agpgart-amd64 not initialized in 2.6.33-rc2
Hi all,
Recently (2.6.33-rc2 kernel, x86_64, 4GB RAM) I found (in dmesg):
[drm:mga_do_agp_dma_bootstrap] *ERROR* Unable to acquire AGP: -19
and there is no /dev/agpgart device on the machine while all is OK if booting 2.6.32.2.
In both kernels I have:
CONFIG_AGP=yCONFIG_AGP_AMD64=y
CONFIG_GART_IOMMU=y
but nevertheless dmesg shows:
PCI-DMA: Using software bounce buffering for IO (SWIOTLB)
due to quirks in via K8T800Pro host bridge.
Looking for the reason I found that agp_amd64_init() appears in:
#ifndef CONFIG_GART_IOMMUmodule_init(agp_amd64_init);module_exit(agp_amd64_cleanup);
#endif
/* so it is not invoked here due to CONFIG_GART_IOMMU=y */
and also appears in:
arch/x86/kernel/pci-gart_64.c, function: gart_iommu_init()
The last one is part of the struct x86_init_ops(struct x86_init_iommu, as a function pointer) only if gart-iommu is successfully detected, which is not the case here due to the quirk, so agp_amd64_init() is not invoked here neither.
Sure, configuring the kernel without CONFIG_GART_IOMMU=y (which is not user selectable)
should solve the problem, but usually users do not know about quirks, so it should work even as set here (and it works up to 2.6.32.2). I believe the bug is introduced with the changes in the order iommu detect/init works for 2.6.33.
I am here for additional info/tests.
Best regards
Marin Mitov
Quoted 39 lines · On Sun, 27 Dec 2009 17:19:39 +0200 Marin Mitov <[email protected]> wrote:
Hi all,
Recently (2.6.33-rc2 kernel,
x86_64, 4GB RAM) I found (in dmesg):[drm:mga_do_agp_dma_bootstrap] *ERROR* Unable to acquire AGP: -19
and there is no
/dev/agpgartdevice on the machine while all is OK if booting 2.6.32.2.In both kernels I have:
CONFIG_AGP=yCONFIG_AGP_AMD64=y
CONFIG_GART_IOMMU=ybut nevertheless dmesg shows:
PCI-DMA: Using software bounce buffering for IO (SWIOTLB)
due to quirks in via K8T800Pro host bridge.
Looking for the reason I found that
agp_amd64_init()appears in:#ifndef
CONFIG_GART_IOMMUmodule_init(agp_amd64_init);module_exit(agp_amd64_cleanup);
#endif/* so it is not invoked here due to
CONFIG_GART_IOMMU=y*/and also appears in:
arch/x86/kernel/pci-gart_64.c, function:gart_iommu_init()The last one is part of the struct
x86_init_ops(struct x86_init_iommu, as a function pointer)only if gart-iommu is successfully detected, which is not the case here due to the quirk, soagp_amd64_init()is not invoked here neither.
Quoted 7 lines
Sure, configuring the kernel without
CONFIG_GART_IOMMU=y(which is not user selectable)should solve the problem, but usually users do not know about quirks, so it should work even as set here (and it works up to 2.6.32.2). I believe the bug is introduced with the changes in the order iommu detect/init works for 2.6.33.
I am here for additional info/tests.
Sorry about the regression. Does this works?
