Re: 64GB option broken in 2.4.2

2001-02-26 Thread Rico Tudor
Problem is not fixed with your patch. Debugging packet is http://patrec.com./rico/vger/diag002.tar.bz2 - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at

Re: 64GB option broken in 2.4.2

2001-02-26 Thread Andrea Arcangeli
On Mon, Feb 26, 2001 at 02:44:03AM -0600, Rico Tudor wrote: > > Hypthesis#2 The bounce buffer code in the Linus tree is known to be > > imperfect. Does 2.4.2ac3 do the same ? > > > No improvement. (In fact, 2.4.2ac3 breaks 3ware IDE RAID support.) The highmem changes in 2.4.2ac3 has a couple

Re: 64GB option broken in 2.4.2

2001-02-26 Thread Alan Cox
> No improvement. (In fact, 2.4.2ac3 breaks 3ware IDE RAID support.) Curiouser and curiouser. The 3ware does have some known problems but you'd see those equally on any SMP 1Ggig/4Gig/64Gig. > While operating this Thunder 2500 (Tyan motherboard, ServerWorks chipset) > is like walking on a

Re: 64GB option broken in 2.4.2

2001-02-26 Thread Rico Tudor
> Hypthesis#2 The bounce buffer code in the Linus tree is known to be > imperfect. Does 2.4.2ac3 do the same ? > No improvement. (In fact, 2.4.2ac3 breaks 3ware IDE RAID support.) While operating this Thunder 2500 (Tyan motherboard, ServerWorks chipset) is like walking on a minefield, the

Re: 64GB option broken in 2.4.2

2001-02-26 Thread Rico Tudor
Hypthesis#2 The bounce buffer code in the Linus tree is known to be imperfect. Does 2.4.2ac3 do the same ? No improvement. (In fact, 2.4.2ac3 breaks 3ware IDE RAID support.) While operating this Thunder 2500 (Tyan motherboard, ServerWorks chipset) is like walking on a minefield, the problem

Re: 64GB option broken in 2.4.2

2001-02-26 Thread Alan Cox
No improvement. (In fact, 2.4.2ac3 breaks 3ware IDE RAID support.) Curiouser and curiouser. The 3ware does have some known problems but you'd see those equally on any SMP 1Ggig/4Gig/64Gig. While operating this Thunder 2500 (Tyan motherboard, ServerWorks chipset) is like walking on a

Re: 64GB option broken in 2.4.2

2001-02-26 Thread Andrea Arcangeli
On Mon, Feb 26, 2001 at 02:44:03AM -0600, Rico Tudor wrote: Hypthesis#2 The bounce buffer code in the Linus tree is known to be imperfect. Does 2.4.2ac3 do the same ? No improvement. (In fact, 2.4.2ac3 breaks 3ware IDE RAID support.) The highmem changes in 2.4.2ac3 has a couple of bugs,

Re: 64GB option broken in 2.4.2

2001-02-26 Thread Rico Tudor
Problem is not fixed with your patch. Debugging packet is http://patrec.com./rico/vger/diag002.tar.bz2 - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at

Re: 64GB option broken in 2.4.2

2001-02-25 Thread Alan Cox
> Hypothesis: > Code to handle PAE has buggy spinlock management. Hypthesis#2 The bounce buffer code in the Linus tree is known to be imperfect. Does 2.4.2ac3 do the same ? Alan - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL

64GB option broken in 2.4.2

2001-02-25 Thread Rico Tudor
Subject: 64GB option broken in 2.4.2 Problem: The 64GB option causes processes to spin in the kernel. Setup: Linux 2.4.2 Slack current 20-feb-2001 ServerWorks III HE 4GB main memory dual 933 Piii: SL47Q (stepping 3), SL4KK (stepping 6) gcc

64GB option broken in 2.4.2

2001-02-25 Thread Rico Tudor
Subject: 64GB option broken in 2.4.2 Problem: The 64GB option causes processes to spin in the kernel. Setup: Linux 2.4.2 Slack current 20-feb-2001 ServerWorks III HE 4GB main memory dual 933 Piii: SL47Q (stepping 3), SL4KK (stepping 6) gcc

Re: 64GB option broken in 2.4.2

2001-02-25 Thread Alan Cox
Hypothesis: Code to handle PAE has buggy spinlock management. Hypthesis#2 The bounce buffer code in the Linus tree is known to be imperfect. Does 2.4.2ac3 do the same ? Alan - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL