Re: 2.6.27 mtrr fixes do not work when X starts

Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]

From: Helge Hafting

Subject: Re: 2.6.27 mtrr fixes do not work when X starts

Date: Monday, August 25, 2008 - 4:10 am

Yinghai Lu wrote:
quoted text> On Fri, Aug 22, 2008 at 3:11 AM, Helge Hafting
>   
[old bug report removed]
quoted text> please boot with "mtrr_spare_reg_nr=3 debug"
>   
I did that, and got:
dmesg:
Command line: root=/dev/sda2 ro mtrr_spare_reg=3 debug
KERNEL supported cpus:
  Intel GenuineIntel
  AMD AuthenticAMD
  Centaur CentaurHauls
BIOS-provided physical RAM map:
 BIOS-e820: 0000000000000000 - 000000000009f000 (usable)
 BIOS-e820: 000000000009f000 - 00000000000a0000 (reserved)
 BIOS-e820: 0000000000100000 - 000000007fe80400 (usable)
 BIOS-e820: 000000007fe80400 - 0000000080000000 (reserved)
 BIOS-e820: 00000000f4000000 - 00000000f8000000 (reserved)
 BIOS-e820: 00000000fec00000 - 00000000fec10000 (reserved)
 BIOS-e820: 00000000fed18000 - 00000000fed1c000 (reserved)
 BIOS-e820: 00000000fed20000 - 00000000fed90000 (reserved)
 BIOS-e820: 00000000feda0000 - 00000000feda6000 (reserved)
 BIOS-e820: 00000000fee00000 - 00000000fee10000 (reserved)
 BIOS-e820: 00000000ffe00000 - 0000000100000000 (reserved)
last_pfn = 0x7fe80 max_arch_pfn = 0x3ffffffff
x86 PAT enabled: cpu 0, old 0x7040600070406, new 0x7010600070106
total RAM coverred: 2047M
 gran_size: 1M  chunk_size: 1M  num_reg: 8      lose RAM: 7M
 gran_size: 1M  chunk_size: 2M  num_reg: 8      lose RAM: 7M
 gran_size: 1M  chunk_size: 4M  num_reg: 8      lose RAM: 7M
 gran_size: 1M  chunk_size: 8M  num_reg: 8      lose RAM: 7M
*BAD* gran_size: 1M     chunk_size: 16M         num_reg: 8      lose 
RAM: -1M
 gran_size: 1M  chunk_size: 32M         num_reg: 8      lose RAM: 0M
 gran_size: 1M  chunk_size: 64M         num_reg: 7      lose RAM: 0M
 gran_size: 1M  chunk_size: 128M        num_reg: 6      lose RAM: 0M
 gran_size: 1M  chunk_size: 256M        num_reg: 5      lose RAM: 0M
 gran_size: 1M  chunk_size: 512M        num_reg: 4      lose RAM: 0M
 gran_size: 1M  chunk_size: 1024M       num_reg: 3      lose RAM: 0M
 gran_size: 1M  chunk_size: 2048M       num_reg: 2      lose RAM: 0M
 gran_size: 1M  chunk_size: 4096M       num_reg: 8      lose RAM: 7M
 gran_size: 2M  chunk_size: 2M  num_reg: 8      lose RAM: 7M
 gran_size: 2M  chunk_size: 4M  num_reg: 8      lose RAM: 7M
 gran_size: 2M  chunk_size: 8M  num_reg: 8      lose RAM: 7M
*BAD* gran_size: 2M     chunk_size: 16M         num_reg: 8      lose 
RAM: -1M
 gran_size: 2M  chunk_size: 32M         num_reg: 8      lose RAM: 1M
 gran_size: 2M  chunk_size: 64M         num_reg: 7      lose RAM: 1M
 gran_size: 2M  chunk_size: 128M        num_reg: 6      lose RAM: 1M
 gran_size: 2M  chunk_size: 256M        num_reg: 5      lose RAM: 1M
 gran_size: 2M  chunk_size: 512M        num_reg: 4      lose RAM: 1M
 gran_size: 2M  chunk_size: 1024M       num_reg: 3      lose RAM: 1M
 gran_size: 2M  chunk_size: 2048M       num_reg: 2      lose RAM: 1M
 gran_size: 2M  chunk_size: 4096M       num_reg: 8      lose RAM: 7M
 gran_size: 4M  chunk_size: 4M  num_reg: 8      lose RAM: 7M
 gran_size: 4M  chunk_size: 8M  num_reg: 8      lose RAM: 7M
*BAD* gran_size: 4M     chunk_size: 16M         num_reg: 8      lose 
RAM: -1M
 gran_size: 4M  chunk_size: 32M         num_reg: 8      lose RAM: 3M
 gran_size: 4M  chunk_size: 64M         num_reg: 7      lose RAM: 3M
 gran_size: 4M  chunk_size: 128M        num_reg: 6      lose RAM: 3M
 gran_size: 4M  chunk_size: 256M        num_reg: 5      lose RAM: 3M
 gran_size: 4M  chunk_size: 512M        num_reg: 4      lose RAM: 3M
 gran_size: 4M  chunk_size: 1024M       num_reg: 3      lose RAM: 3M
 gran_size: 4M  chunk_size: 2048M       num_reg: 2      lose RAM: 3M
 gran_size: 4M  chunk_size: 4096M       num_reg: 8      lose RAM: 7M
 gran_size: 8M  chunk_size: 8M  num_reg: 8      lose RAM: 7M
 gran_size: 8M  chunk_size: 16M         num_reg: 8      lose RAM: 7M
 gran_size: 8M  chunk_size: 32M         num_reg: 8      lose RAM: 7M
 gran_size: 8M  chunk_size: 64M         num_reg: 7      lose RAM: 7M
 gran_size: 8M  chunk_size: 128M        num_reg: 6      lose RAM: 7M
 gran_size: 8M  chunk_size: 256M        num_reg: 5      lose RAM: 7M
 gran_size: 8M  chunk_size: 512M        num_reg: 4      lose RAM: 7M
 gran_size: 8M  chunk_size: 1024M       num_reg: 3      lose RAM: 7M
 gran_size: 8M  chunk_size: 2048M       num_reg: 2      lose RAM: 7M
 gran_size: 8M  chunk_size: 4096M       num_reg: 8      lose RAM: 7M
 gran_size: 16M         chunk_size: 16M         num_reg: 7      lose 
