I just used Uwe's Tyan S1846 tree and everything *just* worked. :-)
However, I don't mean everything worked.
I've attached a log file of the boot for those curious.
Here's Failure #1 - DRAM Verify Failure --- Begin Snip --- DRAM filled DRAM verify: 00000000-04000000 00000000 00010000 00020000 00030000 00040000 00050000 00060000 00070000 00080000 00090000 000a0000 F ail: @0x000a0000 Read value=0xffffffff Fail: @0x000a0004 Read value=0xffffffff Fail: @0x000a0008 Read value=0xffffffff ... (repeated millions of times) ... Fail: @0x000a0400 Read value=0xffffffff Aborting. 000a0400 DRAM did _NOT_ verify! Done. --- End Snip ---
Here's Failure #2 - Cannot Find PCI Bus Operations --- Begin Snip --- Enumerating buses... scan_static_bus for Root Device Finding PCI configuration type. PCI: Using configuration type 1 PCI_DOMAIN: 0000 enabled PCI_DOMAIN: 0000 scanning... PCI: pci_scan_bus for bus 00 PCI: 00:00.0 Cannot find pci bus operations --- End Snip ---
-- Roger http://www.eskimo.com/~roger/index.html Key fingerprint = 8977 A252 2623 F567 70CD 1261 640F C963 1005 1D61
Thu Apr 26 23:54:57 PDT 2007
roger wrote:
I just used Uwe's Tyan S1846 tree and everything *just* worked. :-)
However, I don't mean everything worked.
I've attached a log file of the boot for those curious.
Here's Failure #1 - DRAM Verify Failure --- Begin Snip --- DRAM filled DRAM verify: 00000000-04000000 00000000 00010000 00020000 00030000 00040000 00050000 00060000 00070000 00080000 00090000 000a0000 F ail: @0x000a0000 Read value=0xffffffff Fail: @0x000a0004 Read value=0xffffffff Fail: @0x000a0008 Read value=0xffffffff ... (repeated millions of times) ... Fail: @0x000a0400 Read value=0xffffffff Aborting. 000a0400 DRAM did _NOT_ verify! Done. --- End Snip ---
Known issue, I figured you already knew. Ram init is very much broken on the i440bx northbridge.
-Corey
* Corey Osgood corey_osgood@verizon.net [070427 09:09]:
roger wrote:
I just used Uwe's Tyan S1846 tree and everything *just* worked. :-)
However, I don't mean everything worked.
I've attached a log file of the boot for those curious.
Here's Failure #1 - DRAM Verify Failure --- Begin Snip --- DRAM filled DRAM verify: 00000000-04000000 00000000 00010000 00020000 00030000 00040000 00050000 00060000 00070000 00080000 00090000 000a0000 F ail: @0x000a0000 Read value=0xffffffff Fail: @0x000a0004 Read value=0xffffffff Fail: @0x000a0008 Read value=0xffffffff ... (repeated millions of times) ... Fail: @0x000a0400 Read value=0xffffffff Aborting. 000a0400 DRAM did _NOT_ verify! Done. --- End Snip ---
Known issue, I figured you already knew. Ram init is very much broken on the i440bx northbridge.
0xa0000 looks more like video memory to me.
try checking 0-0x9ffff and 100000-TOM
On Fri, 2007-04-27 at 00:01 -0700, roger wrote:
I just used Uwe's Tyan S1846 tree and everything *just* worked. :-)
However, I don't mean everything worked.
I've attached a log file of the boot for those curious.
Here's Failure #1 - DRAM Verify Failure --- Begin Snip --- DRAM filled DRAM verify: 00000000-04000000 00000000 00010000 00020000 00030000 00040000 00050000 00060000 00070000 00080000 00090000 000a0000 F ail: @0x000a0000 Read value=0xffffffff Fail: @0x000a0004 Read value=0xffffffff Fail: @0x000a0008 Read value=0xffffffff ...
Known issue, I figured you already knew. Ram init is very much broken
on
the i440bx northbridge.
0xa0000 looks more like video memory to me.
try checking 0-0x9ffff and 100000-TOM
[Attached full log of LinuxBIOSv2 output.]
Error #1 of 2 (DRAM Succeeds):
Ok. I've set the following in tyan/s1846/auto.c:
/* Check 64 MB of RAM. */ /*ram_check(0x00000000, 0x04000000);*/ ram_check(0,0x9ffff); /*ram_check(100000,TOM);*/
Should I try also (100000,TOM) or is it satisfactory once DRAM succeeds?
--- Snip of Currect LinuxBIOSv2 Output --- Testing DRAM : 00000000-0009ffff DRAM fill: 00000000-0009ffff 00000000 00010000 00020000 00030000 00040000 00050000 00060000 00070000 00080000 00090000 000a0000 DRAM filled DRAM verify: 00000000-0009ffff 00000000 00010000 00020000 00030000 00040000 00050000 00060000 00070000 00080000 00090000 000a0000 DRAM range verified. Done. --- End of Snip ---
Error #2 of 2 (PCI Bus Error): (I've yet to google for this.)
--- Snip of PCI Bus Error --- Finding PCI configuration type. PCI: Using configuration type 1 PCI_DOMAIN: 0000 enabled PCI_DOMAIN: 0000 scanning... PCI: pci_scan_bus for bus 00 PCI: 00:00.0 Cannot find pci bus operations --- End of Snip ---
-- Roger http://www.eskimo.com/~roger/index.html Key fingerprint = 8977 A252 2623 F567 70CD 1261 640F C963 1005 1D61
Fri Apr 27 11:50:24 PDT 2007
On Fri, 2007-04-27 at 12:13 -0700, roger wrote:
On Fri, 2007-04-27 at 00:01 -0700, roger wrote:
I just used Uwe's Tyan S1846 tree and everything *just* worked. :-)
However, I don't mean everything worked.
