* Re: SMP lockup at boot on Freescale/NXP T2080 (powerpc 64)
From: Christophe Leroy @ 2019-08-08 8:46 UTC (permalink / raw)
To: Chris Packham, linuxppc-dev@lists.ozlabs.org, mpe@ellerman.id.au,
npiggin@gmail.com
Cc: linux-kernel@vger.kernel.org, Grant McEwan
In-Reply-To: <1565141097.19352.12.camel@alliedtelesis.co.nz>
Le 07/08/2019 à 03:24, Chris Packham a écrit :
> On Wed, 2019-08-07 at 11:13 +1000, Michael Ellerman wrote:
>> Chris Packham <Chris.Packham@alliedtelesis.co.nz> writes:
>>>
>>> On Tue, 2019-08-06 at 21:32 +1000, Michael Ellerman wrote:
>>> The difference between a working and non working defconfig is
>>> CONFIG_PREEMPT specifically CONFIG_PREEMPT=y makes my system hang
>>> at
>>> boot.
>>>
>>> Is that now intentionally prohibited on 64-bit powerpc?
>> It's not prohibitied, but it probably should be because no one really
>> tests it properly. I have a handful of IBM machines where I boot a
>> PREEMPT kernel but that's about it.
>>
>> The corenet configs don't have PREEMPT enabled, which suggests it was
>> never really supported on those machines.
>>
>> But maybe someone from NXP can tell me otherwise.
>>
>
> I think our workloads need CONFIG_PREEMPT=y because our systems have
> switch ASIC drivers implemented in userland and we need to be able to
> react quickly to network events in order to prevent loops. We have seen
> instances of this not happening simply because some other process is in
> the middle of a syscall.
>
> One thing I am working on here is a setup with a few vendor boards and
> some of our own kit that we can test the upstream kernels on. Hopefully
> that'd make these kinds of reports more timely rather than just
> whenever we decide to move to a new kernel version.
>
>
The defconfig also sets CONFIG_DEBUG_PREEMPT. Have you tried without
CONFIG_DEBUG_PREEMPT ?
Christophe
^ permalink raw reply
* Re: [PATCH v4 7/9] powerpc/eeh: Add bdfn field to eeh_dev
From: Oliver O'Halloran @ 2019-08-08 9:08 UTC (permalink / raw)
To: Jordan Niethe; +Cc: Sam Bobroff, linuxppc-dev
In-Reply-To: <1da83896323a35fc91312cd8eaf3a82b819f84eb.camel@gmail.com>
On Thu, Aug 8, 2019 at 5:05 PM Jordan Niethe <jniethe5@gmail.com> wrote:
>
> On Wed, 2019-08-07 at 13:44 +1000, Sam Bobroff wrote:
> > From: Oliver O'Halloran <oohall@gmail.com>
> >
> > Preparation for removing pci_dn from the powernv EEH code. The only
> > thing we really use pci_dn for is to get the bdfn of the device for
> > config space accesses, so adding that information to eeh_dev reduces
> > the need to carry around the pci_dn.
> >
> > Signed-off-by: Oliver O'Halloran <oohall@gmail.com>
> > [SB: Re-wrapped commit message, fixed whitespace damage.]
> > Signed-off-by: Sam Bobroff <sbobroff@linux.ibm.com>
> > ---
> > arch/powerpc/include/asm/eeh.h | 2 ++
> > arch/powerpc/include/asm/ppc-pci.h | 2 ++
> > arch/powerpc/kernel/eeh_dev.c | 2 ++
> > 3 files changed, 6 insertions(+)
> >
> > diff --git a/arch/powerpc/include/asm/eeh.h
> > b/arch/powerpc/include/asm/eeh.h
> > index 7f9404a0c3bb..bbe0798f6624 100644
> > --- a/arch/powerpc/include/asm/eeh.h
> > +++ b/arch/powerpc/include/asm/eeh.h
> > @@ -121,6 +121,8 @@ static inline bool eeh_pe_passed(struct eeh_pe
> > *pe)
> > struct eeh_dev {
> > int mode; /* EEH mode */
> > int class_code; /* Class code of the device
> > */
> > + int bdfn; /* bdfn of device (for cfg ops) */
> > + struct pci_controller *controller;
>
> The other members of the structure get a comment, maybe it would be
> more consistant if this one did too?
At some point we need to go through all the EEH documentation /
comments and get rid of everything that is not useful or just straight
up wrong. The comments here are in-offensive, but they mostly just
repeat the information in the variable name so it's hard to see the
value.
^ permalink raw reply
* Re: [PATCH v2 0/3] arm/arm64: Add support for function error injection
From: Leo Yan @ 2019-08-08 9:09 UTC (permalink / raw)
To: Will Deacon
Cc: Song Liu, Alexei Starovoitov, Oleg Nesterov, Paul Mackerras,
H. Peter Anvin, linux-arch, Daniel Borkmann, x86, Russell King,
clang-built-linux, Ingo Molnar, Catalin Marinas, Yonghong Song,
Naveen N. Rao, Arnd Bergmann, Borislav Petkov, Thomas Gleixner,
linux-arm-kernel, netdev, linux-kernel, Masami Hiramatsu, bpf,
linuxppc-dev, Martin KaFai Lau
In-Reply-To: <20190807160703.pe4jxak7hs7ptvde@willie-the-truck>
On Wed, Aug 07, 2019 at 05:07:03PM +0100, Will Deacon wrote:
> On Tue, Aug 06, 2019 at 06:00:12PM +0800, Leo Yan wrote:
> > This small patch set is to add support for function error injection;
> > this can be used to eanble more advanced debugging feature, e.g.
> > CONFIG_BPF_KPROBE_OVERRIDE.
> >
> > The patch 01/03 is to consolidate the function definition which can be
> > suared cross architectures, patches 02,03/03 are used for enabling
> > function error injection on arm64 and arm architecture respectively.
> >
> > I tested on arm64 platform Juno-r2 and one of my laptop with x86
> > architecture with below steps; I don't test for Arm architecture so
> > only pass compilation.
>
> Thanks. I've queued the first two patches up here:
>
> https://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux.git/log/?h=for-next/error-injection
Thank you, Will.
Leo.
^ permalink raw reply
* [PATCH v2] powerpc/fadump: sysfs for fadump memory reservation
From: Sourabh Jain @ 2019-08-08 10:08 UTC (permalink / raw)
To: mpe
Cc: corbet, mahesh, linux-doc, linux-kernel, Sourabh Jain,
linuxppc-dev, hbathini
Add a sys interface to allow querying the memory reserved by
fadump for saving the crash dump.
Add an ABI doc entry for new sysfs interface.
- /sys/kernel/fadump_mem_reserved
Signed-off-by: Sourabh Jain <sourabhjain@linux.ibm.com>
---
v1 -> v2:
- Added ABI doc for new sysfs interface.
---
Documentation/ABI/testing/sysfs-kernel-fadump | 6 ++++++
Documentation/powerpc/firmware-assisted-dump.rst | 5 +++++
arch/powerpc/kernel/fadump.c | 14 ++++++++++++++
3 files changed, 25 insertions(+)
create mode 100644 Documentation/ABI/testing/sysfs-kernel-fadump
diff --git a/Documentation/ABI/testing/sysfs-kernel-fadump b/Documentation/ABI/testing/sysfs-kernel-fadump
new file mode 100644
index 000000000000..003e2f025dcb
--- /dev/null
+++ b/Documentation/ABI/testing/sysfs-kernel-fadump
@@ -0,0 +1,6 @@
+What: /sys/kernel/fadump_mem_reserved
+Date: August 2019
+Contact: linuxppc-dev@lists.ozlabs.org
+Description: read only
+ Provide information about the amount of memory
+ reserved by fadump to saving the crash dump.
diff --git a/Documentation/powerpc/firmware-assisted-dump.rst b/Documentation/powerpc/firmware-assisted-dump.rst
index 9ca12830a48e..a5dfb20d4dc3 100644
--- a/Documentation/powerpc/firmware-assisted-dump.rst
+++ b/Documentation/powerpc/firmware-assisted-dump.rst
@@ -222,6 +222,11 @@ Here is the list of files under kernel sysfs:
be handled and vmcore will not be captured. This interface can be
easily integrated with kdump service start/stop.
+ /sys/kernel/fadump_mem_reserved
+
+ This is used to display the memory reserved by fadump for saving the
+ crash dump.
+
/sys/kernel/fadump_release_mem
This file is available only when fadump is active during
second kernel. This is used to release the reserved memory
diff --git a/arch/powerpc/kernel/fadump.c b/arch/powerpc/kernel/fadump.c
index 4eab97292cc2..cd373d1d4b82 100644
--- a/arch/powerpc/kernel/fadump.c
+++ b/arch/powerpc/kernel/fadump.c
@@ -1514,6 +1514,13 @@ static ssize_t fadump_enabled_show(struct kobject *kobj,
return sprintf(buf, "%d\n", fw_dump.fadump_enabled);
}
+static ssize_t fadump_mem_reserved_show(struct kobject *kobj,
+ struct kobj_attribute *attr,
+ char *buf)
+{
+ return sprintf(buf, "%ld\n", fw_dump.reserve_dump_area_size);
+}
+
static ssize_t fadump_register_show(struct kobject *kobj,
struct kobj_attribute *attr,
char *buf)
@@ -1632,6 +1639,9 @@ static struct kobj_attribute fadump_attr = __ATTR(fadump_enabled,
static struct kobj_attribute fadump_register_attr = __ATTR(fadump_registered,
0644, fadump_register_show,
fadump_register_store);
+static struct kobj_attribute fadump_mem_reserved_attr =
+ __ATTR(fadump_mem_reserved, 0444,
+ fadump_mem_reserved_show, NULL);
DEFINE_SHOW_ATTRIBUTE(fadump_region);
@@ -1663,6 +1673,10 @@ static void fadump_init_files(void)
printk(KERN_ERR "fadump: unable to create sysfs file"
" fadump_release_mem (%d)\n", rc);
}
+ rc = sysfs_create_file(kernel_kobj, &fadump_mem_reserved_attr.attr);
+ if (rc)
+ pr_err("unable to create sysfs file fadump_mem_reserved (%d)\n",
+ rc);
return;
}
--
2.17.2
^ permalink raw reply related
* [Bug 204371] BUG kmalloc-4k (Tainted: G W ): Object padding overwritten
From: bugzilla-daemon @ 2019-08-08 10:53 UTC (permalink / raw)
To: linuxppc-dev
In-Reply-To: <bug-204371-206035@https.bugzilla.kernel.org/>
https://bugzilla.kernel.org/show_bug.cgi?id=204371
David Sterba (dsterba@suse.com) changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |dsterba@suse.com
--- Comment #9 from David Sterba (dsterba@suse.com) ---
I've hit the same problem, on x86_64.
--
You are receiving this mail because:
You are on the CC list for the bug.
^ permalink raw reply
* Re: [PATCH v2] powerpc/fadump: sysfs for fadump memory reservation
From: Hari Bathini @ 2019-08-08 11:17 UTC (permalink / raw)
To: Sourabh Jain, mpe; +Cc: mahesh, linux-doc, linuxppc-dev, linux-kernel, corbet
In-Reply-To: <20190808100833.30367-1-sourabhjain@linux.ibm.com>
On 08/08/19 3:38 PM, Sourabh Jain wrote:
> Add a sys interface to allow querying the memory reserved by
> fadump for saving the crash dump.
>
> Add an ABI doc entry for new sysfs interface.
> - /sys/kernel/fadump_mem_reserved
>
> Signed-off-by: Sourabh Jain <sourabhjain@linux.ibm.com>
> ---
> v1 -> v2:
> - Added ABI doc for new sysfs interface.
> ---
>
> Documentation/ABI/testing/sysfs-kernel-fadump | 6 ++++++
> Documentation/powerpc/firmware-assisted-dump.rst | 5 +++++
> arch/powerpc/kernel/fadump.c | 14 ++++++++++++++
> 3 files changed, 25 insertions(+)
> create mode 100644 Documentation/ABI/testing/sysfs-kernel-fadump
>
> diff --git a/Documentation/ABI/testing/sysfs-kernel-fadump b/Documentation/ABI/testing/sysfs-kernel-fadump
> new file mode 100644
> index 000000000000..003e2f025dcb
> --- /dev/null
> +++ b/Documentation/ABI/testing/sysfs-kernel-fadump
> @@ -0,0 +1,6 @@
> +What: /sys/kernel/fadump_mem_reserved
> +Date: August 2019
> +Contact: linuxppc-dev@lists.ozlabs.org
> +Description: read only
> + Provide information about the amount of memory
> + reserved by fadump to saving the crash dump.
s/to saving/to save/
Rest looks good..
Thanks
Hari
^ permalink raw reply
* [Bug 204371] BUG kmalloc-4k (Tainted: G W ): Object padding overwritten
From: bugzilla-daemon @ 2019-08-08 11:26 UTC (permalink / raw)
To: linuxppc-dev
In-Reply-To: <bug-204371-206035@https.bugzilla.kernel.org/>
https://bugzilla.kernel.org/show_bug.cgi?id=204371
--- Comment #10 from David Sterba (dsterba@suse.com) ---
In my case it happened on 5.3-rc3, with a strestest. The same machine has been
running fstests periodically, with slab debug on, but there are no slab reports
like that.
[ 8516.870046] BUG kmalloc-4k (Not tainted): Poison overwritten
[ 8516.875873]
-----------------------------------------------------------------------------
[ 8516.885864] Disabling lock debugging due to kernel taint
[ 8516.891312] INFO: 0x000000001c70c8c9-0x000000003cd1e164. First byte 0x16
instead of 0x6b
[ 8516.899717] INFO: Allocated in btrfs_read_tree_root+0x46/0x120 [btrfs]
age=1769 cpu=7 pid=8717
[ 8516.908544] __slab_alloc.isra.53+0x3e/0x70
[ 8516.912861] kmem_cache_alloc_trace+0x1b0/0x330
[ 8516.917581] btrfs_read_tree_root+0x46/0x120 [btrfs]
[ 8516.922737] btrfs_read_fs_root+0xe/0x40 [btrfs]
[ 8516.927552] create_reloc_root+0x17f/0x2a0 [btrfs]
[ 8516.932536] btrfs_init_reloc_root+0x72/0xe0 [btrfs]
[ 8516.937686] record_root_in_trans+0xbb/0xf0 [btrfs]
[ 8516.942750] btrfs_record_root_in_trans+0x50/0x70 [btrfs]
[ 8516.948340] start_transaction+0xa1/0x550 [btrfs]
[ 8516.953237] __btrfs_prealloc_file_range+0xca/0x490 [btrfs]
[ 8516.959003] btrfs_prealloc_file_range+0x10/0x20 [btrfs]
[ 8516.964509] prealloc_file_extent_cluster+0x13e/0x2b0 [btrfs]
[ 8516.970447] relocate_file_extent_cluster+0x8d/0x530 [btrfs]
[ 8516.976305] relocate_data_extent+0x80/0x110 [btrfs]
[ 8516.981469] relocate_block_group+0x473/0x720 [btrfs]
[ 8516.986711] btrfs_relocate_block_group+0x15f/0x2c0 [btrfs]
[ 8516.992470] INFO: Freed in btrfs_drop_snapshot+0x832/0xbb0 [btrfs] age=331
cpu=5 pid=8717
[ 8517.000865] kfree+0x29a/0x2d0
[ 8517.004098] btrfs_drop_snapshot+0x832/0xbb0 [btrfs]
[ 8517.009279] clean_dirty_subvols+0xf7/0x120 [btrfs]
[ 8517.014369] relocate_block_group+0x25a/0x720 [btrfs]
[ 8517.019616] btrfs_relocate_block_group+0x15f/0x2c0 [btrfs]
[ 8517.025385] btrfs_relocate_chunk+0x49/0x100 [btrfs]
[ 8517.030557] __btrfs_balance+0xa00/0xdb0 [btrfs]
[ 8517.035365] btrfs_balance+0x3b8/0xbb0 [btrfs]
[ 8517.040011] btrfs_ioctl_balance+0x2d5/0x380 [btrfs]
[ 8517.045176] btrfs_ioctl+0x16db/0x3460 [btrfs]
[ 8517.049772] do_vfs_ioctl+0xa5/0x710
[ 8517.053491] ksys_ioctl+0x70/0x80
[ 8517.056958] __x64_sys_ioctl+0x16/0x20
[ 8517.060845] do_syscall_64+0x5c/0x1d0
[ 8517.064650] entry_SYSCALL_64_after_hwframe+0x49/0xbe
[ 8518.630509] INFO: 0x00000000088ac804-0x00000000600f3eff. First byte 0x17
instead of 0x6b
[ 8518.640015] Object 0000000064763fee: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b
6b 6b 6b kkkkkkkkkkkkkkkk
[ 8518.650047] INFO: Allocated in btrfs_read_tree_root+0x46/0x120 [btrfs]
age=2298 cpu=4 pid=8634
[ 8518.658240] Object 000000001d16ab39: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b
6b 6b 6b kkkkkkkkkkkkkkkk
[ 8518.667744] __slab_alloc.isra.53+0x3e/0x70
[ 8518.667751] kmem_cache_alloc_trace+0x1b0/0x330
[ 8518.676569] Object 000000000f5b2c4b: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b
6b 6b 6b kkkkkkkkkkkkkkkk
[ 8518.686125] btrfs_read_tree_root+0x46/0x120 [btrfs]
[ 8518.686186] btrfs_read_fs_root+0xe/0x40 [btrfs]
[ 8518.690444] Object 000000000e589530: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b
6b 6b 6b kkkkkkkkkkkkkkkk
[ 8518.695159] create_reloc_root+0x17f/0x2a0 [btrfs]
[ 8518.695226] btrfs_init_reloc_root+0x72/0xe0 [btrfs]
[ 8518.704680] Object 00000000e3821ddd: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b
6b 6b 6b kkkkkkkkkkkkkkkk
[ 8518.709851] record_root_in_trans+0xbb/0xf0 [btrfs]
[ 8518.709912] btrfs_record_root_in_trans+0x50/0x70 [btrfs]
[ 8518.714606] Object 000000009552602b: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b
6b 6b 6b kkkkkkkkkkkkkkkk
[ 8518.724164] start_transaction+0xa1/0x550 [btrfs]
[ 8518.724225] btrfs_start_transaction_fallback_global_rsv+0x34/0x1f0 [btrfs]
[ 8518.729096] Object 00000000048bc005: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b
6b 6b 6b kkkkkkkkkkkkkkkk
[ 8518.734242] btrfs_unlink+0x34/0xd0 [btrfs]
[ 8518.734251] vfs_unlink+0x106/0x1f0
[ 8518.743763] Object 00000000e803d7b6: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b
6b 6b 6b kkkkkkkkkkkkkkkk
[ 8518.748767] do_unlinkat+0x2bf/0x330
[ 8518.748775] do_syscall_64+0x5c/0x1d0
[ 8518.754301] Object 00000000774a30d7: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b
6b 6b 6b kkkkkkkkkkkkkkkk
[ 8518.763804] entry_SYSCALL_64_after_hwframe+0x49/0xbe
[ 8518.763864] INFO: Freed in btrfs_drop_snapshot+0x832/0xbb0 [btrfs] age=746
cpu=5 pid=8717
[ 8518.768641] Object 000000007b92411f: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b
6b 6b 6b kkkkkkkkkkkkkkkk
[ 8518.775730] kfree+0x29a/0x2d0
[ 8518.775789] btrfs_drop_snapshot+0x832/0xbb0 [btrfs]
[ 8518.785253] Object 00000000ae532d5f: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b
6b 6b 6b kkkkkkkkkkkkkkkk
[ 8518.789626] clean_dirty_subvols+0xf7/0x120 [btrfs]
[ 8518.789693] relocate_block_group+0x25a/0x720 [btrfs]
[ 8518.793253] Object 000000002df294e8: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b
6b 6b 6b kkkkkkkkkkkkkkkk
[ 8518.802820] btrfs_relocate_block_group+0x15f/0x2c0 [btrfs]
[ 8518.802886] btrfs_relocate_chunk+0x49/0x100 [btrfs]
[ 8518.806528] Object 00000000df2dd63a: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b
6b 6b 6b kkkkkkkkkkkkkkkk
[ 8518.810370] __btrfs_balance+0xa00/0xdb0 [btrfs]
[ 8518.810437] btrfs_balance+0x3b8/0xbb0 [btrfs]
[ 8518.819894] Object 00000000682d1c71: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b
6b 6b 6b kkkkkkkkkkkkkkkk
[ 8518.825135] btrfs_ioctl_balance+0x2d5/0x380 [btrfs]
[ 8518.825202] btrfs_ioctl+0x16db/0x3460 [btrfs]
[ 8518.833522] Object 00000000eb8c2c61: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b
6b 6b 6b kkkkkkkkkkkkkkkk
[ 8518.843038] do_vfs_ioctl+0xa5/0x710
[ 8518.843044] ksys_ioctl+0x70/0x80
[ 8518.846228] Object 00000000574d97aa: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b
6b 6b 6b kkkkkkkkkkkkkkkk
[ 8518.851322] __x64_sys_ioctl+0x16/0x20
[ 8518.851329] do_syscall_64+0x5c/0x1d0
[ 8518.860844] Object 00000000a5c7d1b2: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b
6b 6b 6b kkkkkkkkkkkkkkkk
[ 8518.865870] entry_SYSCALL_64_after_hwframe+0x49/0xbe
[ 8518.865876] INFO: Slab 0x000000001ef6adf1 objects=7 used=7
fp=0x00000000b9747429 flags=0x3ffff000010200
[ 8518.871058] Object 000000009c9435a8: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b
6b 6b 6b kkkkkkkkkkkkkkkk
[ 8518.880559] INFO: Object 0x000000003bdbade7 @offset=8872
fp=0x00000000b9747429
[ 8522.364211] Redzone 00000000be2e5096: bb bb bb bb bb bb bb bb
........
