will cause problems; apps will just pick up the libc version. __isnanf is only present in libm in 9-CURRENT (due to the commit referenced above). Because of symbol version differences, however, removing it will affect apps that were linked under 9-CURRENT AND rely on isnanf AND link against libm before libc. On my system, libwebkit is the only affected binary I could find. ___ 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: r216113 - head/gnu/usr.bin/cc
Author: nwhitehorn
Date: Thu Dec 2 04:58:07 2010
New Revision: 216113
URL: http://svn.freebsd.org/changeset/base/216113
Log:
The driver-XXX.c files used for host CPU detection with -march=native
should not be compiled in the cross-tools case (where -march=native makes
no sense). This fixes cross-building x86 toolchains on non-x86 systems.
Modified:
head/gnu/usr.bin/cc/Makefile.fe
Modified: head/gnu/usr.bin/cc/Makefile.fe
==
--- head/gnu/usr.bin/cc/Makefile.fe Thu Dec 2 04:28:01 2010
(r216112)
+++ head/gnu/usr.bin/cc/Makefile.fe Thu Dec 2 04:58:07 2010
(r216113)
@@ -19,7 +19,8 @@ CFLAGS+= ${DRIVER_DEFINES}
SRCS= gcc.c opts-common.c options.c intl.c prefix.c version.c
-.if exists(${GCCDIR}/config/${GCC_CPU}/driver-${GCC_CPU}.c)
+.if ${TARGET_ARCH} == ${MACHINE_ARCH} && \
+ exists(${GCCDIR}/config/${GCC_CPU}/driver-${GCC_CPU}.c)
SRCS+= driver-${GCC_CPU}.c
.endif
___
svn-src-head@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
Re: svn commit: r209110 - in head/lib/msun: . src
On Wed, Dec 1, 2010 at 8:57 PM, David Schultz wrote: > On Tue, Jun 15, 2010, David Schultz wrote: >> On Tue, Jun 15, 2010, Kostik Belousov wrote: >> > On Sat, Jun 12, 2010 at 05:32:05PM +, David Schultz wrote: >> > > Author: das >> > > Date: Sat Jun 12 17:32:05 2010 >> > > New Revision: 209110 >> > > URL: http://svn.freebsd.org/changeset/base/209110 >> > > >> > > Log: >> > > Introduce __isnanf() as an alias for isnanf(), and make the isnan() >> > > macro expand to __isnanf() instead of isnanf() for float arguments. >> > > This change is needed because isnanf() isn't declared in strict POSIX >> > > or C99 mode. >> > > >> > > Compatibility note: Apps using isnan(float) that are compiled after >> > > this change won't link against an older libm. >> > > >> > > Reported by: Florian Forster >> > >> > May be, it makes sense to remove the default version for the isnan symbol ? >> >> Wouldn't this mean apps that use isnanf() directly will no longer >> compile? isnanf() is a historical BSD interface, and although >> it's been deprecated for many years, it's still declared (if >> __BSD_VISIBLE). >> >> Oops, to complicate matters further, I just noticed that we >> already have isnanf and __isnanf symbols in libc, so maybe the new >> symbol isn't needed. (isnan() and isnanf() are in libc because >> that's where they were historically.) The second version in >> libm looks like a mistake (wrong scope of the #if 0 in s_isnan.c.) >> Perhaps we could just remove the duplicate symbols from libm. > > Any thoughts on removing the isnanf and __isnanf symbols from > libm? Both symbols are already in libc for historical reasons, so > the duplication isn't needed. > > Although we've had the duplicate isnanf symbol in libm for several > releases, I don't believe removing it will cause problems; apps > will just pick up the libc version. __isnanf is only present in > libm in 9-CURRENT (due to the commit referenced above). Because > of symbol version differences, however, removing it will affect > apps that were linked under 9-CURRENT AND rely on isnanf AND link > against libm before libc. On my system, libwebkit is the only > affected binary I could find. Yeah... it's going to be broken according to the manpage (manpage explicitly calls out -lm and math.h) and POSIX (POSIX only calls out math.h) as math.h lives in lib/msun for C apps: $ find /usr/src/ -name math.h /usr/src/contrib/libstdc++/include/tr1/math.h /usr/src/contrib/libstdc++/include/c_compatibility/math.h /usr/src/lib/msun/src/math.h So unless math.h is going to move to libc, I'd leave it alone if I was making the final call. Thanks, -Garrett ___ 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: r209110 - in head/lib/msun: . src