I've attached a log file of the boot for those curious.
Here's Failure #1 - DRAM Verify Failure --- Begin Snip --- DRAM filled DRAM verify: 00000000-04000000 00000000 00010000 00020000 00030000 00040000 00050000 00060000 00070000 00080000 00090000 000a0000 F ail: @0x000a0000 Read value=0xffffffff Fail: @0x000a0004 Read value=0xffffffff Fail: @0x000a0008 Read value=0xffffffff ...
Known issue, I figured you already knew. Ram init is very much broken
on
the i440bx northbridge.
0xa0000 looks more like video memory to me.
try checking 0-0x9ffff and 100000-TOM
[Attached full log of LinuxBIOSv2 output.]
Error #1 of 2 (DRAM Succeeds):
Ok. I've set the following in tyan/s1846/auto.c:
/* Check 64 MB of RAM. */ /*ram_check(0x00000000, 0x04000000);*/ ram_check(0,0x9ffff); /*ram_check(100000,TOM);*/
Should I try also (100000,TOM) or is it satisfactory once DRAM succeeds?
--- Snip of Currect LinuxBIOSv2 Output --- Testing DRAM : 00000000-0009ffff DRAM fill: 00000000-0009ffff 00000000 00010000 00020000 00030000 00040000 00050000 00060000 00070000 00080000 00090000 000a0000 DRAM filled DRAM verify: 00000000-0009ffff 00000000 00010000 00020000 00030000 00040000 00050000 00060000 00070000 00080000 00090000 000a0000 DRAM range verified. Done. --- End of Snip ---
Error #2 of 2 (PCI Bus Error): (I've yet to google for this.)
--- Snip of PCI Bus Error --- Finding PCI configuration type. PCI: Using configuration type 1 PCI_DOMAIN: 0000 enabled PCI_DOMAIN: 0000 scanning... PCI: pci_scan_bus for bus 00 PCI: 00:00.0 Cannot find pci bus operations --- End of Snip ---
According to my searching via google, this looks to be about the current status of this board.
If the above DRAM memory search range is a custom that is related to this specific Tyan S1832DL, how about I provide an svn diff for inclusion into the tree? (If not, will just keep using s1846 tree and is preferable for further code updates.) -- Roger http://www.eskimo.com/~roger/index.html Key fingerprint = 8977 A252 2623 F567 70CD 1261 640F C963 1005 1D61
Fri Apr 27 13:21:36 PDT 2007
On Fri, Apr 27, 2007 at 01:21:47PM -0700, roger wrote:
--- Snip of Currect LinuxBIOSv2 Output --- Testing DRAM : 00000000-0009ffff DRAM fill: 00000000-0009ffff 00000000 00010000 00020000 00030000 00040000 00050000 00060000 00070000 00080000 00090000 000a0000 DRAM filled DRAM verify: 00000000-0009ffff 00000000 00010000 00020000 00030000 00040000 00050000 00060000 00070000 00080000 00090000 000a0000 DRAM range verified. Done. --- End of Snip ---
OK, this _looks_ good, but I'm not sure it actually works. Did you start from a cold boot (power-off)?
Results which you get from merely soft-resetting the system are not reliable.
If the above DRAM memory search range is a custom that is related to this specific Tyan S1832DL, how about I provide an svn diff for inclusion into the tree? (If not, will just keep using s1846 tree and is preferable for further code updates.)
No need for a patch here, I think.
What needs fixing is the code in raminit.c which is generic for all 440BX boards...
Uwe.
On Fri, 2007-04-27 at 23:37 +0200, Uwe Hermann wrote:
OK, this _looks_ good, but I'm not sure it actually works. Did you start from a cold boot (power-off)?
Yupper! Cold as in Cold Beer. :-)
Results which you get from merely soft-resetting the system are not reliable.
If the above DRAM memory search range is a custom that is related to this specific Tyan S1832DL, how about I provide an svn diff for inclusion into the tree? (If not, will just keep using s1846 tree and is preferable for further code updates.)
No need for a patch here, I think.
What needs fixing is the code in raminit.c which is generic for all 440BX boards...
Ok.
I've posted additional info about the 440BX boards I've test here onto my website, along wit a S1832DL prebuilt rom image for others to have fun with.
BTW. I like the WP1 & WP2 indicators for the website documentation on the chips denoting what needs work.
-- Roger http://www.eskimo.com/~roger/index.html Key fingerprint = 8977 A252 2623 F567 70CD 1261 640F C963 1005 1D61
Fri Apr 27 15:57:11 PDT 2007
On Fri, 2007-04-27 at 23:37 +0200, Uwe Hermann wrote:
/src/mainboard/tyan/s1846/auto.c --- Begin of Snip --- /* Check 64 MB of RAM. */ /*ram_check(0x00000000, 0x04000000);*/ ram_check(0,0x9ffff); /*ram_check(100000,TOM);*/ --- End of Snip ---
No need for a patch here, I think.
What needs fixing is the code in raminit.c which is generic for all 440BX boards...
Couldn't we just provide an "IF ram_check failed, then provide This Location" within auto.c?
Or are we talking adding a feature for ram_check for checking alternate ranges as a fallback?
-- Roger http://www.eskimo.com/~roger/index.html Key fingerprint = 8977 A252 2623 F567 70CD 1261 640F C963 1005 1D61
Fri Apr 27 16:41:57 PDT 2007