[ 8522.364214] Padding 000000005d4fac5d: 5a 5a 5a 5a 5a 5a 5a 5a
ZZZZZZZZ
[ 8522.364228] CPU: 3 PID: 2817 Comm: tmux Tainted: G B
5.3.0-rc3-1.ge195904-vanilla+ #474
[ 8522.429558] Hardware name: empty empty/S3993, BIOS PAQEX0-3 02/24/2008
[ 8522.429561] Call Trace:
[ 8522.429581] dump_stack+0x67/0x9b
[ 8522.444139] check_bytes_and_report+0xc9/0xf0
[ 8522.444149] check_object+0x284/0x330
[ 8522.444157] ? __tty_buffer_request_room+0x94/0x1a0
[ 8522.444163] ? __tty_buffer_request_room+0x94/0x1a0
[ 8522.444169] alloc_debug_processing+0x197/0x200
[ 8522.444178] ___slab_alloc+0x500/0x620
[ 8522.470992] ? __tty_buffer_request_room+0x94/0x1a0
[ 8522.471005] ? stack_trace_save+0x70/0x70
[ 8522.480156] ? __tty_buffer_request_room+0x94/0x1a0
[ 8522.480162] ? __slab_alloc.isra.53+0x3e/0x70
[ 8522.489651] __slab_alloc.isra.53+0x3e/0x70
[ 8522.489676] ? __tty_buffer_request_room+0x94/0x1a0
[ 8522.489680] __kmalloc+0x25e/0x370
[ 8522.489689] __tty_buffer_request_room+0x94/0x1a0
[ 8522.507369] tty_insert_flip_string_fixed_flag+0x57/0x130
[ 8522.507412] pty_write+0x52/0x90
[ 8522.507421] n_tty_write+0x402/0x4f0
[ 8522.507433] ? do_wait_intr_irq+0xe0/0xe0
[ 8522.507443] tty_write+0x1a3/0x350
[ 8522.507450] ? process_echoes+0x60/0x60
[ 8522.507459] do_iter_write+0x182/0x1f0
[ 8522.507466] ? import_iovec+0x8b/0xb0
[ 8522.507473] vfs_writev+0x92/0x120
[ 8522.507497] ? do_writev+0xde/0x130
[ 8522.546489] do_writev+0xde/0x130
[ 8522.546500] do_syscall_64+0x5c/0x1d0
[ 8522.553740] entry_SYSCALL_64_after_hwframe+0x49/0xbe
--
You are receiving this mail because:
You are on the CC list for the bug.
^ permalink raw reply
* [PATCH] powerpc/64e: drop stale call to smp_processor_id() which hangs SMP startup
From: Christophe Leroy @ 2019-08-08 12:48 UTC (permalink / raw)
To: Benjamin Herrenschmidt, Paul Mackerras, Michael Ellerman,
Chris Packham
Cc: linuxppc-dev, linux-kernel
Santa commit ebb9d30a6a74 ("powerpc/mm: any thread in one core can be
the first to setup TLB1") removed the need to know the cpu_id in
early_init_this_mmu(), but the call to smp_processor_id() which was
marked __maybe_used remained.
Since commit ed1cd6deb013 ("powerpc: Activate
CONFIG_THREAD_INFO_IN_TASK") thread_info cannot be reached before mmu
is properly set up.
Drop this stale call to smp_processor_id() which make SMP hang
when CONFIG_PREEMPT is set.
Reported-by: Chris Packham <Chris.Packham@alliedtelesis.co.nz>
Fixes: ebb9d30a6a74 ("powerpc/mm: any thread in one core can be the first to setup TLB1")
Link: https://github.com/linuxppc/issues/issues/264
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
Cc: stable@vger.kernel.org
---
arch/powerpc/mm/nohash/tlb.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/arch/powerpc/mm/nohash/tlb.c b/arch/powerpc/mm/nohash/tlb.c
index d4acf6fa0596..bf60983a58c7 100644
--- a/arch/powerpc/mm/nohash/tlb.c
+++ b/arch/powerpc/mm/nohash/tlb.c
@@ -630,7 +630,6 @@ static void early_init_this_mmu(void)
#ifdef CONFIG_PPC_FSL_BOOK3E
if (mmu_has_feature(MMU_FTR_TYPE_FSL_E)) {
unsigned int num_cams;
- int __maybe_unused cpu = smp_processor_id();
bool map = true;
/* use a quarter of the TLBCAM for bolted linear map */
--
2.13.3
^ permalink raw reply related
* Re: SMP lockup at boot on Freescale/NXP T2080 (powerpc 64)
From: Christophe Leroy @ 2019-08-08 13:12 UTC (permalink / raw)
To: Chris Packham, linuxppc-dev@lists.ozlabs.org, mpe@ellerman.id.au,
npiggin@gmail.com
Cc: linux-kernel@vger.kernel.org, Grant McEwan
In-Reply-To: <b0ad453c-3a2b-6dd5-7bfc-5e275aa7bf62@c-s.fr>
Le 08/08/2019 à 10:46, Christophe Leroy a écrit :
>
>
> Le 07/08/2019 à 03:24, Chris Packham a écrit :
>> On Wed, 2019-08-07 at 11:13 +1000, Michael Ellerman wrote:
>>> Chris Packham <Chris.Packham@alliedtelesis.co.nz> writes:
>>>>
>>>> On Tue, 2019-08-06 at 21:32 +1000, Michael Ellerman wrote:
>>>> The difference between a working and non working defconfig is
>>>> CONFIG_PREEMPT specifically CONFIG_PREEMPT=y makes my system hang
>>>> at
>>>> boot.
>>>>
>>>> Is that now intentionally prohibited on 64-bit powerpc?
>>> It's not prohibitied, but it probably should be because no one really
>>> tests it properly. I have a handful of IBM machines where I boot a
>>> PREEMPT kernel but that's about it.
>>>
>>> The corenet configs don't have PREEMPT enabled, which suggests it was
>>> never really supported on those machines.
>>>
>>> But maybe someone from NXP can tell me otherwise.
>>>
>>
>> I think our workloads need CONFIG_PREEMPT=y because our systems have
>> switch ASIC drivers implemented in userland and we need to be able to
>> react quickly to network events in order to prevent loops. We have seen
>> instances of this not happening simply because some other process is in
>> the middle of a syscall.
>>
>> One thing I am working on here is a setup with a few vendor boards and
>> some of our own kit that we can test the upstream kernels on. Hopefully
>> that'd make these kinds of reports more timely rather than just
>> whenever we decide to move to a new kernel version.
>>
>>
>
>
> The defconfig also sets CONFIG_DEBUG_PREEMPT. Have you tried without
> CONFIG_DEBUG_PREEMPT ?
>
Reproduced on QEMU. CONFIG_DEBUG_PREEMPT is the trigger. Due to
smp_processor_id() being called from early_init_this_mmu(), when
CONFIG_DEBUG_PREEMPT is set debug_smp_processor_id() is called instead
of raw_smp_processor_id(), but this is too early for
debug_smp_processor_id()
As this call is useless, just drop it.
Can you test patch at https://patchwork.ozlabs.org/patch/1144005/ ?
Thanks
Christophe
^ permalink raw reply
* [Bug 204479] KASAN hit at modprobe zram
From: bugzilla-daemon @ 2019-08-08 13:57 UTC (permalink / raw)
To: linuxppc-dev
In-Reply-To: <bug-204479-206035@https.bugzilla.kernel.org/>
https://bugzilla.kernel.org/show_bug.cgi?id=204479
--- Comment #4 from Christophe Leroy (christophe.leroy@c-s.fr) ---
We need to identify if the allocation of KASAN shadow area at module allocation
fails, or if kasan accesses outside of the allocated area.
Could you please run again with the below trace:
diff --git a/arch/powerpc/mm/kasan/kasan_init_32.c
b/arch/powerpc/mm/kasan/kasan_init_32.c
index 74f4555a62ba..2bca2bf691a9 100644
--- a/arch/powerpc/mm/kasan/kasan_init_32.c
+++ b/arch/powerpc/mm/kasan/kasan_init_32.c
@@ -142,6 +142,9 @@ void *module_alloc(unsigned long size)
if (!base)
return NULL;
+ pr_err("###### module_alloc(%lx) = %px [%px-%px]\n", size, base,
+ kasan_mem_to_shadow(base), kasan_mem_to_shadow(base + size));
+
if (!kasan_init_region(base, size))
return base;
--
You are receiving this mail because:
You are on the CC list for the bug.
^ permalink raw reply related
* Re: [PATCH 0/2] ftrace: two fixes with func_probes handling
From: Naveen N. Rao @ 2019-08-08 15:15 UTC (permalink / raw)
To: Steven Rostedt; +Cc: linuxppc-dev, linux-kernel
In-Reply-To: <cover.1562249521.git.naveen.n.rao@linux.vnet.ibm.com>
Naveen N. Rao wrote:
> Two patches addressing bugs in ftrace function probe handling. The first
> patch addresses a NULL pointer dereference reported by LTP tests, while
> the second one is a trivial patch to address a missing check for return
> value, found by code inspection.
Steven,
Can you please take a look at these patches?
Thanks,
Naveen
^ permalink raw reply
* Re: [PATCH 0/2] ftrace: two fixes with func_probes handling
From: Steven Rostedt @ 2019-08-08 15:17 UTC (permalink / raw)
To: Naveen N. Rao; +Cc: linuxppc-dev, linux-kernel
In-Reply-To: <1565277255.xyq4vg2zkj.naveen@linux.ibm.com>
On Thu, 08 Aug 2019 20:45:04 +0530
"Naveen N. Rao" <naveen.n.rao@linux.vnet.ibm.com> wrote:
> Naveen N. Rao wrote:
> > Two patches addressing bugs in ftrace function probe handling. The first
> > patch addresses a NULL pointer dereference reported by LTP tests, while
> > the second one is a trivial patch to address a missing check for return
> > value, found by code inspection.
>
> Steven,
> Can you please take a look at these patches?
Thanks for the ping. Yes I will.
-- Steve
^ permalink raw reply
* [PATCH 1/8] dma-mapping: provide a better default ->get_required_mask
From: Christoph Hellwig @ 2019-08-08 15:59 UTC (permalink / raw)
To: iommu, Marek Szyprowski
Cc: linux-xtensa, Michal Simek, Vladimir Murzin, linux-parisc,
linux-sh, Takashi Iwai, linuxppc-dev, Helge Deller, x86,
linux-kernel, linux-m68k, Robin Murphy, linux-arm-kernel
In-Reply-To: <20190808160005.10325-1-hch@lst.de>
Most dma_map_ops instances are IOMMUs that work perfectly fine in 32-bits
of IOVA space, and the generic direct mapping code already provides its
own routines that is intelligent based on the amount of memory actually
present. Wire up the dma-direct routine for the ARM direct mapping code
as well, and otherwise default to the constant 32-bit mask. This way
we only need to override it for the occasional odd IOMMU that requires
64-bit IOVA support, or IOMMU drivers that are more efficient if they
can fall back to the direct mapping.
