Zheng Bao (zheng.bao(a)amd.com) just uploaded a new patch set to gerrit, which you can find at http://review.coreboot.org/1265
-gerrit
commit b485f25c251f66eac763e1d38f93f432f4380709
Author: zbao <fishbaozi(a)gmail.com>
Date: Mon Jul 23 19:52:58 2012 +0800
Remove the misleading 0x100 from the limit.
I dont known if missed something, but why an extra 0x100 was added to limit?
My board would get the wrong memory table entry 7f000000-7fffffff as RAM, which
is higher than TOM.
coreboot memory table:
0. 0000000000000000-0000000000000fff: CONFIGURATION TABLES
1. 0000000000001000-000000000009ffff: RAM
2. 00000000000c0000-000000005e13efff: RAM
3. 000000005e13f000-000000005effffff: CONFIGURATION TABLES
4. 000000005f000000-000000007effffff: RESERVED
5. 000000007f000000-000000007fffffff: RAM
6. 00000000a0000000-00000000afffffff: RESERVED
Change-Id: I3848ed5f23001e5bd61a19833650fe13df26eef3
Signed-off-by: Zheng Bao <zheng.bao(a)amd.com>
Signed-off-by: zbao <fishbaozi(a)gmail.com>
---
src/northbridge/amd/agesa/family15tn/northbridge.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/src/northbridge/amd/agesa/family15tn/northbridge.c b/src/northbridge/amd/agesa/family15tn/northbridge.c
index 12aab33..8ad7841 100644
--- a/src/northbridge/amd/agesa/family15tn/northbridge.c
+++ b/src/northbridge/amd/agesa/family15tn/northbridge.c
@@ -790,7 +790,7 @@ static void domain_set_resources(device_t dev)
if (!(d.mask & 1)) continue;
basek = ((resource_t)(d.base & 0x1fffff00)) << 9; // could overflow, we may lost 6 bit here
- limitk = ((resource_t)((d.mask + 0x00000100) & 0x1fffff00)) << 9 ;
+ limitk = ((resource_t)(d.mask & 0x1fffff00)) << 9 ;
sizek = limitk - basek;
Zheng Bao (zheng.bao(a)amd.com) just uploaded a new patch set to gerrit, which you can find at http://review.coreboot.org/1264
-gerrit
commit 6c9e7d7c6d8eb28b39c9bef1211031a88fd5773e
Author: zbao <fishbaozi(a)gmail.com>
Date: Mon Jul 23 19:49:40 2012 +0800
Limit the device field to 5 bits.
The field device in PCI_ADDRESS only takes 5 bits. So if the device number is
more than 32, it will truncated to 5 bits. Before this patch, other pci devices
will be incorrectly probed as processor node.
Change-Id: I64dcd4f4fda7b7080a9905dce580feb829584b94
Signed-off-by: Zheng Bao <zheng.bao(a)amd.com>
Signed-off-by: zbao <fishbaozi(a)gmail.com>
---
src/northbridge/amd/agesa/family15tn/northbridge.c | 4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/src/northbridge/amd/agesa/family15tn/northbridge.c b/src/northbridge/amd/agesa/family15tn/northbridge.c
index bc3877f..12aab33 100644
--- a/src/northbridge/amd/agesa/family15tn/northbridge.c
+++ b/src/northbridge/amd/agesa/family15tn/northbridge.c
@@ -120,8 +120,8 @@ static void set_mmio_addr_reg(u32 nodeid, u32 linkn, u32 reg, u32 index, u32 mmi
static device_t get_node_pci(u32 nodeid, u32 fn)
{
-#if MAX_NODE_NUMS == 64
- if (nodeid < 32) {
+#if MAX_NODE_NUMS + CONFIG_CDB >= 32
+ if ((CONFIG_CDB + nodeid) < 32) {
return dev_find_slot(CONFIG_CBB, PCI_DEVFN(CONFIG_CDB + nodeid, fn));
} else {
return dev_find_slot(CONFIG_CBB-1, PCI_DEVFN(CONFIG_CDB + nodeid - 32, fn));
On 2012-07-31 06:19, Mark Nelson wrote:
> Hi Peter,
Hi!
> It should be the same as the documentation for the 890FX / SB850.
>
> See the full e-mail trail starting here (for some reason my mail client
> didn't reply properly so it seems to have confused mailman with the
> response...):
> http://www.coreboot.org/pipermail/coreboot/2012-July/070712.html
Cool!
> Hope that helps!
> Mark
It sure does, super thanks!
Best regards
Peter K
the following patch was just integrated into master:
commit 222fd1bbc00af648cf59f3ec34cc684d435328d0
Author: Sven Schnelle <svens(a)stackframe.org>
Date: Sun Jul 29 19:18:03 2012 +0200
Revert "Use broadcast SIPI to startup siblings"
This reverts commit 042c1461fb777e583e5de48edf9326e47ee5595f.
It turned out that sending IPIs via broadcast doesn't work on
Sandybridge. We tried to come up with a solution, but didn't
found any so far. So revert the code for now until we have
a working solution.
Change-Id: I7dd1cba5a4c1e4b0af366b20e8263b1f6f4b9714
Signed-off-by: Sven Schnelle <svens(a)stackframe.org>
Build-Tested: build bot (Jenkins) at Tue Jul 31 06:30:15 2012, giving +1
Reviewed-By: Ronald G. Minnich <rminnich(a)gmail.com> at Tue Jul 31 06:46:01 2012, giving +2
See http://review.coreboot.org/1381 for details.
-gerrit
the following patch was just integrated into master:
commit a76ce3ce25f5480278469a3515742625cb73cf3d
Author: Sven Schnelle <svens(a)stackframe.org>
Date: Sun Jul 29 17:42:52 2012 +0200
Revert "remove CONFIG_SERIAL_CPU_INIT"
This reverts commit 78efc4c36c68b51b3e73acdb721a12ec23ed0369.
The broadcast patch was reverted, so this commit should also
be reverted. The reason for reverting the broadcast patch:
It turned out that sending IPIs via broadcast doesn't work on
Sandybridge. We tried to come up with a solution, but didn't
found any so far. So revert the code for now until we have
a working solution.
Change-Id: I05c27dec55fa681f455215be56dcbc5f22808193
Signed-off-by: Sven Schnelle <svens(a)stackframe.org>
Build-Tested: build bot (Jenkins) at Mon Jul 30 22:45:41 2012, giving +1
Reviewed-By: Ronald G. Minnich <rminnich(a)gmail.com> at Tue Jul 31 05:52:44 2012, giving +2
See http://review.coreboot.org/1380 for details.
-gerrit