RAM: 15M
 gran_size: 16M         chunk_size: 32M         num_reg: 7      lose 
RAM: 15M
 gran_size: 16M         chunk_size: 64M         num_reg: 7      lose 
RAM: 15M
 gran_size: 16M         chunk_size: 128M        num_reg: 6      lose 
RAM: 15M
 gran_size: 16M         chunk_size: 256M        num_reg: 5      lose 
RAM: 15M
 gran_size: 16M         chunk_size: 512M        num_reg: 4      lose 
RAM: 15M
 gran_size: 16M         chunk_size: 1024M       num_reg: 3      lose 
RAM: 15M
 gran_size: 16M         chunk_size: 2048M       num_reg: 2      lose 
RAM: 15M
 gran_size: 16M         chunk_size: 4096M       num_reg: 7      lose 
RAM: 15M
 gran_size: 32M         chunk_size: 32M         num_reg: 6      lose 
RAM: 31M
 gran_size: 32M         chunk_size: 64M         num_reg: 6      lose 
RAM: 31M
 gran_size: 32M         chunk_size: 128M        num_reg: 6      lose 
RAM: 31M
 gran_size: 32M         chunk_size: 256M        num_reg: 5      lose 
RAM: 31M
 gran_size: 32M         chunk_size: 512M        num_reg: 4      lose 
RAM: 31M
 gran_size: 32M         chunk_size: 1024M       num_reg: 3      lose 
RAM: 31M
 gran_size: 32M         chunk_size: 2048M       num_reg: 2      lose 
RAM: 31M
 gran_size: 32M         chunk_size: 4096M       num_reg: 6      lose 
RAM: 31M
 gran_size: 64M         chunk_size: 64M         num_reg: 5      lose 
RAM: 63M
 gran_size: 64M         chunk_size: 128M        num_reg: 5      lose 
RAM: 63M
 gran_size: 64M         chunk_size: 256M        num_reg: 5      lose 
RAM: 63M
 gran_size: 64M         chunk_size: 512M        num_reg: 4      lose 
RAM: 63M
 gran_size: 64M         chunk_size: 1024M       num_reg: 3      lose 
RAM: 63M
 gran_size: 64M         chunk_size: 2048M       num_reg: 2      lose 
RAM: 63M
 gran_size: 64M         chunk_size: 4096M       num_reg: 5      lose 
RAM: 63M
 gran_size: 128M        chunk_size: 128M        num_reg: 4      lose 
RAM: 127M
 gran_size: 128M        chunk_size: 256M        num_reg: 4      lose 
RAM: 127M
 gran_size: 128M        chunk_size: 512M        num_reg: 4      lose 
RAM: 127M
 gran_size: 128M        chunk_size: 1024M       num_reg: 3      lose 
RAM: 127M
 gran_size: 128M        chunk_size: 2048M       num_reg: 2      lose 
RAM: 127M
 gran_size: 128M        chunk_size: 4096M       num_reg: 4      lose 
RAM: 127M
 gran_size: 256M        chunk_size: 256M        num_reg: 3      lose 
RAM: 255M
 gran_size: 256M        chunk_size: 512M        num_reg: 3      lose 
RAM: 255M
 gran_size: 256M        chunk_size: 1024M       num_reg: 3      lose 
RAM: 255M
 gran_size: 256M        chunk_size: 2048M       num_reg: 2      lose 
RAM: 255M
 gran_size: 256M        chunk_size: 4096M       num_reg: 3      lose 
RAM: 255M
 gran_size: 512M        chunk_size: 512M        num_reg: 2      lose 
RAM: 511M
 gran_size: 512M        chunk_size: 1024M       num_reg: 2      lose 
RAM: 511M
 gran_size: 512M        chunk_size: 2048M       num_reg: 2      lose 
RAM: 511M
 gran_size: 512M        chunk_size: 4096M       num_reg: 2      lose 
RAM: 511M
 gran_size: 1024M       chunk_size: 1024M       num_reg: 1      lose 
RAM: 1023M
 gran_size: 1024M       chunk_size: 2048M       num_reg: 1      lose 
RAM: 1023M
 gran_size: 1024M       chunk_size: 4096M       num_reg: 1      lose 
RAM: 1023M
 gran_size: 2048M       chunk_size: 2048M       num_reg: 0      lose 
RAM: 2047M
 gran_size: 2048M       chunk_size: 4096M       num_reg: 0      lose 
RAM: 2047M
Found optimal setting for mtrr clean up
gran_size: 1M   chunk_size: 64M         num_reg: 7      lose RAM: 0M
range0: 0000000000000000 - 000000007c000000
Setting variable MTRR 0, base: 0MB, range: 1024MB, type WB
Setting variable MTRR 1, base: 1024MB, range: 512MB, type WB
Setting variable MTRR 2, base: 1536MB, range: 256MB, type WB
Setting variable MTRR 3, base: 1792MB, range: 128MB, type WB
Setting variable MTRR 4, base: 1920MB, range: 64MB, type WB
range: 000000007c000000 - 0000000080000000
Setting variable MTRR 5, base: 1984MB, range: 64MB, type WB
hole: 000000007ff00000 - 0000000080000000
Setting variable MTRR 6, base: 2047MB, range: 1MB, type UC
x86 PAT enabled: cpu 0, old 0x7040600070406, new 0x7010600070106
After WB checking
MTRR MAP PFN: 0000000000000000 - 0000000000080000
After UC checking
MTRR MAP PFN: 0000000000000000 - 000000000007ff00
After sorting
MTRR MAP PFN: 0000000000000000 - 000000000007ff00
init_memory_mapping
 0000000000 - 007fe00000 page 2M
 007fe00000 - 007fe80000 page 4k
kernel direct mapping tables up to 7fe80000 @ 8000-c000
last_map_addr: 7fe80000 end: 7fe80000
DMI 2.4 present.