Signed-off-by: Christoph Hellwig <hch@lst.de>
---
arch/arm/mm/dma-mapping.c | 3 +++
arch/powerpc/platforms/ps3/system-bus.c | 7 ------
arch/x86/kernel/amd_gart_64.c | 1 +
kernel/dma/mapping.c | 30 +++++++++----------------
4 files changed, 14 insertions(+), 27 deletions(-)
diff --git a/arch/arm/mm/dma-mapping.c b/arch/arm/mm/dma-mapping.c
index 8c0b0baeb398..ad64d32fb39a 100644
--- a/arch/arm/mm/dma-mapping.c
+++ b/arch/arm/mm/dma-mapping.c
@@ -14,6 +14,7 @@
#include <linux/list.h>
#include <linux/init.h>
#include <linux/device.h>
+#include <linux/dma-direct.h>
#include <linux/dma-mapping.h>
#include <linux/dma-noncoherent.h>
#include <linux/dma-contiguous.h>
@@ -192,6 +193,7 @@ const struct dma_map_ops arm_dma_ops = {
.sync_sg_for_cpu = arm_dma_sync_sg_for_cpu,
.sync_sg_for_device = arm_dma_sync_sg_for_device,
.dma_supported = arm_dma_supported,
+ .get_required_mask = dma_direct_get_required_mask,
};
EXPORT_SYMBOL(arm_dma_ops);
@@ -212,6 +214,7 @@ const struct dma_map_ops arm_coherent_dma_ops = {
.map_sg = arm_dma_map_sg,
.map_resource = dma_direct_map_resource,
.dma_supported = arm_dma_supported,
+ .get_required_mask = dma_direct_get_required_mask,
};
EXPORT_SYMBOL(arm_coherent_dma_ops);
diff --git a/arch/powerpc/platforms/ps3/system-bus.c b/arch/powerpc/platforms/ps3/system-bus.c
index 98410119c47b..6818317073b9 100644
--- a/arch/powerpc/platforms/ps3/system-bus.c
+++ b/arch/powerpc/platforms/ps3/system-bus.c
@@ -686,18 +686,12 @@ static int ps3_dma_supported(struct device *_dev, u64 mask)
return mask >= DMA_BIT_MASK(32);
}
-static u64 ps3_dma_get_required_mask(struct device *_dev)
-{
- return DMA_BIT_MASK(32);
-}
-
static const struct dma_map_ops ps3_sb_dma_ops = {
.alloc = ps3_alloc_coherent,
.free = ps3_free_coherent,
.map_sg = ps3_sb_map_sg,
.unmap_sg = ps3_sb_unmap_sg,
.dma_supported = ps3_dma_supported,
- .get_required_mask = ps3_dma_get_required_mask,
.map_page = ps3_sb_map_page,
.unmap_page = ps3_unmap_page,
};
@@ -708,7 +702,6 @@ static const struct dma_map_ops ps3_ioc0_dma_ops = {
.map_sg = ps3_ioc0_map_sg,
.unmap_sg = ps3_ioc0_unmap_sg,
.dma_supported = ps3_dma_supported,
- .get_required_mask = ps3_dma_get_required_mask,
.map_page = ps3_ioc0_map_page,
.unmap_page = ps3_unmap_page,
};
diff --git a/arch/x86/kernel/amd_gart_64.c b/arch/x86/kernel/amd_gart_64.c
index a585ea6f686a..03ed9675f954 100644
--- a/arch/x86/kernel/amd_gart_64.c
+++ b/arch/x86/kernel/amd_gart_64.c
@@ -678,6 +678,7 @@ static const struct dma_map_ops gart_dma_ops = {
.alloc = gart_alloc_coherent,
.free = gart_free_coherent,
.dma_supported = dma_direct_supported,
+ .get_required_mask = dma_direct_get_required_mask,
};
static void gart_iommu_shutdown(void)
diff --git a/kernel/dma/mapping.c b/kernel/dma/mapping.c
index b0038ca3aa92..9c0f6a8eb5cb 100644
--- a/kernel/dma/mapping.c
+++ b/kernel/dma/mapping.c
@@ -233,25 +233,6 @@ int dma_mmap_attrs(struct device *dev, struct vm_area_struct *vma,
}
EXPORT_SYMBOL(dma_mmap_attrs);
-static u64 dma_default_get_required_mask(struct device *dev)
-{
- u32 low_totalram = ((max_pfn - 1) << PAGE_SHIFT);
- u32 high_totalram = ((max_pfn - 1) >> (32 - PAGE_SHIFT));
- u64 mask;
-
- if (!high_totalram) {
- /* convert to mask just covering totalram */
- low_totalram = (1 << (fls(low_totalram) - 1));
- low_totalram += low_totalram - 1;
- mask = low_totalram;
- } else {
- high_totalram = (1 << (fls(high_totalram) - 1));
- high_totalram += high_totalram - 1;
- mask = (((u64)high_totalram) << 32) + 0xffffffff;
- }
- return mask;
-}
-
u64 dma_get_required_mask(struct device *dev)
{
const struct dma_map_ops *ops = get_dma_ops(dev);
@@ -260,7 +241,16 @@ u64 dma_get_required_mask(struct device *dev)
return dma_direct_get_required_mask(dev);
if (ops->get_required_mask)
return ops->get_required_mask(dev);
- return dma_default_get_required_mask(dev);
+
+ /*
+ * We require every DMA ops implementation to at least support a 32-bit
+ * DMA mask (and use bounce buffering if that isn't supported in
+ * hardware). As the direct mapping code has its own routine to
+ * actually report an optimal mask we default to 32-bit here as that
+ * is the right thing for most IOMMUs, and at least not actively
+ * harmful in general.
+ */
+ return DMA_BIT_MASK(32);
}
EXPORT_SYMBOL_GPL(dma_get_required_mask);
--
2.20.1
^ permalink raw reply related
* remove default fallbacks in dma_map_ops v3
From: Christoph Hellwig @ 2019-08-08 15:59 UTC (permalink / raw)
To: iommu, Marek Szyprowski
Cc: linux-xtensa, Michal Simek, Vladimir Murzin, linux-parisc,
linux-sh, Takashi Iwai, linuxppc-dev, Helge Deller, x86,
linux-kernel, linux-m68k, Robin Murphy, linux-arm-kernel
Hi all,
we have a few places where the DMA mapping layer has non-trivial default
actions that are questionable and/or dangerous.
This series instead wires up the mmap, get_sgtable and get_required_mask
methods explicitly and cleans up some surrounding areas. This also means
we could get rid of the ARCH_NO_COHERENT_DMA_MMAP kconfig option, as we
now require a mmap method wired up, or in case of non-coherent dma-direct
the presence of the arch_dma_coherent_to_pfn hook. The only interesting
case is that the sound code also checked the ARCH_NO_COHERENT_DMA_MMAP
symbol in somewhat odd ways, so I'd like to see a review of the sound
situation before going forward with that patch.
Changes since v2:
- fix use of dma_can_mmap in alsa
- improve the CONFIG_* mess a little more
Changes since v1:
- add a dma_can_mmap helper for alsa
^ permalink raw reply
* [PATCH 2/8] dma-mapping: move the dma_get_sgtable API comments from arm to common code
From: Christoph Hellwig @ 2019-08-08 15:59 UTC (permalink / raw)
To: iommu, Marek Szyprowski
Cc: linux-xtensa, Michal Simek, Vladimir Murzin, linux-parisc,
linux-sh, Takashi Iwai, linuxppc-dev, Helge Deller, x86,
linux-kernel, linux-m68k, Robin Murphy, linux-arm-kernel
In-Reply-To: <20190808160005.10325-1-hch@lst.de>
The comments are spot on and should be near the central API, not just
near a single implementation.
Signed-off-by: Christoph Hellwig <hch@lst.de>
---
arch/arm/mm/dma-mapping.c | 11 -----------
kernel/dma/mapping.c | 11 +++++++++++
2 files changed, 11 insertions(+), 11 deletions(-)
diff --git a/arch/arm/mm/dma-mapping.c b/arch/arm/mm/dma-mapping.c
index ad64d32fb39a..b4d65da76393 100644
--- a/arch/arm/mm/dma-mapping.c
+++ b/arch/arm/mm/dma-mapping.c
@@ -880,17 +880,6 @@ static void arm_coherent_dma_free(struct device *dev, size_t size, void *cpu_add
__arm_dma_free(dev, size, cpu_addr, handle, attrs, true);
}
-/*
- * The whole dma_get_sgtable() idea is fundamentally unsafe - it seems
- * that the intention is to allow exporting memory allocated via the
- * coherent DMA APIs through the dma_buf API, which only accepts a
- * scattertable. This presents a couple of problems:
- * 1. Not all memory allocated via the coherent DMA APIs is backed by
- * a struct page
- * 2. Passing coherent DMA memory into the streaming APIs is not allowed
- * as we will try to flush the memory through a different alias to that
- * actually being used (and the flushes are redundant.)
- */
int arm_dma_get_sgtable(struct device *dev, struct sg_table *sgt,
void *cpu_addr, dma_addr_t handle, size_t size,
unsigned long attrs)
diff --git a/kernel/dma/mapping.c b/kernel/dma/mapping.c
index 9c0f6a8eb5cb..41590d003465 100644
--- a/kernel/dma/mapping.c
+++ b/kernel/dma/mapping.c
@@ -136,6 +136,17 @@ int dma_common_get_sgtable(struct device *dev, struct sg_table *sgt,
return ret;
}
+/*
+ * The whole dma_get_sgtable() idea is fundamentally unsafe - it seems
+ * that the intention is to allow exporting memory allocated via the
+ * coherent DMA APIs through the dma_buf API, which only accepts a
+ * scattertable. This presents a couple of problems:
+ * 1. Not all memory allocated via the coherent DMA APIs is backed by
+ * a struct page
+ * 2. Passing coherent DMA memory into the streaming APIs is not allowed
+ * as we will try to flush the memory through a different alias to that
+ * actually being used (and the flushes are redundant.)
+ */
int dma_get_sgtable_attrs(struct device *dev, struct sg_table *sgt,
void *cpu_addr, dma_addr_t dma_addr, size_t size,
unsigned long attrs)
--
2.20.1
^ permalink raw reply related
* [PATCH 3/8] dma-mapping: explicitly wire up ->mmap and ->get_sgtable
From: Christoph Hellwig @ 2019-08-08 16:00 UTC (permalink / raw)
To: iommu, Marek Szyprowski
Cc: linux-xtensa, Michal Simek, Vladimir Murzin, linux-parisc,
linux-sh, Takashi Iwai, linuxppc-dev, Helge Deller, x86,
linux-kernel, linux-m68k, Robin Murphy, linux-arm-kernel
In-Reply-To: <20190808160005.10325-1-hch@lst.de>
While the default ->mmap and ->get_sgtable implementations work for the
majority of our dma_map_ops impementations they are inherently safe
for others that don't use the page allocator or CMA and/or use their
own way of remapping not covered by the common code. So remove the
defaults if these methods are not wired up, but instead wire up the
default implementations for all safe instances.
Fixes: e1c7e324539a ("dma-mapping: always provide the dma_map_ops based implementation")
Signed-off-by: Christoph Hellwig <hch@lst.de>
---
arch/alpha/kernel/pci_iommu.c | 2 ++
arch/ia64/hp/common/sba_iommu.c | 2 ++
arch/ia64/sn/pci/pci_dma.c | 2 ++
arch/mips/jazz/jazzdma.c | 2 ++
arch/powerpc/kernel/dma-iommu.c | 2 ++
arch/powerpc/platforms/ps3/system-bus.c | 4 ++++
arch/powerpc/platforms/pseries/vio.c | 2 ++
arch/s390/pci/pci_dma.c | 2 ++
arch/x86/kernel/amd_gart_64.c | 2 ++
arch/x86/kernel/pci-calgary_64.c | 2 ++
drivers/iommu/amd_iommu.c | 2 ++
drivers/iommu/intel-iommu.c | 2 ++
drivers/parisc/ccio-dma.c | 2 ++
drivers/parisc/sba_iommu.c | 2 ++
kernel/dma/mapping.c | 20 ++++++++++++--------
15 files changed, 42 insertions(+), 8 deletions(-)
diff --git a/arch/alpha/kernel/pci_iommu.c b/arch/alpha/kernel/pci_iommu.c
index 242108439f42..7f1925a32c99 100644
--- a/arch/alpha/kernel/pci_iommu.c
+++ b/arch/alpha/kernel/pci_iommu.c
@@ -955,5 +955,7 @@ const struct dma_map_ops alpha_pci_ops = {
.map_sg = alpha_pci_map_sg,
.unmap_sg = alpha_pci_unmap_sg,
.dma_supported = alpha_pci_supported,
+ .mmap = dma_common_mmap,
+ .get_sgtable = dma_common_get_sgtable,
};
EXPORT_SYMBOL(alpha_pci_ops);
diff --git a/arch/ia64/hp/common/sba_iommu.c b/arch/ia64/hp/common/sba_iommu.c
index 3d24cc43385b..4c0ea6c2833d 100644
--- a/arch/ia64/hp/common/sba_iommu.c
+++ b/arch/ia64/hp/common/sba_iommu.c
@@ -2183,6 +2183,8 @@ const struct dma_map_ops sba_dma_ops = {
.map_sg = sba_map_sg_attrs,
.unmap_sg = sba_unmap_sg_attrs,
.dma_supported = sba_dma_supported,
+ .mmap = dma_common_mmap,
+ .get_sgtable = dma_common_get_sgtable,
};
void sba_dma_init(void)
diff --git a/arch/ia64/sn/pci/pci_dma.c b/arch/ia64/sn/pci/pci_dma.c
index b7d42e4edc1f..12ffb9c0d738 100644
--- a/arch/ia64/sn/pci/pci_dma.c
+++ b/arch/ia64/sn/pci/pci_dma.c
@@ -438,6 +438,8 @@ static struct dma_map_ops sn_dma_ops = {
.unmap_sg = sn_dma_unmap_sg,
.dma_supported = sn_dma_supported,
.get_required_mask = sn_dma_get_required_mask,
+ .mmap = dma_common_mmap,
+ .get_sgtable = dma_common_get_sgtable,
};
void sn_dma_init(void)
diff --git a/arch/mips/jazz/jazzdma.c b/arch/mips/jazz/jazzdma.c
index 1804dc9d8136..a01e14955187 100644
--- a/arch/mips/jazz/jazzdma.c
+++ b/arch/mips/jazz/jazzdma.c
@@ -682,5 +682,7 @@ const struct dma_map_ops jazz_dma_ops = {
.sync_sg_for_device = jazz_dma_sync_sg_for_device,
.dma_supported = dma_direct_supported,
.cache_sync = arch_dma_cache_sync,
+ .mmap = dma_common_mmap,
+ .get_sgtable = dma_common_get_sgtable,
};
EXPORT_SYMBOL(jazz_dma_ops);
diff --git a/arch/powerpc/kernel/dma-iommu.c b/arch/powerpc/kernel/dma-iommu.c
index a0879674a9c8..2f5a53874f6d 100644
--- a/arch/powerpc/kernel/dma-iommu.c
+++ b/arch/powerpc/kernel/dma-iommu.c
@@ -208,4 +208,6 @@ const struct dma_map_ops dma_iommu_ops = {
.sync_single_for_device = dma_iommu_sync_for_device,
.sync_sg_for_cpu = dma_iommu_sync_sg_for_cpu,
.sync_sg_for_device = dma_iommu_sync_sg_for_device,
+ .mmap = dma_common_mmap,
+ .get_sgtable = dma_common_get_sgtable,
};
diff --git a/arch/powerpc/platforms/ps3/system-bus.c b/arch/powerpc/platforms/ps3/system-bus.c
index 6818317073b9..3542b7bd6a46 100644
--- a/arch/powerpc/platforms/ps3/system-bus.c
+++ b/arch/powerpc/platforms/ps3/system-bus.c
@@ -694,6 +694,8 @@ static const struct dma_map_ops ps3_sb_dma_ops = {
.dma_supported = ps3_dma_supported,
.map_page = ps3_sb_map_page,
.unmap_page = ps3_unmap_page,
+ .mmap = dma_common_mmap,
+ .get_sgtable = dma_common_get_sgtable,
};
static const struct dma_map_ops ps3_ioc0_dma_ops = {
@@ -704,6 +706,8 @@ static const struct dma_map_ops ps3_ioc0_dma_ops = {
.dma_supported = ps3_dma_supported,
.map_page = ps3_ioc0_map_page,
.unmap_page = ps3_unmap_page,
+ .mmap = dma_common_mmap,
+ .get_sgtable = dma_common_get_sgtable,
};
/**
diff --git a/arch/powerpc/platforms/pseries/vio.c b/arch/powerpc/platforms/pseries/vio.c
index 6601b9d404dc..3473eef7628c 100644
--- a/arch/powerpc/platforms/pseries/vio.c
+++ b/arch/powerpc/platforms/pseries/vio.c
@@ -605,6 +605,8 @@ static const struct dma_map_ops vio_dma_mapping_ops = {
.unmap_page = vio_dma_iommu_unmap_page,
.dma_supported = dma_iommu_dma_supported,
.get_required_mask = dma_iommu_get_required_mask,
+ .mmap = dma_common_mmap,
+ .get_sgtable = dma_common_get_sgtable,
};
/**
diff --git a/arch/s390/pci/pci_dma.c b/arch/s390/pci/pci_dma.c
index 9e52d1527f71..03d8c1c9f82f 100644
--- a/arch/s390/pci/pci_dma.c
+++ b/arch/s390/pci/pci_dma.c
@@ -668,6 +668,8 @@ const struct dma_map_ops s390_pci_dma_ops = {
.unmap_sg = s390_dma_unmap_sg,
.map_page = s390_dma_map_pages,
.unmap_page = s390_dma_unmap_pages,
+ .mmap = dma_common_mmap,
+ .get_sgtable = dma_common_get_sgtable,
/* dma_supported is unconditionally true without a callback */
};
EXPORT_SYMBOL_GPL(s390_pci_dma_ops);
diff --git a/arch/x86/kernel/amd_gart_64.c b/arch/x86/kernel/amd_gart_64.c
index 03ed9675f954..a6ac3712db8b 100644
--- a/arch/x86/kernel/amd_gart_64.c
+++ b/arch/x86/kernel/amd_gart_64.c
@@ -677,6 +677,8 @@ static const struct dma_map_ops gart_dma_ops = {
.unmap_page = gart_unmap_page,
.alloc = gart_alloc_coherent,
.free = gart_free_coherent,
+ .mmap = dma_common_mmap,
+ .get_sgtable = dma_common_get_sgtable,
.dma_supported = dma_direct_supported,
.get_required_mask = dma_direct_get_required_mask,
};
diff --git a/arch/x86/kernel/pci-calgary_64.c b/arch/x86/kernel/pci-calgary_64.c
index 9d4343aa481b..23fdec030c37 100644
--- a/arch/x86/kernel/pci-calgary_64.c
+++ b/arch/x86/kernel/pci-calgary_64.c
@@ -468,6 +468,8 @@ static const struct dma_map_ops calgary_dma_ops = {
.map_page = calgary_map_page,
.unmap_page = calgary_unmap_page,
.dma_supported = dma_direct_supported,
+ .mmap = dma_common_mmap,
+ .get_sgtable = dma_common_get_sgtable,
};
static inline void __iomem * busno_to_bbar(unsigned char num)
diff --git a/drivers/iommu/amd_iommu.c b/drivers/iommu/amd_iommu.c
index b607a92791d3..2e74ad659985 100644
--- a/drivers/iommu/amd_iommu.c
+++ b/drivers/iommu/amd_iommu.c
@@ -2722,6 +2722,8 @@ static const struct dma_map_ops amd_iommu_dma_ops = {
.map_sg = map_sg,
.unmap_sg = unmap_sg,
.dma_supported = amd_iommu_dma_supported,
+ .mmap = dma_common_mmap,
+ .get_sgtable = dma_common_get_sgtable,
};
static int init_reserved_iova_ranges(void)
diff --git a/drivers/iommu/intel-iommu.c b/drivers/iommu/intel-iommu.c
index bdaed2da8a55..7505c5d9cf78 100644
--- a/drivers/iommu/intel-iommu.c
+++ b/drivers/iommu/intel-iommu.c
@@ -3737,6 +3737,8 @@ static const struct dma_map_ops intel_dma_ops = {
.map_resource = intel_map_resource,
.unmap_resource = intel_unmap_resource,
.dma_supported = dma_direct_supported,
+ .mmap = dma_common_mmap,
+ .get_sgtable = dma_common_get_sgtable,
};
static inline int iommu_domain_cache_init(void)
diff --git a/drivers/parisc/ccio-dma.c b/drivers/parisc/ccio-dma.c
index 217f15aafa4a..1d7125d29bee 100644
--- a/drivers/parisc/ccio-dma.c
+++ b/drivers/parisc/ccio-dma.c
@@ -1024,6 +1024,8 @@ static const struct dma_map_ops ccio_ops = {
.unmap_page = ccio_unmap_page,
.map_sg = ccio_map_sg,
.unmap_sg = ccio_unmap_sg,
+ .mmap = dma_common_mmap,
+ .get_sgtable = dma_common_get_sgtable,
};
#ifdef CONFIG_PROC_FS
diff --git a/drivers/parisc/sba_iommu.c b/drivers/parisc/sba_iommu.c
index 296668caf7e5..fa4df65b7e28 100644
--- a/drivers/parisc/sba_iommu.c
+++ b/drivers/parisc/sba_iommu.c
@@ -1084,6 +1084,8 @@ static const struct dma_map_ops sba_ops = {
.unmap_page = sba_unmap_page,
.map_sg = sba_map_sg,
.unmap_sg = sba_unmap_sg,
+ .mmap = dma_common_mmap,
+ .get_sgtable = dma_common_get_sgtable,
};
diff --git a/kernel/dma/mapping.c b/kernel/dma/mapping.c
index 41590d003465..67b900ad0836 100644
--- a/kernel/dma/mapping.c
+++ b/kernel/dma/mapping.c
@@ -153,11 +153,12 @@ int dma_get_sgtable_attrs(struct device *dev, struct sg_table *sgt,
{
const struct dma_map_ops *ops = get_dma_ops(dev);
- if (!dma_is_direct(ops) && ops->get_sgtable)
- return ops->get_sgtable(dev, sgt, cpu_addr, dma_addr, size,
- attrs);
- return dma_common_get_sgtable(dev, sgt, cpu_addr, dma_addr, size,
- attrs);
+ if (dma_is_direct(ops))
+ return dma_common_get_sgtable(dev, sgt, cpu_addr, dma_addr,
+ size, attrs);
+ if (!ops->get_sgtable)
+ return -ENXIO;
+ return ops->get_sgtable(dev, sgt, cpu_addr, dma_addr, size, attrs);
}
EXPORT_SYMBOL(dma_get_sgtable_attrs);
@@ -238,9 +239,12 @@ int dma_mmap_attrs(struct device *dev, struct vm_area_struct *vma,
{
const struct dma_map_ops *ops = get_dma_ops(dev);
- if (!dma_is_direct(ops) && ops->mmap)
- return ops->mmap(dev, vma, cpu_addr, dma_addr, size, attrs);
- return dma_common_mmap(dev, vma, cpu_addr, dma_addr, size, attrs);
+ if (dma_is_direct(ops))
+ return dma_common_mmap(dev, vma, cpu_addr, dma_addr, size,
+ attrs);
+ if (!ops->mmap)
+ return -ENXIO;
+ return ops->mmap(dev, vma, cpu_addr, dma_addr, size, attrs);
}
EXPORT_SYMBOL(dma_mmap_attrs);
--
2.20.1
^ permalink raw reply related
* [PATCH 4/8] dma-mapping: add a dma_can_mmap helper
From: Christoph Hellwig @ 2019-08-08 16:00 UTC (permalink / raw)
To: iommu, Marek Szyprowski
Cc: linux-xtensa, Michal Simek, Vladimir Murzin, linux-parisc,
linux-sh, Takashi Iwai, linuxppc-dev, Helge Deller, x86,
linux-kernel, linux-m68k, Robin Murphy, linux-arm-kernel
In-Reply-To: <20190808160005.10325-1-hch@lst.de>
Add a helper to check if DMA allocations for a specific device can be
mapped to userspace using dma_mmap_*.
