* 2.6.20-rc2: kernel BUG at include/asm/dma-mapping.h:110!
@ 2006-12-29 20:49 Andreas Schwab
2006-12-29 21:52 ` Benjamin Herrenschmidt
0 siblings, 1 reply; 4+ messages in thread
From: Andreas Schwab @ 2006-12-29 20:49 UTC (permalink / raw)
To: linux1394-devel; +Cc: linuxppc-dev, Stefan Richter
I'm getting this BUG on PowerMac with 2.6.20-rc2 during probing of
firewire devices:
kernel BUG at include/asm/dma-mapping.h:110!
Oops: Exception in kernel mode, sig: 5 [#1]
SMP NR_CPUS=2
Modules linked in: sbp2 snd_aoa_codec_tas ide_cd ohci1394 snd_aoa_fabric_layout snd_aoa cdrom ieee1394 snd_aoa_i2sbus snd_aoa_soundbus sungem sungem_phy uninorth_agp agpgart snd_powermac snd_pcm snd_page_alloc snd_timer snd soundcore sg sata_svw libata sd_mod scsi_mod
NIP: D0000000002F5A90 LR: D0000000002F5A38 CTR: 0000000000000000
REGS: c00000000f447490 TRAP: 0700 Not tainted (2.6.20-rc2)
MSR: 9000000000029032 <EE,ME,IR,DR> CR: 84222482 XER: 00000000
TASK = c000000001139080[2224] 'modprobe' THREAD: c00000000f444000 CPU: 0
GPR00: 0000000000000001 C00000000F447710 D000000000303F38 C00000000F2E00F0
GPR04: D000000000282000 C000000001FD4228 C00000000F517140 0000000000000004
GPR08: 0000000000012800 0000000000000000 D0000000002F7970 D0000000002F7968
GPR12: D00000000005B200 C0000000004D1280 D0000000002EE6C0 0000000000000000
GPR16: D0000000002EE680 0000000000000009 000000001002BD88 D0000000002EDDB2
GPR20: 0000000000000000 0000000000000028 00000000000000C8 D0000000002E8000
GPR24: 0000000000000021 0000000000000022 C00000004E32D450 C00000004E32D400
GPR28: 0000000000000000 C00000004E662FB0 D000000000303A00 C000000001FD4200
NIP [D0000000002F5A90] .sbp2_probe+0x298/0x71c [sbp2]
LR [D0000000002F5A38] .sbp2_probe+0x240/0x71c [sbp2]
Call Trace:
[C00000000F447710] [D0000000002F5A38] .sbp2_probe+0x240/0x71c [sbp2] (unreliable)
[C00000000F4477D0] [C0000000002494A0] .really_probe+0xbc/0x180
[C00000000F447870] [C000000000249858] .__driver_attach+0xb4/0x138
[C00000000F447900] [C00000000024848C] .bus_for_each_dev+0x7c/0xd4
[C00000000F4479C0] [C00000000024928C] .driver_attach+0x28/0x40
[C00000000F447A40] [C000000000248960] .bus_add_driver+0x80/0x1ec
[C00000000F447AE0] [C000000000249C58] .driver_register+0xa8/0xc4
[C00000000F447B60] [D0000000001727F8] .__hpsb_register_protocol+0x48/0xbc [ieee1394]
[C00000000F447C00] [D0000000002F5FB0] .sbp2_module_init+0x9c/0xe0 [sbp2]
[C00000000F447C90] [C000000000071EFC] .sys_init_module+0x17c4/0x198c
[C00000000F447E30] [C0000000000086AC] syscall_exit+0x0/0x40
Instruction dump:
48001839 e8410028 7fe3fb78 3ba0fff4 48000470 e8030240 2fa00000 7c090378
40fe0008 39200000 7d200074 7800d182 <0b000000> e9290000 38800010 38c000d0
Bisecting has identified this commit:
commit 9b7d9c096dd4e4baacc21b2588662bbb56f36c4e
Author: Stefan Richter <stefanr@s5r6.in-berlin.de>
Date: Wed Nov 22 21:44:34 2006 +0100
ieee1394: sbp2: convert from PCI DMA to generic DMA
API conversion without change in functionality
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
I'm only seeing this on ppc64, ppc32 seems to be working fine.
Andreas.
--
Andreas Schwab, SuSE Labs, schwab@suse.de
SuSE Linux Products GmbH, Maxfeldstraße 5, 90409 Nürnberg, Germany
PGP key fingerprint = 58CA 54C7 6D53 942B 1756 01D3 44D5 214B 8276 4ED5
"And now for something completely different."