Xorg.0.log:
(II) Loading sub module "int10"
(II) LoadModule: "int10"
(II) Reloading /usr/lib/xorg/modules//libint10.so
(II) VESA(0): initializing int10
(II) VESA(0): Primary V_BIOS segment is: 0xc000
(II) VESA(0): VESA BIOS detected
(II) VESA(0): VESA VBE Version 3.0
(II) VESA(0): VESA VBE Total Mem: 14336 kB
(II) VESA(0): VESA VBE OEM: NVIDIA
(II) VESA(0): VESA VBE OEM Software Rev: 96.134
(II) VESA(0): VESA VBE OEM Vendor: NVIDIA Corporation
(II) VESA(0): VESA VBE OEM Product: G86 Board - dawson0
(II) VESA(0): VESA VBE OEM Product Rev: Chip Rev  
(II) VESA(0): Splitting WC range: base: 0xfb000000, size: 0xe00000
(II) VESA(0): Splitting WC range: base: 0xfb800000, size: 0x600000
(==) VESA(0): Write-combining range (0xfbc00000,0x200000)
(WW) VESA(0): Failed to set up write-combining range (0xfb800000,0x600000)
(WW) VESA(0): Failed to set up write-combining range (0xfb000000,0xe00000)
(II) VESA(0): virtual address = 0x7f3ccaf43000,
        physical address = 0xfb000000, size = 14680064