Thanks,
drivers/char/agp/amd64-agp.c +5 -4
@@ -725,9 +725,14 @@ static struct pci_driver agp_amd64_pci_driver = {int __init agp_amd64_init (void ) {int err =0 ; +static int done =0 ;if (agp_off)return -EINVAL ; + +if (done++) +return agp_bridges_found ?0 : -ENODEV ; + err =pci_register_driver (&agp_amd64_pci_driver);if (err <0 )return err;
@@ -771,12 +776,8 @@ static void __exit agp_amd64_cleanup(void)pci_unregister_driver (&agp_amd64_pci_driver); } -/* On AMD64 the PCI driver needs to initialize this driver early -for the IOMMU, so it has to be called via a backdoor. */ -#ifndef CONFIG_GART_IOMMU module_init (agp_amd64_init);module_exit (agp_amd64_cleanup); -#endif MODULE_AUTHOR ("Dave Jones <[email protected]>, Andi Kleen" );module_param (agp_try_unsupported, bool,0 );
Quoted 44 lines · FUJITA Tomonori
Quoted 41 lines · On Sun, 27 Dec 2009 17:19:39 +0200 Marin Mitov <[email protected]> wrote:
Hi all,
Recently (2.6.33-rc2 kernel,
x86_64, 4GB RAM) I found (in dmesg):[drm:mga_do_agp_dma_bootstrap] *ERROR* Unable to acquire AGP: -19
and there is no
/dev/agpgartdevice on the machine while all is OK if booting 2.6.32.2.In both kernels I have:
CONFIG_AGP=yCONFIG_AGP_AMD64=y
CONFIG_GART_IOMMU=ybut nevertheless dmesg shows:
PCI-DMA: Using software bounce buffering for IO (SWIOTLB)
due to quirks in via K8T800Pro host bridge.
Looking for the reason I found that
agp_amd64_init()appears in:#ifndef
CONFIG_GART_IOMMUmodule_init(agp_amd64_init);module_exit(agp_amd64_cleanup);
#endif/* so it is not invoked here due to
CONFIG_GART_IOMMU=y*/and also appears in:
arch/x86/kernel/pci-gart_64.c, function:gart_iommu_init()The last one is part of the struct
x86_init_ops(struct x86_init_iommu, as a function pointer)only if gart-iommu is successfully detected, which is not the case here due to the quirk, soagp_amd64_init()is not invoked here neither.Sure, configuring the kernel without
CONFIG_GART_IOMMU=y(which is not user selectable)
Quoted 7 lines
Quoted 5 lines
should solve the problem, but usually users do not know about quirks, so it should work even as set here (and it works up to 2.6.32.2). I believe the bug is introduced with the changes in the order iommu detect/init works for 2.6.33.
I am here for additional info/tests.
Sorry about the regression. Does this works?
Yes, it works for me, thank you.
Marin Mitov
Quoted 33 lines
drivers/char/agp/amd64-agp.c +5
-4@@ -725,9 +725,14 @@ static struct pci_driver agp_amd64_pci_driver = {int __init agp_amd64_init (void ) {int err =0 ; +static int done =0 ;if (agp_off)return -EINVAL ; + +if (done++) +return agp_bridges_found ?0 : -ENODEV ; + err =pci_register_driver (&agp_amd64_pci_driver);if (err <0 )return err;@@ -771,12 +776,8 @@ static void __exit agp_amd64_cleanup(void)pci_unregister_driver (&agp_amd64_pci_driver); } -/* On AMD64 the PCI driver needs to initialize this driver early -for the IOMMU, so it has to be called via a backdoor. */ -#ifndef CONFIG_GART_IOMMU module_init (agp_amd64_init);module_exit (agp_amd64_cleanup); -#endif MODULE_AUTHOR ("Dave Jones <[email protected]>, Andi Kleen" );module_param (agp_try_unsupported, bool,0 );
Quoted 49 lines · Marin Mitov
Quoted 48 lines · FUJITA Tomonori
Quoted 45 lines · On Sun, 27 Dec 2009 17:19:39 +0200 Marin Mitov <[email protected]> wrote:
Hi all,
Recently (2.6.33-rc2 kernel,
x86_64, 4GB RAM) I found (in dmesg):[drm:mga_do_agp_dma_bootstrap] *ERROR* Unable to acquire AGP: -19
and there is no
/dev/agpgartdevice on the machine while all is OK if booting 2.6.32.2.In both kernels I have:
CONFIG_AGP=yCONFIG_AGP_AMD64=y
CONFIG_GART_IOMMU=ybut nevertheless dmesg shows:
PCI-DMA: Using software bounce buffering for IO (SWIOTLB)
due to quirks in via K8T800Pro host bridge.
Looking for the reason I found that
agp_amd64_init()appears in:#ifndef
CONFIG_GART_IOMMUmodule_init(agp_amd64_init);module_exit(agp_amd64_cleanup);
#endif/* so it is not invoked here due to
CONFIG_GART_IOMMU=y*/and also appears in:
arch/x86/kernel/pci-gart_64.c, function:gart_iommu_init()The last one is part of the struct
x86_init_ops(struct x86_init_iommu, as a function pointer)only if gart-iommu is successfully detected, which is not the case here due to the quirk, soagp_amd64_init()is not invoked here neither.Sure, configuring the kernel without
CONFIG_GART_IOMMU=y(which is not user selectable)should solve the problem, but usually users do not know about quirks, so it should work even as set here (and it works up to 2.6.32.2). I believe the bug is introduced with the changes in the order iommu detect/init works for 2.6.33.
Quoted 5 lines
Quoted 3 lines
Quoted 1 line
I am here for additional info/tests.
Sorry about the regression. Does this works?
Yes, it works for me, thank you.
Great! Fujita-san, mind sending a fully changelogged version of the patch with Tested-by and your Signed-off-by?
Dave, do you want to push this fix to Linus, or should we do it via x86/urgent?
Thanks,
Ingo
FUJITA Tomonori
Tested-by Marin MitovQuoted 52 lines · On Mon, 28 Dec 2009 09:10:21 +0100 Ingo Molnar <[email protected]> wrote:
Quoted 49 lines · Marin Mitov
Quoted 48 lines · FUJITA Tomonori
Quoted 45 lines · On Sun, 27 Dec 2009 17:19:39 +0200 Marin Mitov <[email protected]> wrote:
Hi all,
Recently (2.6.33-rc2 kernel,
x86_64, 4GB RAM) I found (in dmesg):[drm:mga_do_agp_dma_bootstrap] *ERROR* Unable to acquire AGP: -19
and there is no
/dev/agpgartdevice on the machine while all is OK if booting 2.6.32.2.In both kernels I have:
CONFIG_AGP=yCONFIG_AGP_AMD64=y
CONFIG_GART_IOMMU=ybut nevertheless dmesg shows:
PCI-DMA: Using software bounce buffering for IO (SWIOTLB)
due to quirks in via K8T800Pro host bridge.
Looking for the reason I found that
agp_amd64_init()appears in:#ifndef
CONFIG_GART_IOMMUmodule_init(agp_amd64_init);module_exit(agp_amd64_cleanup);
#endif/* so it is not invoked here due to
CONFIG_GART_IOMMU=y*/and also appears in:
arch/x86/kernel/pci-gart_64.c, function:gart_iommu_init()The last one is part of the struct
x86_init_ops(struct x86_init_iommu, as a function pointer)only if gart-iommu is successfully detected, which is not the case here due to the quirk, soagp_amd64_init()is not invoked here neither.Sure, configuring the kernel without
CONFIG_GART_IOMMU=y(which is not user selectable)should solve the problem, but usually users do not know about quirks, so it should work even as set here (and it works up to 2.6.32.2). I believe the bug is introduced with the changes in the order iommu detect/init works for 2.6.33.
Quoted 8 lines
Quoted 5 lines
Quoted 3 lines
Quoted 1 line
I am here for additional info/tests.
Sorry about the regression. Does this works?
Yes, it works for me, thank you.
Great! Fujita-san, mind sending a fully changelogged version of the patch with Tested-by and your Signed-off-by?
=
From: FUJITA Tomonori <[email protected]>
Subject: [PATCH] x86/agp: fix agp_amd64_init() initialization with CONFIG_GART_IOMMU enabled
drivers/char/agp/amd64-agp.c has:
#ifndef CONFIG_GART_IOMMUmodule_init(agp_amd64_init);module_exit(agp_amd64_cleanup);
#endif
agp_amd64_init() was called via gart_iommu_init with CONFIG_GART_IOMMU=y agp_amd64_init() was called via module_init with CONFIG_GART_IOMMU=n
The commit 75f1cdf1dda92cae037ec848ae63690d91913eac changes the x86 dma initialization routine: gart_iommu_init() is called only when GART IOMMU is detected. So when GART IOMMU isn't detected, agp_amd64_init isn't called.