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: 2.6.20-rc2: kernel BUG at include/asm/dma-mapping.h:110!
2006-12-29 20:49 2.6.20-rc2: kernel BUG at include/asm/dma-mapping.h:110! Andreas Schwab
@ 2006-12-29 21:52 ` Benjamin Herrenschmidt
2006-12-29 22:47 ` Stefan Richter
0 siblings, 1 reply; 4+ messages in thread
From: Benjamin Herrenschmidt @ 2006-12-29 21:52 UTC (permalink / raw)
To: Andreas Schwab
Cc: linuxppc-dev, linux1394-devel, Stefan Richter, Linux Kernel list
> Bisecting has identified this commit:
>
> commit 9b7d9c096dd4e4baacc21b2588662bbb56f36c4e
> Author: Stefan Richter <stefanr@s5r6.in-berlin.de>
> Date: Wed Nov 22 21:44:34 2006 +0100
>
> ieee1394: sbp2: convert from PCI DMA to generic DMA
>
> API conversion without change in functionality
>
> Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
>
>
> I'm only seeing this on ppc64, ppc32 seems to be working fine.
The patch looks totally bogus to me. It's passing a random struct device
from the hbsp host data structure to the dma_map_* routines. which they
can't do anything about.
The dma_map_* routines only know about some bus types. That's always
been the case (that's why you also can't pass a usb device's struct
device to them for example). Mostly, PCI, possibly others depending on
the platform.
So if you are to pass a struct device pointer to dma_map_*, use the one
inside the pci_dev of the host. Or have the host driver provide you with
the struct device pointer (which is the one from the pci_dev * for PCI
implementations, and others give you what they are on, assuming the
platform can do dma-* on that device).
Ben.
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: 2.6.20-rc2: kernel BUG at include/asm/dma-mapping.h:110!
2006-12-29 21:52 ` Benjamin Herrenschmidt
@ 2006-12-29 22:47 ` Stefan Richter
2006-12-30 12:41 ` Andreas Schwab
0 siblings, 1 reply; 4+ messages in thread
From: Stefan Richter @ 2006-12-29 22:47 UTC (permalink / raw)
To: Benjamin Herrenschmidt; +Cc: linux1394-devel, Linux Kernel list, linuxppc-dev
Benjamin Herrenschmidt wrote:
>> Bisecting has identified this commit:
>>
>> commit 9b7d9c096dd4e4baacc21b2588662bbb56f36c4e
>> Author: Stefan Richter <stefanr@s5r6.in-berlin.de>
>> Date: Wed Nov 22 21:44:34 2006 +0100
>>
>> ieee1394: sbp2: convert from PCI DMA to generic DMA
>>
>> API conversion without change in functionality
>>
>> Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
>>
>>
>> I'm only seeing this on ppc64, ppc32 seems to be working fine.
>
> The patch looks totally bogus to me. It's passing a random struct device
> from the hbsp host data structure to the dma_map_* routines. which they
> can't do anything about.
[...]
> So if you are to pass a struct device pointer to dma_map_*, use the one
> inside the pci_dev of the host. Or have the host driver provide you with
> the struct device pointer (which is the one from the pci_dev * for PCI
> implementations, and others give you what they are on, assuming the
> platform can do dma-* on that device).
[...]
The parent device of my bogus fw-host device should do the trick. Alas I
can't test on ppc64 or with anything else than ohci1394 driven
controllers...
From: Stefan Richter <stefanr@s5r6.in-berlin.de>
Subject: ieee1394: sbp2: fix bogus dma mapping
Need to use a PCI device, not a FireWire host device. Problem found by
Andreas Schwab, mistake pointed out by Benjamin Herrenschmidt.