Signed-off-by: Christoph Hellwig <hch@lst.de>
---
include/linux/dma-mapping.h | 5 +++++
kernel/dma/mapping.c | 23 +++++++++++++++++++++++
2 files changed, 28 insertions(+)
diff --git a/include/linux/dma-mapping.h b/include/linux/dma-mapping.h
index f7d1eea32c78..17271857be5d 100644
--- a/include/linux/dma-mapping.h
+++ b/include/linux/dma-mapping.h
@@ -462,6 +462,7 @@ int dma_get_sgtable_attrs(struct device *dev, struct sg_table *sgt,
int dma_mmap_attrs(struct device *dev, struct vm_area_struct *vma,
void *cpu_addr, dma_addr_t dma_addr, size_t size,
unsigned long attrs);
+bool dma_can_mmap(struct device *dev);
int dma_supported(struct device *dev, u64 mask);
int dma_set_mask(struct device *dev, u64 mask);
int dma_set_coherent_mask(struct device *dev, u64 mask);
@@ -552,6 +553,10 @@ static inline int dma_mmap_attrs(struct device *dev, struct vm_area_struct *vma,
{
return -ENXIO;
}
+static inline bool dma_can_mmap(struct device *dev)
+{
+ return false;
+}
static inline int dma_supported(struct device *dev, u64 mask)
{
return 0;
diff --git a/kernel/dma/mapping.c b/kernel/dma/mapping.c
index 67b900ad0836..64d1de59e133 100644
--- a/kernel/dma/mapping.c
+++ b/kernel/dma/mapping.c
@@ -220,6 +220,29 @@ int dma_common_mmap(struct device *dev, struct vm_area_struct *vma,
#endif /* !CONFIG_ARCH_NO_COHERENT_DMA_MMAP */
}
+/**
+ * dma_can_mmap - check if a given device supports dma_mmap_*
+ * @dev: device to check
+ *
+ * Returns %true if @dev supports dma_mmap_coherent() and dma_mmap_attrs() to
+ * map DMA allocations to userspace.
+ */
+bool dma_can_mmap(struct device *dev)
+{
+ const struct dma_map_ops *ops = get_dma_ops(dev);
+
+ if (IS_ENABLED(CONFIG_ARCH_NO_COHERENT_DMA_MMAP))
+ return false;
+
+ if (dma_is_direct(ops)) {
+ return dev_is_dma_coherent(dev) ||
+ IS_ENABLED(CONFIG_ARCH_HAS_DMA_COHERENT_TO_PFN);
+ }
+
+ return ops->mmap != NULL;
+}
+EXPORT_SYMBOL_GPL(dma_can_mmap);
+
/**
* dma_mmap_attrs - map a coherent DMA allocation into user space
* @dev: valid struct device pointer, or NULL for ISA and EISA-like devices
--
2.20.1
^ permalink raw reply related
* [PATCH 5/8] ALSA: pcm: use dma_can_mmap() to check if a device supports dma_mmap_*
From: Christoph Hellwig @ 2019-08-08 16:00 UTC (permalink / raw)
To: iommu, Marek Szyprowski
Cc: linux-xtensa, Michal Simek, Vladimir Murzin, linux-parisc,
linux-sh, Takashi Iwai, linuxppc-dev, Helge Deller, x86,
linux-kernel, linux-m68k, Robin Murphy, linux-arm-kernel
In-Reply-To: <20190808160005.10325-1-hch@lst.de>
Replace the local hack with the dma_can_mmap helper to check if
a given device supports mapping DMA allocations to userspace.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Takashi Iwai <tiwai@suse.de>
---
sound/core/pcm_native.c | 13 ++++++-------
1 file changed, 6 insertions(+), 7 deletions(-)
diff --git a/sound/core/pcm_native.c b/sound/core/pcm_native.c
index 703857aab00f..9763c18e176a 100644
--- a/sound/core/pcm_native.c
+++ b/sound/core/pcm_native.c
@@ -220,13 +220,12 @@ static bool hw_support_mmap(struct snd_pcm_substream *substream)
{
if (!(substream->runtime->hw.info & SNDRV_PCM_INFO_MMAP))
return false;
- /* architecture supports dma_mmap_coherent()? */
-#if defined(CONFIG_ARCH_NO_COHERENT_DMA_MMAP) || !defined(CONFIG_HAS_DMA)
- if (!substream->ops->mmap &&
- substream->dma_buffer.dev.type == SNDRV_DMA_TYPE_DEV)
- return false;
-#endif
- return true;
+
+ if (substream->ops->mmap ||
+ substream->dma_buffer.dev.type != SNDRV_DMA_TYPE_DEV)
+ return true;
+
+ return dma_can_mmap(substream->dma_buffer.dev.dev);
}
static int constrain_mask_params(struct snd_pcm_substream *substream,
--
2.20.1
^ permalink raw reply related
* [PATCH 6/8] arm-nommu: call dma_mmap_from_dev_coherent directly
From: Christoph Hellwig @ 2019-08-08 16:00 UTC (permalink / raw)
To: iommu, Marek Szyprowski
Cc: linux-xtensa, Michal Simek, Vladimir Murzin, linux-parisc,
linux-sh, Takashi Iwai, linuxppc-dev, Helge Deller, x86,
linux-kernel, linux-m68k, Robin Murphy, linux-arm-kernel
In-Reply-To: <20190808160005.10325-1-hch@lst.de>
Ther is no need to go through dma_common_mmap for the arm-nommu
dma mmap implementation as the only possible memory not handled above
could be that from the per-device coherent pool.
Signed-off-by: Christoph Hellwig <hch@lst.de>
---
arch/arm/mm/dma-mapping-nommu.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/arch/arm/mm/dma-mapping-nommu.c b/arch/arm/mm/dma-mapping-nommu.c
index 52b82559d99b..db9247898300 100644
--- a/arch/arm/mm/dma-mapping-nommu.c
+++ b/arch/arm/mm/dma-mapping-nommu.c
@@ -68,8 +68,9 @@ static int arm_nommu_dma_mmap(struct device *dev, struct vm_area_struct *vma,
if (dma_mmap_from_global_coherent(vma, cpu_addr, size, &ret))
return ret;
-
- return dma_common_mmap(dev, vma, cpu_addr, dma_addr, size, attrs);
+ if (dma_mmap_from_dev_coherent(dev, vma, cpu_addr, size, &ret))
+ return ret;
+ return -ENXIO;
}
--
2.20.1
^ permalink raw reply related
* [PATCH 7/8] parisc: don't set ARCH_NO_COHERENT_DMA_MMAP
From: Christoph Hellwig @ 2019-08-08 16:00 UTC (permalink / raw)
To: iommu, Marek Szyprowski
Cc: linux-xtensa, Michal Simek, Vladimir Murzin, linux-parisc,
linux-sh, Takashi Iwai, linuxppc-dev, Helge Deller, x86,
linux-kernel, linux-m68k, Robin Murphy, linux-arm-kernel
In-Reply-To: <20190808160005.10325-1-hch@lst.de>
parisc is the only architecture that sets ARCH_NO_COHERENT_DMA_MMAP
when an MMU is enabled. AFAIK this is because parisc CPUs use VIVT
caches, which means exporting normally cachable memory to userspace is
relatively dangrous due to cache aliasing.
But normally cachable memory is only allocated by dma_alloc_coherent
on parisc when using the sba_iommu or ccio_iommu drivers, so just
remove the .mmap implementation for them so that we don't have to set
ARCH_NO_COHERENT_DMA_MMAP, which I plan to get rid of.
Signed-off-by: Christoph Hellwig <hch@lst.de>
---
arch/parisc/Kconfig | 1 -
drivers/parisc/ccio-dma.c | 1 -
drivers/parisc/sba_iommu.c | 1 -
3 files changed, 3 deletions(-)
diff --git a/arch/parisc/Kconfig b/arch/parisc/Kconfig
index 6d732e451071..e9dd88b7f81e 100644
--- a/arch/parisc/Kconfig
+++ b/arch/parisc/Kconfig
@@ -52,7 +52,6 @@ config PARISC
select GENERIC_SCHED_CLOCK
select HAVE_UNSTABLE_SCHED_CLOCK if SMP
select GENERIC_CLOCKEVENTS
- select ARCH_NO_COHERENT_DMA_MMAP
select CPU_NO_EFFICIENT_FFS
select NEED_DMA_MAP_STATE
select NEED_SG_DMA_LENGTH
diff --git a/drivers/parisc/ccio-dma.c b/drivers/parisc/ccio-dma.c
index 1d7125d29bee..ad290f79983b 100644
--- a/drivers/parisc/ccio-dma.c
+++ b/drivers/parisc/ccio-dma.c
@@ -1024,7 +1024,6 @@ static const struct dma_map_ops ccio_ops = {
.unmap_page = ccio_unmap_page,
.map_sg = ccio_map_sg,
.unmap_sg = ccio_unmap_sg,
- .mmap = dma_common_mmap,
.get_sgtable = dma_common_get_sgtable,
};
diff --git a/drivers/parisc/sba_iommu.c b/drivers/parisc/sba_iommu.c
index fa4df65b7e28..ed50502cc65a 100644
--- a/drivers/parisc/sba_iommu.c
+++ b/drivers/parisc/sba_iommu.c
@@ -1084,7 +1084,6 @@ static const struct dma_map_ops sba_ops = {
.unmap_page = sba_unmap_page,
.map_sg = sba_map_sg,
.unmap_sg = sba_unmap_sg,
- .mmap = dma_common_mmap,
.get_sgtable = dma_common_get_sgtable,
};
--
2.20.1
^ permalink raw reply related
* [PATCH 8/8] dma-mapping: remove CONFIG_ARCH_NO_COHERENT_DMA_MMAP
From: Christoph Hellwig @ 2019-08-08 16:00 UTC (permalink / raw)
To: iommu, Marek Szyprowski
Cc: linux-xtensa, Michal Simek, Vladimir Murzin, linux-parisc,
linux-sh, Takashi Iwai, linuxppc-dev, Helge Deller, x86,
linux-kernel, linux-m68k, Robin Murphy, linux-arm-kernel
In-Reply-To: <20190808160005.10325-1-hch@lst.de>
CONFIG_ARCH_NO_COHERENT_DMA_MMAP is now functionally identical to
!CONFIG_MMU, so remove the separate symbol. The only difference is that
arm did not set it for !CONFIG_MMU, but arm uses a separate dma mapping
implementation including its own mmap method, which is handled by moving
the CONFIG_MMU check in dma_can_mmap so that is only applies to the
dma-direct case, just as the other ifdefs for it.
Signed-off-by: Christoph Hellwig <hch@lst.de>
---
arch/Kconfig | 3 ---
arch/c6x/Kconfig | 1 -
arch/m68k/Kconfig | 1 -
arch/microblaze/Kconfig | 1 -
arch/sh/Kconfig | 1 -
arch/xtensa/Kconfig | 1 -
kernel/dma/mapping.c | 12 +++++-------
7 files changed, 5 insertions(+), 15 deletions(-)
diff --git a/arch/Kconfig b/arch/Kconfig
index a7b57dd42c26..ec2834206d08 100644
--- a/arch/Kconfig
+++ b/arch/Kconfig
@@ -790,9 +790,6 @@ config COMPAT_32BIT_TIME
This is relevant on all 32-bit architectures, and 64-bit architectures
as part of compat syscall handling.