On Wed, Dec 1, 2010 at 9:16 PM, Garrett Cooper wrote: > On Wed, Dec 1, 2010 at 8:57 PM, David Schultz wrote: >> On Tue, Jun 15, 2010, David Schultz wrote: >>> On Tue, Jun 15, 2010, Kostik Belousov wrote: >>> > On Sat, Jun 12, 2010 at 05:32:05PM +, David Schultz wrote: >>> > > Author: das >>> > > Date: Sat Jun 12 17:32:05 2010 >>> > > New Revision: 209110 >>> > > URL: http://svn.freebsd.org/changeset/base/209110 >>> > > >>> > > Log: >>> > > Introduce __isnanf() as an alias for isnanf(), and make the isnan() >>> > > macro expand to __isnanf() instead of isnanf() for float arguments. >>> > > This change is needed because isnanf() isn't declared in strict POSIX >>> > > or C99 mode. >>> > > >>> > > Compatibility note: Apps using isnan(float) that are compiled after >>> > > this change won't link against an older libm. >>> > > >>> > > Reported by: Florian Forster >>> > >>> > May be, it makes sense to remove the default version for the isnan symbol >>> > ? >>> >>> Wouldn't this mean apps that use isnanf() directly will no longer >>> compile? isnanf() is a historical BSD interface, and although >>> it's been deprecated for many years, it's still declared (if >>> __BSD_VISIBLE). >>> >>> Oops, to complicate matters further, I just noticed that we >>> already have isnanf and __isnanf symbols in libc, so maybe the new >>> symbol isn't needed. (isnan() and isnanf() are in libc because >>> that's where they were historically.) The second version in >>> libm looks like a mistake (wrong scope of the #if 0 in s_isnan.c.) >>> Perhaps we could just remove the duplicate symbols from libm. >> >> Any thoughts on removing the isnanf and __isnanf symbols from >> libm? Both symbols are already in libc for historical reasons, so >> the duplication isn't needed. >> >> Although we've had the duplicate isnanf symbol in libm for several >> releases, I don't believe removing it will cause problems; apps >> will just pick up the libc version. __isnanf is only present in >> libm in 9-CURRENT (due to the commit referenced above). Because >> of symbol version differences, however, removing it will affect >> apps that were linked under 9-CURRENT AND rely on isnanf AND link >> against libm before libc. On my system, libwebkit is the only >> affected binary I could find. > > Yeah... it's going to be broken according to the manpage (manpage > explicitly calls out -lm and math.h) and POSIX (POSIX only calls out > math.h) as math.h lives in lib/msun for C apps: > > $ find /usr/src/ -name math.h > /usr/src/contrib/libstdc++/include/tr1/math.h > /usr/src/contrib/libstdc++/include/c_compatibility/math.h > /usr/src/lib/msun/src/math.h > > So unless math.h is going to move to libc, I'd leave it alone if I > was making the final call. Arg -- please ignore this email. I did another search and you said isnanf, not isnan. It's not mentioned anywhere in POSIX land or in any of our manpages. Sorry bout that, -Garrett ___ 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: r216114 - in head/sys: modules/cc modules/cc/cc_cubic netinet/cc
Author: lstewart
Date: Thu Dec 2 06:05:44 2010
New Revision: 216114
URL: http://svn.freebsd.org/changeset/base/216114
Log:
Import a clean-room implementation of the experimental CUBIC congestion
control
algorithm based on the Internet-Draft "draft-rhee-tcpm-cubic-02.txt". It is
implemented as a kernel module compatible with the recently committed modular
congestion control framework.