Marin Mitov reported this issue:
http://marc.info/?l=linux-kernel&m=126192729110083&w=2
With this patch, agp_amd64_init() is always called via module_init (the above ifndef is removed). If agp_amd64_init() is called via gart_iommu_init() earlier, agp_amd64_init() finishes without doing anything (when it is called via module_init).
--- drivers/char/agp/amd64-agp.c | 9 +++++---- 1 files changed, 5 insertions(+), 4 deletions(-)
drivers/char/agp/amd64-agp.c +5 -4
@@ -725,9 +725,14 @@ static struct pci_driver agp_amd64_pci_driver = {int __init agp_amd64_init (void ) {int err =0 ; +static int done =0 ;if (agp_off)return -EINVAL ; + +if (done++) +return agp_bridges_found ?0 : -ENODEV ; + err =pci_register_driver (&agp_amd64_pci_driver);if (err <0 )return err;
@@ -771,12 +776,8 @@ static void __exit agp_amd64_cleanup(void)pci_unregister_driver (&agp_amd64_pci_driver); } -/* On AMD64 the PCI driver needs to initialize this driver early -for the IOMMU, so it has to be called via a backdoor. */ -#ifndef CONFIG_GART_IOMMU module_init (agp_amd64_init);module_exit (agp_amd64_cleanup); -#endif MODULE_AUTHOR ("Dave Jones <[email protected]>, Andi Kleen" );module_param (agp_try_unsupported, bool,0 );
Signature · 1 line
1.5.6.5
[tip:x86/urgent] x86/agp: Fix agp_amd64_init() initialization with CONFIG_GART_IOMMU enabled
Commit-ID: f405d2c02395a74d3883bd03ded36457aa3697ad Gitweb: http://git.kernel.org/tip/f405d2c02395a74d3883bd03ded36457aa3697ad Author: FUJITA Tomonori <[email protected]> AuthorDate: Mon, 28 Dec 2009 18:11:56 +0900 Committer: Ingo Molnar <[email protected]> CommitDate: Wed, 30 Dec 2009 11:52:04 +0100
x86/agp: Fix agp_amd64_init() initialization with CONFIG_GART_IOMMU enabled
with CONFIG_GART_IOMMU enabled drivers/char/agp/amd64-agp.c has:
#ifndef CONFIG_GART_IOMMU
module_init(agp_amd64_init);
module_exit(agp_amd64_cleanup);
#endif
agp_amd64_init() was called via gart_iommu_init with CONFIG_GART_IOMMU=y agp_amd64_init() was called via module_init with CONFIG_GART_IOMMU=n
The commit 75f1cdf1dda92cae037ec848ae63690d91913eac changes the x86 dma initialization routine: gart_iommu_init() is called only when GART IOMMU is detected. So when GART IOMMU isn't detected, agp_amd64_init isn't called.
Marin Mitov reported this issue:
http://marc.info/?l=linux-kernel&m=126192729110083&w=2
With this patch, agp_amd64_init() is always called via module_init (the above ifndef is removed). If agp_amd64_init() is called via gart_iommu_init() earlier, agp_amd64_init() finishes without doing anything (when it is called via module_init).
LKML-Reference: <[email protected]>
--- drivers/char/agp/amd64-agp.c | 9 +++++---- 1 files changed, 5 insertions(+), 4 deletions(-)
drivers/char/agp/amd64-agp.c +5 -4
@@ -725,9 +725,14 @@ static struct pci_driver agp_amd64_pci_driver = {int __init agp_amd64_init (void ) {int err =0 ; +static int done =0 ;if (agp_off)return -EINVAL ; + +if (done++) +return agp_bridges_found ?0 : -ENODEV ; + err =pci_register_driver (&agp_amd64_pci_driver);if (err <0 )return err;
@@ -771,12 +776,8 @@ static void __exit agp_amd64_cleanup(void)pci_unregister_driver (&agp_amd64_pci_driver); } -/* On AMD64 the PCI driver needs to initialize this driver early -for the IOMMU, so it has to be called via a backdoor. */ -#ifndef CONFIG_GART_IOMMU module_init (agp_amd64_init);module_exit (agp_amd64_cleanup); -#endif MODULE_AUTHOR ("Dave Jones <[email protected]>, Andi Kleen" );module_param (agp_try_unsupported, bool,0 );