-config ARCH_NO_COHERENT_DMA_MMAP
- bool
-
config ARCH_NO_PREEMPT
bool
diff --git a/arch/c6x/Kconfig b/arch/c6x/Kconfig
index b4fb61c83494..e65e8d82442a 100644
--- a/arch/c6x/Kconfig
+++ b/arch/c6x/Kconfig
@@ -20,7 +20,6 @@ config C6X
select OF_EARLY_FLATTREE
select GENERIC_CLOCKEVENTS
select MODULES_USE_ELF_RELA
- select ARCH_NO_COHERENT_DMA_MMAP
select MMU_GATHER_NO_RANGE if MMU
config MMU
diff --git a/arch/m68k/Kconfig b/arch/m68k/Kconfig
index c518d695c376..614b355ae338 100644
--- a/arch/m68k/Kconfig
+++ b/arch/m68k/Kconfig
@@ -8,7 +8,6 @@ config M68K
select ARCH_HAS_DMA_PREP_COHERENT if HAS_DMA && MMU && !COLDFIRE
select ARCH_HAS_SYNC_DMA_FOR_DEVICE if HAS_DMA
select ARCH_MIGHT_HAVE_PC_PARPORT if ISA
- select ARCH_NO_COHERENT_DMA_MMAP if !MMU
select ARCH_NO_PREEMPT if !COLDFIRE
select BINFMT_FLAT_ARGVP_ENVP_ON_STACK
select DMA_DIRECT_REMAP if HAS_DMA && MMU && !COLDFIRE
diff --git a/arch/microblaze/Kconfig b/arch/microblaze/Kconfig
index d411de05b628..632c9477a0f6 100644
--- a/arch/microblaze/Kconfig
+++ b/arch/microblaze/Kconfig
@@ -9,7 +9,6 @@ config MICROBLAZE
select ARCH_HAS_SYNC_DMA_FOR_CPU
select ARCH_HAS_SYNC_DMA_FOR_DEVICE
select ARCH_MIGHT_HAVE_PC_PARPORT
- select ARCH_NO_COHERENT_DMA_MMAP if !MMU
select ARCH_WANT_IPC_PARSE_VERSION
select BUILDTIME_EXTABLE_SORT
select TIMER_OF
diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index 6b1b5941b618..f356ee674d89 100644
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -5,7 +5,6 @@ config SUPERH
select ARCH_HAS_PTE_SPECIAL
select ARCH_HAS_TICK_BROADCAST if GENERIC_CLOCKEVENTS_BROADCAST
select ARCH_MIGHT_HAVE_PC_PARPORT
- select ARCH_NO_COHERENT_DMA_MMAP if !MMU
select HAVE_PATA_PLATFORM
select CLKDEV_LOOKUP
select DMA_DECLARE_COHERENT
diff --git a/arch/xtensa/Kconfig b/arch/xtensa/Kconfig
index ebc135bda921..70653aed3005 100644
--- a/arch/xtensa/Kconfig
+++ b/arch/xtensa/Kconfig
@@ -5,7 +5,6 @@ config XTENSA
select ARCH_HAS_BINFMT_FLAT if !MMU
select ARCH_HAS_SYNC_DMA_FOR_CPU
select ARCH_HAS_SYNC_DMA_FOR_DEVICE
- select ARCH_NO_COHERENT_DMA_MMAP if !MMU
select ARCH_USE_QUEUED_RWLOCKS
select ARCH_USE_QUEUED_SPINLOCKS
select ARCH_WANT_FRAME_POINTERS
diff --git a/kernel/dma/mapping.c b/kernel/dma/mapping.c
index 64d1de59e133..fc17016b0871 100644
--- a/kernel/dma/mapping.c
+++ b/kernel/dma/mapping.c
@@ -186,7 +186,7 @@ int dma_common_mmap(struct device *dev, struct vm_area_struct *vma,
void *cpu_addr, dma_addr_t dma_addr, size_t size,
unsigned long attrs)
{
-#ifndef CONFIG_ARCH_NO_COHERENT_DMA_MMAP
+#ifdef CONFIG_MMU
unsigned long user_count = vma_pages(vma);
unsigned long count = PAGE_ALIGN(size) >> PAGE_SHIFT;
unsigned long off = vma->vm_pgoff;
@@ -217,7 +217,7 @@ int dma_common_mmap(struct device *dev, struct vm_area_struct *vma,
user_count << PAGE_SHIFT, vma->vm_page_prot);
#else
return -ENXIO;
-#endif /* !CONFIG_ARCH_NO_COHERENT_DMA_MMAP */
+#endif /* CONFIG_MMU */
}
/**
@@ -231,12 +231,10 @@ bool dma_can_mmap(struct device *dev)
{
const struct dma_map_ops *ops = get_dma_ops(dev);
- if (IS_ENABLED(CONFIG_ARCH_NO_COHERENT_DMA_MMAP))
- return false;
-
if (dma_is_direct(ops)) {
- return dev_is_dma_coherent(dev) ||
- IS_ENABLED(CONFIG_ARCH_HAS_DMA_COHERENT_TO_PFN);
+ return IS_ENABLED(CONFIG_MMU) &&
+ (dev_is_dma_coherent(dev) ||
+ IS_ENABLED(CONFIG_ARCH_HAS_DMA_COHERENT_TO_PFN));
}
return ops->mmap != NULL;
--
2.20.1
^ permalink raw reply related
* Re: [PATCH] powerpc/kmcent2: update the ethernet devices' phy properties
From: Valentin Longchamp @ 2019-08-08 21:09 UTC (permalink / raw)
To: Madalin-cristian Bucur
Cc: Scott Wood, netdev@vger.kernel.org, linuxppc-dev@lists.ozlabs.org
In-Reply-To: <VI1PR04MB55679AAE8DDC3160B9CCE073ECDC0@VI1PR04MB5567.eurprd04.prod.outlook.com>
Le mar. 30 juil. 2019 à 11:44, Madalin-cristian Bucur
<madalin.bucur@nxp.com> a écrit :
>
> > -----Original Message-----
> >
> > > Le dim. 14 juil. 2019 à 22:05, Valentin Longchamp
> > > <valentin@longchamp.me> a écrit :
> > > >
> > > > Change all phy-connection-type properties to phy-mode that are better
> > > > supported by the fman driver.
> > > >
> > > > Use the more readable fixed-link node for the 2 sgmii links.
> > > >
> > > > Change the RGMII link to rgmii-id as the clock delays are added by the
> > > > phy.
> > > >
> > > > Signed-off-by: Valentin Longchamp <valentin@longchamp.me>
> >
> > I don't see any other uses of phy-mode in arch/powerpc/boot/dts/fsl, and I see
> > lots of phy-connection-type with fman. Madalin, does this patch look OK?
> >
> > -Scott
>
> Hi,
>
> we are using "phy-connection-type" not "phy-mode" for the NXP (former Freescale)
> DPAA platforms. While the two seem to be interchangeable ("phy-mode" seems to be
> more recent, looking at the device tree bindings), the driver code in Linux seems
> to use one or the other, not both so one should stick with the variant the driver
> is using. To make things more complex, there may be dependencies in bootloaders,
> I see code in u-boot using only "phy-connection-type" or only "phy-mode".
>
> I'd leave "phy-connection-type" as is.
So I have finally had time to have a look and now I understand what
happens. You are right, there are bootloader dependencies: u-boot
calls fdt_fixup_phy_connection() that somehow in our case adds (or
changes if already in the device tree) the phy-connection-type
property to a wrong value ! By having a phy-mode in the device tree,
that is not changed by u-boot and by chance picked up by the kernel
fman driver (of_get_phy_mode() ) over phy-connection-mode, the below
patch fixes it for us.
I agree with you, it's not correct to have both phy-connection-type
and phy-mode. Ideally, u-boot on the board should be reworked so that
it does not perform the above wrong fixup. However, in an "unfixed"
.dtb (I have disabled fdt_fixup_phy_connection), the device tree in
the end only has either phy-connection-type or phy-mode, according to
what was chosen in the .dts file. And the fman driver works well with
both (thanks to the call to of_get_phy_mode() ). I would therefore
argue that even if all other DPAA platforms use phy-connection-type,
phy-mode is valid as well. (Furthermore we already have hundreds of
such boards in the field and we don't really support "remote" u-boot
update, so the u-boot fix is going to be difficult for us to pull).
Valentin
>
> Madalin
>
> > > > ---
> > > > arch/powerpc/boot/dts/fsl/kmcent2.dts | 16 +++++++++++-----
> > > > 1 file changed, 11 insertions(+), 5 deletions(-)
> > > >
> > > > diff --git a/arch/powerpc/boot/dts/fsl/kmcent2.dts
> > > > b/arch/powerpc/boot/dts/fsl/kmcent2.dts
> > > > index 48b7f9797124..c3e0741cafb1 100644
> > > > --- a/arch/powerpc/boot/dts/fsl/kmcent2.dts
> > > > +++ b/arch/powerpc/boot/dts/fsl/kmcent2.dts
> > > > @@ -210,13 +210,19 @@
> > > >
> > > > fman@400000 {
> > > > ethernet@e0000 {
> > > > - fixed-link = <0 1 1000 0 0>;
> > > > - phy-connection-type = "sgmii";
> > > > + phy-mode = "sgmii";
> > > > + fixed-link {
> > > > + speed = <1000>;
> > > > + full-duplex;
> > > > + };
> > > > };
> > > >
> > > > ethernet@e2000 {
> > > > - fixed-link = <1 1 1000 0 0>;
> > > > - phy-connection-type = "sgmii";
> > > > + phy-mode = "sgmii";
> > > > + fixed-link {
> > > > + speed = <1000>;
> > > > + full-duplex;
> > > > + };
> > > > };
> > > >
> > > > ethernet@e4000 {
> > > > @@ -229,7 +235,7 @@
> > > >
> > > > ethernet@e8000 {
> > > > phy-handle = <&front_phy>;
> > > > - phy-connection-type = "rgmii";
> > > > + phy-mode = "rgmii-id";
> > > > };
> > > >
> > > > mdio0: mdio@fc000 {
> > > > --
> > > > 2.17.1
> > > >
> > >
> > >
>
^ permalink raw reply
* Re: [PATCH] powerpc/64e: drop stale call to smp_processor_id() which hangs SMP startup
From: Chris Packham @ 2019-08-08 21:12 UTC (permalink / raw)
To: christophe.leroy@c-s.fr, paulus@samba.org, mpe@ellerman.id.au,
benh@kernel.crashing.org
Cc: linuxppc-dev@lists.ozlabs.org, linux-kernel@vger.kernel.org
In-Reply-To: <bef479514f4c08329fa649f67735df8918bc0976.1565268248.git.christophe.leroy@c-s.fr>
Hi Christophe,
On Thu, 2019-08-08 at 12:48 +0000, Christophe Leroy wrote:
> Santa commit ebb9d30a6a74 ("powerpc/mm: any thread in one core can be
> the first to setup TLB1") removed the need to know the cpu_id in
> early_init_this_mmu(), but the call to smp_processor_id() which was
> marked __maybe_used remained.
>
> Since commit ed1cd6deb013 ("powerpc: Activate
> CONFIG_THREAD_INFO_IN_TASK") thread_info cannot be reached before mmu
> is properly set up.
>
> Drop this stale call to smp_processor_id() which make SMP hang
> when CONFIG_PREEMPT is set.
>
> Reported-by: Chris Packham <Chris.Packham@alliedtelesis.co.nz>
> Fixes: ebb9d30a6a74 ("powerpc/mm: any thread in one core can be the
> first to setup TLB1")
> Link: https://github.com/linuxppc/issues/issues/264
> Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
> Cc: stable@vger.kernel.org
Many thanks for your help.
Tested-by: Chris Packham <chris.packham@alliedtelesis.co.nz>
> ---
> arch/powerpc/mm/nohash/tlb.c | 1 -
> 1 file changed, 1 deletion(-)
>
> diff --git a/arch/powerpc/mm/nohash/tlb.c
> b/arch/powerpc/mm/nohash/tlb.c
> index d4acf6fa0596..bf60983a58c7 100644
> --- a/arch/powerpc/mm/nohash/tlb.c
> +++ b/arch/powerpc/mm/nohash/tlb.c
> @@ -630,7 +630,6 @@ static void early_init_this_mmu(void)
> #ifdef CONFIG_PPC_FSL_BOOK3E
> if (mmu_has_feature(MMU_FTR_TYPE_FSL_E)) {
> unsigned int num_cams;
> - int __maybe_unused cpu = smp_processor_id();
> bool map = true;
>
> /* use a quarter of the TLBCAM for bolted linear map
> */
^ permalink raw reply
* [PATCH] soc: fsl: dpio: Add support for QBMan ring bulk enqueue.
From: Youri Querry @ 2019-08-08 20:59 UTC (permalink / raw)
To: Roy Pledge, Leo Li, linux-kernel@vger.kernel.org,
linuxppc-dev@lists.ozlabs.org,
linux-arm-kernel@lists.infradead.org, Ioana Ciocoi Radulescu,
Ioana Ciornei
Cc: Youri Querry
The QBMan frame descriptor enqueuing is changed from array
mode (a single frame enqueue at a time) to bulk ring mode.
This new mode allows the enqueuing of multiple frames in one operation.
The original interface is kept but use the bulk enqueue of one frame
Signed-off-by: Youri Querry <youri.querry_1@nxp.com>
---
drivers/soc/fsl/dpio/dpio-service.c | 69 +++-
drivers/soc/fsl/dpio/qbman-portal.c | 772 ++++++++++++++++++++++++++++++++----
drivers/soc/fsl/dpio/qbman-portal.h | 175 +++++++-
3 files changed, 935 insertions(+), 81 deletions(-)
diff --git a/drivers/soc/fsl/dpio/dpio-service.c b/drivers/soc/fsl/dpio/dpio-service.c
index b9539ef..4eb53ee 100644
--- a/drivers/soc/fsl/dpio/dpio-service.c
+++ b/drivers/soc/fsl/dpio/dpio-service.c
@@ -1,7 +1,7 @@
// SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause)
/*
* Copyright 2014-2016 Freescale Semiconductor Inc.
- * Copyright 2016 NXP
+ * Copyright 2016-2019 NXP
*
*/
#include <linux/types.h>
@@ -435,6 +435,69 @@ int dpaa2_io_service_enqueue_fq(struct dpaa2_io *d,
EXPORT_SYMBOL(dpaa2_io_service_enqueue_fq);
/**
+ * dpaa2_io_service_enqueue_multiple_fq() - Enqueue multiple frames
+ * to a frame queue using one fqid.
+ * @d: the given DPIO service.
+ * @fqid: the given frame queue id.
+ * @fd: the list of frame descriptors enqueued.
+ * @nb: number of frames to be enqueued
+ *
+ * Return the number of enqueued frames (0 if EQCR is busy)
+ * or -ENODEV if there is no dpio service.
+ */
+int dpaa2_io_service_enqueue_multiple_fq(struct dpaa2_io *d,
+ u32 fqid,
+ const struct dpaa2_fd *fd,
+ int nb)
+{
+ struct qbman_eq_desc ed;
+
+ d = service_select(d);
+ if (!d)
+ return -ENODEV;
+
+ qbman_eq_desc_clear(&ed);
+ qbman_eq_desc_set_no_orp(&ed, 0);
+ qbman_eq_desc_set_fq(&ed, fqid);
+
+ return qbman_swp_enqueue_multiple(d->swp, &ed, fd, 0, nb);
+}
+EXPORT_SYMBOL(dpaa2_io_service_enqueue_multiple_fq);
+
+/**
+ * dpaa2_io_service_enqueue_multiple_desc_fq() - Enqueue multiple frames
+ * to different frame queue using a list of fqids.
+ * @d: the given DPIO service.
+ * @fqid: the given list of frame queue ids.
+ * @fd: the list of frame descriptors enqueued.
+ * @nb: number of frames to be enqueued
+ *
+ * Return the number of enqueued frames (0 if EQCR is busy)
+ * or -ENODEV if there is no dpio service.
+ */
+int dpaa2_io_service_enqueue_multiple_desc_fq(struct dpaa2_io *d,
+ u32 *fqid,
+ const struct dpaa2_fd *fd,
+ int nb)
+{
+ int i;
+ struct qbman_eq_desc_min ed[32];
+
+ d = service_select(d);
+ if (!d)
+ return -ENODEV;
+
+ for (i = 0; i < nb; i++) {
+ qbman_eq_desc_min_clear(&ed[i]);
+ qbman_eq_desc_set_no_orp_min(&ed[i], 0);
+ qbman_eq_desc_set_min_fq(&ed[i], fqid[i]);
+ }
+
+ return qbman_swp_enqueue_multiple_desc(d->swp, &ed[0], fd, nb);
+}
+EXPORT_SYMBOL(dpaa2_io_service_enqueue_multiple_desc_fq);
+
+/**
* dpaa2_io_service_enqueue_qd() - Enqueue a frame to a QD.
* @d: the given DPIO service.
* @qdid: the given queuing destination id.
@@ -528,7 +591,7 @@ EXPORT_SYMBOL_GPL(dpaa2_io_service_acquire);
/**
* dpaa2_io_store_create() - Create the dma memory storage for dequeue result.
- * @max_frames: the maximum number of dequeued result for frames, must be <= 16.
+ * @max_frames: the maximum number of dequeued result for frames, must be <= 32.
* @dev: the device to allow mapping/unmapping the DMAable region.
*
* The size of the storage is "max_frames*sizeof(struct dpaa2_dq)".
@@ -543,7 +606,7 @@ struct dpaa2_io_store *dpaa2_io_store_create(unsigned int max_frames,
struct dpaa2_io_store *ret;
size_t size;
- if (!max_frames || (max_frames > 16))
+ if (!max_frames || (max_frames > 32))
return NULL;
ret = kmalloc(sizeof(*ret), GFP_KERNEL);
diff --git a/drivers/soc/fsl/dpio/qbman-portal.c b/drivers/soc/fsl/dpio/qbman-portal.c
index c66f5b7..c892a86 100644
--- a/drivers/soc/fsl/dpio/qbman-portal.c
+++ b/drivers/soc/fsl/dpio/qbman-portal.c
@@ -1,13 +1,14 @@
// SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause)
/*
* Copyright (C) 2014-2016 Freescale Semiconductor, Inc.
