conventional memory replacement
in [Home Built PC]
|
Prev: Need power cord for Compaq laptop
Next: International Journal of Electronics, Information and Systems (IJEIS) Call for Paper
From: cisz on 7 Feb 2010 19:24 Recently, our ethernet card came slightly loose, (it's a card for a tower that's in a smaller computer and can't be screwed in). I mistakenly tried to push it back in without turning off the computer and it seems to have blown some of the conventional memory because we're now getting a message that win xp requires at least 512k of low memory and the computer won't boot. I assume there's a chip for the conventional memory that needs to be replaced? It's a gateway computer with 800MHz cpu. Thanks.
From: Paul on 7 Feb 2010 20:10 cisz wrote: > Recently, our ethernet card came slightly loose, (it's a card for a tower > that's in a smaller computer and can't be screwed in). I mistakenly tried to > push it back in without turning off the computer and it seems to have blown > some of the conventional memory because we're now getting a message that win > xp requires at least 512k of low memory and the computer won't boot. > > I assume there's a chip for the conventional memory that needs to be > replaced? It's a gateway computer with 800MHz cpu. > > Thanks. > First, consider the architecture of your machine. I don't think the memory was physically damaged. It is "too far away" from where you shorted something. (That is, unless the two pieces of silicon shared a common voltage rail. There may be a path they both use in that respect. Otherwise, a fault traveling though that much silicon is less likely.) Northbridge ---- single_channel_memory_bus_for_DIMMs | (Not likely to be broken up here.) | Southbridge | PCI bus | X <--- something shorted here | X I agree that something is not right with your system. The BIOS passes tables to the OS at boot time, and the BIOS is now reporting something bogus to the OS. For example, a certain BIOS call, reports areas of memory that are reserved. It could be, that the BIOS is no longer reporting the correct hardware information. Why that would be, I haven't a clue. Download a copy of memtest86+. It respects the BIOS E820 memory reservation information, and when memtest86+ starts, you can see what amount of non-conventional memory it claims is available. That may give a hint as to how broken the box is. http://www.memtest.org/ (a version to prepare a blank floppy is available) http://www.memtest86.com/tech.html (describes using E820 reservation info) (the two versions of memtest share a common ancestor) (Some of the history of the two versions of memtest) http://en.wikipedia.org/wiki/Memtest86%2B (Article on "conventional memory") http://en.wikipedia.org/wiki/Conventional_memory Maybe you can get some idea what is going on, by using one of those memory test programs. If no memory was available, for example, for the BIOS to use during POST, you would be getting beep codes, and no attempt to boot Windows at all. At least one of the DIMMs on the main memory bus, must be working. And there is a good chance, you can use memtest to learn more. HTH, Paul
From: cisz on 8 Feb 2010 04:19 "Paul" <nospam(a)needed.com> wrote in message news:hkno9p$sp$1(a)news.eternal-september.org... > cisz wrote: >> Recently, our ethernet card came slightly loose, (it's a card for a tower >> that's in a smaller computer and can't be screwed in). I mistakenly tried >> to push it back in without turning off the computer and it seems to have >> blown some of the conventional memory because we're now getting a message >> that win xp requires at least 512k of low memory and the computer won't >> boot. >> >> I assume there's a chip for the conventional memory that needs to be >> replaced? It's a gateway computer with 800MHz cpu. >> >> Thanks. > > First, consider the architecture of your machine. I don't think the > memory was physically damaged. It is "too far away" from where you > shorted something. (That is, unless the two pieces of silicon shared > a common voltage rail. There may be a path they both use in that > respect. Otherwise, a fault traveling though that much silicon is > less likely.) > > Northbridge ---- single_channel_memory_bus_for_DIMMs > | (Not likely to be broken up here.) > | > Southbridge > | > PCI bus > | > X <--- something shorted here > | > X > > I agree that something is not right with your system. The BIOS > passes tables to the OS at boot time, and the BIOS is now reporting > something bogus to the OS. For example, a certain BIOS call, reports > areas of memory that are reserved. It could be, that the BIOS is > no longer reporting the correct hardware information. > > Why that would be, I haven't a clue. > > Download a copy of memtest86+. It respects the BIOS E820 memory > reservation information, and when memtest86+ starts, you can > see what amount of non-conventional memory it claims is available. > That may give a hint as to how broken the box is. > > http://www.memtest.org/ (a version to prepare a blank floppy is > available) > > http://www.memtest86.com/tech.html (describes using E820 reservation > info) > (the two versions of memtest share a > common > ancestor) > > (Some of the history of the two versions of memtest) > > http://en.wikipedia.org/wiki/Memtest86%2B > > (Article on "conventional memory") > > http://en.wikipedia.org/wiki/Conventional_memory > > Maybe you can get some idea what is going on, by using one of those > memory test programs. > > If no memory was available, for example, for the BIOS to use during > POST, you would be getting beep codes, and no attempt to boot Windows > at all. At least one of the DIMMs on the main memory bus, must be working. > And there is a good chance, you can use memtest to learn more. > > HTH, > Paul Thanks. I tried using memtest and, (although I didn't run it for a real long time), it didn't show any errors. Is conventional memory taken from the RAM? (the memory sticks - sdram) The computer runs the memory check at the beginning and isn't finding any problems. I was able to repartition the hard drive using fdisk and reformat. I tried installing win98 and got an error saying that 7340032 bytes of ava memory is needed to install it.