Just like last time, this lack of mtrr is reflected in dmesg:
mtrr: no more MTRRs available
mtrr: no more MTRRs available

$ cat /proc/mtrr
reg00: base=0x00000000 (   0MB), size=1024MB: write-back, count=1
reg01: base=0x40000000 (1024MB), size= 512MB: write-back, count=1
reg02: base=0x60000000 (1536MB), size= 256MB: write-back, count=1
reg03: base=0x70000000 (1792MB), size= 128MB: write-back, count=1
reg04: base=0x78000000 (1920MB), size=  64MB: write-back, count=1
reg05: base=0x7c000000 (1984MB), size=  64MB: write-back, count=1
reg06: base=0x7ff00000 (2047MB), size=   1MB: uncachable, count=1
reg07: base=0xfbc00000 (4028MB), size=   2MB: write-combining, count=1


The problem is still there, I hope the output is useful.
I can test experimental patches if necessary.

Helge Hafting
--
unsubscribe noticeTo unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]

Messages in current thread:

2.6.27-rc1 mtrr fixes do not work, Alexander Huemer, (Wed Aug 6, 7:05 am)

Re: 2.6.27-rc1 mtrr fixes do not work, Rafael J. Wysocki, (Thu Aug 7, 1:14 pm)

RE: 2.6.27-rc1 mtrr fixes do not work, Pallipadi, Venkatesh, (Thu Aug 7, 1:15 pm)

Re: 2.6.27-rc1 mtrr fixes do not work, Yinghai Lu, (Thu Aug 7, 1:23 pm)

Re: 2.6.27-rc1 mtrr fixes do not work, Yinghai Lu, (Thu Aug 7, 1:26 pm)

Re: 2.6.27-rc1 mtrr fixes do not work, Alexander Huemer, (Thu Aug 7, 3:14 pm)

Re: 2.6.27-rc1 mtrr fixes do not work, Yinghai Lu, (Thu Aug 7, 4:16 pm)

Re: 2.6.27-rc1 mtrr fixes do not work, Yinghai Lu, (Thu Aug 7, 4:17 pm)

Re: 2.6.27-rc1 mtrr fixes do not work, Alexander Huemer, (Thu Aug 7, 4:30 pm)

Re: 2.6.27-rc1 mtrr fixes do not work, Yinghai Lu, (Thu Aug 7, 4:58 pm)

Re: 2.6.27-rc1 mtrr fixes do not work, Alexander Huemer, (Thu Aug 7, 5:28 pm)

Re: 2.6.27-rc1 mtrr fixes do not work, Alexander Huemer, (Thu Aug 14, 7:09 am)

Re: 2.6.27-rc1 mtrr fixes do not work, Yinghai Lu, (Thu Aug 14, 10:31 am)

Re: 2.6.27 mtrr fixes do not work, Alexander Huemer, (Thu Aug 21, 7:52 am)

Re: 2.6.27 mtrr fixes do not work, Yinghai Lu, (Thu Aug 21, 8:27 am)

Re: 2.6.27 mtrr fixes do not work, Alexander Huemer, (Thu Aug 21, 11:44 am)

Re: 2.6.27 mtrr fixes do not work when X starts, Helge Hafting, (Fri Aug 22, 3:11 am)

Re: 2.6.27 mtrr fixes do not work when X starts, Yinghai Lu, (Fri Aug 22, 10:24 am)

Re: 2.6.27 mtrr fixes do not work when X starts, Helge Hafting, (Mon Aug 25, 4:10 am)

Re: 2.6.27 mtrr fixes do not work when X starts, Yinghai Lu, (Mon Aug 25, 1:08 pm)

Re: 2.6.27 mtrr fixes do not work when X starts, Alexander Huemer, (Mon Aug 25, 1:19 pm)

Re: 2.6.27 mtrr fixes do not work when X starts, Yinghai Lu, (Mon Aug 25, 1:22 pm)

Re: 2.6.27 mtrr fixes do not work when X starts, Helge Hafting, (Tue Aug 26, 2:32 am)

Re: 2.6.27 mtrr fixes do not work when X starts, Yinghai Lu, (Tue Aug 26, 9:43 am)

Re: 2.6.27 mtrr fixes do not work when X starts, Helge Hafting, (Wed Aug 27, 5:53 am)

Re: 2.6.27 mtrr fixes do not work when X starts, Andi Kleen, (Wed Aug 27, 6:03 am)


linux-kernel:
David Howells	[PATCH] KEYS: Use the variable 'key' in keyctl_describe_key()
Greg Kroah-Hartman	[PATCH 17/36] sysdev: detect multiple driver registrations
Sam Ravnborg	Re: [PATCH] kbuild: fix make V=1
Greg Kroah-Hartman	[PATCH 09/36] driver core: register_memory/unregister_memory clean ups and bugfix
Pierre Ossman	Re: sdio: enhance IO_RW_EXTENDED support
git:
Mark Junker	git on MacOSX and files with decomposed utf-8 file names
Johannes Schindelin	Re: error: cannot lock ref 'refs/remotes/origin/*'
Pat Thoyts	[PATCH] git-gui: use themed tk widgets with Tk 8.5
Johannes Schindelin	Re: [PATCH 2/2] git-svn: support fetch with autocrlf on
Jonathan Nieder	Re: [PATCH v2] git-send-email.perl: fix In-Reply-To for second and subsequent patc...
linux-netdev:
David Miller	Re: [PATCH 32/53] netns xfrm: finding policy in netns
Jean-Louis Dupond	Re: tg3 driver not advertising 1000mbit
Jan Engelhardt	[PATCH 1/3] net: tcp: make hybla selectable as default congestion module
Daniel Schaffrath	Re: tcp bw in 2.6
Matt Mackall	Re: [regression] nf_iterate(), BUG: unable to handle kernel NULL pointer dereference
git-commits-head:
Linux Kernel Mailing List	ipv6: fix an oops when force unload ipv6 module
Linux Kernel Mailing List	tracing: protect reader of cmdline output
Linux Kernel Mailing List	KVM: VMX: Clear CR4.VMXE in hardware_disable
Linux Kernel Mailing List	USB: set correct configuration in probe of ti_usb_3410_5052
Linux Kernel Mailing List	SiS DRM: fix the memory allocator if the SIS FB is built as a module
openbsd-misc:
Stas Miasnikou	Re: Another question: device naming convention
Darrin Chandler	Re: That whole "Linux stealing our code" thing
Community First Financial	Teacher A+ Loan
Jan Stary	Re: audio recording levels
Andrej Elizarov	Re: Web Browsers