- * Copyright 2016 NXP
+ * Copyright 2016-2019 NXP
*
*/
#include <asm/cacheflush.h>
#include <linux/io.h>
#include <linux/slab.h>
+#include <linux/spinlock.h>
#include <soc/fsl/dpaa2-global.h>
#include "qbman-portal.h"
@@ -28,6 +29,7 @@
/* CINH register offsets */
#define QBMAN_CINH_SWP_EQCR_PI 0x800
+#define QBMAN_CINH_SWP_EQCR_CI 0x840
#define QBMAN_CINH_SWP_EQAR 0x8c0
#define QBMAN_CINH_SWP_CR_RT 0x900
#define QBMAN_CINH_SWP_VDQCR_RT 0x940
@@ -51,6 +53,8 @@
#define QBMAN_CENA_SWP_CR 0x600
#define QBMAN_CENA_SWP_RR(vb) (0x700 + ((u32)(vb) >> 1))
#define QBMAN_CENA_SWP_VDQCR 0x780
+#define QBMAN_CENA_SWP_EQCR_CI 0x840
+#define QBMAN_CENA_SWP_EQCR_CI_MEMBACK 0x1840
/* CENA register offsets in memory-backed mode */
#define QBMAN_CENA_SWP_DQRR_MEM(n) (0x800 + ((u32)(n) << 6))
@@ -78,6 +82,12 @@
/* opaque token for static dequeues */
#define QMAN_SDQCR_TOKEN 0xbb
+#define QBMAN_EQCR_DCA_IDXMASK 0x0f
+#define QBMAN_ENQUEUE_FLAG_DCA (1ULL << 31)
+
+#define EQ_DESC_SIZE_WITHOUT_FD 29
+#define EQ_DESC_SIZE_FD_START 32
+
enum qbman_sdqcr_dct {
qbman_sdqcr_dct_null = 0,
qbman_sdqcr_dct_prio_ics,
@@ -90,6 +100,82 @@ enum qbman_sdqcr_fc {
qbman_sdqcr_fc_up_to_3 = 1
};
+/* Internal Function declaration */
+static int qbman_swp_enqueue_ring_mode_direct(struct qbman_swp *s,
+ const struct qbman_eq_desc *d,
+ const struct dpaa2_fd *fd);
+static int qbman_swp_enqueue_ring_mode_mem_back(struct qbman_swp *s,
+ const struct qbman_eq_desc *d,
+ const struct dpaa2_fd *fd);
+static int qbman_swp_enqueue_multiple_direct(struct qbman_swp *s,
+ const struct qbman_eq_desc *d,
+ const struct dpaa2_fd *fd,
+ uint32_t *flags,
+ int num_frames);
+static int qbman_swp_enqueue_multiple_mem_back(struct qbman_swp *s,
+ const struct qbman_eq_desc *d,
+ const struct dpaa2_fd *fd,
+ uint32_t *flags,
+ int num_frames);
+static int
+qbman_swp_enqueue_multiple_desc_direct(struct qbman_swp *s,
+ const struct qbman_eq_desc_min *d,
+ const struct dpaa2_fd *fd,
+ int num_frames);
+static
+int qbman_swp_enqueue_multiple_desc_mem_back(struct qbman_swp *s,
+ const struct qbman_eq_desc_min *d,
+ const struct dpaa2_fd *fd,
+ int num_frames);
+static int qbman_swp_pull_direct(struct qbman_swp *s,
+ struct qbman_pull_desc *d);
+static int qbman_swp_pull_mem_back(struct qbman_swp *s,
+ struct qbman_pull_desc *d);
+
+const struct dpaa2_dq *qbman_swp_dqrr_next_direct(struct qbman_swp *s);
+const struct dpaa2_dq *qbman_swp_dqrr_next_mem_back(struct qbman_swp *s);
+
+static int qbman_swp_release_direct(struct qbman_swp *s,
+ const struct qbman_release_desc *d,
+ const u64 *buffers,
+ unsigned int num_buffers);
+static int qbman_swp_release_mem_back(struct qbman_swp *s,
+ const struct qbman_release_desc *d,
+ const u64 *buffers,
+ unsigned int num_buffers);
+
+/* Function pointers */
+int (*qbman_swp_enqueue_ring_mode_ptr)(struct qbman_swp *s,
+ const struct qbman_eq_desc *d,
+ const struct dpaa2_fd *fd)
+ = qbman_swp_enqueue_ring_mode_direct;
+
+int (*qbman_swp_enqueue_multiple_ptr)(struct qbman_swp *s,
+ const struct qbman_eq_desc *d,
+ const struct dpaa2_fd *fd,
+ uint32_t *flags,
+ int num_frames)
+ = qbman_swp_enqueue_multiple_direct;
+
+int
+(*qbman_swp_enqueue_multiple_desc_ptr)(struct qbman_swp *s,
+ const struct qbman_eq_desc_min *d,
+ const struct dpaa2_fd *fd,
+ int num_frames)
+ = qbman_swp_enqueue_multiple_desc_direct;
+
+int (*qbman_swp_pull_ptr)(struct qbman_swp *s, struct qbman_pull_desc *d)
+ = qbman_swp_pull_direct;
+
+const struct dpaa2_dq *(*qbman_swp_dqrr_next_ptr)(struct qbman_swp *s)
+ = qbman_swp_dqrr_next_direct;
+
+int (*qbman_swp_release_ptr)(struct qbman_swp *s,
+ const struct qbman_release_desc *d,
+ const u64 *buffers,
+ unsigned int num_buffers)
+ = qbman_swp_release_direct;
+
/* Portal Access */
static inline u32 qbman_read_register(struct qbman_swp *p, u32 offset)
@@ -146,6 +232,15 @@ static inline u32 qbman_set_swp_cfg(u8 max_fill, u8 wn, u8 est, u8 rpm, u8 dcm,
#define QMAN_RT_MODE 0x00000100
+static inline u8 qm_cyc_diff(u8 ringsize, u8 first, u8 last)
+{
+ /* 'first' is included, 'last' is excluded */
+ if (first <= last)
+ return last - first;
+ else
+ return (2 * ringsize) - (first - last);
+}
+
/**
* qbman_swp_init() - Create a functional object representing the given
* QBMan portal descriptor.
@@ -156,8 +251,12 @@ static inline u32 qbman_set_swp_cfg(u8 max_fill, u8 wn, u8 est, u8 rpm, u8 dcm,
*/
struct qbman_swp *qbman_swp_init(const struct qbman_swp_desc *d)
{
- struct qbman_swp *p = kmalloc(sizeof(*p), GFP_KERNEL);
+ struct qbman_swp *p = kzalloc(sizeof(*p), GFP_KERNEL);
u32 reg;
+ u32 mask_size;
+ u32 eqcr_pi;
+
+ spin_lock_init(&p->access_spinlock);
if (!p)
return NULL;
@@ -189,22 +288,39 @@ struct qbman_swp *qbman_swp_init(const struct qbman_swp_desc *d)
if ((p->desc->qman_version & QMAN_REV_MASK) >= QMAN_REV_5000)
memset(p->addr_cena, 0, 64 * 1024);
- reg = qbman_set_swp_cfg(p->dqrr.dqrr_size,
- 1, /* Writes Non-cacheable */
- 0, /* EQCR_CI stashing threshold */
- 3, /* RPM: Valid bit mode, RCR in array mode */
- 2, /* DCM: Discrete consumption ack mode */
- 3, /* EPM: Valid bit mode, EQCR in array mode */
- 1, /* mem stashing drop enable == TRUE */
- 1, /* mem stashing priority == TRUE */
- 1, /* mem stashing enable == TRUE */
- 1, /* dequeue stashing priority == TRUE */
- 0, /* dequeue stashing enable == FALSE */
- 0); /* EQCR_CI stashing priority == FALSE */
- if ((p->desc->qman_version & QMAN_REV_MASK) >= QMAN_REV_5000)
+ if ((p->desc->qman_version & QMAN_REV_MASK) >= QMAN_REV_5000) {
+
+ reg = qbman_set_swp_cfg(p->dqrr.dqrr_size,
+ 0, /* Writes Non-cacheable */
+ 1, /* EQCR_CI stashing threshold */
+ 3, /* RPM: RCR in array mode */
+ 2, /* DCM: Discrete consumption ack */
+ 0, /* EPM: EQCR in ring mode */
+ 1, /* mem stashing drop enable */
+ 1, /* mem stashing priority enable */
+ 1, /* mem stashing enable */
+ 1, /* dequeue stashing priority enable */
+ 0, /* dequeue stashing enable */
+ 0); /* EQCR_CI stashing priority enable */
+ } else {
+ reg = qbman_set_swp_cfg(p->dqrr.dqrr_size,
+ 1, /* Writes Non-cacheable */
+ 0, /* EQCR_CI stashing threshold */
+ 3, /* RPM: RCR in array mode */
+ 2, /* DCM: Discrete consumption ack */
+ 2, /* EPM: EQCR in ring mode */
+ 1, /* mem stashing drop enable enable */
+ 1, /* mem stashing priority enable */
+ 1, /* mem stashing enable */
+ 1, /* dequeue stashing priority Tenable */
+ 0, /* dequeue stashing enable enable */
+ 0); /* EQCR_CI stashing priority enable */
+ }
+ if ((p->desc->qman_version & QMAN_REV_MASK) >= QMAN_REV_5000) {
reg |= 1 << SWP_CFG_CPBS_SHIFT | /* memory-backed mode */
1 << SWP_CFG_VPM_SHIFT | /* VDQCR read triggered mode */
1 << SWP_CFG_CPM_SHIFT; /* CR read triggered mode */
+ }
qbman_write_register(p, QBMAN_CINH_SWP_CFG, reg);
reg = qbman_read_register(p, QBMAN_CINH_SWP_CFG);
@@ -225,6 +341,30 @@ struct qbman_swp *qbman_swp_init(const struct qbman_swp_desc *d)
* applied when dequeues from a specific channel are enabled.
*/
qbman_write_register(p, QBMAN_CINH_SWP_SDQCR, 0);
+
+ p->eqcr.pi_ring_size = 8;
+ if ((p->desc->qman_version & QMAN_REV_MASK) >= QMAN_REV_5000) {
+ p->eqcr.pi_ring_size = 32;
+ qbman_swp_enqueue_ring_mode_ptr =
+ qbman_swp_enqueue_ring_mode_mem_back;
+ qbman_swp_enqueue_multiple_ptr =
+ qbman_swp_enqueue_multiple_mem_back;
+ qbman_swp_enqueue_multiple_desc_ptr =
+ qbman_swp_enqueue_multiple_desc_mem_back;
+ qbman_swp_pull_ptr = qbman_swp_pull_mem_back;
+ qbman_swp_dqrr_next_ptr = qbman_swp_dqrr_next_mem_back;
+ qbman_swp_release_ptr = qbman_swp_release_mem_back;
+ }
+
+ for (mask_size = p->eqcr.pi_ring_size; mask_size > 0; mask_size >>= 1)
+ p->eqcr.pi_ci_mask = (p->eqcr.pi_ci_mask << 1) + 1;
+ eqcr_pi = qbman_read_register(p, QBMAN_CINH_SWP_EQCR_PI);
+ p->eqcr.pi = eqcr_pi & p->eqcr.pi_ci_mask;
+ p->eqcr.pi_vb = eqcr_pi & QB_VALID_BIT;
+ p->eqcr.ci = qbman_read_register(p, QBMAN_CINH_SWP_EQCR_CI)
+ & p->eqcr.pi_ci_mask;
+ p->eqcr.available = p->eqcr.pi_ring_size;
+
return p;
}
@@ -378,6 +518,7 @@ enum qb_enqueue_commands {
#define QB_ENQUEUE_CMD_ORP_ENABLE_SHIFT 2
#define QB_ENQUEUE_CMD_IRQ_ON_DISPATCH_SHIFT 3
#define QB_ENQUEUE_CMD_TARGET_TYPE_SHIFT 4
+#define QB_ENQUEUE_CMD_DCA_EN_SHIFT 7
/**
* qbman_eq_desc_clear() - Clear the contents of a descriptor to
@@ -389,6 +530,16 @@ void qbman_eq_desc_clear(struct qbman_eq_desc *d)
}
/**
+ * qbman_eq_desc_min_clear() - Clear the contents of a minimal
+ * enqueue descriptor to
+ * default/starting state.
+ */
+void qbman_eq_desc_min_clear(struct qbman_eq_desc_min *d)
+{
+ memset(d, 0, sizeof(*d));
+}
+
+/**
* qbman_eq_desc_set_no_orp() - Set enqueue descriptor without orp
* @d: the enqueue descriptor.
* @response_success: 1 = enqueue with response always; 0 = enqueue with
@@ -403,6 +554,22 @@ void qbman_eq_desc_set_no_orp(struct qbman_eq_desc *d, int respond_success)
d->verb |= enqueue_rejects_to_fq;
}
+/**
+ * qbman_eq_desc_set_no_orp_min() - Set minimal enqueue descriptor without orp
+ * @d: the enqueue descriptor.
+ * @response_success: 1 = enqueue with response always; 0 = enqueue with
+ * rejections returned on a FQ.
+ */
+void qbman_eq_desc_set_no_orp_min(struct qbman_eq_desc_min *d,
+ int respond_success)
+{
+ d->verb &= ~(1 << QB_ENQUEUE_CMD_ORP_ENABLE_SHIFT);
+ if (respond_success)
+ d->verb |= enqueue_response_always;
+ else
+ d->verb |= enqueue_rejects_to_fq;
+}
+
/*
* Exactly one of the following descriptor "targets" should be set. (Calling any
* one of these will replace the effect of any prior call to one of these.)
@@ -422,6 +589,17 @@ void qbman_eq_desc_set_fq(struct qbman_eq_desc *d, u32 fqid)
}
/**
+ * qbman_eq_desc_set_min_fq() - set the FQ for the minimal enqueue command
+ * @d: the enqueue descriptor
+ * @fqid: the id of the frame queue to be enqueued
+ */
+void qbman_eq_desc_set_min_fq(struct qbman_eq_desc_min *d, u32 fqid)
+{
+ d->verb &= ~(1 << QB_ENQUEUE_CMD_TARGET_TYPE_SHIFT);
+ d->tgtid = cpu_to_le32(fqid);
+}
+
+/**
* qbman_eq_desc_set_qd() - Set Queuing Destination for the enqueue command
* @d: the enqueue descriptor
* @qdid: the id of the queuing destination to be enqueued
@@ -453,41 +631,340 @@ static inline void qbman_write_eqcr_am_rt_register(struct qbman_swp *p,
QMAN_RT_MODE);
}
+#define QB_RT_BIT ((u32)0x100)
/**
- * qbman_swp_enqueue() - Issue an enqueue command
+ * qbman_swp_enqueue_ring_mode_direct() - Issue an enqueue command
* @s: the software portal used for enqueue
* @d: the enqueue descriptor
* @fd: the frame descriptor to be enqueued
*
- * Please note that 'fd' should only be NULL if the "action" of the
- * descriptor is "orp_hole" or "orp_nesn".
+ * Return 0 for successful enqueue, -EBUSY if the EQCR is not ready.
+ */
+static
+int qbman_swp_enqueue_ring_mode_direct(struct qbman_swp *s,
+ const struct qbman_eq_desc *d,
+ const struct dpaa2_fd *fd)
+{
+ int flags = 0;
+ int ret = qbman_swp_enqueue_multiple_direct(s, d, fd, &flags, 1);
+
+ if (ret >= 0)
+ ret = 0;
+ else
+ ret = -EBUSY;
+ return ret;
+}
+
+/**
+ * qbman_swp_enqueue_ring_mode_mem_back() - Issue an enqueue command
+ * @s: the software portal used for enqueue
+ * @d: the enqueue descriptor
+ * @fd: the frame descriptor to be enqueued
*
* Return 0 for successful enqueue, -EBUSY if the EQCR is not ready.
*/
-int qbman_swp_enqueue(struct qbman_swp *s, const struct qbman_eq_desc *d,
- const struct dpaa2_fd *fd)
+static
+int qbman_swp_enqueue_ring_mode_mem_back(struct qbman_swp *s,
+ const struct qbman_eq_desc *d,
+ const struct dpaa2_fd *fd)
{
- struct qbman_eq_desc *p;
- u32 eqar = qbman_read_register(s, QBMAN_CINH_SWP_EQAR);
+ int flags = 0;
+ int ret = qbman_swp_enqueue_multiple_mem_back(s, d, fd, &flags, 1);
- if (!EQAR_SUCCESS(eqar))
- return -EBUSY;
+ if (ret >= 0)
+ ret = 0;
+ else
+ ret = -EBUSY;
+ return ret;
+}
- p = qbman_get_cmd(s, QBMAN_CENA_SWP_EQCR(EQAR_IDX(eqar)));
- memcpy(&p->dca, &d->dca, 31);
- memcpy(&p->fd, fd, sizeof(*fd));
+/**
+ * qbman_swp_enqueue_multiple_direct() - Issue a multi enqueue command
+ * using one enqueue descriptor
+ * @s: the software portal used for enqueue
+ * @d: the enqueue descriptor
+ * @fd: table pointer of frame descriptor table to be enqueued
+ * @flags: table pointer of QBMAN_ENQUEUE_FLAG_DCA flags, not used if NULL
+ * @num_frames: number of fd to be enqueued
+ *
+ * Return the number of fd enqueued, or a negative error number.
+ */
+static
+int qbman_swp_enqueue_multiple_direct(struct qbman_swp *s,
+ const struct qbman_eq_desc *d,
+ const struct dpaa2_fd *fd,
+ uint32_t *flags,
+ int num_frames)
+{
+ uint32_t *p = NULL;
+ const uint32_t *cl = (uint32_t *)d;
+ uint32_t eqcr_ci, eqcr_pi, half_mask, full_mask;
+ int i, num_enqueued = 0;
+
+ half_mask = (s->eqcr.pi_ci_mask>>1);
+ full_mask = s->eqcr.pi_ci_mask;
+
+ if (!s->eqcr.available) {
+ eqcr_ci = s->eqcr.ci;
+ p = s->addr_cena + QBMAN_CENA_SWP_EQCR_CI;
+ s->eqcr.ci = __raw_readl(p) & full_mask;
+
+ s->eqcr.available = qm_cyc_diff(s->eqcr.pi_ring_size,
+ eqcr_ci, s->eqcr.ci);
+ if (!s->eqcr.available)
+ return 0;
+ }
- if ((s->desc->qman_version & QMAN_REV_MASK) < QMAN_REV_5000) {
- /* Set the verb byte, have to substitute in the valid-bit */
- dma_wmb();
- p->verb = d->verb | EQAR_VB(eqar);
- } else {
- p->verb = d->verb | EQAR_VB(eqar);
- dma_wmb();
- qbman_write_eqcr_am_rt_register(s, EQAR_IDX(eqar));
+ eqcr_pi = s->eqcr.pi;
+ num_enqueued = (s->eqcr.available < num_frames) ?