CUBIC was designed for provide increased throughput in fast and long-distance
networks. It attempts to maintain fairness when competing with legacy NewReno
TCP in lower speed scenarios where NewReno is able to operate adequately. The
paper "CUBIC: A New TCP-Friendly High-Speed TCP Variant" provides additional
detail.
In collaboration with:David Hayes and
Grenville Armitage
Sponsored by: FreeBSD Foundation
Reviewed by: rpaulo (older patch from a few weeks ago)
MFC after:3 months
Added:
head/sys/modules/cc/cc_cubic/
head/sys/modules/cc/cc_cubic/Makefile (contents, props changed)
head/sys/netinet/cc/cc_cubic.c (contents, props changed)
head/sys/netinet/cc/cc_cubic.h (contents, props changed)
Modified:
head/sys/modules/cc/Makefile
Modified: head/sys/modules/cc/Makefile
==
--- head/sys/modules/cc/MakefileThu Dec 2 04:58:07 2010
(r216113)
+++ head/sys/modules/cc/MakefileThu Dec 2 06:05:44 2010
(r216114)
@@ -1,5 +1,5 @@
# $FreeBSD$
-SUBDIR=
+SUBDIR=cc_cubic
.include
Added: head/sys/modules/cc/cc_cubic/Makefile
==
--- /dev/null 00:00:00 1970 (empty, because file is newly added)
+++ head/sys/modules/cc/cc_cubic/Makefile Thu Dec 2 06:05:44 2010
(r216114)
@@ -0,0 +1,9 @@
+# $FreeBSD$
+
+.include
+
+.PATH: ${.CURDIR}/../../../netinet/cc
+KMOD= cc_cubic
+SRCS= cc_cubic.c
+
+.include
Added: head/sys/netinet/cc/cc_cubic.c
==
--- /dev/null 00:00:00 1970 (empty, because file is newly added)
+++ head/sys/netinet/cc/cc_cubic.c Thu Dec 2 06:05:44 2010
(r216114)
@@ -0,0 +1,396 @@
+/*-
+ * Copyright (c) 2008-2010 Lawrence Stewart
+ * Copyright (c) 2010 The FreeBSD Foundation
+ * All rights reserved.
+ *
+ * This software was developed by Lawrence Stewart while studying at the Centre
+ * for Advanced Internet Architectures, Swinburne University, made possible in
+ * part by a grant from the Cisco University Research Program Fund at Community
+ * Foundation Silicon Valley.
+ *
+ * Portions of this software were developed at the Centre for Advanced
+ * Internet Architectures, Swinburne University of Technology, Melbourne,
+ * Australia by David Hayes under sponsorship from the FreeBSD Foundation.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ *notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *notice, this list of conditions and the following disclaimer in the
+ *documentation and/or other materials provided with the distribution.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ */
+
+/*
+ * An implementation of the CUBIC congestion control algorithm for FreeBSD,
+ * based on the Internet Draft "draft-rhee-tcpm-cubic-02" by Rhee, Xu and Ha.
+ * Originally released as part of the NewTCP research project at Swinburne
+ * University's Centre for Advanced Internet Architectures, Melbourne,
+ * Australia, which was made possible in part by a grant from the Cisco
+ * University Research Program Fund at Community Foundation Silicon Valley.
More
+ * details are available at:
+ * http://caia.swin.edu.au/urp/newtcp/
+ */
+
+#include
+__FBSDID("$FreeBSD$");
+
+#include
+#include
+#include
+#include
+#include
+#include
+#include
+#include
+
+#include
+
+#include
+#include
+#include
+#include
+
+#include
+#include
+
+static voidcubic_a
svn commit: r216115 - in head/sys: modules/cc modules/cc/cc_htcp netinet/cc
Author: lstewart
Date: Thu Dec 2 06:40:21 2010
New Revision: 216115
URL: http://svn.freebsd.org/changeset/base/216115
Log:
Import a clean-room implementation of the experimental H-TCP congestion
control
algorithm based on the Internet-Draft "draft-leith-tcp-htcp-06.txt". It is
implemented as a kernel module compatible with the recently committed modular
congestion control framework.