http://ozlabs.org/pipermail/linuxppc-dev/2006-December/029595.html
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
---
drivers/ieee1394/sbp2.c | 73 +++++++++++++++++++++-------------------
1 file changed, 40 insertions(+), 33 deletions(-)
Index: linux-2.6.20-rc2/drivers/ieee1394/sbp2.c
===================================================================
--- linux-2.6.20-rc2.orig/drivers/ieee1394/sbp2.c
+++ linux-2.6.20-rc2/drivers/ieee1394/sbp2.c
@@ -490,11 +490,11 @@ static int sbp2util_create_command_orb_p
spin_unlock_irqrestore(&lu->cmd_orb_lock, flags);
return -ENOMEM;
}
- cmd->command_orb_dma = dma_map_single(&hi->host->device,
+ cmd->command_orb_dma = dma_map_single(hi->host->device.parent,
&cmd->command_orb,
sizeof(struct sbp2_command_orb),
DMA_TO_DEVICE);
- cmd->sge_dma = dma_map_single(&hi->host->device,
+ cmd->sge_dma = dma_map_single(hi->host->device.parent,
&cmd->scatter_gather_element,
sizeof(cmd->scatter_gather_element),
DMA_BIDIRECTIONAL);
@@ -516,10 +516,11 @@ static void sbp2util_remove_command_orb_
if (!list_empty(&lu->cmd_orb_completed))
list_for_each_safe(lh, next, &lu->cmd_orb_completed) {
cmd = list_entry(lh, struct sbp2_command_info, list);
- dma_unmap_single(&host->device, cmd->command_orb_dma,
+ dma_unmap_single(host->device.parent,
+ cmd->command_orb_dma,
sizeof(struct sbp2_command_orb),
DMA_TO_DEVICE);
- dma_unmap_single(&host->device, cmd->sge_dma,
+ dma_unmap_single(host->device.parent, cmd->sge_dma,
sizeof(cmd->scatter_gather_element),
DMA_BIDIRECTIONAL);
kfree(cmd);
@@ -601,17 +602,17 @@ static void sbp2util_mark_command_comple
if (cmd->cmd_dma) {
if (cmd->dma_type == CMD_DMA_SINGLE)
- dma_unmap_single(&host->device, cmd->cmd_dma,
+ dma_unmap_single(host->device.parent, cmd->cmd_dma,
cmd->dma_size, cmd->dma_dir);
else if (cmd->dma_type == CMD_DMA_PAGE)
- dma_unmap_page(&host->device, cmd->cmd_dma,
+ dma_unmap_page(host->device.parent, cmd->cmd_dma,
cmd->dma_size, cmd->dma_dir);
/* XXX: Check for CMD_DMA_NONE bug */
cmd->dma_type = CMD_DMA_NONE;
cmd->cmd_dma = 0;
}
if (cmd->sge_buffer) {
- dma_unmap_sg(&host->device, cmd->sge_buffer,
+ dma_unmap_sg(host->device.parent, cmd->sge_buffer,
cmd->dma_size, cmd->dma_dir);
cmd->sge_buffer = NULL;
}
@@ -836,37 +837,37 @@ static int sbp2_start_device(struct sbp2
struct sbp2_fwhost_info *hi = lu->hi;
int error;
- lu->login_response = dma_alloc_coherent(&hi->host->device,
+ lu->login_response = dma_alloc_coherent(hi->host->device.parent,
sizeof(struct sbp2_login_response),
&lu->login_response_dma, GFP_KERNEL);
if (!lu->login_response)
goto alloc_fail;
- lu->query_logins_orb = dma_alloc_coherent(&hi->host->device,
+ lu->query_logins_orb = dma_alloc_coherent(hi->host->device.parent,
sizeof(struct sbp2_query_logins_orb),
&lu->query_logins_orb_dma, GFP_KERNEL);
if (!lu->query_logins_orb)
goto alloc_fail;
- lu->query_logins_response = dma_alloc_coherent(&hi->host->device,
+ lu->query_logins_response = dma_alloc_coherent(hi->host->device.parent,
sizeof(struct sbp2_query_logins_response),
&lu->query_logins_response_dma, GFP_KERNEL);
if (!lu->query_logins_response)
goto alloc_fail;
- lu->reconnect_orb = dma_alloc_coherent(&hi->host->device,
+ lu->reconnect_orb = dma_alloc_coherent(hi->host->device.parent,
sizeof(struct sbp2_reconnect_orb),
&lu->reconnect_orb_dma, GFP_KERNEL);
if (!lu->reconnect_orb)
goto alloc_fail;
- lu->logout_orb = dma_alloc_coherent(&hi->host->device,
+ lu->logout_orb = dma_alloc_coherent(hi->host->device.parent,
sizeof(struct sbp2_logout_orb),
&lu->logout_orb_dma, GFP_KERNEL);
if (!lu->logout_orb)
goto alloc_fail;
- lu->login_orb = dma_alloc_coherent(&hi->host->device,
+ lu->login_orb = dma_alloc_coherent(hi->host->device.parent,
sizeof(struct sbp2_login_orb),
&lu->login_orb_dma, GFP_KERNEL);
if (!lu->login_orb)
@@ -929,32 +930,32 @@ static void sbp2_remove_device(struct sb
list_del(&lu->lu_list);
if (lu->login_response)
- dma_free_coherent(&hi->host->device,
+ dma_free_coherent(hi->host->device.parent,
sizeof(struct sbp2_login_response),
lu->login_response,
lu->login_response_dma);
if (lu->login_orb)
- dma_free_coherent(&hi->host->device,
+ dma_free_coherent(hi->host->device.parent,
sizeof(struct sbp2_login_orb),
lu->login_orb,
lu->login_orb_dma);
if (lu->reconnect_orb)
- dma_free_coherent(&hi->host->device,
+ dma_free_coherent(hi->host->device.parent,
sizeof(struct sbp2_reconnect_orb),
lu->reconnect_orb,
lu->reconnect_orb_dma);
if (lu->logout_orb)
- dma_free_coherent(&hi->host->device,
+ dma_free_coherent(hi->host->device.parent,
sizeof(struct sbp2_logout_orb),
lu->logout_orb,
lu->logout_orb_dma);
if (lu->query_logins_orb)
- dma_free_coherent(&hi->host->device,
+ dma_free_coherent(hi->host->device.parent,
sizeof(struct sbp2_query_logins_orb),
lu->query_logins_orb,
lu->query_logins_orb_dma);
if (lu->query_logins_response)
- dma_free_coherent(&hi->host->device,
+ dma_free_coherent(hi->host->device.parent,
sizeof(struct sbp2_query_logins_response),
lu->query_logins_response,
lu->query_logins_response_dma);
@@ -1445,7 +1446,7 @@ static void sbp2_prep_command_orb_sg(str
cmd->dma_size = sgpnt[0].length;
cmd->dma_type = CMD_DMA_PAGE;
- cmd->cmd_dma = dma_map_page(&hi->host->device,
+ cmd->cmd_dma = dma_map_page(hi->host->device.parent,
sgpnt[0].page, sgpnt[0].offset,
cmd->dma_size, cmd->dma_dir);
@@ -1457,8 +1458,8 @@ static void sbp2_prep_command_orb_sg(str
&cmd->scatter_gather_element[0];
u32 sg_count, sg_len;
dma_addr_t sg_addr;
- int i, count = dma_map_sg(&hi->host->device, sgpnt, scsi_use_sg,
- dma_dir);
+ int i, count = dma_map_sg(hi->host->device.parent, sgpnt,
+ scsi_use_sg, dma_dir);
cmd->dma_size = scsi_use_sg;
cmd->sge_buffer = sgpnt;
@@ -1508,7 +1509,8 @@ static void sbp2_prep_command_orb_no_sg(
cmd->dma_dir = dma_dir;
cmd->dma_size = scsi_request_bufflen;
cmd->dma_type = CMD_DMA_SINGLE;
- cmd->cmd_dma = dma_map_single(&hi->host->device, scsi_request_buffer,
+ cmd->cmd_dma = dma_map_single(hi->host->device.parent,
+ scsi_request_buffer,
cmd->dma_size, cmd->dma_dir);
orb->data_descriptor_hi = ORB_SET_NODE_ID(hi->host->node_id);
orb->misc |= ORB_SET_DIRECTION(orb_direction);
@@ -1626,10 +1628,11 @@ static void sbp2_link_orb_command(struct
size_t length;
unsigned long flags;
- dma_sync_single_for_device(&hi->host->device, cmd->command_orb_dma,
+ dma_sync_single_for_device(hi->host->device.parent,
+ cmd->command_orb_dma,
sizeof(struct sbp2_command_orb),
DMA_TO_DEVICE);
- dma_sync_single_for_device(&hi->host->device, cmd->sge_dma,
+ dma_sync_single_for_device(hi->host->device.parent, cmd->sge_dma,
sizeof(cmd->scatter_gather_element),
DMA_BIDIRECTIONAL);
@@ -1655,14 +1658,15 @@ static void sbp2_link_orb_command(struct
* The target's fetch agent may or may not have read this
* previous ORB yet.