+ s->eqcr.available : num_frames;
+ s->eqcr.available -= num_enqueued;
+ /* Fill in the EQCR ring */
+ for (i = 0; i < num_enqueued; i++) {
+ p = (s->addr_cena + QBMAN_CENA_SWP_EQCR(eqcr_pi & half_mask));
+ /* Skip copying the verb */
+ memcpy(&p[1], &cl[1], EQ_DESC_SIZE_WITHOUT_FD - 1);
+ memcpy(&p[EQ_DESC_SIZE_FD_START/sizeof(uint32_t)],
+ &fd[i], sizeof(*fd));
+ eqcr_pi++;
}
- return 0;
+ dma_wmb();
+
+ /* Set the verb byte, have to substitute in the valid-bit */
+ eqcr_pi = s->eqcr.pi;
+ for (i = 0; i < num_enqueued; i++) {
+ p = (s->addr_cena + QBMAN_CENA_SWP_EQCR(eqcr_pi & half_mask));
+ p[0] = cl[0] | s->eqcr.pi_vb;
+ if (flags && (flags[i] & QBMAN_ENQUEUE_FLAG_DCA)) {
+ struct qbman_eq_desc *d = (struct qbman_eq_desc *)p;
+
+ d->dca = (1 << QB_ENQUEUE_CMD_DCA_EN_SHIFT) |
+ ((flags[i]) & QBMAN_EQCR_DCA_IDXMASK);
+ }
+ eqcr_pi++;
+ if (!(eqcr_pi & half_mask))
+ s->eqcr.pi_vb ^= QB_VALID_BIT;
+ }
+
+ s->eqcr.pi = (s->eqcr.pi + num_enqueued) & full_mask;
+
+ return num_enqueued;
+}
+
+/**
+ * qbman_swp_enqueue_multiple_mem_back() - Issue a multi enqueue command
+ * using one enqueue descriptor
+ * @s: the software portal used for enqueue
+ * @d: the enqueue descriptor
+ * @fd: table pointer of frame descriptor table to be enqueued
+ * @flags: table pointer of QBMAN_ENQUEUE_FLAG_DCA flags, not used if NULL
+ * @num_frames: number of fd to be enqueued
+ *
+ * Return the number of fd enqueued, or a negative error number.
+ */
+static
+int qbman_swp_enqueue_multiple_mem_back(struct qbman_swp *s,
+ const struct qbman_eq_desc *d,
+ const struct dpaa2_fd *fd,
+ uint32_t *flags,
+ int num_frames)
+{
+ uint32_t *p = NULL;
+ const uint32_t *cl = (uint32_t *)(d);
+ uint32_t eqcr_ci, eqcr_pi, half_mask, full_mask;
+ int i, num_enqueued = 0;
+ unsigned long irq_flags;
+
+ spin_lock(&s->access_spinlock);
+ local_irq_save(irq_flags);
+
+ half_mask = (s->eqcr.pi_ci_mask>>1);
+ full_mask = s->eqcr.pi_ci_mask;
+ if (!s->eqcr.available) {
+ eqcr_ci = s->eqcr.ci;
+ p = s->addr_cena + QBMAN_CENA_SWP_EQCR_CI_MEMBACK;
+ s->eqcr.ci = __raw_readl(p) & full_mask;
+ s->eqcr.available = qm_cyc_diff(s->eqcr.pi_ring_size,
+ eqcr_ci, s->eqcr.ci);
+ if (!s->eqcr.available) {
+ local_irq_restore(irq_flags);
+ spin_unlock(&s->access_spinlock);
+ return 0;
+ }
+ }
+
+ eqcr_pi = s->eqcr.pi;
+ num_enqueued = (s->eqcr.available < num_frames) ?
+ s->eqcr.available : num_frames;
+ s->eqcr.available -= num_enqueued;
+ /* Fill in the EQCR ring */
+ for (i = 0; i < num_enqueued; i++) {
+ p = (s->addr_cena + QBMAN_CENA_SWP_EQCR(eqcr_pi & half_mask));
+ /* Skip copying the verb */
+ memcpy(&p[1], &cl[1], EQ_DESC_SIZE_WITHOUT_FD - 1);
+ memcpy(&p[EQ_DESC_SIZE_FD_START/sizeof(uint32_t)],
+ &fd[i], sizeof(*fd));
+ eqcr_pi++;
+ }
+
+ /* Set the verb byte, have to substitute in the valid-bit */
+ eqcr_pi = s->eqcr.pi;
+ for (i = 0; i < num_enqueued; i++) {
+ p = (s->addr_cena + QBMAN_CENA_SWP_EQCR(eqcr_pi & half_mask));
+ p[0] = cl[0] | s->eqcr.pi_vb;
+ if (flags && (flags[i] & QBMAN_ENQUEUE_FLAG_DCA)) {
+ struct qbman_eq_desc *d = (struct qbman_eq_desc *)p;
+
+ d->dca = (1 << QB_ENQUEUE_CMD_DCA_EN_SHIFT) |
+ ((flags[i]) & QBMAN_EQCR_DCA_IDXMASK);
+ }
+ eqcr_pi++;
+ if (!(eqcr_pi & half_mask))
+ s->eqcr.pi_vb ^= QB_VALID_BIT;
+ }
+ s->eqcr.pi = eqcr_pi & full_mask;
+
+ dma_wmb();
+ qbman_write_register(s, QBMAN_CINH_SWP_EQCR_PI,
+ (QB_RT_BIT)|(s->eqcr.pi)|s->eqcr.pi_vb);
+ local_irq_restore(irq_flags);
+ spin_unlock(&s->access_spinlock);
+
+ return num_enqueued;
+}
+
+/**
+ * qbman_swp_enqueue_multiple_desc_direct() - Issue a multi enqueue command
+ * using multiple enqueue descriptor
+ * @s: the software portal used for enqueue
+ * @d: table of minimal enqueue descriptor
+ * @fd: table pointer of frame descriptor table to be enqueued
+ * @num_frames: number of fd to be enqueued
+ *
+ * Return the number of fd enqueued, or a negative error number.
+ */
+static
+int qbman_swp_enqueue_multiple_desc_direct(struct qbman_swp *s,
+ const struct qbman_eq_desc_min *d,
+ const struct dpaa2_fd *fd,
+ int num_frames)
+{
+ uint32_t *p;
+ const uint32_t *cl;
+ uint32_t eqcr_ci, eqcr_pi, half_mask, full_mask;
+ int i, num_enqueued = 0;
+
+ half_mask = (s->eqcr.pi_ci_mask>>1);
+ full_mask = s->eqcr.pi_ci_mask;
+ if (!s->eqcr.available) {
+ eqcr_ci = s->eqcr.ci;
+ p = s->addr_cena + QBMAN_CENA_SWP_EQCR_CI;
+ s->eqcr.ci = __raw_readl(p) & full_mask;
+ s->eqcr.available = qm_cyc_diff(s->eqcr.pi_ring_size,
+ eqcr_ci, s->eqcr.ci);
+ if (!s->eqcr.available)
+ return 0;
+ }
+
+ eqcr_pi = s->eqcr.pi;
+ num_enqueued = (s->eqcr.available < num_frames) ?
+ s->eqcr.available : num_frames;
+ s->eqcr.available -= num_enqueued;
+ /* Fill in the EQCR ring */
+ for (i = 0; i < num_enqueued; i++) {
+ p = (s->addr_cena + QBMAN_CENA_SWP_EQCR(eqcr_pi & half_mask));
+ cl = (uint32_t *)(&d[i]);
+ /* Skip copying the verb */
+ memcpy(&p[1], &cl[1], EQ_DESC_SIZE_WITHOUT_FD - 1);
+ memcpy(&p[EQ_DESC_SIZE_FD_START/sizeof(uint32_t)],
+ &fd[i], sizeof(*fd));
+ eqcr_pi++;
+ }
+
+ dma_wmb();
+
+ /* Set the verb byte, have to substitute in the valid-bit */
+ eqcr_pi = s->eqcr.pi;
+ for (i = 0; i < num_enqueued; i++) {
+ p = (s->addr_cena + QBMAN_CENA_SWP_EQCR(eqcr_pi & half_mask));
+ cl = (uint32_t *)(&d[i]);
+ p[0] = cl[0] | s->eqcr.pi_vb;
+ eqcr_pi++;
+ if (!(eqcr_pi & half_mask))
+ s->eqcr.pi_vb ^= QB_VALID_BIT;
+ }
+ s->eqcr.pi = (s->eqcr.pi + num_enqueued) & full_mask;
+
+ return num_enqueued;
+}
+
+/**
+ * qbman_swp_enqueue_multiple_desc_mem_back() - Issue a multi enqueue command
+ * using multiple enqueue descriptor
+ * @s: the software portal used for enqueue
+ * @d: table of minimal enqueue descriptor
+ * @fd: table pointer of frame descriptor table to be enqueued
+ * @num_frames: number of fd to be enqueued
+ *
+ * Return the number of fd enqueued, or a negative error number.
+ */
+static
+int qbman_swp_enqueue_multiple_desc_mem_back(struct qbman_swp *s,
+ const struct qbman_eq_desc_min *d,
+ const struct dpaa2_fd *fd,
+ int num_frames)
+{
+ uint32_t *p;
+ const uint32_t *cl;
+ uint32_t eqcr_ci, eqcr_pi, half_mask, full_mask;
+ int i, num_enqueued = 0;
+
+ half_mask = (s->eqcr.pi_ci_mask>>1);
+ full_mask = s->eqcr.pi_ci_mask;
+ if (!s->eqcr.available) {
+ eqcr_ci = s->eqcr.ci;
+ p = s->addr_cena + QBMAN_CENA_SWP_EQCR_CI_MEMBACK;
+ s->eqcr.ci = __raw_readl(p) & full_mask;
+ s->eqcr.available = qm_cyc_diff(s->eqcr.pi_ring_size,
+ eqcr_ci, s->eqcr.ci);
+ if (!s->eqcr.available)
+ return 0;
+ }
+
+ eqcr_pi = s->eqcr.pi;
+ num_enqueued = (s->eqcr.available < num_frames) ?
+ s->eqcr.available : num_frames;
+ s->eqcr.available -= num_enqueued;
+ /* Fill in the EQCR ring */
+ for (i = 0; i < num_enqueued; i++) {
+ p = (s->addr_cena + QBMAN_CENA_SWP_EQCR(eqcr_pi & half_mask));
+ cl = (uint32_t *)(&d[i]);
+ /* Skip copying the verb */
+ memcpy(&p[1], &cl[1], EQ_DESC_SIZE_WITHOUT_FD - 1);
+ memcpy(&p[EQ_DESC_SIZE_FD_START/sizeof(uint32_t)],
+ &fd[i], sizeof(*fd));
+ eqcr_pi++;
+ }
+
+ /* Set the verb byte, have to substitute in the valid-bit */
+ eqcr_pi = s->eqcr.pi;
+ for (i = 0; i < num_enqueued; i++) {
+ p = (s->addr_cena + QBMAN_CENA_SWP_EQCR(eqcr_pi & half_mask));
+ cl = (uint32_t *)(&d[i]);
+ p[0] = cl[0] | s->eqcr.pi_vb;
+ eqcr_pi++;
+ if (!(eqcr_pi & half_mask))
+ s->eqcr.pi_vb ^= QB_VALID_BIT;
+ }
+
+ s->eqcr.pi = eqcr_pi & full_mask;
+
+ dma_wmb();
+ qbman_write_register(s, QBMAN_CINH_SWP_EQCR_PI,
+ (QB_RT_BIT)|(s->eqcr.pi)|s->eqcr.pi_vb);
+
+ return num_enqueued;
}
/* Static (push) dequeue */
@@ -645,7 +1122,7 @@ void qbman_pull_desc_set_channel(struct qbman_pull_desc *d, u32 chid,
}
/**
- * qbman_swp_pull() - Issue the pull dequeue command
+ * qbman_swp_pull_direct() - Issue the pull dequeue command
* @s: the software portal object
* @d: the software portal descriptor which has been configured with
* the set of qbman_pull_desc_set_*() calls
@@ -653,7 +1130,7 @@ void qbman_pull_desc_set_channel(struct qbman_pull_desc *d, u32 chid,
* Return 0 for success, and -EBUSY if the software portal is not ready
* to do pull dequeue.
*/
-int qbman_swp_pull(struct qbman_swp *s, struct qbman_pull_desc *d)
+static int qbman_swp_pull_direct(struct qbman_swp *s, struct qbman_pull_desc *d)
{
struct qbman_pull_desc *p;
@@ -671,18 +1148,48 @@ int qbman_swp_pull(struct qbman_swp *s, struct qbman_pull_desc *d)
p->dq_src = d->dq_src;
p->rsp_addr = d->rsp_addr;
p->rsp_addr_virt = d->rsp_addr_virt;
+ dma_wmb();
+ /* Set the verb byte, have to substitute in the valid-bit */
+ p->verb = d->verb | s->vdq.valid_bit;
+ s->vdq.valid_bit ^= QB_VALID_BIT;
- if ((s->desc->qman_version & QMAN_REV_MASK) < QMAN_REV_5000) {
- dma_wmb();
- /* Set the verb byte, have to substitute in the valid-bit */
- p->verb = d->verb | s->vdq.valid_bit;
- s->vdq.valid_bit ^= QB_VALID_BIT;
- } else {
- p->verb = d->verb | s->vdq.valid_bit;
- s->vdq.valid_bit ^= QB_VALID_BIT;
- dma_wmb();
- qbman_write_register(s, QBMAN_CINH_SWP_VDQCR_RT, QMAN_RT_MODE);
+ return 0;
+}
+
+/**
+ * qbman_swp_pull_mem_back() - Issue the pull dequeue command
+ * @s: the software portal object
+ * @d: the software portal descriptor which has been configured with
+ * the set of qbman_pull_desc_set_*() calls
+ *
+ * Return 0 for success, and -EBUSY if the software portal is not ready
+ * to do pull dequeue.
+ */
+static int qbman_swp_pull_mem_back(struct qbman_swp *s,
+ struct qbman_pull_desc *d)
+{
+ struct qbman_pull_desc *p;
+
+ if (!atomic_dec_and_test(&s->vdq.available)) {
+ atomic_inc(&s->vdq.available);
+ return -EBUSY;
}
+ s->vdq.storage = (void *)(uintptr_t)d->rsp_addr_virt;
+ if ((s->desc->qman_version & QMAN_REV_MASK) < QMAN_REV_5000)
+ p = qbman_get_cmd(s, QBMAN_CENA_SWP_VDQCR);
+ else
+ p = qbman_get_cmd(s, QBMAN_CENA_SWP_VDQCR_MEM);
+ p->numf = d->numf;
+ p->tok = QMAN_DQ_TOKEN_VALID;
+ p->dq_src = d->dq_src;
+ p->rsp_addr = d->rsp_addr;
+ p->rsp_addr_virt = d->rsp_addr_virt;
+
+ /* Set the verb byte, have to substitute in the valid-bit */
+ p->verb = d->verb | s->vdq.valid_bit;
+ s->vdq.valid_bit ^= QB_VALID_BIT;
+ dma_wmb();
+ qbman_write_register(s, QBMAN_CINH_SWP_VDQCR_RT, QMAN_RT_MODE);
return 0;
}
@@ -690,14 +1197,14 @@ int qbman_swp_pull(struct qbman_swp *s, struct qbman_pull_desc *d)
#define QMAN_DQRR_PI_MASK 0xf
/**
- * qbman_swp_dqrr_next() - Get an valid DQRR entry
+ * qbman_swp_dqrr_next_direct() - Get an valid DQRR entry
* @s: the software portal object
*
* Return NULL if there are no unconsumed DQRR entries. Return a DQRR entry
* only once, so repeated calls can return a sequence of DQRR entries, without
* requiring they be consumed immediately or in any particular order.
*/
-const struct dpaa2_dq *qbman_swp_dqrr_next(struct qbman_swp *s)
+const struct dpaa2_dq *qbman_swp_dqrr_next_direct(struct qbman_swp *s)
{
u32 verb;
u32 response_verb;
@@ -740,10 +1247,7 @@ const struct dpaa2_dq *qbman_swp_dqrr_next(struct qbman_swp *s)
QBMAN_CENA_SWP_DQRR(s->dqrr.next_idx)));
}
- if ((s->desc->qman_version & QMAN_REV_MASK) < QMAN_REV_5000)
- p = qbman_get_cmd(s, QBMAN_CENA_SWP_DQRR(s->dqrr.next_idx));
- else
- p = qbman_get_cmd(s, QBMAN_CENA_SWP_DQRR_MEM(s->dqrr.next_idx));
+ p = qbman_get_cmd(s, QBMAN_CENA_SWP_DQRR(s->dqrr.next_idx));
verb = p->dq.verb;
/*
@@ -785,6 +1289,98 @@ const struct dpaa2_dq *qbman_swp_dqrr_next(struct qbman_swp *s)
}
/**
+ * qbman_swp_dqrr_next_mem_back() - Get an valid DQRR entry
+ * @s: the software portal object
+ *
+ * Return NULL if there are no unconsumed DQRR entries. Return a DQRR entry
+ * only once, so repeated calls can return a sequence of DQRR entries, without
+ * requiring they be consumed immediately or in any particular order.
+ */
+const struct dpaa2_dq *qbman_swp_dqrr_next_mem_back(struct qbman_swp *s)
+{
+ u32 verb;
+ u32 response_verb;
+ u32 flags;
+ struct dpaa2_dq *p;
+
+ /* Before using valid-bit to detect if something is there, we have to
+ * handle the case of the DQRR reset bug...