H-TCP was designed to provide increased throughput in fast and long-distance
networks. It attempts to maintain fairness when competing with legacy NewReno
TCP in lower speed scenarios where NewReno is able to operate adequately. The
paper "H-TCP: A framework for congestion control in high-speed and
long-distance
networks" provides additional detail.
In collaboration with:David Hayes and
Grenville Armitage
Sponsored by: FreeBSD Foundation
Reviewed by: rpaulo (older patch from a few weeks ago)
MFC after:3 months
Added:
head/sys/modules/cc/cc_htcp/
head/sys/modules/cc/cc_htcp/Makefile (contents, props changed)
head/sys/netinet/cc/cc_htcp.c (contents, props changed)
Modified:
head/sys/modules/cc/Makefile
Modified: head/sys/modules/cc/Makefile
==
--- head/sys/modules/cc/MakefileThu Dec 2 06:05:44 2010
(r216114)
+++ head/sys/modules/cc/MakefileThu Dec 2 06:40:21 2010
(r216115)
@@ -1,5 +1,6 @@
# $FreeBSD$
-SUBDIR=cc_cubic
+SUBDIR=cc_cubic \
+ cc_htcp
.include
Added: head/sys/modules/cc/cc_htcp/Makefile
==
--- /dev/null 00:00:00 1970 (empty, because file is newly added)
+++ head/sys/modules/cc/cc_htcp/MakefileThu Dec 2 06:40:21 2010
(r216115)
@@ -0,0 +1,9 @@
+# $FreeBSD$
+
+.include
+
+.PATH: ${.CURDIR}/../../../netinet/cc
+KMOD= cc_htcp
+SRCS= cc_htcp.c
+
+.include
Added: head/sys/netinet/cc/cc_htcp.c
==
--- /dev/null 00:00:00 1970 (empty, because file is newly added)
+++ head/sys/netinet/cc/cc_htcp.c Thu Dec 2 06:40:21 2010
(r216115)
@@ -0,0 +1,521 @@
+/*-
+ * Copyright (c) 2007-2008
+ * Swinburne University of Technology, Melbourne, Australia
+ * Copyright (c) 2009-2010 Lawrence Stewart
+ * Copyright (c) 2010 The FreeBSD Foundation
+ * All rights reserved.
+ *
+ * This software was developed at the Centre for Advanced Internet
+ * Architectures, Swinburne University, by Lawrence Stewart and James Healy,
+ * made possible in part by a grant from the Cisco University Research Program
+ * Fund at Community Foundation Silicon Valley.
+ *
+ * Portions of this software were developed at the Centre for Advanced
+ * Internet Architectures, Swinburne University of Technology, Melbourne,
+ * Australia by David Hayes under sponsorship from the FreeBSD Foundation.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ *notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *notice, this list of conditions and the following disclaimer in the
+ *documentation and/or other materials provided with the distribution.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ */
+
+/*
+ * An implementation of the H-TCP congestion control algorithm for FreeBSD,
+ * based on the Internet Draft "draft-leith-tcp-htcp-06.txt" by Leith and
+ * Shorten. Originally released as part of the NewTCP research project at
+ * Swinburne University's Centre for Advanced Internet Architectures,
Melbourne,
+ * Australia, which was made possible in part by a grant from the Cisco
+ * University Research Program Fund at Community Foundation Silicon Valley.
More
+ * details are available at:
+ * http://caia.swin.edu.au/urp/newtcp/
+ */
+
+#include
+__FBSDID("$FreeBSD$");
+
+#include
+#include
+#include
+#include
+#include
+#include
+#include
+#include
+#includ
