initcall warnings in 2.6.17
Subsystems (4)
I still get
initcall at 0xffffffff807414c2: pci_iommu_init+0x0/0x501(): returned with error code -1 initcall at 0xffffffff80748b4d: init_autofs4_fs+0x0/0xc(): returned with error code -16 initcall at 0xffffffff803c7d5c: init_netconsole+0x0/0x6b(): returned with error code -22 initcall at 0xffffffff80249307: software_resume+0x0/0xcf(): returned with error code -2
I'm not sure it was that good an idea to enable this warning by default in 2.6.17. It will be still in the release and probably generate some user queries.
Might be better to disable it for 2.6.17 again and only reenable for 2.6.18 after some auditing?
-Andi
Quoted 4 lines · Andi Kleen
I still get
initcall at
0xffffffff807414c2: pci_iommu_init+0x0/0x501(): returned with error code -1
Should be returning -ENODEV.
Quoted 1 line
initcall at
0xffffffff80748b4d: init_autofs4_fs+0x0/0xc(): returned with error code -16
hm. Why'd that happen?
Quoted 1 line
initcall at
0xffffffff803c7d5c: init_netconsole+0x0/0x6b(): returned with error code -22
Yeah. I think netconsole is just being wrong here. If it wasn't enabled there's no error.
Quoted 1 line
initcall at
0xffffffff80249307: software_resume+0x0/0xcf(): returned with error code -2
Similarly, there's no resume file configured so should we really consider this an error?
Quoted 6 lines
I'm not sure it was that good an idea to enable this warning by default in 2.6.17. It will be still in the release and probably generate some user queries.
Might be better to disable it for 2.6.17 again and only reenable for 2.6.18 after some auditing?
Yes, I'm inclined to agree. It's finding some real but very minor problems. Probably too minor to be scaring users. I'll cook up a patch to move it under `initcall_debug'.
Quoted 14 lines · Andrew Morton
Quoted 4 lines · Andi Kleen
I still get
initcall at
0xffffffff807414c2: pci_iommu_init+0x0/0x501(): returned with error code -1Should be returning
-ENODEV.Quoted 1 line
initcall at
0xffffffff80748b4d: init_autofs4_fs+0x0/0xc(): returned with error code -16hm. Why'd that happen?
Quoted 1 line
initcall at
0xffffffff803c7d5c: init_netconsole+0x0/0x6b(): returned with error code -22
Quoted 7 lines
Yeah. I think netconsole is just being wrong here. If it wasn't enabled there's no error.
Quoted 1 line
initcall at
0xffffffff80249307: software_resume+0x0/0xcf(): returned with error code -2Similarly, there's no resume file configured so should we really consider this an error?
That one is easier to solve. But "configured but image not present" will also produce warning I guess. If I return -ENODEV, will it keep debugging system quiet?
Pavel
Signature · 1 line
Thanks for all the (sleeping) penguins.
Hi!
This should fix
Quoted 1 line
Quoted 1 line
initcall at
0xffffffff80249307: software_resume+0x0/0xcf(): returned with error code -2
warning.
kernel/power/disk.c +1 -1
@@ -168,7 +168,7 @@ static int software_resume(void)if (!swsusp_resume_device) {if (!strlen (resume_file)) {up (&pm_sem); -return -ENOENT ; +return 0 ; } swsusp_resume_device =name_to_dev_t (resume_file);pr_debug ("swsusp: Resume From Partition %s\n" , resume_file);
Signature · 1 line
Thanks for all the (sleeping) penguins.
PATCH 1/1 clean-up initcall warning for netconsole
In reply to Andrew MortonQuoted 2 lines · Andrew Morton
Yeah. I think netconsole is just being wrong here. If it wasn't enabled there's no error.
Here's a fix:
drivers/net/netconsole.c +1 -1
@@ -107,7 +107,7 @@ static int init_netconsole(void)if (!configured) {printk ("netconsole: not configured, aborting\n" ); -return -EINVAL ; +return 0 ; }if (netpoll_setup (&np))
Signature · 1 line
Mathematics is the supreme nostalgia of our time.