+ */
+ if (unlikely(s->dqrr.reset_bug)) {
+ /*
+ * We pick up new entries by cache-inhibited producer index,
+ * which means that a non-coherent mapping would require us to
+ * invalidate and read *only* once that PI has indicated that
+ * there's an entry here. The first trip around the DQRR ring
+ * will be much less efficient than all subsequent trips around
+ * it...
+ */
+ u8 pi = qbman_read_register(s, QBMAN_CINH_SWP_DQPI) &
+ QMAN_DQRR_PI_MASK;
+
+ /* there are new entries if pi != next_idx */
+ if (pi == s->dqrr.next_idx)
+ return NULL;
+
+ /*
+ * if next_idx is/was the last ring index, and 'pi' is
+ * different, we can disable the workaround as all the ring
+ * entries have now been DMA'd to so valid-bit checking is
+ * repaired. Note: this logic needs to be based on next_idx
+ * (which increments one at a time), rather than on pi (which
+ * can burst and wrap-around between our snapshots of it).
+ */
+ if (s->dqrr.next_idx == (s->dqrr.dqrr_size - 1)) {
+ pr_debug("next_idx=%d, pi=%d, clear reset bug\n",
+ s->dqrr.next_idx, pi);
+ s->dqrr.reset_bug = 0;
+ }
+ prefetch(qbman_get_cmd(s,
+ QBMAN_CENA_SWP_DQRR(s->dqrr.next_idx)));
+ }
+
+ p = qbman_get_cmd(s, QBMAN_CENA_SWP_DQRR_MEM(s->dqrr.next_idx));
+ verb = p->dq.verb;
+
+ /*
+ * If the valid-bit isn't of the expected polarity, nothing there. Note,
+ * in the DQRR reset bug workaround, we shouldn't need to skip these
+ * check, because we've already determined that a new entry is available
+ * and we've invalidated the cacheline before reading it, so the
+ * valid-bit behaviour is repaired and should tell us what we already
+ * knew from reading PI.
+ */
+ if ((verb & QB_VALID_BIT) != s->dqrr.valid_bit) {
+ prefetch(qbman_get_cmd(s,
+ QBMAN_CENA_SWP_DQRR(s->dqrr.next_idx)));
+ return NULL;
+ }
+ /*
+ * There's something there. Move "next_idx" attention to the next ring
+ * entry (and prefetch it) before returning what we found.
+ */
+ s->dqrr.next_idx++;
+ s->dqrr.next_idx &= s->dqrr.dqrr_size - 1; /* Wrap around */
+ if (!s->dqrr.next_idx)
+ s->dqrr.valid_bit ^= QB_VALID_BIT;
+
+ /*
+ * If this is the final response to a volatile dequeue command
+ * indicate that the vdq is available
+ */
+ flags = p->dq.stat;
+ response_verb = verb & QBMAN_RESULT_MASK;
+ if (response_verb == QBMAN_RESULT_DQ &&
+ (flags & DPAA2_DQ_STAT_VOLATILE) &&
+ (flags & DPAA2_DQ_STAT_EXPIRED))
+ atomic_inc(&s->vdq.available);
+
+ prefetch(qbman_get_cmd(s, QBMAN_CENA_SWP_DQRR(s->dqrr.next_idx)));
+
+ return p;
+}
+
+/**
* qbman_swp_dqrr_consume() - Consume DQRR entries previously returned from
* qbman_swp_dqrr_next().
* @s: the software portal object
@@ -872,7 +1468,7 @@ void qbman_release_desc_set_rcdi(struct qbman_release_desc *d, int enable)
#define RAR_SUCCESS(rar) ((rar) & 0x100)
/**
- * qbman_swp_release() - Issue a buffer release command
+ * qbman_swp_release_direct() - Issue a buffer release command
* @s: the software portal object
* @d: the release descriptor
* @buffers: a pointer pointing to the buffer address to be released
@@ -880,8 +1476,53 @@ void qbman_release_desc_set_rcdi(struct qbman_release_desc *d, int enable)
*
* Return 0 for success, -EBUSY if the release command ring is not ready.
*/
-int qbman_swp_release(struct qbman_swp *s, const struct qbman_release_desc *d,
- const u64 *buffers, unsigned int num_buffers)
+int qbman_swp_release_direct(struct qbman_swp *s,
+ const struct qbman_release_desc *d,
+ const u64 *buffers, unsigned int num_buffers)
+{
+ int i;
+ struct qbman_release_desc *p;
+ u32 rar;
+
+ if (!num_buffers || num_buffers > 7)
+ return -EINVAL;
+
+ rar = qbman_read_register(s, QBMAN_CINH_SWP_RAR);
+ if (!RAR_SUCCESS(rar))
+ return -EBUSY;
+
+ /* Start the release command */
+ if ((s->desc->qman_version & QMAN_REV_MASK) < QMAN_REV_5000)
+ p = qbman_get_cmd(s, QBMAN_CENA_SWP_RCR(RAR_IDX(rar)));
+ else
+ p = qbman_get_cmd(s, QBMAN_CENA_SWP_RCR_MEM(RAR_IDX(rar)));
+ /* Copy the caller's buffer pointers to the command */
+ for (i = 0; i < num_buffers; i++)
+ p->buf[i] = cpu_to_le64(buffers[i]);
+ p->bpid = d->bpid;
+
+ /*
+ * Set the verb byte, have to substitute in the valid-bit
+ * and the number of buffers.
+ */
+ dma_wmb();
+ p->verb = d->verb | RAR_VB(rar) | num_buffers;
+
+ return 0;
+}
+
+/**
+ * qbman_swp_release_mem_back() - Issue a buffer release command
+ * @s: the software portal object
+ * @d: the release descriptor
+ * @buffers: a pointer pointing to the buffer address to be released
+ * @num_buffers: number of buffers to be released, must be less than 8
+ *
+ * Return 0 for success, -EBUSY if the release command ring is not ready.
+ */
+int qbman_swp_release_mem_back(struct qbman_swp *s,
+ const struct qbman_release_desc *d,
+ const u64 *buffers, unsigned int num_buffers)
{
int i;
struct qbman_release_desc *p;
@@ -904,19 +1545,10 @@ int qbman_swp_release(struct qbman_swp *s, const struct qbman_release_desc *d,
p->buf[i] = cpu_to_le64(buffers[i]);
p->bpid = d->bpid;
- if ((s->desc->qman_version & QMAN_REV_MASK) < QMAN_REV_5000) {
- /*
- * Set the verb byte, have to substitute in the valid-bit
- * and the number of buffers.
- */
- dma_wmb();
- p->verb = d->verb | RAR_VB(rar) | num_buffers;
- } else {
- p->verb = d->verb | RAR_VB(rar) | num_buffers;
- dma_wmb();
- qbman_write_register(s, QBMAN_CINH_SWP_RCR_AM_RT +
- RAR_IDX(rar) * 4, QMAN_RT_MODE);
- }
+ p->verb = d->verb | RAR_VB(rar) | num_buffers;
+ dma_wmb();
+ qbman_write_register(s, QBMAN_CINH_SWP_RCR_AM_RT +
+ RAR_IDX(rar) * 4, QMAN_RT_MODE);
return 0;
}
diff --git a/drivers/soc/fsl/dpio/qbman-portal.h b/drivers/soc/fsl/dpio/qbman-portal.h
index f3ec5d2..f4b6e47 100644
--- a/drivers/soc/fsl/dpio/qbman-portal.h
+++ b/drivers/soc/fsl/dpio/qbman-portal.h
@@ -9,6 +9,13 @@
#include <soc/fsl/dpaa2-fd.h>
+#define QMAN_REV_4000 0x04000000
+#define QMAN_REV_4100 0x04010000
+#define QMAN_REV_4101 0x04010001
+#define QMAN_REV_5000 0x05000000
+
+#define QMAN_REV_MASK 0xffff0000
+
struct dpaa2_dq;
struct qbman_swp;
@@ -67,6 +74,22 @@ enum qbman_pull_type_e {
#define QBMAN_FQ_XOFF 0x4e
/* structure of enqueue descriptor */
+struct qbman_eq_desc_min {
+ u8 verb;
+ u8 dca;
+ __le16 seqnum;
+ __le16 orpid;
+ __le16 reserved1;
+ __le32 tgtid;
+ __le32 tag;
+ __le16 qdbin;
+ u8 qpri;
+ u8 reserved[3];
+ u8 wae;
+ u8 rspid;
+ __le64 rsp_addr;
+};
+
struct qbman_eq_desc {
u8 verb;
u8 dca;
@@ -132,8 +155,48 @@ struct qbman_swp {
u8 dqrr_size;
int reset_bug; /* indicates dqrr reset workaround is needed */
} dqrr;
+
+ struct {
+ u32 pi;
+ u32 pi_vb;
+ u32 pi_ring_size;
+ u32 pi_ci_mask;
+ u32 ci;
+ int available;
+ u32 pend;
+ u32 no_pfdr;
+ } eqcr;
+
+ spinlock_t access_spinlock;
};
+/* Function pointers */
+extern
+int (*qbman_swp_enqueue_ring_mode_ptr)(struct qbman_swp *s,
+ const struct qbman_eq_desc *d,
+ const struct dpaa2_fd *fd);
+extern
+int (*qbman_swp_enqueue_multiple_ptr)(struct qbman_swp *s,
+ const struct qbman_eq_desc *d,
+ const struct dpaa2_fd *fd,
+ uint32_t *flags,
+ int num_frames);
+extern
+int (*qbman_swp_enqueue_multiple_desc_ptr)(struct qbman_swp *s,
+ const struct qbman_eq_desc_min *d,
+ const struct dpaa2_fd *fd,
+ int num_frames);
+extern
+int (*qbman_swp_pull_ptr)(struct qbman_swp *s, struct qbman_pull_desc *d);
+extern
+const struct dpaa2_dq *(*qbman_swp_dqrr_next_ptr)(struct qbman_swp *s);
+extern
+int (*qbman_swp_release_ptr)(struct qbman_swp *s,
+ const struct qbman_release_desc *d,
+ const u64 *buffers,
+ unsigned int num_buffers);
+
+/* Functions */
struct qbman_swp *qbman_swp_init(const struct qbman_swp_desc *d);
void qbman_swp_finish(struct qbman_swp *p);
u32 qbman_swp_interrupt_read_status(struct qbman_swp *p);
@@ -158,29 +221,25 @@ void qbman_pull_desc_set_wq(struct qbman_pull_desc *d, u32 wqid,
void qbman_pull_desc_set_channel(struct qbman_pull_desc *d, u32 chid,
enum qbman_pull_type_e dct);
-int qbman_swp_pull(struct qbman_swp *p, struct qbman_pull_desc *d);
-
-const struct dpaa2_dq *qbman_swp_dqrr_next(struct qbman_swp *s);
void qbman_swp_dqrr_consume(struct qbman_swp *s, const struct dpaa2_dq *dq);
int qbman_result_has_new_result(struct qbman_swp *p, const struct dpaa2_dq *dq);
void qbman_eq_desc_clear(struct qbman_eq_desc *d);
+void qbman_eq_desc_min_clear(struct qbman_eq_desc_min *d);
void qbman_eq_desc_set_no_orp(struct qbman_eq_desc *d, int respond_success);
+void qbman_eq_desc_set_no_orp_min(struct qbman_eq_desc_min *d,
+ int respond_success);
void qbman_eq_desc_set_token(struct qbman_eq_desc *d, u8 token);
void qbman_eq_desc_set_fq(struct qbman_eq_desc *d, u32 fqid);
+void qbman_eq_desc_set_min_fq(struct qbman_eq_desc_min *d, u32 fqid);
void qbman_eq_desc_set_qd(struct qbman_eq_desc *d, u32 qdid,
u32 qd_bin, u32 qd_prio);
-int qbman_swp_enqueue(struct qbman_swp *p, const struct qbman_eq_desc *d,
- const struct dpaa2_fd *fd);
-
void qbman_release_desc_clear(struct qbman_release_desc *d);
void qbman_release_desc_set_bpid(struct qbman_release_desc *d, u16 bpid);
void qbman_release_desc_set_rcdi(struct qbman_release_desc *d, int enable);
-int qbman_swp_release(struct qbman_swp *s, const struct qbman_release_desc *d,
- const u64 *buffers, unsigned int num_buffers);
int qbman_swp_acquire(struct qbman_swp *s, u16 bpid, u64 *buffers,
unsigned int num_buffers);
int qbman_swp_alt_fq_state(struct qbman_swp *s, u32 fqid,
@@ -194,6 +253,61 @@ void qbman_swp_mc_submit(struct qbman_swp *p, void *cmd, u8 cmd_verb);
void *qbman_swp_mc_result(struct qbman_swp *p);
/**
+ * qbman_swp_enqueue() - Issue an enqueue command
+ * @s: the software portal used for enqueue
+ * @d: the enqueue descriptor
+ * @fd: the frame descriptor to be enqueued
+ *
+ * Return 0 for successful enqueue, -EBUSY if the EQCR is not ready.
+ */
+static inline int
+qbman_swp_enqueue(struct qbman_swp *s, const struct qbman_eq_desc *d,
+ const struct dpaa2_fd *fd)
+{
+ return qbman_swp_enqueue_ring_mode_ptr(s, d, fd);
+}
+
+/**
+ * qbman_swp_enqueue_multiple() - Issue a multi enqueue command
+ * using one enqueue descriptor
+ * @s: the software portal used for enqueue
+ * @d: the enqueue descriptor
+ * @fd: table pointer of frame descriptor table to be enqueued
+ * @flags: table pointer of QBMAN_ENQUEUE_FLAG_DCA flags, not used if NULL
+ * @num_frames: number of fd to be enqueued
+ *
+ * Return the number of fd enqueued, or a negative error number.
+ */
+static inline int
+qbman_swp_enqueue_multiple(struct qbman_swp *s,
+ const struct qbman_eq_desc *d,
+ const struct dpaa2_fd *fd,
+ uint32_t *flags,
+ int num_frames)
+{
+ return qbman_swp_enqueue_multiple_ptr(s, d, fd, flags, num_frames);
+}
+
+/**
+ * qbman_swp_enqueue_multiple_desc() - Issue a multi enqueue command
+ * using multiple enqueue descriptor
+ * @s: the software portal used for enqueue
+ * @d: table of minimal enqueue descriptor
+ * @fd: table pointer of frame descriptor table to be enqueued
+ * @num_frames: number of fd to be enqueued
+ *
+ * Return the number of fd enqueued, or a negative error number.
+ */
+static inline int
+qbman_swp_enqueue_multiple_desc(struct qbman_swp *s,
+ const struct qbman_eq_desc_min *d,
+ const struct dpaa2_fd *fd,
+ int num_frames)
+{
+ return qbman_swp_enqueue_multiple_desc_ptr(s, d, fd, num_frames);
+}
+
+/**
* qbman_result_is_DQ() - check if the dequeue result is a dequeue response
* @dq: the dequeue result to be checked
*
@@ -504,4 +618,49 @@ int qbman_bp_query(struct qbman_swp *s, u16 bpid,
u32 qbman_bp_info_num_free_bufs(struct qbman_bp_query_rslt *a);
+/**
+ * qbman_swp_release() - Issue a buffer release command
+ * @s: the software portal object
+ * @d: the release descriptor
+ * @buffers: a pointer pointing to the buffer address to be released
+ * @num_buffers: number of buffers to be released, must be less than 8
+ *
+ * Return 0 for success, -EBUSY if the release command ring is not ready.
+ */
+static inline int qbman_swp_release(struct qbman_swp *s,
+ const struct qbman_release_desc *d,
+ const u64 *buffers,
+ unsigned int num_buffers)
+{
+ return qbman_swp_release_ptr(s, d, buffers, num_buffers);
+}
+
+/**
+ * qbman_swp_pull() - Issue the pull dequeue command
+ * @s: the software portal object
+ * @d: the software portal descriptor which has been configured with
+ * the set of qbman_pull_desc_set_*() calls
+ *
+ * Return 0 for success, and -EBUSY if the software portal is not ready
+ * to do pull dequeue.
+ */
+static inline int qbman_swp_pull(struct qbman_swp *s,
+ struct qbman_pull_desc *d)
+{
+ return qbman_swp_pull_ptr(s, d);
+}
+
+/**
+ * qbman_swp_dqrr_next() - Get an valid DQRR entry
+ * @s: the software portal object
+ *
+ * Return NULL if there are no unconsumed DQRR entries. Return a DQRR entry
+ * only once, so repeated calls can return a sequence of DQRR entries, without
+ * requiring they be consumed immediately or in any particular order.
+ */
+static inline const struct dpaa2_dq *qbman_swp_dqrr_next(struct qbman_swp *s)
+{
+ return qbman_swp_dqrr_next_ptr(s);
+}
+
#endif /* __FSL_QBMAN_PORTAL_H */
--
2.7.4
^ permalink raw reply related
* [Bug 204479] KASAN hit at modprobe zram
From: bugzilla-daemon @ 2019-08-08 21:55 UTC (permalink / raw)
To: linuxppc-dev
In-Reply-To: <bug-204479-206035@https.bugzilla.kernel.org/>
https://bugzilla.kernel.org/show_bug.cgi?id=204479
Erhard F. (erhard_f@mailbox.org) changed:
What |Removed |Added
----------------------------------------------------------------------------
Attachment #284177|0 |1
is obsolete| |
--- Comment #5 from Erhard F. (erhard_f@mailbox.org) ---
Created attachment 284271
--> https://bugzilla.kernel.org/attachment.cgi?id=284271&action=edit
kernel .config (5.3-rc3, PowerMac G4 DP)
--
You are receiving this mail because:
You are on the CC list for the bug.
^ permalink raw reply
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox