FreeBSD

I have ported Matt’s patch to 4-STABLE however Matthew Seaman reports that there is still panic, while different, after the patch and I believe I should port Matt’s VM post-morti diagnostic tools to HEAD.

I have been suffered by the speed of GNOME because it usually gets “jitter”.

It seems that marcus@ have the same problem. I have tried HZ=100 but it seems that nothing was changed. Then I have disabled ACPI and it seems the issue was worked around.

What happend in ACPI, so? Still under investigate.

Matt has committed a critical VM fix and kindly reminded that FreeBSD has also the issue. I have ported the patch here and requested a review by Dr. Alan Cox. I’d like to commit it at once it is approved.

I have started to clean up certain code of FreeBSD userland I interest or use at company, to make them WARNS=6 clean. A patchset separated by categories would be posted on my home directory at freefall right after I have a preliminary test.

Ruslan has suggested that I should have used ``make universe’’ to test my patches and I’ll try this on some 1750 boxes - my laptop is too slow and its hard disk is too small to finish a full make universe build :-)

Today NetApplience folks has told me that Linux performance is twice better than FreeBSD when using em(4) driver, and they pointed out that the driver FreeBSD is using does not do a good job at this time. I have tried to tune some certain part of the kernel however the result is still not so good at the time I am writing.

http://people.freebsd.org/~delphij/patch-pppd

It seems that FreeBSD and NetBSD have a lot of difference on busdma interfaces and this has blocked me from porting OpenBSD’s firmware upload support on mpt(4). Will have another attempt this weekend.

Today, JINMEI Tatuya / 神明達哉 has submitted a patch against sys/kern/uipc_socket2.c, which is:

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14

Index: src/sys/kern/uipc_socket2.c
diff -u src/sys/kern/uipc_socket2.c:1.138 src/sys/kern/uipc_socket2.c:1.139
--- src/sys/kern/uipc_socket2.c:1.138	Tue Oct 19 15:13:30 2004
+++ src/sys/kern/uipc_socket2.c	Wed Oct 27 05:37:58 2004
@@ -1253,7 +1253,7 @@
 
 	if (CMSG_SPACE((u_int)size) > MCLBYTES)
 		return ((struct mbuf *) NULL);
-	if (CMSG_SPACE((u_int)size > MLEN))
+	if (CMSG_SPACE((u_int)size) > MLEN)
 		m = m_getcl(M_DONTWAIT, MT_CONTROL, 0);
 	else
 		m = m_get(M_DONTWAIT, MT_CONTROL);

I have fighted against the issue that LJ1000 can’t work correctly with FreeBSD however I have failed.

I got “ulpt0: Input/Output Error” when doing cat foo > /dev/ulpt0.

Of course, cups and friends won’t run with this state.

This problem exists in FreeBSD 5.3-RC, 4.10-STABLE. However, interestingly Windows works correctly.

So, why??

Today DragonFly has buryed GNU awk. G’bye, amen! :-D