From: Peter on 8 Feb 2010 07:19 In article <hkokv102k8v(a)news1.newsguy.com>, cisz.x(a)yahoo.com says... > > "Paul" <nospam(a)needed.com> wrote in message > news:hkno9p$sp$1(a)news.eternal-september.org... > > cisz wrote: > > I tried using memtest and, (although I didn't run it for a real long time), > it didn't show any errors. > > Is conventional memory taken from the RAM? (the memory sticks - sdram) The > computer runs > the memory check at the beginning and isn't finding any problems. > > I was able to repartition the hard drive using fdisk and reformat. I tried > installing win98 > and got an error saying that 7340032 bytes of ava memory is needed to > install it. > > > Does the computer show how much memory it has at POST? It might be worth resetting the BIOS to defaults. -- Pete Ives Remove All_stRESS before sending me an email
From: cisz on 10 Feb 2010 03:27
"Paul" <nospam(a)needed.com> wrote in message news:hkt7dl$r8c$1(a)news.eternal-september.org... > You could compare the following, to what you find, to see if there is > some significant difference. > > I booted a Knoppix 5.0.1 CD, and using the "dmesg" tool, and this is what > it reports > for my motherboard. I tried a later version of Knoppix, and it didn't dump > the > E820 reserved memory areas like this CD does. 0xA0000 is the hex > equivalent > of 640K, and in this case, the "0x9FC00 usable" is the 639K mark. So it > seems > virtually the entire DOS area is listed as available. > > The release date of Knoppix 5.0.1 is listed as "2 June 2006", so that > seems to be > a pretty close match for the "May 10" reported in the kernel message. > > http://en.wikipedia.org/wiki/Knoppix > > The other part of the BIOS communications to the OS, is the ACPI > information. > On an older motherboard, I suppose that could be missing, but then the > OS installation could not support ACPI power management if that was the > case. > > ******* > Linux version 2.6.17 (root(a)Knoppix) (gcc version 4.0.4 20060507 > (prerelease) > (Debian 4.0.3-3)) #4 SMP PREEMPT Wed May 10 13:53:45 CEST 2006 > > BIOS-provided physical RAM map: > BIOS-e820: 0000000000000000 - 000000000009fc00 (usable) > BIOS-e820: 000000000009fc00 - 00000000000a0000 (reserved) > BIOS-e820: 00000000000e6000 - 0000000000100000 (reserved) > BIOS-e820: 0000000000100000 - 000000007ffb0000 (usable) > BIOS-e820: 000000007ffb0000 - 000000007ffc0000 (ACPI data) > BIOS-e820: 000000007ffc0000 - 000000007fff0000 (ACPI NVS) > BIOS-e820: 000000007fff0000 - 0000000080000000 (reserved) > BIOS-e820: 00000000fec00000 - 00000000fec01000 (reserved) > BIOS-e820: 00000000fee00000 - 00000000fee01000 (reserved) > BIOS-e820: 00000000ff380000 - 0000000100000000 (reserved) > 1151MB HIGHMEM available. > 896MB LOWMEM available. > found SMP MP-table at 000ff780 > On node 0 totalpages: 524208 > DMA zone: 4096 pages, LIFO batch:0 > Normal zone: 225280 pages, LIFO batch:31 > HighMem zone: 294832 pages, LIFO batch:31 > DMI present. > ACPI: RSDP (v000 ACPIAM ) @ 0x000f8860 > ACPI: RSDT (v001 A_M_I OEMRSDT 0x02000904 MSFT 0x00000097) @ 0x7ffb0000 > ACPI: FADT (v002 A M I OEMFACP 0x12000601 MSFT 0x00000097) @ 0x7ffb0200 > ACPI: MADT (v001 A_M_I OEMAPIC 0x02000904 MSFT 0x00000097) @ 0x7ffb0390 > ACPI: MCFG (v001 A_M_I OEMMCFG 0x02000904 MSFT 0x00000097) @ 0x7ffb0410 > ACPI: OEMB (v001 A_M_I AMI_OEM 0x02000904 MSFT 0x00000097) @ 0x7ffc0040 > ACPI: AAFT (v001 A_M_I OEMAAFT 0x02000904 MSFT 0x00000097) @ 0x7ffb4fc0 > ACPI: DSDT (v001 4CDS2 4CDS2213 0x00000213 INTL 0x02002026) @ 0x00000000 > ACPI: PM-Timer IO Port: 0x808 > ACPI: Local APIC address 0xfee00000 > ACPI: LAPIC (acpi_id[0x01] lapic_id[0x00] enabled) > Processor #0 6:15 APIC version 20 > ACPI: LAPIC (acpi_id[0x02] lapic_id[0x01] enabled) > Processor #1 6:15 APIC version 20 > ACPI: LAPIC (acpi_id[0x03] lapic_id[0x82] disabled) > ACPI: LAPIC (acpi_id[0x04] lapic_id[0x83] disabled) > ACPI: IOAPIC (id[0x02] address[0xfec00000] gsi_base[0]) > IOAPIC[0]: apic_id 2, version 3, address 0xfec00000, GSI 0-23 > ACPI: IOAPIC (id[0x03] address[0xfecc0000] gsi_base[24]) > IOAPIC[1]: apic_id 3, version 3, address 0xfecc0000, GSI 24-47 > ACPI: INT_SRC_OVR (bus 0 bus_irq 0 global_irq 2 dfl dfl) > ACPI: INT_SRC_OVR (bus 0 bus_irq 9 global_irq 9 low level) > ACPI: IRQ0 used by override. > ACPI: IRQ2 used by override. > ACPI: IRQ9 used by override. > Enabling APIC mode: Flat. Using 2 I/O APICs > Using ACPI (MADT) for SMP configuration information > ******* > > Anyway, that is as close as I can get, to understanding what the > BIOS is passing to the OS. And what may be preventing a normal > boot on your machine. > > Paul I was able to reformat the harddrive using gparted. I'll try doing the "dmesg" and check what it says. I was thinking of trying to install debian. (I've got debian on a different computer). |