*/
- dma_sync_single_for_cpu(&hi->host->device, last_orb_dma,
+ dma_sync_single_for_cpu(hi->host->device.parent, last_orb_dma,
sizeof(struct sbp2_command_orb),
DMA_TO_DEVICE);
last_orb->next_ORB_lo = cpu_to_be32(cmd->command_orb_dma);
wmb();
/* Tells hardware that this pointer is valid */
last_orb->next_ORB_hi = 0;
- dma_sync_single_for_device(&hi->host->device, last_orb_dma,
+ dma_sync_single_for_device(hi->host->device.parent,
+ last_orb_dma,
sizeof(struct sbp2_command_orb),
DMA_TO_DEVICE);
addr += SBP2_DOORBELL_OFFSET;
@@ -1790,10 +1794,11 @@ static int sbp2_handle_status_write(stru
else
cmd = sbp2util_find_command_for_orb(lu, sb->ORB_offset_lo);
if (cmd) {
- dma_sync_single_for_cpu(&hi->host->device, cmd->command_orb_dma,
+ dma_sync_single_for_cpu(hi->host->device.parent,
+ cmd->command_orb_dma,
sizeof(struct sbp2_command_orb),
DMA_TO_DEVICE);
- dma_sync_single_for_cpu(&hi->host->device, cmd->sge_dma,
+ dma_sync_single_for_cpu(hi->host->device.parent, cmd->sge_dma,
sizeof(cmd->scatter_gather_element),
DMA_BIDIRECTIONAL);
/* Grab SCSI command pointers and check status. */
@@ -1931,10 +1936,11 @@ static void sbp2scsi_complete_all_comman
while (!list_empty(&lu->cmd_orb_inuse)) {
lh = lu->cmd_orb_inuse.next;
cmd = list_entry(lh, struct sbp2_command_info, list);
- dma_sync_single_for_cpu(&hi->host->device, cmd->command_orb_dma,
+ dma_sync_single_for_cpu(hi->host->device.parent,
+ cmd->command_orb_dma,
sizeof(struct sbp2_command_orb),
DMA_TO_DEVICE);
- dma_sync_single_for_cpu(&hi->host->device, cmd->sge_dma,
+ dma_sync_single_for_cpu(hi->host->device.parent, cmd->sge_dma,
sizeof(cmd->scatter_gather_element),
DMA_BIDIRECTIONAL);
sbp2util_mark_command_completed(lu, cmd);
@@ -2059,11 +2065,12 @@ static int sbp2scsi_abort(struct scsi_cm
spin_lock_irqsave(&lu->cmd_orb_lock, flags);
cmd = sbp2util_find_command_for_SCpnt(lu, SCpnt);
if (cmd) {
- dma_sync_single_for_cpu(&hi->host->device,
+ dma_sync_single_for_cpu(hi->host->device.parent,
cmd->command_orb_dma,
sizeof(struct sbp2_command_orb),
DMA_TO_DEVICE);
- dma_sync_single_for_cpu(&hi->host->device, cmd->sge_dma,
+ dma_sync_single_for_cpu(hi->host->device.parent,
+ cmd->sge_dma,
sizeof(cmd->scatter_gather_element),
DMA_BIDIRECTIONAL);
sbp2util_mark_command_completed(lu, cmd);
--
Stefan Richter
-=====-=-==- ==-- ===-=
http://arcgraph.de/sr/
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: 2.6.20-rc2: kernel BUG at include/asm/dma-mapping.h:110!
2006-12-29 22:47 ` Stefan Richter
@ 2006-12-30 12:41 ` Andreas Schwab
0 siblings, 0 replies; 4+ messages in thread
From: Andreas Schwab @ 2006-12-30 12:41 UTC (permalink / raw)
To: Stefan Richter; +Cc: linux1394-devel, Linux Kernel list, linuxppc-dev
Stefan Richter <stefanr@s5r6.in-berlin.de> writes:
> The parent device of my bogus fw-host device should do the trick. Alas I
> can't test on ppc64 or with anything else than ohci1394 driven
> controllers...
Successfully tested on ppc64.
Thanks, Andreas.
--
Andreas Schwab, SuSE Labs, schwab@suse.de
SuSE Linux Products GmbH, Maxfeldstraße 5, 90409 Nürnberg, Germany
PGP key fingerprint = 58CA 54C7 6D53 942B 1756 01D3 44D5 214B 8276 4ED5
"And now for something completely different."
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2006-12-30 12:41 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2006-12-29 20:49 2.6.20-rc2: kernel BUG at include/asm/dma-mapping.h:110! Andreas Schwab
2006-12-29 21:52 ` Benjamin Herrenschmidt
2006-12-29 22:47 ` Stefan Richter
2006-12-30 12:41 ` Andreas Schwab
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).